We integrated a rigid optical trap into a tunable pulsed cavity ringdown spectroscopy (OT-CRDS) system to characterize the extinction of single airborne particles in the UV spectral region (306-315 nm). Single solid particles from a multi-walled carbon nanotube (MWCNT), Bermuda grass smut spore, carbon microsphere, and blackened polyethylene microsphere were trapped in air based on the photophoretic force. The improved OT-CRDS system was highly sensitive and able to resolve extinctions of single particles from different materials and sizes at a given wavelength. Further, we successfully manipulated the number of particles, e.g., 1, 2 or more particles, in the trap and measured their distinguishable extinctions using the OT-CRDS. We also show that the particle size and extinction have a good linear correlation from the measurements of 24 single MWCNT particles. Material- and wavelength-dependent extinctions of the four types of airborne particles were also characterized. Results reveal that single airborne particles regardless of their differences in material and size, due to their heterogeneous morphology, have individual-particle dependent extinctions and that dependence can be resolved and characterized using the OT-CRDS technique.
© 2017 Optical Society of America
Aerosol is related to a wide range of research disciplines, including meteorology, atmospheric chemistry, planetary science, combustion, health, etc. Current studies of aerosol properties mainly rely on the measurements of aerosol particle ensembles or agglomerates [1,2]. A particle ensemble not only has great complexity of properties, such as heterogeneous mass distribution, various chemical makeup, physical phase, and water concentration, but also contains intricate interactions among particles in the ensemble. A single aerosol particle is the fundamental unit of the particle ensembles or agglomerates. Single aerosol particle studies offer better understanding of the aerosol’s physical and chemical properties at the fundamental level. Characterization of single particles requires two basic techniques for: 1) trapping, manipulating, and transporting a single particle and 2) highly sensitive and selective measurement. Among the various techniques for particle trapping, optical trapping is an attractive tool to manipulate single particles [3–5]. One of the optical trapping techniques is the optical tweezers, which relies on the radiation pressure force (RPF) to trap non-absorbing particles, typically in a solution . The technique employs high numerical aperture (NA) optics to tightly focus a laser beam and creates a strong gradient force to stabilize the trapped particles. Another optical trapping technique is the photophoretic trapping. Unlike the optical tweezers, the photophoretic trapping is based on the photophoresis phenomenon that is induced by interactions between an absorbing particle and its surrounding gas molecules. Typically, a photophoretic force (PPF) consists of two components, ΔT- and Δα-forces that govern complex motions of the particles inside a trapping (or illuminating) beam, such as positive, negative, circular, helix, or any combined motions [7,8]. The ΔT-force originates from the temperature gradient along the particle surface; the Δα-force is caused by the inhomogeneous thermal accommodation coefficients of the particles. In general, the amplitude of the PPF is 2-4 orders of magnitude larger than the RPF for an absorbing particle. This property makes the PPF particularly suitable for trapping and manipulating absorbing particles including airborne particles . The development of optical trapping and manipulating techniques using the PPF has gained a strong momentum over the last decades [3,10–18].
Comparatively, the development of measurement techniques for single particles is still in its infancy. The spectroscopy based technologies, for example, Raman and cavity ringdown spectroscopy (CRDS), are highly sensitive as well as chemically selective. They are particularly attractive in characterization of single aerosol particles [5,19]. The first Raman spectra of single particles in air were measured from a levitated glass sphere in 1984 . Later, the optical trap-Raman spectroscopy (OT-RS) was introduced to characterize single carbon nanotube particles [13,14], bioaerosol particles , droplets [18,22], and non-absorbing solid particles (glass sphere, silica beads, etc.) [e.g., 15].
CRDS has been applied to aerosol characterization in both laboratory studies and field measurements . The early CRDS measurements of aerosol particles were made by introducing aerosol particles into a closed ringdown cavity, and the optical extinctions of the aerosol particles were determined [23–27]. Later, the CRDS technique was extended to the measurements of extinction efficiency (Qext) , single scattering albedo (SSA) [29,30], complex refractive indices (RI) , etc. However, the CRDS technique had not been applied to measure single aerosol particles until Orr-Ewing and associates reported the first CRDS measurements of single droplets flowing through the ringdown beam . Later, Reid and associates optically trapped single droplets used a tightly focused Bessel beam and measured their extinction efficiencies using the continuous wave (cw)-CRDS at 532 nm . Using the similar optical trap, recently, the same group has extended their CRDS measurements to the dynamic change of the optical cross section of single droplets , variations of extinction cross sections and refractive indices of individual hydroscopic droplets with single or binary chemical components , and the accuracy of refractive index (RI) retrievals . More recently, Wang et al have demonstrated CRDS measurements of large, irregularly shaped, single solid particles photophoretically trapped in air in the UV spectral region . The combination of the CRDS technique with an imaging system not only measures single particle extinction, but also reveals the single particle’s oscillations associated with the thermal relaxation of the particle under photophoresis.
In the present study, we extended the newly developed OT-CRDS technique to characterize four types of single solid airborne particles: multi-walled carbon nanotube (MWCNT), Bermuda grass smut spore, carbon microsphere, and blackened polyethylene microsphere (PE microsphere). In order to do so, we integrated a more efficient, robust, and flexible optical trap into the CRDS system. The new OT-CRDS system offers a tunable wavelength range of 306.0-315.0 nm, with good ringdown stability and reproducibility. The rigid optical trap facilities the trapped particle manipulations and control, e.g., precisely transport a trapped particle from one location to another and control the number of trapped particles in the trap. The high stability and sensitivity of the CRDS system enables measurements of the individual-particle dependent extinctions that are differentiated by each individual particle regardless of its material or size. Finally, the wavelength-dependent extinction of single particles was measured by scanning the tunable OT-CRDS system in the range of 306.0 - 315.0 nm at a step of 0.5 nm.
2. Experimental setup
Figure 1(b) illustrates the experimental configuration of the improved optical trap-cavity ringdown spectroscopy (OT-CRDS) system. Compared to the system used in the previous work , more robust particle trapping and higher flexibility of the particle manipulation were achieved using a confocal hollow beam trapping configuration. The trapping laser beam (along the x-axis) was a near TEM00 mode Gaussian beam from a cw 532 nm laser with the maximum power of 3.0 W. The Gaussian beam was converted to a hollow beam using an axicon (cone angle = 170°, Thorlabs). The hollow beam diverged after the axicon and was then focused by an aspheric lens (NA = 0.66, diameter = 25 mm, back focal length = 12 mm, Edmund optics). Size of the incident hollow beam was controlled by the distance between the axicon and the aspheric lens. The thickness of the hollow ring was adjusted by an iris that was placed before the axicon, which changed the incident Gaussian beam size. In this work, the diameter of the hollow beam was ~18.0 mm before being focused by the aspheric lens. The ring thickness was ~1.0 mm. A concave mirror (f = 10.0 mm) was placed behind the aspheric lens to reflect and re-converge the hollow beam. Both the aspheric lens and the concave mirror were mounted on 3D translation stages, whose spatial resolution was as high as 25.4 µm/div. The foci of the aspheric lens and the concave mirror were separated by tens of microns, and a symmetric hollow trapping zone was therefore formed, as shown in Fig. 1(a). In the experiments, the separation distance can be fine tuned in order to manipulate the number of particles in the trap.
The particles were trapped in the center of a custom designed quartz cuvette (120 × 40 × 12.5 mm3). The ringdown beam passed the trapping cell via two round holes (mm) opened in both ends. The round holes had a larger size than the two rectangular slits used in the previous setup . This modification provided more spatial flexibility to tune a trapped particle’s position with respect to the ringdown beam (along the x-axis). All the trapping optics were integrated on a rail and then mounted onto a 2D translation stage. The design enabled the spatial tuning of the trapped particle along the x- or z-axis. The y-axis was set to be approximately overlapped with the optical axis of the cavity and orthogonal to the trapping beam axis. The origin of the coordinates was set at the common center of the cavity and the ringdown beam.
The trapped particles were observed by a microscopic imaging system from the top of the trapping cell (along the z-axis). The imaging optics included a micro objective (NA = 0.42, 20 × , Mitutoyo), a set of tube lens (InfiniTube, Edmund Optics), and a fast CMOS camera (pixel pitch = 3.45 µm, Point Grey), and all of the parts were integrated into a compact tube system. The camera used in this study had a much faster frame rate and a higher sensitivity than that used in the previous work . The present imaging system enables observations of fast particle movements. It also provides twice larger magnification and a better correction of image aberrations, which gives more accurate results on particle size and shape estimation. The shutter speed of camera was typically set at 1.0 ms. The pixel resolution of this image system was equivalent to 0.18 ± 0.01 µm/pixel, which was calibrated using a single mode optical fiber (diameter = 125 µm) and a standard microscope calibration slide with a fine scale of 10 µm/div. Here the particle size was interpreted as the diameter d (for spherical shape) or geometric cross section (the area of the closed contour of the particle image, for irregular shape). Therefore, the size of trapped particles can be estimated by counting the pixels in the particle images.
A typical CRDS system consists of a pulsed laser source, a high finesse cavity, a signal detector, and a data processing system. A laser pulse is injected into the cavity that is usually formed by two high reflectivity mirrors (R > 99.9%). The laser pulse is then reflected back and forth inside the cavity for up to thousands times; and each time a small fraction of the laser pulse escapes from the cavity and is detected by the detector. The amplitude of the escaped light follows an exponential decay under a normal condition, whose decay constant is called ringdown time. Without particles inside the cavity, the baseline ringdown time is determined by the reflectivity of the mirrors and the scattering loss of the background molecules inside the cavity. When a single aerosol particle is present inside the ringdown beam, the absorption and scattering of the particle will cause an extra optical loss and change the ringdown time to a lower value . Then the dimensionless extinction (the sum of the absorption and scattering) of a single particle can be determined by the following formula [32,37]:38]. Therefore the extinction merely depends on the particle’s cross section , and the local ringdown beam intensity i at the position (x, z) where the particle is trapped. The extinction is defined as [32,33]:Eq. (2) can be written as:Eq. (3) to reveal the extinction cross section of the single particle. When the geometric cross section of the particle () is obtained, the extinction efficiency is therefore determined .
The laser source of our CRDS system was from the frequency doubling of a tunable dye laser system (Narrowscan, Radiant), which was pumped by a Q-switch Nd:YAG laser (DLS8020, Continuum) at 532 nm. The laser pulse repetition rate was 20 Hz. The pulse width was 10 ns. The minimum scanning step of the laser was 0.0001 nm. The dye laser was able to operate in the wavelength range of 607.0-676.0 nm using the DCM dye. We selected a narrow range between 612.0 and 630.0 nm for the better performance. A tunable wavelength range of 306.0-315.0 nm was generated by the second harmonic generator. The ringdown beam was reshaped to be ~1.0 mm in diameter prior to its injection into the ringdown cavity. The mirror reflectivity was R ≈99.95% at 308 nm. The two mirrors, with a radius of curvature of 1.0 m, were separated by a distance of 1.0 m. The estimated beam waist inside the ringdown cavity was ~500 ± 150 µm. The detector was a photomultiplier tube module (PMT, H9375, Hamamatsu). A narrow band pass filter (the central wavelength = 308 nm, FWHM = 12 ± 2 nm, Newport) was placed before the PMT. The ringdown decay was viewed by an oscilloscope. 128 ringdown events were averaged to generate one data point of ringdown time.
The particle samples used in the experiments were multi-walled carbon nanotubes (MWCNT, outside diameter = 30-50 nm, length = 10-20 µm, US Research Nanomaterials), Bermuda grass smut spores (5-7 µm, Greer), carbon microspheres (glassy carbon spherical particles, 2-12 µm, Sigma-Aldrich), and blackened polyethylene microspheres (PE microsphere, 10-20 µm, Cospheric). The black coating on the PE microspheres is paramagnetic iron manganese oxide (~30%). The particles were injected to the trapping cell by a disposal transfer pipette (Fisher Scientific). Thousands of particles were introduced into the trapping zone each time. It took several minutes for the floating particles to fall down while a single particle or a cluster of particles was stably trapped.
3. Results and discussion
3.1 Particles trapping and manipulation
The optical trap used in the present study enabled us to trap different types of particles and manipulate the number of the trapped particles inside the trap. Due to the structure of the optical field around the trapping zone, the optical trap was also able to trap multiple particles or a particle cluster. In general, we used two methods to manipulate the number of particles in the trap: altering the optical field of the trapping zone and introducing an external perturbation (turbulence, vibration or airflow). On the one hand, the trapping zone is formed by the overlap of the vertices of two confocal hollow beams, and the size of trapping zone can be adjusted by changing the separation distance between the aspheric lens and the concave mirror (see Fig. 1(a)). The strength of the optical field around the trapping zone can be altered by changing the trapping beam power. On the other hand, the optical trap has different trapping robustness and stability for different particles in the trap. An external turbulence, mechanical vibration or gentle airflow may kick out some relatively-loosely trapped particles. The strength of the external perturbation was qualitatively controlled in the experiment. Mechanical vibrations were introduced by tapping the translation stages with different forces; the airflow was generated by gently waving a business card with different speeds near the trapping cell. In practice, it required a careful handling to manipulate the number of particles in the trap. In many cases, all the methods above were used in combination to achieve the control of the particle number in the trap.
With the fine tuning of the trapping zone or introduction of an external turbulence, the size, particle number, and distribution of the trapped particles inside the trap can be selected for OT-CRDS measurements. Although it is difficult to completely control an exact number as desired, it is possible to alter the overall properties of the particle clusters. Figure 2 shows that a cluster of trapped Bermuda grass smut spores was trapped and the number of trapped particles in the cluster was manipulated from 4 particles initially to 0 particle finally (see Visualization 1). At each stage, the trapped particle cluster stayed for up to hours, but the extinction measurements using the OT-CRDS were conducted for about two minutes. In Fig. 2(a), the four trapped particles were located in the different positions along the z-axis; therefore two of them were not in the same focal plane of the imaging system, which led to the blurring images and certain scattering patterns. In Fig. 2(b), the two particles were kicked out by using the manually controlled turbulence. Another two relatively more rigidly trapped particles remained in the trap, but they still had different trapping robustness as they were located in the different positions inside the trap. Visualization 1 shows the process of the top-left particle seeking a rigid position in the trapping zone, while the bottom-right particle was jittering. After they were stabilized, their extinctions were measured. With a stronger external turbulence introduced, the bottom-right particle was kicked out, and a single particle remained trapped, as shown in Fig. 2(c). Finally, the last particle was knocked by using even stronger external turbulence. If the trapped particle was trapped too rigid to be kicked out by external turbulence, then it was removed by blocking the trapping beam. Figure 2(d) shows no particle was trapped in the trapping zone, in which the measured ringdown time was the baseline . The video for each stage in Visualization 1 was truncated to reduce the file size.
The trapped particles may have oscillations caused by the external air turbulence and/or different morphologies of the particles . Figure 3 shows typical oscillation amplitudes for each type of particles. The particle’s motion was observed by the microscopic imaging system. The trapped particles were located at the center of image and their positions along the x and y axes were tracked frame by frame using an in-house Matlab program. The frame rate of the videos for each particle varies from 346 to 770 frames per second due to the size of the image frames and speed of the data transmission. The length of the recording was over 15.0 s. Since the overall oscillation trends are quite uniform, only part of the recorded oscillations of each particle is plotted in Fig. 3. Both the oscillation amplitudes (the x- and y-components) and the oscillation patterns (frequency and uniformity) are different for different particles. For instance, the radial and axial oscillations of the single Bermuda particle in Fig. 3(b) are nearly in phase, while the two oscillation components for other particles are out of phase. The amplitude and phase differences between radial and axial oscillations imply the orientation of the oscillation path. In Fig. 3(b), the uniform oscillation data suggested a diagonal path of the oscillation and that was confirmed by the recorded video. However, the oscillation amplitude along the y-axis (radial) is smaller than that along the x-axis (axial) in all cases, which means that the radial displacement is tightly confined by the optical field near the trapping zone. Most of the oscillation amplitudes are smaller than the particle size.
3.2 OT-CRDS sensitivity and extinction of single particles
The stability, the repeatability of the extinction measurements, and the sensitivity of this OT-CRDS system were evaluated. Figure 4(a) shows a stable ringdown time baseline of the OT-CRDS system at 308.0 nm. Each ringdown time was determined by the averaging over 128 ringdown events. The averaged ringdown time from a 30 minute test is which yields the baseline stability of based on the 1-σ criteria. Although the ringdown baseline changed over the tunable range (306-315 nm) due to the wavelength-dependent coatings of the ringdown mirrors, the variations in the ringdown baseline stability were small, e.g., 0.41 ± 0.20% over the wavelength range.
Figure 4(b) shows reproducible extinction measurements of a single Bermuda grass smut spore trapped. The upper dots in Fig. 4(b) are the maximum extinctions measured when the particle was trapped at the center of the ringdown beam; the lower dots denote the extinction baseline when the particle was completely outside the ringdown beam. The trapped particle was moved in or out of the ringdown beam by the translation stage along the x-axis. The height of the trapped particle was fixed at , where the extinction reached a peak along the vertical orientation. The x position of the particle was freely manipulated at a fine step of 25.4 µm, which was read from the scale of the micrometers on the translation stage. The resolution of the position manipulation can be further improved by using a differential micrometer with a resolution of 0.1 µm.
The extinction of a single particle depends on the particle’s position in the ringdown beam (see Eq. (3)). Figure 4(c) shows the measured extinctions of a single MWCNT particle at different locations inside the ringdown beam. The total scanning range of the particle’s position along the x-axis was 1778.0 µm with a scanning step of 127.0 µm. The square dots indicate the measured extinctions. Each data point represents the mean value of 20 measured results. The red curve is the fitting result using Eq. (3), which yields the extinction cross section and the beam waist w = 0.61 ± 0.03 µm. The geometric cross section of this MWCNT particle was estimated to be . Therefore, the extinction efficiency is .
The current CRDS system is sufficiently sensitive to be able to distinguish different numbers of particles trapped in the trap. Figure 4(d) shows the distinct extinctions measured with different numbers of the particles in the trap, corresponding to the cases shown in Fig. 2. The first step of the extinction in Fig. 4(d) was measured when the 4 particles were trapped, as shown in Fig. 2(a); the next three steps denote the measured extinctions of the trapping states shown in Figs. 2(b)-2(d), respectively. When all the particles were kicked out, the extinction reproducibly went back to the baseline at the last step. The extinction does not decrease linearly as the number of particles decreases, which can be explained by the different sizes, morphologies and trapping positions of each particle.
3.3 Size-dependent extinction of single particles
Excluding the factors of particle material and laser wavelength, we measured the extinctions of 24 trapped MWCNT particles at 308 nm and their geometric cross sections . The curve of the extinction efficiency () versus the size parameter (, r is the radius of the particle) exhibits an interference structure and approaches to the limit of ~2.0 when the x is sufficiently large (e.g., , namely, ) due to the “extinction paradox” . Therefore, when the particle diameter d is larger than , the total extinction of the particle is proportional to its geometric cross section. Figure 5 shows the scatter graph of the total extinction versus the geometric cross section of the particles. The extinctions were measured for more than three minutes when the particles were stably trapped. After the extinction measurement, the particle was kicked out by blocking the OT beam, and the ringdown time baseline was measured subsequently. Therefore the extinction of each particle can be calculated using Eq. (1), and the vertical error bars of extinction were determined by the stability of measured ringdown times. The geometric cross section of a single particle was calculated by counting the pixels occupied by the particle in the image. The horizontal error bars of geometric cross section were obtained from the fittings of several different images of the same particle. The results of the 24 MWCNT particles show a good linear relation between the total extinction and the geometric cross section (). As the trapped particles are not spherical, the geometric cross section imaged from the dimensions of the particle in the x-y plane is slightly different from that in the x-z plane where the extinction cross section was measured. This factor also contributes some uncertainty to the linear correlation.
3.4 Material-dependent extinction of single particles
Figure 5 shows the repeatedly measured extinctions of the four types of particles at the single wavelength 308.0 nm that was used without any preference. The extinctions of each particle were measured by moving the translation stage back and forth to place the particle inside and outside the center of the ringdown beam. Figures 6(a)-6(d) show the images of single trapped MWCNT, Bermuda, carbon microsphere, and PE microsphere, respectively. The images were taken by using the microscopic imaging system with a shutter speed of 1.0 ms. The images of some spherical particles are not exactly circular. For example, the image of PE microspheres in Fig. 6(d) shows an elliptical shape, which may be from a package of two or more microspheres. From the previous section, larger particles tend to have a higher extinction for the same particle materials. However, the extinctions of the particles in Figs. 6(a)-6(d) do not follow the sole dependence on the particle size because the distinct particle materials also play a role. For example, the single Bermuda particle in Fig. 6(b) has larger size but significantly lower extinction than the single MWCNT particle, as shown in Fig. 6(a), because Bermuda particles are a type of smut spores with a dark brown color, but MWCNT particles contain more than 99.9% carbon nanotubes.
Furthermore, the properties of particle materials also affect the morphology of the particle, which is also one of the important factors related to the extinctions of the single particle. For example, the carbon microspheres are solid microspheres, while the MWCNT consists of long nanotubes with a length of 10-20 µm and outer diameter of 30-50 nm. Therefore, each trapped MWCNT particle might be a tightly or loosely tangled cluster of nanotubes, which leads to a totally different shape, surface structure and thermal property as compared to carbon microspheres. Therefore, besides the particle size, the particle materials also account for the difference in extinction among those particles shown in Fig. 6.
3.5 Wavelength-dependent extinction of single particles
The wavelength-dependent optical properties of aerosols are imperative since the sunlight is not monochromatic and it occupies a wide spectral range. However, most of the extinctions of aerosol particles are measured at fixed wavelengths in the literatures (e.g., 355 nm [23,40], 405 nm , 532 nm [23–26,29,33], 1064 nm , etc.). The wavelength resolved extinction of droplets over 540-570 nm has been reported very recently . The capability of measuring extinction spectra of single particles in air at two UV wavelengths has been demonstrated in our previous work . In this paper, we extended the extinction measurements to the full range of the wavelength (306-315 nm) at a relatively high spectral resolution of 0.5 nm and to four different types of particles, so that the resolving power of the OT-CRDS system with respect to an individual particle and wavelength was fully examined.
Figure 7 shows the wavelength-resolved extinctions of the four types of particles in the UV range of 306-315 nm with a wavelength resolution of 0.5 nm. About 10 ringdown time data (1280 ringdown events) were averaged to generate one data point at each wavelength. The single particles were stably trapped and placed at the origin of the coordinates, namely, at the center of the cavity and the ringdown beam. The baseline data set of an empty cavity (without particles) were measured before the particle was trapped and right after the particle was kicked out. The wavelength-dependent extinctions were determined using Eq. (1). It took ~30 minutes to collect one set of the wavelength scanning data, and at least 2.5 hours to complete the measurement of one type of particle. It was the robust optical trapping and the stable ringdown system that allowed us to perform the time-consuming measurements of the wavelength-dependent extinctions.
Figure 7(a) shows the three sets of extinction data for a single MWCNT particle. The microscopic image shows a loose internal structure of the trapped particle. The scale bar on the image is 10.0 µm. The extinction curves show a slowly decreasing tendency in the range of 306.0-315.0 nm. Figure 7(b) shows the three sets of extinction data for a single Bermuda grass smut spore, which reveals an insensitive response to the wavelength variations in the range of 306.0-315.0 nm. The internal structure of the Bermuda particle is solid while the extinction tends to be smaller than that of the MWCNT, which may be explained by the fact that the material of Bermuda grass smut spores is less absorptive than carbon nanotubes in this wavelength range. Figure 7(c) shows the four sets of extinction data for a single carbon microsphere. Although the main compositions of carbon microsphere and MWCNT are the same, they have different wavelength-dependent extinctions. Different morphology may influence the scattering of the single particle, and different density may have impact on the absorption and optical depth. Figure 7(d) shows the three sets of extinction data for a single PE microsphere. The decreasing pattern of the extinction curves, which have a relative flat region from 308.5 to 313.0 nm, is different from that of the MWCNT (Fig. 7(a)). The single PE microsphere has much larger extinction than other types of particles, which may be due to the larger size of PE microspheres.
Several factors affect the spectral features from the different types of particles. In the UV region, the size parameter of a large particle may influence the amplitude of extinction only; but the particle material and morphology might be the essential factors that account for the spectral features, mainly, in the absorption part of the extinction. For example, the MWCNT and Bermuda particles, shown in Figs. 7(a) and 7(b), have similar size and shape; however, their wavelength-dependent extinction patterns are significantly different. Further, the particle’s morphology may mainly influence the scattering part. The MWCNT and carbon microspheres in Figs. 7(a) and 7(c) are both carbon based materials but have different morphologies, thus the spectral features of their extinction curves are different from each other. Therefore, when we study the extinction spectra of single particles, we need to consider the compound effect of their morphology and materials on the scattering and absorption components in the total extinction.
We improved the OT-CRDS by integrating a rigid optical trap into a tunable pulsed CRDS system to characterize single airborne particles. With the benefits from the photophoretic trap and highly sensitive CRDS technique, we were able to trap and manipulate different types of single particles in air and measure their extinctions. Four types of particles (MWCNT, Bermuda grass smut spore, carbon microsphere, and PE microsphere) were measured using the improved OT-CRDS. The optical trap using a confocal hollow beam was capable of stably trapping single particles of each type. This robust trapping configuration allowed us to manipulate the spatial position of trapped particles (along the x- and z-axes) as well as the number of particles in the trap (from 4 to 0 particles). Small oscillations of single trapped particles were viewed using the microscopic imaging system which shows that the oscillation amplitude along the axial direction (the x-axis) is always larger than that of the radial direction (the y-axis). The OT-CRDS system has a tunable range of 306-315 nm. The ringdown baseline stability is 0.41% at 308 nm with the average over 128 ringdown events. The system is highly sensitive and able to resolve extinctions from the trapped particles’ size, material, or morphology at a given laser wavelength or from the same trapped particle in different wavelengths at a wavelength resolution of 0.5 nm. The results of the size-, material-, and wavelength-dependent extinctions measured using the OT-CRDS system reveal that the single particle’s extinction is governed not only by the size, material and wavelength, but also by the particle’s morphology. The extinction is individual-particle dependent. The dependence can be resolved and characterized by this improved tunable, pulsed OT-CRDS technique.
This research is supported by the U.S. Army Research Office Grants W911NF-13-1-0429 and W911NF-16-1-0483.
References and links
1. U. K. Krieger, C. Marcolli, and J. P. Reid, “Exploring the complexity of aerosol particle properties and processes using single particle techniques,” Chem. Soc. Rev. 41(19), 6631–6662 (2012). [CrossRef] [PubMed]
2. P. H. McMurry, “A review of atmospheric aerosol measurements,” Atmos. Environ. 34(12-14), 1959–1999 (2000). [CrossRef]
4. A. Ashkin, “History of optical trapping and manipulation of small-neutral particle, atoms, and molecules,” IEEE J. Sel. Top. Quantum Electron. 6(6), 841–856 (2000). [CrossRef]
7. O. Jovanovic, “Photophoresis-light induced motion of particles suspended in gas,” J. Quant. Spectrosc. Radiat. Transf. 110(11), 889–901 (2009). [CrossRef]
8. H. Horvath, “Photophoresis – a forgotten force?” KONA Powder Part. J. 31(0), 181–199 (2014). [CrossRef]
12. P. Zhang, Z. Zhang, J. Prakash, S. Huang, D. Hernandez, M. Salazar, D. N. Christodoulides, and Z. Chen, “Trapping and transporting aerosols with a single optical bottle beam generated by moiré techniques,” Opt. Lett. 36(8), 1491–1493 (2011). [CrossRef] [PubMed]
13. Y.-L. Pan, S. C. Hill, and M. Coleman, “Photophoretic trapping of absorbing particles in air and measurement of their single-particle Raman spectra,” Opt. Express 20(5), 5325–5334 (2012). [CrossRef] [PubMed]
15. L. Rkiouak, M. J. Tang, J. C. J. Camp, J. McGregor, I. M. Watson, R. A. Cox, M. Kalberer, A. D. Ward, and F. D. Pope, “Optical trapping and Raman spectroscopy of solid particles,” Phys. Chem. Chem. Phys. 16(23), 11426–11434 (2014). [CrossRef] [PubMed]
17. A. E. Carruthers, J. S. Walker, A. Casey, A. J. Orr-Ewing, and J. P. Reid, “Selection and characterization of aerosol particle size using a bessel beam optical trap for single particle analysis,” Phys. Chem. Chem. Phys. 14(19), 6741–6748 (2012). [CrossRef] [PubMed]
18. R. J. Hopkins, L. Mitchem, A. D. Ward, and J. P. Reid, “Control and characterisation of a single aerosol droplet in a single-beam gradient-force optical trap,” Phys. Chem. Chem. Phys. 6(21), 4924–4927 (2004). [CrossRef]
19. R. E. H. Miles, A. E. Carruthers, and J. P. Reid, “Novel optical techniques for measurements of light extinction, scattering and absorption by single aerosol particles,” Laser Photonics Rev. 5(4), 534–552 (2011). [CrossRef]
20. R. Thurn and W. Kiefer, “Raman-microsampling technique applying optical levitation by radiation pressure,” Appl. Spectrosc. 38(1), 78–83 (1984). [CrossRef]
21. C. Wang, Y. Pan, S. C. Hill, and B. Redding, “Photophoretic trapping-Raman spectroscopy for single pollens and fungal spores trapped in air,” J. Quant. Spectrosc. Radiat. Transf. 153, 4–12 (2015). [CrossRef]
22. L. Mitchem, J. Buajarern, R. J. Hopkins, A. D. Ward, R. J. J. Gilham, R. L. Johnston, and J. P. Reid, “Spectroscopy of growing and evaporating water droplets: exploring the variation in equilibrium droplet size with relative humidity,” J. Phys. Chem. A 110(26), 8116–8125 (2006). [CrossRef] [PubMed]
24. J. D. Smith and D. B. Atkinson, “A portable pulsed cavity ring-down transmissometer for measurement of the optical extinction of the atmospheric aerosol,” Analyst (Lond.) 126(8), 1216–1220 (2001). [CrossRef] [PubMed]
25. A. Pettersson, E. R. Lovejoy, C. A. Brock, S. S. Brown, and A. R. Ravishankara, “Measurement of aerosol optical extinction at 532 nm with pulsed cavity ring down spectroscopy,” J. Aerosol Sci. 35(8), 995–1011 (2004). [CrossRef]
26. H. Moosmüller, R. Varma, and W. P. Arnott, “Cavity ring-down and cavity-rnhanced detection techniques for the measurement of aerosol extinction,” Aerosol Sci. Technol. 39(1), 30–39 (2005). [CrossRef]
27. A. W. Strawa, R. Castaneda, T. Owano, D. S. Baer, and B. A. Paldus, “The measurement of aerosol optical properties using continuous wave cavity ring-down techniques,” J. Atmos. Ocean. Technol. 20(4), 454–465 (2003). [CrossRef]
28. V. Bulatov, M. Fisher, and I. Schechter, “Aerosol analysis by cavity-ring-down laser spectroscopy,” Anal. Chim. Acta 466(1), 1–9 (2002). [CrossRef]
30. K. D. Dial, S. Hiemstra, and J. E. Thompson, “Simultaneous measurement of optical scattering and extinction on dispersed aerosol samples,” Anal. Chem. 82(19), 7885–7896 (2010). [CrossRef] [PubMed]
31. A. A. Riziq, C. Erlick, E. Dinar, and Y. Rudich, “Optical properties of absorbing and non-absorbing aerosols retrieved by cavity ring down (CRD) spectroscopy,” Atmos. Chem. Phys. 7(6), 1523–1536 (2007). [CrossRef]
32. T. J. A. Butler, J. L. Miller, and A. J. Orr-Ewing, “Cavity ring-down spectroscopy measurements of single aerosol particle extinction. I. The effect of position of a particle within the laser beam on extinction,” J. Chem. Phys. 126(17), 174302 (2007). [CrossRef] [PubMed]
34. M. I. Cotterell, B. J. Mason, T. C. Preston, A. J. Orr-Ewing, and J. P. Reid, “Optical extinction efficiency measurements on fine and accumulation mode aerosol using single particle cavity ring-down spectroscopy,” Phys. Chem. Chem. Phys. 17(24), 15843–15856 (2015). [CrossRef] [PubMed]
35. B. J. Mason, M. I. Cotterell, T. C. Preston, A. J. Orr-Ewing, and J. P. Reid, “Direct measurements of the optical cross sections and refractive indices of individual volatile and hygroscopic aerosol particles,” J. Phys. Chem. A 119(22), 5701–5713 (2015). [CrossRef] [PubMed]
36. M. I. Cotterell, T. C. Preston, A. J. Orr-Ewing, and J. P. Reid, “Assessing the accuracy of complex refractive index retrievals from single aerosol particle cavity ring-down spectroscopy,” Aerosol Sci. Technol . 6826, 1077–1095 (2016)
37. C. Wang, Z. Gong, Y.-L. Pan, and G. Videen, “Optical trap-cavity ringdown spectroscopy (OT-CRDS) as a single-aerosol-particle-scope,” Appl. Phys. Lett. 107(24), 241903 (2015). [CrossRef]
38. B. J. Mason, J. S. Walker, J. P. Reid, and A. J. Orr-Ewing, “Deviations from plane-wave Mie scattering and precise retrieval of refractive index for a single spherical particle in an optical cavity,” J. Phys. Chem. A 118(11), 2083–2088 (2014). [CrossRef] [PubMed]
39. C. Wang, Z. Gong, Y.-L. Pan, and G. Videen, “Laser pushing or pulling of absorbing airborne particles,” Appl. Phys. Lett. 109(1), 011905 (2016). [CrossRef]
40. J. M. Flores, R. Z. Bar-Or, N. Bluvshtein, A. Abo-Riziq, A. Kostinski, S. Borrmann, I. Koren, I. Koren, and Y. Rudich, “Absorbing aerosols at high relative humidity: linking hygroscopic growth to optical properties,” Atmos. Chem. Phys. 12(12), 5511–5521 (2012). [CrossRef]
41. R. E. H. Miles, S. Rudić, A. J. Orr-Ewing, and J. P. Reid, “Measurements of the wavelength dependent extinction of aerosols by cavity ring down spectroscopy,” Phys. Chem. Chem. Phys. 12(15), 3914–3920 (2010). [CrossRef] [PubMed]