We demonstrate the use of holographic optical tweezers for trapping particles in air, specifically aerosol droplets. We show the trapping and manipulation of arrays of liquid aerosols as well as the controlled coagulation of two or more droplets. We discuss the ability of spatial light modulators to manipulate airborne droplets in real time as well as highlight the difficulties associated with loading and trapping particles in such an environment. We conclude with a discussion of some of the applications of such a technique.
© 2006 Optical Society of America
Optical manipulation techniques have matured considerably in the 35 years since Ashkin first demonstrated the use of radiation pressure to guide particles . The field was firmly established by the demonstration of the optical tweezers technique by Ashkin et al , and these have developed into a tool that is routinely used to probe biological function [3,4], colloidal dynamics [5,6], properties of light beams [7,8], and to facilitate the stable trapping and manipulation of particles at the micron scale in a wide range of disciplines. Almost the entire body of work on optical manipulation is carried out on particles suspended in a liquid medium. This medium acts to damp out the motion of particles as they are trapped and this fact combined with the buoyancy of particles in such samples makes their trapping relatively straightforward. To trap particles in the absence of such a damping medium (such as air) is more difficult and less relevant for studies to date. Ashkin studied the optical levitation of airborne droplets  and others have used these techniques to probe the size and composition of aerosols [10,11] but it is only relatively recently that their optical tweezing (that is gradient force trapping) has been demonstrated [12,13].
The ability to trap and interrogate airborne particles in a controlled manner offers much to those who wish to study aerosol properties, composition and dynamics. The localization of droplets in this way has recently been used to probe the Raman spectra of a trapped seawater droplet  as well as the controlled oxidation of oleic acid within the droplet. The stable trapping of the droplet also allowed the reaction and growth dynamics of the process to be followed. In a related experiment  the cavity enhanced Raman signal from a trapped droplet acting as a cavity was used to size droplet diameters to +/- 2nm and using a dual beam system two droplets could be controllably coagulated. Such experiments should allow systematic studies of processes on and within aerosols to be carried out with the atmospheric sciences being the main beneficiaries.
In this paper we outline the first use of holographic optical tweezers [16,17] for the manipulation of airborne particles, in this case aerosol droplets. Making use of such devices allows for arrays of particles to be trapped simultaneously, as well as allowing their controlled xyz translation. We show that controlled coagulation can be achieved and discuss the performance of the spatial light modulator in this context as well as outlining future applications of such a technique.
2. Experimental setup
Holographic tweezers are now a well established technique for the simultaneous manipulation of many particles. Recent work using these tools has explored applications in microfluidics [18,19], Raman imaging  and the creation of complex colloidal structures . Our system makes use of a Holoeye LC-R 2500 spatial light modulator (SLM) to modulate the phase of 532nm cw light from a Laser Quantum Finesse laser (maximum output 4W). A schematic of our system is shown in Fig. 1. We first rotate the laser’s plane of polarisation to optimize the efficiency of the phase modulation, then having expanded the beam via a telescope (L1 and L2) it is incident on, and covers, the SLM. Two 4f (L3 and L4, and L5 and L6) imaging systems are placed directly after the SLM to reduce the beam size to slightly overfill the back aperture of the microscope objective (Nikon CFI E Plan Achromat 100X oil, N.A. 1.25). The 4f imaging systems also make the SLM and back aperture planes conjugate. Having focused the light into the trapping plane the same objective is used for imaging.
We generate our holograms making use of a custom LabVIEW program implementing an adaptive-additive algorithm  which treats the SLM as a secondary microdisplay.
The aerosol droplets are created using an ultrasonic nebuliser (Omron U22(NE-U22-E)), which produces aerosols with a mass median aerodynamic diameter (MMAD) between approximately 3 and 5 microns . Custom glassware is attached to the nebuliser allowing the droplets to be fed accurately into a sealed glass cube, with an inlet and outlet hole in either side, which acts as our trapping chamber. The bottom of the cube is formed from a glass coverslip having been soaked in a 2% weight by volume solution of sodium dodecyl sulphate, for reasons explained in the discussion. To help control the humidity inside the cell we place a small water soaked paper tissue in the cube. In the experiments making use of one type of aerosol we block the second hole to avoid unnecessary air currents.
For the work outlined here we make use of water aerosols. However, water droplets generally only exist in oversaturated atmospheres (that is with a relative humidity greater than 100%) so we use aqueous solutions of NaCl to decrease the vapour pressure of the droplets which allows them to form in relative humidity <100% . The majority of the work has been carried out with 0.34817 Molar concentrations  but others have been tested, for example 40mM and seawater, which provide similar results, although they do alter the droplet dynamics. The ambient laboratory temperature is 22.5 ± 1°C.
3. Experimental results
We are able to trap arrays of aerosol droplets as shown in Fig. 2, which also outlines one of the difficulties of trapping in air. In a liquid medium it can be hard to reliably load a large array of trap sites with colloidal particles. One has to actively seek out the particles, but this can result in particles falling out of the trap or in multiple particles being trapped at a single site. Alternatively the trap can be static with particles flown in to try and fill the sites and this can lead to jamming, or particles knocking each other out. In air we have no choice but to wait for particles to fall into the traps. In the experimental geometry trap loading is far from optimized, and the images shown illustrate that although we have a relatively simple trap pattern, which would be easy to fill in a liquid, we may be unable to fill the sites at all if no aerosol flies into the trapping region of that site.
The current system allows xy translation of droplets in single jumps of up to 3.40 ± 0.02μm, which combined with a high refresh rate of ~25 frames per second allows xy translation at speeds up to 85μms-1. The precision to which we can translate droplets in x and y is 227 ± 1nm with control of droplets over the entire field of view; 45 by 60μm. The apparatus also allows z control of the droplets as shown in Fig. 3.
We are able to trap a range of droplet sizes from ~2.5 to 12μm in diameter and determine that the minimum power required to trap an aerosol is 0.38 ± 0.02mW and an axial Q value of 0.22 ± 0.01, which are all comparable to results in conventional tweezers . The droplet size is dependent on power with larger droplets requiring greater powers, as shown in Fig. 4.
This stable trapping and precise manipulation of water droplets allows not only their individual control but also the ability to coagulate multiple droplets both of which are demonstrated in Fig. 5.
One may imagine the Brownian motion of aerosol particles is too high, and the SLM refresh rate too slow to enable such precise control of water droplets. However, by calculating the displacement due to Brownian diffusivity using Eq. (1):
where k is Boltzmann’s constant, T is the temperature in Kelvin, Cc is the slip correction factor, μ is the viscosity of the surrounding fluid (air), Dp is the droplet diameter, and t is the time. Plotting the results, Fig 6, it is clear that for the droplet sizes studied here the displacement due to Brownian motion after the SLM response time of 16ms is significantly below a single droplet radius .
4. Discussion and conclusion
Clearly from Figs. 3, 4, and 6 a major hurdle in quantitative analysis is the inability to accurately size droplets from video images alone, due not only to their dynamic nature but also to the poor definition of their edges on the video output. However, precise droplet size measurement techniques have already been demonstrated  allowing for future accuracy improvement and more precise quantitative results.
The relationship between laser power and droplet size combined with the ability of holographic optical tweezers to control each individual trap site’s intensity will eventually allow some degree of size selective trapping.
One problem encountered was the formation of water ‘puddles’ from parts of the aerosol cloud settling onto the microscope slide. Measuring tens of microns across they create sufficient aberrations in the beam to cause the loss of droplets from traps unless they are sited directly over the puddle’s centre where curvature is at a minimum. As a solution we applied a detergent surfactant to the slide thus increasing the wetting and reducing the formation of puddles.
The quantitative results shown above are specific to one concentration of sodium chloride solution, with any alteration in the aerosol composition significantly altering the droplet’s properties. This obviously increases the complexity over simple monodispersed or even polydispersed microspheres as they have fixed dimensions and properties where as droplets can condense, evaporate, absorb, and constantly interact with their surroundings.
The experiments described above show for the first time the optical trapping and control of arrays of aerosol particles. Further they demonstrate the applicability of using the slow (in comparison to acousto-optical deflection techniques) phase only SLMs to manipulate airborne particles and to coagulate them. However issues remain with the techniques as described for some applications. Better kinoform generation algorithms need to be implemented  in order to reduce the effects of ghost traps and increase the intensity uniformity of the pattern. As we are using a flow of aerosols from a nebuliser source we mimic how atmospheric aerosols might behave, and for real sampling strategies we must make robust devices that can cope with a range of particle sizes and velocities. To some extent this is what we have demonstrated. For other applications, such as airborne digital microfluidics, where we wish to have less random choices about which particles we trap and carry out droplet reactions with we must come up with better loading strategies for the traps. We envisage using single droplet maker devices (such as ink jet printer heads) to position the drops ready for accurate loading. Combining this sort of technique with the random process would also allow better implementation of digital microfluidics for real time sampling.
We envisage many applications for airborne optical manipulation and the initial type of work in this area is outlined in  and  with regard to atmospheric chemistry. Other application areas include the sampling and characterisation of aerosols [27,28] such as bioaerosol sampling for anti-bioterrorism related work  and combustion studies . There also exist many opportunities in more traditional optical manipulation studies, for example studying particle dynamics in the absence of a damping medium, optical rotation studies and even esoteric suggestions such as building neutrino detectors .
We would like to thank both Jonathan Reid and Laura Mitchem from Bristol University for helpful discussions and Kishan Dholakia (St. Andrews) for the use of the SLM. DRB would like to thank EPSRC for support, while DM is a Royal Society University Research Fellow. The work was partly funded by NERC and EPSRC.
References and Links
1. A. Ashkin, “Acceleration and trapping of particles by radiation pressure,” Phys. Rev. Lett. 24, 156 (1970). [CrossRef]
4. S. Dumont, W. Cheng, V. Serebrov, R. K. Beran, I. Tinoco Jr., A. M. Pyle, and C. Bustamante “RNA translocation and unwinding mechanism of HCV NS3 helicase and its coordination by ATP,” Nature 439, 105–108 (2006). [CrossRef] [PubMed]
5. P. T. Korda, G. C. Spalding, and D. G. Grier, “Evolution of a colloidal critical state in an optical pinning potential landscape,” Physical Review B 66, 024504 (2002). [CrossRef]
6. P. M. Hansen, J. K. Dreyer, J. Ferkinghoff-Borg, and L. Oddershede “Novel optical and statistical methods reveal colloid-wall interactions inconsistent with DLVO and Lifshitz theories,”, Journal of Colloid and Interface Science, 287561–571 (2005). [CrossRef]
7. V. Garcés-Chávez, D. McGloin, M. J. Padgett, W. Dultz, H. Schmitzer, and K. Dholakia, “Observation of the transfer of the local angular momentum density of a multi-ringed light beam to an optically trapped particle,” Phys. Rev. Lett. 91, 093602 (2003). [CrossRef] [PubMed]
8. N. B. Simpson, K. Dholakia, L. Allen, and M. J. Padgett, “The mechanical equivalence of the spin and orbital angular momentum of light: an optical spanner,” Opt. Lett. 22, 52–54 (1997). [CrossRef] [PubMed]
10. A. Biswas, H. Lati, R. L. Armstrong, and R.G. Pinnick, “Double-resonance stimulated Raman scattering from optically levitated glycerol droplets,” Phys. Rev. A 40, 7413–7416 (1989). [CrossRef] [PubMed]
11. R. Thurn and W. Kiefer, “Raman microsampling technique applying optical levitation by radiation pressure,” Appl. Spectrosc. 38, 78–83 (1984). [CrossRef]
13. N. Magome, M.I. Kohira, E. Hayata, S. Mukai, and K. Yoshikawa, “Optical Trapping of a Growing Water Droplet in Air” J. Phys. Chem. B 107, 3988–3990 (2003). [CrossRef]
14. M.D. King, K.C. Thompson, and A.D. Ward, “Laser Tweezers Raman Study of Optically Trapped Aerosol Droplets of Seawater and Oleic Acid Reacting with Ozone: Implications for Cloud-Droplet Properties,” J. Am. Chem. Soc. 126, 16710–16711 (2004). [CrossRef] [PubMed]
15. 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, 4924–4927 (2004). [CrossRef]
16. M. Reicherter, T. Haist, E. U. Wagemann, and H. J. Tiziani “Optical particle trapping with computer generated holograms written on a liquid crystal display,” Opt. Lett. 24, 608–610 (1999). [CrossRef]
17. J. A. Curtis, B. A. Koss, and D. G. Grier, “Dynamic Holographic Optical Trapping,” Opt. Commun. 207, 169–175 (2002). [CrossRef]
19. K. Ramser, J. Enger, M. Goksör, D. Hanstorp, K. Logg, and M. Käll, “A microfluidic system enabling Raman measurements of the oxygenation cycle in single optically trapped red blood cells,” Lab on a Chip 5, 431–436 (2005). [CrossRef] [PubMed]
22. V. Soifer, V. Kotlyar, and L. Doskolovich, Iterative Methods for Diffractive Optical Elements Computation (Taylor and Francis, London, 1997).
23. J. H. Dennis, C. A. Pieron, and K. Asai, “Aerosol Output and Size from Omron NE-U22 nebulizer,” in Proceedings of the 14th International Congress International Society for Aerosols in Medicines, Baltimore June 14-18 2003. Journal of Aerosol Medicine 16:2 213, 2003.
24. J. H. Seinfeld and S. N. Pandis, Atmospheric Chemistry and Physics: Air Pollution to Climate Change (John Wiley and Sons Inc., 1997).
25. Laura Mitchem, Particle Dynamics Group, School of Chemistry, University of Bristol, Bristol, BS8 1TS (personal communication, 2006).
27. P. Kaye, W. R. Stanley, E. Hirst, E. V. Foot, K. L. Baxter, and S. J. Barrington, “Single particle multichannel bio-aerosol fluorescence sensor,” Opt. Express 13, 3583–3593 (2005). [CrossRef] [PubMed]
28. K. Davitt, Y. -K. Song, W. Patterson III, A. Nurmikko, M. Gherasimova, J. Han, Y. -L. Pan, and R. Chang, “290 and 340 nm UV LED arrays for fluorescence detection from single airborne particles,” Opt. Express 13, 9548–9555 (2005). [CrossRef] [PubMed]
30. E. F. Mikhailov, S. S. Vlasenko, Lutz Krämer, and Reinhard Niessner, “Interaction of soot aerosol particles with water droplets: influence of surface hydrophilicity,” J. Aero. Sci. 32, 697–711 (2001). [CrossRef]