We demonstrate the use of surface acoustic wave nebulization (SAWN) to load optical traps. We show that the droplets sizes produced can be tuned by altering the RF frequency applied to the devices, which leads to more control over the sizes of trapped particles. Typically the size distribution of the liquid aerosols delivered using SAWN is smaller than via a standard commercial nebulization device. The ability to trap a range of liquids or small solid particles, not readily accessible using other ultrasonic devices, is also demonstrated both in optical tweezers and dual beam fiber traps.
© 2013 Optical Society of America
The generation of aerosols is familiar to many through the use of household objects such as deodorant, air freshener and asthma inhalers. They also play an important role in atmospheric processes, such as cloud formation  and in the Earth’s radiative balance . Aerosol research has a wide scope, examining processes in drug delivery  combustion science  pollution  and weather  and is carried out over a wide range of size scales. To obtain a full understanding of basic aerosol properties it is essential to be able to dynamically interrogate single particles over long time scales. A number of technologies have been used to look at single or small groups of airborne particles. These include electrostatic traps (the electrodynamic balance)  and acoustic levitation . Perhaps the most versatile technique in terms of mechanical control of trapped particles is that of optical manipulation, and this has been receiving growing attention in recent years [9–17].
The ability to load such traps typically relies on a single droplet  or nebulizer source . A simple and cheap method is to use a medical nebulizer, which are primarily used to deliver drugs to the lungs. As most of these are designed for drugs suspended in an aqueous solution they are not compatible with some non-aqueous liquids, such as solvents. In addition such nebulizers are often bulky and do not have a facility to tune the size of particle produced. To date the majority of work making use of optical tweezers to trap airborne particles has been restricted to particle sizes larger than 1µm, which makes them less useful for trapping many atmospherically relevant particles. This is primarily due to the size distribution of the cheap medical nebulizers used as aerosol sources.
In this paper we introduce the use of SAW nebulizers [9–21, 23] (SAWNs) for the loading of particles in the optical trap devices. In a more general sense they should be suitable for loading single particle traps of any kind, including solid particles . SAWNs have the advantage of being very compact, amenable to a wider range of fluids and materials, and the size distribution of the droplets produced can be tuned by altering the applied RF frequency.
2. Materials and methods
Surface acoustic wave devices: In our experiments we make use of a surface acoustic wave nebulizer that has previously been described [25, 26]. Briefly, a surface acoustic wave can be set up in a lithium niobate substrate by the application of a radio frequency wave with (in our case) a frequency around 9.5MHz. The surface waves are able to excite capillary waves on the surface of the droplet. The wave amplitude is related to a number of parameters including the RF power supplied and will result in mechanical manipulation of a droplet placed on the substrate through to jetting and finally the breakup of the droplet into small droplets (nebulization). The SAW device used in our experiments is shown in Fig. 1(a). An example of nebulization is shown in Fig. 1(b). Mixing and nebulization are shown in Media 1 and Media 2.
The surface waves are generated by application of RF field (generated by an Agilent 33250A arbitrary waveform generator and a Mini Circuits ZHL-5W-1, 5-500MHz amplifier) to an inter-digitated transducer (IDT) with 20 pairs of 100µm wide gold electrodes. The device aperture is 1cm. These are deposited using UV photolithography, on a single crystal 127.68° Y-cut, X-propagation lithium niobate (LiNbO3) crystal. The wavelength of the SAW is determined by the width of the IDT fingers and also the gaps between them. In our case we set the dimensions to generate waves with a wavelength of 400µm, setting the theoretical resonant frequency at 10MHz. In practice, due to manufacturing tolerances, the presence of the liquid to be nebulized, self-heating, degradation of the IDT over time, the operating frequency will shift slightly from the design frequency. In the examples used for the trapped experiments presented here the resonance was found to be in the range 9.51 - 9.6 MHz, measured using a network analyzer. The SAWs were created by use of a continuous wave RF field: SAWs are generated in both directions away from the IDT, thus splitting the power in each wave. Typically we use input powers of 1-2W in each direction.
We operated the SAW in continuous wave (CW) operation, so that droplets are produced continuously. This can lead to adverse heating in the SAW, which can result in unwanted evaporation of the liquid after prolonged usage. The RF can also be applied in a pulsed mode, which can overcome this . For our experimental runs, small bursts of CW power were able to produce sufficient aerosol plumes to enable particles to be readily trapped.
Optical tweezers: Our optical tweezers system makes use of a 1070nm fibre laser (IPG YLM-5-1070-LP) and of a conventional inverted microscope setup, shown in Fig. 2. The major difference from normal optical tweezers is the use of a sample cell that can be used to collect the aerosols. The cell is placed over the end of the lithium niobate substrate to place the nebulization area close to the trapping site. We used a coverslip over the top of the microscope objective, treated  so as to allow a thin layer of water to wet the surface, which prevents optical aberrations appearing due to droplets of water forming on the coverslip. Humidity is kept sufficiently high to avoid droplet evaporation by placing a piece of wet tissue paper inside the chamber. We made use of a Nikon oil immersion Plan Apochromatic microscope (100X, numerical aperture of 1.25). Although not shown in the inset in Fig. 2, we could constantly pump liquid onto the SAW substrate with a syringe pump (Pico Plus, Harvard Apparatus) with the flow rate chosen so that solution formed a continuous layer on the SAWN. This enabled, if required, a continuous flow of aerosols into our devices allowing experiments over an extended period of time. This was necessary, for example, in the case of a liquid such as dodecane which nebulises very quickly.
The size of an inhaled aerosol plays a significant part in defining where in the respiratory system the aerosol will reach before being deposited . This is an important performance indicator for a medical nebuliser that will be designed to deliver a specific drug to a specific location e.g. either to the mouth, nose and throat or deep into the lungs. Generally, the diameter of an aerosol is in the same order of the instability capillary wavelength, λ where k is a coefficient that has been used to fit the formula (Eq. (1)) to experimental data sets. The instability capillary wavelength based on Kelvin’s Equation is given by Eq. (1) 10]. We examined nebulization of water, ethanol and dodecane.
The devices produce a range of droplet sizes in separate distribution regions. Some of these include very large droplets with sizes of up to 1000µm. These are not relevant for our studies or small droplet trapping and so the data in Fig. 3 considers only the smallest of the size distributions. The data are normalised against the commonest droplet size for each device in that distribution. Note that the percentage of droplets produced in this distribution for the medical nebulizer is an order of magnitude larger than that of the SAW. The medical nebulizer is designed specifically to produce droplets in this region, so this is perhaps unsurprising. These large volume mode distributions from the SAWN device appear due to other destabilization phenomena, namely jetting, and are different from the Rayleigh instability . The main finding of this paper in the context of optical manipulation is that our SAWs can readily make smaller droplets in the case of water (when compared with the devices used in our previous work), and the size distribution of the produced droplets can be tuned by altering applied RF frequency. The full width half maximum (FWHM) of the medical nebulizer produced distribution is 10.11µm compared to 1.08µm for the SAWN (note the x axis of Fig. 3 is a log scale). The production of these much smaller droplets is of particular interest for our trapping experiments, with a view to trapping a full range of atmospherically relevant particle sizes.
We also measured the nebulization rate for the three different liquids, (Fig. 4) finding that water is the most difficult liquid to nebulize, due to its higher surface tension and the better wetting of the substrate by the ethanol and dodecane. The nebulisation is initialised by the destabilisation of the liquid air interface through the creation of capillary waves which cause the aerosol droplets to start to be released from the surface of the liquid . As such the surface tension, along with the liquid's viscosity, plays an important role in determining the rate of nebulisation with higher surface tension resulting in lower rates of nebulisation. This is shown in Fig. 4 where water, which has a relatively high surface tension of 72.8mN/m, is nebulised at a lower rate to dodecane and ethanol that have lower surface tensions of 25.0mN/m and 22.4mN/m respectively.
We examined the size distribution as a function of applied frequency. Figure 5 shows the aerosol size of water generated by SAW devices with four different applied RF frequencies, 9.55, 12.23, 13.33 and 19.87 MHz, compared with the theoretical values expected from the Kelvin Equation. The aerosol size decreases as the applied RF frequency increases providing tuneable control over the droplet sizes. Using these mean aerosol diameters and the properties of the water (density, ρ = 1000kgm−3 and surface tension, g = 72.8mNm−1) give values of k between 0.82 to 1.26, which are in agreement with previous work [29, 30].
The inset in Fig. 5 shows the frequency response of the 1cm aperture IDT, determined through performing an S11 measurement with a network analyser (E5071C, Agilent Technologies). The reflected power drops at certain frequencies that correspond to resonances in the SAWN device primarily relating to the spacing of the transducers (notably at 9.6MHz). This response can be used to gauge the efficiency of the SAWN device at different frequencies with a lower valley corresponding to a frequency where the device will produce acoustic waves more efficiently and could hence nebulise liquid more effectively.
4. Optical manipulation of droplets produced using SAWN
By integrating the SAWN chip into the trapping chamber we are able to create a very compact system. We are able to trap a range of liquids (e.g. water, salt water, ethanol and dodecane), as well as solid particles (4.32µm silica beads, initially dispersed in ethanol) . This included the trapping of submicron water droplets (measured using video microscopy) around 0.98 ± 0.05µm. Such particles are smaller than the typical particles trapped using the medical nebulizers  and are an indication that we are producing these particles by SAWN. Finally we were also able to trap a collection of 100nm diameter gold nanoparticles dispersed in water. These could be trapped for several seconds, before, we assume, heating processes took over and the droplet became unstable. Data related to the trapping of salt water droplets is shown in Fig. 6. At higher RF frequency the size and the velocity of the nebulised droplets are also higher. This may be the reason that the different size of droplets of the same liquid nebulized at different frequencies are trapped at the same power.
We were also able to load droplets (salt water and dodecane droplets) into a dual beam fibre trap  by placing the SAWN below the trapping region (SAWN-fibre distance was 38mm), allowing a much simpler and more compact geometry than in previous experiments. In this experiment we used a 1064nm laser coupled into a single mode fibre (Thorlabs 1060XP) with a 6.2µm mean field diameter at 1060nm. The power used to trap was 4mW in each arm and the fibre separation distance was 160µm for water and 130µm for dodecane.
5. Discussion and conclusions
As the field of airborne optical tweezers matures, robust loading strategies will become more important. Here we present a versatile method for nebulizing a wide variety of liquids and particle types and introducing them into optical traps of different types. We have shown that tuning of the applied RF frequency gives rise to some size tunability of the droplets produced when nebulizing a liquid sample. The sizes of aerosols are consistent with Kelvin’s Equation. There exist other methods of loading particles, making use of ultrasonic transducers  which are useful for solid particles; electrostatic droplet generators, which are good for monodisperse and repeatable droplets of larger diameters  and of course medical nebulizers . Often these techniques are limited to single types of particle (liquid or solid) and have limitations in terms of size production, integration with the trap, the types of chemical that they can use, or the cost. Although the SAWN device is not yet a precision technique, importantly it offers size selection, is readily integrable with the trap and is very flexible over the type of chemicals that can be nebulized.
In addition to the flexibility of the nebulization technique, SAWN devices also offer the possibility of easy integration into integrated aerosol traps  to allow all on-chip delivery of a sample into an optical trap, which offers great opportunities for miniaturized sampling and analysis devices in the future. The throughput of the approach can be extended by using arrays of droplets, containing for example different substances to be nebulised that can be addressed using a recently developed phononic technique .
Our work on the loading of the dual beam fibre trap also opens up the possibility of using such geometry to try and improve the loading of mass spectrometers, making use of the SAWN in combination with optical guiding from a single beam towards the inlet of a spectrometer device. This could be a very valuable way of increasing the loading efficiency of valuable, small volume samples. Work is currently under way to examine this application.
One issue that we encountered was that it is not straightforward to swap between different chemical types to explore, for example, mixing and chemical reactions. One solution to this may be the integration of twin/quad SAWN devices into the trapping chamber to load from either end/side. More attractively, recent work using phononic lattices shows the potential to integrate different functions on a disposable device, that could provide a complete microfluidic system where reagents are mixed, moved and nebulised .
We have demonstrated the use of a surface acoustic wave nebulization for the loading of aerosol optical traps. The increasing flexibility of SAWN devices, with focusing effects using phononic crystals , for example, offer great potential for controlled loading of very small volumes of sample and also better integration of the devices with tweezers.
We thank the EPSRC Engineering Instrument Loan pool for the loan of the Malvern Spraytec aerosol sizer. SA thanks the Schlumberger Faculty for the Future program for support. SN thanks the support of the RAEng. DM thanks the Royal Society and EPSRC, grants EP/H004238/1 and EP/G007713/1, for support. We thank Rab Wilson for his advice and support in setting up the SAWNs in Dundee.
References and links
1. Z. Li, F. Niu, J. Fan, Y. Liu, D. Rosenfeld, and Y. Ding, “The long-term impacts of aerosols on the vertical development of clouds and precipitation,” Nat. Geosci. 4(12), 888–894 (2011). [CrossRef]
2. A. Metzger, B. Verheggen, J. Dommen, J. Duplissy, A. S. Prevot, E. Weingartner, I. Riipinen, M. Kulmala, D. V. Spracklen, K. S. Carslaw, and U. Baltensperger, “Evidence for the role of organics in aerosol particle formation under atmospheric conditions,” Proc. Natl. Acad. Sci. U.S.A. 107(15), 6646–6651 (2010). [CrossRef] [PubMed]
4. J. M. Williams, J. M. Jones, L. Ma, and M. Pourkashanian, “Pollutants from the combustion of solid biomass fuels,” Pror. Energy Combust. Sci. 38(2), 113–137 (2012). [CrossRef]
6. J. F. Kok, “A scaling theory for the size distribution of emitted dust aerosols suggests climate models underestimate the size of the global dust cycle,” Proc. Natl. Acad. Sci. U.S.A. 108(3), 1016–1021 (2011). [CrossRef] [PubMed]
7. L. Treuel, S. Pederzani, and R. Zellner, “Deliquescence behaviour and crystallisation of ternary ammonium sulfate/dicarboxylic acid/water aerosols,” Phys. Chem. Chem. Phys. 11(36), 7976–7984 (2009). [CrossRef] [PubMed]
8. G. Kaduchak, D. N. Sinha, and D. C. Lizon, “Novel cylindrical, air-coupled levitation/concentration device,” Rev. Sci. Instrum. 73(3), 1332–1336 (2002). [CrossRef]
9. 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]
11. M. D. Summers, D. R. Burnham, and D. McGloin, “Trapping solid aerosols with optical tweezers: A comparison between gas and liquid phase optical traps,” Opt. Express 16(11), 7739–7747 (2008). [CrossRef] [PubMed]
12. J. R. Butler, J. B. Wills, L. Mitchem, D. R. Burnham, D. McGloin, and J. P. Reid, “Spectroscopic characterisation and manipulation of arrays of sub-picolitre aerosol droplets,” Lab Chip 9(4), 521–528 (2009). [CrossRef] [PubMed]
13. D. R. Burnham and D. McGloin, “Radius measurements of optically trapped aerosols through Brownian motion,” New J. Phys. 11(6), 063022 (2009). [CrossRef]
15. D. R. Burnham and D. McGloin, “Modelling of optical traps for aerosols,” J. Opt. Soc. Am. B 28(12), 2856–2864 (2011). [CrossRef]
17. R. E. H. Miles, J. S. Walker, D. R. Burnham, and J. P. Reid, “Retrieval of the complex refractive index of aerosol droplets from optical tweezers measurements,” Phys. Chem. Chem. Phys. 14(9), 3037–3047 (2012). [CrossRef] [PubMed]
18. C. Esen, T. Weigel, V. Sprynchak, and G. Schweiger, “Raman spectroscopy on deformed droplets: theory and experiments,” J. Quant. Spectrosc. Radiat. Transf. 89(1-4), 79–85 (2004). [CrossRef]
19. M. Kurosawa, T. Watanabe, A. Futami, and T. Higuchi, “Surface acoustic wave atomizer,” Sens. Actuators A Phys. 50(1-2), 69–74 (1995). [CrossRef]
20. K. Chono, N. Shimizu, Y. Matsui, J. Kondoh, and S. Shiokawa, “Development of Novel Atomization System Based on SAW Streaming,” Jpn. J. Appl. Phys. 43(5B), 2987–2991 (2004). [CrossRef]
21. J. W. Kim, Y. Yamagata, M. Takasaki, B. H. Lee, H. Ohmori, and T. Higuchi, “A device for fabricating protein chips by using a surface acoustic wave atomizer and electrostatic deposition,” Sens. Actuators B Chem. 107(2), 535–545 (2005). [CrossRef]
23. L. Y. Qi, L. Y. Yeo, and J. R. Friend, “Interfacial destabilization and atomization driven by surface acoustic waves,” Phys. Fluids 20(7), 074103 (2008). [CrossRef]
24. J. Reboud, R. Wilson, Y. Zhang, M. H. Ismail, Y. Bourquin, and J. M. Cooper, “Nebulisation on a disposable array structured with phononic lattices,” Lab Chip 12(7), 1268–1273 (2012). [CrossRef] [PubMed]
25. S. R. Heron, R. Wilson, S. A. Shaffer, D. R. Goodlett, and J. M. Cooper, “Surface Acoustic Wave Nebulization of Peptides As a Microfluidic Interface for Mass Spectrometry,” Anal. Chem. 82(10), 3985–3989 (2010). [CrossRef] [PubMed]
27. J. Friend and L. Y. Yao, “Microscale acoustofluidics: Microfluidics driven via acoustics and ultrasonics,” Rev. Mod. Phys. 83(2), 647–704 (2011). [CrossRef]
28. A. Qi, J. R. Friend, L. Y. Yeo, D. A. V. Morton, M. P. McIntosh, and L. Spiccia, “Miniature inhalation therapy platform using surface acoustic wave microfluidic atomization,” Lab Chip 9(15), 2184–2193 (2009). [CrossRef] [PubMed]
29. R. J. Lang, “Ultrasonic Atomization of Liquids,” J. Acoust. Soc. Am. 34(1), 6–8 (1962). [CrossRef]
30. F. Barreras, H. Amaveda, and A. Lozano, “Transient High-Frequency Ultrasonic Water Atomization,” Exp. Fluids 33(3), 405–413 (2002). [CrossRef]
33. R. Wilson, J. Reboud, Y. Bourquin, S. L. Neale, Y. Zhang, and J. M. Cooper, “Phononic crystal structures for acoustically driven microfluidic manipulations,” Lab Chip 11(2), 323–328 (2011). [CrossRef] [PubMed]