Anal. Chem. 2007, 79, 6027-6030
Controlling Charge on Levitating Drops Ryan T. Hilger, Michael S. Westphall, and Lloyd M. Smith*
Department of Chemistry, University of WisconsinsMadison, 1101 University Avenue, Madison, Wisconsin 53706-1396
Levitation technologies are used in containerless processing of materials, as microscale manipulators and reactors, and in the study of single drops and particles. Presented here is a method for controlling the amount and polarity of charge on a levitating drop. The method uses singleaxis acoustic levitation to trap and levitate a single, initially neutral drop with a diameter between 400 µm and 2 mm. This drop is then charged in a controllable manner using discrete packets of charge in the form of charged drops produced by a piezoelectric drop-on-demand dispenser equipped with a charging electrode. The magnitude of the charge on the dispensed drops can be adjusted by varying the voltage applied to the charging electrode. The polarity of the charge on the added drops can be changed allowing removal of charge from the trapped drop (by neutralization) and polarity reversal. The maximum amount of added charge is limited by repulsion of like charges between the drops in the trap. This charging scheme can aid in micromanipulation and the study of charged drops and particles using levitation. Containerless processing of drops using levitation is utilized in order to avoid the negative effects of container walls including adsorption and nucleation. Container walls can interfere with certain chemical and physical processes and, in some cases, may also hinder one’s ability to observe or manipulate the system. Increasing emphasis on miniaturization has resulted in numerous containerless trapping techniques becoming recognized for their ability to handle small volumes to serve as micromanipulators and microreactors.1-3 Among these trapping techniques, single-axis acoustic levitation has gained wide acceptance because there are relatively few restrictions on the properties of the sample.4 An overview of the principles underlying single-axis acoustic levitation of drops has been presented by Yarin et al.5 Charging of drops allows for the use of electric fields as a convenient means of manipulation. Electrostatic levitation of charged drops has been described,6 and hybrid acoustic/ electrostatic levitation has been developed for use in the study of * To whom correspondence should be addressed. Phone: 608-262-9207. Fax: 608-265-6780. E-mail:
[email protected]. (1) Vandaele, V.; Lambert, P.; Delchambre, A. Precis. Eng. 2005, 29, 491505. (2) Santesson, S.; Andersson, M.; Degerman, E.; Johansson, T.; Nilsson, J.; Nilsson, S. Anal. Chem. 2000, 72, 3412-3418. (3) Kavouras, A.; Krammer, G. Rev. Sci. Instrum. 2003, 74, 4468-4473. (4) Welter, E.; Neidhart, B. Fresenius’ J. Anal. Chem. 1997, 357, 345-350. (5) Yarin, A. L.; Pfaffenlehner, M.; Tropea, C. J. Fluid Mech. 1998, 356, 6591. (6) Rhim, W.; Chung, S. K. Methods 1990, 1, 118-127. 10.1021/ac070413j CCC: $37.00 Published on Web 06/20/2007
© 2007 American Chemical Society
drop arrays7,8 and crystallization.9 Acoustic levitation of charged drops has been used to form clusters of levitated drops10 as well as in the study of drop shape oscillation frequencies in superimposed acoustic and electric fields.11 Electrodynamic levitation has been used to study fission of desolvating, charged drops12-15 in order to better understand progeny drop and ion production from charged drops.16-18 Prior work utilizing levitation of charged drops has typically separated drop charging from drop levitation. Usually a method is employed to create or release a charged drop which is subsequently trapped and levitated. Manipulation of the amount of charge on a trapped drop is not possible using most such techniques. Apfel and co-workers demonstrated fusion of clusters of charged drops within an acoustic/electrostatic levitator.7,8 Davis and co-workers were able to reverse the polarity of a charged particle trapped in an electrodynamic levitator by combining it with a drop19 or particle20 possessing a greater but opposite charge. We are unaware of any work demonstrating charging of an initially neutral, levitating drop. The present work demonstrates charging of a neutral, levitating drop by addition of charged drops. The number of drops added can be controlled as can their charge. This method allows for precise control over the amount of added charge. Removal of charge (by neutralization) and polarity reversal are also demonstrated. EXPERIMENTAL SECTION Our single-axis acoustic levitator is similar to that used by Xie et al.21,22 It uses a CL4 ultrasonic converter (Misonix, Farmingdale, (7) (8) (9) (10)
(11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22)
Tian, Y.; Apfel, R. E. J. Aerosol Sci. 1996, 27, 721-737. Apfel, R. E.; Zheng, Y.; Tian, Y. J. Acoust. Soc. Am. 1999, 105, L1-L6. Chung, S. K.; Trinh, E. H. J. Cryst. Growth 1998, 194, 384-397. Liu, S.; Ruff, G. A. In Proceedings of the Fifth International Microgravity Combustion Workshop; National Aeronautics and Space Administration: Hanover, MD, 1999; pp 285-288. Trinh, E. H.; Holt, R. G.; Thiessen, D. B. Phys. Fluids 1996, 8, 43-61. Duft, D.; Achtzehn, T.; Mu ¨ ller, R.; Huber, B. A.; Leisner, T. Nature 2003, 421, 128. Taflin, D. C.; Ward, T. L.; Davis, E. J. Langmuir 1989, 5, 376-384. Richardson, C. B.; Pigg, A. L.; Hightower, R. L. Proc. R. Soc. London, Ser. A 1989, 422, 319-328. Achtzehn, T.; Mu ¨ ller, R.; Duft, D.; Leisner, T. Eur. Phys. J. D 2005, 34, 311-313. Feng, X.; Bogan, M. J.; Agnes, G. R. Anal. Chem. 2001, 73, 4499-4507. Bakhoum, S. F. W.; Agnes, G. R. Anal. Chem. 2005, 77, 3189-3197. Bakhoum, S. F. W.; Bogan, M. J.; Agnes, G. R. Anal. Chem. 2005, 77, 3461-3465. Widmann, J. F.; Aardahl, C. L.; Johnson, T. J.; Davis, E. J. J. Colloid Interface Sci. 1998, 199, 197-205. Vehring, R.; Aardahl, C. L.; Davis, E. J.; Schweiger, G.; Covert, D. S. Rev. Sci. Instrum. 1997, 68, 70-78. Xie, W. J.; Cao, C. D.; Lu ¨ , Y. J.; Wei, B. Phys. Rev. Lett. 2002, 89, 104304. Xie, W. J.; Wei, B. Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys. 2002, 66, 026605.
Analytical Chemistry, Vol. 79, No. 15, August 1, 2007 6027
Figure 2. Top view of the apparatus showing the image acquisition components and the current probe. Figure 1. Schematic showing a side view of the levitation apparatus.
NY) driven by an ultrasonic processor XL model XL2015 also from Misonix. The XL2015 outputs a 20 kHz sine wave with adjustable amplitude. The maximum power output of the XL2015 is 475 W. The output is fed into the ultrasonic converter, the tip of which has been modified with a flat disk of radius 15.9 mm (0.935λ where λ is the acoustic wavelength) constructed of PEEK polymer. This surface serves as the acoustic wave emitter. The aluminum reflector is spherically curved with a section radius of 21.6 mm (1.27λ) and a curvature radius of 33.5 mm (1.97λ). The vertical position of the reflector can be adjusted using a micrometer screw. Figure 1 shows a side view of the levitation region. The transducer is mounted using rubber gaskets and washers to minimize stray vibrations in the levitation region. The levitation region is surrounded by a plexiglass tube (50.6 mm o.d., 44.4 mm i.d.) which shields it from wind currents and provides a mounting surface for electrodes and drop dispensers. The levitation force is tuned by adjusting the sound pressure level by means of the transducer input voltage and by adjusting the emitter-reflector spacing. Power used for drop levitation is maintained near 24 W corresponding to an input voltage amplitude of 245 Vp-p. Emitterreflector spacing is maintained near λ (17 mm) so that two pressure nodes exist between the emitter and the reflector with the drop being levitated in the node nearest the emitter. The levitated drop is visualized using a TM-200 CCD camera (JAI Pulnix, Sunnyvale, CA) equipped with a Precise Eye 2.4× long working distance lens (Navitar, Rochester, NY). Lighting is provided by an LED powered by a 0-20 V, 500 mA power supply. Figure 2 shows a top view of the apparatus which includes the image acquisition components. The camera is interfaced to a Windows XP personal computer using an Avid Liquid Pro breakout box (Avid Technology, Tewksbury, MA). The images are viewed and processed using Avid Liquid version 7 software. Diameters of drops are measured by calibrating the images using images of lines with known spacing. The piezoelectric drop dispenser used in this work is similar to that used by Berggren et al.23 It is constructed by securing a pulled glass capillary (o.d. ) 1 mm, length ) 53 mm, orifice diameter ) 30 µm; World Precision Instruments, Sarasota, FL) (23) Berggren, W. T.; Westphall, M. S.; Smith, L. M. Anal. Chem. 2002, 74, 3443-3448.
6028 Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
Figure 3. Characteristics of the trapezoid wave used to drive the piezoelectric drop dispensers. Amplitude ranges from 0 to 400 V; rise and fall times range from 0 to 100 µs; pulse width ranges from 0 to 200 µs.
inside a piezoelectric tube (length ) 20 mm, o.d. ) 2.2 mm, i.d. ) 1 mm; PI Ceramic, Lederhose, Germany) using cyanoacrylate glue. This dispenser element is inserted into a plexiglass holder which is glued to the side of the plexiglass tube surrounding the levitation region; a small hole allows the dispenser access to the levitation region. The holder includes two small brass screws which are used to both secure the dispenser in place and make electrical connections to its electrodes. The voltage pulse used to drive the piezoelectric element is created using a program written in LabView 7 Express version 7.0 (National Instruments, Austin, TX) on a Windows XP PC. The computer employs a National Instruments PCI-6713 DAQ card to output a low-voltage pulse to a National Instruments SCB-68 I/O module. From there the pulse is amplified using a PZD 350 dual channel amplifier (Trek, Medina, NY) and is sent to the piezoelectric element. The system outputs a trapezoid wave with characteristics depicted in Figure 3. The PC also outputs a square pulse which is delayed by 0-1000 µs from the dispenser drive pulse. This pulse is used to trigger a TGP110 pulse generator (Tenma, Springboro, OH) which outputs a pulse that is used to drive an LED for stroboscopic illumination of the dispenser tip. Varying the LED delay using the software enables visualization of drop production at the tip at various time points after the pulse is applied to the piezoelectric element using a CCD camera system similar to that described above. The diameters of the dispensed drops are estimated to be around 50
µm based on the orifice diameter. Solution is gravity fed from a reservoir to the dispenser through a length of 0.64 mm i.d. silicone tubing. A 35 mm length of platinum wire (d ) 368 µm) is inserted into the glass capillary through the silicone tubing. The voltage on this wire, supplied by a model PS350 power supply (Stanford Research Systems, Sunnyvale, CA), controls the magnitude and polarity of the charge on the dispensed drops. In order to demonstrate our control over the charge on the dispensed drops, current measurements were made on the drops emitted from a piezoelectric dispenser operated at 120 Hz. The tip of the dispenser was positioned 2 mm in front of a stainless steel disk connected to ground through a model 6485 picoammeter (Keithley, Cleveland, OH) operated in zero-corrected mode with a 100 ms integration time. The voltage on the dispenser electrode was varied, and the measured current at each voltage was taken as the average of 200 readings. The noise was similarly measured at each dispenser electrode voltage with the dispenser off and subtracted from the drop current. The current reading was allowed to stabilize prior to acquisition of data in order to avoid artifacts resulting from activating either the dispenser or the picoammeter. The measurements were made over as short of a period of time as possible in order to minimize error due to variation in the drop dispenser. Deionized water was the liquid used in all experiments. Two piezoelectric dispensers operating at a drive frequency of 30 Hz were used to introduce drops into the levitation region. Each dispenser was positioned orthogonal to the levitator axis and on opposite sides of the levitation region (see Figures 1 and 2). One dispenser was used to produce neutral drops, and the other was used to produce charged drops. The magnitude and polarity of the charge on the dispensed drops was examined by collecting them on a probe connected to the picoammeter. The probe consisted of a length of RG-174 coaxial cable with the last half inch stripped to expose the center conductor. The probe was inserted into the levitation region in order to collect the dispensed drops (see Figure 2). Voltages of 30-100 V applied to the dispenser electrode resulted in drop currents in the range of 0-10 pA. The image acquisition system was used to record videos of the levitation region. In this manner, the interactions of charged and neutral drops as well as those of like charged and oppositely charged drops could be recorded. The picoammeter was also employed to measure the charge on a levitating drop by inserting the current probe into the levitation region and touching the drop. The current trace was recorded by digitizing the analog output from the picoammeter using a DI-718B data logger (DATAQ Instruments, Akron, OH) equipped with a DI-8B51-02 amplifier and interfaced to a Windows XP PC running WinDaq/HS version 3.17 (DATAQ Instruments). The sampling rate of the data logger was 4.8 kHz. The current traces were integrated using the composite Simpson’s rule24 to determine the charges and their error bounds. RESULTS AND DISCUSSION Charging and polarity reversal of a levitating drop were observed visually using the video acquisition system. The experiment began by using the neutral drop dispenser to create a large (d ) 550 µm) drop in the acoustic trap by fusion of many (24) Atkinson, K. E. An Introduction to Numerical Analysis, 2nd ed.; Wiley: New York, 1989.
Figure 4. Column A displays images of the initial drops before exposure to charged drops. Column B displays the same drops after exposure. The drop in row i was exposed to drops whose polarity was changed as a function of time. The drop in row ii was exposed to drops of a fixed polarity for the same period of time. Drop A(i) coalesced with drops each time the polarity of these drops was changed and thus increased in size as seen in panel B(i). Drop A(ii) did not increase in size as shown in panel B(ii) because repulsion of like charges kept the incoming drops away.
dispensed drops. The neutral drop dispenser was then turned off, and the charged drop dispenser was used to produce either positively or negatively charged drops using a voltage of (30 V applied to the dispenser electrode. The incoming, charged drops initially fused with the large drop, but after about 5 s they ceased fusing, and instead, small numbers of them fused with each other forming positively charged satellite drops (d ∼50-100 µm) which would remain trapped with the large drop for a second or two before falling out of the trap. The polarity of the charge on the incoming drops was then reversed, and they once again added to the large drop for about 5 s. After this they once again stopped adding and began forming the satellite drops. This cycle was repeated by switching the polarity of the charge on the incoming drops three more times. A control experiment was performed in which positively charged drops were continuously sprayed at a levitating drop. This was done for a time period similar to that required for the previously described experiment (about 1 min). A comparison of the two experiments provided additional evidence for the charging and polarity reversal. During the first experiment the large drop grew in diameter from 550 to 780 µm as the polarity of the incoming drops is repeatedly changed causing them to fuse with the large drop. By contrast, during the control experiment the diameter of the large drop actually shrank slightly (480-440 µm) due to evaporation as the incoming drops refused to add after the initial charging (see Figure 4). Charging and polarity reversal of levitating drops were also demonstrated by measuring the charges on levitated drops using the picoammeter. Figure 5 shows current traces obtained from three different levitated drops. Figure 5A depicts a trace of a “neutral” drop which was found to have minimal charge (1.61 ( 0.14 pC). Figure 5B shows a trace from an initially neutral drop which was charged to 21.8 ( 2.6 pC by exposure to positively charged drops. Lastly, Figure 5C depicts a trace from an initially neutral drop which was exposed to positively charged drops and then exposed to negatively charged drops and charged to -71.7 ( 2.5 pC. These results, together with the visual evidence Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
6029
Figure 5. Current traces obtained from three different levitated drops. (A) A neutral drop. (B) An initially neutral drop that was charged positively to 21.8 ( 2.6 pC. (C) An initially neutral drop that was charged positively then charged negatively to a final value of -71.7 ( 2.5 pC.
Figure 6. Drop current vs voltage applied to the charging electrode. The dispenser drive frequency was 120 Hz.
described above, demonstrate charging and polarity reversal of levitating drops. Varying the charge on the dispensed drops provides a way of controlling the sizes of the “packets” of charge added to the levitating drop. This is accomplished by varying the voltage applied to the dispenser’s charging electrode. Measurements of the drop current emitted from a dispenser as a function of the voltage applied to the charging electrode are depicted in Figure 6. In this experiment the dispenser drive frequency was 120 Hz. With the use of this data, the charge on a single dispensed drop was calculated to be 29.3 ( 6.5 fC when 30 V was applied to the charging electrode by dividing the measured drop current by the dispenser drive frequency. This amount of charge approximates the size of a “packet” of charge that may be added to the levitating drop. The results described above demonstrate controlled charging and polarity reversal of acoustically levitating drops by fusion of charged drops. Initially, incoming positively charged drops fuse with the large neutral drop and charge it positively. Eventually, the surface charge density of the large drop builds up to the point (25) Adamiak, K. IEEE Trans. Ind. Appl. 2002, 38, 1001-1007. (26) Gomez, A.; Tang, K. Phys. Fluids 1994, 6, 404-414. (27) Smith, J. N.; Flagan, R. C.; Beauchamp, J. L. J. Phys. Chem. A 2002, 106, 9957-9967. (28) Switzer, G. L. Rev. Sci. Instrum. 1991, 62, 2765-2771. (29) Petersson, M.; Nilsson, J.; Wallman, L.; Laurell, T.; Johansson, J.; Nilsson, S. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 1998, 714, 39-46.
6030
Analytical Chemistry, Vol. 79, No. 15, August 1, 2007
at which repulsion of like charges prevents the incoming drops from fusing. Upon reversal of the polarity of the charge on the incoming drops they once again fuse with the large drop initially neutralizing it and then charging it negatively. Eventually repulsion of like charges again prevents the incoming drops from fusing. The maximum amount of charge that can be added to a levitating drop using this method is presumably limited by repulsion of like charges between the large drop and the incoming drops.7,8 More charge can be added to a larger drop since the increased surface area lowers the surface charge density. Other methods of increasing the maximum amount of charge could include polarization of the trapped drop by an electric field25 and increasing the depth of the acoustic potential well in order to provide additional acoustic force to counteract the repulsion of like charges. While the maximum amount of added charge may be low compared with other methods of producing charged drops such as induction6 or electrospray,26,27 the method presented here does possess some advantages. Unlike charging by induction, this method does not require the solution to possess significant ionic strength as the method does not depend upon polarization of the liquid. Modern drop-on-demand piezoelectric dispensers can dispense drops with highly uniform size and velocity distributions28 and could therefore be used to produce drops with highly uniform charge. This fact, coupled with the ability of acoustic levitators to be tuned to capture the incoming drops with near unity efficiency,29 enables extremely precise control over the amount of charge added to the levitating drop. The method described here thus gives one the ability to add and remove charge at will and in very precise increments as well as to reverse the polarity of the charge on a drop. CONCLUSIONS A method for controlling the charge on acoustically levitating drops has been presented. Charge is added to a levitating drop by injecting charged drops which are trapped by the acoustic field and fuse with the already trapped drop thereby charging it. The magnitude and polarity of the charge on the incoming drops may be controlled thereby permitting fine control over the amount of added charge. Charge can also be removed from the levitating drop through neutralization, and polarity reversal is possible. This new charging scheme extends the capabilities of levitation based methods for the study of single drop/particle systems. Controlled charging may also be useful for micromanipulation where attractive and repulsive electrostatic forces could be exploited. ACKNOWLEDGMENT We thank Dr. Brian Frey for helpful conversations and for his help preparing the manuscript. This work was supported by NIH/ NHLBI Contract No. N01-HV-28182 (Proteomics Initiative) and NIH/NIDDK Grant No. R33 DK070297. R.T.H. was supported by NIH/NHGRI predoctoral training Grant No. 5T32HG002760.
Received for review February 28, 2007. Accepted April 27, 2007. AC070413J
