Anal. Chem. 2005, 77, 1253-1260
In Situ Visualization and Characterization of Aerosol Droplets in an Inductively Coupled Plasma Kaveh Jorabchi, Kaveh Kahen, Callum Gray,† and Akbar Montaser*
Department of Chemistry, George Washington University, Washington, D.C. 20052
Laser-scattering techniques are utilized for the first time to visualize the aerosol droplets in an inductively coupled plasma (ICP) torch from the nebulizer tip to the site of analytical measurements. The resulting images provide key information about the spatial distribution of the aerosol introduced by direct injection and conventional sample introduction devices: (1) a direct injection highefficiency nebulizer (DIHEN); (2) a large-bore DIHEN; and (3) a MicroFlow PFA nebulizer with a PFA Scott-type spray chamber. Moreover, particle image velocimetry is used to study the in situ behavior of the aerosol before interaction with the plasma, while the individual surviving droplets are explored by particle tracking velocimetry. Directly introduced aerosols are highly scattered across the plasma torch as a result of their radial motion, indicating less than optimum sample consumption efficiency for the current direct injection devices. Further, the velocity distribution of the surviving droplets demonstrates the importance of the initial droplet velocities in complete desolvation of the aerosol for optimum analytical performance in ICP spectrometries. These new observations are critical in the design of the next-generation direct injection devices for lower sample consumption, higher sensitivity, lower noise levels, suppressed matrix effects, and developing smart spectrometers. Inductively coupled plasma (ICP) spectrometry is the current method of choice for elemental and isotopic analysis.1-3 Despite years of research and considerable improvements in instrumentation, sample introduction is still the main bottleneck of this powerful analytical tool. In pneumatic nebulization, by far the most common method of solution introduction in ICP spectrometry, an aerosol is produced by interactions between the liquid sample and a gas flow at the nebulizer nozzle.4-8 The nebulizer is typically coupled to a spray chamber to remove the coarse droplets prior * To whom correspondence should be addressed. Telephone: 202-994-6480. Fax: 202-994-2298. E-mail:
[email protected]. † Current address: LaVision Inc., Ypsilanti, MI 48197. (1) Montaser, A., Ed. Inductively Coupled Plasma Mass Spectrometry; VCHWiley: New York, 1998. (2) Montaser, A., Golightly, D. W., Eds. Inductively Coupled Plasmas in Analytical Atomic Spectrometry, 2nd ed.; Wiley-VCH: New York, 1992. (3) Montaser, A.; McLean, J. A.; Liu, H.; Mermet, J.-M. An introduction to ICP spectrometries for elemental analysis. In Inductively Coupled Plasma Mass Spectrometry; Montaser, A., Ed.; Wiley-VCH: New York, 1998: pp 1-31. (4) Montaser, A.; Minnich, M. G.; McLean, J. A.; Liu, H.; Caruso, J. A.; McLeod, C. W. Sample introduction in ICPMS. In Inductively Coupled Plasma Mass Spectrometry; Montaser, A., Ed.; Wiley-VCH: New York, 1998; pp 83-264. 10.1021/ac048576k CCC: $30.25 Published on Web 01/22/2005
© 2005 American Chemical Society
to introduction of aerosol into the plasma. However, spray chambers suffer from low transport efficiencies (typically 10-30% at 0.1 mL/min; 1-3% at 1 mL/min), loss of volatile analyte, and memory and transient acid effects.4,8-10 To alleviate these drawbacks, test solution is injected directly into the ICP through devices such as the direct injection nebulizer and the direct injection high-efficiency nebulizer (DIHEN).11-13 Furthermore, direct injection devices offer reduced dead volume for chromatography and capillary electrophoresis, minimizing postcolumn broadening and improving the separation efficiency.14-17 In both direct injection and conventional sample introduction methods, the quality of the aerosol determines its fate inside the plasma, profoundly affecting the analytical performance. Ideally, the droplets must be small and slow, uniform in size and velocity (unless they are so small that they evaporate quickly), and confined to the central channel of the ICP. These properties lead to optimum conditions for efficient desolvation, vaporization, atomization, excitation, and ionization of the analyte in the plasma, resulting in continuous atomic/ionic clouds and eventually a steady signal for the best analytical performance. Common pneumatic nebulizers, however, generate polydisperse aerosol.18-25 This spreads the desolvation point of the droplets in the plasma (5) Montaser, A.; Minnich, M. G.; Liu, H.; Gustavson, A. G. T.; Browner, R. F. Fundamental aspects of sample introduction in ICP spectrometry. In Inductively Coupled Plasma Mass Spectrometry; Montaser, A., Ed.; WileyVCH: New York, 1998; pp 335-420. (6) McLean, J. A.; Minnich, M. G.; Iacone, L. A.; Liu, H.; Montaser, A. J. Anal. At. Spectrom. 1998, 13, 829-842. (7) Gustavson, A. G. T. Liquid sample introduction into plasmas. In Inductively Coupled Plasma in Analytical Atomic Spectrometry, 2nd ed.; Montaser, A., Golightly, D. W., Eds.; Wiley-VCH: New York, 1992; pp 690-720. (8) Browner, R. F. Fundamental aspects of aerosol generation and transport. In Inductively Coupled Plasma Emission Spectroscopy’ Boumans, P. W. J. M., Ed.; Wiley-VCH: New York, 1987: Part II, pp 244-288. (9) Todoli, J.-L.; Mermet, J.-M. Spectrochim. Acta 1999, 54B, 895-929. (10) Bjorn, E.; Frech, W. J. Anal. At. Spectrom. 2001, 16, 4-11. (11) Lawrence, K. E.; Rice, G. W.; Fassel, V. Anal. Chem. 1984, 56, 289-292. (12) Wiederin, D. R.; Smith, F. G.; Houk, R. S. Anal. Chem. 1991, 63, 219-225. (13) McLean, J. A.; Zhang, H.; Montaser, A. Anal. Chem. 1998, 70, 1012-1020. (14) Majidi, V.; Quarnstrom, J.; Tu, Q.; Frech, W.; Thomassen, Y. J. Anal. At. Spectrom. 1999, 14, 1933-1935. (15) Acon, B. W.; McLean, J. A.; Montaser, A. J. Anal. At. Spectrom. 2001, 16, 852-857. (16) Wind, M.; Eisenmenger, A.; Lehmann, W. D. J. Anal. At. Spectrom. 2002, 17, 21-26. (17) Todoli, J.-L.; Mermet, J.-M. J. Anal. At. Spectrom. 2001, 16, 514-520. (18) Clifford, R. H.; Ishii, I.; Montaser, A.; Meyer, G. A. Anal. Chem. 1990, 62, 390-394. (19) Shum, S. C. K.; Johnson, S. K.; Pang, H.-M.; Houk, R. S. Appl. Spectrosc. 1993, 47, 575-583. (20) Liu, H.; Montaser, A. Anal. Chem. 1994, 66, 3233-3242. (21) Olesik, J. W.; Kinzer, J. A.; McGowan, G. J. Appl. Spectrosc. 1997, 51, 607616.
Analytical Chemistry, Vol. 77, No. 5, March 1, 2005 1253
over a large area, and in fact, some droplets never completely desolvate. The incompletely desolvated droplets in the analytical zone of the plasma (10-15-mm downstream the load coil where atomic ions and neutrals are abundant) act as local heat sinks, introducing fluctuations in plasma properties (i.e., excitation and ionization temperatures, atom and ion number densities). As a result, the noise level and matrix effects are elevated in addition to the loss of analyte by the incompletely desolvated droplets for both ICP atomic emission spectroscopy and ICP mass spectrometry (ICPMS).26-31 The fate of the droplets in the plasma may be studied by highspeed photography or time-resolved spectroscopic techniques.26-36 The latter can provide the axial velocity of the atomic or ionic clouds around the droplets and analyte particles in the ICP. However, the cited clouds are typically much larger (a few millimeters in diameter) than the droplets (a few micrometers in diameter), are entrained by the gas flow and travel at the local gas velocity (20-25 m/s), and thus cannot offer direct information on the size and velocity of the droplets inside the plasma. Recently, planar drop sizing and particle image velocimetry (PIV) as well as interferometric imaging and particle tracking velocimetry (PTV) have been utilized to characterize the size and velocity of the droplets and particles in glow discharge plasmas and flames.37-39 Also, a laser-based imaging method has been developed to measure the size of the droplets in a thermal reactor.40 However, the source gas temperature in these studies was much lower than that of typical argon ICP, and the droplets were at least 1 order of magnitude larger than those encountered in ICP spectrometries. Small droplets (25 m/s) for direct injection devices, which survive the high temperature plasma because of their short residence time. Note that these droplets travel at considerably higher velocities than the local gas velocity in the axial channel of ICP; therefore, this effect cannot be attributed to the acceleration of the aerosol in the plasma. In contrast to the PFA-spray chamber arrangement, the high-velocity droplets are observed in the velocity distributions produced by PDPA for both the DIHEN and LB-DIHEN in the absence of the plasma,46 clearly demonstrating the importance of low initial droplet velocity for efficient desolvation in the plasma. Note, however, that the high-velocity droplets are not captured in PIV images25 due to ensemble averaging. CONCLUSIONS AND FUTURE DIRECTIONS A laser-based imaging technique is introduced to visualize and contrast the properties of droplets in a high-temperature ICP using direct injection and conventional pneumatic sample introduction
techniques. The technique is novel because it provides simultaneous information on the in situ location, velocity distribution, and number of the droplets in the high-temperature plasma, revealing their eventual contribution to analytical signal or noise in ICP spectrometries. Despite 100% transport efficiency, the current direct injection devices offer less than optimum spatial focusing of the aerosol into the central channel of the ICP, resulting in signal loss, elevated noise levels, and matrix effects. Compared to this, conventional sample introduction methods suffer from lower transport efficiencies, large dead volume, and spray chamberinduced matrix and memory effects.4,8-10 However, for conventional sample introduction, the droplets strictly travel in the axial channel thanks to the injector tube of the ICP torch. In both cases, the droplet velocities do not match the gas velocity in the axial channel. Conventionally introduced droplets lag behind the gas flow. Direct injection results in some droplets exceeding the local gas velocity in the analytical zone of the plasma. In pneumatic nebulization, higher gas flow rates are required to produce fine droplets; however, such high flows yield fast droplets that may survive the plasma. This tradeoff is particularly important for direct injection devices where no droplet size filtering is performed on the aerosol before their introduction into the ICP. For the best analytical performance, small droplets must be focused into the axial channel of the ICP at low velocities (