1
Principles of Charged Aerosol Detection

Paul H. Gamache1 and Stanley L. Kaufman2

1 Thermo Fisher Scientific, Chelmsford, MA, USA

2 Retired from TSI Inc., Shoreview, MN, USA

CHAPTER MENU

1. 1.1 Summary
2. 1.2 History and Introduction to the Technology
3. 1.3 Charged Aerosol Detection Process
4. 1.4 CAD Response Model
5. 1.5 Performance Characteristics
6. References

1.1 Summary

This chapter provides a brief history and detailed overview of charged aerosol detection (CAD) and a semiempirical model describing its response and expected performance under various analytical conditions. CAD and other evaporative aerosol detectors involve the same successive steps of primary spray droplet formation from an eluent stream, conditioning by inertial impaction to remove droplets too large to evaporate during passage through the instrument, and evaporation of remaining droplets to form residue particles each comprised of nonvolatile background impurities and any nonvolatile analyte present. Detection of the residue particles produces the detector signal. In CAD, the aerosol is given a charge dependent on the particle size, and the total charge carried by the aerosol is measured as a current; in ELSD, the aerosol is detected by its light scattering properties. Both detection methods produce a response that is approximately mass-flow dependent. The analyte dry bulk density affects the residue particle size for a given eluting mass, which has a minor effect on the mass sensitivity of both detector types. Other analyte properties in particular optical properties (e.g., refractive index (RI)) for the ELSD likewise affect the sensitivity. Detection selectivity for evaporative aerosol detectors is based on differences in vaporization of components within an eluent. Accordingly, these techniques are expected to have very similar detection scope, eluent requirements, and solvent dependency of response. The unique characteristics of CAD are due to the aerosol measurement technique, which includes diffusion charging of residue particles and detection of the current due to deposition of particles with their charge in an aerosol-electrometer filter. Aerosol charging by diffusion mechanisms is well known to have only a minor dependence on particle material (i.e., analyte properties), which is the basis for uniform response capabilities of CAD. Like ELSD, CAD response (e.g., peak area vs. mass injected (minj)) can be described by a power law function with a variable exponent b. Linear response, never perfectly achieved by either methods, would correspond to b?=?1. For both techniques, the exponent b is at its maximum at the lowest minj and decreases with increasing minj. This is attributed to smaller residue particles that have a higher power law exponent (ß1) of response and are more prevalent with low minj and the low concentration that occurs near the edges of any peak. For ELSD this corresponds to Rayleigh light scattering for particle diameters (d) typically?<?50?nm where ß1?=?6, while for CAD corresponds to aerosol charging of d?<?~9?nm where ß1?~?2.25. For CAD, the lower d transition and smaller ß1 (closer to 1) underlies the widely observed lower detection limits, wider dynamic range, and less complex response curve than ELSD. Newer CAD designs produce an even smaller relative proportion of residue particles of d?<?~9?nm, thus further simplifying the response curve, enabling lower sensitivity limits and a wider quasi-linear response range.

1.2 History and Introduction to the Technology

The technique that is now most commonly called charged aerosol detection (CAD) was first described in 2001 by Kaufman at TSI Inc. in a provisional patent application that ultimately led to US patent 6,568,24 [1]. This device was termed an evaporative electrical detector (EED) and was based on coupling liquid chromatography (LC) and other separation techniques with TSI's well-established electrical aerosol measurement (EAM) technology [2]. Around the same time, Dixon and Peterson at California State University were pursuing a similar avenue of innovation with a laboratory-built device that coupled LC with an earlier generation of TSI's EAM instruments. Dixon and Peterson described their device, termed aerosol charge detector (ACD), in the Journal of Analytical Chemistry in 2002 [3]. In both instances, the primary objective was to exploit the advantages, well described in aerosol science literature [4], of EAM over direct light scattering for measuring the very small (i.e., low nm diameter range) particles typically produced by LC detectors. The approach was therefore mainly geared toward addressing some of the limitations of evaporative light scattering detection (ELSD), which at the time had been used for LC detection for about 20?years. Subsequent collaboration between TSI and ESA Biosciences, Inc. led to the introduction of the first commercial instrument, the Corona® CAD®, in 2005 [5]. While there are some differences among these early EAM-based LC devices and with newer commercial instruments, the basic detection process remains the same. Therefore, Kaufman's patent disclosure and Dixon's article are acknowledged as the primary theoretical descriptions of CAD.

Since its commercial introduction in 2005, CAD has been widely adopted for a broad range of chromatographic applications. CAD and other aerosol techniques, including ELSD and condensation nucleation light scattering detection (CNLSD) [6], are described as "universal" since response depends primarily on aerosol particle size and number concentration (e.g., number of particles/cubic centimeter of gas) rather than individual analyte properties. These "common property" measurement characteristics provide significant advantages over other devices whose detection scope (viz., range of chemicals for which a useful response can be obtained) and sensitivity (viz., signal output per unit mass or per unit concentration) are highly dependent on analyte nature such as optical properties (e.g., ultraviolet (UV) absorption, fluorescence (Fl)) or propensity to form gas-phase ions (e.g., electrospray ionization with mass spectrometry (ESI-MS)). While UV detection remains a primary technique for many LC analyses, its detection scope is limited to compounds with a sufficient UV chromophore, and its sensitivity varies widely among analytes. Likewise, the detection scope and sensitivity of MS strongly depends on ion source and operational conditions, analyte (e.g., basicity, surface activity), and eluent conditions (e.g., pH). In many respects, CAD is still considered to be relatively new, and its performance for a given application is not completely understood or fully explained by current theory. The objective of this chapter is to provide an updated description of CAD theory that draws upon the primary references, aerosol science literature, insights presented throughout this book, and more than 200 literature references and review articles that describe its applications in chromatographic analysis [7-11]. Relevant theory and background of aerosol particle measurement technology in the context of non-LC applications will be briefly discussed. Because these non-LC applications typically involve measurement of steady-state or slowly changing aerosols as opposed to the rapidly changing aerosols encountered with LC, simplified models that address transient versus steady-state measurements will be introduced to help describe and predict the expected performance of CAD.

Understanding the properties of aerosols is of great significance in many fields, including those within environmental, industrial, health, and medical disciplines. To a practicing chromatographer, the properties and behaviors of aerosols may be unfamiliar or seem counterintuitive. Fortunately, significant knowledge exists within the field of aerosol science. For readers interested in additional background on aerosol technology, the most recent editions of Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles [4] and Aerosol Measurement: Principles, Techniques, and Applications [12] are highly recommended. To facilitate more detailed descriptions of theory and practice of CAD, basic information about aerosols along with some definitions and conventions used for other topics within this chapter are described in the succeeding text.

An aerosol is defined as a suspension of solid particles and/or liquid droplets in a gas and is most commonly described as a two-phase system (gas and condensed phase). For LC detection, the gas includes the flow that is supplied to the detector and any component from the liquid eluent that is vaporized. The evaporation process with LC detection involves a transition from liquid droplets to, typically, solid particles, but, in some cases, the "stable particles" that remain after evaporation are still in liquid form (e.g., oils). In describing LC detection, we will assume that evaporation is complete before the aerosol exits the evaporation tube and, unless otherwise specified, will use the terms droplet for species before evaporation and residue particle for those after evaporation, irrespective of the...