Aerosol-based detectors function in three steps: nebulization of the HPLC effluent, evaporation of the solvent leaving aerosol particles, and detection of the aerosol, with the last step depending on the type of aerosol-based detector. Evaporative light scattering detection (ELSD), which measure the light-scattering ability of the particles, has been commercially available for decades. The use of condensation nucleation counters, which "grow" and detect particles, has resulted in a more sensitive detection method known as condensation nucleation light scattering detection (CNLSD). More on CNLSD is available at John Koropchak's internet site. Aerosol-based detection methods are considered to be "universal".

We have been investigating the construction of sensitive and potentially economical HPLC detectors using aerosol charging for detecting the aerosol. We have used the name aerosol charge detection (ACD) although the commercial instrument (described more below) is called the charged aerosol detector (CAD). The first ACD instrument was built using nebulizers, a spray chamber, heated tubing and an Electrical Aerosol size Analyzer (EAA) for aerosol detection. In the flow injection analysis (FIA) mode (i.e. without a chromatography column) and using water as the solvent, detection limits around 10 ppb were achieved with linear response observed from 0.1 to around 10 to 100 ppm (depending on the nebulizer used). When using typical HPLC solvents, the sensitivity often is limited by non-volatile impurities present in solvents. It is possible to improve the signal to noise by using the EAA to remove small particles. Detection limits of about 20 ppb (similar to CNLSD and more sensitive than ELSD) and good reponse linearity were observed for weakly retained compounds separated using reverse-phase HPLC. This detection method is documented in the Dixon and Peterson paper in the publications section of my homepage. To see an example chromatogram set (using both AC detection and UV detection) showing the separation and detection of 5 ppm ammonium sulfate, uracil, sulfanilamide, nicotinamide, and adenosine, click here.

A commerical instrument using aerosol charging, but through the use of colliding ion and aerosol jets, has been developed by ESA Biosciences. This instrument won a Silver Editors' Choice Award at the 2005 Pittsburgh Convention and an R&D100 Award. More information on the instrument can be found here. Sensitivity of the CAD instrument is similar to our first ACD instrument. A number of papers have come out documenting the CAD and some applications.

Current work is focused on creation of related detectors utilizing other mechanism of aerosol charging and involving Bill DeGraffenreid (CSUS Physics and Astronomy Department). Work also has been undertaken to use ACD/CAD for practical applications. This has involved projects describe below and is starting to involve other CSUS faculty.

1. Applications related to my research in determination of aerosol chemistry- Atmospheric aerosols are suspended particles in the atmosphere. The collection of particles making up aerosols is also know as particulate matter. The composition of organic compounds found in atmospheric aerosols is not well known. The focus on determining the composition in the past has focused on using gas chromatography (GC) with more recent interest in real-time mass spectrometry. Analysis of many polar compounds by GC is difficult unless compounds are derivatized to less polar/more volatile compounds. We have focused on developing methods for methods using HPLC-ACD for analyzing such compounds. We have worked on methods to analyze dicarboxylic acids and sugars. The work on dicarboxylic acids has given relatively high detection limits (around 1 ppm) due to high baselines and volatility of the smaller dicarboxylic acids. Application of HPLC-ACD to analysis of sugars, specifically monosaccharide anhydrides (MAs) such as levoglucosan, have been more successful. We have been using a Ca impregnated cation exchange column for the sugar separations with 100% water eluent. The initial work (see J. Chromatogr. article in puclications section) gave detection limits of about 0.1 ppm or 5 ng injected). This method has been applied to a number of sample sets (described in section 3) with further improvements reducing detection limits to 40 ppb or 2 ng injected (documented in the recent Atmos. Environ. article in the publications section) and more recently to under 10 ppb or 1 ng injected. The California Air Resources Board plans to use of a variant of this method for measurement of levoglucosan.
2. Use of FIA-ACD for sensitive measurement of non-volatile solute- A number of specific applications, such as measurment of dissolved solids in water or hexane soluble materials in soil, require the measurement of non-volatile solute in samples. Traditionally this has been done by tedious gravimetric methods that are not very sensitive (based on typical 0.1 mg uncertainties in balances). While many of the specific applications have been replaced by more modern methods that tend to correlate well with the measured quantity, there is no universal method for non-volatile solute measurement. For example, water quality often is determined by electrical conductivity which generally correlates with dissolved solid content. The FIA-ACD response is expected to be correlated with non-volatile solute mass. In collaboration with other CSUS faculty who are providing us with samples, we are investigating using FIA-ACD for analysis of dissovled solids in river water, motor oil contamination of soils, and lipids in fish eggs. We also have analyzed aerosol samples to estimate the water-soluble aerosol concentrations.
3. Quantitation of oligosaccharides- in collaboration with Tom Peavy (CSUS Biological Sciences). Oligosaccharides are important constituents in molecular recognition in cells. We have been working to develop a method to quantitate different oligosaccharides using HPLC-ACD. Because of the complexity of oligosaccharides, qualitative and quantitive analysis of individual oligosaccharides is difficult. In this project, we are focusing on using hydrophillic interaction HPLC-ACD to quantitate various oligosaccharides (actual identification will require other techniques) present in amphibian egg samples.

Atmospheric aerosols are known to adversely affect health, to decrease visibility, and to affect the earth's climate. Knowledge of the composition of atmospheric aerosols is useful in understanding sources of particulate matter as well as for understanding how particles behave (e.g. scatter light or nucleate cloud droplets).

We have worked on using different techniques of using ACD to analyze atmospheric aerosols. Initial interest was in using FIA-ACD to determine water soluble aerosol mass concentrations. When combined with other measurements this may be useful for better characterizing water soluble organic mass.

Because our method for analysis of MAs has been the most successful, we have applied this to the largest sets of aerosol samples. Analysis has consisted of a small set of samples from New Mexico, UC Davis, and CSUS for method development. We were able to detect levoglucosan in all winter samples collected, indicating the presense of woodsmoke. We also were able to see forest fire smoke from samples collected in August, 2002. Beyond these initial samples, we also have investigated a set of samples from Montana collected in August and September, 2003, which was a time of great forest fire smoke influence, and a set of samples from cities in and near the Central Valley of California collected by the CA Air Resources Board in January, 2001. The results from the set of samples from Montana have been publised recently (see Ward et al. paper in publications section). In the Central Valley samples, aerosol levoglucosan concentrations of up to about 10 µg/m3 were observed in winter in communities strongly affected by wood burning. We have used the concentrations of MAs to estimate the source of particulate matter from wood smoke in these communities. We have started analyzing additional samples from cities affected by the August 2002 forest fires in southern Oregon.

Some of the samples we have or are analyzing are associated with the following forest fires:

• Star Fire (Lake Tahoe Region), August/Sept., 2001.
• Southern Oregon fires of August, 2002.
• Forrest fires around Missoula, Montana, August, 2003.