A review of atmospheric aerosol measurements

https://doi.org/10.1016/S1352-2310(99)00455-0Get rights and content

Abstract

Recent developments in atmospheric aerosol measurements are reviewed. The topics included complement those covered in the recent review by Chow (JAWMA 45: 320–382, 1995) which focuses on regulatory compliance measurements and filter measurements of particulate composition. This review focuses on measurements of aerosol integral properties (total number concentration, CCN concentration, optical coefficients, etc.), aerosol physical chemical properties (density, refractive index, equilibrium water content, etc.), measurements of aerosol size distributions, and measurements of size-resolved aerosol composition. Such measurements play an essential role in studies of secondary aerosol formation by atmospheric chemical transformations and enable one to quantify the contributions of various species to effects including light scattering/absorption, health effects, dry deposition, etc. Aerosol measurement evolved from an art to a science in the 1970s following the development of instrumentation to generate monodisperse calibration aerosols of known size, composition, and concentration. While such calibration tools permit precise assessments of instrument responses to known laboratory-generated aerosols, unquantifiable uncertainties remain even when carefully calibrated instruments are used for atmospheric measurements. This is because instrument responses typically depend on aerosol properties including composition, shape, density, etc., which, for atmospheric aerosols, may vary from particle-to-particle and are often unknown. More effort needs to be made to quantify measurement accuracies that can be achieved for realistic atmospheric sampling scenarios. The measurement of organic species in atmospheric particles requires substantial development. Atmospheric aerosols typically include hundreds of organic compounds, and only a small fraction (∼10%) of these can be identified by state-of-the-art analytical methodologies. Even the measurement of the total particulate organic carbon mass concentration is beset by difficulties including the unknown extent of evaporative losses during sampling, adsorption of gas-phase organic compounds onto sampling substrates, and the unknown relationship between carbon mass and mass of the particulate organics. The development of improved methodologies for such measurements should be a high priority for the future. Mass spectrometers that measure the composition of individual particles have recently been developed. It is not clear that these instruments will provide quantitative information on species mass concentrations, and more work is needed to routinely interpret the vast quantities of data generated during field sampling. Nevertheless, these instruments substantially expand the range of atmospheric aerosol issues that can be explored experimentally. These instruments represent the most significant advance in aerosol instrumentation in recent years.

Introduction

Atmospheric aerosol particles range in size over more than four orders of magnitude, from freshly nucleated clusters containing a few molecules to cloud droplets and crustal dust particles up to tens of microns in size. Average particle compositions vary with size, time, and location, and the bulk compositions of individual particles of a given size also vary significantly, reflecting the particles’ diverse origins and atmospheric processing. Particle surface composition is also an important characteristic since it affects interfacial mass transfer and surface reactions, which play a role in atmospheric chemical transformations. Such transformations can be significant both for their effects on gas-phase composition, as in stratospheric ozone depletion, and for their effects on particle composition. The production of fine (sub 2.5 μm) sulfates by liquid transformations in clouds is an example of a process that involves gas-to-particle mass transfer of species including water, sulfur dioxide, and oxidants.

An aerosol is defined as a suspension of liquid or solid particles in a gas. In reviewing aerosol measurement it is important to remember the gas. While atmospheric particles contain nonvolatile species such as salt, soot, metals, and crustal oxides, they also contain semivolatile compounds such as nitrates and many organic compounds. The distribution of such semivolatile compounds between the gas and particle phases varies with the amount of available particulate matter on which they can accumulate, the thermodynamic properties of the semivolatile compounds, and the gas and particle composition. Furthermore, fine (<2.5 μm) atmospheric particles are mostly hygroscopic and the water mass fraction in the condensed phase increases with relative humidity. Water typically constitutes more than half of the atmospheric fine particle mass at relative humidities exceeding roughly 80%. Thus, particle composition is inextricably linked with the composition of the gas phase, adding to the challenge of adequately characterizing the aerosol. Furthermore, sampling and/or measurement can change the thermodynamic environment or gas-phase composition thereby causing changes in particle composition before measurements are carried out.

In his visionary articles Friedlander, 1970, Friedlander, 1971 introduced a conceptual framework for characterizing instruments used for aerosol measurement. In these articles, he defined the aerosol size-composition probability density function g(v,n1,…,nk−1) for an aerosol containing k chemical species. This function is defined such that the fraction of the total number concentration N having particle volume between v and v+dv, and molar composition of species i between ni and ni+dni at time t isdNN=g(t,v,n1,…nk−1)dvdn1dnk−1.Only k−1 species are specified as independent variables because particle volume depends on the species’ molar composition:v=i=1kniv̄i,where v̄i is the partial molar volume of species i. This formulation does not explicitly account for particle charge states, surface composition, morphologies, phase composition, etc., but it could in principle be generalized to include such information. Gas-phase compositions are implicitly coupled through the dependence of particle composition ni on the gas phase.

Knowledge of Ng(t,v,n1,…nk−1) would provide a comprehensive characterization of the size-resolved aerosol composition, including variations in composition among particles of a given size. Advances in single-particle mass spectrometry during the past several years have moved us closer to making such information a reality. Most aerosol measurements, however, provide integrals over time, size, and/or composition.

Fig. 1, adapted from Friedlander (1971), illustrates the type of information provided by various aerosol instruments in terms of Ng(t,v,n1,…nk−1). The following notation is used to indicate integrations over size, time, and composition:t1t2v1v2Wgdn1dnk−1dvdtt2−t1v1v2∫Wgdnidv.The weighting factor, W(v), for continuous integral measurements depends on the integral aerosol property being measured. Examples of weighting factors include:W(v)=1.0forCNCs,W(v)=πD2p4·Kspforintegratingnephelometers,W(v)=ρp·vformassmeasurement,where Dp is the particle diameter, Ksp the single-particle scattering efficiency, and ρp the particle density. Additional information on integral measurements is available from various sources (e.g., Friedlander, 1977; Hinds, 1982; Seinfeld, 1986).

Because the available instruments use a variety of approaches to measure particle size, different sizes can be reported for the same particle. For example, the “aerodynamic size” obtained with impactors and aerodynamic particle sizers depends on particle shape, density, and size, while the “electrical mobility size” obtained by electrostatic classification depends on particle shape and size but not on density. “Optical sizes”, which are determined from the amount of light scattered by individual particles, depend on particle refractive index, shape, and size. These sizes can be quite different from the “geometric” or “Stokes” sizes that would be observed in a microscope. Converting from one measure of size to another typically involves significant uncertainty. Such conversions, however, are often essential in utilizing aerosol measurements. These observations underline the importance of understanding the means used to measure sizes and of developing techniques to measure such properties including shape, density, and refractive index.

Laboratory calibrations can provide a misleading impression of accuracies that can be achieved when an instrument is used to measure atmospheric aerosols. Similar instruments that have been carefully calibrated in the laboratory may disagree when used for ambient aerosol measurements due to subtle difference in size cuts, or different sensitivities to aerosol hygroscopic properties, particle density or hygroscopicity. Therefore, rather than provide a misleading table of measurement precision and accuracy, I have discussed factors that affect measurement accuracy when discussing individual measurement techniques.

This review of aerosol instrumentation is organized according to the categories suggested by Friedlander with the order of presentation following Fig. 1. We first discuss measurements that provide a single piece of information integrated over size and composition and progress towards instruments that provide more detailed resolution with respect to size and time. We then follow a similar progression for instruments that measure aerosol chemical composition.

A previous comprehensive review on ambient particulate measurements was written by Chow (1995). Chow's paper focuses on fixed-site sampling and includes comprehensive discussions of size-selective inlets, flow measurement, filter media, methods and sensitivities of analytical methodologies, etc. Much of the material that was discussed in Chow's article is pertinent to the NARSTO review, and the reader is referred to her paper for an in-depth critical review of measurements used for compliance monitoring. The present paper complements this earlier review.

Section snippets

Aerosol sampling inlets

The ideal aerosol sampling inlet would draw in 100% of the particles in a specified size range and would transport them all without modification to the detector or collector. Unfortunately, obtaining representative samples of aerosols can be difficult. The efficiency with which particles enter the inlet can be more or less than 100% and varies with particle size, wind speed, and direction. Particles can be lost en route from the inlet to the measurement device, and thermodynamic changes in the

Integral measurements

Instruments that provide totals (integrals) of specified variables over a given size range are often used for aerosol measurement. For example, condensation nucleus counters provide the total number concentration of particles larger than a minimum size, and cloud condensation nuclei counters measure the subset of particles that can form cloud droplets when exposed to water vapor at a specified supersaturation. Filter samplers are often used to measure total mass concentrations, integrated with

Off-line measurements

Measurements of particle composition typically involve the chemical analysis of deposited particles in a laboratory some time after sample collection. Filters are the most commonly used collection substrates, but a variety of films and foils have been used with impactors to collect size-resolved samples. Sampling times vary with ambient loadings, sampling rates, substrate blanks, and analytical sensitivities but typically vary from several hours in urban areas to a day or more under clean

Calibration of atmospheric aerosol instrumentation

A review of techniques used to produce calibration aerosols is given by Chen (1993). In this section the techniques that are most commonly used to calibrate atmospheric aerosol instrumentation are discussed. Significant progress has been made since 1970 in the development of techniques for generating calibration aerosols, but the measurement of atmospheric aerosols introduces challenges that are not all resolved by these tools.

As was mentioned in the introduction, measured particle “sizes”

Federal reference method

The original National Ambient Air Quality Standard (NAAQS) for particulate matter was for “total suspended particulate matter” (TSP) and was in force from 1970 to 1987, when it was replaced by a standard for particles smaller than 10 μm aerodynamic diameter (PM10) (Register, 1987). More recently, an additional fine particle (PM2.5) has been proposed (Register, 1997). These methods are briefly reviewed in light of the preceding discussion on measurement methodologies.

The PM10 standard defines

Summary of significant advances

  • Instrumentation for producing laboratory calibration aerosols of known size, composition and concentration became available about 25 years ago. This instrumentation is now widely used to characterize the response of aerosol instrumentation to known aerosols. These calibration techniques have facilitated a steady advance in the quality of atmospheric aerosol measurements.

  • Mass spectrometers that can measure the composition of individual atmospheric particles in real time are now available. These

Summary of future aerosol measurement needs

  • Gravimetric techniques that are used for regulatory compliance purposes involve filtration. While such methods are relatively simple and inexpensive to implement, they require manual operation, provide only rough time and spatial resolution, and are subject to sampling errors that cannot be quantified. Real-time techniques for accurate measurement of mass that avoid such sampling errors are needed.

  • The response of aerosol instruments depends on particle properties including density, complex

Acknowledgements

Preparation of this review was supported in part by the Electric Power Research Institute through Grant No. EPRI W09116-08/W04105-01 and in part by the Department of Energy through Grant No. DE-FG02-91ER61205. Colleagues too numerous to mention have readily responded to my requests for information. Thank you all.

References (395)

  • S.M Buhr et al.

    Development of a semi-continuous method for the measurement of nitric acid vapor and particulate nitrate and sulfate

    Atmospheric Environment

    (1995)
  • S.H Cadle et al.

    Problems in the Sampling and Analysis of Carbon Particulate

    Atmospheric Environment

    (1983)
  • D.C Camp et al.

    Intercomparison of concentration results from fine particle sulfur monitors

    Atmospheric Environment

    (1982)
  • R.J Charlson et al.

    On the generality of correlation of atmospheric aerosol mass concentration and light scatter

    Atmospheric Environment

    (1968)
  • Y.S Cheng et al.

    Theory of a screen-type diffusion battery

    Journal of Aerosol Science

    (1980)
  • D.W Cooper et al.

    The inversion matrix and error estimation in data inversionApplication to diffusion battery measurements

    Journal of Aerosol Science

    (1990)
  • B.E Dahneke et al.

    Properties of continuum source particle beams. I. Calculation methods and results

    Journal of Aerosol Science

    (1979)
  • G.D Danilatos

    Foundations of environmental scanning electron microscopy

    Advances in Electronics and Electron Physics

    (1988)
  • L de Juan et al.

    High resolution size-analysis of nanoparticles and ionsRunning a DMA of near optimal length at Reunolds numbers up to 5000

    Journal of Aerosol Science

    (1998)
  • F De Wiest et al.

    Sur la validite des determinations du benzo(a)pyrene atmospherique pendant les mois d'ete (Short Communication)

    Atmospheric Environment

    (1976)
  • D.J Eatough et al.

    A multiple-system, multi-channel diffusion denuder sampler for the determination of fine-particulate organic material and the atmosphere

    Atmospheric Environment

    (1993)
  • H.J Fissan et al.

    Determination of particle size distributions by means of an electrostatic classifier

    Journal of Aerosol Science

    (1983)
  • R.G Flocchini et al.

    Characterization of particles in the arid west

    Atmospheric Environment

    (1981)
  • V.L Foltescu et al.

    Corrections for particle losses and sizing errors during aircraft aerosol sampling using a Rosemount inlet and the PMS LAS-X

    Atmospheric Environment

    (1995)
  • S.K Friedlander

    The characterization of aerosols distributed with respect to size and chemical composition

    Journal of Aerosol Science

    (1970)
  • M Adachi et al.

    Electrical neutralization of charged aerosol particles by bipolar ions

    Journal of Chemical Engineering of Japan

    (1983)
  • K.M Adams

    Real-time in situ measurements of atmospheric optical absorption in the visible via photoacoustic spectroscopy. 1evaluation of the photoacoustic cells

    Applied Optics

    (1988)
  • Aden, G.D., Buseck, P.R., 1983. Aminicomputer procedure for quantitative EDS analyses of small particles. Microbeam...
  • Aitken, J., 1890–1891. On a simple pocket dust-counter. Proceedings of the Royal Society of Edinburgh XVIII,...
  • Allen, G.A., Turner, W.A., Wolfson, J.M., Spengler, J.D., 1984. Description of a continuous sulfuric acid/sulfate...
  • G.S Almasi et al.

    A heat-flow problem in electron-beam microprobe analysis

    Journal of Applied Physics

    (1965)
  • G.P Ananth et al.

    Theoretical analysis of the performance of the TSI aerodynamic particle sizer

    Journal of Aerosol Science

    (1988)
  • J.R Anderson et al.

    Chemistry of individual aerosol particles from Chandler, Arizona, an Arid Urban environment

    Environmental Science and Technology

    (1988)
  • T.L Anderson et al.

    Performance characteristics of a high-sensitivity, three wavelength, total scatter/backscatter nephelometer

    Journal of Atmospheric and Oceanic Technology

    (1996)
  • M.O Andreae et al.

    Internal mixture of sea salt, silicates, and excess sulfate in marine aerosols

    Science

    (1986)
  • B.R Appel et al.

    Atmospheric particulate nitrate sampling errors due to reactions with particulate and gaseous strong acids

    Atmospheric Enviroment

    (1981)
  • J.T Armstrong et al.

    Quantitative chemical analysis of individual microparticles using the electron microprobe

    Analytical Chemistry

    (1975)
  • W.P Arnott et al.

    Thermoacoustic enhancement of photoacoustic spectroscopytheory and measurements of the signal to noise ratio

    Review of Scientific Instruments

    (1995)
  • W.P Arnott et al.

    Photoacoustic spectrometer for measuring light absorption by aerosol: Instrument description

    Atmospheric Environment

    (1999)
  • P Artaxo et al.

    Trace elements and individual particle analysis of atmospheric aerosols from the antarctic peninsula

    Tellus

    (1992)
  • P.A Baron

    Calibration and use of the aerodynamic particle sizer (APS 3300)

    Aerosol Science and Technology

    (1986)
  • Baron, P.A., Mazumder, M.K., Cheng, Y.S., 1993. Direct-reading techniques using optical particle detection. Aerosol...
  • D Baumgardner et al.

    Refractive, indices of aerosols in the upper troposphere and lower stratosphere

    Geophysical Research Letters

    (1996)
  • D Baumgardner et al.

    The Multiangle Aerosol Spectrometer Probe (MASP)A New Aerosol Probe for Airborne Stratospheric Research

    (1993)
  • R.N Berglund et al.

    Generation of monodisperse aerosol standards

    Environmental Science and Technology

    (1973)
  • R.G Beuttell et al.

    Instruments for the measurement of the visual Range

    Journal of Science and Instruction

    (1949)
  • W Birmili et al.

    Determination of differential mobility analyzer transfer functions using identical instruments in series

    Aerosol Science and Technology

    (1997)
  • P Biswas et al.

    Distortion of size distributions by condensation and evaporation in aerosol instruments

    Aerosol Science and Technology

    (1987)
  • W.D Bowers et al.

    Surface acoustic wave piezoelectric crystal aerosol mass monitor

    Review of Scientific Instrument

    (1989)
  • J Bricard et al.

    Detection of ultra-fine particles by means of a continuous flux condensation nuclei counter

  • Cited by (684)

    View all citing articles on Scopus

    Prepared for the NARSTO assessment.

    View full text