This snapshot taken on 16/07/2006, shows web content selected for preservation by The National Archives. External links, forms and search boxes may not work in archived websites.
Methods of Measurement of Airborne Particles2
Summary
Appropriate measurement of airborne particulate matter presents problems, since it is not known for sure which of its various components are responsible for its observed harmful effects. Ideally, the method of measurement should represent this toxic fraction or component. Particle concentrations can be measured as mass (weight) of particles within a certain size range; as the number of particles; as the mass of one or several chemicals; or as a specific surface area, all in a given volume of air. Most measurements in the past and presently have been of mass, either indirectly as Black Smoke or directly as the weight of particles below a certain size or as the weight of sulphate. Lately, there has been some interest in counting the number of particles below a certain diameter.
All the commonly used measuring instruments have advantages and disadvantages, discussed in this chapter. At present that most widely used in the UK, the TEOM, underestimates the mass of PM10 and of PM2.5, the latter to a significantly greater extent. This is mainly a consequence of loss of volatile matter, particularly nitrate, at the temperature at which the TEOM operates. Whether this is of significance from a regulatory point of view is uncertain. Other samplers are also subject to measurement errors, but generally to a lesser degree. However, until the toxic component of particulate pollution is more clearly defined it will not be possible to recommend an ideal method of measurement. The underestimation of particle mass by the TEOM does not imply any lessening of protection of public health since the air quality standard recommended by this Panel took account of the results of an epidemiological study conducted in the UK which linked adverse health outcomes with TEOM-derived PM10 measurements.
21. Airborne particle monitoring instruments generally share a number of common features. A pump draws air at a known flow rate through a validated sampling head which selects a particular size fraction of the ambient airborne particles. The collected particles are then analysed in different ways, depending on the metric being measured. For example, the mass of particles can be determined gravimetrically, when particles are deposited on a filter over a sampling period of normally 24 hours. The mass of particles collected on the filter is determined by weighing after conditioning at controlled temperature and relative humidity. The collected sample can also be analysed chemically to determine the composition of the sampled particles. In direct reading instruments, the selected particles are either deposited on a filter, with continuous assessment of the change of a property of the filter due to their presence, or passed through an optical sensing device. The Black Smoke methodology relies on the drawing of air through a funnel, collecting particles on a filter and then assessing the “blackness” of the stain by measurement of light reflectance from the surface of the filter. There is increasing interest in the so-called ultrafine particles, for which the common metric is the particle number concentration. Total particle number concentrations are measured using a Condensation Particle Counter (CPC). A supersaturated vapour is condensed onto the particles, causing them to grow into larger droplets. A laser light source can then detect and count these droplets. Fractionated particle counts may also be carried out using a Scanning Mobility Particle Sizer (SMPS). An electrostatic classifier is used to fractionate the particles according to size before they are counted by the CPC.
22. Because of the diverse nature of the chemical and physical properties of particles in the atmosphere, there are many possible measurement procedures. Other pollutants, such as sulphur dioxide, may also be measured by several different methods, but in principle all should yield the same measurement of concentration. In the case of particulate matter, however, the different techniques quantify different properties of the airborne particles, and hence a number of different methods are available for expressing the concentration of airborne particulate matter. This can provide problems in that conversion from one metric to another is rarely possible, as the relationships are site- and time-specific. On the other hand, this variety offers opportunities for selection of the metric which will best describe the property of the particles to be investigated. Whilst the European Committee for Standardisation has designated three reference methods (EN 12341), these relate only to the measurement of particle mass and do not take account of other metrics which might relate more closely to particle toxicity.
23. Until the early 1990s, almost all measurements of particulate matter in the UK atmosphere were derived from the Black Smoke method. This is based upon the collection of particles by drawing air through a filter, the particle concentration being estimated from the blackness of the filter, which is evaluated by measuring the efficiency with which the filter reflects light shone upon it (Figure 5). The method is sensitive only to the light-absorbing components of the particulate matter, which tend to be dominated by carbon in its elemental form, present in soots from combustion processes. Components of the particles that do not absorb visible light, typified by the secondary materials ammonium nitrate and ammonium sulphate, are not measured by this procedure. Because the method is responsive to only one major component of the particulate matter, a standard calibration curve has to be used to convert the reflectance to a total mass concentration of particles in the atmosphere. The calibration curve in use today was determined in the 1960s at a time when coal smoke still dominated the particle loading of the UK urban atmosphere (QUARG 1993). Now that this is no longer the case, Black Smoke measurements do not provide a reliable quantification of the mass concentration of particles in the atmosphere, but rather reflect the contribution of combustion sources, most particularly diesel vehicles. The British Black Smoke method is closely related to the OECD method upon which the European Directive on smoke and sulphur dioxide is based; the two methods are correlated, the measurements differing by a constant factor 0.85. The coefficient of haze measurement, at one time used in North America, although not identical with the Black Smoke measurement, is also based on the absorptive properties of particles. The air sampler traditionally used for Black Smoke in the UK has been shown to sample only particles smaller than about 4.5 µm (Figure 5; McFarland et al 1982). This is sufficient to include almost all particles formed in combustion but will exclude coarser particulate matter. Black Smoke data provide a valuable record of past combustion-generated air pollutant concentrations and can be a great asset in epidemiological studies.
24. The main alternatives to Black Smoke methods are the gravimetric methods of particle measurement. Such methods depend upon weighing the mass of particles collected by drawing a known volume of air through a filter paper. For many years this was carried out in North America using the high volume sampler method (Figure 6). This device samples air at a rate of 1-2 cubic metres per minute and therefore collects milligram quantities of particles over a 24-hour period. After the filter has been conditioned at controlled temperature and humidity, the weight change during sampling reveals the mass of airborne particles. Originally, this method was used with air samplers having an unsymmetrical apex roof above the filter, with a wide slot below the roof through which airborne particles were drawn. With this sampler configuration the measurement is known as total suspended particulate matter (abbreviated to TSP). Investigation of the inlet characteristics of the TSP sampler showed them to be highly sensitive both to orientation with respect to wind direction and to wind speed, particles in excess of 20 µm diameter being sampled efficiently under some conditions, whilst under others, efficiency was poor even for particles below 10 µm diameter (Wedding et al 1977).
25. The International Standards Organisation (ISO) has defined three criteria for biologically-relevant size-selective aerosol sampling, primarily for studies of workers exposed to hazardous dust in industrial situations: respirable dust; thoracic dust and inhalable dust (ISO 1994; Figure 7). Inhalable dust is the mass fraction of total airborne particles which is inhaled through the nose and mouth, the thoracic fraction is the mass fraction of aerosol which penetrates beyond the larynx, and the respirable fraction is the mass of inhaled particles penetrating to the unciliated, smallest, airways of the lung. Other arbitrary size-selective sampling criteria have been defined for studies of general populations, notably PM10 and PM2.5 - particulate matter sampled through an inlet with a 50% cutoff at 10 µm and 2.5 µm, respectively. PM10 closely approximates to the ISO thoracic sampling criterion while the respirable criterion would be met by a PM4 sampler (PM3.5 under an older US Standard for respirable dust sampling).
26. The poor characteristics of the TSP sampler led the US Environmental Protection Agency to promote the development of size-selective samplers, and after some years of trials the PM10 sampler was adopted as standard (US EPA 1987). This has a symmetrical inlet, the properties of which are independent of wind direction and, largely, wind speed, which excludes larger particles by having a 50% cut-off point at 10 µm (Figure 8). Thus, to a good approximation, it samples all particles with an aerodynamic diameter of less than 10 µm. Whilst the original PM10 inlets were designed for use with high volume samplers, they are now also available for samplers operating at lower volume flow rates. Samplers can also be fitted with inlets that incorporate a cyclone or impactor giving a 50% cut-off point at either 2.5 µm or 1 µm, hence collecting size fractions referred to as PM2.5 and PM1.0. Reference to Figure 3 (page 7) shows that the respective samplers with inlet cut-off points of 10, 2.5 and 1 µm sample different components of the particulate matter. Thus the PM1.0 inlet collects particles in the nucleation mode together with the smaller accumulation mode particles; the PM2.5 inlet, the nucleation and accumulation modes; whilst the PM10 inlet samples both of these modes together with a proportion of the coarse particle mode. The choice of a 2.5 µm cut-off point between fine and coarse particles may not be optimal for separation of the accumulation and coarse particle modes, but has been adopted internationally.
27. In the UK, the national automatic air quality monitoring networks employ mainly tapered element oscillating microbalance (TEOM) PM10 samplers, together with some TEOM PM2.5 instruments and one beta gauge3 (Table 1). The TEOM instruments were adopted because of their ability to provide measurement data in real time, as compared to measurements from traditional gravimetric methods, which are available only after the period required for collection and weighing. TEOM measurements therefore allow the provision of real time information to the public and input into research to identify sources of particulate matter. In the TEOM sampler, particles are collected on a filter that is attached to the vibrating element of a tapered element oscillating microbalance (Figure 8). The element vibrates at a precise frequency that changes according to the mass of particles on the filter, allowing a direct determination of collected mass. In theory such measurements should be directly equivalent to those of a gravimetric procedure using the same inlet characteristics, but in practice divergences occur in the presence of semi-volatile components of the particulate matter (Figure 9). This is because the TEOM instrument, as used in the national networks, pre-heats the air to 50OC prior to particle collection in order to drive off associated water and other components which might lead to inconsistent and variable mass measurements. This causes loss of some semi-volatile particulate components (e.g. ammonium nitrate and some organic compounds). In the UK atmosphere TEOM instruments give lower measurements of particle mass than both high and low volume samplers that do not involve pre-heating the sampled air, although these too will lose some volatile material (APEG 1999). Because the differences depend upon the chemical composition of the particulate matter, they are dependent on both site and season, and no universally applicable correction is available. In measuring PM10 the divergence is typically of the order of 20%, but rarely exceeds 40%. It should be noted that the air quality standard for PM10 recommended by this Panel took account of the results of an epidemiological study conducted in Birmingham that used data from the national network TEOM sampler. The under-reading of the instrument does not, therefore, imply any weakening of the protection of public health.
28. Gravimetric PM10 monitoring using low volume samplers is currently being carried out at seven national network monitoring stations in the UK. At six of these sites, intercomparison trials with TEOM instruments are also being carried out with one of the European transfer reference samplers, the Kleinfiltergerat, which is also equipped with a size-selective inlet. A further eight PM10 monitoring stations are equipped with low volume PM2.5 samplers.
29. Table 2 shows results from the UK air pollution monitoring networks in 1998. The recent report of the Airborne Particles Expert Group (APEG 1999) analysed the available measurement data on PM10 and PM2.5 in the UK. Concentrations are highest at roadside sites and decline between these and urban background and thence from suburban to rural sites. Thus the average concentrations of PM10 from June 1997 until February 1998 at four of the sites were: at an urban roadside site at London Marylebone Road, 37.90 µg/m3; at a central urban background site at London Bloomsbury, 26.12 µg/m3; at a suburban site at Rochester, 19.32 µg/m3; and at a rural site at Harwell, 16.60 µg/m3 . At each of these sites measurements were made also of PM2.5. The two measurements are correlated, with correlation coefficients for hourly data in the range 0.59 to 0.92. Coarse particle concentrations were also correlated with PM10, but less strongly so (correlation coefficient ranges from 0.30 to 0.61 for hourly data). The proportion of PM10 mass comprising the finer PM2.5 particles is typically within the range of 50 to 80%, with variations between sites and seasons. An analysis of data from paired roadside and urban background sites for 1997 showed traffic generated increments in PM10 concentration between 2 and 13 µg/m3. Careful analysis of the data from the Marylebone Road and London Bloomsbury sites indicates that this road traffic increment is split about half and half between fine and coarse particles, the former arising from exhaust emissions and the latter presumably coming predominantly from the suspension of road surface dust.
30. Sulphate is of low volatility (and therefore conserved in air samplers) and is a major contributor to fine particle mass. Daily mean particulate sulphate concentrations are currently measured at eight rural sites across the UK. Measurements at Eskdalemuir started in 1978 and the full network was operational by 1987. At these sites the sulphate concentration on filters is determined by ion chromatography. These measurements represent sulphate particles of diameter of less than about 4.5 µm. In North America, widespread use has been made of measurements of sulphate in particulate matter. For many years the United States has operated a network of dichotomous air samplers which use a virtual impaction principle to collect separate fine (less than 2.5 µm) and coarse (10-2.5 µm) particle fractions. As well as recording the mass concentration of these fractions, the fine fraction within which most of the sulphate resides is dissolved in water and analysed for its sulphate content. Comparisons with co-located measurements collected using dichotomous samples in the early 1980s showed that most of the sulphate collected using non-automatic sulphur dioxide bubblers was non-sea salt sulphate in particles of diameter of less than 2.5 µm (Irwin 1982).
31. Daily Total Inorganic Nitrate concentrations were measured at two UK rural sites between 1989 and 1999 using an open-faced cellulose filter impregnated with sodium hydroxide. This measurement represents the sum of particulate nitrate and nitric acid. These measurements have been replaced by monthly measurements at twelve UK sites, which commenced during 1999, to provide separate measurements of these two components. Daily measurements were carried out at one site in the south east of England until late 2000.
32. A number of other metrics are available for quantifying airborne particles, but are less commonly applied. Condensation particle counters (CPCs) depend upon condensing a vapour (often n-butanol) onto particles, which are then counted as they pass through a light beam (Booker et al 1998). If the CPC is equipped with an electrostatic classifier, then size-fractionated particle counts can be carried out, as with the scanning mobility particle sizer (SMPS). Instruments are available which count all particles down to 0.003 µm (3 nm) diameter. In practice, the number count is dominated by particles of less than 0.1 µm diameter, hence reflecting freshly formed or emitted nucleation mode particles together with part of the accumulation mode. Instruments such as the epiphaniometer (Gaggeler et al 1989) are also available for measurement of the surface area of airborne particles. To date, however, they have not been used in routine applications and available data are limited to a few research investigations.
33. Measurements of size-fractionated particle counts are currently being undertaken at three national network sites, at London Bloomsbury, London Marylebone Road and Harwell, with CPC particle counts at a further seven UK locations (Belfast Centre, Birmingham Centre, Glasgow Centre, London Marylebone Road, London North Kensington, Manchester Piccadilly and Port Talbot).
References
APEG. Source apportionment of airborne particulate matter in the United Kingdom. Report of the Airborne Particles Expert Group. London: Department of the Environment, Transport and the Regions, 1999.
Booker DR, Griffiths WD, Lyons CP, Mark D, Upton SL. Aerosol sampling guidelines. AL Nichols (Co-ordinating Editor), Cambridge: Royal Society of Chemistry, 1998.
Gaggeler HW, Baltensperger U, Emmenegger M, Jost DT, Schmidt OH, Haller P, Hoffman M. The Epiphaniometer, a new device for continuous aerosol monitoring. J Aerosol Sci, 1989; 20: 557-564.
International Standards Organisation. Air Quality - particle size fraction definitions for health-related sampling, IS 7708, Geneva: ISO, 1994.
Irwin JG. Atmospheric particulate measurements using dichotomous samplers and ion chromatography. EMEP expert meeting on chemical matters, Geneva, 1982. Lillestrom: Norwegian Institute of Air Research, 1982.
McFarland AR, Ortiz CA, Rodes CE. Wind tunnel evaluation of the British Smoke shade sampler. Atmos Environ, 1982; 16: 325-328.
QUARG. Urban air quality in the United Kingdom. First report of the Quality of Urban Air Review Group. London: Department of the Environment, 1993.
U.S. E.P.A. Federal Register, 40CFR, Part 53, 1987.
Wedding JB, McFarland AR, Cermak JE. Large particle collection characteristics of ambient aerosol samples. Environ Sci Technol, 1977; 11: 387-390.
Figure 5. Components of the British Black Smoke sampler together with its sampling efficiency.
Figure 6. High Volume sampler together with its sampling efficiency.
Figure 7. ISO Health Related Particle Sampling Convention (ISO 1994).
Figure 8. Rupprecht and Patashnick TEOM ambient aerosol monitor and sampling efficiency of the low volume sampling head.
Table 1. Summary of measurement methods used by DETR and the Devolved Administrations in the United Kingdom air monitoring networks for particulate matter (at November 2000).
Measurement metric Measurement method Comments PM10 TEOM4, b-Gauge, EU reference method (Kleinfiltergerat) and Partisol Plus gravimetric instruments PM10 monitored since 1992. Currently 50 TEOM (+ 8 proposed by Jan 2001), 6 Kleinfiltergerat, 6 Partisol Plus gravimetric (+ 7 proposed by Feb 2001) and 1 b-gauge sites in national network PM2.5 TEOM and Partisol Plus gravimetric instruments PM2.5 monitored since 1997. 4 TEOM and 6 Partisol Plus gravimetric sites (+ 2 proposed by Feb 2001) Numbers CPC5total particle numbers (7-1000 nm) SMPS6fractionated particle numbers (11-450 nm) Used in national network since 1997. 7 CPC sites and 3 SMPS sites Black Smoke Reflectance of smoke stain on filter paper Used in national network since 1961. Currently 178 sites in the national network Sulphate Pre-filter on 8 port bubbler - daily Used in acid deposition network since 1987. 8 sites PM10 gravimetric collection using Partisol Plus instruments with analysis by ion chromatography -daily 4 proposed sites (Dec 2000) Nitrate monthly/daily denuder Acid deposition network: 12 monthly sites and 1 daily site (particulate nitrate) Flash vaporisation of PM2.5 fraction followed by continuous NOx measurements Automatic urban network: 2 proposed sites (Dec 2000) Total and elemental carbon Direct thermal CO2 analysis. Organic and elemental carbon differentiated by oxidation 4 proposed sites (Dec 2000) Metals Collection on 0.8 mm pore filters,over 1 week followed by analysis by ICP-OAS for Cd, Cr, Cu, Fe, Mn, Ni, Pb, Zn and V Used in lead and multi-element national network since 1976: 18 heavy metals sites Weekly PM10 gravimetric collection using Partisol 2000 equipment followed by ICP-AES or ICP-MS analysis for Pb, Cd, As, Ni and Hg 30 sites for 12 month monitoring only Polycyclic aromatic hydrocarbons Vapour and particulate phases collected using modified PM10 sampling head (~PM12-18) Currently 15 sites and 10 proposed sites (Dec 2000) Table 2. 1998 Monitoring data for urban roadside, urban background and rural sites for different measures of the concentrations of airborne suspended particulate matter in the United Kingdom.
Month Urban roadside PM2.5 (µg/m3) Urban roadside PM10–2.5 (µg/m3) Urban roadside PM10 (µg/m3) Urban background PM2.5 (µg/m3) Urban background PM10–2.5 (µg/m3) Urban background PM10 (µg/m3) Urban background SO4(µg/m-3) Rural PM2.5 (µg/m3) Rural PM10–2.5 (µg/m3) Rural PM2.5 (µg/m3) January 19.3 9.7 29.0 13.5 9.5 23.0 12.4 4.9 17.3 February 26.6 13.4 40.0 20.1 6.9 27.0 13.3 4.6 17.9 March 21.1 11.9 33.0 15.5 10.5 26.0 11.8 5.4 17.2 April 18.3 10.7 29.0 14.6 5.4 20.0 6.8 8.6 3.1 11.7 May 20.7 12.3 33.0 20.3 7.7 28.0 5.3 14.0 5.3 19.3 June 20.3 12.7 33.0 13.7 6.3 20.0 6.8 8.4 3.8 12.2 July 18.5 12.5 31.0 12.2 6.8 19.0 7.7 8.5 4.7 13.2 August 19.6 12.4 32.0 14.9 9.1 24.0 10.6 10.1 6.0 16.1 September 24.0 12.0 36.0 18.6 5.4 24.0 7.5 13.2 5.4 18.6 October 18.2 10.8 29.0 12.8 7.2 20.0 8.9 7.5 3.7 11.2 November 22.3 11.7 34.0 17.4 6.6 24.0 11.9 10.4 3.5 13.9 December 18.8 10.2 29.0 14.2 5.8 20.0 7.4 8.4 3.6 12.0 Notes:
- PM2.5 and PM10 data were taken for 1998 for London Marylebone Road, London Bloomsbury and Harwell monitoring sites.
- Monthly mean hourly concentrations are shown from the DETR TEOM monitors.
- PM10 -2.5 data were obtained by subtraction of the monthly mean quantities.
- Sulphate data were taken from London Bridge Place.
2The report seeks only to indicate some of the equipment choices available to those wishing to measure airborne particulate matter. It makes no attempt to be exhaustive, and some commercial instruments have not been included. Omission of an instrument should not be inferred as conveying any opinion by the Expert Panel and equally, inclusion does not constitute an endorsement by EPAQS.
3Beta Gauge: an instrument which evaluates the mass of particles collected on a filter from their ability to attentuate a beam of beta particles
4 TEOM - Tapered Element Oscillating Microbalance
5 CPC - Condensation Particle Counter
6 SMPS - Scanning Mobility Particle Sizer
Published 17 May 2001
Expert Panel on Air Quality Standards
Air and Envirnonmental Quality
Environmental Protection Index
Defra Home Page
