This chapter describes the wide range of methods available for particulate monitoring and provides guidance on the suitability of methods for different purposes.
5.1 Monitoring for different fractions
There are a number of approaches and methods for measuring particulates in the air, and many of these have been used in New Zealand. Different approaches measure different properties of atmospheric particulates, and therefore care must be taken before selecting a monitoring method or attempting to compare the results of different methods.
PM10 is atmospheric particulate matter less than or equal to 10 micrometres (µm) in diameter. This is the fraction of atmospheric particulates that are small enough to penetrate deep into the human lung. To monitor PM10, the sample air enters a size-selective inlet which has at least 50 per cent efficiency cut-off at a 10 µm aerodynamic diameter. The resulting air stream contains particulate matter generally less than 10 µm (with a small proportion of particulate matter greater than 10 µm).
PM10 can arise from a wide range of sources, but can generally be separated into three categories:
primary combustion particulates – produced directly from combustion, such as domestic heating, road transport, power stations and industrial processes
secondary particulates – aggregates in the atmosphere following their release as gases (include nitrates and sulphates)
coarse particulates – from non-combustion sources such as re-suspended road dust, construction work, mineral extraction, wind-blown dust and soil, and sea salt.
PM2.5 is particulate matter less than or equal to 2.5 µm in diameter. PM2.5 is emitted from primary combustion processes and requires the appropriate size-selective inlet for sampling.
Current research indicates that it has a greater effect on health than PM10. The AAQG include a monitoring guideline of 25 µg/m3 as a 24-hour average for PM2.5. This is equivalent to the World Health Organization (WHO) ambient air quality guideline for PM2.5 as a 24-hour average. WHO further provides an ambient air quality guideline of 10 µg/m3 as an annual average for PM2.5.
Standards for monitoring PM2.5 include:
AS/NZS 3580.9.10:2006, Methods for sampling and analysis of ambient air – Determination of particulate matter – PM2.5low volume sampler – Gravimetric method
US Code of Federal Regulations Title 40, Part 50 Appendix L, Reference method for the determination of fine particulate matter as PM2.5in the atmosphere.
Systems complying with the US specification are notified in the Federal Register, v. 63, p. 18911, 16 April 1998, and v. 63, p. 31991, 11 June 1998. These cover a variety of 24-hour average, low-volume systems fitted with a PM2.5 inlet. Some of the units are fitted with automated filter-changing systems, which allow for unattended operation over extended periods of time.
5.1.3 Total suspended particulate (TSP)
TSP can be considered as anything smaller than 100 µm in diameter. A considerable quantity of inhaled TSP can be removed quite effectively from the human body, although only particles smaller than 10 µm (eg, PM10) achieve any significant degree of lung penetration. For this reason TSP concentrations are not suited to monitoring in relation to health effects, but are more suited to the appraisal of dust nuisance. A detailed approach to monitoring dust nuisance can be found in the Good Practice Guide for Assessing and Managing the Environmental Effects of Dust Emissions (Ministry for the Environment, 2001) and Amenity Effects of PM10and TSP Concentrations in New Zealand (Ministry for the Environment, 2003b).
5.1.4 Dust deposition
Deposited dust can cause significant nuisance effects at locations close to sources such as unpaved roads, railways, site works, quarries and various industrial sites. Simple deposition gauges usually measure it, with results presented as deposited mass per unit area per time period (usually grams per square metre per 30 days). Monitoring methods used in New Zealand include the AS/NZS 3580.10.1:2003 Methods for sampling and analysis of ambient air – Determination of particulate matter – Deposited matter – Gravimetric method.
Deposition monitoring is a cheap and easy method for monitoring dust nuisance. However, the results can be difficult to interpret because of the poor time resolution of the method (typically 30 days). There is usually too much variation in weather conditions and other factors such as source emissions over this time to allow any sensible correlation with the monitoring results. The method is best used as a means of comparing overall nuisance dust levels in different locations.
There are several variants to the deposition gauge, including:
alternative dust measurement techniques (eg, dust deposition (soiling) meter and the use of sticky plates).
More information about methods for monitoring ambient dust is provided in the Ministry’s Good Practice Guide for Assessing and Managing the Environmental Effects of Dust Emissions (Ministry for the Environment, 2001). This report can be downloaded from the Ministry’s website: http://www.mfe.govt.nz/publications/air/dust-guide-sep01.pdf
5.2 Methods for particulate monitoring
The USEPA categorises particulate monitoring methodologies as either reference or equivalent methods. Reference methods are gravimetric (eg, direct measurement by weight), and equivalent methods are alternative methodologies that have been granted (following stringent inter-comparison studies) equivalency to the reference methods.
5.2.1 Beta attenuation monitor (BAM)
Particle mass density is measured using beta radiation attenuation. A pump draws ambient air through a paper-band filter and the reduction in intensity of beta radiation measured at the detector is proportional to the mass of particulate deposited on the filter. As the mass of PM10 increases, the beta count is reduced. The relationship between the decrease in count and particulate mass is computed according to a known equation (the Beer–Lambert law, as for CO and O3).
Monitors can be set to operate for 15-minute to 24-hour cycles, with intermediate averages if selected. Some samplers will automatically take a measurement and feed the filter tape if the filter loading reaches a predetermined level.
This method allows for unattended operation over extended periods of time, with a time resolution of about 0.5 to 2 hours. The response of the instrument depends on the beta absorption coefficient of the particulate, and this can vary with chemical composition. As with the tapered element oscillating microbalance (TEOM, see section 5.2.2), the requirement to heat the air inlet also results in the loss of some semi-volatiles. However, because the collected material does not remain on the filter for long periods before being measured, the volatile loss is not normally as significant as for the TEOM.
Under-reporting of approximately 23–24 per cent at 50 µg/m3 has been recorded by BAM monitoring at Taupo and Tokoroa when compared with gravimetric results.1 Most comparisons were under-reporting by around 6 per cent when compared with gravimetric methods at 50 µg/m3. The variation in semi-volatile components, both across an airshed and even within an airshed on a seasonal basis, means that correcting data to gravimetric equivalent is not presently recommended.
Recommendation 10: Operation of a beta attenuation monitor
- Enclosure temperatures should be maintained at 25°C ± 3°C to avoid moisture collecting on filter paper.
- Regular maintenance in accordance with the operation manual is critical. Irregular and/or inadequate maintenance can result in up to 20 per cent variation.
The sample heater should be switched off for at least one hour before calibration.
The condition of the radioactive source should be checked twice a year (R2 count). It may be necessary to adjust after two to three years due to decay in the radioactive source.
Inlet temperature should be set to 40°C.
Equipment should be maintained in accordance with the operation manual (refer also to AS/NZS 3580.9.11:2008).
Data correction to gravimetric equivalent is not generally recommended.
Where possible, humidity should be logged along with appropriate meteorological data.
5.2.2 Tapered element oscillating microbalance (TEOM)
The TEOM is a proprietary system that determines particulate concentration by continuously weighing particles deposited on a filter. The filter is attached to a hollow tapered element, which vibrates at its natural frequency of oscillation, ‘f’. The frequency changes by an amount proportional to the mass deposited, ‘m’, as particles progressively collect on the filter:
m = k/f2
where k is a constant determined during calibration of the TEOM. The mass measurement system is also known as the mass transducer. The flow rate of air through the system is controlled using thermal mass flow controllers and automatically measured to determine mass concentration. The TEOM analyser consists of a sample inlet head attached to the sensor unit, a control unit containing the mass flow controllers and system software, and a carbon vane pump. The total flow of 16.67 litres per minute through the sampling head is divided using a flow splitter to give 3 litres per minute through the filter cartridge and an auxiliary flow of 13.67 litres per minute. This provides direct mass measurement (USEPA, 2002) and, in conjunction with measured flow rate, concentration can be calculated, providing 10-minute averages.
The mass concentration, oscillation frequency, filter loading, flow rates, temperature and other diagnostic information can be displayed on the controller’s liquid crystal display (LCD) screen. In addition, mass concentration and filter loading, and possibly other parameters, are output to the datalogger as analogue voltages or through the RS232 interface.
The system allows for continuous unattended monitoring over extended periods of time, and has a high level of precision. It is classified as an equivalent method for PM10 monitoring (but not PM2.5) when operated in accordance with 40 CFR Part 50, Appendix J. Alternatively, TEOMs can be operated in accordance with AS/NZS 3580.9.6:2003 (Methods for sampling and analysis of ambient air – Determination of suspended particulate matter – PM10high volume sampler with size selective inlet – Gravimetric method).
Comparative studies of the TEOM against gravimetric methods have shown that the heated inlet, designed to remove unwanted water vapour from the sample, inadvertently causes the loss of volatile particulates (such as ammonium nitrate), both in the sample train and on the filter itself. This effect can be reduced by operating the sampler at 40°C instead of 50°C. This is standard recommended practice in New Zealand.
For the measurement of PM10 a high-volume sampler should be co-located with the TEOM for one year to establish an adjustment factor for volatile loss. The difference between TEOM and gravimetric methods varies with season and location. Greater differences are likely to occur where wood smoke comprises a large portion of the PM10. Wood smoke contains a significant fraction of low molecular weight volatile organic compounds, which are volatilised by the TEOM’s heated inlet.
It is not necessary to determine an adjustment factor if the TEOM is fitted with an FDMS (filter dynamics measurement system). In simple terms, the FDMS alternates between sampling aerosol-laden and aerosol-purged air. Any decrease in filter mass as a result of being purged of aerosols is added back to the unpurged mass to take account of the volatilised component.
It is strongly recommended that TEOMs be fitted with FDMS when used for compliance monitoring. FDMS can be retrofitted to Series 1400a analysers. Data should be annotated to show that it is from a TEOM-FDMS monitor. Alternatively, the TEOM could be used without the FDMS by using a correction factor. This factor is determined by co-locating the TEOM with a gravimetric monitor for at least one year.
Recommendation 11: Operation of a tapered element oscillating microbalance (TEOM)
It is recommended that TEOMs be fitted with a filter dynamics measurement system (FDMS) when monitoring for national standards.
Alternatively, TEOMs can be used without the FDMS by using a correction factor that is determined by co-locating the TEOM with a gravimetric monitor for at least one year.
Partisols are available as sequential gravimetric samplers with size-selective inlets that can monitor different particle size fractions (PM10 and PM2.5). Some partisols have reference method designation (USGPO, 1998a and 1998b) making them suitable for compliance monitoring. Other samplers may be used when operated in accordance with US 40 CFR Part 50, Appendix J.
Partisols operate with either a hub (and satellite) or filter cassette system. Hub systems incorporate two or more filters and can switch sample flow to a new filter to allow daily sampling. The filter cassette system is capable of loading up to 16 filters, which are changed automatically at a predetermined time (AEA Technology plc, 2003). The conditioning of the filters and calculation of results is the same as for a hi-vol sampler (see below). This system can be left unattended up to two weeks and its progress can be monitored remotely by telemetry.
Some samplers are also able to store meteorological parameters such as average ambient temperature, pressure and relative humidity.
5.2.4 High-volume (hi-vol) gravimetric method
A hi-vol sampler draws ambient air at a constant flow rate (66–68 m3/hour) onto a glass-fibre filter, which has been pre-weighed after being conditioned at constant relative humidity and temperature for at least 24 hours. The filter is exposed for a 24-hour period and then reweighed after being conditioned again under the same conditions of relative humidity and temperature. The total volume of air sampled is determined from the flow rate and the sampling time. The mass concentration is calculated as the mass of the sample collected, divided by the volume of air sampled.
Hi-vols either have a mass flow controller or volumetric flow control, which maintains a constant flow rate as the filter loading increases during sampling. Care must be taken to use appropriate filters that will not become overburdened during the sampling period. For the purposes of PM10 monitoring a size-selective inlet is required.
Under Schedule 1 of the NES for air quality, a 24-hour mean is calculated every 24 hours at midnight for the preceding 24 hours. While hi-vols can be used for compliance monitoring when operated in accordance with US 40 CFR Part 50, Appendix J, the requirement to manually change filters at midnight makes this impractical.
Recommendation 12: The importance of filter conditioning
Manual methods for particulate monitoring are all based on weighing material collected on a filter. It is therefore important to recognise that the pre- and post-conditioning of the filter and the filter weighing techniques are just as important as the selection and use of the sampling equipment.
Most filters will absorb moisture from the atmosphere, so filter weight will vary in accordance with the surrounding humidity. Particulate matter collected on the filters will also behave in the same way. It is therefore essential that the filters be carefully conditioned and weighed under conditions of constant temperature and humidity, both before and after sampling.
Detailed procedures for filter handling, conditioning and weighing are given in the relevant standard method specifications. For example, the USEPA recommends that filters be conditioned for at least 24 hours at a humidity between 20 to 45%, ± 5%, and a temperature of 15 to 30°C, ± 3°C (40 CFR Part 50, Appendix J).
Appendix C lists various types of sampling filters and their applications.
5.2.5 Light-scattering instruments
Light-scattering instruments have been available for many years, but mainly for use in monitoring workplace dust exposures. Over the past few years some of these instruments have been adapted for ambient monitoring, with variable degrees of success. The ‘workplace’ units are relatively cheap and portable, and give a direct readout of particle concentrations. Their measurement precision and sensitivity, however, are not appropriate for compliance monitoring (see section 4.1). As such, they are more suitable for research or low-level survey work.
The main limitation with light-scattering instruments is that the instrument response depends on both the size distribution and the number of particles, rather than the total mass of airborne particulate. This can be overcome to some extent by carrying out periodic calibrations using manual filter sampling.
Ultimately, there is no direct relationship between light scattering and mass, and the method is not suitable for compliance monitoring. Some light-scattering instruments also give an indication of particle size distributions, which may be of value in specific investigations.
5.2.6 Low-volume (low-vol) gravimetric method
There are a number of low-vol methodologies available. These are designed with specific flow rates sufficient for the size-selective inlet to collect the particle size fraction being monitored. Low-vol systems include:
minivol (with a flow rate of 5 litres/minute)
microvol (with a flow rate of 3 litres/minute).
Minivols and microvols do not comply with US 40 CFR Part 50, Appendix J, and are not suitable for compliance monitoring. However, they are useful screening tools, being small, portable, battery powered and easy to deploy. It should be noted that special care needs to be taken when handling the particulate filters for minivol and microvol samplers. The weighing of these filters requires a five decimal place microbalance, and it may also be necessary to install vibration isolators and take special anti-static precautions. At low concentrations (below 15 µg/m3), it is not unusual for minivol samplers to be ± 50 per cent of the reported concentration (Northland Regional Council, unpublished).
Appendix D summarises the widely used particulate monitoring instruments in New Zealand.
1 Minutes from the Beta Attenuation Monitor Workshop, Hawke’s Bay Regional Council, Napier, 17 March 2008.