Suspended particulate matter (SPM) in ambient air remains a critical focus in environmental monitoring because inhalable particles such as PM10 and PM2.5 can carry toxic elements and contribute to adverse health outcomes. Reliable elemental analysis of air filter samples is therefore essential for laboratories supporting air quality studies, regulatory compliance, and source apportionment work. US EPA method IO-3.3 has long provided a framework for the elemental characterization of particulate matter collected on filters using energy-dispersive X-ray fluorescence spectroscopy (EDXRF).
Air filter analysis presents a demanding combination of requirements: laboratories must quantify a broad range of elements, achieve low detection limits, maintain consistency across samples, and preserve sample integrity for repeat analysis when needed. These demands become even more important when working with trace-level analytes distributed across thin-film particulate deposits. In practice, analytical performance depends not only on detector sensitivity, but also on optimized excitation conditions, robust spectral deconvolution, and calibration against suitable reference materials.
EDXRF is widely used for elemental analysis of air filters because it is a non-destructive technique capable of measuring multiple elements across the periodic table in a single workflow. In the context of EPA IO-3.3, the method is designed for the analysis of particulate matter deposited on filters and supports the determination of dozens of elements relevant to environmental and toxicological assessment. For laboratories evaluating atmospheric aerosols, filter-based elemental data can help identify contributions from crustal dust, industrial emissions, combustion sources, and other pollution pathways.
The gated asset explores the use of the Revontium bench-top EDXRF spectrometer for this application. The system configuration described includes an X-ray tube, primary and secondary filters, four high-resolution silicon drift detectors, a spinner, and an automatic sample changer for batch analysis. Measurements were performed under vacuum conditions, supporting improved sensitivity for light elements as well as broader multi-element coverage.
The full asset outlines a calibration strategy developed from 96 single-element air filter standards. To quantify 45 elements efficiently, five optimized measurement conditions were used, each tailored to a specific elemental group. These conditions varied in tube settings, filters, and counting time in order to balance excitation efficiency, spectral quality, and throughput. The reported total measurement time per sample was 21 minutes, with flexibility to adjust individual conditions when application requirements call for different sensitivity or productivity targets.
The methodology also highlights the importance of spectral processing. A deconvolution algorithm was used to resolve overlapping peaks and determine net intensities, while selected elements such as Pb, Ag, and Cd were evaluated using a region-of-interest approach. The source also notes a practical consideration for aluminum analysis in thin-film air filter samples: when aluminum is included, Teflon foil coverage is required to minimize scatter, but this can increase detection limits for Na, Mg, and Si.
According to the source material, the study compared measured values with certified concentrations for two reference air filter standards, UCD ME-212 and UCD ME-213, using three consecutive measurements per sample. The reported outcome was good agreement between certified and measured concentrations across the certified elements presented. The study also reports that calculated lower limits of detection (LLD), based on replicate blank measurements and expressed at the 1-sigma level, were below the EPA IO-3.3 limits for the analyzed application at a total measurement time of 21 minutes per sample.
These findings are relevant for laboratories that need confidence in both accuracy and sensitivity when evaluating particulate matter samples. They also suggest that non-destructive EDXRF workflows can provide practical advantages for repeat measurements and long-term preservation of valuable air filter standards and samples.
For environmental, industrial, and research laboratories, the ability to perform non-destructive elemental analysis on air filters can streamline routine monitoring while preserving flexibility for remeasurement or follow-up studies. Bench-top EDXRF platforms are particularly relevant when organizations need a combination of multi-element coverage, laboratory efficiency, and regulatory-method alignment without relying on destructive sample preparation. This makes the approach meaningful for applications in ambient air monitoring, particulate characterization, environmental compliance, and method development.
Register to access the complete asset for a deeper review of the analytical workflow, optimized measurement conditions, calibration design, reference material results, and detection-limit performance. The full document provides the technical detail needed to evaluate the suitability of this EDXRF approach for air filter analysis under EPA IO-3.3 requirements.
Complete the registration form to read the full application note and explore the full methodology and results.
The presence of toxic pollutants in the air has been a subject of research for many years in countries around the world. In the United States, air quality standards are governed by the ‘Clean Air Act’ and administered by the US Environmental Protection Agency (EPA). One of the key areas of concern for the US EPA is the Suspended Particulate Matter (SPM) of air. Research on the health effects of SPM in ambient air has focused increasingly on particles that can be inhaled into the respiratory system, i.e. particles of aerodynamic diameter of < 10 µm. These particles are referred to as PM10 (2.5 – 10 µm) and PM2.5 (<2.5 µm). In addition to chemical toxicity, these particles are a significant threat to health.
The elemental analysis of the SPM on air filters is traditionally performed by energy dispersive X-ray fluorescence spectroscopy (EDXRF) using EPA method IO-3.3, which outlines the protocol of the analysis of 45 elements.
This application note demonstrates the capability of Revontium, a high-performance bench top EDXRF spectrometer, as an analytical tool for the analysis of air filters according to the US EPA method IO-3.3.
Measurements were performed using a Revontium™ Heavy Element edition energy-dispersive XRF spectrometer, equipped with an X-ray tube (50 W, 5 mA, 60 kV, Ag anode), primary and secondary filters, four high-resolution silicon drift detectors, a spinner, and a sample changer for automatic batch measurements. All measurements were performed in vacuum by using an oil-free (dry) vacuum pump.
In total, 96 single-element air filter standards from Micromatter Co and UC Davies were used to set up the calibrations.
Five measurement conditions were used to quantify 45 elements in the air filters, each one optimizing the excitation of a specific group of elements (Table 1). The total measurement time per sample was only 21 minutes. The measurement time for each condition can be optimized according to specific needs.
The SuperQ v7 software features a powerful deconvolution algorithm, which analyzes the sample spectrum and determines the net intensities of element peaks, even when they overlap one another. This is essential to perform accurate trace element analysis. For the elements Pb, Ag and Cd, a region of interest (ROI) was used to determine the intensities instead of the traditional deconvolution method.
In this study, Aluminum was not analyzed because of the focus on the other light elements Na, Mg and Si in the air filters. When aluminum is also quantified in these thin film samples, each air filter must be covered with a Teflon foil to minimize the scatter into the XRF spectrum. Consequently, the detection limits (LLD) for Na, Mg and Si will increase by a factor of 2 when using this method of sample introduction.
| Elements | kV | mA | Medium | Filter | Measuring time (min) |
|---|---|---|---|---|---|
| Na, Mg, Si, P, S, Cl | 9 | 5.000 | vacuum | Ti | 5 |
| K, Ca, Sc, Ti, V, Te, I, Cs, Ba, La | 12 | 4.166 | vacuum | Al-thin | 2 |
| Cr, Mn, Fe, Co | 20 | 2.500 | vacuum | Al-thick | 2 |
| Ni, Cu, Zn, Ga, Ge, As, Se, Br, Rb, Sr, Y, Zr, Mo, W, Au, Hg, Pb, U | 60 | 0.833 | vacuum | Ag | 10 |
| Rh, Pd, Ag, Cd, In, Sn, Sb | 60 | 0.833 | vacuum | Cu | 2 |
Table 1. Five optimized measurement conditions for quantifying 45 elements in air filter samples
To evaluate the accuracy of the method, two reference air filter standards UCD ME-212 and ME-213 were analyzed. Each sample was measured 3 times consecutively. For each element in the sample, the average concentration and standard deviation (SD, 1 sigma) of the measurements were compared with the certified value reported on the certificate (see Table 2). Not all 45 elements are reported in Table 2 since not all elements were certified in the reference standards.
The results show a good agreement between the certified and measured concentrations of both air filter standards.
| UCD ME-212 | UCD ME-213 | |||
|---|---|---|---|---|
| Certified conc. ± SD (ng/cm2) | Measured conc. ± SD (ng/cm2) | Certified conc. ± SD (ng/cm2) | Measured conc. ± SD (ng/cm2) | |
| Na | 1900 ± 380 | 1840 ± 12 | 2700 ± 540 | 2447 ± 14 |
| Mg | < 1000 | 588 ± 1 | < 1000 | 721 ± 1 |
| Si | 1200 ± 120 | 1082 ± 1 | 1700 ± 170 | 1390 ± 1 |
| P | < 100 | 53 ± 0.4 | < 100 | 69.1 ± 0.1 |
| S | 3800 ± 380 | 3379 ± 7 | 5300 ± 530 | 4560 ± 3 |
| Cl | < 100 | 0 ± 0.1 | < 100 | 0 ± 0.1 |
| K | 640 ± 64 | 618 ± 1 | 880 ± 88 | 860 ± 2 |
| Ca | 580 ± 58 | 569 ± 1 | 820 ± 82 | 789 ± 1 |
| Sc | < 100 | 2.5 ± 0.4 | < 100 | 1.1 ± 0.2 |
| Ti | 43 ± 8.6 | 57.2 ± 0.2 | 61 ± 12.2 | 72.3 ± 0.3 |
| V | 96 ± 9.6 | 118.2 ± 0.3 | 140 ± 14 | 163.6 ± 0.3 |
| Cr | 39 ± 3.9 | 41.5 ± 0.1 | 55 ± 5.5 | 51.1 ± 0.1 |
| Mn | 50 ± 5 | 53 ± 0.4 | 70 ± 7 | 74.1 ± 0.1 |
| Fe | 560 ± 56 | 622 ± 1 | 790 ± 79 | 859 ± 1 |
| Co | 70 ± 7 | 70 ± 0.1 | 97 ± 9.7 | 98.7 ± 0.5 |
| Ni | 29 ± 2.9 | 26 ± 1 | 40 ± 4 | 35.6 ± 0.5 |
| Cu | 90 ± 9 | 90 ± 1 | 120 ± 12 | 130 ± 0.7 |
| Zn | 39 ± 4 | 38 ± 1 | 55 ± 5.5 | 55.9 ± 0.3 |
| As | 350 ± 70 | 380 ± 1 | 490 ± 98 | 534.9 ± 0.4 |
| Se | 42 ± 4.2 | 38.2 ± 0.6 | 59 ± 5.9 | 54.9 ± 0.2 |
| Rb | 70 ± 7 | 68.2 ± 0.5 | 97 ± 9.7 | 96.1 ± 0.5 |
| Sr | 70 ± 7 | 74.3 ± 0.1 | 97 ± 9.7 | 105.2 ± 0.6 |
| Zr | 43 ± 8.6 | 40.5 ± 0.6 | 61 ± 12.2 | 58.4 ± 0.2 |
| Mo | 43 ± 8.6 | 39.2 ± 0.4 | 61 ± 12.2 | 57.3 ± 0.7 |
| Cd | 120 ± 12 | 99 ± 18 | 180 ± 36 | 170 ± 9 |
| Ba | 170 ± 130 | 136 ± 1 | 250 ± 50 | 211 ± 2 |
| Pb | 96 ± 9.6 | 91.6 ± 0.3 | 140 ± 14 | 132.4 ± 0.8 |
Table 2. Accuracy results of UCD ME-212 and ME-213 air filter standards against it certified concentrations.
Detection limits are an important measure of an instrument’s performance. The detection limits for this application were calculated from 20 replicate measurements of 2 Teflon blank samples and a cellulose blank sample. As specified in method IO-3.3, the detection limits are calculated based on 1-sigma standard deviation and are shown in Figure 1. As a comparison, the LLD limits reported in the EPA method are also shown in the overview. With a measurement time of only 21 minutes per sample, all LLD values are smaller than the LLD limits set by the EPA method (see Fig 1).
![[AN260618-figure1.png] AN260618-figure1.png](https://dam.malvernpanalytical.com/dfa02656-0be3-476f-afa0-b46d0074f1e2/AN260618-figure1_Original%20file.png)
The Revontium can analyze particulate matter on air filters according to US EPA IO-3.3 with a high degree of accuracy for a wide range of elements across the periodic table.
In contrast to WD-XRF, the non-destructive nature of the ED-XRF method makes it possible to increase measurement times in case an even lower detection limit is needed. Furthermore, samples can be measured repeatedly without damage, ensuring the longevity of air filter standards.