Abstract
The magnetic susceptibility (κ) of particulate matter (PM) is a useful tool in estimation concentration of iron-rich particles and provides useful information on the emission sources and pathways of spread of PM in the atmosphere. However, there is currently no established protocol for measuring the magnetic susceptibility of PM collected on filters used in standard monitoring of PM concentration. This paper presents a step-by-step process for collecting PM on filters in automatic samplers and measuring their κ. The procedure outlines requirements for data quality, measurement uncertainty, exposure time and conditions, and the amount of material collected on the filters. The study analyzed a 2-year dataset of magnetic susceptibility measurements by MFK-1 kappabridge (Agico, Czech Republic) for PM10 and PM2.5 collected at two locations, Warsaw and Cracow, in Poland using low-volume PM samplers. By strictly following the procedure for conditioning filters, measuring magnetic susceptibility and mass of PM, the study found that it is possible to obtain repeatable data with good measurement accuracy and acceptable errors. This makes magnetic susceptibility an additional reliable parameter for tracking of emission sources of iron-rich particles. Successful implementation of this magnetic method as a standard procedure for monitoring PM in addition to the PM mass collected on filters could be used to analyze sources of emission of Fe-particles and their contribution to the PM mass, especially in urban and industrial environments.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
In recent times, there has been an increase in social and scientific interest in improving air quality in urban areas due to the health risks posed by pollutants like particulate matter (PM), sulfur dioxide, ozone, benzene, nitrogen oxides, and carbon monoxide (Zheng et al. 2015, Thurston et al. 2016; Weichenthal et al. 2017; Wu et al. 2018; Harrison et al. 2017; Strak et al. 2017; Čabanová et al. 2019; Pope et al. 2020; Rachwał et al. 2020; Hammond et al. 2022; Adamiec et al. 2022; Loaiza-Ceballos et al. 2022). Of these pollutants, airborne particles with aerodynamic diameters less than 10 µm (PM10), less than 2.5 µm (PM2.5), and smaller are particularly harmful to human health as they can easily penetrate deep into the lungs and circulatory system, leading to serious respiratory and cardiovascular diseases, cancer, and even mortality (Maher et al. 2016; Thurston et al. 2016; Miller et al. 2017; Weichenthal et al. 2017; Bové et al. 2019; Calderón-Garcidueñas et al. 2020; Nadali et al. 2022).
Urban aerosols can be made up of natural dust from events like resuspension of soil particles, long-range transport of natural dust from deserts, volcanoes, geothermal and seismic eruptions, as well as anthropogenic particles (Sagnotti et al. 2006). The latter poses a greater danger to human health because they contain potentially toxic metals emitted into the atmosphere by various urban and industrial activities such as industrial technological processes, fossil fuel combustion from heat and power plants, traffic emissions, and low stack emission, among others (Hwang et al. 2016; Bourliva et al. 2017; Sung et al. 2018; Abdulaziz et al. 2022; Górka-Kostrubiec et al. 2023). Public and the scientific interest requires the development of additional methods and techniques that can provide information on the origin of PM from distinct natural and anthropogenic sources. In the case of exceeding the threshold limits of PM10 and PM2.5 concentrations in the ambient air established by the European Parliament and of the Council ( Directive 1999/30/EC), the authorities are obliged to introduce actions and measures to counteract the increase in air pollution. However, PM concentration limits may be exceeded due to particle contributions from natural events which cause a relatively lower health risk for citizens than particles from a source associated with human activity. Therefore, the monitoring of PM concentrations alone appears to be insufficient for better understand the spreading mechanism of PM related to particles from anthropogenic and natural processes.
Research teams studying environmental magnetism have shown promising results in the study of environmental pollution. Magnetometry (magnetic techniques or methods), which is widely used for rock-magnetic studies, is an inexpensive, fast, and precise technique for assessing and monitoring pollution in different environmental systems (Kapper et al. 2020), including soil (e.g., ** such a procedure would allow for a harmonized assessment of air pollution through magnetic susceptibility, facilitating identification of changing patterns or sources of pollutant emissions. Ultimately, successful implementation of magnetic susceptibility as a standard parameter for monitoring of PM could lead to establish the strategy and policy to reduce pollutant emissions from various sources in urban and industrial environments.
The primary objective of this research was to develop a standardized protocol for measuring magnetic susceptibility as a reliable parameter in tracking of emission sources of air pollutants. The research aimed to achieve this goal by (i) develo** and refining the protocol for measuring the magnetic susceptibility of PM collected on filters, (ii) assessing the accuracy of measurements for magnetic susceptibility of PM, and (iii) indicating the sampling parameters—exposure time for collecting the PM on the filters and the error of PM mass measurement that affect the accuracy of determination of the mass-specific magnetic susceptibility. The study also identified the data quality requirements and especially the measurement uncertainties for magnetic susceptibility.
Magnetic susceptibility
Several magnetic techniques are employed to characterize minerals based on their magnetic properties, with magnetic susceptibility being a common method. Volume magnetic susceptibility (κ) is defined as the ratio of the vector of applied magnetic field \(\overline{H}\) (in A/m) and the vector of induced magnetization \(\overline{M}\) (in A/m) in the material \(\overline{M}=\kappa \overline{H}\), where κ is the second-rank tensor. In environmental studies, the anisotropic effects are neglected and the mean value of κ is used (Tauxe 1998, 2002). The unit of κ is dimensionless. Mass-specific magnetic susceptibility (χ) is another parameter that is commonly used. It is defined as the κ divided by the density (ρ) of the material (\(\chi =\kappa /\rho\)), and its unit is cubic meters per kilogram (Thompson and Oldfield 1986).
Diamagnetic materials such as quartz, calcite, and silicon exhibit relatively low negative values of χ, while paramagnetic materials such as aluminum, sodium, and oxygen have values of that are strongly temperature-dependent and linearly dependent on the intensity of the applied magnetic field. Ferromagnetic materials, such as iron, nickel, and cobalt, achieve saturation easily (i.e., alignment of all atomic moments), even at relatively low magnetic fields. They exhibit a hysteresis effect, which is related to the nonlinear relationship between magnetic field and magnetization.
To minimize the contribution of the paramagnetic fraction to magnetic susceptibility and obtain mainly ferromagnetic components that saturate at relatively low fields, it is standard practice to measure the magnetic susceptibility of environmental samples at low fields, typically in the range of 200–700 A/m (Evans and Heller 2003; Thompson and Oldfield 1986).
Magnetic susceptibility is dependent on the concentration of magnetic particles, their mineralogy, and grain size distribution. In urban and industrial dust, the mineralogy of the magnetic fraction is dominated by magnetite or maghemite, which exhibit strong magnetic properties, while a minor contribution of weakly magnetic hematite is also observed. Magnetic susceptibility is a fast and direct method for quantifying the content of magnetic particles in such samples. Sagnotti et al. (2006) developed an experimental protocol for the use of magnetic properties as reliable proxies for the identification of the natural and anthropogenic sources of PM10. They showed that a magnetic fingerprint (mineralogy and domain state of ferrimagnetic carriers) of fine atmospheric particles may be associated to distinct natural and anthropogenic sources. Additionally, magnetic susceptibility can be useful parameter in monitoring heavy metal contamination in PM, as anthropogenic particles are more efficient in absorbing and transferring heavy metals due to their strongly defected crystal structure (** the box used for transport by the manufacturer and placing them in a desiccator containing silica gel as the moisture-absorbing material. The humidity and temperature in the desiccator were monitored using a digital meter. The filters were left in the desiccator for approximately 1–2 days, and the humidity of the air inside the desiccator was controlled to ensure that the filters were well-conditioned, as per the regulations established by EN 12341:2014 for measuring the mass of PM collected on filters. Once the humidity reached between 40 and 50%, the filters were assumed to be well-conditioned, and their mass was measured. Filters removed from the desiccator for mass measurement were assumed to absorb any neglected moisture. After measuring the mass, each filter was assigned a unique ID number and placed into a sampling cartridge (Fig. 1a), which was then arranged on top of each other in a sampling container (Fig. 1b). Each container can hold up to eighteen cartridges with clean filters.
After exposure in the PM sampler, the container with the filters was brought to the laboratory and placed in a desiccator for conditioning, following the same procedure as for the clean filters. Well-conditioned filters were removed from the sampling cartridges and reweighed using the established procedure for measuring filter mass (see the “Procedure for measuring the mass of PM collected on the filters” section). To prevent loss of material captured by the filter, each filter was folded in half with the dusty side inwards and placed in a paper envelope of appropriate dimensions (57 × 50 mm) with the ID number of each filter and its mass before and after exposure recorded on the envelope. The envelopes with filters were then placed into lockable cardboard boxes, with 100 filters in each box, and stored in a desiccator until the measurement of magnetic susceptibility was started. Tweezers were used when moving the filters to the weighing pan, cartridges, and envelopes. The sampling cartridge was wiped with cellulose swabs moistened with alcohol and high-purity isopropyl alcohol whenever a clean filter was placed in it.
Procedure for collecting PM on filters using low-flow samplers
We used an automatic sampling system with double magazine PNS 18 T-3.1-DM (PNS TDM) for monitoring PM10 and PM2.5 (Atmoservice, Poznań Poland and Comde-Derenda GmbH, Stahnsdorf, Germany) to collect PM on the filters (Fig. 1c). This sampling system is a reference for monitoring suspended PM in accordance with German Air Quality Standards (TA Luft 2002; Comde-Derenda 2014). The PNS TDM device consists of a low volume sampler (LVS 3.1), an automatic filter changer with a suction tube, and a head for collecting PM10 or PM2.5 in a stainless steel cabinet. The PM fractions are collected on filters in accordance with the EN 12341:2014 standard (Comde-Derenda 2014). The sampling process involves drawing in ambient air by a rotary-vane vacuum pump and fractionating the particles according to their aerodynamic diameter in the head. The air containing the desired PM fraction then passes through a filter on which the particles are captured. The automatic filter changer with a Geneva drive and two filter containers allows for sequential sampling of a series of 18 filters. Two cylindrical containers are used for collecting one series of samples. The first container (left in Fig. 1b) contains sampling cartridges with clean filters arranged on top of each other, while the second container (right in Fig. 1b) stores the filters after exposure. During the filter collection process, the lowest cartridge with a filter from the first container is transferred to the sampling position, and the air with the desired PM fraction is passed through the filter for a specified time. Then, the cartridge with the filter after exposure is transferred to the second container. In each container, filter no. 18 is not exposed to dust collection but is used as a comparative filter to determine whether there is dust deposition on the walls of the container. The containers have tight covers to prevent the cartridges with filters from falling out and to prevent contamination of filters with foreign particles during the sampling of PM and transport of containers to and from the laboratory (Comde-Derenda 2014).
The control unit allows for various parameters to be set, such as the volumetric flow rate, sampling periods, time of day and data, and the number of cartridges with filters. The EN 12341:2014 standard specifies a volume flow rate of 2.3 m3/h with an accuracy better than 1% deviation from the set-point value, which is measured at an orifice plate located between the filter and the pump (Comde-Derenda 2014). To ensure that the collected PM sample is sufficient for magnetic studies, a sampling period of 72 h per filter was set for monitoring magnetic susceptibility in Warsaw and Cracow cities. The determination of the sampling time will be discussed in more detail in the “Result and discussion” section. The PNS TDM sampler is equipped with an external sensor that continuously registers temperature in a range from − 40 to + 80 °C with an accuracy of ± 0.5 °C and relative humidity in a range from 0 to 100% with an accuracy of ± 3%. The controller records the current number of filters in the magazine, datum and sampling period, volume flow rate, temperature, and relative humidity, which are stored on a Secure Digital memory card (Comde-Derenda 2014).
Procedure for measuring the magnetic susceptibility of PM collected on filters
The multifunctional kappabridge MFK1-FA (AGICO, Brno, Czech Republic) was utilized to measure the magnetic susceptibility of PM collected on filters. This laboratory instrument is highly sensitive and is commonly used to measure magnetic susceptibility in weak magnetic fields. The magnetic field strength less than 500 A/m (~ 0.625 mT) is used to minimize the effect that magnetic susceptibility of ferromagnetic minerals sensu lato does not obey Rayleigh Law (Néel 1955; Hrouda et al. 2006). The magnetic susceptibility of PM was measured at a frequency of 976 Hz (factory set value for MFK1-FA unit) and a magnetic field strength of 200 A/m (~ 0.25 mT), which can be adjusted by the operator. The sensitivity of the magnetic susceptibility measurement is 2 × 10−8 SI, according to the technical specifications of the MFK1-FA kappabridge (Agico 2009). The field homogeneity at 976 Hz is ~ 0.5%. The measurement of the magnetic susceptibility of PM was conducted in accordance with the standard procedure recommended by the manufacturer of the device (Agico 2009). Prior to commencing measurements, the MFK1-FA kappabridge was stabilized for approximately 1 h to achieve temperature stabilization, which is necessary for the correct operation of the device at maximum sensitivity. The SAFYR7 software (Agico 2022) was used to control the functions of the MFK1-FA kappabridge, acquire data, and calculate the results of individual measurements. The software also enables calibration of the kappabridge, automatic start of measurement, and control of its course. The calibration of the MFK1-FA kappabridge is performed using a magnetic susceptibility standard, which is measured once with an absolute accuracy of ± 3%. This value is controlled and automatically saved in the software parameters file. The magnetic susceptibility (\({\kappa }_{h}\)) of the holder is measured three times, and its mean value and standard deviation (SD) are calculated using the standard holder correction procedure adopted for the MFK1-FA kappabridge. If the values are too high, the operator is notified. The mean value of \({\kappa }_{h}\) is automatically saved in the software parameters file for further data processing, but its SD is not stored. The determination of the magnetic susceptibility of the holder is a required procedure before measuring each set of filters, which typically consists of approximately 18 filters from a single sampling container. In our case, the standard holder for manual measurements of magnetic susceptibility was adapted to filter measurements. The lower part used to hold the cubic samples was replaced with a thin sheet of plastic film. Other solutions can also be used to minimize the effect of the holder on the measurement.
The process of measuring magnetic susceptibility involves placing a folded in half filter with the dust side facing inwards into a holder and initiating the measurement option in SAFYR7 software. The software automatically conducts a sequence of ten measurements for each filter and saves the raw data with the filter ID in an output file. A measurement session for 18 filters takes approximately 2 h, and the output file for a batch of 18 filters contains ten records for each filter, including the filter ID, magnetic susceptibility of the holder, and magnetic susceptibility of the filter. The magnetic susceptibility of the filter can be further processed in a spreadsheet (Excel), where the mean values and SD of the magnetic susceptibility of each filter can be calculated from the ten measurements to assess the accuracy and quality of the measurements.
An important step in the measurement process involves determining the magnetic susceptibility of clean filters (\({\kappa }_{c}\)). This value, along with the \({\kappa }_{h}\), should be subtracted from the magnetic susceptibility of each filter after exposure. In this study, the \({\kappa }_{c}\) of each clean filter was not measured prior to exposure to PM. Instead, due to the similar magnetic susceptibility values of clean filters and time constraints, the following procedure was implemented: for each new set of 100 filters (as standard, 100 filters are packed in one box by the manufacturer), five filters were randomly selected and the magnetic susceptibility was measured 10 times for each filter. During this process, the \({\kappa }_{h}\) was automatically subtracted, and the resulting values for clean filters were stored in the output file. The averaged value from the 10 measurements performed for the five filters was used as the \({\kappa }_{c}\) of the clean filter for further calculations. The SD of the mean value (\(\Delta {\kappa }_{c}\)) was considered the measurement error of the \({\kappa }_{c}\).
Procedure for measuring the mass of PM collected on the filters
The MYA 4Y.F PLUS microbalance (Radwag, Radom, Poland) was utilized to determine the mass of clean filters and filters after exposure. The microbalance is equipped with a specially designed pan dedicated to measuring the mass of filters with a diameter of up to 50 mm. The accuracy of the MYA 4Y.F PLUS microbalance is 1 μg after temperature stabilization has been achieved (Radwag 2023). Prior to measuring each series of filters, the microbalance was calibrated using a professional 100 mg mass standard, class E2, following the Radwag procedure (Radwag 2023). The process for measuring the mass of clean filters and filters after exposure was identical. During a single measurement session, a set of filters stored in one sampling container was weighed. The mass of each filter was measured twice, following the procedure outlined in the “Procedure for the conditioning, preparation and storage of filters used for monitoring of magnetic susceptibility of PM” section. In general, to ensure sufficient accuracy in determining the mass of both clean and exposed filters, it is necessary to select an appropriate, very sensitive balance.
PM mass and magnetic susceptibility and error analysis
PM mass and its error
The output file contains the raw data for filter mass. To determine the mass of PM, the mean values of mass for each clean filter and filter after exposure were calculated from two measurements and labeled as \({\overline{m}}_{c}\) and \({\overline{m}}_{e}\), respectively. The mass of PM (\({m}_{PM}\)) was obtained by subtracting the \({\overline{m}}_{c}\) from the \({\overline{m}}_{e}\):
The error of a complex variable, which depends on many variables, can be calculated using the Taylor series expansion of the function while ignoring higher-order terms. If the complex variable (Z) is a function of many variables \(f=({x}_{1}\), \({x}_{2}\), \({x}_{3}\),…), and their values \({x}_{1}\), \({x}_{2}\), \({x}_{3}\),…, and their errors \({\Delta x}_{1}\), \({\Delta x}_{2}\), \({\Delta x}_{3}\),…. are known, then the maximum error of the complex variable (\(\Delta Z\)) can be calculated using the following formula:
where \(\Delta Z\) is the maximum error of the complex variable Z, which shows how the errors of individual variables (\(\Delta {x}_{1}\), \(\Delta {x}_{2}\), \(\Delta {x}_{3}\), …) affect the final error of the complex variable (Taylor 1982).
The error of the PM mass was calculated using the following formula:
which is an extension into the Taylor series of the function describing the PM mass (Eq. 1) and depends only on the mass of the clean filter and the mass of the filter after exposure. The errors of \({m}_{c}\) and \({m}_{e}\) were determined as root mean square errors expressed by multiplying the SD and the Student’s t-factor of 1.84 at the confidence level of 0.683 for the number of measurements of n = 2.
Magnetic susceptibility of PM and its error
The raw data of magnetic susceptibility was processed as follows. For each filter, the mean values of magnetic susceptibility (\(\overline{\kappa }\)) and the SD (\(\Delta \overline{\kappa }\)) were calculated from ten individual measurements. According to the measurement procedure (see the “Procedure for measuring the magnetic susceptibility of PM collected on filters” section), the final value of the magnetic susceptibility of PM (\({\kappa }_{PM}\)) was calculated by subtraction of \({\kappa }_{c}\) the clean filter and \({\kappa }_{h}\) the holder from the \(\overline{\kappa }\) as follows:
The maximum error of the \({\kappa }_{PM}\) determined from Eq. (4) in accordance with the Eq. (2) is described by the following formula:
where \(\Delta {\kappa }_{c}\) and \(\Delta {\kappa }_{h}\) are the error of measuring the magnetic susceptibility of the clean filter and the holder, respectively. According to our procedure, the κ of the clean filter and the holder are not determined for each filter but measured earlier therefore their measurement errors are independent and should be taken into account.
The mass-specific magnetic susceptibility of PM (\({\chi }_{{\text{PM}}}\)) is defined as the volume \({\kappa }_{{\text{PM}}}\) divided by the mass of PM collected on the filter and normalized by the calibration constant \({V}_{o}\), which is 1 × 10−5 m3 (10 cm3) for the MFK1-FA kappabridge:
The magnetic susceptibility normalized per unit volume of air (V) pumped through the filter during its exposure (\({\kappa }_{{\text{V}}}\)) is defined as follows:
The maximum errors of \({\Delta \chi }_{{\text{PM}}}\) and \({\Delta \kappa }_{{\text{V}}}\) determined from Eqs. (6) and (7) in accordance with the Eq. (2) are described by the following formulas:
and
where \(\Delta V\) is the error of \(V.\)
Data repository
The raw data, including magnetic susceptibility data and meteorological data, are stored in files in the repository of the Laboratory for Paleomagnetic and Environmental Studies of the Institute of Geophysics of the Polish Academy of Sciences (IG PAS) in Warsaw. This data is not publicly available, but it can be shared upon request. Processed data for each station location and year are stored in separate files created in the Excel program. These files contain columns with the following data: ID of filter, start and end date of filter exposure in the PM sampler, average mass of clean filter, average mass of exposed filter, calculated average mass of PM, average magnetic susceptibility of PM, average mass-specific magnetic susceptibility, average magnetic susceptibility normalized on the volume of pumped air, exposure time, and meteorological data such as average temperature and pressure. The processed data for the years 2016–2020 are stored in the CIBAL repository database, which can be accessed through the dataportal.igf.edu.pl web page in the Magnetic Susceptibility Monitoring folder.
Result and discussion
Examples of the sets of mass and magnetic susceptibility data for PM
The mass and magnetic susceptibility of PM10 and PM2.5 collected from three stations located in IG PAS, Gabrieli Zapolskiej street, Cracow and IG PAS, Księcia Janusza 64 street, Warsaw, respectively, were analyzed to test the procedure of PM collection and determination of mass and magnetic susceptibility. The data collected over a long period of time, from 72-h sampling periods, were analyzed and summarized in Table 1.
For the IGF_W station in Warsaw, the average κ for PM10 was 3.58 × 10−6 SI, with a median value of 2.93 × 10−6 SI. The minimum and maximum values of κ were 0.49 × 10−6 SI and 21.54 × 10−6 SI, respectively. The average and median mass of PM10 collected on the filter were 4.10 and 3.65 mg, respectively. The minimum and maximum values of PM mass were 0.87 and 18.77 mg, respectively.
In comparison, for the IGF_K station in Cracow, both the average and median values of κ for PM10 were higher than those for Warsaw, at 8.10 × 10−6 SI and 5.90 × 10−6 SI, respectively. The minimum and maximum values of κ were 1.25 × 10−6 SI and 33.80 × 10−6 SI, respectively. The average and median mass of PM10 collected on the filter were about 40% higher than those obtained for the IGF_W station in Warsaw (Table 1).
When comparing the magnetic susceptibility of PM2.5 with PM10 collected at the same location in Warsaw, the average and median values of κ for PM2.5 were 32% and 34% of the respective values obtained for PM10. The average and median mass of PM2.5 were respectively 28% and 30% lower than the mass of PM10 collected in the same location in Warsaw (Table 1).
To create the histograms in Fig. 2a–c, the maximum range of κ of PM10 (Fig. 2a) and PM2.5 (Fig. 2b) collected at the IGF_W station in Warsaw and PM10 (Fig. 2c) collected at the IGF_K station in Cracow was divided into equal intervals of 0. 5 × 10−6 SI. The histograms displayed in Fig. 2a–c reveal that the distribution of magnetic susceptibility values for PM10 and PM2.5 is not symmetrical. Most of the samples possess κ values above the average. For the IGF_W station in Warsaw (Fig. 2a), the set of the lowest values from 0.5 × 10 − 6 SI to 1.0 × 10 − 6 SI is above the sensitivity threshold of the MFK1-FA kappabridge. Moreover, the measurement error for the filter with the minimum κ value of 0.54 × 10 − 6 SI (AP-303 filter) did not exceed 11% (Table S1, Supplementary material). In contrast, for filters with an average κ value (AP-294, κ = 3.67 × 10 − 6 SI) and maximum κ value (AP-280, κ = 21.54 × 10 − 6 SI), the absolute errors were relatively low (~ 0.06 × 10 − 6 SI), with a percentage error of 1.5 and 0.3% for the AP-294 and AP-280, respectively (Table S1, Supplementary material). Similarly, for the IGF_K station in Cracow, the histogram also has a right skew. However, the lowest κ values are higher than 1.25 × 10 − 6 SI. For the filter with the minimum κ value (KP-71), the relative percentage error was about 5%, while for filters with average (KP-70) and maximum (KP-165) κ values, the errors did not exceed 1% (Table S1, Supplementary material). Regarding PM2.5 samples collected at the IGF_W station in Warsaw, most of the filters (n = 185) had κ values within the range of 0.5–1.5 × 10 − 6 SI. The relative percentage error for the DP-105 filter within this range was approximately 6%. However, for the DP-171 filter, which had the lowest κ value of 0.28 × 10 − 6 SI, the relative percentage error was around 21% (Table S1, Supplementary material). In general, the errors of measurement were relatively low, suggesting that the procedure of collecting and measuring PM and its magnetic susceptibility was reliable and robust.
Our analysis indicates that for Warsaw, even samples showing the lowest κ value, accounting for about 2% of all PM10 samples, had κ values falling within an acceptable sensitivity range of the MFK1-FA kappabridge, with a relative error of no more than 11%. Our analysis indicates that for Warsaw, even samples showing the lowest κ, accounting for about 2% of all PM10 samples, had the κ values falling in an acceptable range of sensitivity of the MFK1-FA kappabridge, with a relative error of no more than 11%. This suggests that an exposure time of 72 h for a single filter is sufficient to obtain satisfactory magnetic susceptibility values. For Cracow, the lowest magnetic susceptibility values of PM10 (in the range of 1.0–1.5 × 10−6 SI) were also measured in an acceptable range of sensitivity of the MFK1-FA kappabridge, with an error of less than 5.5%. The differences in the distribution of magnetic susceptibility of PM10 between Cracow and Warsaw (with Cracow showing a shift towards higher values of κ) may be due to an additional source of anthropogenic magnetic particles, such as low-stack emissions from private home furnaces (Górka-Kostrubiec and Dudzisz 2023). The enrichment of PM10 with strongly magnetic particles observed in Cracow results in higher values of magnetic susceptibility. In this case, it may be possible to shorten the exposure time of individual filters to obtain more accurate information on the level of contamination over a shorter period of time. In contrast, the distribution of magnetic susceptibility of PM2.5 in Warsaw showed a relatively large number of samples (~ 14%) with low values falling barely above the sensitivity range of the MFK1-FA kappabridge. For collecting PM with low concentrations of magnetic particles, exposure time for a single filter can be extended to obtain a more satisfactory error of κ. However, this approach provides information about the level of pollution averaged over a longer period of time.
Figure 3 displays histograms of the distribution of PM mass collected at the IGF_W station in Warsaw for PM10 (Fig. 3a) and PM2.5 (Fig. 3b) and PM10 collected at the IGF_K station in Cracow (Fig. 3c). The histograms were plotted by dividing the maximum range of the PM mass obtained for each collection into equal intervals of 1 mg. The histograms indicate that the distribution of PM mass is skewed to the right. For both PM10 collections, the majority of samples have a mass in the range of 2–5 mg, while for PM2.5, the mass range is 2–4 mg. In any case, the PM mass collected on the filters is determined with good accuracy and an acceptable error.
Time exposure for PM collected on filters
When determining the PM exposure time on a single filter, the environmental conditions in which the monitoring is carried out should be taken into account. In the case of monitoring the quality of air in an area with a relatively small number of anthropogenic sources of magnetic particles, even extending the exposure time will not allow us to measure the magnetic susceptibility of PM with sufficient sensitivity. Excessively extending the exposure time to collect a larger mass of PM can result in “filling up the filter.” The filter heavily saturated with dust particles will block the flow of air, and as a consequence, the constant airflow velocity required in the measurement procedure will not be maintained. On the other hand, in areas with a relatively large number of sources of magnetic particles, satisfactory values of the magnetic susceptibility of PM can be obtained even with a relatively short exposure time. In this case, the limitation may be the accuracy of determining the mass of PM, i.e., having a balance with a sufficiently good measurement sensitivity. It should be emphasized that the mass of PM is determined by the balance between the mass of the filter after exposure and the clean filter, whose mass is much greater than the mass of collected PM.
Magnetic susceptibility of empty holder and clean filters and their errors
The procedure for measuring magnetic susceptibility should aim to minimize the measurement error as much as possible. Therefore, several factors have to be considered that may affect the accuracy of determination of the magnetic susceptibility. The first factor that can affect the accuracy of a magnetic susceptibility meter is the noise from the surrounding space, which is due to temperature instability, spikes from power supply, the presence of computer displays, various iron-rich objects, etc. To check the noise coming from the surrounding space, a special procedure called sigma test in the SAFYR7 software, for noise measurements can be performed. The sigma test consists of measuring the magnetic susceptibility of empty coils with the up/down mechanism turned off and without a sample holder. During our experiments, the noise was lower than 1.2 × 10−9 SI, with an SD of 9.4 × 10−9 SI (Table S2, Supplementary material). These values are low compared to the factory-set accuracy for a single measurement, which is 2 × 10−8 SI. The sigma test does not take into account the influence of noise from moving parts of the pickup unit. This effect must be taken into account in the procedure of measuring the empty sample holder. When measuring the magnetic susceptibility of an empty sample holder, any noise generated by the up/down mechanism (e.g., the motor mechanism moving the sample in and out of the coil) is also taken into account. Since the magnetic susceptibility of the filter with the collected PM is relatively low (in the range of 10−6 SI), the sample holder should be selected so that its magnetic susceptibility and error are at an acceptable level. For the purposes of our measurements, we used a part of the standard KLY CUB20 holder with a filter adapter, which has the lowest value of magnetic susceptibility of all holders dedicated to the MFK1-FA kappabridge (see Table S2, Supplementary material).
The magnetic susceptibility of such prepared holder was measured repeatedly 88 times to evaluate the SD for that series of measurements. The magnetic susceptibility of the holder ranged from − 3.09 × 10−7 to − 0.2 × 10−7 SI, with an average value of − 1.64 × 10−7 SI, and an SD of 0.5 × 10−7 SI (Table S2, Supplementary material).
In order to assess the distribution of magnetic susceptibility of clean (unexposed) filters and their measurement errors, two tests were performed. The first test involved randomly selecting 20 filters manufactured by Hahnemühle, for which the magnetic susceptibility was measured 10 times. The values of κ were low and in the range from − 1.0 × 10−7 to 4.3 × 10−8 SI, with an average value of − 1.54 × 10−8 SI and an SD of 4.1 × 10−8 SI (Table S2, Supplementary material). In the second test, we analyzed the magnetic susceptibility of clean filters, which was measured for five randomly selected filters after opening a new box containing a collection of 100 filters (according to the procedure described in Section 3.3). In this case, the magnetic susceptibility varied from − 1.5 × 10−8 to 3.6 × 10−8 SI, with an average value of 3.1 × 10−8 SI, and the measurement error did not exceed 9.0 × 10−8 SI (Table S2, Supplementary material). We can conclude that the magnetic susceptibility values of clean filters and their measurement errors are close to the measurement accuracy of the MFK1-FA, given by the manufacturer as 2 × 10−8 SI. Measuring the magnetic susceptibility of each clean filter before exposing it to PM will not improve the accuracy of the measurements. It is due to the distribution of magnetic susceptibility of clean filters is within the value of the standard deviation determined from a ten-fold measurement of the magnetic susceptibility of a single clean filter.
In order to obtain a general view of the measurement error of κ filters after exposure, an analysis of the entire measurement series (2016–2020) collected at three studied stations was performed. According to the established procedure, for each exposed filter, the magnetic susceptibility was measured ten times and the SD was calculated. For each collection of filters collected at the stations located in Warsaw and Cracow, the SD values were below 5 × 10−8 SI for 84–88% of the studied filters. In addition, the values of SD were also below 15 × 10−8 SI for 99% filters with PM10 and 99% filters with PM2.5 (Table S3, Supplementary material). This analysis shows that the estimated measurement error of κ of the exposed filters is relatively low and depends mainly on the sensitivity of the used apparatus (in our case it was 2 × 10−8 SI). However, to assess the maximum error of the magnetic susceptibility of PM, according to formula (5), the \({\Delta \kappa }_{{\text{PM}}}\) must be a sum of the errors of the exposed filter, the holder, and the clean filter. As was shown above, for an empty holder, the \({\Delta \kappa }_{h}\) is 5.0 × 10−8 SI, and for a clean filter, \({\Delta \kappa }_{c}\) is approximately 4.1 × 10−8 SI (Table S3, Supplementary material). The calculation of the maximum error of magnetic susceptibility of PM (\({\Delta \kappa }_{PM}\)) was performed for each collection of filters, and the results were the same, and they are listed in the last row of Table S3 (Supplementary material). Thus, in our case the maximum error \({\Delta \kappa }_{PM}\) did not exceed 24 × 10−8 SI for 99% of all the studied filters (and was below 14 × 10−8 SI for 88% of filters, details in Table S3, Supplementary material). Such low errors of \({\Delta \kappa }_{PM}\) are crucial for obtaining reliable data in the procedure of monitoring of magnetic susceptibility of PM collected on filters.
The value of the maximum error of magnetic susceptibility of 24 × 10−8 SI is valid for our collections of data. In each other case depending on the collected data series, the procedure has to be followed independently. It is influenced by the equipment accuracy as well as the magnetic susceptibility of clean and exposed filters. If the mass-specific magnetic susceptibility (χPM) is also calculated, in addition, the error of mass has to be taken into account. However, due to high accuracy of the mass determination, it usually less affects the final value of the error of mass-specific magnetic susceptibility. For each station, calculations of errors were performed for selected three filters with high, moderate, and low values of κPM based on Eq. (8) (Table S1, Supplementary material). Results indicate that percentage error of ΔχPM was below 15% and below 22% for the filters with low magnetic susceptibility for PM10 and PM2.5, respectively, while, for filters with average magnetic susceptibility, the values of ΔχPM were below 4% and below 7% for PM10 and PM2.5, respectively (Table S1, Supplementary material). Our analysis shows that a critical element of data evaluation in the case of PM collected on filters is the sensitivity of the equipment used to measure magnetic susceptibility and mass.
Atmospheric condition during mass measurements
The accuracy of the measurement of the mass of clean filters and PM collected on the filters can be affected by atmospheric conditions such as temperature, humidity, and pressure. The EN12341:2014 standard describes in detail the physical phenomena that affect the total error of mass measurement, with moisture adsorption being the most significant. This is especially important when measuring the mass near the balance's sensitivity limit. To avoid issues with moisture absorption, the EN12341:2014 standard specifies the ranges of meteorological parameters that must prevail in the room during weighing and in the desiccator during storage (i.e., temperature of 19–21 °C with an accuracy of ≤ 0.2 °C and humidity of 45–50% with an accuracy of ≤ 2%). When these requirements are met, the maximum error of mass of clean filters and filters with PM should not exceed 40 and 60 μg, respectively, assuming a filter exposure time of 24 h.
Since the goal of this study was not to monitor the mass of PM according to the standards applicable to the network of air quality monitoring stations, the exposure time of each filter was extended to 72 h to collect sufficient PM mass to determine the magnetic susceptibility with satisfactory sensitivity. In this case, the maximum error for the mass measurement of clean filters and filters after 72 h PM exposure was approximately 60 and 80 μg, respectively.
Outlook for the future
Automatic optical light scattering systems are commonly used in many network monitoring stations to measure PM concentrations with 1-min readings and hourly averages. For such stations, no PM filters are collected and our method cannot be applied. Although PM collected by low-volume dust samplers show average changes in PM concentration over a longer period of time, they can be used to analyze PM sources and their contribution to the PM mass. We believe that the magnetic susceptibility of PM is an additional parameter that enriches the analysis, because magnetically strong dust particles mainly come from anthropogenic sources.
Conclusions
-
1.
To effectively use magnetic susceptibility for PM monitoring, it is essential to follow a standard procedure for conditioning filters, measuring magnetic susceptibility and PM mass, to ensure repeatable values and measurement accuracy with acceptable errors.
-
2.
The exposure time for collecting PM on the filter should be selected based on the distribution of magnetic susceptibility of the PM, which is mainly affected by the contribution of magnetic particles to the PM mass, taking into account environmental conditions such as the number and intensity of sources emitting magnetic particles into the atmosphere.
-
3.
The following conditions must be met to use magnetic susceptibility as a tool to track the contribution of iron-rich anthropogenic particles to PM.
-
a.
The magnetic susceptibility of PM collected on the filters can be determined with satisfactory accuracy provided that a magnetic susceptibility meter with a sensitivity of 5–10 × 10−8 SI is used.
-
b.
Magnetic susceptibility measurement error can be minimized by precisely measuring an empty sample holder, a clean filter, and reducing the noise from the surrounding space.
-
c.
In order to obtain the mass-specific magnetic susceptibility of PM, it is necessary to precisely determine the PM mass in accordance with the EN12341:2014 standard.
-
a.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
Abbasi S (2022) Magnetic particles weight as an indicator for heavy metals concentration. Lett Appl NanoBioSci 11:3770–3779. https://doi.org/10.33263/LIANBS113.37703779
Abdulaziz M, Alshehri A, Yadav ICh, Badri H (2022) Pollution level and health risk assessment of heavy metals in ambient air and surface dust from Saudi Arabia: a systematic review and meta-analysis. Air Qual Atmos Health 15:799–810. https://doi.org/10.1007/s11869-022-01176-1
Adamiec E, Jarosz-Krzemińska E, Bilkiewicz-Kubarek A (2022) Adverse health and environmental outcomes of cycling in heavily polluted urban environments. Sci Rep 12:148. https://doi.org/10.1038/s41598-021-03111-3
Agico (2009) MFK1-FA / CS4 / CSL, MFK1-A / CS4 / CSL, MFK1-FB MFK1-B User’s Guide, Modular system for measuring magnetic susceptibility, anisotropy of magnetic susceptibility and temperature variation of magnetic susceptibility. Advanced Geoscience Instruments Co., Brno, Czech Republic
Agico (2022) SAFYR7 Kappabridge control software version 7.5.04, User manual, AGICO Advanced Geoscience Instruments Co., Brno, Czech Republic
Anis N, Kumar A, Kumar Arya A (2023) Assessment of concentration and distribution of contaminants using magnetic susceptibility measurements. Pollution 9(1):139–149. https://doi.org/10.22059/poll.2022.341263.1488
Bourliva A, Papadopoulou L, Aidona E, Giouri K, Simeonidis K, Vourlias G (2017) Characterization and geochemistry of technogenic magnetic particles (TMPs) in contaminated industrial soils: assessing health risk via ingestion. Geoderma 295:86–97. https://doi.org/10.1016/j.geoderma.2017.02.001
Bové H, Bongaerts E, Slenders E, Bijnens EM, Saenen ND, Gyselaers W, Van Eyken P, Plusquin M, Roeffaers MBJ, Ameloot M, Nawrot TS (2019) Ambient black carbon particles reach the fetal side of human placenta. Nat Commun 10:3866. https://doi.org/10.1038/s41467-019-11654-3
Bućko MS, Magiera T, Johanson B, Petrovský E, Pesonen LJ (2011) Identification of magnetic particulates in road dust accumulated on roadside snow using magnetic, geochemical and micro-morphological analyses. Environ Pollut 159:1266–1276. https://doi.org/10.1016/j.envpol.2011.01.030
Čabanová K, Hrabovská K, Matějková P, Dědková K, Tomášek V, Dvořáčková J, Kukutschová J (2019) Settled iron-based road dust and its characteristics and possible association with detection in human tissues. Environ Sci Pollut Res 26:2950–2959. https://doi.org/10.1007/s11356-018-3841-x
Calderón-Garcidueñas L, González-Maciel A, Reynoso-Robles R, Hammond J, Kulesza R, Lachmann I, Torres-Jardón R, Mukherjee PS, Maher BA (2020) Quadruple abnormal protein aggregates in brainstem pathology and exogenous metal-rich magnetic nanoparticles. The substantia nigrae is a very early target in young urbanities and the gastrointestinal tract likely a key brainstem portal. Environ Res 191:110139. https://doi.org/10.1016/j.envres.2020.110139
Comde-Derenda (2014) Environmental Monitoring Systems Automatic dust sampling system for collecting particular matter PM10 or PM2.5 or PM1, Types: PNS-DM 3.1 / PNS-DM 6.1 Instruction manual, Comde-Derenda GmbH, Stahnsdorf, Germany
Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe. Official Journal of the European Communities, L 152/1, 11.06.2008. Found at: http://data.europa.eu/eli/dir/2008/50/oj
Directive 1999/30/EC of the European Parliament and of the Council of 22 April 1999 relating to limit values for sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter and lead in ambient air, Official Journal of the European Communities, 163, 29.6.1999, p. 41–60. Found at: http://data.europa.eu/eli/dir/2008/50/oj
EN 12341:2014 Ambient air - Standard gravimetric measurement method for the determination of the PM10 or PM2.5 mass concentration of suspended particulate matter. Found at: https://standards.iteh.ai/catalog/standards/cen/9e212b76-3171-40b4-9d69-07409bc6bf75/en-12341-2023
Evans ME, Heller F (2003) Environmental magnetism: principles and applications of enviromagnetics. Elsevier Science, Academic Press, San Diego (USA)
Gonet T, Maher BA (2019) Airborne, Vehicle-derived Fe-bearing nanoparticles in the urban environment: a review. Environ Sci Technol 53:9970–9991. https://doi.org/10.1021/acs.est.9b01505
Gonet T, Maher BA, Kukutschová J (2021) Source apportionment of magnetite particles in roadside particulate matter. Sci Total Environ 752:141828. https://doi.org/10.1016/j.scitotenv.2020.141828
Górka-Kostrubiec B, Dudzisz K (2023) Effect of COVID-19 pandemic restrictions on air pollution at a local scale in urban areas affected by high-intensity vehicle traffic in Poland. Acta Geophys 71:1109. https://doi.org/10.1007/s11600-023-01026-3
Górka-Kostrubiec B, Król E, Jeleńska M (2012) Dependence of air pollution on meteorological conditions based on magnetic susceptibility measurements: a case study from Warsaw. Stud Geophys Geod 56:861–877. https://doi.org/10.1007/s11200-010-9094-x
Górka-Kostrubiec B, Magiera T, Dudzisz K, Dytłow S, Wawer M, Winkler A (2020) Integrated magnetic analyses for the discrimination of urban and industrial dusts. Minerals 10:1056. https://doi.org/10.3390/min10121056
Górka-Kostrubiec B, Świetlik R, Szumiata T, Dytłow S, Trojanowska M (2023) Integration of chemical fractionation, Mössbauer spectrometry, and magnetic methods for identification of Fe phases bonding heavy metals in street dust. J Environ Sci 124:875–891. https://doi.org/10.1016/j.jes.2022.02.015
Hammond J, Maher BA, Gonet T, Bautista F, Allsop D (2022) Oxidative stress, cytotoxic and inflammatory effects of urban ultrafine road-deposited dust from the UK and Mexico in human epithelial lung (Calu-3) cells. Antioxidants 11:1814. https://doi.org/10.3390/antiox11091814
Harikrishnan N, Chandrasekaran A, Ravisankara R, Alagarsamyc R (2018) Statistical assessment to magnetic susceptibility and heavy metal data for characterizing the coastal sediment of East coast of Tamilnadu, India. Appl Radiat Isot 135:177–183. https://doi.org/10.1016/j.apradiso.2018.01.030
Harrison RM, Bousiotis D, Mohorjy AM, Alkhalaf AK, Shamy M, Alghamdi M, Khoder M, Costa M (2017) Health risk associated with airborne particulate matter and its components in Jeddah, Saudi Arabia. Sci Total Environ 590–591:531–539. https://doi.org/10.1016/j.scitotenv.2017.02.216
Hofman J, Maher BA, Muxworthy AR, Wuyts K, Castanheiro A, Samson R (2017) Biomagnetic monitoring of atmospheric pollution: a review of magnetic signatures from biological sensors. Environ Sci Technol 51:6648–6664. https://doi.org/10.1016/j.scitotenv.2017.02.216
Hrouda F, Chlupáčová M, Pokorný J (2006) Low-field variation of magnetic susceptibility measured by the KLY-4S Kappabridge and KLF-4A magnetic susceptibility meter: accuracy and interpretational programme. Stud Geophys Geod 50:283–299. https://doi.org/10.1007/s11200-006-0016-x
Hwang H-M, Fiala MJ, Park D, Wade TL (2016) Review of pollutants in urban road dust and stormwater runoff: part 1. Heavy metals released from vehicles. Int J Urban Sci 23:445–463. https://doi.org/10.1080/12265934.2016.1193041
Kapper KL, Bautista F, Goguitchaishvili A, Bógalo MF, Cejudo-Ruíz R, Cervantes Solano M (2020) The use and misuse of magnetic methods to monitor environmental pollution in urban areas. Bol Soc Geol Mex 72(1):A111219. https://doi.org/10.18268/bsgm2018v72n1a111219
Lagler F, Barbiere M, Borowiak A (2019) Evaluation of the field comparison exercise for PM10 and PM2.5, Ispra, January 18th – March 14th, 2018, EUR 29939 EN, Publications Office of the European Union, Luxembourg. https://doi.org/10.2760/32013,JRC118170
Loaiza-Ceballos MC, Marin-Palma D, Zapata W, Hernandez JC (2022) Viral respiratory infections and air pollutants. Air Qual Atmos Health 15:105–114. https://doi.org/10.1007/s11869-021-01088-6
TA Luft (2002) Technical instructions on air quality control – TA Luft (Technische Anleitung zur Reinhaltung der Luft), Federal Ministry for Environment, Nature Conservation and Nuclear Safety, First General Administrative Regulation Pertaining the Federal Immission Control Act, GMBl. (Gemeinsames Ministerialblatt - Joint Ministerial Gazette, 30 July 2002), Germany. Found at: https://www.verwaltungsvorschriften-im-internet.de/bsvwvbund_18082021_IGI25025005.htm
Magiera T, Parzentny H, Róg L, Chybiorz R, Wawer MS (2015) Spatial variation of soil magnetic susceptibility in relation to different emission sources in southern Poland. Geoderma 255–256:94–103. https://doi.org/10.1016/j.geoderma.2015.04.028
Magiera T, Górka-Kostrubiec B, Szumiata T, Wawer M (2021) Technogenic magnetic particles from steel metallurgy and iron mining in topsoil: indicative characteristic by magnetic parameters and Mössbauer spectra. Sci Total Environ 775:145605. https://doi.org/10.1016/j.scitotenv.2021.145605
Magiera T, Górka-Kostrubiec B, Szumiata T, Bućko MS (2023) Technogenic magnetic particles in topsoil: characteristic features for different emission sources. Sci Total Environ 865:161186. https://doi.org/10.1016/j.scitotenv.2022.161186
Maher BA, Ahmed IAM, Karloukovski VV, MacLaren DA, Foulds PG, Allsop D, Mann DMA, Torres-Jardón R, Calderón-Garcidueñas L (2016) Magnetite pollution nanoparticles in the human brain. Proc Natl Acad Sci USA 113(39):10797–10801. https://doi.org/10.1073/pnas.1605941113
Mantovani L, Tribaudino M, Solzi M, Barraco V, De Munari E, Pironi C (2018) Magnetic and SEM-EDS analyses of Tilia cordata leaves and PM10 filters as a complementary source of information on polluted air: results from the city of Parma (Northern Italy). Environ Pollut 239:777–787. https://doi.org/10.1016/j.envpol.2018.04.055
Miller MR, Raftis JB, Langrish JP, McLean SG, Samutrtai P, Connell SP, Wilson S, Vesey AT, Fokkens PHB, Boere AJF, Krystek P, Campbell CJ, Hadoke PWF, Donaldson K, Cassee FR, Newby DE, Duffin R, Mills NL (2017) Inhaled nanoparticles accumulate at sites of vascular disease. ACS Nano 11:4542–4552. https://doi.org/10.1021/acsnano.6b08551
Muxworthy AR, Schmidbauer E, Petersen N (2002) Magnetic properties and Mössbauer spectra of urban atmospheric particulate matter: a case study from Munich, Germany. Geophys J Int 150(2):558–570. https://doi.org/10.1046/j.1365-246X.2002.01725.x
Muxworthy AR, Matzka J, Davila AF, Petersen N (2003) Magnetic signature of daily sampled urban atmospheric particles. Atmos Environ 37(29):4163–4169. https://doi.org/10.1016/S1352-2310(03)00500-4
Nadali A, Leili M, Karami M, Bahrami A, Afkhami A (2022) Correction to: The short-term association between air pollution and asthma hospitalization: a time-series analysis. Air Qual Atmos Health 15:901. https://doi.org/10.1007/s11869-021-01111-w
Néel L (1955) Some theoretical aspects of rock-magnetism. Adv Phys 4(14):191–243. https://doi.org/10.1080/00018735500101204
Petrovský E, Kapička A, Grison H, Kotlík B, Miturová H (2020) Negative correlation between concentration of iron oxides and particulate matter in atmospheric dust: case study at industrial site during smoggy period. Environ Sci Eur 32:134. https://doi.org/10.1186/s12302-020-00420-8
Pope CA III, Coleman N, Pond ZA, Burnett RT (2020) Fine particulate air pollution and human mortality: 25+ years of cohort studies. Environ Res 183:108924. https://doi.org/10.1016/j.envres.2019.108924
Prajith A, Rao VP, Kessarkar PM (2015) Magnetic properties of sediments in cores from the Mandovi estuary, western India: inferences on provenance and pollution. Mar Pollut Bull 99:338–345. https://doi.org/10.1016/j.marpolbul.2015.07.034
Rachwał M, Wawer M, Magiera T, Steinnes E (2017) Integration of soil magnetometry and geochemistry for assessment of human health risk from metallurgical slag dumps. Environ Sci Pollut Res 24:26410–26423. https://doi.org/10.1007/s11356-017-0218-5
Rachwał M, Wawer M, Jabłońska M, Rogula-Kozłowska W, Rogula-Kopiec P (2020) Geochemical and mineralogical characteristics of airborne particulate matter in relation to human health risk. Minerals 10:866. https://doi.org/10.3390/min10100866
Radwag (2023) 5Y series balances (UYA 5Y ultra-microbalances, MYA 5Y microbalances, XA 5Y.M microbalances, XA 5Y analytical balances, XA 5Y.F analytical balances for filters, 5Y PM precision balances, 5Y HRP balances) IMMU-111-10-02-23-ENG, 2023, User manual, Radwag, Radom, Poland
Sagnotti L, Macrì P, Egli R, Mondino M (2006) Magnetic properties of atmospheric particulate matter from automatic air sampler stations in Latium (Italy): toward a definition of magnetic fingerprints for natural and anthropogenic PM sources. J Geophys Res 111:B12S22. https://doi.org/10.1029/2006JB004508
Saragnese F, Lanci L, Lanza R (2011) Nanometric-sized atmospheric particulate studied by magnetic analyses. Atmos Environ 45:450–459. https://doi.org/10.1016/j.atmosenv.2010.09.057
Strak M, Janssen N, Beelen R, Schmitz O, Vaartjes I, Karssenberg D, van den Brink C, Bots ML, Dijst M, Brunekreef B, Hoeka G (2017) Long-term exposure to particulate matter, NO2 and the oxidative potential of particulates and diabetes prevalence in a large national health survey. Environ Int 108:228–236. https://doi.org/10.1016/j.envint.2017.08.017
Sung JH, Oh I, Kim A, Lee J, Sim ChS, Yoo Ch, Park SJ, Kim G-G, Kim Y (2018) Environmental and body concentrations of heavy metals at sites near and distant from industrial complexes in Ulsan, Korea. J Korean Med Sci 33:33. https://doi.org/10.3346/jkms.2018.33.e33
Szczepaniak-Wnuk I, Górka-Kostrubiec B, Dytłow S, Szwarczewski P, Kwapuliński P, Karasiński J (2020) Assessment of heavy metal pollution in Vistula river (Poland) sediments by using magnetic methods. Environ Sci Pollut Res 27:24129–24144. https://doi.org/10.1007/s11356-020-08608-4
Szuszkiewicz M, Magiera T, Kapička A, Petrovský E, Grison H, Gołuchowska B (2015) Magnetic characteristics of industrial dust from different sources of emission: a case study of Poland. J Appl Geophys 16:84–92. https://doi.org/10.1016/j.jappgeo.2015.02.027
Tauxe L (1998) Paleomagnetic principles and practice; Kluwer Academic Publishers, Dordrecht, (Nederland). https://doi.org/10.1007/0-306-48128-6
Taylor JR (1982) An introduction to error analysis: the study of uncertainties in physical measurements. University Science Books, Sausalito, California
Thompson R, Oldfield F (1986) Environmental magnetism. Allen & Unwin Publishers Ltd, London, (U.K). https://doi.org/10.1007/978-94-011-8036-8
Thurston GD, Ahn J, Cromar KR, Shao Y, Reynolds HR, Jerrett M et al (2016) Ambient particulate matter air pollution exposure and mortality in the NIH-AARP Diet and Health cohort. Environ Health Perspect 124:484–490. https://doi.org/10.1289/ehp.1509676
Wang G, Liu Y, Chen J, Ren F, Chen Y, Ye F, Zhang W (2018a) Magnetic evidence for heavy metal pollution of topsoil in Shanghai, China. Front Earth Sci 12:125–133. https://doi.org/10.1007/s11707-017-0624-5
Wang L, Hu Sh, Ma M, Wang X, Wang Q, Zhang Z, Shen J (2018b) Responses of magnetic properties to heavy metal pollution recorded by lacustrine sediments from the Lugu Lake, Southwest China. Environ Sci Pollut Res 25:26527–26538. https://doi.org/10.1007/s11356-018-2725-4
Weichenthal S, Bai L, Hatzopoulou M, van Ryswyk K, Kwong JC, Jerrett M, van Donkelaar A, Martin RV, Burnett RT, Lu H, Chen H (2017) Long-term exposure to ambient ultrafine particles and respiratory disease incidence in Toronto, Canada: a cohort study. Environ Health 16:64. https://doi.org/10.1186/s12940-017-0276-7
Winkler A, Contardo T, Vannini A, Sorbo S, Basile A, Loppi S (2020) Magnetic emissions from brake wear are the major source of airborne particulate matter bioaccumulated by lichens exposed in Milan (Italy). Appl Sci 10:2073. https://doi.org/10.3390/app10062073
Wu W, Yuefei JY, Carlsten Ch (2018) Inflammatory health effects of indoor and outdoor particulate matter. J Allergy Clin Immunol 141:833–844. https://doi.org/10.1016/j.jaci.2017.12.981
**a D, Wang B, Yu Y, Jia J, Nie Y, Wang X, Xu S (2014) Combination of magnetic parameters and heavy metals to discriminate soil-contamination sources in Yinchuan—A typical oasis city of Northwestern China. Sci Total Environ 485:83–92. https://doi.org/10.1016/j.scitotenv.2014.03.070
Zhang W, Dong Ch, Hutchinson SM, Ge C, Wang F, Feng H (2018) Recent Applications of Mineral Magnetic Methods in Sediment Pollution Studies: a Review. Curr Pollut Rep 4:1–7. https://doi.org/10.1007/s40726-018-0075-y
Zheng S, Pozzer A, Cao CX, Lelieveld J (2015) Long-term (2001–2012) concentrations of fine particulate matter (PM2.5) and the impact on human health in Bei**g, China. Atmos Chem Phys 15:5715–5725. https://doi.org/10.5194/acp-15-5715-2015
Acknowledgements
We would like to acknowledge the support of the EPOS-PL project. This project contributed to the provision of the laboratory facilities, including the PM samplers and Radwag microbalance, used in this study at IG PAS. We also acknowledge the anonymous reviewer and Eduard Petrovský for their constructive comments and suggestions that improved the manuscript.
Funding
This work was supported by the statutory activity No. 3841/E-41/S/2023 from the Ministry of Science and Higher Education of Poland and the EPOS-PL project (No. POIR.04.02.00–14-A003/16), which was co-financed by the European Union from the funds of the European Regional Development Fund.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by B. Górka-Kostrubiec, T. Werner, and G. Karasiński. The first draft of the manuscript was written by B. Górka-Kostrubiec, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethical approval
Not applicable.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Conflict of interest
The authors declare no competing interests.
Additional information
Responsible Editor: Gerhard Lammel
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Highlights
• Develo** and refining the protocol for measuring the κ of PM collected on filters.
• Assessing the accuracy of measurements for magnetic susceptibility of PM.
• Outlining and refining a procedure for collecting PM on filters for measuring κ.
• Indicating the time exposure of filters to PM.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Górka-Kostrubiec, B., Werner, T. & Karasiński, G. Measuring magnetic susceptibility of particulate matter collected on filters. Environ Sci Pollut Res 31, 4733–4746 (2024). https://doi.org/10.1007/s11356-023-31416-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11356-023-31416-5