Particulate matter can have negative effects on health when it enters the human body. Therefore, measures have been taken against the exposure to particulate matter. This poses the question of how effective these measures truly are. In this research, the effectiveness of extraction systems, catalytic converters in exhausts and face masks is studied. Test setups were designed to measure particulate matter emission, with and without measures, and to measure the filtration rate of face masks. The conducted research estimated that extraction systems and catalytic converters in exhaust systems are effective measures and that face masks are effective, but mostly in ideal, sealed off, situations. These findings support previously conducted studies. Further research is needed in which fitting and human breathing are taken into account to determine the effectiveness of a face mask more accurately.
Particulate matter (abbreviated: PM) is defined as all particles smaller than 10 μm suspended in the air around us. Often, particulate matter is divided in 3 groups based on dimensions; PM10, PM2.5 and PM0.1. PM10 encompasses all particles smaller than 10 μm, PM2.5 encompasses particles smaller than 2.5 μm and PM0.1 all particles smaller than 0.1 μm. PM0.1 are also called ultrafine particles (UFPs); it is difficult to separate this group from PM2.5, and in this paper no distinction will be made between PM0.1 and PM2.5.
The time particulate matter remains airborne depends on the dimensions of the particles; bigger particles remain in the air for a shorter period of time since they are deposited more easily and travel shorter distances, smaller particles remain airborne for a longer period of time, thus they travel longer distances .
Even though the exact effects of particulate matter are hard to determine because of the proportions and measurability of particulate matter, some effects are known:
Particulate matter in the atmosphere has an impact on the environment in several ways. Particulate matter with a dimension of PM 2.5 or with a dimension of PM10 has a possibility of scattering and reflecting light. This light will then not be able to reach Earth’s surface, but instead the light will be cast back into space. This, ultimately, means that particulate matter has the potential to cool Earth   .
It is complicated to estimate the exact effects of this effect as a result of the variation in concentrations of particulate matter between different locations.
In addition, water molecules attach to particulate matter suspended in the air. This increases the number of cloud condensation nuclei, so when a cloud is formed, the water will form more and smaller droplets, since the water is distributed over these nuclei. This results in an optically thicker cloud which reflects more light. The smaller droplets detach not as quickly, so the rainfall from the cloud also decreases because of the particulate matter .
Besides the impact particulate matter has on the climate, it also has an impact on public health. Particulate matter can access the human body via the respiratory system which can lead to a variety of health complications . Because of the lack of research about the effects of particulate matter, it is hard to determine the short- and long term effects of particulate matter. A study by Burnett et al. states that exposure to particulate matter is an important cause for hospitalisation from cardiovascular diseases . Other studies show that an increased concentration of PM10 in the air increases the risk of coronary thrombosis . In 2005 the ‘Rijksinstituut voor Volksgezondheid en Milieu’ (RIVM) estimated that roughly 2 to 5 percent of disease cases in the Netherlands could be linked to exposure to particulate matter .
There is difficulty in distinguishing the long term effects of particulate matter from the short term effects since the exposure to particulate matter is continuous . Multiple studies suggest that exposure to particulate matter decreases the life expectancy of a human , consequently a reduction of the concentration of particulate matter in the air will cause an increase in the average life expectancy, since the exposure to particulate matter is reduced .
Thirty percent of the total exposure to particulate matter that people experience is due to cooking in and around the house . Another big contributor to the total exposure people experience to particulate matter is transportation over land. Transportation of people and goods over land produces thirty percent of the total particulate matter emissions in the Netherlands , thus having a significant contribution to the exposure to particulate matter.
Because of the negative impact particulate matter has on public health, it is important to minimise the exposure to particulate matter. To effectuate this, measures are taken to reduce the emission of particulate matter or to reduce the duration of people’s exposure to particulate matter. For instance exhausts are equipped with new technologies, better filters and catalytic converters so the emission is reduced . Another measure is that houses have Air duct cleaning Minneapolis done and are ventilated better to reduce the duration of exposure to particulate matter inside the house . In some cities people wear face masks to protect themselves from smog and exhaust gases, and thus reduce their exposure to particulate matter. These masks may not be as effective for this purpose since these masks do not fit properly on people’s faces . A different study suggests that cloth masks filter out a percentage of particulate matter, but that cloth masks are not as effective as disposable surgical masks .
The question this leaves us with is: to what extent are these measures and precautions functional and how significantly do they reduce the exposure to particulate matter and protect us from the dangers particulate matter may cause? In this research the functionality of various measures was examined. This was done by answering three separate sub-questions:
The first sub-question is: What is the effect of an extraction system on the concentration of particulate matter in the air in a kitchen while cooking? To reduce the exposure to particulate matter, some buildings are equipped with extraction and ventilation systems . With this question the effectiveness of these extraction systems is researched.
The second sub-question is: What is the difference in particulate matter emissions between a moped equipped with a catalytic converter and a moped without a catalytic converter? As mentioned before, in some countries, exhausts have to be equipped with catalytic converters to reduce the total emissions of particulate matter . With this question the effectiveness of these catalytic converters is tested.
The third and last sub-question is: What percentage of particulate matter is filtered out by different types of face masks in an ideal situation? Particulate matter can cause problems when it enters the body via the respiratory system , in some places people wear masks to prevent this. With this question the effectiveness of these masks is researched.
A study by Zhao, Azimi and Stephens shows that certain air filtration systems are successful in the reduction of the particulate matter concentration in buildings . Therefore it is expected that the effect of an extraction system will be a decrease in the concentration of particulate matter in the air and show lower maximum values compared to the concentration of particulate matter in the air when no use is made of an extraction system.
Since 2010 the emissions of particulate matter in the Netherlands have been reduced by 55 percent, partly because of the use of catalytic converters . In addition to this, a study by Khair, Di Silverio, shows that new technologies in exhausts decrease the emission of particulate matter by vehicles . Therefore, it is expected that a moped equipped with a catalytic converter will emit less particulate matter than a moped without a catalytic converter. This is expected to be perceived in a difference between the peaks of concentrations of particulate matter in exhaust gasses of a moped equipped with a catalytic converter and a moped without; in which the peak of a moped equipped with a catalytic converter is expected to significantly lower than the peak of a moped without. Thus a catalytic converter will reduce the emission of particulate matter of a moped.
A study by Cherrie et al. suggests that many face masks are not effective in protecting against particulate matter . Therefore, it is expected that the masks will not filter out a high percentage of particulate matter. Another study by Skakya, et al., suggests that cloth face masks are effective to a certain extent since they filter out a percentage of the particulate matter, but that disposable surgical masks are more effective in doing so . This is expected to be measured by a higher percentage of particulate matter being filtered out by a disposable mask than the percentage filtered out by a cloth mask.
Materials and methods
With all three test setups, the concentration of particulate matter in the air was measured with a Laser PM 2.5 Sensor . This device measures all particulate matter that has a size between 0.3 and 10 μm and it reads out the concentration of PM 10 and PM 2.5 every 10 seconds. The PM 10 values are all the particulate matter the device measures minus the PM 2.5 values. The PM 2.5 value is all the particulate matter that has a size between 0.3 and 2.5 μm. A python script was written to read these values and sort them in an Excel sheet. This device makes use of the laser scattering principle . The laser in the measuring area of the device, sends out light that is scattered on impact with a particle. The scattered light is transferred into electrical signals. The diameter and amount of particulate matter can be measured with this data, since there is a relationship between the signal waveforms and the diameter of the particulate matter. The relationship between waveforms and the diameter is: smaller particulate matter scatters light at larger angles and bigger particulate matter scatters light at smaller angles.
For the first experiment a kitchen with an extraction system will be used. The kitchen could easily be sealed off from the rest of the house. The kitchen was cleaned and cleared, then it was sealed off. For this, an airtight tarp was used. Two sensors were used during the experiment. One was placed close to the emission source and the other was placed one meter away from the emission source. Figure 1 is a schematic drawing of this test setup. (Photos of the whole test setup are included in the appendix)
Figure 1: Schematic drawing test setup for the extraction system (front view)
When executing the experiment, an egg was fried to create a source of PM. This was done in a pan; with every experiment the stove was set to the same temperature and the same amount of butter, twelve ml, was added to the pan; after every measurement the pan was cleaned so there would not be any remnants of food or particulate matter.
The measurements started with turning on the sensors, then the pan was heated for sixty seconds. After these sixty seconds, cooking butter was added and after thirty more seconds the egg was put in. The egg was fried for 330 seconds. After these six minutes, the stove was turned off and the sensor remained on for another three minutes. The PM levels during the frying of an egg were measured eight times; four times with the extraction system switched on and four times without the extraction system.
All the data was automatically gathered in an Excel spreadsheet. After this data was all combined, the averages of all the 4 repetitions of the experiment were calculated and used for a visualisation, in the form of a diagram. In this diagram the difference in peaks can be compared, so the impact of an extractor system on the concentration of particulate matter in the air over time can be determined.
The second experiment was conducted outside. For this a Hanway raw 50 moped was used. This moped has a removable exhaust pipe with a catalytic converter. As for the preparations for this experiment an aluminum tube was used to redirect the exhaust gases. This tube was two meters long and was fixed to a closed container; a test chamber of thirty liters on one end. This was done by drilling a hole in the test chamber and putting the tube through it. This test chamber was sealed on the bottom so the PM could not leave the test chamber. In the test chamber two holes were drilled for the sensors, one in the top of the test chamber and one in the bottom. This was done because the gasses out of the exhaust pipe are hot. Hot gasses move up which means that the air in the test chamber was not homogeneous and it is most likely that the sensor on top measures a higher concentration than the sensor at the bottom. Figure 2 is a schematic drawing of this test setup. (Photos of the whole test setup are included in the appendix)
Figure 2: Schematic drawing of the test setup for the catalytic converter (top view)
When conducting the experiment the first step executed was starting the sensors. These ran for twenty seconds before the moped was started and was set into neutral. After one minute the moped was turned off and the sensor remained on for another five minutes. After this time, the experiment was finished and the sensors were stopped. After the experiment was done, the test chamber was turned upside down so all the PM would exit the chamber. This test was done 5 times with a catalytic converter and 3 times without a catalytic converter. Since the person conducting this experiment was exposed to the exhaust gases of the moped, the experimenter needed to wear a protective mask at all times during the experiment.
The data from this second experiment was gathered in an Excel spreadsheet just like the previous experiment. The averages of all five repetitions with catalytic converter and three repetitions without catalytic converter were calculated and used for a visualisation, in the form of a diagram. In this diagram the differences in the concentration of particulate matter in exhaust gas when using a catalytic converter and when not using a catalytic converter, can be compared.
The third experiment was conducted outside. The test chamber and the aluminium tube which were used in the test setup for the catalytic converter were also used in the experiment. At the end of the tube inside of the chamber, the mask or fabric which is going to be tested is fixed. The vacuum cleaner was not attached directly to the tube, so the vacuum cleaner would not create too much suction and thus a vacuum. To make sure every test had the same amount of particulate matter at the start, a hole was drilled in the side of the test chamber so a match could be lit inside the test chamber. This hole could be covered up with duct tape during the test. (Photos of the whole test setup are included in the appendix). Figure 3 is a schematic drawing of this test setup.
Figure 3: Schematic drawing of the test setup for face masks (front view)
When conducting the experiment the first step executed was turning on the vacuum cleaner. Then the sensors were turned on and one match was lit and put into the test chamber. After the match burned out, the hole for the match was closed. Five minutes after turning on the sensors, the experiment was finished.
This experiment was performed two times for every setup; two times with the disposable mask, two times with the fabric mask and two times without a mask.
The data from this third experiment was gathered in an Excel spreadsheet. An average was calculated of the different takes. All the average values were added and a percentage decrease/increase compared to the sensor inside the chamber was calculated and displayed in a graph.
As shown in Figure 4 and Figure 5, when not using an extraction system, the average increase in PM in the air is higher than when using an extraction system. These increases were most substantial with PM 10. The PM 10 values at the start of the experiment averaged out at 24.17 μg/m³ PM 10 . The PM 2.5 values at the start of the experiment were on average around 15.08 μg/m³ PM 2.5. It is shown that the concentration of PM10 in the air whilst not using an extraction system had a higher peak value than whilst using an extraction system. The difference between these peaks was less with the PM2.5. Sensor 2 (denoted by ) was positioned further away from the stove and the extraction system: this resulted in lower PM concentrations measured by this sensor and also resulted in the sensor measuring a more gradual increase and decrease of PM.
As shown in Figure 6 and Figure 7, there is a difference between the content of PM in exhaust gasses of a moped when the moped’s exhaust has a catalytic converter versus when the moped’s exhaust is not equipped with a catalytic converter. When the moped was equipped with a catalytic converter, the average concentration of PM in the exhaust gasses was lower, compared to the average concentration of PM in the exhaust gasses when the moped was not equipped with a catalytic converter. The peaks of the content of PM 10 in the exhaust gas without the catalytic converter were on average roughly 3.18 (sensor 1) and 3.42 (sensor 2) times as high compared to the content of PM 10 in the exhaust gas with a catalytic converter. The peaks of the content of PM 2.5 in the exhaust gas without the catalytic converter were on average roughly 4.47 (sensor 1) and 5.55 (sensor 2) times as high compared to the content of PM 2.5 in the exhaust gas with a catalytic converter.
During test 3, the difference between the concentration of PM behind a mask and the concentration in the surrounding area was measured. The values in Table 1 were calculated by adding up all the measured values of PM behind the mask minus the sum of all the measured values of PM in front of the mask, and dividing this by the sum of all values of PM in front of the mask. As shown in Table 1, there is a decrease in the concentration of PM (particulate matter) when a face mask is used. There also appears to be a difference in percentage decrease between a fabric mask and a disposable mask. The disposable masks filtered out most of the PM; it filtered out 98.13 percent of the PM 2.5 and 97.33 percent of the PM10. While the fabric mask filtered out 77.26 percent of the PM 2.5 and 71.68 percent of the PM10.
For the raw data, please get in contact with us.
Figure 4: The average content PM 2.5 whilst baking an egg with and without a kitchen extractor hood.
Figure 5: The average content PM 10 whilst baking an egg with and without a kitchen extractor hood.
Figure 6: The average content PM 2.5 in the exhaust gasses of a moped with and without a catalyst measured with 2 sensors
Figure 7: The average content PM 10 in the exhaust gasses of a moped with and without a catalyst measured with 2 sensors
Table 1: Percentage decrease/increase in concentration Particulate Matter in the air using different kinds of masks and a control group.
|Type of mask→||No mask||Fabric mask||Disposable mask|
|Type of PM ↓|
Discussion and Conclusion
During the first experiment, the effectiveness of an extraction system on the removal of PM in the air during cooking was tested. The hypothesis states that it is expected that the effect of an extraction system will be that an extraction system decreases the concentration of PM in the air and show lower maximum values compared to the concentration of PM in the air when no use is made of an extraction system. The results from the conducted experiment supports this hypothesis; during the tests which were conducted with an extraction system, the peaks of the concentration of PM in the air were lower than when an extraction system was not used. Figures 4 and 5 show this, since the concentration PM in the air remained fairly constant during the test when the extraction system was used, while the concentration of PM in the air increased notably when the extraction system was not used.
It is possible to determine the percentage increase between the average amount of PM and the starting amount of PM. However, because the starting values fluctuated substantially, this gives unclear data which would not be useful to draw conclusions from. Instead, it is better to examine the difference between maximum values. As is mentioned before, the maximum values when the extraction system was used were lower than the maximum values when the extraction system was not used; hence, it can be concluded that the extraction system was effective in reducing the concentration of PM in the air during cooking.
A study conducted by Zhao, Azimi and Stephens shows that certain air filtration systems are successful in the reduction of the PM concentration in buildings . An extraction system does not filter the air, and therefore does not filter out PM. However, it does create under pressure in the room, resulting in new fresh outside air flowing into the kitchen. When this air contains less PM than the air previously in the kitchen, a decrease in concentration of PM in the air is expected. The results found with the conducted experiment correspond to the research by Zhao et al. and support the study’s claim. However, these results also indicate that the test setup might not have been sealed off perfectly. Still, the data is usable since the clear differences between using an extraction system and not using an extraction system were both tested in the same setup, and the difference between using an extraction system and not using one was notable.
During the experiment, the amount of PM in the outside air fluctuated quite a bit. This resulted in the starting values of PM in the air to fluctuate between experiments, which in turn resulted in it being difficult to draw conclusions from the data. Furthermore, not all eggs were exactly the same size; they were all medium size eggs, but it is possible that there was a slight difference in weight. It is not clear whether this caused more or less PM to form. In follow-up research, maintaining the starting values would be essential to make sure the data can be interpreted better and to form a clearer conclusion. Also, in future research, the cooking ingredient that is being tested, in this case eggs, should be exactly the same size to make sure this does not affect the amount of PM emitted during the cooking process. Another follow-up research that would be useful is a research in which different types of food are used during the cooking process, so possible differences in PM emission during the cooking process for different types of food can be shown.
During the second experiment the effectiveness of a catalytic converter on decreasing the PM emission of a moped was tested. The hypothesis for this experiment states that it is expected that a moped equipped with a catalytic converter will emit less PM than a moped without a catalytic converter. The results mentioned before support this hypothesis because they show a notably higher peak of PM without a catalytic converter than with a catalytic converter.
The starting points during this experiment were constant, because of a better sealed test setup, so a comparison could be made between the peaks of the concentrations of particulate matter in exhaust gas. The average peaks were determined by getting the average of the few highest data points and thereafter calculating the average of the attempts. After this, because the amount of PM in the air was substantially higher when no catalytic converter was used, the multiplication factor between the start and the peak concentration of particulate matter in exhaust gas when using a catalytic converter and when not using a catalytic converter was calculated. It would have been possible to add up all data points and then calculate the multiplication factor between them, but because of fluctuating data these values would not give a reliable image of the effect of a catalytic converter on the concentration of PM in the exhaust gas of a moped. Nonetheless, as mentioned before, the average peaks of PM in the air when not using a catalytic converter were substantially higher than when not using a catalytic converter, hence it can be concluded the catalytic converter ensures a moped emits less PM.
Multiple studies, a study by Khair, Di Silverio  and a study by RIVM , suggest that new technologies in exhausts like catalytic converters are effective in reducing the concentration of PM in the exhaust gases emitted. The conducted experiment supports these claims and corresponds to the studies and literature already available on the effectiveness of new technologies in exhausts in reducing the concentration of PM emitted in the exhaust gasses.
Originally, it was planned to test both variables, with or without a catalytic converter, 5 times, but the experiment had to be cut short due to noise complaints.
During the experiments, the test chamber which contained the exhaust gases might not have been perfectly sealed off, since the necessary materials were not at disposal; the expectation had been that the experiment would be conducted at the University of Utrecht, but this was not possible because of the covid-19 virus. It is expected that only a small amount of PM could escape the test setup. A follow-up experiment could be done with a fully sealed test setup so more accurate data could be acquired. Another follow-up research which should be conducted is one into other techniques in exhausts and their impact on PM emission, difference in emission between different types of exhausts, and differences in emission between exhausts of different vehicles than a moped.
During the third experiment, the percentage decrease of PM using face masks in an ideal situation was determined. The hypothesis for this last experiment states that a fask mask will not filter out a high percentage of PM and the difference between a cloth mask and a disposable surgical mask is that a disposable face mask is expected to filter out a larger percentage of PM than a cloth face mask. The results mentioned before support only half of this claim, showing that both types of mask filter out a lot of the PM; this was a 97.33 percentage of PM10 and a 98.13 percentage of PM2.5 for the surgical face mask; and a 71.68 percentage of PM10 and a 77.26 percentage of PM2.5 for the cloth face mask. Furthermore the difference between a surgical face mask and a cloth mask was that a surgical mask filters around 26.65 percent more PM10 and 20.87 percent more PM2.5 than a cloth mask; this result does support the hypothesis that disposable masks are more effective in filtering out particulate matter than cloth masks.
The percentage decrease (PM10 and/or PM2.5) between inside the test tube (behind the mask) and inside the test chamber (in front of the mask) was determined by adding up all the measurements per conducted experiment and calculating the percentage decrease/increase between the two sensors. The choice was made to add up all the values because the two sensors collected data at the same time. So, it was possible to calculate the average increase/decrease between the two sensors. Another option would have been to measure the differences between the maximum values; however, an average over time gives a more accurate depiction of how much PM actually gets filtered out of the air by a mask.
Studies about the effectiveness of face masks in filtering out PM differ from each other; a study by Cherrie et al.  suggests that face masks are not effective in filtering out PM, mostly because of the poor fit. Another study, a study by Skakya et al. suggests that face masks are to some extent effective in filtering out PM and that cloth masks are less effective in doing so than disposable, surgical masks. The conducted research does not fully correspond with these studies. First of all, the experiments suggested that masks are very effective in filtering out PM, which does not correspond with the study done by Cherrie et al. Furthermore, the experiments proved that cloth masks were less effective than disposable masks; this does correspond with another study, i.e., that by Shakya et al.
The experiment does not correspond with the study by Cherrie et al., possibly because in the designed test setup, how the face masks fit was not taken into account. In the test setup, the face mask completely sealed off the tube through which the air flowed, thus the test setup was not suited to answering the question about the protection against exposure to PM of people. To answer this question, a test setup needs to be designed that mimics human breathing and where fitting is taking into account. Because of the way an airflow was created, by sucking out air, a small vacuum could occur. As a result of this, suction would fluctuate and thus a measurement error would occur. In future research, an airflow should be created in another way so data would be more accurate. During the experiment, it was noticed that when not using a face mask the second sensor (behind the masks) recorded on average 50 percent higher PM 2.5 values than the first sensor (in front of the mask). The second sensor did not, on average, record higher PM 10 values than the first sensor. It was possible that the increased airflow, as a result of not using a mask to block the airflow, caused these faulty readings. However, this does not explain why PM 10 did not go up, so it is uncertain what caused the increase. In a follow up research, it would be necessary to maintain a steady airflow to prevent faulty reading by the sensor.
The effect of an extraction system on the concentration of particulate matter in the air while cooking is that the concentration of particulate matter in air does not peak as much as the concentration of particulate matter in air without an extraction system. Therefore, it is estimated that an extraction system is an effective measure against exposure to particulate matter.
The difference in particulate matter emissions between a moped equipped with a catalytic converter and one without a catalytic converter is that the concentration of particulate matter in the exhaust gas of a moped with a catalytic converter is lower than the concentration of particulate matter in the exhaust gas of a moped without a catalytic converter. So, a catalytic converter reduces the particulate matter emission of a moped and therefore is estimated to be an effective measure against the exposure to particulate matter.
Face masks could be effective in filtering out particulate matter in an ideal, sealed off, situation. A cloth mask filters out 77.26 percent of PM 2.5 and 71.68 percent of PM 10, while a disposable mask filters out 98.13 percent of PM2.5 and 97.33 percent of PM10. According to the conducted research in an ideal situation, a surgical mask with a perfect fit is estimated to be an effective measure against the exposure to particulate matter. Moreover a disposable mask is estimated to be a more effective measure against the exposure to particulate matter than a cloth mask.
The experimenters would like to thank: Dr. R. Holzinger, associate professor of Institute for Marine and Atmospheric Research (IMAU) Atmospheric Physics and Chemistry Group (APCG) at the University of Utrecht, for his assistance during the process of writing this article, in assisting with the design of the test setups and in answering questions the experimenters had concerning particulate matter; Dr. G. de Vries, teacher of physics and thesis mentor at the U-talent academy, for helping experimenters with general questions about writing an article; Roy Meinen, technician for Marine and Atmospheric Research (IMAU) Atmospheric Physics and Chemistry Group (APCG) at the University of Utrecht, for explaining and providing the sensors.
 Iijama, S. (1985). Electron Microscopy of Small Particles, Journal of Electron Microscopy; Volume 34, Issue 4, December 1985,, Retrieved from: https://doi.org/10.1093/oxfordjournals.jmicro.a050518 Consulted on: 3 September 2020
 RIVM (19 november 2018). Fijnstof. Retrieved from: https://www.rivm.nl/fijn-stof Consulted on: 3 September 2020
 RIVM (Januari 2013). Dossier Fijnstof, 1: Stof, hoe en wat. Dossier fijnstof, chapter 1, page 3 and page 7
Retrieved from https://www.rivm.nl/documenten/stof-hoe-en-wat Consulted on: 10 September 2020
 RIVM (Januari 2013). Dossier Fijnstof, 2: Emissies. Dossier fijnstof, chapter 2, page 4 and page 6
Retrieved from https://www.rivm.nl/documenten/emissies Consulted on: 10 September 2020
 US Environmental Protection Agency (n.d.) Particulate Matter (PM) basics. Retrieved from: https://www.epa.gov/pm-pollution/particulate-matter-pm-basics Consulted: 27 October 2020
 World Health Organization Europe (2006) Health risks of particulate matter from long-range transboundary air pollution. Chapter 2
Retrieved from https://www.euro.who.int/__data/assets/pdf_file/0006/78657/E88189.pdf
Consulted on: 27 October 2020
 Jacob, D. (1999) Introduction to Atmospheric Chemistry, Chapter 8, Princeton press. Retrieved from http://acmg.seas.harvard.edu/publications/jacobbook/bookchap8.pdf Consulted on 22 September 2020
 Jacobson, M. Z (2002) .: Control of fossil-fuel particulate black carbon and organic matter, possibly the most effective method of slowing global warming, Retrieved from: Journal of geographical research: atmoshphers, volume 107, issue D19, 4047 doi:10.1029/2001JD001376
Consulted: 24 September 2020
 Ten Hoeve, J.E., Remer, L.A., Jacobson, M.Z. (2011) Microphysical and radiative effects of aerosols on warm clouds during the Amazon biomass burning season as observed by MODIS: impacts of water vapor and land cover Retrieved from: Atmospheric Chemistry and Physics, issue 11, 3021–3036, https://doi.org/10.5194/acp-11-3021-2011 Consulted: 24 September 2020
 Zhang, R., Wang, G., Guo, S., Zamora, M.L., Ying, Q., Lin, Y., Wang, W., Hu, M., Wang, Y. (2015) Formation of urban fine particulate matter. Chemical reviews, volume 115, issue 10, pages 3803–3855 Retrieved from: https://www.researchgate.net/profile/Song_Guo21/publication/275894973_Formation_of_Urban_Fine_Particulate_Matter/links/5e7457ad458515c677c63a9d/Formation-of-Urban-Fine-Particulate-Matter.pdf Consulted: 24 September 2020
 Intergewestelijke Cel voor het Leefmilieu – IRCEL (2009) Gezondheidsimpact van fijn stof: is er een verminderde levensverwachting? Retrieved from:
Consulted on: 4 September 2020
 Burnett, R.T., Smith-Doiron, M., Stieb, D., Cakmak, S., Brook, J.R. (1999) Effects of
particulate and gaseous air pollution on cardiorespiratory hospitalizations. Retrieved from: Archives of Environmental Health, issue 54, 130–139. Consulted on: 4 September 2020
Morris, R.D. (2001) Airborne particulates and hospital admissions for cardiovascular
disease: a quantitative review of the evidence. Retrieved from: Environmental Health Perspectives, issue 109,
issue 4, 495–500. Consulted on: 4 September 2020
 Knol, A.B., Staatsen, B.A.M. (2005) Trends in the environmental burden of disease in the Netherlands. Retrieved from: https://www.rivm.nl/bibliotheek/rapporten/500029001.html Consulted on: 4 September 2020
 Polichetti, G., Stefania, C., Spinali, A., Trimarco, V., Nunzatia, A., (2009) Effects of particulate matter (PM10, PM2.5 and PM1) on the cardiovascular system Elsevier, volume 261, issue 1, 1-8 Retrieved from:
Consulted on: 4 September 2020
 Englert, N., (2004) Fine particles and human health—a review of epidemiological studies. Retrieved from: Toxicology Letters, volume 149, 235–242 Consulted on: 4 September 2020
 Brunekeef, B. (1997) Air pollution and life expectancy: is there a relation? Occupational and Environmental Medicine, volume 54, 781-784 Retrieved from
https://oem.bmj.com/content/oemed/54/11/781.full.pdf Consulted on 4 september 2020
 Apte, J., Brauer, M., Ezzati, M.,Pope, C. (2018) PM 2.5 Reduces Global and Regional Life Expectancy Environmental science & technology letters, volume 5, issue 9, 546–551 Retrieved from:
https://www.researchgate.net/publication/327167942_Ambient_PM_25_Reduces_Global_and_Regional_Life_Expectancy Consulted on 4 September 2020
 Pope, C.A., Ezzati, M., Dockery, D.W. (2009) Fine-particulate air pollution and life expectancy in the United States The new England Journal of medicine, volume 360, 376-386 Retrieved from:
https://pubmed.ncbi.nlm.nih.gov/19164188/ Consulted on 4 september 2020
 Jacobs, P., Kornaat, W., Borsboom, W.A. (2017) Fijnstof bij het koken: het effect van kookafzuiging op fijnstofconcentraties in woningen TU delft, TNO publications Retrieved from
https://www.narcis.nl/publication/RecordID/oai:tudelft.nl:uuid%3A6e023376-8efa-482b-9568-3c867305e003 Consulted on 10 September 2020
 Centraal Bureau van Statistiek (2016) Retrieved from;
Consulted on 10 September 2020
 Khair, M.K., Di Silverio, W.D. (1997) Exhaust gas recirculation system for simultaneously reducing NOx and particulate matter Retrieved from: https://patents.google.com/patent/US5806308A/en Consulted on 10 September 2020
 Zhao, D., Azimi, P., Stephens, B. (2015) Evaluating the Long-Term Health and Economic Impacts of Central Residential Air Filtration for Reducing Premature Mortality Associated with Indoor Fine Particulate Matter (PM2.5) of Outdoor Origin. International Journal of Environmental Research and Public Health, volume 12, issue 7, 8448-8479 Retrieved from:
https://www.mdpi.com/1660-4601/12/7/8448 Consulted on 10 September 2020
 Cherrie, J.W., Apsley, A., Cowie, H., et al. (2017) Effectiveness of face masks used to protect Beijing residents against particulate air pollution Occupational and Environmental Medicine volume 75, 446-452. Retrieved from:
Consulted on 23 November 2020
 Shakya, K.M., Noyes, A., Kallin, R., Peltier, R.E. (2017) Evaluating the efficacy of cloth facemasks in reducing particulate matter exposure Journal of exposure science & environmental epidemiology, volume 27, 352–357 Retrieved from:
https://pubmed.ncbi.nlm.nih.gov/27531371/ Consulted on 23 November 2020
 Nova Fitness Co., Ltd. (2015) Laser PM2.5 Sensor specification Retrieved from:
https://cdn-reichelt.de/documents/datenblatt/X200/SDS011-DATASHEET.pdf Consulted on 28 oktober
 Malvern panalytical (n.d.) Laser Diffraction (LD) – Particle size distributions from nanometers to millimeters.
Consulted on 28 Oktober 2020
The full test setup to test the catalytic converter The full setup to test the effect of an extraction system
Close up of the test chamber