(SPM) and fine particulate matter

Air Quality, Atmosphere & Health https://doi.org/10.1007/s11869-018-0645-6

Correlation between inorganic pollutants in the suspended particulate matter (SPM) and fine particulate matter (PM2.5) collected from industrial and residential areas in Greater Cairo, Egypt Abdallah A. Shaltout 1,2 K. Eleftheriadis 4

&

Salwa K. Hassan 3 & Sultan E. Alomairy 2 & M. Manousakas 4 & Andreas G. Karydas 5 &

Received: 24 May 2018 / Accepted: 1 November 2018 # Springer Nature B.V. 2018

Abstract Simultaneous sampling collection of suspended particulate matter (SPM) and fine aerosol particles with an aerodynamic diameter equal or less than 2.5 μm (PM2.5) from industrial and residential areas of Greater Cairo, Egypt, has been carried out during two different seasons namely autumn 2014 and winter 2014/2015. The average mass concentrations of both SPM and PM2.5 samples are higher than the annual mean levels, especially for the samples collected from the industrial area. In addition, the mass concentrations of SPM are much higher than the PM2.5 mass concentrations during the whole sampling period. The ratios of the mass concentration between the SPM and PM2.5 were determined to be equal to 20 ± 6 and 17 ± 4 for the residential and industrial areas, respectively, and these ratios seem to be constant during the two mentioned seasons. Based on our previous elemental analysis results using multiple secondary target energy dispersive X-ray fluorescence (EDXRF), 18 elements in both SPM and PM2.5 samples have been quantified. Remarkable variations in the elemental concentrations between the SPM and PM2.5 samples were obtained. Comparison and statistical analysis of the elemental composition of both SPM and PM2.5 have been investigated. The PMF model EPA 5.0 was utilized for source identification on both PM fractions. Seven sources were identified and their relative contributions in the two areas of the study were investigated. Keywords Suspended particulates matters (SPM) . PM2.5 . Positive matrix factorization (PMF) . Greater Cairo, Egypt

Introduction Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11869-018-0645-6) contains supplementary material, which is available to authorized users. * Abdallah A. Shaltout [email protected] 1

Spectroscopy Department, Physics Division, National Research Centre, El-Behooth St., Dokki, Cairo 12622, Egypt

2

Physics Department, Faculty of Science, Taif University, P.O. Box 888, Taif 21974, Kingdom of Saudi Arabia

3

Air Pollution Research Department, Environmental Research Division, National Research Centre, El-Behooth St., Dokki, Cairo 12622, Egypt

4

E.R.L., Institute of Nuclear and Radiological Sciences and Technology, Energy and Safety, NCSR BDemokritos^, 15310 Ag. Paraskevi, Athens, Attiki, Greece

5

Institute of Nuclear and Particle Physics, NCSR BDemokritos^, 15310 Ag. Paraskevi, Athens, Greece

Air pollution including different types of pollutants in Greater Cairo, Egypt, is a matter of serious concern. High mass concentrations of air particulates including fine and coarse particles, dust, soot, hydrocarbons, and heavy metal compounds were identified in Greater Cairo which may cause serious respiratory diseases and have a carcinogenic effect due to inhaling these particulates (Abd El Maksoud 2011; Abu-Allaban et al. 2002, 2007, 2009; Hassan 2006; Khoder 1997, 2007; Shakour et al. 2011). The growth of the population and the black cloud over the city, as well as the increase of the industrial activities, are the reasons for the air quality to deteriorate continuously in the city. Moreover, pollutant removal rates in Greater Cairo are very low due to the lack of rain during the whole year, while long residence times give rise to the photochemical production of secondary aerosol mass (Lazaridis et al. 2005). This increases the probability of exposure of the public to high air pollution. The total suspended particulate matter (SPM) has a different particle size ranging from fine and ultrafine particulate up

Air Qual Atmos Health

to the coarse particulates and it has long-term effects on plants, animals, and humans (Đorđević et al. 2004; Ostro et al. 2015; Stafoggia et al. 2017). The monthly and seasonal contributions of the SPM samples collected from different locations in Greater Cairo have been investigated and the elemental analysis has been quantified (Abd El Maksoud 2011; Anwar 2003; Hassan 2000; Hindy et al. 1990; Saleh 2002). The monthly and the seasonal mean concentrations of the SPM were found to be much higher than the annual mean level given by European air quality standards as well as the Egyptian ambient air quality standards and it reaches in some cases to 1054.56 μg/m3 at the industrial location south of the city (El-Tebeen) (Saleh 2002). At southeast of the city which is considered as an industrial location for cutting, polishing, and processing marble and granite (Shaq Al-Teban), the average SPM mass concentrations reach up to 932 μg/m3 and 1325 μg/m3 in winter and autumn, respectively (Abd El Maksoud 2011). The high mass concentration of SPM in the industrial areas may be attributed to the presence of anthropogenic sources such as industrial dust and industrial fine particulates originated from industrial activities, cement factories, coke, iron and steel factory, nonferrous metallurgical factory, and re-suspension of the deposited dust during the course of the day (Athanasopoulou et al. 2010). By utilizing three secondary targets, a tri-axial polarization geometry energy dispersive X-ray fluorescence (EDXRF) spectrometer working under vacuum environment was used for the elemental analysis of 19 elements in the present SPM samples namely Na, Al, Si, S, Cl, K, Ca, Ti, V, Mn, Fe, Ni, Cu, Zn, Pb, Se, Br, Rb, and Sr (Shaltout et al. 2018a). The average mass concentrations of the SPM were 531 ± 198 μg/m3 and 912 ± 230 μg/m3 at residential and industrial areas, respectively, and these values are relatively higher than the annual mean level. On the other hand, the fine particulate matter with an aerodynamic diameter equal or less than 2.5 μm (PM2.5) represents the most hazardous particles for the human health because they are stable in air and can be transported through the movable air masses over several hundreds of kilometers (Solomon 2012; Terrouche et al. 2016). It should be mentioned that the deposition of PM2.5 aerosols in the respiratory system has a higher probability. Once inside the lungs, they can aggravate respiratory conditions, which might lead to a series of respiratory diseases. Epidemiological studies have focused on the connection between concentrations of fine aerosols and daily mortality (Englert 2004; Goldberg et al. 2013; Green and Armstrong 2003). Intensification studies have been carried out to characterize the PM2.5 aerosols collected from Greater Cairo as well as other cities in the Middle East region using different spectroscopic techniques (AbuAllaban et al. 2002, 2007; Hassan and Khoder 2017; Saliba et al. 2010; Shaltout et al. 2013, 2014, 2015, 2017, 2018b). Recently, the concentration of 18 elements including light elements has been determined in PM2.5 samples by means of

EDXRF spectrometer utilizing polarization geometry and three different secondary targets (CaF 2 , Ge, and Mo) (Shaltout et al. 2018b). The average mass concentrations of PM2.5 aerosols collected from Greater Cairo were higher than the annual mean limit value of the European Air quality standard (EU 2017) and the World Health Organization (WHO) guideline limit for PM2.5. A clear similarity between SPM and PM2.5 particulates was declared in terms of the source apportionments as well as their origin. Both SPM and PM2.5 occurs in the atmosphere from natural sources such as dust storms and anthropogenic sources such as the burning of fossil fuels in vehicles, power plants, and various industrial processes contribute. In addition, the SPM aerosols comprise the PM2.5 fine aerosol. Therefore, the present work focuses on studying the relationship between the SPM and PM2.5 aerosols collected from industrial and residential areas in Greater Cairo during two different seasons (autumn 2014 and winter 2014/2015) simultaneously. Positive matrix factorization (PMF) analysis was conducted for the SPM and PM2.5 in order to identify and quantify the sources of the aforementioned fractions (Khan et al. 2017). The results of the analysis indicated that the major particulate matter (PM) sources in the two areas of the study (industrial and residential) were seven. These sources are heavy oil combustion, industry, sea salt, soil, construction/cement, traffic, and a Pbrelated source. The sources that are related to industrial activities present much higher mass contributions in the industrial area. The most prominent source in the residential area for both PM fractions was traffic.

Experimental Sampling Simultaneous sampling of suspended particulate matter (SPM) and fine aerosol particles with an aerodynamic diameter equal or less than 2.5 μm (PM2.5) from residential and industrial locations in Greater Cairo has been carried out during two different seasons namely autumn 2014 and winter 2014/2015. At each location, one SPM sample and one PM2.5 sample were collected per week for 24 h simultaneously. The PM2.5 aerosols had been collected on polycarbonate filters (25 mm diameter, pore size 0.4 μm, Whatman, Maidstone, UK) loaded inside a Dewell-Higgins type cyclone (Casella CEL, Bedford, UK). The SPM aerosols had been collected on membrane filters (37 mm diameter, pore size 0.45 μm, Whatman, Maidstone, UK) loaded inside a homemade cyclone. For SPM and PM2.5 aerosols, 12 samples/season/location were collected and analyzed. The industrial location represents the most polluted area in Greater Cairo which is called BHelwan district^ and it is located south of the city. A great number of industrial


activities are existing such as cement industries, iron industries, and many mining locations. The residential sampling location is close to the most famous tourist location in the city which is the pyramid whereas it is also close to the desert. More information about the sampling location can be found in our previous work (Shaltout et al. 2018a; Shaltout et al. 2018b).

Description of positive matrix factorization For the identification of the sources for the two PM fractions, positive matrix factorization (PMF) was applied using EPA PMF 5.0 software (Norris and Brown 2014). The basic equation that refers to the solution of the mass balance problem is common for all the utilized multivariate receptor models including PMF: X ¼ GF þ E

ð1Þ

where G is the source contribution matrix with p sources, and F a source profile matrix. G and F are constrained to obtain only non-negative values. Equation (1) in index notation is: p

X ij ¼ ∑ g ik f kj þ eij k¼1

ð2Þ

where Xij is the concentration of species j measured on sample i, p is the number of factors contributing to the samples, fkj is the concentration of species j in factor profile k, gik is the relative contribution of factor k to sample i, and eij is error of the PMF model for the j species measured on sample i. The Fig. 1 The ratios of mass concentrations of SPM to PM2.5 (SPM/PM2.5) collected from industrial and residential areas weekly and monthly

goal is to find the gik, fkj, and p values that best reproduce Xij. The values of gik and fkj are adjusted until a minimum value of Q for a given p is found. Q is defined as: m

n

Q¼ ∑ ∑

e2ij

2 j¼1 i¼1 sij

ð3Þ

where sij is the uncertainty of the jth species concentration in sample i, n is the number of samples, and m is the number of species. The correct determination of the number of factors is a crucial step when applying PMF.

Results and discussion The relation between mass concentrations of SPM and PM2.5 In the present work, the mass concentrations of the SPM and PM2.5 aerosols were weighed before and after sampling and the mass concentrations were calculated in micrograms per cubic meter. A detailed discussion about the mass concentrations of SPM and PM2.5 aerosols has been investigated earlier (Shaltout et al. 2018a; b). Figure 1 illustrates the ratios of mass concentrations of SPM to PM2.5 (SPM/PM2.5) collected from industrial and residential areas weekly and monthly. The standard deviation is only available during months whereas sampling was performed once per week for each location. Based on the calculation of the ratios of SPM and PM2.5 aerosols, it was found that the ratios of SPM/PM2.5 at the residential areas


vary from 7.39 to 55.46 with an average of 21.68 ± 11.79. This means that the average mass concentrations of SPM aerosols exceed the average mass concentration of PM2.5 samples by a factor of 22, which is remarkably high. In the case of industrial area, the SPM/PM2.5 ratios vary from 11.37 to 35.19 with an average value of 18.70 ± 5.93. Similarly, in the industrial area, the average mass concentration of SPM of PM2.5 samples is by a factor of 19 which is also very high and close to the average ratio found in the residential area. The SPM/ PM2.5 ratios for the samples collected from residential and industrial areas during autumn and winter seasons were calculated. A slight increase of SPM/PM2.5 ratios at the residential area was recognized during the winter season, which means the increasing rate to SPM is a little bit higher than their increasing rate for PM2.5 aerosols. On the other hand, a slight decrease of SPM/PM2.5 ratios at the industrial area was found during the autumn season, which means the increasing rate to PM2.5 is a little bit higher than their increasing rate for SPM aerosols. However, the average values of the SPM/PM2.5 ratios are 20 ± 6 and 17 ± 4 for the residential and industrial areas, respectively, and these ratios are approximately close to the SPM/PM2.5 ratios during autumn and winter at the same location. Therefore, the SPM/PM2.5 ratios seem to be constant for each location. This could be an indication for the same natural and anthropogenic sources that influence the SPM and PM2.5 simultaneously. Based on our previous work, both SPM and PM2.5 aerosols increased remarkably at industrial location regardless of the seasons. At the same time, the SPM/PM2.5 ratios at the industrial areas are a little bit lower Fig. 2 An example of SPM and PM2.5 EDXRF spectra collected from the industrial area in Greater Cairo, Egypt, using CaF2, Ge, and Mo secondary targets. For each secondary target, the respective spectra were acquired under similar experimental conditions

than those found at the residential area. This could be an indication that the anthropogenic particulates in PM2.5 aerosols are much higher than that found at the residential area. In addition, the anthropogenic particulates tend to PM2.5 aerosols more than SPM. These anthropogenic particulates originate from automobile exhausts, stationary point sources, exhaust emission from vehicles, waste burning, metal smelting, and cement industries (Hassan 2000, 2006; Hassan et al. 2017; Hassan and Khoder 2017). In addition, these differences are expected in large urban areas and have been also observed elsewhere like in the Athens metropolitan area (Eleftheriadis et al. 2014).

X-ray fluorescence spectra of SPM and PM2.5 Figure 2 illustrates an example of energy dispersive X-ray fluorescence (EDXRF) spectra for SPM and PM2.5 samples collected from the industrial location during the winter season. The represented spectra consist of three subplots, which represent EDXRF spectra coming from three different secondary targets namely CaF2, Ge, and Mo for low, medium, and high Z elements, respectively. At medium and high Z elements, Cr and W lines are originated from the contaminations either from the X-ray tube or from the instrument itself and were marked with an asterisk. The complete description of EDXRF analysis of SPM and PM2.5 samples has been investigated earlier (Shaltout et al. 2018a; b), and up to 19 elements were determined. However, the relationship between both of them did not discuss and need further investigation to be


understandable. As shown in Fig. 2, SPM samples have relatively the same pattern of PM2.5 samples and the same element could be found in both type of samples. However, the elemental concentrations in SPM samples are higher regardless of sampling time or sampling location. This was clear from Fig. 2, where the characteristic lines of SPM samples are always higher than those found in the PM2.5 samples.

The relationship between the elemental concentrations of SPM and PM2.5 Based on the elemental quantitative analysis, the SPM/PM2.5 ratios were calculated for each element at industrial and residential areas during autumn 2014 and winter 2014/2015 seasons. Figure 3 depicts the calculated ratios of SPM/PM2.5 ratios for each element. Generally, all the elements detected in SPM and PM2.5 in the industrial locations are higher than that found in the residential area during autumn and winter with the exception of Ni and V. At the industrial area, the SPM/PM2.5 ratios during autumn vary from 1.1 ± 0.1 to 5.5 ± 2.5 with an average of 3.6 ± 0.9 and vary from 1.0 ± 0.1 to 5.7 ± 6.8 with an average of 3.4 ± 0.9 during winter. In the case of residential area, the SPM/PM2.5 ratios during autumn vary from 1.01 ± 0.0 to 5.4 ± 1.4 with an average of 2.4 ± 0.5 and vary from 1.6 ± 0.3 to 55.9 ± 91.1 with an average of 5.3 ± 6.0 during winter. As illustrated in Fig. 3, the SPM/PM2.5 ratios at residential area are around 2 with the exception of the elements Ca, Pb, Rb, and Sr where the SPM/PM2.5 ratios become more than 3. Fig. 3 The ratios of SPM/PM2.5 mass concentration versus detected elements for samples collected from residential and industrial areas in Greater Cairo during autumn and winter seasons

On the other hand, the SPM/PM2.5 ratios for most of the detected elements from the industrial area are greater than 3 except Al, Ni, Ta, and Cl. The SPM/PM2.5 ratios for the elements increase as the concentration of these elements increases in the SPM samples. Moreover, the SPM/PM2.5 ratios of the elements decrease, as the elemental concentration in the PM2.5 aerosols is comparable with SPM samples as shown for most of the elements found in the residential area. The SPM/ PM2.5 ratio for Ni is close to unity, which indicates that there is no change in the Ni concentrations between the SPM and PM2.5 samples. Probably, Ni particulates could be found only in fine and ultrafine aerosols. Similarly, approximate unity ratio appeared again for V for the samples collected from the residential area during the winter season. This indicates that V is also contained in fine aerosols.

The relationship between the elemental mass concentrations of SPM and PM2.5 Looking at the relationship between the percentages of the total elemental mass concentration detected in the SPM and PM2.5 samples, Fig. 4a, b illustrates the relationship between the elemental mass concentrations in percent versus months for the aerosol samples collected from industrial and residential areas. According to the evaluated value of the total mass concentration of the detected elements in SPM samples, it was found that the percentages of the mass concentration of the detected elements in the SPM samples are 3 ± 1% and 11 ± 3% of the total mass concentration collected from the residential

Air Qual Atmos Health Fig. 4 Monthly variation of the elemental mass concentration of a industrial area, b residential area, and c, d SPM/PM2.5 percentages ratios (monthly and seasonally)

and industrial areas, respectively. In the residential areas, the percentages of the elemental mass concentrations in winter are a little bit higher than that values during the autumn. At the industrial site, the percentages of the elemental mass concentration seem to be constant and close to the average values mentioned above. Although the percentages of the elemental mass concentration at residential and industrial sites seem to be very low at SPM samples, the percentages of the elemental mass concentrations found at the industrial areas are three times higher than the values found at the residential areas. In the case of PM2.5 samples, the percentages of the mass concentration of the detected elements are 35 ± 16% and 55 ± 9% of the total mass concentration collected from the residential and industrial areas, respectively. The main elements (Si, Al, S, Cl, Ca, K, and Fe) could be responsible for high contribution especially in the industrial area. It is clear that the total mass concentration of the detected elements at PM2.5 samples represents a high fraction of the mass of the samples, which reaches to more than the half of the total mass concentration of the elements at the industrial site. In addition, the percentages of the elemental mass concentration at the industrial site are higher than the one found at the residential site. Similar to SPM samples, the percentages of the mass concentration of the detected elements at winter are higher than that found in autumn with values of 20% and 8% at

residential and industrial areas, respectively. In the case of industrial area, the mass concentrations represent more than 50% of the total mass, the percentages for PM2.5 samples, and 11% for the SPM sample. Based on the calculated ratios of SPM/PM2.5 for the percentage elemental mass concentrations, Fig. 4c, d depicts the SPM/PM2.5 ratios versus months and seasons as well as the average value. The maximum percentages of elemental mass concentrations of PM2.5 and SPM represent approximately 55% and 11%, respectively. Therefore, the ratios of SPM/PM2.5 for the percentages elemental mass concentrations at residential and industrial areas are less than unity. The lower ratios of SPM/PM2.5 at the residential area and the higher ratios of SPM/PM2.5 at industrial area confirm our hypothesis that most of the inorganic pollutants mainly exist in the fine and ultrafine particulate matters and these increase remarkably at the industrial areas. This could lead to another important hypothesis that the main anthropogenic sources in the industrial area produce fine and ultrafine aerosols.

Positive matrix factorization In order to obtain the sources that affect PM concentrations, a range of solutions was examined with a different number of factors (4–9), but seven factors were the maximum number of


factors with physical meaning. If the factors were increased, some profiles were split creating profiles with no physical meaning, while the rotational instability of the solution increased significantly. Following the procedure described elsewhere (Xie et al. 1999), the scaled residuals were also inspected. All of the scaled residuals were between − 3.0 and 3.0. The number of variables inserted to the model was 17 (PM2.5, Na, Cl, Al, Si, S, K, Ca, Ti, V, Mn, Fe, Ni, Cu, Zn, Pb, and Br). The elements defined as Bweak^ variables were Na, V, and Ni. A variable is defined to be weak if its S/N is between 0.2 and 2 (Paatero and Hopke 2003). The PM2.5 mass was set as a total variable. The number of the model runs was set to 20. A large number of different starting points were used to ensure that the Q was the global and not just a local minimum. The same settings were used for both PM fractions (separate runs), while the industrial and residential samples were combined into a single dataset, in order to produce a matrix of sufficient size to be used for PMF. The stability of the solution was tested using two of the diagnostic tools offered by PMF 5.0. Bootstrapping (BS) was set to 100 runs with a minimum correlation of 0.6. BS results were very good, revealing reproducibility > 80% for all factors in both cases. Displacement (DIS) analysis showed no factor swaps for 0.5% dQ change and very low Q decrease (< 0.1%). Because of the physical meaning of the solution and the excellent results of the other two diagnostic tests, the solution was considered robust. Finally, the model predicted and the real PM mass concentration had a very good correlation (R2 = 0.85 for the SPM and R2 = 0.98) indicating a good fit, which describes well the real contribution of the PM sources in the area. The uncertainties of the measurements were calculated according to the methodology described in (Manousakas et al. 2017). The seven sources that were identified for PM2.5 and SPM were namely industry, heavy oil, sea salt/waste burning, soil/marble and granite, construction/cement, traffic, and a Pb-rich source. The percentages of the elements in each factor are presented in (Supplementary

Fig. 5 PM2.5 source contributions in the two areas of the study

material; Tables 1 and 2). Since the residential and industrial samples were combined into a single dataset, the factor profiles for the two areas are common for both fractions. The source contributions to PM2.5 and SPM mass concentration are presented in Figs. 5 and 6. The factor that corresponds to the source identified as Bindustry^ is identified by the high abundance of Na, S, Zn, Cu, and Mn in the PM2.5 factor and Mn, Cu, and Zn in the SPM factor. The contribution of the source even though it represents the same percentage of PM2.5 mass in the two areas has a much higher mass contribution in the industrial area (2.3 μg/m3 vs 5.4 μg/m3). That indicates that the contribution of the source is more profound in the industrial area but it is still an important source in the residential area as well, as it is responsible for 9% of the PM2.5 levels in the area. Regarding the contribution, the same picture applies for the SPM fraction as well. The percentage is identical in the two locations (10%), while the mass contribution is much higher in the industrial area (1.1 μg/m3 vs 5.0 μg/m3). The mass contributions are similar in both fractions. This observation can be explained by the fact that the PM which originates from industrial processes is usually fine. In the ideal situation, the contributions would be identical for both fractions, but it is not possible to quantify the contributions with very high precision in PMF analysis, especially when a small number of species is used and for factors with low contributions (Manousakas et al. 2018). The second factor represents heavy oil combustion and it is traced by Ni, V, and S in both fractions. The percentage contribution is higher in the residential area in both cases, which might indicate use for residential heating or use in vehicles, but the mass contribution is similar in both sites. This is an important finding for selecting air pollution mitigation measures in the city of Cairo. The high contribution for this source can be reduced to the levels found elsewhere in the East Mediterranean (Diapouli et al. 2017) if cleaner fuels like diesel are used. The factor that represents sea salt and waste burning


Fig. 6 SPM source contributions in the two areas of the study

is traced by Na, Cl, Br, and Zn for both fractions. The contribution of this source is much higher in the industrial site, indicating that waste burning and incinerating processes indeed affect the source. The reason that this factor comes as a combination of two sources might be the common tracers of the two sources such as Cl and Br. Soil/marble and granite source is traced by Al, Si, K, Ti, and Fe. Similarly, in this case, the factor is a combination of two different sources because of the common tracers those two sources share. As expected, the contribution is much higher in the industrial area for both fractions. This source has the highest contribution to the industrial area for PM2.5. The source that represents emissions from construction activities and cement factories is traced by the high loadings of Ca in the factor for both fractions. As it is the case with the other industrial related activities, the contribution for both fractions is much higher in the industrial area. As it appears, the residential area is not very affected by this source, while it provides the highest contribution in the industrial area for the SPM. Traffic is traced by Ni, Cu, and Zn. This source represents the highest contribution in the residential areas for both fractions. Even though the relative contribution of this source is lower in the industrial area, it holds a large share of the PM contribution for both fractions. The last factor is the source that is traced by the high loadings of Pb. The processes that can be related to this source are heavy oil combustion in cement and brick factories and emissions from iron and steel factories. The relative contribution is similar to the two areas for both fractions, with the mass concentration being much higher in the industrial area. The higher relative contributions in the PM2.5 fraction indicate that this source affects this fraction more than SPM.

(0.199 ± 0.03) seems to be twice that value found at the residential area (0.11 ± 0.04). The lower ratios of SPM/ PM2.5 at the residential area and the higher ratios of SPM/PM2.5 at industrial area confirm our hypothesis that the most of the inorganic pollutants are emitted in the fine and ultrafine particulate matters and it increases remarkably at the industrial areas. Depending on the total elemental mass concentration, the average elemental mass concentrations represent only 3% and 11% elements for SPM samples collected from residential and industrial areas, respectively. Source apportionment by PMF confirmed some of the aforementioned observations. The number and the type of the sources were the same for both PM fractions and areas. The seven identified PM sources were namely heavy oil combustion, industry, sea salt, soil, construction/cement, traffic, and a Pb-related source. The sources that are related to industrial activities present much higher mass contributions in the industrial area. The most prominent PM2.5 sources in the residential area were traffic and soil in the industrial, while in the SPM fraction, they were traffic in the residential area and sea salt/waste burning in the industrial. The study revealed the importance of source apportionment studies in sensitive areas, from the environmental point of view, such as Greater Cairo area, in that a number of different anthropogenic and natural sources contribute to the exceptionally increased PM concentration levels. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References Conclusions Based on the total mass concentration, the average value the SPM/PM2.5 percentage ratios at the industrial area

Abd El Maksoud AS (2011) Study of air pollution in the industrial area of Shaq El-Teeban in Cairo. PhD Thesis, Cairo University Abu-Allaban M, Gertler AW, Lowenthal DH (2002) A preliminary apportionment of the sources of ambient PM10, PM2.5, and VOCs in

Air Qual Atmos Health Cairo. Atmos Environ 36:5549–5557. https://doi.org/10.1016/ s1352-2310(02)00662-3 Abu-Allaban M, Lowenthal DH, Gertler AW, Labib M (2007) Sources of PM (10) and PM (2.5) in Cairo’s ambient air. Environ Monit Assess 133:417–425. https://doi.org/10.1007/s10661-006-9596-8 Abu-Allaban M, Lowenthal DH, Gertler AW, Labib M (2009) Sources of volatile organic compounds in Cairo’s ambient air. Environ Monit Assess 157:179–189. https://doi.org/10.1007/s10661-008-0526-9 Anwar WA (2003) Environmental health in Egypt. Int J Hyg Environ Health 206:339–350. https://doi.org/10.1078/1438-4639-00230 Athanasopoulou E, Tombrou M, Russell AG, Karanasiou A, Eleftheriadis K, Dandou A (2010) Implementation of road and soil dust emission parameterizations in the aerosol model CAMx: applications over the greater Athens urban area affected by natural sources. J Geophys Res-Atmos 115: D17301. https://doi.org/10.1029/2009JD013207 Diapouli E, Manousakas M, Vratolis S, Vasilatou V, Maggos T, Saraga D, Grigoratos T, Argyropoulos G, Voutsa D, Samara C, Eleftheriadis K (2017) Evolution of air pollution source contributions over one decade, derived by PM10 and PM2.5 source apportionment in two metropolitan urban areas in Greece. Atmos Environ 164:416–430. https://doi.org/10.1016/j.atmosenv.2017.06.016 Đorđević D, Vukmirović Z, Tošić I, Unkašević M (2004) Contribution of dust transport and resuspension to particulate matter levels in the Mediterranean atmosphere. Atmos Environ 38:3637–3645. https:// doi.org/10.1016/j.atmosenv.2004.04.001 Eleftheriadis K, Ochsenkuhn KM, Lymperopoulou T, Karanasiou A, Razos P, Ochsenkuhn-Petropoulou M (2014) Influence of local and regional sources on the observed spatial and temporal variability of size resolved atmospheric aerosol mass concentrations and watersoluble species in the Athens metropolitan area. Atmos Environ 97: 252–261. https://doi.org/10.1016/j.atmosenv.2014.08.013 Englert N (2004) Fine particles and human health—a review of epidemiological studies. Toxicol Lett 149:235–242. https://doi.org/10.1016/ j.toxlet.2003.12.035 EU (2017) Air quality standards. European Commission of air quality. http://ec.europa.eu/environment/air/quality/standards.htm. Accessed in 07/11/2018 Goldberg MS, Burnett RT, Stieb DM, Brophy JM, Daskalopoulou SS, Valois M-F, Brook JR (2013) Associations between ambient air pollution and daily mortality among elderly persons in Montreal, Quebec. Sci Total Environ 463-464:931–942. https://doi.org/10. 1016/j.scitotenv.2013.06.095 Green LC, Armstrong SR (2003) Particulate matter in ambient air and mortality: toxicologic perspectives. Regul Toxicol Pharmacol 38: 326–335. https://doi.org/10.1016/S0273-2300(03)00099-0 Hassan SK (2000) A study on indoor air quality in Greater Cairo vol M.Sc. Master Thesis, Cairo University, Cairo, Egypt Hassan SK (2006) Atmospheric polycyclic aromatic hydrocarbons and some heavy metals in suspended particulate matter in urban, industrial and residential areas in Greater Cairo. PhD Thesis, Ain Shams University Hassan SK, Khoder MI (2017) Chemical characteristics of atmospheric PM2.5 loads during air pollution episodes in Giza, Egypt. Atmos Environ 150:346–355. https://doi.org/10.1016/j. atmosenv.2016.11.026 Hassan SK, Abozied AM, Shaltout AA, Boman J (2017) Elemental composition of fine aerosol in the atmosphere of industrial area in Greater Cairo, Egypt X-Ray Spectrometry submitted Hindy KT, Farag SA, El-Taieb NM (1990) Monthly and seasonal trends of total suspended particulate matter and smoke concentration in industrial and residential areas in Cairo. Atmos Environ B 24:343– 353. https://doi.org/10.1016/0957-1272(90)90041-R Khan MF, Hwa SW, Hou LC, Mustaffa NIH, Amil N, Mohamad N, Sahani M, Jaafar SA, Nadzir MSM, Latif MT (2017) Influences of inorganic and polycyclic aromatic hydrocarbons on the sources of PM2.5 in the Southeast Asian urban sites air quality. Atmosphere & Health 10:999–1013. https://doi.org/10.1007/s11869-017-0489-5

Khoder MI (1997) Effect of weather condition on the chemical constituent of total suspended particulate. PhD thesis, Cairo University Khoder MI (2007) Ambient levels of volatile organic compounds in the atmosphere of Greater Cairo. Atmos Environ 41:554–566. https:// doi.org/10.1016/j.atmosenv.2006.08.051 Lazaridis M, Spyridaki A, Solberg S, Smolík J, Zdímal V, Eleftheriadis K, Aleksanropoulou V, Hov O, Georgopoulos PG (2005) Mesoscale modeling of combined aerosol and photo-oxidant processes in the Eastern Mediterranean. Atmos Chem Phys 5:927–940 Manousakas M, Papaefthymiou H, Diapouli E, Migliori A, Karydas AG, Bogdanovic-Radovic I, Eleftheriadis K (2017) Assessment of PM2.5 sources and their corresponding level of uncertainty in a coastal urban area using EPA PMF 5.0 enhanced diagnostics. Sci Total Environ 574:155–164. https://doi.org/10.1016/j.scitotenv. 2016.09.047 Manousakas M, Diapouli E, Papaefthymiou H, Kantarelou V, Zarkadas C, Kalogridis AC, Karydas AG, Eleftheriadis K (2018) XRF characterization and source apportionment of PM10 samples collected in a coastal city. X-Ray Spectrom 47:190–200. https://doi.org/10.1002/ xrs.2817 Norris G, Brown S (2014) EPA positive matrix factorization (PMF) 5.0 Fundamentals and User Guide, 5.0 edn Ostro B, Tobias A, Karanasiou A, Samoli E, Querol X, Rodopoulou S, Basagaña X, Eleftheriadis K, Diapouli E, Vratolis S, Jacquemin B, Katsouyanni K, Sunyer J, Forastiere F, Stafoggia M, MEDPARTICLES Study Group (2015) The risks of acute exposure to black carbon in southern Europe: results from the MEDPARTICLES project. Occup Environ Med 72:123–129. https://doi. org/10.1136/oemed-2014-102184 Paatero P, Hopke PK (2003) Discarding or downweighting high-noise variables in factor analytic models. Anal Chim Acta 490:277–289. https://doi.org/10.1016/S0003-2670(02)01643-4 Saleh EA (2002) Air quality and effects of air pollution on materials of different structures in Cairo atmosphere, 2002, Ph.D. thesis, Faculty of Science, Ain Shams University, Cairo, Egypt., PhD Thesis, Ain Sham University Saliba NA, El Jam F, El Tayar G, Obeid W, Roumie M (2010) Origin and variability of particulate matter (PM10 and PM2.5) mass concentrations over an Eastern Mediterranean city. Atmos Res 97:106–114. https://doi.org/10.1016/j.atmosres.2010.03.011 Shakour A, El-Shahat M, El-Taieb N, Hassanein M, Mohamed A (2011) Health impacts of particulate matter in Greater Cairo, Egypt. J Am Sci 7:840–848 Shaltout AA, Boman J, Al-Malawi D-aR, Shehadeh ZF (2013) Elemental composition of PM2. 5 particles sampled in industrial and residential areas of Taif, Saudi Arabia. Aerosol Air Qual Res 13:1356–1364 Shaltout AA, Boman J, Welz B, Castilho INB, Al Ashkar EA, Gaita SM (2014) Method development for the determination of Cd, Cu, Ni and Pb in PM 2.5 particles sampled in industrial and urban areas of Greater Cairo, Egypt, using high-resolution continuum source graphite furnace atomic absorption spectrometry. Microchem J 113:4–9 Shaltout AA, Boman J, Shehadeh ZF, Al-Malawi DR, Hemeda OM, Morsy MM (2015) Spectroscopic investigation of PM 2.5 collected at industrial, residential and traffic sites in Taif, Saudi Arabia. J Aerosol Sci 79:97–108 Shaltout AA, Boman J, Alsulimane ME (2017) Identification of elemental composition of PM2.5collected in Makkah, Saudi Arabia, using EDXRF. X-Ray Spectrom 46:151–163. https:// doi.org/10.1002/xrs.2732 Shaltout AA, Hassan SK, Karydas AG, Harfouche M, Abd-Elkader OH, Kregsamer P, Wobrauschek P, Streli C (2018a) EDXRF analysis of suspended particulate matter (SPM) from residential and industrial areas in Cairo, Egypt. X-Ray Spectrom 47:223–230. https://doi.org/ 10.1002/xrs.2830

Air Qual Atmos Health Shaltout AA, Hassan SK, Karydas AG, Zaki ZI, Mostafa NY, Kregsamer P, Wobrauschek P, Streli C (2018b) Comparative elemental analysis of fine particulate matter (PM 2.5 ) from industrial and residential areas in Greater Cairo-Egypt by means of a multi-secondary target energy dispersive X-ray fluorescence spectrometer. Spectrochim Acta B At Spectrosc 145:29–35. https://doi.org/10.1016/j.sab. 2018.04.003 Solomon PA (2012) Introduction: special issue of air quality, atmosphere and health for air pollution and health: bridging the gap from sources-to-health outcomes. Air Qual Atmos Health 5:3–8. https:// doi.org/10.1007/s11869-011-0165-0 Stafoggia M, Schneider A, Cyrys J, Samoli E, Andersen ZJ, Bedada GB, Bellander T, Cattani G, Eleftheriadis K, Faustini A, Hoffmann B, Jacquemin B, Katsouyanni K, Massling A, Pekkanen J, Perez N,

Peters A, Quass U, Yli-Tuomi T, Forastiere F, UF&HEALTH Study Group (2017) Association between short-term exposure to ultrafine particles and mortality in eight European urban areas. Epidemiology (Cambridge, Mass) 28:172–180. https://doi.org/10.1097/ede. 0000000000000599 Terrouche A, Ali-Khodja H, Kemmouche A, Bouziane M, Derradji A, Charron A (2016) Identification of sources of atmospheric particulate matter and trace metals in Constantine, Algeria. Air Qual Atmos Health 9:69–82. https://doi.org/10.1007/s11869-014-0308-1 Xie Y-L, Hopke PK, Paatero P, Barrie LA, Li S-M (1999) Identification of source nature and seasonal variations of arctic aerosol by positive matrix factorization. J Atmos Sci 56:249–260. https://doi.org/10. 1175/1520-0469(1999)0562.0.co;2