Regional Inhaled Deposited Dose of Indoor Combustion-Generated Aerosols in Jordanian Urban Homes

Indoor combustion processes associated with cooking, heating, and smoking are a major source of aerosols in Jordanian dwellings. To evaluate human exposure to combustion-generated aerosols in Jordanian indoor environments, regional inhaled deposited dose rates of indoor aerosols (10 nm to 25 μm) were determined for different scenarios for adult occupants. The inhaled deposited dose rate provides an estimate of the number or mass of inhaled aerosol that deposits in each region of the respiratory system per unit time. In general, sub-micron particle number (PN1) dose rates ranged from 109 to 1012 particles/h, fine particle mass (PM2.5) dose rates ranged from 3 to 216 μg/h, and coarse particle mass (PM10) dose rates ranged from 30 to 1600 μg/h. Dose rates were found to be dependent on the type and intensity of indoor combustion processes documented in the home. Dose rates were highest during cooking activities using a natural gas stove, heating via natural gas and kerosene, and smoking (shisha/tobacco). The relative fraction of the total dose rate received in the head airways, tracheobronchial, and alveolar regions varied among the documented indoor combustion (and non-combustion) activities. The significant fraction of sub-100 nm particles produced during the indoor combustion processes resulted in high particle number dose rates for the alveolar region. Suggested approaches for reducing indoor aerosol dose rates in Jordanian dwellings include a reduction in the prevalence of indoor combustion sources, use of extraction hoods to remove combustion products, and improved ventilation/filtration in residential buildings.


Introduction
Inhalation exposure to indoor air pollution is a significant factor affecting human respiratory and cardiovascular health. As people spend the majority of their time in indoor environments, it is important to conduct comprehensive assessments of indoor air quality and exposure [1][2][3]. This is especially needed in many urban areas of the world that lack reliable indoor air pollution data for evaluation of human health outcomes. In general, particulate and gaseous indoor air pollutant concentrations depend on the dynamic relationship between pollutant source and loss processes

Inhaled Deposited Dose Rate of Indoor Combustion-Generated Aerosols
The ICRP and MPPD models divide the respiratory tract into three regions: head/throat, tracheobronchial (TB), and pulmonary/alveolar (P/Alv). Following our previous methodology as described by Hussein et al. [7,19,20], we can calculate the regional inhaled deposited dose for a specific particle diameter range (D p1 -D p2 ) during a one-hour exposure period as a dose rate: where V E is the minute ventilation (volume of air breathed as reported by Holmes [21], Table S1), DF(D p ) is the aerosol deposition fraction in a particular region of the respiratory tract ( Figure S1 as reported by Löndahl et al. [22]), n N 0 (D p ) (particles/cm 3 ) is the particle number size distribution (i.e., dN/dlog(D p )), and f is a metric conversion for the aerosol concentration (i.e., it is 1 for particle number and for particle mass = ρ p D p 3 π/6, where ρ p is the particle effective density). The deposition fraction (DF) and the particle number size distribution (n) are functions of particle diameter (D p ). Dose rates were calculated for adult male and female occupants reflecting different activity levels (resting, exercising, housework) and exposures to different indoor combustion sources (heating, cooking, smoking). The combination of subjects, activities, and combustion processes reflect common exposure scenarios in Jordanian dwellings during the winter. The indoor aerosol exposure assessment was adopted from our prior observations reported for eight homes in Amman, Jordan.

Indoor Aerosol Concentrations and Size Distributions in Jordanian Urban Homes
Particle number size distributions were adopted from a previous field measurement campaign reported by Hussein et al. [23] for eight homes from 23 December 2018 to 12 January 2019 (i.e., winter season) in Amman, Jordan ( Figure S2, Table S2). Indoor aerosol measurements were made using two portable condensation particle counters (CPC 3007 and P-Trak 8525, TSI Inc., Minnesota, USA) and a handheld optical particle counter (AeroTrak 9306-V2, TSI Inc., Minnesota, USA); see Table S3. The use of this combination of portable aerosol instrumentation provides a basis to derive the particle number size distribution (0.01-25 µm) with 8 bins or size fractions [20,[23][24][25][26][27][28][29][30][31]. The particle effective density is needed when converting a particle number size distribution to a particle mass size distribution. Here, we assumed the effective density to be similar to what was reported for urban air in Asian cities [32] as we are investigating exposure to aerosols originating from fossil fuel combustion, i.e., natural gas heaters and stoves. The calculation of size-fractionated particle number and mass concentrations (i.e., PN 1 , PM 2.5 , and PM 10 ) are described in the Supplementary Material.

Indoor Aerosol Concentrations during Background Periods
Investigating aerosol concentrations during periods without indoor activities (i.e., no active indoor emission sources) is necessary to benchmark the background conditions in each dwelling. We identified such periods and calculated the mean, median, and quartiles for total particle number (PN) and mass (PM 2.5 and PM 10 ) concentrations (Tables 1 and 2 and Figure 1). Each of these periods was characterized during the nighttime when occupants were asleep. For background PN concentrations (Table 1), the lowest was found in home apartment A1 (mean 4000 ± 300 cm −3 ) and the highest in ground floor apartment GFA2 (mean 15,800 ± 5800 cm −3 ). The corresponding PM 2.5 and PM 10 concentrations were the highest in home H2 (approximately 21.7 ± 21.2 µg/m 3 and 82.4 ± 90.8 µg/m 3 , respectively) and the lowest in A1 (approximately 4.1 ± 0.1 µg/m 3 and 5.3 ± 0.5 µg/m 3 , respectively) ( Table 2). In general, indoor aerosol concentrations during these periods reflect outdoor concentrations as the primary indoor aerosol source is infiltration via indoor-outdoor air exchange.   Variations in the magnitude of the mean indoor particle number size distributions during background conditions were observed among the eight homes ( Figure 2). This is due in part to variations in outdoor particle size distributions at each site, infiltration and ventilation rates, and prevalence of indoor emission sources prior to the background periods during the nighttime. The mean particle number size distributions generally exhibit similarity in shape, with pronounced nucleation and Aitken modes. Ultrafine particles (D p ≤ 0.1 µm) dominate the number size distributions for each home. Homes duplex D1, house H1, and GFA2 were associated with the highest sub-micron particle concentrations, likely due to indoor smoking activities that occurred prior to the background periods. Accumulation mode aerosols emitted during smoking typically have a low deposition rate to indoor surfaces, and thus, have a long residence time in indoor air.

Indoor Activity Categories for Inhaled Deposited Dose Analysis
We selected the following activities for the inhaled deposited dose analysis: heating (kerosene, natural gas, central heating system, air conditioning split unit (AC)), cooking (microwave, water heater jug, combustion using natural gas stove), and smoking (shisha, tobacco). The activity-specific aerosol concentrations are listed in Table S5 and the mean particle number size distributions categorized per event and home are presented in Figure S3. The events are categorized as follows:  TYPE I: Non-combustion cooking activities (i.e., microwave, water heater jug).  TYPE II: Intensive cooking activities by combustion (i.e., natural gas stove) combined with noncombustion heating (central or AC).  TYPE III: Cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas).  TYPE IV: Cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas) and smoking (shisha and/or tobacco).
The overall mean size-fractionated particle number concentrations for each category and background conditions are listed in Tables S6-S10 and the particle concentrations are presented in Figure 3 and Table 3. The corresponding particle number size distributions are presented in Figure 4. Particle concentrations were the lowest for TYPE I indoor activities (Table S7 and Table 3), which did not include any combustion processes. The overall mean PN concentrations were approximately 16,400 ± 17,200 cm −3 and the corresponding PM2.5 and PM10 concentrations were 16 ± 11 µg/m 3 and 82 ± 53 µ g/m 3 , respectively. This was slightly higher than the overall mean background concentrations for all homes with PN = 9900 ± 4900 cm −3 , PM2.5 = 13 ± 15 µ g/m 3 , and PM10 = 33 ± 53 µ g/m 3 .

Indoor Activity Categories for Inhaled Deposited Dose Analysis
We selected the following activities for the inhaled deposited dose analysis: heating (kerosene, natural gas, central heating system, air conditioning split unit (AC)), cooking (microwave, water heater jug, combustion using natural gas stove), and smoking (shisha, tobacco). The activity-specific aerosol concentrations are listed in Table S5 and the mean particle number size distributions categorized per event and home are presented in Figure S3. The events are categorized as follows: • TYPE I: Non-combustion cooking activities (i.e., microwave, water heater jug). • TYPE II: Intensive cooking activities by combustion (i.e., natural gas stove) combined with non-combustion heating (central or AC). • TYPE III: Cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas). • TYPE IV: Cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas) and smoking (shisha and/or tobacco).
The overall mean size-fractionated particle number concentrations for each category and background conditions are listed in Tables S6-S10 and the particle concentrations are presented in Figure 3 and Table 3. The corresponding particle number size distributions are presented in Figure 4. Particle concentrations were the lowest for TYPE I indoor activities (Table S7 and Table 3), which did not include any combustion processes. The overall mean PN concentrations were approximately Atmosphere 2020, 11, 1150 6 of 17 16,400 ± 17,200 cm −3 and the corresponding PM 2.5 and PM 10 concentrations were 16 ± 11 µg/m 3 and 82 ± 53 µg/m 3 , respectively. This was slightly higher than the overall mean background concentrations for all homes with PN = 9900 ± 4900 cm −3 , PM 2.5 = 13 ± 15 µg/m 3 , and PM 10 = 33 ± 53 µg/m 3 .    Cooking was reported on either a stove (natural gas) or using non-combustion appliances (i.e., water jug heater, microwave, etc.); the cooking intensity was indicated. The legend refers to the home ID and indoor activities.

Inhaled Deposited Dose Scenarios
The primary goal of this study was to quantify the regional deposited dose rate of combustiongenerated aerosols in the respiratory tract for exposure during four indoor activity types (TYPEs I, II, III, IV) and background conditions. Such analyses have yet to be made for indoor environments in the Middle East, which include a mixture of western and eastern living styles with respect to heating, cooking, and other indoor activities. The dose rate calculations were made for the following scenarios:

Inhaled Deposited Dose Scenarios
The primary goal of this study was to quantify the regional deposited dose rate of combustion-generated aerosols in the respiratory tract for exposure during four indoor activity types (TYPEs I, II, III, IV) and background conditions. Such analyses have yet to be made for indoor environments in the Middle East, which include a mixture of western and eastern living styles with respect to heating, cooking, and other indoor activities. The dose rate calculations were made for the following scenarios: • Housework activities: effort equivalent to yardwork.

•
Moving activities: effort corresponding to running at 8 km/h and walking at 4 km/h.

•
Resting activities: standing and sitting.
3.2.1. Regional Inhaled Deposited Dose Rates for Background Condition Scenarios The regional inhaled deposited dose rate calculated based on exposure to mean sub-micron particle number concentrations (i.e., PN 1 ) was the highest in the alveolar region and the lowest in the head airways during indoor background conditions (Table 4 and Figure 5). The total PN 1 dose rate was in the range of 2.9 × 10 9 -1.9 × 10 10 particles/h for an adult male and 2.3 × 10 9 -1.7 × 10 10 particles/h for an adult female. The highest dose rate was received during running (i.e., indoor exercising) and the lowest during sitting due to a higher minute ventilation at increased effort for the former. Approximately 75% of the total PN 1 dose rate was received in the alveolar region and approximately 7.5% was received in the head region for adult males performing running, walking, and working activities. As for standing and sitting, adult males received approximately 62% of the total PN 1 dose rate in the alveolar region and approximately 14% in the head region. When compared to an adult male, an adult female received a slightly lower PN 1 dose rate fraction in the alveolar region (about 73%) and a slightly higher dose rate fraction in the head region (about 8%) during running, walking, and working activities. Adult females received 53% of the total PN 1 dose rate in the alveolar region and 16% in the head region for standing and sitting. Table 4. Mean regional dose rates for aerosol exposure during indoor background conditions. Note that yardwork is assumed to be equivalent to housework and running is equivalent to indoor exercising.

Scenario
Adult   5 and PM10). The color legend is: blue-head airways (head), redtracheobronchial (TB), and gray-alveolar (Alv). Exposure is based on mean concentrations. Note that yardwork is assumed to be equivalent to housework and running is equivalent to indoor exercising. Regional inhaled deposited dose rates calculated for different activities during indoor background conditions for: (a) sub-micron particle number concentrations (PN 1 ) and (b,c) particle mass concentrations (PM 2.5 and PM 10 ). The color legend is: blue-head airways (head), red-tracheobronchial (TB), and gray-alveolar (Alv). Exposure is based on mean concentrations. Note that yardwork is assumed to be equivalent to housework and running is equivalent to indoor exercising.
The regional dose rate for mean fine particle mass (i.e., PM 2.5 ) followed a similar pattern as that for sub-micron particle number (i.e., PN 1 ). For example, the PM 2.5 dose rate was the highest in the alveolar region (range: 1.3-9 µg/h for adult males and 0.9-7.7 µg/h for adult females) and the lowest in the head region (range: 0.4-1 µg/h for adult males and 0.3-0.8 µg/h for adult females). The highest PM 2.5 dose rate was received during running and the lowest during sitting (Table 4 and Figure 5).
An adult male performing running, walking, and working activities would receive approximately 75% of the total PM 2.5 dose rate in the alveolar region and about 8% in the head region. As for standing and sitting, adult males received approximately 50% of the total PM 2.5 dose rate in the alveolar region and about 38% in the head region. When compared to an adult male, an adult female received rather similar fractions in the head region. However, the fractions were slightly lower for the alveolar region (about 46% during standing and sitting and about 74% during running, walking, and working).
The PM 10 regional dose rate pattern was different than those for PN 1 and PM 2.5 . However, the pattern was similar for adult males and females. For example, the PM 10 dose rate fraction was approximately 35% in the head region during yardwork, running, and walking and it was about 68% during standing and sitting. The corresponding fraction in the alveolar region was about 32% and 22%, respectively. The total PM 10 dose rate in the head region was in the range of 8-24 µg/h for adult males and 6-21 µg/h for adult females. As for the alveolar region, the dose rate was in the range of 3-22 µg/h for adult males and 2-20 µg/h for adult females (Table 4 and Figure 5). The higher dose rate in the head airways for particles larger than 2.5 µm is explained by a higher deposition efficiency via impaction and gravitational settling for large particles.

Regional Inhaled Deposited Dose Rates for TYPE I Scenarios
This category of indoor activities includes indoor aerosol emissions during non-combustion processes. In general, the dose rates received during TYPE I indoor activities were 1.6-, 1.4-, and 2.9-fold higher than what was received during indoor background conditions; respectively for PN 1 , PM 2.5 , and PM 10 . Adult females received lower aerosol dose rates than adult males, primarily due to a lower minute ventilation (Table 5 and Figure 6). Table 5. Mean regional dose rates (head, tracheobronchial (TB), alveolar (Alv), and total) for aerosol exposure during TYPE I indoor activities. Note that yardwork is assumed to be equivalent to housework and running is equivalent to indoor exercising.

Scenario
Adult was expected because during TYPE I scenarios, concentrations of coarse mode particles increased in part due to human movement-driven settled dust resuspension. The total PM10 dose rate was in the range of 35-210 µ g/h and 27-183 µ g/h; respectively for adult males and females. For adult males, the dose rate received was 26-87 µ g/h in the head region and 6-50 µ g/h in the alveolar region; with lower values during sitting and higher values during indoor exercising. For adult females, the corresponding values in the head and alveolar regions were 20-78 µ g/h and 4-44 µ g/h, respectively. Figure 6. Regional inhaled deposited dose rates calculated for each activity type (TYPEs I, II, III, IV) and background conditions for: (a) sub-micron particle number concentrations (PN1) and (b,c) particle mass concentrations (PM2.5 and PM10). The color legend is: blue-yardwork equivalent activities, yellow-walking activities, and red-sitting and resting.  Figure 6. Regional inhaled deposited dose rates calculated for each activity type (TYPEs I, II, III, IV) and background conditions for: (a) sub-micron particle number concentrations (PN 1 ) and (b,c) particle mass concentrations (PM 2.5 and PM 10 ). The color legend is: blue-yardwork equivalent activities, yellow-walking activities, and red-sitting and resting.

Regional Inhaled
PN 1 and PM 2.5 regional deposition patterns (fractions in each respiratory region) were similar as that received during indoor background conditions. The total PN 1 dose rate received in the head region was in the range of 4.6 × 10 9 -3.1 × 10 10 particles/h for an adult male and 3.6 × 10 9 -2.7 × 10 10 particles/h for an adult female. The corresponding total PM 2.5 mass dose rate was in the range of 4-17 µg/h and 3-15 µg/h; respectively for adult males and females. In contrast to PN 1 and PM 2.5 , the PM 10 deposition pattern was different from that received during background conditions. Specifically, the PM 10 dose rate fraction was approximately 42% in the head region during yardwork, running, and walking and it was about 75% during standing and sitting. The corresponding fraction in the alveolar region was approximately 24% and 16%, respectively. The change in the PM 10 dose pattern was expected because during TYPE I scenarios, concentrations of coarse mode particles increased in part due to human movement-driven settled dust resuspension. The total PM 10  For these categories of indoor activities, fine particle number concentrations were much higher than what was observed during the background conditions and TYPE I activities (Table 3 and Figures 3  and 4). This was reflected in the dose rate calculations. For instance, the dose rates received during TYPE II indoor activities were 7.9-, 10.2-, and 12.9-fold higher than what was received during indoor background conditions (Table 6 and Figure 6) respectively for PN 1 , PM 2.5 , and PM 10 . As for TYPE III indoor activities, the corresponding ratios were 22.6, 14.6, and 32.9 (Table 7 and Figure 6), respectively.
Smoking activities (shisha, tobacco) were associated with ratios of 21.3, 17.5, and 46.6 ( Table 8 and Figure 6), respectively. Adult females received lower dose rates in all regions and scenarios as compared to adult males. Interestingly, the impact of combustion processes was more pronounced for the PM 10 dose rates than for the PN 1 and PM 2.5 dose rates, as determined from the ratios relative to background scenarios. Table 6. Mean regional dose rates (head, tracheobronchial (TB), alveolar (Alv), and total) for aerosol exposure during TYPE II indoor activities. Note that yardwork is assumed to be equivalent to housework and running is equivalent to indoor exercising.

Regional Inhaled Deposited Dose Rates Based on Median Particle Number Size Distributions
Measured indoor particle concentrations were not normally distributed and had periods with high peak values (see Tables S7-S10 in the Supplementary Material). Thus, the mean concentration values were typically higher than median values for the exposure scenarios. Since air quality guidelines are based on mean values, most of our dose rate analysis is based on mean particle number and mass concentrations. Mean values are also the relevant measure for calculation of the inhaled aerosol dose over time. Nevertheless, it is important to compare the mean and the median particle concentrations in order to assess the influence of concentration peaks on the overall dose rate analysis. As presented in Tables 1 and 2 (also Figure 1) for background concentrations inside each dwelling individually, the percentage difference between the mean and the median values for the number concentrations was between 1.5% and 3.8% for the first six dwellings; only dwellings GFA2 and GFA3 had large difference between the mean and the median (23% and 25%, respectively). As for PM 2.5 , the difference between the mean and the median values was less than 9% for all dwellings, aside from the fifth and eighth dwellings (H2 and GFA3). For PM 10 , the difference between the mean and the median values was less than 7% for all dwellings, aside from the fifth and seventh dwellings (H2 and GFA2).
The particle size distributions used in the dose rate calculations (i.e., for all scenarios including background conditions) were combined from the data obtained across all eight dwellings. Therefore, differences between the mean and the median concentrations were more significant (Table 3 and Figure 3). For instance, the difference in the number concentrations was between 3% and 27%. The differences in PM 2.5 and PM 10 ranged from 17%-41% and 5%-61%, respectively. Thus, the inhaled dose rates were recalculated based on the median particle size distributions and reported in the Supplementary Material (Section S5). The calculated dose rates were generally lower when using the median values. Considering the total inhaled deposited dose rate for PN 1 for adult males and females, it was lower by approximately 9%, 28%, 21%, 8%, and 14%, respectively during background conditions and TYPEs I-IV scenarios. The corresponding difference in PM 2.5 dose rates was lower by 44%, 21%, 18%, 14%, and 23%, respectively. The largest variation in the difference was found in the PM 10 dose rates as 69%, 16%, 0%, 41%, and 10%, respectively. Periods with very high particle concentrations can therefore have a significant impact on the estimated dose rates.

Conclusions
In this study, regional inhaled deposited dose rates of indoor combustion-generated aerosols were evaluated based on mean indoor particle number size distributions (0.01-25 µm) measured in Jordanian dwellings. Dose rates were also calculated in terms of PM 2.5 and PM 10 based on mean particle mass size distributions, which were estimated from the particle number size distributions. An important outcome of this investigation is extending dose rate calculations to common exposure scenarios inside Jordanian dwellings. Exposure was classified according to four activity types: TYPE I for non-combustion cooking activities (i.e., microwave), TYPE II for intensive cooking activities by combustion (i.e., natural gas stove) combined with non-combustion heating (central or AC), TYPE III for cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas), and TYPE IV for cooking activities by combustion (i.e., natural gas stove) combined with combustion heating (kerosene or natural gas) and smoking (shisha, tobacco). The activities were classified into three main categories: yardwork equivalent activities, moving activities (running at 8 km/h and walking at 4 km/h), and resting activities (standing and sitting).
The indoor aerosol dose rate calculations were based on: (1) characteristics of particle number size distributions, (2) activity type (exercise versus rest), (3) gender, and (4) particle concentration metric (number versus mass) and the particle diameter range (PN 1 , PM 2.5 , PM 10 ). Regardless of gender, the PM 10 dose rate fraction during rest was mostly in the head airways (~70%) and the least in the