# Measurement and prediction of the effect of students' activities on airborne particulate concentration in a classroom.

IntroductionA number of studies have reported that exposure to high concentrations of particles are associated with adverse health effects (Pope et al. 1995; Annesi-Maesano et al. 2007). In particular, fine particles (e.g., [PM.sub.1] and [PM.sub.2.5]) are known to have a bigger health impact (Owen et al. 1992), and these particles have the ability to reach deeper parts of the respiratory tract (Berico et al. 1997). Some previous research has also considered the risk resulting from short-term exposure to airborne contaminants (Delfino et al. 1998; Gold et al. 2000). It has been reported that exposure to a level of 200 [micro]g [m.sup.-3] of fine particles for 2 h could influence the respiratory system (WHO 2006).

The number of hours people spend indoors is much higher than those they spend outdoors, and people can spend about 85-90% of their time in various indoor environments. Recently, much consideration has been invested in examining air quality in environments such as homes (e.g., Massey et al. 2009), offices (e.g., Zuraimi and Tham, 2009), classrooms (e.g., Awbi and Pay, 1995), hospitals (e.g., Wang et al. 2006), and vehicles (e.g., Song et al. 2009).

The classroom environment is very important for students' health and performance (Bako-Biro et al. 2008). The air quality in classrooms is particularly important due to the large amount of time teachers and students spend within them (Hoppe 2002). It has been estimated that more than 50% of students have some kind of allergy or asthma, and therefore, there is a great need to consider ventilation in schools (Karimipanah et al. 2007). Nevertheless, there is little information concerning pollutants concentration in these premises, especially in adult education rooms.

Previous studies have shown that resuspension of residential dust particles due to human activity increases the level of indoor particle concentrations (Thatcher and Layton 1995; Qian et al. 2008; Raja et al. 2010). However, the results of particle resuspension were not conclusive, and most publications have recommended further research to quantify this factor within the indoor microenvironment. To the best knowledge of the authors, no results of particle resuspension in classrooms have been published.

In this article, the particle concentration in a university classroom was assessed experimentally with different occupancy periods (occupied and unoccupied) using Grimm aerosol spectrometers. The objectives of this work were to obtain quantitative information on the concentration of different particle size ranges in a classroom and how this is influenced by occupants' behavior. The aim was also to find a correlation between the particle concentration in a classroom and occupancy patterns of students and the relationship with outdoor particle concentrations. Furthermore, an investigation of the resuspension due to students' activities within the classroom was carried out to identify particle size ranges that are influenced by such activities.

Methods

Location of the sampling site

Airborne particle mass concentrations measurement was carried out for four weeks in a university classroom during winter (two weeks), spring (one week), and summer (one week) periods of 2010 at the University of Reading, UK. The university campus is about 2 km (1.24 miles) from Reading's town center and approximately 60 km (37.28 miles) west of London in the United Kingdom. The classroom is situated on the first floor (second level) with a floor area of 60 [m.sup.2] (645.83 [ft.sup.2]), a volume of 211 [m.sup.3] (7451.39 [ft.sup.3]), and a capacity of about 60 students. The classroom was naturally ventilated by 12 double-glazed windows under the occupants' control and a windcatcher. The windcatcher is a commercial design and was installed on the roof incorporating four quadrants with an opening area of 1.44 [m.sup.2] (15.50 [ft.sup.2]). The classroom had carpeted floor.

Instrumentation

Concentrations in the classroom and outdoors were monitored using two units of Grimm Portable Laser Aerosol Spectrometer model 1.108. This device uses a light-scattering technique for measuring particle concentration and gives real-time measurements of particles in the size range 0.3-[greater than or equal to]20 [micro]m. The air sampled was continuously drawn into the instrument by a pump with a flow rate of 0.0012 [m.sup.3]/min (0.0424 cf/min). The instruments were calibrated by the manufacturer on December 2009, and the calibrations are valid until the end of February 2011. Both units were set side by side for a period of 24 h before the experiment was conducted to ensure that they provided similar results.

Sampling method

Classroom measurements

Measurements of particle mass concentrations in the classroom were made during school time for three weeks (two weeks in the winter and one week in the spring). Additional measurements were conducted in the classroom for one week in the summer during nonschool time (holiday). The purpose of the summer measurements was to investigate the impact of outdoor particles on the particulate matter (PM) in the classroom. The Grimm unit was set to simultaneously measure airborne pollutants in the size range 0.3-[greater than equal to]20 [micro]m and to simultaneously sample [PM.sub.1], [PM.sub.2.5], and [PM.sub.10] fractions (particles with an aerodynamic diameter equal or less than 1, 2.5, and 10 [micro]m, respectively) in the classroom. Class time was usually from 09:00 to 18:00. Therefore, the measurements time during school time for each day was usually from 07:30 to 20:00 to cover the occupied and unoccupied periods.

The optical particle monitor was placed in the middle of the classroom at a height 1.20 m (3.94 ft) above floor level, which represents the breathing level of a sedentary seated adult of average height. The measuring unit was set to a 1-min data logging interval (each reading was an average of 1 min) because with such short intervals, students' activity showed a clear influence on indoor concentration, but for longer intervals this effect could not be captured. It has been reported that with long intervals of particle samplings, significant variations in contaminant concentrations with indoor activity were not observed (Tippayawong et al. 2009).

The researcher was in the classroom during the school-time measurements to record the conditions in the classroom. Examples of the recorded conditions are presented in table 1.

Outdoor measurement

Outdoor measurement of PM mass concentration levels was conducted simultaneously whenever it was possible, for example, when the ambient weather conditions were suitable for leaving the sampling unit outside. The outdoor samples were collected at a distance of about 15 m (49.21 ft) away from the classroom. Ambient PM samples were taken when the outside relative humidity (RH) did not exceed 80% and the outdoor temperature did not go below 0 [degrees]C in order to protect the optical parts of the spectrometer from damage. Seven days were possible for measuring outdoor particle concentrations during the whole school-time measurement. With respect to nonschool-time measurements, the outdoor samplings were carried out for six days. The sampling point of the outdoor monitoring was 2 m (6.56 ft) above ground level. The device was set to provide an average value of 1 min of particle mass concentrations in the size range 0.3-[greater than or equal to]20 [micro]m, [PM.sub.1], [PM.sub.2.5], and [PM.sub.10].

Air change rate measurements

Tracer gas measurements with S[F.sub.6] gas (sulphur hexafluoride) was used to measure the ventilation rate after students had left the classroom. Before the gas injection, the windows, windcatcher, and door of the classroom were closed, and three mixing fans located at different locations were turned on for 10-15 min to mix the air. The tracer gas was then injected at several points in the classroom. After this period, the selected ventilation method was activated and the measurement of concentration decay commenced. Thus, the average air change rate was calculated from the following equation:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (1)

where

[bar.a] is the average air change rate ([h.sup.-1]); t and [t.sub.0] are the final and initial time (h), respectively; and

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the concentrations of the tracer gas (ppm) at time t and [t.sub.0], respectively.

Calculating resuspension of particles

In general, the parameters influencing the concentration of contaminants indoors are indoor pollutant sources, deposition rate of particles, the ventilation rate, and the penetration of outdoor particles. In classrooms, the contribution from students' activities to the PM concentration in classrooms is significant particularly for particles larger than 1 [micro]m (e.g., Branis et al. 2005, 2009). In this study, resuspension was estimated when high peaks were obtained in the occupied classroom, which were the periods of students entering and leaving the classroom. Several peaks were observed, and the size-specific mass balance approach for indoor particle (e.g., Hussein et al. 2005) was modified and simplified to estimate the resuspension per student in the classroom for [PM.sub.1], [PM.sub.2.5], and [PM.sub.10] as follows:

V d[C.sub.in] / dt = VP[bar.a][bar.[C.sub.out]] - V[bar.a][bar.[C.sub.in]] - V[bar.[KC.sub.in]] + [bar.R]n, (2)

where

V is the volume of the classroom ([m.sup.3]);

[bar.[C.sub.in]] and [bar.[C.sub.out]] are the average indoor and outdoor particle concentrations ([micro]g [m.sup.-3]) during students entering or leaving the classroom, respectively;

P is the penetration factor (dimensionless),

[bar.a] and [bar.K] are the average air change rate and deposition rate ([h.sup.-1]), respectively;

[bar.R] is the average resuspension, micrograms per student; and

n is the number of students entering or leaving the classroom per unit time.

In the above equation, the bar over the symbol indicates a mean value throughout a particular period, but other values are instantaneous. The penetration factor P in the equation depends on a number of parameters, such as construction of the building, air change rate, and particle size. Thatcher and Layton (1995) showed that P was equal to 1 for PMs equal or less than 10 [micro]m. Wallace (1996) reported that P is generally assumed to be close to one for fine ([less than or equal]2.5 [micro]m) and coarse ([greater than or equal]2.5 [micro]m) particles. Other studies (e.g., Long et al. 2001; Mosley et al. 2001) reported that the penetration factor can be below 1. It is also noted that the penetration factor is generally higher for fine particles than for coarse ones (Abt et al. 2000). Thus, in equation 2, the penetration factor was taken to be 0 when the ventilation openings were closed and 1 when open for [PM.sub.2.5] and [PM.sub.10], as a precise value would be difficult to obtain, as was also observed by others (e.g., Ferro et al. 2004). For [PM.sub.1], P[bar.[C.sub.out]] in Equation 2 can be approximated by the background (unoccupied) concentration under both conditions in the classroom. It has been reported by a previous study that during the absence of indoor PM sources, the indoor particle concentration can be approximated by ambient particle concentration (Morawska 2001). Note that Equation 2 assumes that the effects of coagulation, evaporation, and condensation are insignificant.

The particle concentrations in the occupied zone during lectures and when the classroom was unoccupied were found to be well mixed under different ventilation regimes. The well-mixed air during occupancy was thought to be due to students' activities; however, during unoccupied periods, the concentrations represented background levels only. This was checked by placing two dust monitors at different places in the classroom during different occupancy periods, and the results showed no significant differences between their readings within the occupied zone; hence, the well-mixed condition in the classroom was confirmed for occupied and unoccupied periods.

Under the conditions of low-ventilation rates (windows and windcatcher closed) and immediately after the students take their seats or leave the classroom, the particle concentrations in the classroom reach the maximum value due to movement effects. The concentrations under those conditions were assumed to be much larger than those outdoors, and the resuspension of particles during the decay periods could be neglected. Therefore, time-dependent Equation 2 can be used to estimate the average deposition loss rate ([bar.K]) as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

Equation 3 can also be written as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (4)

where

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the initial particle concentrations ([micro]g [m.sup.-3]) just after students take their seats or leave the classroom (peak),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the final particle concentrations ([micro]g [m.sup.-3]), and

[bar.a + K] is the average total loss rate ([h.sup.-1]).

Thus, by subtracting the average air change rate ([bar.a]) from the average total loss rate, the average deposition rate [bar.K] can be obtained. By integrating Equation 2 and solving for the average resuspension, the following equation is obtained:

[bar.R] = V[[C.sub.peak] - [C.sub.initial] / [DELTA]t + [bar.[aC.sub.in]] + [bar.[KC.sub.in]] - P[bar.[aC.sub.out]] / n (5)

where

[C.sub.initial] is the concentration ([micro]g [m.sup.-3]) just before students enter or leave the classroom (background),

[C.sub.peak] is the concentration ([micro]g [m.sup.-3]) just after they take their seats or leave the classroom (peak), and

[DELTA]t is the time difference between initial and peak concentrations (min)

Here, the units of [bar.a] and [bar.K] in Equation 5 were converted to [min.sup.-1] instead of [h.sup.-1] to match the unit of the time step and the unit of the number of students entering and/or leaving the classroom. It is important to mention that when solving Equation 5 for the resuspension of [PM.sub.2.5] and [PM.sub.10], [bar.[C.sub.out]] in the equation was neglected whenever the outdoor measurements were not possible. This will not significantly affect the resuspension calculated for [PM.sub.2.5] and [PM.sub.10] since the days that outdoor samplings were not conducted there was very little ventilation (windows and windcatcher were closed) in the classroom; thus, the contribution of outdoor particles to the classroom was considered to be small. In addition, most of the resuspension periods represented a short time interval (<15 min) during which the movement of particles in the classroom and the concentration values at the indoor sampling point (center of the classroom) were not disturbed by the flow of outdoor air. As mentioned earlier, for the resuspension of [PM.sub.1], P[bar.[C.sub.out]] in Equation 5 was approximated by the particle concentrations in the classroom before the occupancy periods. The resuspension calculated from Equation 5 was used to predict the peaks of the three PM fractions ([PM.sub.1], [PM.sub.2.5], and [PM.sub.10]) determined in the classroom during different lecture periods by using the following equation:

[P.sub.p] = ([bar.R]N) / V + [B.sub.g], (6)

where

[P.sub.p] is the predicted peak ([micro]g [m.sup.-3]),

[B.sub.g] is the background concentration in the classroom ([micro]g [m.sup.-3]), and

N is the total number of students entering or leaving the classroom during the observed peak.

Equation 6 is a simplified equation that does not account for the effect of outdoor particles, air change rate, and deposition rate on the concentration in the classroom.

Statistical analysis

SPSS[R] 17 statistical software was used to analyze the data obtained, including descriptive statistical analysis, paired samples t-test, linear regression, and Pearson correlation coefficient. EXCEL 10 software was also employed to estimate the resuspension factor for the three PM fractions and to predict peak concentrations in the classroom.

Results

Measurements of air change rate in the classroom under different ventilation scenarios

The average air change rate was measured in the classroom under varying window and windcatcher operations. As shown in Table 2, the measured air change rate was strongly related to the ventilation method and outdoor weather conditions.

Particle concentrations in the classroom during occupied and unoccupied periods

The average number of students attending a typical lecture was 34. The mean particle mass concentration measurements in all size classes taken in the classroom when it was unoccupied and occupied are illustrated in Table 3. Note that unoccupied periods were defined as the periods preceding the first lecture of each day (typically 20 min) plus the periods after about 1 h following the last lecture of each day. As shown in the table, the concentrations during occupied periods are higher than those when the classroom was unoccupied for all size ranges, with a mean ratio for all size ranges greater than 1.

A mean comparison using a paired samples t-test was conducted on the mean particle concentrations in all size ranges during occupied and unoccupied periods. The results of the t-test showed that the impact of students' presence on the particle concentrations becomes more significant for particles in the size range of 0.65-0.8 [micro]m and larger (p-value <0.05). Table 3 shows that the standard deviation for each particle size is large when compared with the mean value for the whole measuring period. This large variation is related to the way the classroom was used (e.g., number of people, occupancy period, etc.). This is also true for the unoccupied periods, as the particle concentration during these periods is also influenced by the proceeding periods.

Figure 1 shows atypical spectrum of particle concentration over a whole day, i.e., for occupancy and nonoccupancy periods. When students entered the classroom at the beginning of each lecture, the particle concentration levels rapidly rose to reach maximum values (e.g., the period between 08:50 and 9:03). In the following period, when the students took their seats, the movements stopped and the decay commenced (e.g., the period between 09:03 and 10:45). At the end of each lecture, the students left the classroom; then the concentrations start to increase gradually due to movements' effect (e.g., the period between 12:34 and 12:36); thereafter, the peak decreased gradually due to the absence of indoor sources (e.g., the period between 12:36 and 13:40). This clearly shows the large variation in particle concentration for a whole day, which may explain the large standard deviation values given in Table 3 and the subsequent related figures.

[FIGURE 1 OMITTED]

Figure 2 shows the mean mass concentrations of the three PM fractions measured in the classroom during the unoccupied and occupied periods. These values are for all particle sizes up to and smaller than the size shown in the figure. It is clear from the figure that the mean concentration of the [PM.sub.10] fraction for the measurement duration showed the highest correlation with students' activities, where the mean concentration when the classroom was occupied (79.14 [+ or -] 19.21 [micro]g [m.sup.-3]) is about six times higher than when it was unoccupied (13.52 [+ or -] 7.29 [micro]g [m.sup.-3]). Although the influence of students' attendance on the [PM.sup.2.5] fraction was not as high as [PM.sup.10] fraction, [PM.sup.2.5] gave a significant correlation (p-value < 0.05) with students' activities. The finest fraction ([PM.sup.1]) gave similar average values within the two periods measured, which revealed that the relationship with students' presence was not significant (p-value = 0.065).

Figure 3 illustrates the mean concentration levels of [PM.sub.1], [PM.sub.2.5], and [PM.sub.10] measured in the classroom during unoccupied and different occupied periods (sitting and walking). A sitting period is defined as the period when the students have been seated. These periods cover the last 20 min of each lecture (sufficient time after entering and just before leaving) in order to reduce the effect of students' movements on the results. A walking period is defined as the period during which students enter or leave the classroom. This period covers the time students spend to enter or leave the classroom. Again the [PM.sub.10] fraction presented significant differences (p-value < 0.05) among the mean of unoccupied, sitting, and walking periods (14.04 [+ or -] 8.05 [micro]g [m.sup.-3], 52.07 [+ or -] 14.12 [micro]g [m.sup.-3], and 97.10 [+ or -] 16.52 [micro]g [m.sup.-3], respectively). On the other hand, the [PM.sub.2.5] fraction showed significant differences between the mean of unoccupied and walking periods (p-value < 0.05), but it failed to show significant differences between unoccupied and sitting periods and between walking and sitting periods (p-value = 0.136 and 0.119). The [PM.sub.1] fraction did not show any significant differences among the three periods.

[FIGURE 2 OMITTED]

Effect of occupancy periods on indoor/outdoor (I/O) ratio

Figure 4 shows the influence of different occupancy activities (walking and sitting) under different ventilation methods on the I/O ratios for the three PM fractions during two typical days of school-time measurements in the classroom. It is obvious from the figure that the I/O ratio under both ventilation methods for [PM.sub.10] changed significantly during the different periods. During walking periods, the I/O ratio reached its maximum values, but during the sitting periods, the I/O ratio decreased gradually. This confirms the effect of students' presence on the [PM.sub.10] fraction with an I/O ratio always higher than 1.0 during occupancy. Also, it can be seen from the figure that occupancy periods did not cause any important changes to the I/O ratio for [PM.sub.1] and [PM.sub.2.5] fractions.

Simple linear regression was employed to determine the correlation between the number of students walking (entering and/or leaving) in the classroom and the concentration of the three particle fractions (Figure 5). In order to reduce the impact of deposition rates on the particle concentration during students entering or leaving, the data obtained within short interval periods was analyzed (<20 min). By taking PM fractions as the dependent variables and the number of students walking as the independent variable, Pearson correlation analysis confirmed that the number of students walking in the classroom make a positive contribution to the classroom [PM.sub.10] concentrations ([R.sup.2] = 0.47; significant at the 0.01 level). However, the correlations for the effect of the number of students on the other fractions are not clear ([R.sup.2] = 0.14 and 0.057 for [PM.sub.1] and [PM.sub.2.5], respectively).

[FIGURE 3 OMITTED]

Resuspension of the three PM fractions

The average resuspension per student for [PM.sub.1], [PM.sub.2.5], and [PM.sub.10] due to students' activities in the classroom calculated using Equation 5 is given in Table 4. It is important to mention that resuspension occurred at the time the students enter and/or leave the classroom. For example, in general, particles do not resuspend after students take their seats. On average, it took about 16 min for students to be settled (PM reaches the peak). As was expected, the [PM.sub.10] fraction has the highest average resuspension value. The two finer fractions ([PM.sub.1] and [PM.sub.2.5]) resuspension values were small.

[FIGURE 4 OMITTED]

As mentioned earlier, peaks were observed at the beginning and the end of the classes, which were related to students walking. The decays were determined after students were seated and/or the classroom was vacated (see Figure 1). Using the resuspension values (microgram per student) determined for the three particulate fractions, strong correlations were obtained between measured and predicted peaks in the classroom for the three particle fractions ([R.sup.2] = 0.99 for [PM.sub.1], [R.sup.2] = 0.87 for [PM.sub.2.5], and [R.sup.2] = 0.80 for [PM.sub.10]). Pearson correlations ([R.sup.2]) are significant at the 0.01 level for the three PM fractions (see Figure 6). The predicted peaks were based on the average resuspension for all the measurements during this study. This shows that the resuspension per student of the three particle fractions is very similar for different occupancy patterns.

Outdoor and classroom particle concentrations

Figures 7 and 8 illustrate the relationship between indoor and outdoor PM fractions under different ventilation scenarios, including windcatcher open, two windows open, and windows closed. The data of the classroom analyzed for this investigation was obtained from the measurements during holiday periods in order to avoid the influence of indoor sources (students) on the correlations. When the windcatcher was open, there was some influence of outdoor concentrations, particularly for fine fractions, as the outdoor values were increasing during the measurements periods. This was due to the windcatcher allowing outdoor particles to enter the classroom. Under the condition of opening two windows (Figure 8a), the correlation between indoor and outdoor concentration is much stronger, particularly for [PM.sub.1] and [PM.sub.2.5] ([R.sup.2] = 0.96 and 0.92, respectively). Under this and the previous condition, a relatively good correlation was determined for indoor and outdoor [PM.sub.10] concentrations ([R.sup.2] = 0.46 and 0.55, respectively). However, with all windows closed (Figure 8b), and the fact that the outdoor PM fractions were essentially constant, the indoor PM fractions were almost at constant concentrations; thus, the correlation between indoors and outdoors was not clear under this condition.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

Discussion

Exposure to PM is associated with harmful effects on human health. The effect of PM on people is dependent on the size range of particles. Thus, examining different particle size classes is very important in personal exposure studies. The results obtained in this study show that particle concentrations during occupancy are higher than those when the classroom was unoccupied for all size ranges. The reason for the difference is the presence of students. When the students occupy the classroom, resuspension of particles larger than 1 [micro]m is significant, which leads to higher levels than those when they are not in the classroom. This finding is consistent with previous findings in other situations. A recent study conducted in a bus showed an increase in coarse particles due to resuspension from passenger movements in the bus (Song et al. 2009). Another investigation carried out in a house for particle diameters of 0.18-18 [micro]m and >18 [micro]m reported that particles larger than 1 [micro]m were more likely to be resuspended by human activity (Qian et al. 2008). A previous finding by Branis et al. (2009) documented that the indoor coarse fractions are associated with the number of students in the microenvironment. These studies confirm the findings of this study. On the other hand, the effect of occupancy periods on the fine particles in this study might be explained by the fact that the students sometimes opened the windows when they were in the classroom, which led to an increase in air change with the outdoor air and a possible increase in the concentrations of fine particle (<1 [micro]m) in the classroom, hence an increase in the levels of these size classes during occupancy and, therefore, an increase in the occupancy to nonoccupancy ratios. It has previously been found that indoor fine particles are well correlated with ambient concentrations because they have higher penetration efficiency than for larger ones (e.g., Long et al. 2001; Fromme et al. 2007; Guo et al 2010).

When particle size range increases, the occupancy and nonoccupancy ratio increases as well. This confirms that larger particles have a strong relationship with human activates. This finding is supported by the findings of Thatcher and Layton (1995), which showed that the activity and nonactivity ratio for the PM size range of 0.5-25 [micro]m increases with particle size range. The ratios obtained in the current study were higher than those found in their study. This could be due to the difference in the number of occupants and their activities between the two studies, causing greater resuspension of the particles in the current case, as their study was conducted in a residence. The high ratios observed in the present study for the particle size classes 15-20 [micro]m and >20 [micro]m (68.70 and 78.03, respectively) might be explained by the fact that the concentrations of these size ranges during the nonoccupancy periods were usually very low (<0.5 [micro]g [m.sup.-3]), so when the classroom was occupied, the ratio became very high.

Ambient air quality standards (AAQs) with 24-h and annual averaging times have been established for [PM.sub.2.5] and [PM.sub.10] to define the maximum amount of particles that can be present in outdoor air without threatening public health. This article focused on the three PM fractions ([PM.sub.1], [PM.sub.2.5], and [PM.sub.10]). The average of [PM.sub.10] levels when the classroom was occupied was 79.14 [+ or -] 19.21 [micro]g [m.sup.-3].This value does not meet the World Health Organization (WHO 2006) standards for outdoor [PM.sub.10] 24-h and annual averages, which have been set at 50 [micro]g [m.sub.-3] and 20 [micro]g [m.sup.-3], respectively. It is also above the suggested indoor guideline value (40 [micro]g [m.sup.-3])for the [PM.sub.10] 24-h average as prescribed for Flanders, Belgium (Flemish Government 2004). The average of [PM.sub.10] measured in the classroom, which is well associated with students' movements, was 97.10 [+ or -] 16.52 [micro]g [m.sup.-3]--far beyond the above-mentioned guideline values. In addition, the average of [PM.sub.10] measured in the classroom when the students were seated (no movements) was higher than those given by the standards above (52.07 [+ or -] 14.12 [micro]g [m.sup.-3]).

Also, the average of [PM.sub.2.5] during the occupied periods was 16.15 [+ or -] 8.76 [micro]g [m.sup.-3], which does not satisfy the WHO (2006) standard for ambient [PM.sub.2.5] annual average that has been set at 10 [micro]g [m.sup.-3]. A number of peaks of [PM.sub.2.5] in the classroom exceeded the outdoor limit of the 24-h average for [PM.sub.2.5] (peaks >25 [micro]g [m.sup.-3]). The peaks of [PM.sub.2.5] also exceeded the indoor air quality guideline values for the [PM.sub.2.5] annual average (15 [micro]g [m.sup.-3]) as proposed for Flanders (Flemish Government 2004). With respect to [PM.sub.1], there is no guideline set for this size fraction.

The average [PM.sub.10] levels measured in this classroom during school time (79.14 [micro]g [m.sup.-3])is high in comparison with the previous findings in the same classroom during school time, which was 53.12 [micro]g [m.sup.-3] (Alshitawi et al. 2009). This could be due to the fact that the number of students and the number of classes during the current study were more than those reported in the previous investigation. It was shown by a study conducted in a number of classrooms in primary and secondary schools that during periods when classrooms were occupied, the mean [PM.sub.10] and [PM.sub.2.5] in winter were 105.0 [micro]g [m.sup.-3] and 23.0 [micro]g [m.sup.-3], respectively, and in summer were 71.7 [micro]g [m.sup.-3] and 13.5 [micro]g [m.sup.-3], respectively (Fromme et al. 2007). Another study conducted in a university classroom found that the mean concentration of [PM.sub.10],[PM.sub.2.5,] and [PM.sub.1] during school time was equal to 42.3 [micro]g [m.sup.-3], 21.9 [micro]g [m.sup.-3], and 13.7 [micro]g [m.sup.-3], respectively (Branis et al. 2005). The difference between the findings here and those by others could be due to the different number of students in classrooms, different number of classes, and finally different ventilation practices. For the purpose of calculating the particle concentrations during occupancy periods, parts of the spectrum related to the unoccupied periods were excluded. The results of PM fractions during occupancy periods are generally in agreement with above-mentioned published data.

The influence of student attendance on the three PM fractions was examined in this study. The effect of students on [PM.sub.10] was very clear; the difference between occupancy and nonoccupancy periods was significant, which was about six times higher than when the classroom was unoccupied. This observation agrees with previous studies (e.g., Janssen et al. 1999). In addition, [PM.sub.10] concentrations significantly differed within different occupancy periods (walking and sitting), where the average concentrations during walking and sitting periods were 97.10 [+ or -] 16.52 [micro]g [m.sup.-3] and 52.07 [+ or -] 14.12 [micro]g [m.sup.-3], respectively. This implies that the high concentration detected in the classroom was due to students' movements (entering and/or leaving the classroom). This result appears to be consistent with the result found by Thatcher and Layton (1995), where 2 min of continuous walking and sitting by one person resuspended more particles than four people doing normal activity for 5 min. The correlations of fine fractions with occupancy periods were not as strong as those for [PM.sub.10], although the concentrations during occupancy periods were clear (particularly for [PM.sub.2.5]).

The I/O ratio is often applied to detect the presence of indoor sources, e.g., I/O ratio above 1 (e.g., Jones 1999; Liu et al. 2004; Gemenetzis et al. 2006). The I/O ratios measured in this study for [PM.sub.10] were usually higher than 1 during occupied periods. In the following, the I/O ratios for different periods representing different ventilation methods during two typical days of measurements in the classroom will be presented. The I/O ratios for [PM.sub.10] ranged from 0.71 to 1.04 (mean 0.86 [+ or -] 0.09), 0.73 to 1.58 (mean 1.04 [+ or -] 0.17), and 1.24 to 4.51 (mean 2.72 [+ or -] 0.71) during nonoccupancy, sitting, and walking periods, respectively, when all windows and the windcatcher were closed. It ranged from 0.27 to 1.23 (mean 0.64 [+ or -] 0.16), 0.47 to 1.74 (mean 1.16 [+ or -] 0.32), and 0.79 to 6.62 (mean 3.30 [+ or -] 1.51) during nonoccupancy, sitting, and walking periods, respectively, under the operation of the windcatcher. On both days, the presence of students influenced the I/O ratios of [PM.sub.10] and the highest ratios observed during walking periods, as these periods are considered to have more contribution to classroom particles than other periods. When the classroom was not occupied, the mean I/O ratios of [PM.sub.10] were recorded below 1, which was expected because of the absence of indoor sources. Although the average number of students per lecture during the day with no ventilation (25 students per lecture) were higher than for the day with the windcatcher open (9 students per lecture), the mean I/O ratios of [PM.sub.10] obtained in the classroom during occupancy periods during the operation of the windcatcher were higher than those found under the condition of closed windows. This was due to the fact that during windcatcher operation, the average outdoor [PM.sub.10] concentration (17 [+ or -] 6.14 [micro]g [m.sup.-3) was much lower than when windows were closed (44.54 [+ or -] 6.31 [micro]g [m.sup.-3]). In contrast, no significant difference in the mean I/O ratios for the fine fractions was recorded during the three periods.

The I/O ratios of [PM.sub.10] measured in the present study are now comparable with those in previous studies. A number of studies determined the effect of human activity on the I/O ratio. For example, Song et al. (2009) monitored [PM.sub.10] concentrations in buses and showed that I/O ratio varied between 3.73 and 7.66. Another research study investigated the I/O ratio of [PM.sub.10] in a number of classrooms and found that I/O ratio varied between 2.34 and 4.62 (Janssen et al. 1997). A previous investigation conducted in residential flats concluded that I/O ratio of [PM.sub.10] ranged from 1.2 to 3.9 (Jones et al. 2000). With respect to the fine fractions, a recent study examined the influence of occupancy periods in classrooms on I/O ratio of fine particles (Guo et al. 2010). It was concluded that there was no significant difference in I/O ratio between occupied and unoccupied periods in the classrooms. Thus, this study's I/O ratios during occupancy periods are consistent with that found in other microenvironments.

The good correlation ([R.sup.2] = 0.47) obtained in this study between the number of students walking (entering and/or leaving) in the classroom and the concentration of [PM.sub.10] in the classroom during walking periods is more evidence of the strong relationship between coarse particles and student activity. On the other hand, fine factions failed to show such a clear correlation. Therefore, it can be concluded that coarse particles are the size range most influenced by students, and although fine particles (particularly for [PM.sub.2.5]) showed reasonable difference between occupancy and nonoccupancy periods, the effect of students' presence on this size fraction was not conclusive. This could be due to the strong association between fine particles and ambient particles. This conclusion is supported by number of previous studies (e.g., Thatcher and Layton 1995; Ferro et al. 2004; Guo et al. 2010).

The results of airborne particle levels in this study suggested that the classroom required adequate ventilation by means of opening windows at the beginning and at the end of each class, because on these occasions, the concentrations reach maximum values. This action may protect occupants from the risk of short-term exposure to airborne contaminants.

This study was able to quantify the resuspension per student in the classroom for a range of particle sizes. The resuspension for the three particle fractions were discussed. The resuspension values obtained were used to simulate peaks in the classroom at different lectures and under different ventilation modes. The predicted peak concentrations for PM fractions covered over 79% (see Figure 6) of the variation in measured peaks concentration in the classroom. It is important to mention that the resuspension values could be affected by rapid deposition, especially for larger particles, which, in turn, would underestimate the resuspension coefficient. In contrast, the resuspension coefficient of fine particles can be influenced by the outdoor particles and the ventilation rate.

The formula used for the prediction of PM fractions was only used to simulate the peak concentrations in the classroom. This was because, in this study, the parameters governing the concentrations of airborne particles indoor (e.g., outdoor particles, deposition rates, etc.) were not taken into consideration to enable the prediction of the concentration spectrum. For future work, the limitation in the current study will be extended by comparing the resuspension values for PM fractions determined during this study with resuspension values for the same size fractions obtained in a new study that was conducted under control environment conditions (environmental chamber). In addition, the mass balance model for modeling indoor airborne particles will be utilized using the resuspension values and taking into consideration the important factors affecting the airborne PM concentration indoors, hence simulating the concentration profiles in the classrooms due to occupancy.

Positive correlation between outdoor and indoor PM mass concentrations was reported by previous studies (Branis et al. 2005; Wang et al. 2006; Massey et al. 2009; Kuo and Shen, 2010). This study also investigated the contribution from the outdoor airborne particles to the classroom particles under different ventilation scenarios. The impact of outdoor particulates depends on a number of factors (e.g., ventilation scenarios and particle size ranges). In general, the correlation is stronger when the outdoor concentration varies (increasing or decreasing), a sufficient air change between indoor and outdoor exists, and during the absence of indoor sources. The last two conditions can be met, but the first is outside of control. This investigation was able to obtain data that included the conditions above.

Under the conditions of operation of the windcatcher and the opening of two windows, a strong positive relationship between the outdoor fine fractions levels and those in the classroom was recorded. The correlation with the two windows open was higher than that with the windcatcher open. This might be explained by the difference in ventilation rate between the two scenarios. The air change rate measured when two windows were open was higher than that under the operation of the windcatcher. The correlation for the [PM.sub.10] fraction was lower than for fine fractions under both ventilation methods. This confirms this study's findings (mentioned earlier), where fine particles are influenced strongly by outdoor particles but coarse particles are affected strongly by human activity. Under no ventilation conditions (windows closed), the correlation between the classroom and the outdoor PM fractions was not clear due to a low air change rate and/or outdoor particles concentration in a steadystate situation.

Conclusion

This study measured three PM fractions ([PM.sub.1], [PM.sub.2.5], and [PM.sub.10]) and size-resolved airborne particles in the range of 0.3 to [greater than or equal to]20 [micro]m in a classroom to determine the impact of students' presence on the concentrations of different PM classes. The concentration levels of the coarse particles were strongly affected by students walking into the classroom. The effect of students' activity after they had settled in their seats was very low. Thus, this study suggests further investigation into the effect of students movements (e.g., entering and/or leaving the classroom) because at these moments, the concentrations reach the highest value, and the key parameters (resuspension and deposition rate) for modeling particle concentrations in classrooms need to be derived. The resuspension per student calculated in this study for the prediction of the three PM fractions ([PM.sub.1], [PM.sub.2.5], and [PM.sub.10]) showed more than 79% correlation between measured and predicted peaks.

DOI: 10.1080/10789669.2011.583708

References

Abt, E., H.H. Suh, P. Catalano, and P. Koutrakis. 2000. Relative concentration of outdoor and indoor particle sources to indoor concentrations. Environmental Science and Technology 34:3579-87.

Alshitawi, M., H. Awbi, and N. Mahyuddin. 2009. Particulate matter mass concentration ([PM.sub.10]) under different ventilation methods in classrooms. International Journal of Ventilation 8(2):93-108.

Annesi-Maesano, I., D. Moreau, D. Caillaud, F. Lavaud, Y. Le Moullec, A. Taytard, G. Pauli, and D. Charpin. 2007. Residential proximity fine particles related to allergic sensitisation and asthma in primary school children. Respiratory Medicine 101(8):1721-9.

Awbi, H., and A. Pay. 1995. A study of air quality in classrooms. Proceedings of the Second International Conference on Indoor Air Quality, Ventilation and Energy Conservation in Buildings, Montreal, Canada 93-104.

Bako-Biro, Z., N. Kochhar, D. Clements-Croome, H. Awbi, and M. Williams. 2008. Ventilation rates in schools and pupil's performance using computerized assessment tests. Paper ID:880. Proceedings of the International Conference on Indoor Air, Copenhagen, Denmark.

Berico, M., A. Luciani, and M. Formignani. 1997. Atmospheric aerosol in an urban area: Measurements of TSP and [PM.sub.10] standards and pulmonary deposition assessments. Atmospheric Environment 31(21):3659-65.

Branis, M., P. Rezacova, and M. Domasova. 2005. The effect of outdoor air and indoor human activity on mass concentrations of [PM.sub.10], [PM.sub.2.5], and [PM.sub.1] in a classroom. Environmental Research 99(2):143-9.

Braniss, M., J. safranek, and A. Hytychova. 2009. Exposure of children to airborne particulate matter of different size fractions during indoor physical education at school. Building and Environment 44:1246-52.

Delfino, R.J., R.S. Zeiger, J.M. Seitzer, and D.H Street. 1998. symptoms in pediatric asthmatics and air pollution: Differences in effects by symptom severity, anti-inflammatory medication use and particulate averaging time. Environmental Health Perspectives 106:751-61.

Ferro, A.R., R.J. Kopperud, and L.M. Hildemann. 2004. Source strength for indoor human activity that resuspend particulate matter. Environmental Science and Technology 38:1759-64.

Flemish Government. 2004. Decree of the Flemish Government of 11 June 2004 providing for measures aiming at controlling health risks caused by indoor pollution. Belgium Official Journal 19/10/2004:72556.

Fromme, H., D. Twardella, S. Dietrich, D. Heitmann, R. Schierl, B. Liebl, and H. Ruden. 2007. Particulate matter in the indoor air of classrooms--exploratory results from Munich and surrounding area. Atmospheric Environment 41(4):854-66.

Gemenetzis, P., P. Moussas, A. Arditsoglou, and C. samara. 2006. Mass concentration and elemental composition of indoor [PM.sub.2.5] and [PM.sub.10] in university rooms in Thessaloniki, northern Greece. Atmospheric Environment 40(17):3195-206.

Gold, D.R., A. Litonjua, J. Schwartz, E. Lovett, A. Larson, B. Nearing, G. Allen, M. Verrier, R. Cherry, and R. Verrie. 2000. Ambient pollution and heart rate variability. Circulation 101:1267-73.

Guo, H., L. Morawska, C. He, Y.L. Zhang, G. Ayoko, and M. Cao. 2010. Characterization of particle number concentrations and [PM.sub.2.5] in a school: influence of outdoor air pollution on indoor air. Environmental Science and Pollution Research 17:1268-78.

Hoppe, P. 2002. Different aspects of assessing indoor and outdoor thermal comfort. Energy and Buildings 34(6):661-5.

Hussein, T., K. Hameri, M.S.A. Heikkinen, and M. Kulmala. 2005. Indoor and outdoor particle size characterization at a family house in Espoo-Finland. Atmospheric Environment 39:3697-709.

Janssen, N.A., G. Hoek, B. Brunekreef, and H. Harssema. 1999. Mass concentration and elemental composition of [PM.sub.10] in classrooms. Occupational and Environmental Medicine 56:482-7.

Janssen, N.A., G. Hoek, H. Harssema, and B. Brunekreef. 1997. Childhood exposure to [PM.sub.10]: Relation between personal, classroom, and outdoor concentrations. Occupational and Environmental Medicine 54:888-94.

Jones, A. 1999. Indoor air quality and health. Atmospheric Environment 33:4535-64.

Jones, N.C., C.A. Thornton, D. Mark, and R.M. Harrison. 2000. Indoor/outdoor relationships of particulate matter in domestic homes with roadside, urban and rural locations. Atmospheric Environment 34:2603-12.

Karimipanah, T., H. Awbi, M. Sandberg, and C. Blomqvist. 2007. Investigation of air quality, comfort parameters and effectiveness for two floor-level air supply systems in classrooms. Building and Environment 42(2):647-55.

Kuo, H., andH. Shen. 2010. Indoor and outdoor [PM.sub.2.5] and [PM.sub.10] concentrations in the air during a dust storm. Building and Environment 45:610-4.

Liu, Y., R. Chen, X. Shen, and X. Mao. 2004. Wintertime indoor air levels of [PM.sub.10], PM2.5, and [PM.sub.1] at public places and their contributions to TSP. Environment International 30(2):189-97.

Long, C.M., H.H. Suh, P.J. Catalano, and P. Koutrakis. 2001. Using time-and size-resolved particulate data to quantify indoor penetration and deposition behavior. Environmental Science and Technology 35(10):2089-99.

Massey, D., J. Masih, A. Kulshrestha, M. Habil, and A. Taneja. 2009. Indoor/outdoor relationship of fine particles less than 2.5 [micro]m ([PM.sub.2.5]) in residential homes locations in central Indian region. Building and Environment 44:2037-45.

Morawska, L., C. He, J. Hitchins, D. Gilbert, and S. Parappukkaran. 2001. The relationship between indoor and outdoor airborne particles in the residential environment. Atmospheric Environment 35:3463-73.

Mosley, R.B., D.J. Greenwell, L.E. Sparks, Z. Guo, W.G. Tucker, R. Fortmann, and C. Whitfield. 2001. Penetration of ambient fine particles into the indoor environment. Aerosol Science & Technology 34:127-36.

Owen, M.K., D.S. Ensor, and L.E. Sparks. 1992. Airborne particle sizes and sources found in indoor air. Atmospheric Environment. Part A. General Topics 26(12):2149-62.

Pope, C.A., D.V. Bates, and M.E. Raizenne. 1995. Health effects of particulate air pollution: Time for reassessment? Environmental Health Perspective 103(5):472-80.

Qian, J., A.R. Ferro, and K.R. Fowler. 2008. Estimating the resuspension rate and residence time of indoor particles. Journal of the Air & Waste Management Association 58:502-16.

Raja, S., Y. Xu, A.R. Ferro, P.A. Jaques, and P.K. Hopke. 2010. Resuspension of indoor aeroallergens and relationship to lung inflammation in asthmatic children. Environment International 36:8-14.

Song, W.W., M.R. Ashmore, and A.C. Terry. 2009. The influence of passenger activities on exposure inside buses. Atmospheric Environment 43:6271-8.

Thatcher, T., and D. Layton. 1995. Deposition, resuspension, and penetration of particles within a residence. Atmospheric Environment 29(13):1487-97.

Tippayawong, N., P. Khuntong, C. Nitatwichit, Y. Khunatorn, and C. Tantakitti. 2009. Indoor/outdoor relationships of size resolved particle concentrations in naturally ventilated school environments. Building and Environment 44(1):188-97.

Wallace, L. 1996. Indoor particles: A review. Journal of the Air and Waste Management Association 46:98-126.

Wang, X., X. Bi, G. Sheng, and J. Fu. 2006. Hospital indoor [PM.sub.10]/[PM.sub.2.5] and associated trace elements in Guangzhou, China. Science of the Total Environment 366:124-35.

WHO. 2006. Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide Global update 2005: Summary of risk assessment WHO/SDE/PHE/OEH/06.02. Geneva, Switzerland.

Zuraimi, M.S., and K.W. Tham. 2009. Reducing particle exposures in a tropical office building using electrostatic precipitators. Building and Environment 44:2475-85.

Received December 6, 2010; accepted April 14, 2011

Mohammed S. Alshitawi, MS, is Lecturer. Hazim B. Awbi, PhD, Member ASHRAE, is Professor.

Mohammed S. Alshitawi * and Hazim B. Awbi

School of Construction Management and Engineering, University of Reading, Reading, UK

* Corresponding author e-mail: m.s.alshitawi@reading.ac.uk

Table 1. Summary of the conditions in the occupied classroom in each day of the measurements. Day (time) (a) Number of Number of Vacuuming lectures students (c) 18-1-10 (09:00 to 16:30) 1 (b) 57 No 19-1-10 (09:00 to 17:30) 4 28 Yes (d) 20-1-10 (10:00 to 16:00) 3 36 No 21-1-10 (10:30 to 18:00) 3 19 No 22-1-10 (09:00 to 17:00) 3 40 No 25-1-10 (09:00 to 16:00) 1b 52 No 26-1-10 (09:00 to 18:00) 4 25 No 27-1-10 (10:00 to 14:00) 3 39 Yes (d) 28-1-10 (10:00 to 18:00) 4 40 No 29-1-10 (09:00 to 15:40) 2 32 No 19-4-10 (e) No lecture 0 No 20-4-10 (12:00 to 15:30) 3 9 No 21-4-10 No lecture 0 No 22-4-10 (11:00 to 15:50) 3 31 No 23-4-10(13:00 to 14:00) (e) 1 9 n/a (a) Day of the measurements (class time in each day). (b) Seminar for the whole day. (c) Average number of students per lecture. (d) Vacuuming took place early morning before classes started. (e) Particle number concentration is measured instead of mass concentrations, and the results are not presented in this article. Table 2. Average air change rate measured under different ventilation methods in the classroom. Ventilation scenario Opening size ACH ([h.sup.-1]) Three windows opened Two were fully open 2.85 and one was partially open Two windows opened Fully open 2.44/3.08 (a) Four windows opened Fully open 6.06 One window opened Half open 2.02 One window opened Partially open 1.02 No windows opened Closed 0.72/0.59/0.41 (b) Windcatcher Fully open 1 .8/2.1 (c) (a) Two different days. (b) Three different days. (c) Two different days. Table 3. Mean particles concentrations during unoccupied and occupied periods in the classroom for whole measurements. Particle Non-occupancy (a) Occupancy (a) Mean size range ([micro][gm.sup.-3]) ([micro][gm.sup.-3]) ratio (b) ([micro]m) 0.3-0.4 2.53 [+ or -] 2.37 3.41 [+ or -] 3.03 1.35 0.4-0.5 1.64 [+ or -] 1.91 2.41 [+ or -] 2.46 1.47 0.5-0.65 1.09 [+ or -] 1.30 1.81 [+ or -] 1.76 1.66 0.65-0.8 0.48 [+ or -] 0.44 0.88 [+ or -] 0.65 1.83 0.8-1.0 0.46 [+ or -] 0.27 0.94 [+ or -] 0.37 2.04 1.0-1.6 0.47 [+ or -] 0.22 1.21 [+ or -] 0.37 2.57 1.6-2.0 0.87 [+ or -] 0.35 2.63 [+ or -] 0.79 3.02 2.0-3.0 2.32 [+ or -] 0.88 10.62 [+ or -] 3.57 4.58 3.0-4.0 1.99 [+ or -] 0.83 15.30 [+ or -] 5.09 7.69 4.0-5.0 2.0 [+ or -] 0.89 18.72 [+ or -] 5.98 9.36 5.0-7.5 2.97 [+ or -] 1.68 36.95 [+ or -] 11.11 12.44 7.0-10 1.39 [+ or -] 1.13 28.42 [+ or -] 7.08 20.45 10-15 1.60 [+ or -] 1.34 51.21 [+ or -] 11.02 32 15-20 0.31 [+ or -] 0.42 21.30 [+ or -] 4.34 68.70 > 20 0.31 [+ or -] 0.40 24.19 [+ or -] 5.96 78.03 (a) Average (mean) with standard deviation in ([micro][gm.sup.-3]) from day to day. (b) Occupancy/nonoccupancy ratio. Table 4. Average resuspension values due to students' activities calculated for the three particulate fractions in the classroom during the measurements. Size fraction Resuspension (a) ([bar.R]), ([micro]m) microgram per student [PM.sub.1] 8.22 [+ or -] 2.33 [PM.sub.2.5] 50.63 [+ or -] 18.94 [PM.sub.10] 847.31 [+ or -] 208.14 (a) Average resuspension with standard deviation in (microgram per student).

Printer friendly Cite/link Email Feedback | |

Author: | Alshitawi, Mohammed S.; Awbi, Hazim B. |
---|---|

Publication: | HVAC & R Research |

Date: | Jul 1, 2011 |

Words: | 9031 |

Previous Article: | Spatial distribution of human respiratory droplet residuals and exposure risk for the co-occupant under different ventilation methods. |

Next Article: | Method to assess quality and quantity of outdoor ventilation airflow received by occupants. |

Topics: |