Airflow characteristics and pollution distribution around a thermal manikin--impact of specific personal and indoor environmental factors.
People spend most of their time indoors and they are constantly exposed to pollution that affects their health, comfort, and productivity. Due to strong economic and environmental pressures to reduce building energy consumption, low air velocity design is gaining popularity; hence buoyancy flows generated by building occupants are gaining more prominent influence in space airflow formation and in the indoor environment overall. This will especially be the case in the future because it is reasonable to expect increased occupant space density due to an overall population growth and urbanization, while the airflows generated by thermal sources such as lighting and equipment will be less important due to the increasing use of low-power devices. In such spaces with low air supply velocity, air mixing is minimized and the pollution is nonuniformly distributed. Large spatial differences in pollution concentration mean that personal exposure, rather than average space concentration, determines the risk of elevated exposure. Current room air distribution design practice does not take into account the air movement induced by the thermal flows from occupants, which often results in inaccurate exposure prediction. This highlights the importance of a detailed understanding of the complex airflow interactions that take place in the close proximity of the human body and their impact on personal exposure.
Factors Influencing Airflow Characteristics Around the Human Body
The recommended design indoor air temperature ranges from 20[degrees]C to 26[degrees]C (68[degrees]F to 78.8[degrees]F) (ISO 2005; ASHRAE 2013), which is about 7[degrees]C to 13[degrees]C (12.6[degrees]F to 23.4[degrees]F) colder than the human skin temperature. Consequently, temperature gradients exist between the room air and the surface of the skin that cause a steady natural process of convective heat loss to the surrounding space. The convective heat loss from the human body induces upward movement of the surrounding air, thus forming a convective boundary layer (CBL) around it and a free-convection thermal plume above the shoulders and the head. The human CBL has been investigated in the past by using several measurement techniques, such as hot wire anemometry (Lewis et al. 1969; Homma and Yakiyama 1988), schlieren photography (Settles 2001; Lewis et al. 1969), laser doppler anemometry (Melikov and Zhou 1996), and particle image velocimetry (Marr et al. 2005; Craven and Settles 2006; Melikov et al. 2011; Licina et al. 2014).
The velocity and the thickness of the human CBL can strongly differ under different ambient conditions and at different body heights. A definition of the CBL physical thickness is somewhat arbitrary since the transition from zero velocity at the surface (assuming no-slip condition) to the velocity in the free-stream outside the CBL is nonlinear. Homma and Yakiyama (1988) reported that at the room air temperature range from 19[degrees]C to 21[degrees]C (66.2[degrees]F to 69.8[degrees]F) the thickness of the CBL in the lower leg region varies from 0.01 to 0.03 m (0.03 to 0.01 ft), the flow regime is laminar, and the velocity varies from 0.10 to 0.15 m/s (0.33 to 0.49 ft/s). As the ambient air becomes entrained into the CBL, it spreads to about 0.075 m (0.25 ft) in the mid-chest region of a standing nude person (i.e., 0.15 m [0.49 ft] of the standing clothed person) with a turbulent flow regime and a wide velocity range from 0.05 to 0.25 m/s (0.16 to 0.82 ft/s). The human CBL is naturally warmer in the proximity of the skin and becomes cooler with increasing the distance from the body (Homma and Yakiyama 1988). The distance at which the temperature of the air enveloping the human body equals or closely approaches the temperature of undisturbed room air defines the outer edge (thickness) of a temperature boundary layer (TBL). Several researchers investigated the air temperature distribution around the human subject by means of infrared thermography (Homma and Yakiyama 1988; Voelker et al. 2014) and thermocouples (Melikov and Zhou 1996).
The development of the CBL and TBL is influenced by a range of factors. Apart from the room air temperature (Licina et al. 2014) and body posture (Clark and Toy 1975; Homma and Yakiyama 1988; Clark and de Galcina-Goff 2009; Licina et al. 2014), important parameters are clothing insulation (Homma and Yakiyama 1988; McCullough et al. 1994; Licina et al. 2014), table positioning (Bolashinkov et al. 2010; Licina et al. 2014), and room ventilation flow (Melikov and Zhou 1996; Liu et al. 2009; Licina et al. 2015a). In addition, the human respiration flow affects the air distribution around the human body (Ozcan et al. 2005; Melikov and Kaczmarczyk 2007; Tang et al. 2013). Our previous findings (Licina et al. 2014; Licina et al. 2015a, 2015b) investigated the influence of the room air temperature, body posture, clothing insulation, table positioning, and ventilation flow on the development of the human CBL and its ability to transport the pollution around the human body. That work provides a basis for the present study that focuses on studying the relationships between the airflow characteristics (velocity and temperature) and concentration distribution in the microclimate around a thermal manikin in relation to a specific set of personal (body posture, clothing insulation, and table positioning) and environmental factors (room air temperature and ventilation flow).
It should be emphasized that the present study relies on the results that have already been published in the literature (Licina et al. 2014; Licina et al. 2015a, 2015b) and adds a new set of experimental results to enhance overall understanding in the field of airflow characteristics (velocity and temperature) and pollution distribution in the occupied spaces.
Human CBL as a Transport Mechanism for Ambient Air Pollution
Apart from contributing to the convective heat loss from the human body, the CBL has an ability to transport the pollution/ clean air from the lower room levels up towards the breathing zone. Its ability to transport the pollution around the human body is especially pronounced in rooms that operate with little air mixing, such as rooms equipped with displacement air distribution, where the spatial variability of the pollution is high. A number of studies in the past showed that the existing concentration may be elevated in the occupied spaces, compared to the concentration of the room surroundings, and this phenomenon is known as a "personal cloud" (Rodes et al. 1991; Bjorn and Nielsen 2002). Therefore, it is obvious that the well-mixed mass balance models may underestimate human exposure prediction, which puts the primarily focus on evaluation of concentrations in the occupied spaces, rather than evaluation of average space concentrations.
Lewis et al. (1969) documented that content of microorganisms is substantially higher (30% to 400%) in the microenvironment of the nude standing man than in the ambient air. Melikov (2004) reported that the biggest portion of inhaled polluted air for sedentary person originates from the human CBL. Similarly, Zhu et al. (2005) reported that in a stagnant indoor environment, a person inhales air that mostly comes from the lower part of the CBL. Inhaled pollutants can be entrained by the CBL from the surrounding air or they can originate from the human body itself (i.e., bioeffluents). Therefore, it is important to study the influence of different source locations on pollutant concentration in the breathing zone. In addition, factors that influence the development of the CBL, such as room air temperature, body posture, and table positioning would also impact the pollution distribution around the human body, and, therefore, it is necessary to establish the mutual relationships between them and personal exposure. Furthermore, the surrounding environment and airflow patterns can modify the CBL (Melikov and Zhou 1996; Liu et al. 2009; Licina et al. 2015a), which makes it important in studying pollutant distribution around the human body. Rim and Novoselac (2009) examined the influence of two different indoor environments on the transport of pollutants in the proximity of a thermal manikin. It was found that in the stratified environment, the inhaled concentration of pollutants located 0.5 m (1.64 ft) behind the occupant and 0.15 m (0.49 ft) above the floor was four times higher than the ambient concentration. In this case, the CBL at the lower leg entrained the pollution and transported it to the breathing zone. In the same study, the authors found that in a highly mixed environment, the role of the CBL diminishes and the concentration around the occupant is more uniform. Similar findings have been reported in an additional set of numerical simulations performed by Rim and Novoselac (2010). These studies suggest that understanding interactions between the ventilation flow and the CBL are important for understanding the pollution spread around the human body. From the literature reviewed, it is obvious that there is a need for establishing relationships between airflow characteristics and pollution distribution around the human body.
Experimental Facilities and Apparatus
The air velocity measurements were performed in an environmental chamber with dimensions 11.1 x 8 x 2.6 [m.sup.3] (36.42 x 26.25 x 8.53 [ft.sup.3]). The chamber was ventilated with a low ventilation rate (1 ach), and the air was introduced through six displacement ventilation diffusers at the floor level and exhausted via six ceiling-mounted grills (Figure 1, left). A non-breathing and non-sweating thermal manikin was used to resemble a realistic human body. The manikin had a complex female body shape of 1.23 m (4.04 ft) height in the sitting and 1.68 m (5.51 ft) height in the standing posture. The total sensible heat load released from the manikin was equal to the heat loss from an average human body in a state of thermal comfort, for the each scenario studied. When the manikin was dressed, we used a tight-fitted clothing (0.7 clo) described in Table 1. The thermal manikin was calibrated prior to experiments.
[FIGURE 1 OMITTED]
The manikin was positioned in the center of the chamber. To measure instantaneous velocity vectors in front of the manikin, particle image velocimetry (PIV) was used. The PIV system, described by Licina et al. (2014) consisted of a laser, synchronizer, computer, and CCD camera with an image acquisition frequency of 10 Hz. The mixture of air and olive oil particles created a neutrally buoyant composition that was used as a seeding material. The olive oil particles were atomized in a six jet atomizer (Model 9306; TSI Inc., Shoreview, MN) and released at the constant emission rate.
The impact of the ventilation flow on the velocity distribution in the breathing zone and on personal exposure was investigated with a custom-made airflow generator that supplied a uniform isothermal airflow against the surface of the manikin. The airflow generator (1.8 x 1.0 x 0.2 [m.sup.3]) (5.91 x 3.28 x 0.66 [ft.sup.3]) had a fine-tuning frequency regulator capable of providing a desired velocity within a range 0 to 1 m/s (0 to 3.28 ft/s). To create a uniform flow from four different directions with respect to the manikin, the airflow generator was positioned at four locations, as shown in Figure 1 (right). Several uniformity tests were undertaken to assure that the airflow generator provided a uniform flow at a constant target velocity close to the surface of the manikin. More details about the airflow generator can be found in Licina et al. (2015b). The impact of the ventilation flow on personal exposure was assessed when the pollution originated from the feet of the manikin. As a source of pollution, we used the same neutrally buoyant mixture of air and olive oil particles that was used to seed the flow in PIV experiments.
The air temperature and pollution concentration in front of the thermal manikin were measured in a climate chamber with dimensions 4.7 x 6.0 x 2.5 [m.sup.3] (15.42 x 19.69 x 8.2 [ft.sup.3]) shown in Figure 2 (top). The chamber was ventilated with an upward piston flow (100% outdoor air) supplied at a low velocity through the floor, which was built of a porous sheet with a steel floor grating. The chamber was constructed to provide equilibrium between the room air temperature and the mean radiant temperature. Below the manikin that was seated in the center of the chamber, a wooden horizontal plate (2.0 x 1.54 [m.sup.2] [6.56 x 5.05 [ft.sup.2]]) was placed to prevent the supply airflow from affecting the CBL induced by the thermal manikin. The velocity measured with omnidirectional thermal anemometers (SENSOR, accuracy [+ or -]0.02 m/s [[+ or -]0.07 ft/s]) at numerous locations around the unheated manikin was less than 0.05 m/s (0.16 ft/s), which indicated that quiescent indoor conditions had been achieved, as suggested by Murakami et al. (2000). The air was exhausted from the chamber through the reduced free flow area of the ceiling (2.4 x 2.4 [m.sup.2]) (7.87 x 7.87 [ft.sup.2]) located directly above the manikin (Figure 2, top). During the measurements, the thermal manikin was the only heat source present in the chamber. A minimal vertical thermal stratification of 0.07 K/m (0.038[degrees]F/ft) was recorded during the experiments.
The air temperature was measured with thermistors with time constant of 0.16 s (63% value) and the mean sampling frequency of 16 Hz. The sensors were not overheated by the measuring current (5 [micro]A), which gave dissipation power of 2.4 [micro]W (8.19 [micro]Btu/h). The sensors were calibrated prior to the experiments with a reference to the medical (mercury) thermometer, and the measuring error was less than [+ or -]0.1 K ([+ or -] 0.18[degrees]F). Figure 2 (bottom) shows the air temperature measurement locations along the horizontal distance in the breathing zone of the manikin. In some experiments, the air temperature was also measured at the chest and stomach in a same way as in the breathing zone. The surface temperature was measured by means of the thermal manikin's control system, which calculated the average surface temperature of the individual body segment by measuring resistance of the nickel wire placed below the skin. Therefore, the surface temperature at the particular point of interest was determined based on the average surface temperature of that body segment.
[FIGURE 2 OMITTED]
The concentration profile was measured in a similar way to the temperature profile in the breathing zone but with more coarse sampling distances (see Figure 2, bottom). The first sampling point was located at the upper lip of the manikin (Melikov and Kaczmarczyk 2007) at 0 mm (0 ft) distance from the surface, to represent the sampling of inhaled air and personal exposure. The tracer gas was sampled through the sampling tubes and sent to two calibrated Innova multi-gas samplers (Model 1303) and analyzers (Model 1312) placed outside the chamber. A source of pollution was simulated with nitrous oxide ([N.sub.2]O) tracer gas that was injected through a sponge ball (diameter 0.05 m [0.16 ft]) at a steady emission rate. The [N.sub.2]O is approximately 1.5 times heavier than the room air; however, the resulting density difference was considered insignificant when the pollution source was released within the CBL. In our study, the gaseous pollution was released at the chest and groin of the manikin.
Experimental Design and Analyses
The influence of the room air temperature on the airflow characteristics in the breathing zone of the manikin and its personal exposure to gaseous pollution was examined under two temperatures 20[degrees]C and 26[degrees]C (68[degrees]F and 78.8[degrees]F). To examine the influence of the body posture on airflow characteristics and personal exposure to gaseous pollution, the manikin was (1) standing, (2) sitting upright, (3) leaning forward, and (4) leaning backwards. The airflow characteristics were examined when the manikin was unclothed and dressed in a tight/ thin outfit, while personal exposure to gaseous pollution was examined when the source was located at the chest outside and inside of the T-shirt. Furthermore, the presence of the table was examined when the front edge of table was positioned at 10 and 0 cm (0.33 and 0 ft) distance from the abdomen of the manikin. Finally, the influence of ventilation flow on personal exposure to particles released at the feet was studied (1) when the manikin was heated and unheated, (2) in relation to the direction of the laminar airflow, and (3) for three velocities supplied through the airflow generator: 0.175, 0.30 and 0.425 m/s (0.57, 0.98 and 1.39 ft/s). Impact of body posture, clothing insulation, table positioning, and ventilation flow was examined at the room air temperature of 23 [degrees]C (73.4 [degrees]F). More experimental details are given in Table 1.
In this study, each measurement location was aligned with the central vertical plane of the manikin and all the results presented are averaged values. Total measurement time for the velocity measurements took close to 1 minute, which corresponds to 540 image pairs. It was found that this number of image pairs was a representative number above which variation in velocity became insignificant. The air temperature measuring time at each point was 150 s. For the temperature measurements, one sensor was used at the chest, stomach, and mouth, and was precisely moved from the manikin by means of a custom-built traverse system controlled from outside of the chamber. Thus, the experimental time was drastically shortened since the steady conditions were not disturbed in-between two consecutive measurements. Total sampling time for measurements of personal exposure to particulate pollutants from the feet was 120 s. The pollutant transmission was considered only through airborne particles within the optical diameter range from 0.5 to 0.65 [micro]m (1.64 to 2.13 [micro]ft). When the pollution originated from the feet, the results were recorded and averaged over 120 seconds (sampling rate 1 Hz). Total sampling time for measurements of personal exposure to gaseous pollutants took 3 hours, which corresponded to 45 samples of gas in each point. It was found that 45 samples was a representative number above which the variation in the concentration became insignificant (<5%). A very low amount of [N.sub.2]O was found in the supply air due to leakage in ducts that was subsequently deducted from the results to minimize measurement inaccuracy.
RESULTS AND DISCUSSION
This section presents the results and discusses the relationships between airflow characteristics in the occupied spaces and personal exposure. These results are presented and discussed as a function of several parameters studied such as the room air temperature, body posture, clothing insulation, table positioning, and ventilation flow. The practical implications are discussed in relation to the optimal air distribution in occupied spaces with emphasis on indoor air quality, thermal comfort, and energy.
The Comparison of the Velocity, Temperature, and Concentration Profiles
Figure 3 presents temperature, velocity, and concentration profiles in the breathing zone of a seated thermal manikin at the room air temperature of 23[degrees]C (73.4[degrees]F). As seen, typical velocity and temperature profiles are developed that are inherent for airflows near a vertical heated surfaces. The highest velocity gradients are recorded within 10 to 20 mm (0.03 to 0.06 ft) distance from the mouth due to increased convective heat transfer. After reaching its peak at 17 mm (0.055 ft) distance from the surface, the velocity steadily dropped until it reached the velocity of undisturbed room air (not shown), which is defined with a value of 0.05 m/s (0.16 ft/s) in our study. The temperature decrease was also the highest within 10 to 20 mm (0.03 to 0.06 ft) next to the surface due to increased heat exchange but dropped at substantially slower pace beyond that distance. In our study, the thickness of the TBL extended to a distance where the air temperature dropped 90% from the temperature of the surface of the body to the temperature of the room air. This suggests that the thickness of the CBL is substantially bigger than the thickness of the TBL. The subsequent results of this study show that both the velocity and temperature profiles near the heated surface can be modified by a number of factors such as height of the body, room air temperature, body posture, clothing insulation, table positioning, and ventilation flow.
[FIGURE 3 OMITTED]
The results in Figure 3 (bottom) show the impact of pollution sources located at the chest and groin on development of the pollution boundary layer in the breathing zone of the manikin. For both source locations, the highest pollution concentration was achieved at the surface of the mouth. As seen, the concentration profile cannot be easily defined, which is not the case with the velocity and temperature profiles. This is because the pollution concentration profile depends on both the airflow characteristics of the CBL and the source location. As seen in Figure 3 (bottom), when the pollution was emitted at the chest, the sharp concentration decay profile occurred near the mouth. When the pollution originated at the groins, the concentration dropped relatively little until it reached the region of unpolluted surrounding air. These results suggest that the human CBL is important from the cross-contamination point of view because there is a discrepancy between the extent to which the CBL spreads around the human body and the extent to which the pollution spreads within the CBL. A general distance at which the interaction between CBL flows of two occupants occurs is 0.4 to 0.5 m (1.31 to 1.64 ft) (Clark and de Calcina-Goff 2009). Therefore, as the extent to which pollution spreads within the CBL can vary, the interaction between convection flows of occupants positioned close by may not cause cross-contamination. This is crucial to consider in densely occupied environments (schools, offices, vehicles, etc.), where people are physically close to one another.
[FIGURE 4 OMITTED]
Influence of the Room Air Temperature
The room air temperature is important not only from the point of view of thermal comfort, but also because it impacts the airflow characteristics around the human body and the pollution distribution. Figure 4 presents the results of peak velocity distribution with the height of the sitting manikin (left), as well as the air temperature and pollution distribution in the breathing zone of the thermal manikin when the room air temperature setpoint was kept at 20[degrees]C and 26[degrees]C (68[degrees]F and 78.8[degrees]F) (right and bottom, respectively). As seen, at the lower room air temperature, the air temperature drop was higher, while the CBL had a higher velocity, which increased its ability to transport the pollution to the breathing zone. The increase of the room air temperature from 20[degrees]C to 26[degrees]C (68[degrees]F to 78.8[degrees]F) reduced the peak velocity in the breathing zone from 0.24 to 0.16 m/s (0.79 to 0.52 m/s) (Figure 4, left), while the same temperature change reduced the ability of the CBL to transport the pollution from the chest by 30%, because the inhaled concentration dropped from 908 to 639 ppm (Figure 4, right). Furthermore, the increased room air temperature also increased the spread of the pollution within the CBL, which could increase the probability of cross-flow and cross-contamination among occupants. In practical terms, the pollution emitted from the groins or chest of a seated occupant is pulled upwards by the CBL and, due to its increased diffusion to the surroundings at elevated room air temperatures, it may more easily end up in the inhalation zone of an occupant seated nearby.
On the other hand, if the source of clean air resides in the lower levels of the room, the CBL can entrain it and increase the amount of clean air being inhaled. Rooms equipped with displacement air distribution are designed based on this principle. Thereby, the lower temperature of the supply air will induce the stronger CBL, which will further increase the inhaled air quality. In practice, however, the displacement ventilation systems are often coupled with chilled ceilings (Riffat et al. 2004) that allow higher dry-bulb temperature design without compromising thermal comfort (Jeong and Mumma 2006). As a result, the ability of the CBL to transport the clean air to the breathing zone decreases, as the chilled ceiling increases the radiant heat loss from the human body up to 50% (Kulpmann 1993). This reduces the human convective heat loss (weaker CBL), which further reduces the amount of the clean air transported to the human breathing zone. These findings suggest that the room air temperature can be used to control the amount of the transported pollution or clean air to the breathing zone. In practice, however, the room air temperature setpoint usually aims to satisfy the thermal comfort requirements without considering the human exposure.
Influence of the Body Posture
Licina et al. (2014) reported that the standing occupants create a narrow CBL in front of the body (0.09 m [0.3 ft]) which is five times less thick compared to sedentary occupants (0.45 m [1.48 ft]) at 20[degrees]C (8[degrees]F) room air temperature. The narrower CBL has a smaller entrainment area for the surrounding air, which suggests that it is likely to entrain a smaller amount of polluted/clean surrounding air and transport it to the breathing zone. In parallel, due to a narrow CBL of a standing person, the transient flow of exhalation can more easily penetrate the CBL and spread across the space. Understanding the airflow and pollution transport around a seated person is more beneficial as most of the building occupants are predominantly seated during the day (Pokora and Melikov 2014).
Figure 5 shows the results of air velocity and temperature (top) and pollution distribution (bottom) in the breathing zone of the thermal manikin for different body postures. As seen, the standing manikin created a relatively similar velocity profile in the breathing zone compared to the seated manikin, within 140 mm (0.46 ft) distance from the mouth. Changing the inclination angle of a seated body showed more prominent influence on the velocity in the breathing zone. Changing the sitting posture from upright to 25[degrees] backward body inclination simultaneously increased the peak velocity in the breathing zone by 8.8% and the mean air temperature by 0.5 K (0.9[degrees]F) (Figure 5, top) and personal exposure by 15.5% when the pollution was located at the chest. In addition, when the manikin was leaned backwards, the pollution spread more compared to an upward body posture. In practical terms, for an optimal delivery of the personalized ventilation air supplied from the front of the occupant, the backward seated body inclination will require a higher supply flow rate to penetrate the breathing zone due to increased velocity of the CBL, compared to the forward body inclination. In addition, the pollutants that originate from the human body or those that are entrained by the CBL will be more easily transported to the breathing zone if the occupant is leaned backwards. The findings obtained based on three seated body postures suggest that the seated body inclination angle could be used to control the amount of the transported pollution/clean air to the breathing zone.
The seated body inclination angle is challenging to control in practice. Considering predominantly static computer workstations and furniture, the majority of the occupants in the office need to adapt to them visually and physically. Occupants doing computer-related work spend about 75% of the time in forward-leaning or upright-seated positions (Dowell et al. 2001). On the other hand, the research has shown that leaning backwards is the most preferred sitting posture because it offers several health benefits (Andersson and Ortengren 1974). In practice, it is possible to design a dynamic workstation with an ability to position the furniture and computers at the distance that will optimally control the strength of the convection flow as the occupant changes its posture. For instance, the occupant that is inclining backwards simultaneously moves the table surface towards itself, along with the computer display, keyboard, and personalized ventilation air terminal device. For an optimal design of the personalized ventilation, the supply diffusers can be also designed in such a way to follow the occupants' body movement at the table in order to create a desired microclimate around the human body.
[FIGURE 5 OMITTED]
Influence of the Clothing Thermal Insulation
Similar to the room air temperature, the clothing insulation usually aims to satisfy thermal comfort requirements as well as cultural aspects, which sometimes can be counterproductive to human body thermoregulation. Figure 6 (top) presents the air velocity distribution in the breathing zone and air temperature distribution at the stomach of the nude and clothed manikin. As the clothing insulation naturally reduces the heat loss from the human body, it also reduces the velocity of the CBL. These findings, however, need to be carefully interpreted, because the results of this study are obtained at the fixed room air temperature (23[degrees]C [73.4[degrees]F]). In practice, the clothing insulation decreases at elevated room air temperatures, while the increase in clothing insulation occurs when the indoor temperature setpoint decreases. As earlier seen in Figure 4, lowering the room air temperature from 26[degrees]C to 20[degrees]C (78.8[degrees]F to 68[degrees]F) increases the velocity of the natural convective flow around the human body. This contradicts the velocity reduction due to increased clothing insulation. The results show that adding the clothing insulation to a nude body reduced the peak velocity in the breathing zone by 17% (from 0.205 to 0.17 m/s [0.67 to 0.56 ft/s]). In addition, the body posture and presence of the chair can affect the insulation level provided by the clothing. For instance, a seated occupant has lower thermal insulation due to compression of air layers within the clothing (ASHRAE 2013). This decrease in insulation can be compensated for by the additional insulation provided by the chair. In addition, body movement decreases clothing insulation by increasing air movement within the clothing, which was not the focus of this study. Therefore, more research on this topic is needed that focuses on a mutual influence of factors such as clothing and chair insulation and design, room air temperature, body posture, activity level, etc.
[FIGURE 6 OMITTED]
Our results suggest that the clothing insulation could also be important for personal exposure considerations. As the increase of the clo value decreases the strength of the CBL, it also diminishes its ability to transport the pollution. On the other hand, clothing itself can be a major source of pollution. In addition, the source location with respect to the clothing can play important role. As seen in Figure 6 (bottom), the pollution emitted below the T-shirt substantially reduced personal exposure (from 807 to 588 ppm), compared to the case when pollution is released outside the clothes. This suggests that the additional clothing could serve as a protection against the airborne pollutants released from the human skin (skin oils, bioeffluents, etc.). In some cases, the clothing surface can reduce concentration of some pollutants in the breathing zone, such as in case of chemical reactions between ozone and skin/ clothing surfaces (Rim and Novoselac 2009). In this case, ozone reacts with the clothing and the CBL becomes depleted of ozone and enriched with ozone reaction products. On the other hand, as the thickness of the CBL increases with the increase of clothing insulation (Homma and Yakiyama 1988), this may intensify mixing of the pollution within the CBL and increase the pollution spread across the room. What needs to be studied in the future is the effect of air movement between the skin and the clothing on personal exposure (i.e. the effect of different clothing designs). Although the clothing insulation cannot be easily controlled in practice in relation to exposure control, the results indicate that it should be carefully considered in full-scale experiments because it modifies the airflow characteristics close to the body surface. The same is important for numerical predictions, since many researches tend to neglect simulating clothing insulation level and the hair on the head, and thus disregard its impact on the CBL formation. In a near future, it is reasonable to expect appearance of multipurpose clothing designs that, apart from reducing the heat loss by providing an additional insulation, can also cool and disinfect the surrounding air, provide thermal comfort, and enhance inhaled air quality.
Influence of the Table Positioning
Figure 7 (top) demonstrates the influence of the table positioning on the velocity in the breathing zone and air temperature profile at the chest of the manikin. As seen, the presence of the table can greatly affect airflow characteristics in the breathing zone. Adding the table in front of an occupant at 10 and 0 cm (0.33 and 0 ft) distances reduced the velocity from 0.17 to 0.141 m/s (0.56 to 0.46 ft/s) and 0.111 m/s (0.36 ft/s), respectively. Although the velocity decreased due to the presence of the table, the ability of the CBL to transport the pollution did not decrease. This can be attributed to a blocking effect of the table and pollution source location. As seen in Figure 7 (bottom), the table located at 10 cm (0.33 ft) distance from the occupant increased personal exposure regardless of whether the pollution was located below or above the table height. This again suggests that for personal exposure it is important to consider not only the velocity of the CBL but also how the pollution is distributed within the CBL. Closing the gap between the table and the occupant reduced the air temperature in front of the chest and created a new and weaker CBL that reduced the transport of pollution released from the chest (by 63.5%) and especially the transport of pollution released from the groin (by 80%). A table contiguous to occupant's body could be beneficial from the point of view of indoor air quality, if the source of clean air is designed to approach the breathing zone from the front of the body (e.g., personalized ventilation, stratum ventilation.) or from the above (e.g., downward ceiling diffuse ventilation, active chilled beams) due to an easier penetration into the weaker CBL. On the other hand, our results suggest that having a source of clean air that entrains the CBL at the lower leg region (e.g., upward piston ventilation flow, displacement ventilation, underfloor ventilation), the distance of 10 cm (0.33 ft) between the edge of the table and occupant's abdomen would enhance the air quality, compared to the case with no table or especially if the table touches occupant's abdomen.
[FIGURE 7 OMITTED]
Based on our findings, it is possible to design and position the table in such a way to be optimized for the desired application, especially in rooms where the location of the pollution or clean air is known. In practice, occupants are not expected to be positioned in a way that their abdomen keeps a fixed distance from the table all the time. It is possible, however, to control the rising convection flow by attaching the additional board with spring system or sponge/foam plate to the edge of the table that gently presses the abdomen of the occupant (Bolashikov et al. 2010). In the offices of the future, it is reasonable to expect evolution of ergonomic design that will make the workstations more dynamic and capable of providing individual micro-environmental control that can be used for enhanced airflow and pollutant distribution around the human body.
Influence of the Ventilation Flow
Figure 8 shows the results of a mean peak velocity and turbulence intensity in the breathing zone of the sitting thermal manikin placed in (1) quiescent indoor environment and (2) uniform surrounding airflow field supplied from three different directions and magnitudes. As shown in Figure 8, the increased velocity of the assisting flow from below surprisingly decreased the mean peak velocity in the breathing zone. This is attributable to a blocking effect of the chair, as already reported by (Licina et al. 2015a). Figure 9 presents the results of mean personal exposure to feet-released particles when the sitting, non-breathing manikin is placed in (1) a quiescent indoor environment and (2) a room ventilated with uniform surrounding flow from four different directions and magnitudes. As seen in Figure 9, comparing the cases "No CBL" and "CBL" suggests that the CBL was able to transport the pollution from the feet to the breathing zone, which is in line with several other studies. Exposing the manikin to assisting flow from below and transverse flow from the back linearly decreased the exposure with increased supply air velocity. The relationship between personal exposure and the supply air velocity was not linear when the manikin was exposed to opposing flow from above. Increasing the velocity of downward flow from 0.175 to 0.3 m/s (0.57 to 0.98 ft/s) increased personal exposure by 85% due to increased particle residence time in the breathing zone, which is attributed to the collision of downward ventilation and the upward CBL flows (Licina et al. 2015a). This can also be seen from Figure 8 that shows lower peak velocity and increased turbulence intensity in the breathing zone due to collision of upward CBL and downward ventilation flow. Finally, the manikin was exposed to negligible amount of particles from the feet when transverse flow from the front was applied, even at the minimum velocity of 0.175 m/s (0.57 ft/s). This is because transverse flow can easily penetrate the upward CBL movement, as suggested by Licina et al. (2015b). The results presented in Figure 9 show that personal exposure can be controlled/minimized by proper understanding of the airflow interactions in occupied spaces.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
The results in Figure 9 suggest that the downward ventilation approach may be inefficient in terms of reducing personal exposure, based on limited range of velocities studied. Since a high velocity of 0.425 m/s (1.39 ft/s) is required to penetrate the thermal plume and protect the occupant from the rising pollution, the downward ventilation approach can cause thermal discomfort and increase energy input. Although our results do not directly imply it, the uniform ventilation flow supplied horizontally from behind a person could also be inefficient as it can increase the exposure to pollutants approaching from in front of a person (e.g., pollution from an occupant seated opposite that is talking, coughing, sneezing, etc.), as suggested by Licina et al. (2015a). Our results obtained with a non-breathing thermal manikin indicate that when the airborne pollution is located on the human skin/clothing, supplying the clean air horizontally in front of the person may most effectively reduce personal exposure. If the supplied airflow is maintained at the velocity of 0.175 m/s (0.57 ft/s), this airflow direction can also provide thermal comfort at minimum energy input. A further increase of the velocity at the room air temperature below 23[degrees]C (73.4[degrees]F) may also cause thermal discomfort (ISO 2005; ASHRAE 2013). Nevertheless, when the room air temperature increases, the flow supplied horizontally from the front can provide thermal comfort even at elevated air movements, which is one of the strategies to save energy recommended by some standards. An exception that allows even higher supply air velocity is granted if the velocity of the supplied air can be individually controlled.
The results from Figure 9 demonstrate that in specific scenarios, increasing the velocity of the supply air can increase personal exposure if the airflow interactions in occupied spaces are not taken into account. The results also show that a further increase of the velocity beyond a certain threshold may not provide additional benefit in terms of personal exposure reduction (case transverse flow from the front). This suggests an existence of a nonlinear relationship between the velocity of supplied ventilation flow and personal exposure in certain indoor environments. To minimize personal exposure without compromising thermal comfort and energy savings, it is necessary to understand airflow interaction in the occupied spaces and to be able to control it. The current total volume air distribution principles commonly used in buildings are generally not able to satisfy these requirements. Therefore, a change in the approach to how rooms are ventilated is clearly needed. Based on our limited set of measurements, a need for an advanced air delivery systems can be suggested that can supply the air from the front side of the person at the velocity that can be optimally controlled to satisfy the requirements for inhaled air quality, thermal comfort, and energy. One such system that can be implemented in practice is personalized ventilation system that can timely supply the air with optimal direction and quantity to the breathing zone for enhanced inhaled air quality and minimized energy input.
SUMMARY AND STUDY LIMITATIONS
This study presented the summary of experimental measurement of relationships between airflow characteristics around the human body and personal exposure to two types of airborne pollutants. These relationships were examined in relation to a specific set of personal (body posture, clothing insulation, and table positioning) and environmental factors (room air temperature and ventilation flow). The main findings of this study can be summarized as follows:
* The CBL generated by the thermal manikin influenced the airflow characteristics and pollution distribution in the breathing zone. When the pollution source originated at the feet, groin, or chest, the CBL contributed, to a great extent, to an increase of personal exposure by transporting the pollution to the breathing zone.
* The room air temperature affected the airflow characteristics around the thermal manikin and the pollution transport. At the low room air temperature of 20[degrees]C (68[degrees]F), the CBL around the seated manikin had the peak velocity of 0.24 m/s (0.79 ft/s). Increasing the room air temperature to 26[degrees]C (78.8[degrees]F) reduced the peak velocity of the CBL to 0.16 m/s (0.52 ft/s). Reducing the room air temperature also intensified the transport of the pollution to the breathing zone and reduced the pollution spread across the room.
* The standing manikin created a relatively similar velocity profile in the breathing zone compared to the sedentary manikin. A backward inclination of a seated manikin increased the velocity and air temperature in the breathing zone. In addition, personal exposure increased when the pollution originated from the chest and groin of a manikin leaned backwards in the chair. The opposite effect was observed when the seated manikin was leaning forward.
* Addition of clothing thermal insulation of 0.7 clo had a measurable influence on the airflow in the manikin's breathing zone, and it should be carefully considered in CFD simulations and in physical experiments on exposure. Adding clothing to a nude body can reduce the peak velocity in the breathing zone from 0.205 to 0.17 m/s (0.67 to 0.58 ft/s). The clothing insulation may be important for personal exposure consideration because it behaves as a barrier for the pollutant released from below the clothing and reduced personal exposure by 27%, for a specific clothing ensemble.
* The presence of the table blocked the rising manikin's convection flow and reduced its peak velocity from 0.17 to 0.111 m/s (0.56 to 0.36 ft/s). Keeping the distance between the table and the manikin at 10 cm (0.33 ft) can potentially increase the amount of pollution/clean air transported from beneath. Closing the gap between the edge of the table and the manikin could be used to reduce the pollution transport to the breathing zone.
* The direction and magnitude of uniform ventilation flow proved to be important for airflow distribution in the breathing zone of the thermal manikin, and it can be used to reduce personal exposure to airborne pollutants and thermal discomfort, as well as to save energy. Under the specific set of conditions studied, the downward airflow from above created the most unfavorable velocity field that increased personal exposure by 85%. The most favorable airflow pattern in preventing feet pollution from reaching the breathing zone was transverse flow from the front as it minimized the exposure at the minimum supply air velocity. This direction should therefore be considered in the ventilation design, when the source location originates from the lower CBL.
* The limited set of experiments suggest that there may be a nonlinear dependence between the amount of the supplied air to the room and personal exposure. Rather than relying on the outdoor air supply rate as a sole indicator of indoor air quality, more attention should be paid to understanding the room air distribution.
Practical implications of the results obtained should be carefully interpreted, given a number of study limitations. The ability of the CBL to transport the pollution can drastically change in spaces with a high degree of air mixing, because the buoyant flows in occupied spaces and a nonuniform pollutant concentration is less prominent. Another limitation of the present study is that it considers personal exposure only to those small particles and gases that naturally trace the room airflows. Although it gives valuable information on small particle transport mechanisms around a human body, it is difficult to draw a general conclusion about the role of the human CBL in transporting large particles. Although the CBL has been shown to have an effect on personal exposure, in reality, a person does not always keep still, which is very likely to modify personal exposure. Nevertheless, there are many spaces where occupants perform sedentary tasks and move little during the day, which makes findings of this study relevant. Apart from the occupants' movement, the effect of the human respiration flow have been disregarded. The study done by Melikov and Kaczmarczyk (2007) reported that the temperature, the relative humidity, and the concentration of tracer gas measured in the air inhaled by the breathing thermal manikin were almost the same as those measured close to the upper lip of the non-breathing thermal manikin. Nevertheless, our measurements cannot indicate to what extend the results would be changed if the respiratory flow was applied. The expiration would certainly disturb the airflow in the breathing zone; however, the same would presumably quickly reestablish during inhalation phase. Hence, the airflow distribution, and thus, the pollution concentration in the breathing zone during inhalation phase would presumably not differ substantially from the results that we reported. Apart from these limitations, the direction for the future work should also include a more comprehensive set of indoor environmental conditions and factors, such as room air temperatures, ratios between convective and radiant heat exchange from the human body (including the effect of skin wetness, clothing levels, and body postures), velocities and airflow directions provided by mechanical ventilation, physical room configurations and furniture design, building occupancy types, exposure routes such as the one through dermal absorption, and types and location of pollution sources. Although some of these topics require further investigation, several important conclusions can be drawn.
For an adequate exposure description, future studies need to go through the paradigm shift from the coarse total room volume description of the environment to the description that takes into account flow interactions in the human microclimate. In addition, the paradigm shift is needed from a total volume to advanced air distribution that can minimize personal exposure, remove the pollution at the source, and provide each occupant possibility to control its own micro-environment, as suggested by Melikov (2012). For the advanced air delivery systems, understanding the CBL will be especially important since they can substantially modulate microclimate and create environments that are not well-mixed. Future research directions should therefore include developing technologies for enhanced micro-environmental control that can control the airflow and pollution distribution around the human body, locally exhaust or clean the pollution, enhance occupant-system interaction, create dynamic furniture/equipment, and provide detached/wearable micro-environmental systems.
Andersson, B.J., and R. Ortengren. 1974. Lumbar disc pressure and myoelectric back muscle activity during sitting. 1. Studies on an office chair. Scandinavian Journal of Rehabilitation Medicine.
ASHRAE. 2013. ANSI/ASHRAE Standard 55-2013, Thermal Environmental Conditions for Human Occupancy. Atlanta: ASHRAE.
Bjorn, E., and P.V. Nielsen. 2002. Dispersal of exhaled air and personal exposure in displacement ventilated rooms. Indoor Air 12:147-64. DOI: 10.1034/j.1600-0668.2002.08126.x.
Bolashikov, Z., A. Melikov, and M. Krenek. 2010. Control of the free convective flow around the human body for enhanced inhaled air quality: Application to a seat-incorporated personalized ventilation unit. HVAC&R Research 16:2.
Clark, R.P., and N. Toy. 1975. Natural convection around the human head. The Journal of the Physiology 244:283-93.
Clark, R.P., and M.L. de Galcina-Goff. 2009. Some aspects of the airborne transmission of infection. Journal of the Royal Society Interface 6:S767-82.
Craven, B.A., and G.S. Settles. 2006. A computational and experimental investigation of the human thermal plume. Journal of Fluids Engineering 128:1251.
Dowell, W.R., F. Yuan, and B.H. Green. 2001. Office seating behaviors: An investigation of posture, task, and job type. Proceedings of the Human Factors and Ergonomics Society 45th Annual Meeting, HFES, CA: Santa Monica.
Homma, H., and M. Yakiyama. 1988. Examination of free convection around occupant's body caused by its metabolic heat. ASHRAE Transactions 94(1):104-24.
ISO. 2005. ISO Standard 7730-2005, Ergonomics of the thermal environment--Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria. Geneva: International Organization for Standardization.
Jeong, J.W., and S. Mumma. 2006. Designing a dedicated outdoor air system with ceiling radiant cooling panels. ASHRAE Journal 48.
Kulpmann, R.W. 1993. Thermal comfort and air quality in rooms with cooled ceilings--Results of scientific investigation. ASHRAE Transactions 99(2):488-502.
Lewis, H.E., A.R. Foster, B.J. Mullan, R.N. Cox, and R.P. Clark. 1969. Aerodynamics of the human microenvironment. The Lancet 322_7609:1273-1277.
Licina, D., A. Melikov. J. Pantelic. C. Sekhar, and K.W. Tham. 2014. Experimental investigation of the human convective boundary layer in a quiescent indoor environment. Building and Environment 75:79-91.
Licina, D., A. Melikov, C. Sekhar, and K.W. Tham. 2015a. Human convection flow in spaces with and without ventilation: personal exposure to floor-released particles and cough-released droplets. Indoor Air; DOI:10.1111/ ina.12177.
Licina, D., A. Melikov, C. Sekhar, and K.W. Tham. 2015b. Human convective boundary layer and its interaction with room ventilation flow. Indoor Air 25: 21-35; DOI: 10.1111/ina.12120.
Liu, L, P.V. Nielsen, Y. Li, R.L. Jensen, M. Litewnicki, and J. J. Zajas. 2009. The thermal plume above a standing human body exposed to different air distribution strategies. Proceeding of Healthy Buildings 2009: 9th International Conference & Exhibition, September 13-17, 2009, NY: Syracuse.
Marr, D., T. Khan, M. Glauser, H. Higuchi, and Z.D. Jianshun. 2005. On particle image velocimetry (PIV) measurements in the breathing zone of a thermal breathing manikin. ASHRAE Transactions 111(2):299.
McCullough, E.A., B.W. Olesen, and S. Hong. 1994. Thermal insulation provided by chairs. ASHRAE Transactions 100(1):795-802.
Melikov, A.K., Z.D. Bolashikov, H. Nagano, C. Velte, and K. E. Meyer. 2011. Airflow characteristics at the breathing zone of a seated person: Active control over the interaction of the free convection flow and locally applied airflow from front for personalized ventilation application. Proceedings of Roomvent 2011, Norway, Trondheim, Paper ID: 278.
Melikov, A., and G. Zhou. 1996. Air movement at the neck of the human body. Proceedings of Indoor Air 1996, Nagoya, Japan, Vol.1, 209-14.
Melikov, A.K. 2004. Personalized ventilation. Indoor Air 14 (Suppl. 7):157--67.
Melikov, A., and J. Kaczmarczyk. 2007. Measurement and prediction of indoor air quality using a breathing thermal manikin. Indoor Air 17: 50-59.
Melikov, A.K. 2011. Advanced air distribution. ASHRAE Journal 73-78.
Melikov, A.K. 2012. Creating benefits for all in the design of the indoor environment. Yearly report of the Department of Civil Engineering, Technical University of Denmark, 19.
Murakami, S., S. Kato and J. Zeng. 2000. Combined simulation of airflow, radiation and moisture transport for heat release from a human body. Building and Environment 35:489-500.
Ozcan, O., K.E. Mayer, and A.K. Melikov. 2005. A visual description of the convective flow field around the head of a human. Journal of Visualization 8(1): 23-31.
Pokora, P., and A. Melikov. 2014. Occupant body movement and seat occupancy rate for design of desk microenvironment. Proceedings of Roomvent 2014, Paper ID 130, Sao Paolo, Brazil.
Riffat, S.B., X. Zhao, and P.S. Dohert. 2004. Review of research into and application of chilled ceilings and displacement ventilation systems in Europe. International Journal of Energy Research 28:257-86.
Rim, D., and A. Novoselac. 2009. Transport of particulate and gaseous pollutants in the vicinity of a human body. Building Environment 44:1840-49.
Rim, D. and A. Novoselac. 2010. Occupational exposure to hazardous airborne pollutants: effects of air mixing and source characteristics. Journal of Occupational and Environmental Hygiene 7: 683-92.
Rodes, C.E., R.M. Kamens, and R.W. Wiener. 1991. The significance and characteristics of the personal activity cloud on exposure assessment measurements for indoor contaminants. Indoor Air 1:123-45. DOI: 10.1111/ j.1600-0668.1991.03-12.x.
Settles, G. 2001. Schlieren and shadowgraph techniques: Visualizing phenomena in transparent media. Berlin, Germany, 2001: Springer-Verlag.
Tang, J.W., A.D. Nicolle, C.A. Klettner, J. Pantelic, L. Wang, A.B. Suhaimi, A.Y.L. Tan, G.W.X. Ong, R. Su, C. Sekhar, D.W. Cheong, and K.W. Tham. 2013. Airflow Dynamics of Human Jets: Sneezing and Breathing--Potential Sources of Infectious Aerosols. PLoS ONE 8(4): e59970. doi:10.1371/journal.pone.0059970.
Voelker, C., S. Maempel, and O. Kornadt. 2014. Measuring the human body's microclimate using a thermal manikin. Indoor Air. DOI: 10.1111/ina.12112.
Zhu, S., S. Kato, S. Murakami and T. Hayashi. 2005. Study on inhalation region by means of CFD analysis and experiment. Building and Environment 40:1329-36.
Student Member ASHRAE
Kwok Wai Tham, PhD
Arsen Melikov, PhD
Chandra Sekhar, PhD
Dusan Licina is a postdoctoral researcher in the department of Civil and Environmental Engineering, University California Berkeley, Berkeley, CA. Arsen Melikov is a professor in the International Centre for Indoor Environment and Energy, Department of Civil Engineering, Technical University of Denmark, Lyngby, Denmark. K.W. Tham is an associate professor and S.C. Sekhar is a professor in the Department of Building, School of Design and Environment, National University of Singapore, Singapore.
Table 1. Summary of the Experiments with 16 Different Scenarios Parameter Scenario Details/Comments Room air 20[degrees]C Nude manikin temperature (68[degrees]F) 26[degrees]C Nude manikin (78.8[degrees]F) Body posture Standing Nude manikin Sitting upright Sitting leaned 25[degrees] from the vertical forward axis Sitting leaned 25[degrees] from the vertical backwards axis Clothing Nude insulation Dressed Thin-tight clothing: Thin socks, thin tight trousers, thin tight short-sleeve T-shirt, underwear and socks Table No table positioning Table 10 cm Table dimensions, 1 x 0.8 x 0.75 (0.33 ft) m (3.28 x 2.62 x 2.46 ft) (W x L x H) Table 0 cm Table dimensions, 1 x 0.8 x 0.75 (0 ft) m (3.28 x 2.62 x 2.46 ft) (W x L x H) Ventilation Quiescent The room air velocity, <0.05 m/s flow * (0.164 ft/s) Assisting Uniform flow velocities, 0.175, 0.3, and 0.425 m/s (0.57, 0.98 and 1.39 ft/s) Opposing Uniform flow velocities, 0.175, 0.3, and 0.425 m/s (0.57, 0.98, and 1.39 ft/s) Transverse front Uniform flow velocities, 0.175, 0.3, and 0.425 m/s (0.57, 0.98, and 1.39 ft/s) Transverse back Uniform flow velocities, 0.175, 0.3, and 0.425 m/s (0.57, 0.98, and 1.39 ft/s) * In case of transverse flow supplied from the back, the velocity was not examined.
|Printer friendly Cite/link Email Feedback|
|Author:||Licina, Dusan; Tham, Kwok Wai; Melikov, Arsen; Sekhar, Chandra|
|Date:||Jan 1, 2016|
|Previous Article:||Reverse relief airflow prevention and building pressurization with a decoupled relief air damper in air-handling units.|
|Next Article:||Hot-water use in hotels: part 3--business hotel hot-water system monitoring results.|