Environ Sci Pollut Res DOI 10.1007/s11356-014-2544-1
CHEMISTRY IN A SUSTAINABLE SOCIETY
Indoor air quality (IAQ) assessment in a multistorey shopping mall by high-spatial-resolution monitoring of volatile organic compounds (VOC) M. Amodio & P. R. Dambruoso & Gianluigi de Gennaro & L. de Gennaro & A. Demarinis Loiotile & A. Marzocca & F. Stasi & L. Trizio & M. Tutino
Received: 10 July 2013 / Accepted: 9 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract In order to assess indoor air quality (IAQ), two 1week monitoring campaigns of volatile organic compounds (VOC) were performed in different areas of a multistorey shopping mall. High-spatial-resolution monitoring was conducted at 32 indoor sites located in two storehouses and in different departments of a supermarket. At the same time, VOC concentrations were monitored in the mall and parking lot area as well as outdoors. VOC were sampled at 48-h periods using diffusive samplers suitable for thermal desorption. The samples were then analyzed with gas chromatography–mass spectrometry (GC–MS). The data analysis and chromatic maps indicated that the two storehouses had the highest VOC concentrations consisting principally of terpenes. These higher TVOC concentrations could be a result of the low efficiency of the air exchange and intake systems, as well as the large quantity of articles stored in these small spaces. Instead, inside the supermarket, the food department was the most critical area for VOC concentrations. To identify potential emission sources in this department, a continuous VOC analyzer was used. The findings indicated that the highest total VOC concentrations were present during cleaning activities and that these activities were carried out frequently in the food department. The study highlights the importance of conducting both high-spatial-resolution monitoring and high-temporal-resolution monitoring. The former was able to identify critical issues in environments with a
Responsible editor: Constantini Samara M. Amodio : P. R. Dambruoso : G. de Gennaro (*) : L. de Gennaro : A. D. Loiotile : A. Marzocca : F. Stasi : L. Trizio : M. Tutino Department of Chemistry, University of Bari, Bari, Apulia Region 70126, Italy e-mail: [email protected]
complex emission scenario while the latter was useful in interpreting the dynamics of each emission source. Keywords Indoor air quality . Volatile organic compounds . Shopping mall . Articles . Food department . Storehouses
Introduction In recent decades, shopping malls have become such an integral part of daily urban life that they may be defined as multifunctional public urban spaces. People like to spend their leisure time in shopping malls, and even tourists will visit them on vacations. People use shopping malls for a variety of different activities (shopping, strolling, meeting people, etc.) in different hours of the day and in different days of the week. The climatically controlled, clean, and comfortable interior environment of shopping malls is often preferred over outdoor areas. Even if shopping malls are profit-oriented private properties, people can spend a whole day in them without doing any actual shopping. Therefore, these environments can be considered a new type of public space. Commercial buildings, such as offices, shopping malls, and mixed residential–commercial buildings, account for 20 % of all construction output in the EU and represent over 20 million m2 of floor space each year. In Italy, there are about 776 shopping malls covering an area (the area of sales and service) equal to 6.8 million m2 (Biasi 2012). Most contemporary shopping malls are microcities with street-like passageways full of shops, intersections, plazas, fountains, signs, and multilevel atria. A supermarket is usually at the heart of these shopping centers, and it is the largest and most populated store in the mall. This is the place offering the largest variety of articles and consumer goods and where different activities,
Environ Sci Pollut Res
such as food preparation and cooking, are carried out. In order to meet the growing demands of visitors, these newly designed and constructed malls have increased the number of services and recreational areas such as beauty salons, laundries, restaurants, playrooms, and coffeehouses. An increase in the number of visitors and the amount of time they spend in these environments has led to great interest in the indoor air quality (IAQ) of these environments. In recent years, numerous scientific studies have highlighted the fact that people who spend most of their time indoors are more exposed to pollution than those passing greater amounts of time outdoors (Bruno et al. 2008; Pegas et al. 2011). Yet, a wide spectrum of symptoms and illnesses can be related to nonindustrial indoor air pollution; hence, the quality of indoor air may have a considerable impact on public health (Jones 1999; Kotzias 2005; Wolkoff 2013). The phenomena of indoor air pollution can be influenced by many variables. For example, while pollutant concentrations may vary widely from one indoor environment to another, those in a specific environment may also vary as a function of location and time (Batterman et al. 2007). The extent of these variations depends on factors such as the emission characteristics of a source, an occupant’s behavior, and the microclimatic and ventilation conditions (UNI EN ISO 16000-1 2006). The detection of volatile organic compounds (VOC) is essential for assessing parameters for indoor air quality because of their ubiquitous presence in the atmosphere and their impact on human health. Recent epidemiological studies have demonstrated that VOC can be associated with disorders of many of the body’s organs (Chang et al. 2013). Short-term health effects include eye and respiratory tract irritation, headaches, dizziness, visual disorders, and memory impairment. Long-term health effects include damage to the liver, kidneys, and central nervous system. Some VOC are known to cause cancer in humans (Rumchev et al. 2007; Kampa and Castanas 2008; Zhou et al. 2011; Ramírez et al. 2012). Moreover, VOC have an important role in the “sick building syndrome” described by the World Health Organization (Brinke et al. 1998). Numerous studies conducted in nonresidential indoor environments have shown that indoor VOC concentrations often exceed those found outdoors (Bruno et al. 2008; Eklund et al. 2008). A preliminary assessment of the air quality of indoor environments such as shopping malls, which are sometimes quite large, involves identifying all areas of concern. The detection of potential sources can be facilitated with high-spatial-resolution monitoring campaigns using silent, low-cost samplers that are lightweight and without encumbrance. Diffusive samplers meet these requirements (UNI EN ISO 16000-1 2006; UNI EN ISO 16000-5 2007; UNI EN ISO 16017-2 2007; UNI EN 14412 2005). The use of thermal desorbable diffusive samplers further reduces the cost and time of analysis and allows automation by the use of an autosampler (Bruno et al. 2005).
Once the critical areas have been highlighted, high-temporalresolution monitoring can be conducted to obtain more in depth data and interpret the dynamics of the emission sources. The present work reports the results of two 1-week high-spatialresolution monitoring campaigns carried out to detect VOCs in a multistorey shopping mall. In order to assess the IAQ, indoor sites located in two storehouses and a supermarket were monitored using Radiello® diffusive samplers for thermal desorption and gas chromatography–mass spectrometry (GC–MS) analysis. At the same time, VOC concentrations were monitored in the mall and parking lot area as well as outdoors. Hence, it was possible to produce chromatic maps of TVOC (calculated according to ISO 16000-6:2011) concentrations to determine the emissions produced by articles and common goods present in the different areas and to identify the most critical zones. Finally, a high-temporal-resolution monitoring of VOC was conducted in the most critical area of the supermarket, and the results of this sampling enabled the identification of the potential emission sources.
Materials and method Study area and sampling sites The shopping mall investigated in this study is located in the suburbs of Bari, Italy. The mall has four floors altogether, and the area of each floor exceeds 20,000 m2. Three of these floors are underground. The second and third underground floors are used as car parks while the first underground floor is occupied by a supermarket, a department store, and other small shops. The monitoring campaigns were conducted on this floor, predominantly, inside the supermarket which covers an area of 10,000 m2 and in the two storehouses nearby whose total area is 5,500 m2. The mall is centrally air-conditioned, and smoking is not allowed. It opens at 9:00 am and closes at 9:00 pm every day. During each experimental period (which included Sunday), there were about 100,000 customers visiting the mall daily and about 3,000 clients visiting the supermarket daily. Two 1-week VOC monitoring campaigns were performed in the multistorey shopping mall. The first was conducted from 23 to 29 November 2010 and the second one from 8 to 16 September 2011. Each campaign was subdivided into three different periods with a sampling duration of 48 h. A monitoring grid was drawn on the map of the mall (supermarket and two storehouses) in order to obtain meshes of where to locate diffusive samplers. The most homogenous areas were selected by considering microenvironments with similar merchandise and lane width. Monitoring was planned taking into account the guidelines of the UNI EN ISO 16000-1 (2006) which describe the procedure (positioning, representativeness of the site, etc.) for carrying out a preliminary evaluation of the spatial distribution of volatile
Environ Sci Pollut Res
pollutants in indoor air. Moreover, the density of people was considered in the selection and location of the sites. These observations were made in order to obtain data as representative as possible of the mean concentration levels of pollutants in each area. It was not easy to define or to check the changes in air quality because of the large number of people that frequented these environments every day. For this reason, sampler placement took into account the distance from doors communicating with storehouses and potential heat sources. Figure 1 shows the 32 sampling sites (red square) in the different indoor microenvironments inside the supermarket. During the first monitoring campaign, a total of 96 air samples were collected. A second monitoring campaign was conducted in order to confirm findings obtained during the first campaign as well as to gain more insight into the critical areas highlighted. In particular, other 10 indoor sampling sites (blue squares in Fig. 2) and parallel outdoor sites were monitored. A total of 149 samples were collected during the second monitoring campaign. Figure 3 shows the distribution of the different departments in the monitoring areas. Sampling and analytical method VOCs were sampled with Radiello® diffusive samplers (Fondazione Salvatore Maugeri, Padova, Italy) suitable for Fig. 1 Map of the sampling sites during the first monitoring campaign
thermal desorption. The sampling system was made up of a cylindrical adsorbing cartridge housed coaxially inside a cylindrical diffusive body of microporous polyethylene 5 mm thick and average porosity 10±2 μm. The cartridges consisted of a stainless steel net cylinder, with 3×8 μm mesh opening and 4.8 mm diameter, packed with 350±10 mg of graphitized charcoal (Carbograph 4) with particle size of 35–50 mesh. Before sampling, the cartridges were conditioned and analyzed to verify the blank levels (Bruno et al. 2005, 2008). Each sampler was exposed for 2 days. After sampling, it was placed in a sealed glass tube and brought to the laboratory for analysis. The analyses were carried out using a thermal desorber (Markes International Ltd, Unity™) equipped with an autosampler (Markes mod. ULTRA™ TD) with 100 positions and coupled with a gas chromatograph (Agilent GC6890 Plus) and a mass selective detector (Agilent MS5973N). The thermal desorber provided a two-stage mechanism: first, the analytes were desorbed from the sample tube and refocused into a cold trap; second, they were desorbed from the trap and carried into the GC column (Bruno et al. 2005; UNI EN ISO 16017-2 2007). To quantify the samples, the calibration curves were prepared by injecting 1 μL of standard solutions (Ultra Scientific Cus-5997) into a tube; the spiked adsorbent tubes were then thermally desorbed in the same conditions of time, gas flow, and split ratio as the
Environ Sci Pollut Res Fig. 2 Map of the sampling sites during the second monitoring campaign. The new sites are in blue
samples. The sampling rates, Q values supplied by the manufacturer, were used to calculate the real concentration of the compound in the atmosphere (C) by GC quantification of the analyte’s mass, m. Q was the function of the diffusive coefficient D, which was the thermodynamic property of each chemical substance. The sampling rate had the dimensions of a gaseous flow: if m is expressed in micrograms, the sampling period in minutes, and C in micrograms per liter, Q is expressed in liters per minute (Radiello 2013). A highFig. 3 Map of the different departments of the supermarket
temporal-resolution monitoring of total volatile organic compounds was also performed in the food department. The detection of total VOC concentrations was carried out with a PhoCheck® Tiger (Ion Science Ltd, UK) that used photoionization technology to detect a large range of VOC. This was factory calibrated against isobutylene, and thus, the concentration of total VOCs was equivalent to this gas. During the sampling period, temperature was measured at both the indoor and outdoor sites. Indoor air temperatures and relative
Environ Sci Pollut Res Table 1 List of compounds detected during the two 1-week monitoring campaigns
Detected compounds Benzene Cyclohexanone Chlorobenzene Methyl cyclohexane 3-Methylpentane
Cyclohexane 2-Isobutylene Naphthalene n-Butyl acetate 1-Octene
Styrene 1-Ethyl- 3,5- dimethyl-benzene Heptane Alpha-pinene Limonene
Ethyl acetate 1,4-Dichlorobenzene Isobutyl acetate 1,2,3-Trimethylbenzene Ethyl benzoate Ethyl-benzene 1-Methyletenylbenzene 1-Methylbenzoate 2-Methyl-butane Undecane
2-Propenylbenzene m-Xylene n-Decane 1,2,4-Trimethylbenzene Dodecane Cyclopentane Propyl-benzene Acetophenone 3-Methyl-hexane 2-Pentanone
Camphene Toluene 1,3,5-Trimethylbenzene 2-Ethyltoluene 2,4-Dimethyl-hexane 1,2-Dimethyl-4-ethyl-benzene Tetrachloroethylene 1,4-Dimethyl-benzene Hexanal
humidity were recorded and saved using portable thermohygrometric probes (LSI-LASTEM s.r.l., Milano, Italy). Outdoor temperatures were obtained from the nearest station of air-quality monitoring network in Bari.
Results and discussion Forty-four compounds were detected during the two monitoring campaigns conducted in the multistorey shopping mall (see Table 1). During the first campaign, benzene, m,p-xylene, styrene, n-ep tane, alpha -pinene , tolu ene, 1,2 ,3trimethylbenzene, tetrachloroethylene, limonene, 1,4-dichlorobenzene, n-decane, butylacetate, camphene, and ethylbenzene were the most abundant VOC. Meteorological data analysis indicated that air temperature and relative humidity ranged from 19.5 to 23.0 °C and from 34 to 64 % in the different departments, respectively. The minimum and maximum concentrations of VOC detected during each monitoring campaign are listed in Table 2. Toluene, limonene, camphene, and a-pinene were detected in higher concentrations during the first campaign. Average concentrations of TVOC (the mean of three consecutive monitoring periods) were also calculated according to ISO 16000-6 (2011) which means that the sum of VOC eluting between and including n-hexane and n-hexadecane was quantified by converting the total area of the chromatogram in that analytical window to toluene equivalents. This procedure was carried out for each monitored area in order to draw chromatic isoconcentration maps. MATLAB software provided visualization of the data (see Fig. 4). An analysis of the chromatic maps enabled us to identify the most critical areas of the supermarket and two storehouses. In particular, it was found that TVOC concentrations detected
in the storehouses were higher than those monitored in other departments; the average concentrations ranged from 330 to 550 μg/m3. The same finding was found for single compounds. In particular, toluene, a-pinene, camphene, and limonene were detected in two storehouses in concentrations three, five, seven, and five times higher than those obtained in the supermarket, respectively. Moreover, the chromatic map shows that the two storehouses (storehouses 1 and 2) yielded different levels of TVOC (different colors); these differences can be linked to the type of goods stored. In fact, storehouse 1 contained mostly grocery goods, whereas storehouse 2 held Table 2 Range of concentrations (in micrograms per cubic meter) for the most abundant compounds quantified during the first and second monitoring campaigns Compounds
First campaign Min–max (μg/m3)
Second campaign Min–max (μg/m3)
Benzene Heptane Toluene Tetrachloroethylene n-Butyl acetate Ethyl-benzene
0.10–5.28 2.28–10.93 0.06–44.26 0.06–0.91 0.19–5.62 0.18–8.30
0.60–9.14 0.99–7.51 0.27–54.69 0.27–5.30 0.73–6.34 0.24–10.40
m,p-Xylene Styrene Alpha-pinene Camphene Decane 1,4-Dichlorobenzene 1,2,3-Trimethylbenzene Limonene
0.62–13.60 0.12–8.01 0.72–83.31 1.22–121.11 0.05–21.37 0.08–0.47 0.06–0.77 0.20–43.05
0.44–35.20 0.14–7.08 0.06–161.76 0.12–116.72 0.14–25.72 0.10–4.14 0.02–5.45 0.03–65.74
Environ Sci Pollut Res Fig. 4 Chromatic map of isoconcentration obtained considering average TVOC concentrations detected during the first monitoring campaign in the supermarket and two storehouses
detergents and clothes. The higher TVOC concentrations detected in the storehouses may be due to the lower efficiency of air exchange and intake systems, as well as the large quantity of articles stored in these small spaces. In order to highlight the most abundant compounds at the most critical sites in the storehouses, Fig. 5 shows the contribution of each pollutant to the total concentration of quantified VOC. Concentrations of limonene, camphene, a-pinene, and toluene were higher than the other VOC. In particular, in storehouse 1 (sites A4 and B5), there were higher concentrations of camphene and limonene, whereas in storehouse 2, there were higher concentrations of a-pinene and toluene. Limonene, camphene, and a-pinene are odorous compounds mainly used to give a pleasant fragrance and flavoring to food,
cosmetics, personal care products, and household cleaning products (Carslaw 2013; Vu et al. 2013). Limonene is a common component found in consumer goods ranging from beverages to cleaning compounds. Camphene and α-pinene are emitted from liquid cleaners/disinfectants, cleaning products, and air fresheners (Nazaroff et al. 2004; Carslaw 2013). Toluene is used in making paints, paint thinners, fingernail polish, lacquers, adhesives, and rubber, as well as in printing operations, leather tanning, and polymerization as a replacement for benzene (Iregren 2000; Bruno et al. 2009). The critical area inside the supermarket is indicated in Fig. 6. Here, there is a map of the average TVOC concentrations, without taking into consideration those obtained in the storehouses. The food department was the most critical area
180
Fig. 5 Contributions of each detected compound to TVOC concentration in all sites of two storehouses
Limonene
1,2,3-trimethylbenzene
160
1,4-dichlorobenzene
140
n-Decane Camphene
120
(µg/m3)
Alfa-pinene
100
Styrene m-Xylene
80
Ethylbenzene
60
Bytyl acetate Tetrachloroetylene
40
Toluene
20
n-Eptane Cyclohexane
0
B12
B15
B5
A4
Benzene
Environ Sci Pollut Res Fig. 6 Chromatic map of isoconcentration of the supermarket obtained considering average TVOC concentrations detected during the first campaign in the supermarket
with TVOC concentrations ranging from 200 to 250 μg/m3. Among quantified compounds, camphene, limone, and apinene were the most abundant. Cleaning activities occur more frequently in the food department than in other areas of the supermarket and can be considered the main source of indoor VOC in this department (Kim et al. 2001; Lee et al. 2002; Loh et al. 2006). VOC concentrations in the other departments were quite homogeneous; this is probably due to the fact that the ventilation systems allowed an efficient
dispersion of VOC inside the supermarket. Therefore, it can be assumed that occasional events such as the staging of various departments and the packaging of consumer goods were the main emission sources of VOC inside the supermarket (Tang et al. 2005). This first result suggested that the mapping of TVOC concentrations in the monitored areas enabled the ranking of investigated areas. Moreover, the chromatic maps provided useful information to support supermarket management in their efforts of mitigating IAQ. The
Table 3 Range of outdoor concentrations (in micrograms per cubic meter) and of indoor/outdoor ratios (I/O) calculated for each monitored area (supermarket, storehouses). The last column lists the average values of I/O obtained during the three sampling periods in the parking lot area Compounds
Outdoor
Supermarket
Storehouses
Parking lot area
Min–max (μg/m3)
Min–max (I/O)
Min–max (I/O)
Mean I/O
Benzene Heptane Toluene Tetrachloroethylene n-Butylacetate
1.09 0.36 1.57 0.15 0.07
2.76 1.91 6.15 1.10 2.91
0.4 1.1 1.1 1.2 1.2
2 6 11 4 19
0.6 2 2 2 10
0.9 5 7 3 19
3 3 2 1.5 3
Ethyl-benzene m,p-Xylene Styrene Alpha-pinene Camphene Decane 1,4-Dichlorobenzene 1,2,3-Trimethylbenzene Limonene
0.50 1.72 0.14
CHEMISTRY IN A SUSTAINABLE SOCIETY
Indoor air quality (IAQ) assessment in a multistorey shopping mall by high-spatial-resolution monitoring of volatile organic compounds (VOC) M. Amodio & P. R. Dambruoso & Gianluigi de Gennaro & L. de Gennaro & A. Demarinis Loiotile & A. Marzocca & F. Stasi & L. Trizio & M. Tutino
Received: 10 July 2013 / Accepted: 9 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract In order to assess indoor air quality (IAQ), two 1week monitoring campaigns of volatile organic compounds (VOC) were performed in different areas of a multistorey shopping mall. High-spatial-resolution monitoring was conducted at 32 indoor sites located in two storehouses and in different departments of a supermarket. At the same time, VOC concentrations were monitored in the mall and parking lot area as well as outdoors. VOC were sampled at 48-h periods using diffusive samplers suitable for thermal desorption. The samples were then analyzed with gas chromatography–mass spectrometry (GC–MS). The data analysis and chromatic maps indicated that the two storehouses had the highest VOC concentrations consisting principally of terpenes. These higher TVOC concentrations could be a result of the low efficiency of the air exchange and intake systems, as well as the large quantity of articles stored in these small spaces. Instead, inside the supermarket, the food department was the most critical area for VOC concentrations. To identify potential emission sources in this department, a continuous VOC analyzer was used. The findings indicated that the highest total VOC concentrations were present during cleaning activities and that these activities were carried out frequently in the food department. The study highlights the importance of conducting both high-spatial-resolution monitoring and high-temporal-resolution monitoring. The former was able to identify critical issues in environments with a
Responsible editor: Constantini Samara M. Amodio : P. R. Dambruoso : G. de Gennaro (*) : L. de Gennaro : A. D. Loiotile : A. Marzocca : F. Stasi : L. Trizio : M. Tutino Department of Chemistry, University of Bari, Bari, Apulia Region 70126, Italy e-mail: [email protected]
complex emission scenario while the latter was useful in interpreting the dynamics of each emission source. Keywords Indoor air quality . Volatile organic compounds . Shopping mall . Articles . Food department . Storehouses
Introduction In recent decades, shopping malls have become such an integral part of daily urban life that they may be defined as multifunctional public urban spaces. People like to spend their leisure time in shopping malls, and even tourists will visit them on vacations. People use shopping malls for a variety of different activities (shopping, strolling, meeting people, etc.) in different hours of the day and in different days of the week. The climatically controlled, clean, and comfortable interior environment of shopping malls is often preferred over outdoor areas. Even if shopping malls are profit-oriented private properties, people can spend a whole day in them without doing any actual shopping. Therefore, these environments can be considered a new type of public space. Commercial buildings, such as offices, shopping malls, and mixed residential–commercial buildings, account for 20 % of all construction output in the EU and represent over 20 million m2 of floor space each year. In Italy, there are about 776 shopping malls covering an area (the area of sales and service) equal to 6.8 million m2 (Biasi 2012). Most contemporary shopping malls are microcities with street-like passageways full of shops, intersections, plazas, fountains, signs, and multilevel atria. A supermarket is usually at the heart of these shopping centers, and it is the largest and most populated store in the mall. This is the place offering the largest variety of articles and consumer goods and where different activities,
Environ Sci Pollut Res
such as food preparation and cooking, are carried out. In order to meet the growing demands of visitors, these newly designed and constructed malls have increased the number of services and recreational areas such as beauty salons, laundries, restaurants, playrooms, and coffeehouses. An increase in the number of visitors and the amount of time they spend in these environments has led to great interest in the indoor air quality (IAQ) of these environments. In recent years, numerous scientific studies have highlighted the fact that people who spend most of their time indoors are more exposed to pollution than those passing greater amounts of time outdoors (Bruno et al. 2008; Pegas et al. 2011). Yet, a wide spectrum of symptoms and illnesses can be related to nonindustrial indoor air pollution; hence, the quality of indoor air may have a considerable impact on public health (Jones 1999; Kotzias 2005; Wolkoff 2013). The phenomena of indoor air pollution can be influenced by many variables. For example, while pollutant concentrations may vary widely from one indoor environment to another, those in a specific environment may also vary as a function of location and time (Batterman et al. 2007). The extent of these variations depends on factors such as the emission characteristics of a source, an occupant’s behavior, and the microclimatic and ventilation conditions (UNI EN ISO 16000-1 2006). The detection of volatile organic compounds (VOC) is essential for assessing parameters for indoor air quality because of their ubiquitous presence in the atmosphere and their impact on human health. Recent epidemiological studies have demonstrated that VOC can be associated with disorders of many of the body’s organs (Chang et al. 2013). Short-term health effects include eye and respiratory tract irritation, headaches, dizziness, visual disorders, and memory impairment. Long-term health effects include damage to the liver, kidneys, and central nervous system. Some VOC are known to cause cancer in humans (Rumchev et al. 2007; Kampa and Castanas 2008; Zhou et al. 2011; Ramírez et al. 2012). Moreover, VOC have an important role in the “sick building syndrome” described by the World Health Organization (Brinke et al. 1998). Numerous studies conducted in nonresidential indoor environments have shown that indoor VOC concentrations often exceed those found outdoors (Bruno et al. 2008; Eklund et al. 2008). A preliminary assessment of the air quality of indoor environments such as shopping malls, which are sometimes quite large, involves identifying all areas of concern. The detection of potential sources can be facilitated with high-spatial-resolution monitoring campaigns using silent, low-cost samplers that are lightweight and without encumbrance. Diffusive samplers meet these requirements (UNI EN ISO 16000-1 2006; UNI EN ISO 16000-5 2007; UNI EN ISO 16017-2 2007; UNI EN 14412 2005). The use of thermal desorbable diffusive samplers further reduces the cost and time of analysis and allows automation by the use of an autosampler (Bruno et al. 2005).
Once the critical areas have been highlighted, high-temporalresolution monitoring can be conducted to obtain more in depth data and interpret the dynamics of the emission sources. The present work reports the results of two 1-week high-spatialresolution monitoring campaigns carried out to detect VOCs in a multistorey shopping mall. In order to assess the IAQ, indoor sites located in two storehouses and a supermarket were monitored using Radiello® diffusive samplers for thermal desorption and gas chromatography–mass spectrometry (GC–MS) analysis. At the same time, VOC concentrations were monitored in the mall and parking lot area as well as outdoors. Hence, it was possible to produce chromatic maps of TVOC (calculated according to ISO 16000-6:2011) concentrations to determine the emissions produced by articles and common goods present in the different areas and to identify the most critical zones. Finally, a high-temporal-resolution monitoring of VOC was conducted in the most critical area of the supermarket, and the results of this sampling enabled the identification of the potential emission sources.
Materials and method Study area and sampling sites The shopping mall investigated in this study is located in the suburbs of Bari, Italy. The mall has four floors altogether, and the area of each floor exceeds 20,000 m2. Three of these floors are underground. The second and third underground floors are used as car parks while the first underground floor is occupied by a supermarket, a department store, and other small shops. The monitoring campaigns were conducted on this floor, predominantly, inside the supermarket which covers an area of 10,000 m2 and in the two storehouses nearby whose total area is 5,500 m2. The mall is centrally air-conditioned, and smoking is not allowed. It opens at 9:00 am and closes at 9:00 pm every day. During each experimental period (which included Sunday), there were about 100,000 customers visiting the mall daily and about 3,000 clients visiting the supermarket daily. Two 1-week VOC monitoring campaigns were performed in the multistorey shopping mall. The first was conducted from 23 to 29 November 2010 and the second one from 8 to 16 September 2011. Each campaign was subdivided into three different periods with a sampling duration of 48 h. A monitoring grid was drawn on the map of the mall (supermarket and two storehouses) in order to obtain meshes of where to locate diffusive samplers. The most homogenous areas were selected by considering microenvironments with similar merchandise and lane width. Monitoring was planned taking into account the guidelines of the UNI EN ISO 16000-1 (2006) which describe the procedure (positioning, representativeness of the site, etc.) for carrying out a preliminary evaluation of the spatial distribution of volatile
Environ Sci Pollut Res
pollutants in indoor air. Moreover, the density of people was considered in the selection and location of the sites. These observations were made in order to obtain data as representative as possible of the mean concentration levels of pollutants in each area. It was not easy to define or to check the changes in air quality because of the large number of people that frequented these environments every day. For this reason, sampler placement took into account the distance from doors communicating with storehouses and potential heat sources. Figure 1 shows the 32 sampling sites (red square) in the different indoor microenvironments inside the supermarket. During the first monitoring campaign, a total of 96 air samples were collected. A second monitoring campaign was conducted in order to confirm findings obtained during the first campaign as well as to gain more insight into the critical areas highlighted. In particular, other 10 indoor sampling sites (blue squares in Fig. 2) and parallel outdoor sites were monitored. A total of 149 samples were collected during the second monitoring campaign. Figure 3 shows the distribution of the different departments in the monitoring areas. Sampling and analytical method VOCs were sampled with Radiello® diffusive samplers (Fondazione Salvatore Maugeri, Padova, Italy) suitable for Fig. 1 Map of the sampling sites during the first monitoring campaign
thermal desorption. The sampling system was made up of a cylindrical adsorbing cartridge housed coaxially inside a cylindrical diffusive body of microporous polyethylene 5 mm thick and average porosity 10±2 μm. The cartridges consisted of a stainless steel net cylinder, with 3×8 μm mesh opening and 4.8 mm diameter, packed with 350±10 mg of graphitized charcoal (Carbograph 4) with particle size of 35–50 mesh. Before sampling, the cartridges were conditioned and analyzed to verify the blank levels (Bruno et al. 2005, 2008). Each sampler was exposed for 2 days. After sampling, it was placed in a sealed glass tube and brought to the laboratory for analysis. The analyses were carried out using a thermal desorber (Markes International Ltd, Unity™) equipped with an autosampler (Markes mod. ULTRA™ TD) with 100 positions and coupled with a gas chromatograph (Agilent GC6890 Plus) and a mass selective detector (Agilent MS5973N). The thermal desorber provided a two-stage mechanism: first, the analytes were desorbed from the sample tube and refocused into a cold trap; second, they were desorbed from the trap and carried into the GC column (Bruno et al. 2005; UNI EN ISO 16017-2 2007). To quantify the samples, the calibration curves were prepared by injecting 1 μL of standard solutions (Ultra Scientific Cus-5997) into a tube; the spiked adsorbent tubes were then thermally desorbed in the same conditions of time, gas flow, and split ratio as the
Environ Sci Pollut Res Fig. 2 Map of the sampling sites during the second monitoring campaign. The new sites are in blue
samples. The sampling rates, Q values supplied by the manufacturer, were used to calculate the real concentration of the compound in the atmosphere (C) by GC quantification of the analyte’s mass, m. Q was the function of the diffusive coefficient D, which was the thermodynamic property of each chemical substance. The sampling rate had the dimensions of a gaseous flow: if m is expressed in micrograms, the sampling period in minutes, and C in micrograms per liter, Q is expressed in liters per minute (Radiello 2013). A highFig. 3 Map of the different departments of the supermarket
temporal-resolution monitoring of total volatile organic compounds was also performed in the food department. The detection of total VOC concentrations was carried out with a PhoCheck® Tiger (Ion Science Ltd, UK) that used photoionization technology to detect a large range of VOC. This was factory calibrated against isobutylene, and thus, the concentration of total VOCs was equivalent to this gas. During the sampling period, temperature was measured at both the indoor and outdoor sites. Indoor air temperatures and relative
Environ Sci Pollut Res Table 1 List of compounds detected during the two 1-week monitoring campaigns
Detected compounds Benzene Cyclohexanone Chlorobenzene Methyl cyclohexane 3-Methylpentane
Cyclohexane 2-Isobutylene Naphthalene n-Butyl acetate 1-Octene
Styrene 1-Ethyl- 3,5- dimethyl-benzene Heptane Alpha-pinene Limonene
Ethyl acetate 1,4-Dichlorobenzene Isobutyl acetate 1,2,3-Trimethylbenzene Ethyl benzoate Ethyl-benzene 1-Methyletenylbenzene 1-Methylbenzoate 2-Methyl-butane Undecane
2-Propenylbenzene m-Xylene n-Decane 1,2,4-Trimethylbenzene Dodecane Cyclopentane Propyl-benzene Acetophenone 3-Methyl-hexane 2-Pentanone
Camphene Toluene 1,3,5-Trimethylbenzene 2-Ethyltoluene 2,4-Dimethyl-hexane 1,2-Dimethyl-4-ethyl-benzene Tetrachloroethylene 1,4-Dimethyl-benzene Hexanal
humidity were recorded and saved using portable thermohygrometric probes (LSI-LASTEM s.r.l., Milano, Italy). Outdoor temperatures were obtained from the nearest station of air-quality monitoring network in Bari.
Results and discussion Forty-four compounds were detected during the two monitoring campaigns conducted in the multistorey shopping mall (see Table 1). During the first campaign, benzene, m,p-xylene, styrene, n-ep tane, alpha -pinene , tolu ene, 1,2 ,3trimethylbenzene, tetrachloroethylene, limonene, 1,4-dichlorobenzene, n-decane, butylacetate, camphene, and ethylbenzene were the most abundant VOC. Meteorological data analysis indicated that air temperature and relative humidity ranged from 19.5 to 23.0 °C and from 34 to 64 % in the different departments, respectively. The minimum and maximum concentrations of VOC detected during each monitoring campaign are listed in Table 2. Toluene, limonene, camphene, and a-pinene were detected in higher concentrations during the first campaign. Average concentrations of TVOC (the mean of three consecutive monitoring periods) were also calculated according to ISO 16000-6 (2011) which means that the sum of VOC eluting between and including n-hexane and n-hexadecane was quantified by converting the total area of the chromatogram in that analytical window to toluene equivalents. This procedure was carried out for each monitored area in order to draw chromatic isoconcentration maps. MATLAB software provided visualization of the data (see Fig. 4). An analysis of the chromatic maps enabled us to identify the most critical areas of the supermarket and two storehouses. In particular, it was found that TVOC concentrations detected
in the storehouses were higher than those monitored in other departments; the average concentrations ranged from 330 to 550 μg/m3. The same finding was found for single compounds. In particular, toluene, a-pinene, camphene, and limonene were detected in two storehouses in concentrations three, five, seven, and five times higher than those obtained in the supermarket, respectively. Moreover, the chromatic map shows that the two storehouses (storehouses 1 and 2) yielded different levels of TVOC (different colors); these differences can be linked to the type of goods stored. In fact, storehouse 1 contained mostly grocery goods, whereas storehouse 2 held Table 2 Range of concentrations (in micrograms per cubic meter) for the most abundant compounds quantified during the first and second monitoring campaigns Compounds
First campaign Min–max (μg/m3)
Second campaign Min–max (μg/m3)
Benzene Heptane Toluene Tetrachloroethylene n-Butyl acetate Ethyl-benzene
0.10–5.28 2.28–10.93 0.06–44.26 0.06–0.91 0.19–5.62 0.18–8.30
0.60–9.14 0.99–7.51 0.27–54.69 0.27–5.30 0.73–6.34 0.24–10.40
m,p-Xylene Styrene Alpha-pinene Camphene Decane 1,4-Dichlorobenzene 1,2,3-Trimethylbenzene Limonene
0.62–13.60 0.12–8.01 0.72–83.31 1.22–121.11 0.05–21.37 0.08–0.47 0.06–0.77 0.20–43.05
0.44–35.20 0.14–7.08 0.06–161.76 0.12–116.72 0.14–25.72 0.10–4.14 0.02–5.45 0.03–65.74
Environ Sci Pollut Res Fig. 4 Chromatic map of isoconcentration obtained considering average TVOC concentrations detected during the first monitoring campaign in the supermarket and two storehouses
detergents and clothes. The higher TVOC concentrations detected in the storehouses may be due to the lower efficiency of air exchange and intake systems, as well as the large quantity of articles stored in these small spaces. In order to highlight the most abundant compounds at the most critical sites in the storehouses, Fig. 5 shows the contribution of each pollutant to the total concentration of quantified VOC. Concentrations of limonene, camphene, a-pinene, and toluene were higher than the other VOC. In particular, in storehouse 1 (sites A4 and B5), there were higher concentrations of camphene and limonene, whereas in storehouse 2, there were higher concentrations of a-pinene and toluene. Limonene, camphene, and a-pinene are odorous compounds mainly used to give a pleasant fragrance and flavoring to food,
cosmetics, personal care products, and household cleaning products (Carslaw 2013; Vu et al. 2013). Limonene is a common component found in consumer goods ranging from beverages to cleaning compounds. Camphene and α-pinene are emitted from liquid cleaners/disinfectants, cleaning products, and air fresheners (Nazaroff et al. 2004; Carslaw 2013). Toluene is used in making paints, paint thinners, fingernail polish, lacquers, adhesives, and rubber, as well as in printing operations, leather tanning, and polymerization as a replacement for benzene (Iregren 2000; Bruno et al. 2009). The critical area inside the supermarket is indicated in Fig. 6. Here, there is a map of the average TVOC concentrations, without taking into consideration those obtained in the storehouses. The food department was the most critical area
180
Fig. 5 Contributions of each detected compound to TVOC concentration in all sites of two storehouses
Limonene
1,2,3-trimethylbenzene
160
1,4-dichlorobenzene
140
n-Decane Camphene
120
(µg/m3)
Alfa-pinene
100
Styrene m-Xylene
80
Ethylbenzene
60
Bytyl acetate Tetrachloroetylene
40
Toluene
20
n-Eptane Cyclohexane
0
B12
B15
B5
A4
Benzene
Environ Sci Pollut Res Fig. 6 Chromatic map of isoconcentration of the supermarket obtained considering average TVOC concentrations detected during the first campaign in the supermarket
with TVOC concentrations ranging from 200 to 250 μg/m3. Among quantified compounds, camphene, limone, and apinene were the most abundant. Cleaning activities occur more frequently in the food department than in other areas of the supermarket and can be considered the main source of indoor VOC in this department (Kim et al. 2001; Lee et al. 2002; Loh et al. 2006). VOC concentrations in the other departments were quite homogeneous; this is probably due to the fact that the ventilation systems allowed an efficient
dispersion of VOC inside the supermarket. Therefore, it can be assumed that occasional events such as the staging of various departments and the packaging of consumer goods were the main emission sources of VOC inside the supermarket (Tang et al. 2005). This first result suggested that the mapping of TVOC concentrations in the monitored areas enabled the ranking of investigated areas. Moreover, the chromatic maps provided useful information to support supermarket management in their efforts of mitigating IAQ. The
Table 3 Range of outdoor concentrations (in micrograms per cubic meter) and of indoor/outdoor ratios (I/O) calculated for each monitored area (supermarket, storehouses). The last column lists the average values of I/O obtained during the three sampling periods in the parking lot area Compounds
Outdoor
Supermarket
Storehouses
Parking lot area
Min–max (μg/m3)
Min–max (I/O)
Min–max (I/O)
Mean I/O
Benzene Heptane Toluene Tetrachloroethylene n-Butylacetate
1.09 0.36 1.57 0.15 0.07
2.76 1.91 6.15 1.10 2.91
0.4 1.1 1.1 1.2 1.2
2 6 11 4 19
0.6 2 2 2 10
0.9 5 7 3 19
3 3 2 1.5 3
Ethyl-benzene m,p-Xylene Styrene Alpha-pinene Camphene Decane 1,4-Dichlorobenzene 1,2,3-Trimethylbenzene Limonene
0.50 1.72 0.14
