**Cooking**/**Window Opening and Associated Increases of Indoor PM2.5 and NO2 Concentrations of Children's Houses in Kaohsiung, Taiwan**

#### **Yu-Chuan Yen 1, Chun-Yuh Yang 1, Kristina Dawn Mena 2, Yu-Ting Cheng <sup>1</sup> and Pei-Shih Chen 1,3,4,5,\***


Received: 29 August 2019; Accepted: 2 October 2019; Published: 14 October 2019

**Abstract:** High concentrations of air pollutants and increased morbidity and mortality rates are found in industrial areas, especially for the susceptible group, children; however, most studies use atmospheric dispersion modeling to estimate household air pollutants. Therefore, the aim of this study was to assess the indoor air quality, e.g., CO, CO2, NO2, SO2, O3, particulate matter with aerodynamic diameter less than 2.5 μm (PM2.5), and their influence factors in children's homes in an industrial city. Children in the "general school", "traffic school", and "industrial school" were randomly and proportionally selected. Air pollutants were sampled for 24 h in the living rooms and on the balcony of their houses and questionnaires of time–microenvironment–activity-diary were recorded. The indoor CO concentration of the traffic area was significantly higher than that of the industrial area and the general area. In regard to the effects of window opening, household NO2 and PM2.5 concentrations during window opening periods were significantly higher than of the reference periods. For the influence of cooking, indoor CO2, NO2, and PM2.5 levels during the cooking periods were significantly higher than that of the reference periods. The indoor air quality of children in industrial cities were affected by residential areas and household activities.

**Keywords:** indoor air quality; children's house; industrial city; window opening; cooking

#### **1. Introduction**

According to the Environmental White Paper of Taiwan Environmental Protection Agency (Taiwan EPA), the annual average concentrations of ambient CO, NO2, SO2, and O3 in 2008 were 0.47 ppm, 16.90 ppb, 4.35 ppb, and 29.09 ppb, respectively. The Kaohsiung–Pingtung area was the worse polluted area in Taiwan and accounted for 5.93% of station-days of the Pollutant Standards Index (PSI) > 100. Especially, Kaohsiung is a heavy industrial city. In industrial areas, high concentrations of air pollutants and increased morbidity and mortality rates are found, depending on the types of industrial activities and exposure concentrations in residential areas [1,2]. Children are more susceptible to the health effects of air pollution than adults due to not having full development of their pulmonary metabolic capacity [3]. Long-term exposure of air pollution may affect children's lung development [4]. Previously, most of the studies revealed that ambient pollution such as particulate matter with aerodynamic diameter less than 10 and 2.5 μm (PM10 and PM2.5), sulfur dioxide (SO2),

nitrogen dioxide (NO2), volatile organic compounds (VOCs), etc. in industrial areas may increase the risk of respiratory symptoms, and attacks of asthma in children [1,5,6]. Therefore, indoor air quality of children's homes may be very important to children's health, especially in industrial cities, since children spend most of their time at home [7].

Indoor air quality may be affected by indoor human activities such as cooking, smoking, cleaning, etc. and the infiltration of outdoor pollutants produced from the traffic or industrial sources [8–11]. For example, SO2, NOx, PM2.5, and carbon monoxide (CO), the major conventional air pollutant in steel plants, oil refineries, and vehicular exhaust emissions [12–14], may enter a house through cracks and windows [15,16]. In addition, if the indoor air is not well ventilated, the air pollutants may accumulate in the indoor environment, and then seriously affects the health of the inhabitants [17].

Previously, atmospheric dispersion modeling was used to estimate the household concentrations of indoor air pollutants in industrial areas [1,18,19]. Only a few studies actually measured individual exposure [20] and household concentrations [21–23], and these studies only focused on PM mass concentrations, elemental composition, and VOCs concentrations. However, other air pollutants e.g., CO, carbon dioxide (CO2), NO2, SO2, and ozone (O3) in households in industrial cities also need to be considered. Therefore, the main aim of this study was to assess the indoor air quality including CO, CO2, NO2, SO2, O3, and PM2.5, temperature and relative humidity, and their influence factors (e.g., window opening and cooking) in children's homes in an industrial city—Kaohsiung City. To our knowledge, this is the first study to assess the indoor air quality including CO, CO2, NO2, SO2, and O3 in children's homes in an industrial city. In addition, the second aim was to evaluate potential determinants of indoor air pollutants levels of occupants' activities, including cooking and window opening, etc. It is also the first study to reveal the differences of air pollutants between cooking periods/window opening periods and reference periods through a time–microenvironment–activity-diary via a questionnaire in one-hour time segments.

#### **2. Materials and Methods**

#### *2.1. Study Area*

Kaohsiung City (22◦38 N, 120◦17 E), located in southern Taiwan and with the population density of 9962.6/km<sup>2</sup> in 2010, is the largest industrialized harbor city in Taiwan with intense traffic and heavy industries including the largest steel plant (the China Steel Corporation, which also ranked the 19th steel mill in the world in 2005), the largest oil refinery (the CPC Corporation), the largest international shipbuilding (it ranked 6th in the world in 2005) in Taiwan, and many petrochemical industries.

#### *2.2. Study Design*

In April 2010, we selected three elementary schools in Kaohsiung City. One elementary school had a general air quality monitoring station of Taiwan EPA on the roof of the 4th floor, so we called this school a "general school". Another elementary school was 0.33 km from Taiwan EPA's traffic air quality monitoring station and was regarded as a "traffic school". The "industrial school" was an elementary school located near the Xiaogang Industrial Zone in Kaohsiung City and about 0.30 km from Taiwan EPA's air monitoring station. The study population was limited to children who attended these schools. The number of students in the "general school", "traffic school", and "industrial school" were 1669, 987, and 960, respectively. After obtaining the assented of the child and the permission of the parents, we recorded the subjects who agreed to home visits for environmental sampling. Children were randomly and proportionally selected from each school to participate in this study. Finally, the home visits of 32, 16, and 12 participants in the "general school", "traffic school", and "industrial school", respectively, were completed between April 2010 and October 2010.

#### *2.3. Air Sampling*

Indoor air pollutants including CO, CO2, NO2, SO2, O3, PM2.5, temperature, and relative humidity were measured by real-time monitoring equipment for 24 h in the living rooms. We also measured the atmospheric CO, CO2, NO2, SO2, O3, and PM2.5 on the balcony as outdoor concentrations. All instruments were placed on the bench at a height of approximately 1 m above the ground. The PM was measured by a real-time optical scattering instrument (DUSTTRAK™ DRX Aerosol Monitor Models 8533, TSI Incorporated, Shoreview, MN, USA) and the measurements were taken every 1 s by the flow rate of 3.0 L/min with detectable concentration from 0.001 to 150 mg/m3. The CO, CO2, NO2, SO2, indoor temperature, and relative humidity were also recorded (KD-airboxx, KD Engineering, Blaine, WA, USA) every 15 s with the measuring range of 0 to 500 ppm, 0 to 10,000 ppm, 0 to 20 ppm, 0 to 20 ppm, 0 to 50 ◦C, and 5% to 95%, respectively. The accuracy of CO, CO2, NO2, and SO2 were ±3% of reading or 2 ppm (whichever was greater), ±5% of reading or 60 ppm (whichever was greater), 0.25 ppm, and 0.25 ppm, respectively. The resolution of CO, CO2, NO2, and SO2 were 0.1 ppm, 1 ppm, 0.01 ppm, and 0.01 ppm, respectively. In terms of O3, it was detected by a real-time monitoring (Model 202 Ozone monitorTM, 2B Technologies Inc, Boulder, CO, USA) every 5 min with the measuring range of 0 to 250 ppm.

All real-time monitors were manufacturer-calibrated for the study in the beginning of this study and every six months. Before every field sampling, the DUSTTRAK™ DRX Aerosol Monitor Models 8533 was calibrated using emery oil aerosol and nominally adjusted to the respirable mass of standard ISO 12103-1, A1 test dust, (Arizona Dust); and the KD-airboxx, the Model 202 Ozone monitorTM were calibrated using zero gas and span gas. In addition, the zero calibrators of instruments were carried out, and the flow rate of sampling pump also was adjusted by Gilian Gilibrator-2NIOSH Primary Standard Air Flow Calibrator (Sensidyne, St. Petersburg, FL, USA) before every household sampling.

#### *2.4. Household Characteristics*

In addition, household characteristics including the number of occupants, air-conditioning use, smoking, incense burning, etc. were also recorded in the questionnaires. In addition to household characteristics, data on potential determinants of indoor air pollutants levels of occupants' activities, including cooking and window opening, etc. were obtained through a time–microenvironment–activity-diary via a questionnaire in one-hour time segments. We also actually evaluated the effects of window opening and cooking on indoor air pollutants. The window opening periods were defined from a time–microenvironment–activity-diary and two one-hour periods before and after window opening periods were defined as the reference periods. In terms of cooking, the cooking periods were the periods recorded by participants as cooking from a time–microenvironment–activity-diary and the reference periods were defined as the one-hour periods before the cooking periods.

#### *2.5. Ethics*

This study was approved by the Institutional Review Board of the Kaohsiung Medical University Chung-Ho Memorial Hospital (the protocol number was KMU-IRB-990045). Informed written consent was obtained from each child (the phonetic version of the consent form that the children read and signed) and their legal guardians.

#### *2.6. Statistical Analyses*

Statistical analyses in this study were performed using SAS version 9.3 (SAS Institute of Taiwan Ltd, Taipei, Taiwan). Descriptive statistics were used to describe the 24-hour of average of exposure data (indoor/outdoor air pollutant concentrations, temperature, and relative humidity). The concentrations of air pollutants were not normally distributed (data not shown), therefore we analyzed our data by nonparametric statistics, also known as distribution-free statistics. A paired Student's *t*-test was used to

assess the difference in the average concentration of air pollutants between indoor and outdoor, between window opening periods and reference periods, and between cooking periods and reference periods. With the objective of evaluating significant differences among the areas (general, traffic, and industry) for all air pollutants variables, data were analyzed using one-way analysis of variance (ANOVA) with Scheffe multiple comparison test. The generalized estimating equations (GEE) is a general statistical method in a longitudinal study with small samples for adjusting time interference, in which each time point is an independent event. Finally, the relationships between the 24-hour average concentrations of indoor air pollutants (dependent variable) and household characteristics (independent variable) were analyzed using GEE, adjusting for other household characteristics, and time interference. A *p*-value of less than 0.05 was considered significant.

#### **3. Results**

Table 1 shows the descriptive statistics of 24-h average indoor and outdoor air pollutants, temperature, and relative humidity in 60 houses. When indoor air pollutants were paired with outdoors within the same home, we found that the 24-hour average concentrations of indoor CO, CO2, and NO2 were significantly higher than the 24-hour average of outdoors concentrations, whereas, outdoor O3 and PM2.5 concentrations were significantly higher than indoor concentrations (all *p* < 0.01). The average distance between homes of subjects and their school were 0.86 km, 0.94 km, and 1.46 km in general, traffic, and industrial areas, respectively, as well as, the average distance between homes of subjects and the nearest air monitoring station were 1.07 km, 0.97 km, and 1.46 km in general, traffic, and industrial areas, respectively.


**Table 1.** Descriptive statistics of 24-h average indoor and outdoor air pollutants, temperature, and relative humidity in 60 houses.
