Household air pollution and personal exposure to air pollutants in rural China – A review

Environmental Pollution 237 (2018) 625e638

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Household air pollution and personal exposure to air pollutants in rural China e A review* Wei Du a, Xinyue Li a, Yuanchen Chen b, *, Guofeng Shen a, ** a

Laboratory of Earth Surface Processes, College of Urban and Environmental Science, Peking University, Beijing, 100871, China Key Laboratory of Microbial Technology for Industrial Pollution Control of Zhejiang Province, College of Environment, Research Centre of Environmental Science, Zhejiang University of Technology, Hangzhou, 310014, China

b

a r t i c l e i n f o

A b s t r a c t

Article history: Received 9 October 2017 Received in revised form 11 January 2018 Accepted 18 February 2018

Solid fuels, an important source of severe Household Air Pollution (HAP) linked to many adverse health outcomes, has been widely consumed around the world. China consumes large amounts of solid fuels and suffers from serious indoor and outdoor air pollution. Though global HAP issues had been reviewed in previous literatures, peer-reviewed Chinese publications were seldom included in those reviews. We conducted a literature review on the studies of HAP and personal exposure in rural China with inputs from peer-reviewed publications in both English and Chinese. A total of 36,572 articles were retrieved, 294 were read in full text, of which 92 were included in final data extraction and in-depth analysis. Although HAP is a very serious issue in China, studies on either HAP or personal exposure assessment were very limited. From existing studies, levels of air pollutants including carbon monoxide, sulfur dioxide, particulate matter (PM), organic carbon, elemental carbon, polycyclic aromatic hydrocarbons (PAHs), etc., in indoor and ambient air were analyzed for their temporal and spatial variations, and the differences across different fuel types were compared. The studies showed that PM and PAHs levels in most rural homes exceeded the World Health Organization (WHO) and Chinese National Standards, especially during the heating season in northern China. Replacing traditional fuels with cleaner ones (such as liquid petroleum gas (LPG), biogas or electricity) was considered as the most appropriate way to mitigate HAP. The daily exposure to PM and PAHs from using LPG, biogas or electricity was considerably lower than that from using traditional solid fuels. However, the level was still higher than the guideline values for PM and PAHs set by WHO to protect human health. To achieve a more effective control, the current data gap need to be closed and suggestions for future research were discussed in this review. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Solid fuel Household air pollution Inhalation exposure

1. Introduction There are still nearly 3 billion people worldwide relying on traditional solid fuels for cooking and heating. Most of the population are in developing countries, particularly in rural areas with relatively low incomes (Bonjour et al., 2013). Inefficient burnings of these solid fuels produced high emissions of various air pollutants like CO, SO2, particulate matter (PM), black carbon (BC), polycyclic aromatic hydrocarbons (PAHs), etc., which caused severe pollution in not only indoor but outdoor air (Clark et al., 2013; Chafe et al., 2014). This issue of residential combustion is widely recognized

*

as Household Air Pollution (HAP) nowadays, rather than Indoor Air Pollution (IAP). Inhalation exposure to severe HAP has been documented to be associated with various diseases and premature deaths (Clark et al., 2013; Smith, 1993; Zhang and Smith, 2007; Lim et al., 2012). Globally, around 2.8 million premature deaths were estimated to be due to exposure to HAP, and in China the number was about 1.0 million (Cohen et al., 2017). There have been several systematic reviews of literatures on HAP and the health impacts (Ezzati and Kammen, 2002a, 2002b; Kim et al., 2011; Mehta et al., 2013; Smith et al., 2000; Zhang and Smith, 2007), on barriers and enablers to promote clean fuels and

This paper has been recommended for acceptance by Eddy Y. Zeng. * Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (Y. Chen), [email protected] (G. Shen). https://doi.org/10.1016/j.envpol.2018.02.054 0269-7491/© 2018 Elsevier Ltd. All rights reserved.

626

W. Du et al. / Environmental Pollution 237 (2018) 625e638

stoves to alleviate HAP problem (Puzzolo et al., 2016; Lewis and Pattanayak, 2012; Rehfuess et al., 2014; Shen et al., 2015), and on changes in HAP and health benefits through interventions (Pope et al., 2017; Quansah et al., 2017). Recognizing that most of these reviews offered the perspective on a global scale, the specific research status in China was brought to our interest. As the largest developing country, China consumes large amounts of solid fuels such as coals and biomass fuels. Traditional solid fuels are the dominant household energies in most rural areas. The wide use of those solid fuels has contributed largely to indoor and ambient air pollution in the country. Although residential combustion was often overlooked in the air pollution control in China relative to other pollution sources such as power plant and vehicle emissions (Liu et al., 2016), there are growing concerns on HAP and its impacts in China, yielding more publications. The research questions we started out to form this review were: 1) what's the current status of research on HAP and inhalation exposure measurement in rural China; and 2) what were the range, temporal-spatial variations as well as the differences in groups burning solid fuels (e.g. coal, crop waste, and wood) and non-solid household energies (e.g. biogas, LPG, and electricity) for air pollutants like CO, SO2, PM, PAHs, etc. For the first question, we were interested to know: how many studies were available in the literature, and when/where these studies were conducted; and, was there an increasing trend in publications on this research topic as the problem gains growing attention among researchers, the public, and the policy makers. In the second question, it was realized that various factors, including fuel type, stove type, burning frequency, household characteristics and ventilation condition, ambient meteorological condition, and other pollution sources, can affect air pollution. In this study, we specifically focused on differences between the traditional solid fuels and non-solid ones, given that many studies and projects were interested in alleviating HAP to protect human health by deploying cleaner fuels into rural homes; and fuel was the most frequently mentioned factor in the literature, while discussions on other influencing factors were limited and only available in few studies. The relatively rich and complete information on fuel allowed us to compile and summarize the data and make comparisons with fairly low bias and uncertainties. Eligible studies were first identified through a literature search and then subject to screening and full-text evaluation. Quantitative data were extracted and synthesized from final included studies. In this review, we analyzed the research status of HAP and personal measurement in rural China (Section 3.1), characteristics of HAP (including pollution levels, indoor-outdoor difference, seasonal change, and fuel difference) for different pollutants (Section 3.2), and daily exposure to air pollutants (concentration, temporal and spatial changes, and difference in different fuel groups) (Section 3.3). A discussion on the implications and limitations of this review and suggestions on future studies were presented in the last section. In the preparation of this manuscript, we noticed a recently published review by Li et al. (2017) on residential solid fuel combustion in China and its impacts on indoor and ambient air quality. Different from Li et al. (2017) who had main foci on how influencing factors like fuels, stove types, and ventilation conditions, affect air quality by referring to evidences in past studies, the present review extended the literature search to have a compilation on available peer-reviewed publications, in either English or Chinese, on HAP and inhalation exposure studies in rural China. The present study was expected to provide readers a general picture on the research progress of this important topic, and characteristics of air pollution and exposure levels (e.g. pollution levels, seasonal change, and variations in homes using different household energies) in rural China.

2. Method 2.1. Literature search Standard web-based searches were conducted to access available studies in electronic databases including Web of Science, Science Direct, and Springer Link for English papers, and China National Knowledge Infrastructure, Wanfang Data, and VIP Journal Integration Platform for publications in Chinese. Although a wellstructured PICO (Population, Intervention, Comparison, and Outcome) framework may increase precision and satisfaction of searched results, the use of a web-based search interface was acceptable (Booth et al., 2000; Schardt et al., 2007; Cheng, 2004). The search terms were (“Indoor air pollution” or “household air pollution” or “pollution” or “inhalation exposure” or “portable carried samplers”) AND (“solid fuels” or “biomass” or “wood” or “crop residues” or “coal” or “clean fuels” or “clean energy” or “household energy” or “electricity” or “biogas” or “LPG” or “stove” or “cookstove” or “gasifier stoves” or “traditional stoves” or “improved stoves”) AND (“China” or “rural China” or “village” or “rural residents” or “rural population”) in the TOPIC. The search was done by two reviewers (WD and XL). 2.2. Study selection and data extraction Only HAP measurements in the rural areas and inhalation exposure for the rural population were included in the present study. The outcome included quantitative measures of daily average concentrations of indoor and outdoor air pollutants in the rural areas, and/or measured or estimated personal exposure to air pollutants. As the review was primarily to understand the status of research and characteristics (e.g. contamination levels, temporalspatial distribution, and fuel impacts) of HAP and personal exposure studies, there was no restriction on intervention and comparison methods in the literature search. There was also no restriction on the measurement methods, but studies finally included in data extraction must have clear methodology on site information, sampling, quality assurance and controls. The search yielded a total of 36,572 papers after exclusion of duplicates from the electronic databases. A reviewer (WD) first checked the relevance of these papers by screening titles and keywords. Abstracts of papers that were potentially relevant were then read by three reviewers independently (WD, GS, and YC). Only peer-reviewed publications that were highly relevant to the review scope were downloaded for a full-text review. Any discordant classifications were discussed. To meet the inclusion criteria, the full text of papers must be peer-reviewed, report HAP and/or personal inhalation in rural China, have clear research aims, study design, methodologies and quality controls. This was assessed by three reviewers independently. A flowchart showing search results of identification, screening, eligibility, and inclusion is illustrated in Fig. S1. 92 peer-reviewed papers were finally included and extracted for data analysis in this review. Data extraction was carried out on all papers that passed selection using the same data extraction form. Information extracted included title, journal, year published, author(s), study year, province/county, site (kitchen, bedroom, or outdoor yard), season, fuel, stove, measurement methodology, number of sample size, pollutant type and statistical results (range, arithmetic mean, median, geometric means, standard deviation, geometric standard deviation) of the HAP and/or inhalation exposure measurements (where applicable). Risk of bias could be from selection bias, study design, con-founders, blinding, data collection, and withdrawals and dropouts. Given the specific objective of the present review, the quality of eligible studies was evaluated for the selection bias

W. Du et al. / Environmental Pollution 237 (2018) 625e638

(study area and representative measurement group), measurement and data analysis method, and withdrawals and dropouts (e.g. valid samples between 22 and 26 h). The quality assessment was performed in a group discussion with all four reviewers. The final 92 papers were representative of the study area and population in the area, adopted typical measurement technologies to quantify HAP and daily exposure, and had quality assurance and controls on measurement accuracy, sample validation, and statistical analysis. Table S1 summarized basic information and main findings from these studies. 2.3. Analysis of eligible studies and data synthesis Studies were classified into two groups: 1) HAP study that measured the indoor and/or outdoor air pollution by stationary samplers; and 2) inhalation exposure study that measured daily exposure using portable samplers and/or estimated daily exposure level by calculating the time-weighted average exposure based on human activity pattern and stationary sampler measurement results. Most studies reported arithmetic means of the measured pollutants, and some also showed other statistical results like geometric means, median values, interquartile range, etc. In data synthesis, arithmetic means from all the literature were compiled to recalculate the range and “overall” means of air pollutants in different seasons, fuel groups, and/or exposure populations. 3. Results 3.1. Status of HAP and personal exposure measurements in rural China There was a general increasing trend in the number of published HAP studies in rural China, particularly after the year 2000 (Fig. 1). Of the 92 peer-reviewed publications, the number of papers published before the year 2000 (1980s-1990s), during the 2000s (20012010) and during 2010-2016 was 8, 38 and 46 (8.7%, 41.3% and 50.0% of the total publication number), respectively. Most studies published their measurement results 2-3 years after the field campaign with a few results published >5 years after the field campaign. The lag included a period for laboratory measurements, data analysis, manuscript prepare, as well as the manuscript submission and peer-review processes before publication. Most of the previous studies used stationary samplers, while some more recent studies have adopted portable samplers in evaluation of daily exposure levels. About 40% of the studies only

627

investigated indoor air pollution, another 30% simultaneously measured indoor and outdoor air pollution, while the others quantified not only HAP but also personal exposure/inhalation exposure. For personal exposure measurement, not only the number of publications was very limited (25 papers) but also the numbers of subjects (sample sizes) were relatively small in those existing studies. Most of the field measurements in the studies were conducted in one single province, with only a few (<10%) done in multiple provinces. As seen in Fig. 2, there were more than 10 publications each for provinces like Yunnan, Guizhou, and Shanxi, between 5 and 10 for Sichuan, Hubei, Hebei, Henan, Tibet, and Shaanxi provinces, while no studies have yet been reported in provinces like Fujian, Hainan, Qinghai, Ningxia, and Shanghai. Nearly one-third of the studies were in North China. The majority of studies only investigated HAP or personal exposure in one single season with a few compared results for two seasons in one publication. Winter and summer periods have appeared to be the two most frequently studied seasons in these publications. Traditional solid fuels like coal and biomass fuels were main household energies in most homes in these studies. There were 20 papers reported HAP from homes using non-solid fuels like biogas, LPG and/or electricity. Three studies had personal exposure measurement for the population using non-solid clean fuels. Regarding stove types studied, one generalized classification in literature is “traditional” versus “improved” stoves. In China, during the National Improved Stove Program (NISP) (1980s-1990s), many rural stoves were technically improved like installation of a chimney or outdoor flue, usage of a grate, optimization of combustion chamber, and adjustment of the distance between ground, grate and bottom of the wok to change primary and/or secondary air supply. Nowadays, though some traditionally simple stoves might be still found in several very poor rural homes, those so-called “improved” stoves became typical and common in most areas. Some biomass gasifier or semi-gasifier stoves (Shen et al., 2012; Du et al., 2017a) and improved coal stoves are available in rural homes. Instead of using the terms of “traditional” and “improved” in stove classification, we prefer to “typical” and “new” stoves in the present analysis. The “typical” stoves are those widely used in the studied region, or stoves used before an intervention program was initiated, while “new” stoves are those with newly developed/improved coal stoves, biomass gasifier or semi-gasifier stoves, or stoves used to replace old stoves in pilot intervention programs. Most stoves are ventilated ones. As seen in Fig. S2, in most HAP and/or personal exposure studies, solid fuels were burned in typical stoves. CO, SO2, PM with different diameters (e.g. TSP, PM10, PM4, PM2.5, and PM1.0), and PAHs were the most frequently measured species (Fig. S3). Some studies quantified metals (including As, Cd, Cr, Pb, Ni, and Hg) and fluorosis contamination. Carbonaceous fractions of PM (EC and OC) and BC were only reported in three publications. 3.2. Household air pollution

Fig. 1. Number of HAP publications in different years when the HAP study was conducted and published.

3.2.1. CO CO is one typical gas pollutant from incomplete combustion process. Most studies available measured CO in homes using solid fuels in summer and/or winter. The CO level ranged from ~1.3 mg/ m3 to ~38 mg/m3, varying in studied site and sampling period as a result of different fuel and stove type, and meteorological conditions (Fig. 3). CO was higher in winter than in summer, although the difference was statistically not significant. And, the concentration was generally lower in the bedroom than the kitchen. The kitchen CO ranged from 0.1 to 35 mg/m3 in homes burning solid fuels, with the

628

W. Du et al. / Environmental Pollution 237 (2018) 625e638

Fig. 2. Geographic distribution of the number of HAP study.

Fig. 3. CO in indoor and outdoor air in winter and other seasons from homes using solid fuels. Data shown are the reported means with standard deviation from studies available in literature.

arithmetic mean of 7.4 mg/m3 in summer, and in winter the kitchen CO concentration ranged from 0.4 to 30 mg/m3, with the arithmetic mean of 9.3 mg/m3, respectively. The daily average CO concentration in bedroom ranged from 1.0 to 3.9 mg/m3 in summer, and in the range of 1.4 to 21 mg/m3 in winter. The overall arithmetic means were 2.5 and 7.0 mg/m3 in the two seasons, respectively. However, Edwards et al. (2007) observed a high CO level in the

bedroom than the CO in the kitchen in a winter, 2002e2003 in rural Shanxi, north China. Thus, the seasonal change and site differences in indoor CO pollution are highly variable among research areas, depending on factors like household characterization, season and stove and fuel types. There is a large variation in CO across homes using different energies, and the difference between homes using solid and nonsolid fuels was statistically not significant, when the comparison was not adjusted for other influencing factors. For homes using gas fuels, the kitchen CO ranged from 2.8 to 12 mg/m3. For example, Wei et al. (2004) found that indoor CO concentration was 11.7 mg/ m3 in homes using biogas in rural Sichuan, southwest China, which was close to Wang et al. (2010)’s study in rural Guizhou, southwest China, for homes using biogas (7.6 mg/m3 ranging from 2.7 to 9.8 mg/m3). The study by Sinton et al. (2004) in three provinces (Shaanxi, Hubei and Zhejiang) reported indoor CO was around 25 ± 31 mg/m3 when LPG was used. Relatively low indoor CO levels were reported by Gong et al. (2014)’s measurement in five provinces during 2009-2010, which reported average indoor CO concentrations were 4.0 ± 3.1 and 2.8 ± 3.3 mg/m3 in homes using biogas and LPG, respectively. For homes using traditional solid fuels, CO had a large variation across different studies. Wang et al. (2010) in rural Guizhou, southwest China reported that the indoor CO levels were 5.5 ± 0.7 and 11 ± 0.8 mg/m3 in homes burning coal and wood, respectively. The study by Wei et al. (2006) in rural Sichuan, southwest China showed that indoor CO concentrations were 4.6, 5.3, and 14.2 mg/m3 when crop straw, wood and coal were combusted.

W. Du et al. / Environmental Pollution 237 (2018) 625e638

3.2.2. SO2 SO2 contamination is a serious issue particularly in homes burning coal. The SO2 concentration ranged from 6.05 to 7400 mg/ m3. The concentrations of SO2 in rural China of different stationary sampling sites, seasons, and energy used are graphed in Fig. 4. For the seasonal and site difference, it can be found that in the summer period, the SO2 concentrations in the kitchen, bedroom and outdoor air were in the range of 24-7400, 7-70, and 6-80 mg/m3, with an overall arithmetic means of 2200, 29, and 36 mg/m3, respectively. While in the winter season, the ranges were 10-3200, 11-1300, and 10-780 mg/m3, respectively. And, the overall arithmetic means of SO2 in the kitchen, bedroom, and outdoor air were 850, 560, and 270 mg/m3, respectively. The contamination level was considerably higher in the kitchen compared to the bedroom and outdoor air, while the concentration in the bedroom was comparable to that in the outdoor air. And, it also reveals that SO2 during the winter is much higher than that during the summer, probably due to widely coal uses for space heating in cold winter. The seasonal differences were more obvious in bedroom and outdoor air compared to the difference in kitchen. For households burning non-solid fuels, Zhang et al. (2007)’s study in rural homes using LPG in Guangzhou, south China reported that the indoor SO2 was 18.4 ± 9.0 mg/m3 in summer. But much high levels were reported by Wei et al. (2004) in rural Sichuan, southwest China, where the kitchen SO2 when burning biogas was as high as 270 mg/m3. The levels were considerably lower than those in

Fig. 4. SO2 in indoor and outdoor air in winter and other seasons from homes using coal (a) and biomass fuels (b). Data shown are the reported means with standard deviation from studies available in literature.

629

homes burning solid fuels. As expected, SO2 concentration was higher in homes burning coals than those burning biomass fuels, in both summer and winter seasons. 3.2.3. PM TSP or PM10 are often studied in past studies, and nowadays more researches are available on fine particles (PM2.5, PM1.0) and even ultrafine particles. In this summary analysis section, data for TSP and PM10 are combined to analyze pollution of coarse particles (Fig. 5), assuming the difference between TSP and PM10 levels is relatively small. Also, results for PM4 and PM2.5 are combined to analysis fine particle concentration (Fig. 6). For TSP and/or PM10 in kitchen, high levels are expectedly reported in winter (935 mg/m3 as an overall arithmetic mean) compared to other seasons (341 mg/m3). In the bedroom, relatively higher concentrations were also found in winter (414 mg/m3 as the overall arithmetic mean) than that in summer (273 mg/m3), but the difference was not as large as that in the kitchen. The seasonal variation for outdoor air was similar to that in the kitchen and bedroom. Regarding the site difference, generally there were higher levels in the kitchen than that in the bedroom, and the lower concentration was in outdoor air. The outdoor PM10 ranged from 23 to 770 mg/m3, with an arithmetic mean of 261 mg/m3. Although the level was lower than that in kitchen and bedroom, it was nearly 1.7 times of the China National Air Quality Standard of 150 mg/m3, and more than 5 times of the standard of 50 mg/m3 set by World Health Organization (WHO). When LPG or biogas were burned, the kitchen PM10 or TSP reported in literature ranged from 104 to 770 mg/m3, with the mean of 422 mg/m3. The level was lower than the kitchen concentration in homes using traditional solid fuels by a factor of ~1.5. It was noted that in one study in Hebei, north China (Zhong et al., 2012), the bedroom concentration in homes using LPG was comparable to the homes using solid fuels, mainly due to the large variances in field measurements and influences of other sources. For fine particles (PM2.5 or PM4), the overall average concentrations in kitchen and bedroom were 338 and 275 mg/m3, respectively. Similar to PM10 (or TSP), the concentrations were higher in the kitchen, but statistically not significant. In outdoor air, the concentration ranged from 12 to 460 mg/m3, with the overall average mean of 152 mg/m3. The ambient PM2.5 exceeded the China National Standard of 75 mg/m3 (MEP, 2012) by a factor of 2, and more than 6 times of the WHO standard of 25 mg/m3 (WHO, 2010). The reported levels in literature for homes burning solid fuels ranged from 62 to 1944 mg/m3 for the kitchen PM2.5, and the range was 63-2334 mg/m3 in bedroom. In homes using biogas or LPG, the kitchen PM2.5 averaged at 174 mg/m3, with the range of reported means in 47-460 mg/m3. The level was about half of the level in homes burning solid fuel. 3.2.4. PAHs PAHs consist of numerous kinds of individual compounds with multiple aromatic rings, existing in either gaseous or particulate phases. Few studies reported both gaseous and particulate phase PAH concentrations. Large quantity of studies only quantified particle-bound PAHs. Most studies available are confined to only U.S. EPA 16 priority PAHs. Chen et al. (2016) measured 28 parent PAHs including 16 priority PAHs and 12 non-priority ones, in rural Shanxi, north China. Though the mass concentration of the 12 nonpriority ones comprised to only about 16%, 8%, and 1% of targeted 28 PAHs in the kitchen, bedroom and outdoor, respectively, it is necessary to note that some non-priority ones are even more toxic than those priority ones. Contamination, transformation, and impacts of these toxic non-priority PAHs in rural areas should be investigated in the future. For the total PAHs of both gaseous and particulate phases, high

630

W. Du et al. / Environmental Pollution 237 (2018) 625e638

Fig. 5. TSP or PM10 in indoor and outdoor air in winter and other seasons from homes using solid (a) and non-solid fuels (b). Data shown are the reported means with standard deviation from studies available in literature.

Fig. 6. PM2.5 in indoor and outdoor air in winter and other seasons from homes using solid (a) and non-solid fuels (b). Data shown are the reported means with standard deviation from studies available in literature.

PAH concentrations were reported by Lv et al. (2010) in rural Yunnan, southwest China, where the indoor and outdoor 16 priority PAH concentration were 8096 and 3260 ng/m3 in spring. But also in rural Yunnan, the study by Lv et al. (2009) in winter reported relatively lower levels of 1716 and 821 ng/m3. The difference

suggested significant spatial and seasonal differences in the studied counties. High contamination levels of PAHs were also reported in norther area, compared to the south area, even in the summer season. For instance, Chen et al. (2016)’s measurement in rural Shanxi, north China, showed that the levels were 7590 ± 6510, 2390 ± 1980, and 2779 ± 3900 ng/m3 in the kitchen, bedroom and outdoor air during a summer season in 2011. Some papers only reported 15 priority PAHs excluding naphthalene, because it is volatile and has large uncertainties compared to other parent PAHs. Duan et al. (2014) reported that the total 15 PAH concentrations in rural Shanxi, north China during winter were 863 and 123 ng/m3 in indoor and outdoor air, respectively. Ding et al. (2012)’s study in rural Hebei, north China during winter found relatively higher levels. The average concentrations of the total 15 PAHs in kitchen, bedroom and outdoor air were 6100 ± 3100, 1900 ± 520, and 2500 ± 1400 ng/m3 in winter, and even in summer the concentrations were as high as 2400 ± 1600, 980 ± 110, and 450 ± 210 ng/m3, respectively (Ding et al., 2012), indicating HAP by PAHs were much severer during wintertime. Despite of the large variations, both studies indicated a notable difference between the indoor and outdoor air. Shen et al. (2014)’s study in rural Jiangsu, east China in fall, reported levels of 546 ± 95, 390 ± 41, and 439 ± 25 ng/m3 in the kitchen, bedroom and outdoor air, respectively. Compared to Ding et al. (2012)’s study in summer in rural Hebei, the indoor concentrations were significantly lower, but the outdoor concentration was comparable. Differences in household structure and meteorological conditions that causing distinct indoor-outdoor air exchange and regional area concentration explained the indoor-outdoor differences in the two regions. As mentioned, many studies only quantified particle-bound PAHs in their measurement. The total 16 PAHs in particle for homes burning solid fuels ranged from 10 to 2329, 5.0 to 754, and 3.4 to 451 ng/m3 in kitchen, bedroom and outdoor air (Fig. 7), with the arithmetic means of 449, 225, and 120 ng/m3, respectively. In households using electricity or LPG, the overall average concentrations were 331, 101, and 179 ng/m3, respectively. Therefore, higher concentrations were found in the kitchen compared to the bedroom and outdoor air, and generally the levels were higher in homes burning solid fuels than those using non-solid ones, although the difference between fuel type was statistically not significant before adjusting other influencing factors. The BaP concentrations (mainly in particulate phase) in the kitchen, bedroom, and outdoor air at homes burning solid fuels

W. Du et al. / Environmental Pollution 237 (2018) 625e638

Fig. 7. The concentration of particulate priority PAHs in winter and other seasons from homes using solid (a) and non-solid fuels (b). Data shown are the reported means with standard deviation from studies available in literature.

ranged from 2.0-901, 0.8-12, and 0.5-141 ng/m3 (Fig. 8), with the overall means of 150, 4.6, and 30 ng/m3, respectively. In homes burning LPG or biogas, the BaP concentration in kitchen, bedroom, and outdoor air ranged from 0.2-13, 2-17, and 3-17 ng/m3, with the overall means and standard deviations of 8.4, 9.6 and 10 ng/m3, respectively. As expected, the indoor BaP level was significantly lower at homes burning LPG compared to that in homes burning solid fuels. Note that the daily average BaP limit in National Air Quality Standard is 2.5 ng/m3 in outdoor air, and 1.0 ng/m3 in indoor air. Thus even in homes using LPG, the BaP level was still above the limits, indicating that inhalation exposure to BaP at this level would cause considerably adverse impacts on human health like lung cancer. Besides parent PAHs, contamination and adverse health outcomes of PAHs derivatives, including nitrated and oxygenated PAHs, had been also documented in HAP studies in China. Yu et al. (1986)’s study conducted during a spring period in Xuanwei, southwest China, quantified nitrated PAHs in TSP in indoor and yard in households burning coal and wood. The average concentrations of total nitrated PAHs in TSP were 14.3 and 2.6 ng/m3 in indoor air, and were 0.313 and 0.015 ng/m3 in yard, for homes burning coal and wood, respectively. However, the second study on PAH derivatives we found is Chen et al. (2017)’s study that was nearly thirty years after Yu et al. (1986)’s study. Chen et al. (2017)’s study in rural Shanxi, north China during a summer season, showed that the total nitrated PAHs including both gaseous and particulate phases were 2.2 ± 2.5, 0.73 ± 0.48, and

631

Fig. 8. BaP in winter and other seasons from homes using solid (a) and non-solid fuels (b). Data shown are the reported means with standard deviation from studies available in literature. The red and dark lines are national air quality standards for indoor and outdoor BaP at 1.0 and 2.5 ng/m3. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

0.33 ± 0.21 ng/m3 in the kitchen, bedroom and outdoor, respectively, and the total oxygenated PAHs were 230 ± 520, 61 ± 120, and 9.9 ± 1.8 ng/m3 in the kitchen, bedroom and outdoor, respectively. The levels of oxygenated PAHs were approximately 1-2 orders of magnitude higher than the nitrated ones. The two studies both verified much higher levels of PAH derivatives in indoor air compared to the corresponding outdoor air (p<0.05). 3.2.5. Other pollutants 3.2.5.1. BC and EC. Shan et al. (2014) measured BC concentration in kitchen and outdoor air in rural Sichuan, southwest China during the fall in 2012. The daily average BC concentrations in kitchen and outdoor air were 3.5 (0.7-11.8 as range) and 0.8 (0.5-1.2) mg/m3, respectively. In contrast, thermal-optical method was used to measure EC concentration in particles in other studies. The reported mean EC concentration ranged from 1.4 ± 1.0 to 29 ± 17 mg/ m3. Zhong et al. (2012) reveal that the indoor EC concentrations in rural Hebei, north China in 2010 were 10 ± 1 and 27 ± 10 mg/m3 in the summer and winter, respectively, comparable to the outdoor EC at 12 ± 4 and 29 ± 17 mg/m3, respectively. A similar indoor-outdoor and seasonal difference was reported by Zhu et al. (2012)’s study in rural Shaanxi, northwest China, in 2008, though the levels were lower than that in rural Hebei. It reported that the average indoor EC concentrations were 2.8 and 6.2 mg/m3 in summer and winter, and the outdoor EC concentrations were 3.2 and 6.6 mg/m3,

632

W. Du et al. / Environmental Pollution 237 (2018) 625e638

respectively. Wang et al. (2010) reported that the indoor EC concentrations in rural Guizhou, Southwest China in spring were 11.9 ± 1.9, 7.1 ± 3.0, and 1.4 ± 1.0 mg/m3 in homes burning coal, wood and biogas, respectively. Relatively lower EC levels were observed when clean fuels were used like biogas. 3.2.5.2. OC. OC is often simultaneously measured with EC. Similar to EC, the indoor-outdoor difference was small and higher levels were in winter than that in summer. For example, the indoor and outdoor OC concentrations in rural Hebei, north China were 28 ± 6 and 28 ± 5 mg/m3; in summer, and 91 ± 30 and 136 ± 83 mg/m3 in winter, respectively (Zhong et al., 2012), and in rural Shaanxi, northwest China, the indoor and outdoor OC concentrations were 18.0 and 17.6 mg/m3 in summer, and 61.1 and 58.6 mg/m3 in winter (Zhu et al., 2012). The OC concentration in homes burning biogas was lower than that of using traditional solid fuel households. The concentrations of indoor OC in homes burning coal, wood and biogas in rural Guizhou, southwest China, were 70.8 ± 3.5, 85.8 ± 22, and 22.9 ± 5.9 mg/m3, respectively (Wang et al., 2010). The OC/EC ratio was in 2-16, with relatively higher ratios in winter than that in summer. 3.2.5.3. Fluoride. Fluorine contamination HAP is serious, particularly in regions using low quality coals (Ando et al., 1998). Fluorine was usually measured by the ion selective electrode method in literature. The concentration of fluoride in kitchen ranged from 1.7 to 134 mg/m3 in homes burning coal. The highest pollution level (134 ± 187 mg/m3) was found in the households in rural Sichuan which were using typical coal stoves (Wang et al., 2010), and the level was reduced to 9.1 ± 3.9 mg/m3 after the stove improvement program. Significant reduction in ambient fluoride pollution through coal stove improvement was also reported in rural Guizhou, southwest China (Zhang et al., 2006), Henan, central China (Chen et al., 2003), and Shanxi, north China (Hu and Yang, 2000), reducing from 23 ± 25 to 4.8 ± 4.4, from 10.4 ± 12.9 to 1.6 ± 1.5, and from 69.6 ± 4.2 to 3.9 ± 1.9 mg/m3, respectively, with 80e95% of reduction on average. The fluoride HAP was relatively lower in bedroom and outdoor air compared to that in the kitchen. The concentration of fluoride was 0.1e3.6 and 0.3e1.6 mg/m3 in the bedroom and outdoor air at homes burning coals. For homes consuming biomass fuels, the fluoride concentration was 0.06e1.4 mg/m3, which was also lower than the kitchen fluoride pollution in coal-use households. One study conducted in rural Zhejiang, east China, reported that in homes using LPG, the indoor fluoride was 13.6 ± 8.3 mg/m3 (Zhou et al., 2006), which was considerably highly polluted, even though the concentration was lower than that in homes burning coals in the same region (33 ± 21 mg/m3). 3.2.5.4. Heavy metals. Arsenic (As), Lead (Pb), Cadmium (Cd), and Mercury (Hg) were mostly reported in indoor and outdoor particles from rural Chinese households. Other heavy metals such as Cr, Fe, Mn, and Zn were sometimes measured and reported as well. As concentration ranged from 5.4-80, 4.5-51, and 4.7-56 pg/m3 when biomass, coal and LPG were combusted, respectively (Fig. S4). The arithmetic and geometric means were 22 and 15 pg/m3 in homes burning biomass, 23 and 20 pg/m3 in households using coal, and 25 and 18 pg/m3 in homes burning LPG. There was no significant difference found among these three different fuel-use households (p>0.05), and statistically no significant differences between the kitchen, bedroom and outdoor air (p>0.05). For Cd, Wu et al. (2015) reported that the concentration in household air in rural Henan, central China in 2012 was in the range of 2.2 ± 0.8 to 10.0 ± 15.8, 2.0 ± 0.9 to 5.8 ± 2.3, and 1.5 ± 0.7 to 13.7 ± 18.5 pg/m3,

respectively, when biomass, coal, and LPG/electricity were used. The concentrations increased in winter compared to the fall, which was consistent with the results of seasonal difference for PM2.5. Relatively higher concentrations were observed in the kitchen, compared to the living room and outdoor air, but the difference was statistically not significant (p>0.05). Pb concentration ranged from 47-380, 95-400, and 39-375 pg/ m3 for homes burning biomass, coal and LPG/electricity, respectively (Fig. S5). Relatively higher levels were observed in kitchen compared to the living room and outdoor air, and the concentration increased in winter. When adjusted for the seasonal difference, the Pb concentration in homes burning coal was higher than that burning biomass, however, the levels between homes burning coals and those using LPG/electricity were comparable, which was probably attributable to other sources such as cooking oil fumes, incense burning and smoking instead of fuel combustion emission (Wu et al., 2015). For mercury (Hg), Du et al. (2010) reported that the indoor and outdoor concentrations in rural Hebei, north China, using coals were 516 and 72.4 pg/m3, which were higher than those using straws in rural Liaoning, northeast China (55.1 and 35.1 pg/m3), and those burning wood in rural Shandong, north China (73.9 and 83.7 pg/m3). High Hg levels in area burning coals were also reported by Zhao et al. (2012) in rural Hebei, north China, during 2009e2010 (321 ± 143 to 1631 ± 413 pg/m3 as ranges, and higher in winter than that in summer). 3.2.5.5. NOx. Seow et al. (2016) reported that kitchen NOx concentration ranged from 80 to 164 mg/m3, varying in fuel and stove types, in rural Yunnan, southwest China during 2008-2009. NOx concentration usually increases during the cooking period. In a measurement in rural Hubei, central China in the winter of 2013, it was found that during the cooking period, the NOx concentration increased by 2-7 times (Xiang et al., 2016). During the non-cooking period, indoor NOx concentration was close to the outdoor level, and similar in homes using different fuel-stove combinations. 3.3. Daily inhalation exposure To better quantify daily inhalation exposure to HAP, some previous studies estimated average exposure based on measured average area air concentrations and time-spent in different microenvironments. However, due to temporal variations in area concentrations (such as peak concentration during a cooking period) and many different microenvironments, there were considerable biases and uncertainties in this estimation compared to the “true” exposure concentrations. In a summertime study in rural Shanxi, north China, Huang et al. (2017) found that though the calculated time-weighted average was positively correlated with directly measured inhalation exposure level, the former was about 30% lower. A notable discrepancy was also reported by Du et al. (2017b) despite the positive correlation. Personal portable samplers are expected to have a more accurate assessment on daily inhalation exposure and have gained popularity in more recent studies, though such studies are still very limited in number. 3.3.1. CO There are three studies that estimated personal exposure to CO. One study by Pan et al. (2001) estimated daily exposure to CO was 2.3 ± 1.6 mg/m3 for the male and 2.5 ± 2.4 mg/m3 for the female, based on area concentrations using stationary samplers while the other two more recent studies used portable samplers. Fischer and Koshland (2007) measured inhalation exposure to CO among rural residents in Jilin, north China by using colorimetric diffusion tubes during the heating seasons of 2001-2003. The primary cook's daily

W. Du et al. / Environmental Pollution 237 (2018) 625e638

inhalation exposure to CO was 9.6 ± 2.8 mg/m3, with median and geometric means being 4.8 and 5.0 mg/m3, respectively. This exposure was not significantly different from the kitchen area concentration, and a moderate association between paired area and personal exposure measurements was observed. Alnes et al. (2014) examined inhalation exposure to CO among over 1700 women residents in the rural Guizhou, southwest China, in 2009. The overall arithmetic and geometric means were 2.3 and 1.6 mg/m3, separately, with a maximum exposure of 39 mg/m3. The exposure was associated with a variety of factors including stove and fuel type, ventilation, kitchen configuration, secondhand smoke, time spent outdoor/indoor, and ambient temperature. The measured inhalation exposure to CO was weakly correlated with indoor CO concentration in the living room of wood users, but not the kitchen. After adjusting for temperature (seasonal difference), the correlation between inhalation exposure and the area concentration in the living room was significant (r¼0.57, p<0.01). No significant correlation was found between the CO inhalation exposure and indoor CO level for coal users. Both studies had used portable samplers yet the reported numbers differed by 4-fold on average, which could be attributed to the region difference, the distinctive fuel-stove systems, different household characteristics, the population's occupation, as well as the study season and year. For example, the northeast region usually had worse air pollution when compared to the southwest region, consequently, the inhalation exposure for the population in the northeast region was usually higher than that for those from southwest. Alnes et al. (2014) had more measurements in the spring, while Fischer and Koshland (2007) conducted measurements only in winter heating periods. In addition, Alnes et al. (2014) found a unit increase in temperature was associated with 4% reduction in CO exposure for the population in rural Guizhou, southwest China. 3.3.2. SO2 Exposure to SO2 was only reported in two studies by Pan et al. (2001) and Liu et al. (2011). Pan et al. (2001) estimated daily inhalation exposure from measured area concentration and the amount of time rural residents spent indoor and outdoor during the winter in Anhui, east China. The exposure concentrations to SO2 were 23 ± 67 and 25 ± 70 mg/m3 for the male and the female, respectively. Using personal portable absorbing tubes, Liu et al. (2011) compared inhalation exposure to SO2 between the population that adopted new stoves and those still used typical ones during the 2006e2007 in rural Guizhou, southwest China. The results showed that the use of new stoves had lowered daily average inhalation exposure to SO2 from 970 ± 1380 mg/m3 to 680 ± 670 mg/m3. Overall, SO2 exposure was much higher in rural Guizhou than that in Anhui, which was believed to be driven by the widespread use of coal in Guizhou, southwest China whereas biomass fuels like crop straw and wood were more often used in Anhui, east China. 3.3.3. PM The number of studies on daily inhalation exposure to PM was limited. In the study by Pan et al. (2001), daily inhalation exposure to PM10 of the population using biomass in the rural areas of Anhui, east China was estimated to be about 556 ± 535 and 659 ± 646 mg/ m3 for the male and female separately during a winter, based on measured area concentration and time spent in different microenvironments (e.g., kitchen, bedroom, living room and outdoor). Gao et al. (2009) conducted a measurement in Tibet, southwest China during 2006-2007. The estimated daily exposure to PM2.5 was 119 (80-157, as 95% confidence interval) mg/m3 for the population using solid fuels, and was 68 (50-87 as 95% CI) mg/m3 for the population using biogas.

633

There were relatively more studies using portable samplers to evaluate inhalation exposure to PM with a preponderance to PM2.5. Most of the studies were conducted after 2008, and published after 2010. The exposure to PM2.5 ranged from ~30 mg/m3 to ~600 mg/m3, varying by different fuels and stove types, cooking behaviors, sampling seasons, and study locations, etc. The lowest exposure level (31 ± 10 mg/m3) was found in rural Yunnan, southwest China, during a summer season, and the highest exposure (590 ± 220 mg/ m3) was reported in rural Hebei, north China in a cold winter season when solid fuels were widely combusted for daily cooking and heating. These field measurement campaigns were mainly in north and northeast China regions (Hebei, Shanxi, Liaoning and Inner Mongolia provinces), and southwest China region (Yunnan, Sichuan and Guizhou). The exposure in the Southwest was generally lower than that in the North and Northeast area. Winter and summer were the most frequently studied seasons and only one study reported inhalation exposure among the cooking women in fall (Shan et al., 2014). As expected, the exposure was higher in winter compared to summer. Based on compiled data, the overall geometric means of exposure to PM2.5 among rural residents in summer and winter of north and northeast China were 121 and 359 mg/m3 separately. For the population in southwest area, the daily exposure levels were 160 and 69 mg/m3 in winter and other seasons (spring, summer and fall), respectively (Fig. 9). The usage of clean energies like electricity, natural gas, biogas and LPG was found to lower HAP, which in turn resulted in lower inhalation exposure to air pollutants, compared to the use of solid fuels (Baumgartner et al., 2011; Du et al., 2017; Hu et al., 2014; Huang et al., 2017). Huang et al. (2017) reported that the daily exposure was about 30e40% higher for the residents who burned solid fuels (coal, peat and wood), compared to those used clean energies like electricity and gas. Among residents who used electricity and/or gas, ventilation instrument had an impact on inhalation exposure to particulate matters. The daily average exposure to PM2.5 and PM1.0 were 96 ± 61 and 78 ± 61 mg/m3 when the kitchen was not equipped with a fan or smoke exhaust system, which could be decreased to 81 ± 39 and 60 ± 40 mg/m3 when a ventilation instrument was used. Note that this level was still higher than the WHO guideline for ambient PM2.5 which was “requisite” to protect public health. The daily inhalation exposure would be much higher in cold winter, even when clean energies were used. Du et al. (2017b) reported the daily inhalation exposure to PM2.5 and PM1.0 in a winter season in rural Guizhou, southwest China were 166 ± 45 and 141 ± 41 mg/m3 even when using electricity, which was lower than 184 ± 83 and 161 ± 73 mg/m3 for residents burning wood but not statistically significant (p > 0.05). The risk of adverse health outcomes due to PM exposure were not only associated with the mass concentration, but also with factors like size distribution and the particle's chemical compositions. Huang et al. (2017) used a four-stage cascade impactor to evaluate size distribution of inhaled PM among rural residents in Shanxi, north China during a summer season. The overall average inhalation exposure to TSP, PM2.5, PM1.0 and PM0.25 were 144 ± 71, 98 ± 52, 77 ± 47, and 48 ± 32 mg/m3, respectively. The average mass fraction of PM2.5 in TSP was around 70%, and PM1.0 contributed to nearly 90% of the PM2.5 mass. The same instrument was also used by Du et al. (2017b) in a study in wintertime. The daily average inhalation exposure to TSP, PM2.5, PM1.0 and PM0.25 were 214 ± 86, 176 ± 69, 153 ± 61, and 73 ± 46 mg/m3, for residents in rural Guizhou, southwest China, and were 517 ± 318, 451 ± 301, 375 ± 248, and 215 ± 202 mg/m3, for residents in rural Shanxi, north China. The mass fraction of PM2.5 in TSP was over 80%, and again in PM2.5, a majority was PM1.0. The presence of higher fine particles could potentially cause more severe health outcomes owing to their ability to penetrate deeper into lung tissue.

634

W. Du et al. / Environmental Pollution 237 (2018) 625e638

Fig. 9. Personal daily exposure to PM2.5 for the population using solid (a) and non-solid fuels (b) in the north and southwest regions where relatively more personal exposure measurements were available in literature. Data points are means and standard deviations collected from the studies, and the red lines are recalculated overall means for each group. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

There was only one study we found quantifying detailed particle chemical compositions. In this study, Secrest et al. (2016) not only measured the mass of PM2.5 but also investigated chemical compositions including water-soluble organic matter, ions, metals and some non-metals, and identified sources of particles in inhaled air among the studied rural women population. Such studies provided vital information for identifying the source of inhalation exposure and more similar studies are expected in the future to provide direct and important supports for air pollution control. 3.3.4. BC Daily inhalation exposure to BC was measured by Baumgartner et al. (2011) in Yunnan, southwest China from 2008 to 2009, and in Sichuan, southwest China and Inner Mongolia, northwest China in 2012 (Secrest et al., 2016; Shan et al., 2014). The geometric mean exposure concentrations for rural women from Yunnan in summer and those from Sichuan in fall were 4.0 (2-14 as range) and 3.4 (1.57.8 as range) mg/m3 respectively. In winter, the geometric mean exposure increased to 6.0 (2-44 as range) mg/m3 for rural women in Yunnan and 8.3 (2.6-27 as range) mg/m3 for rural women from Inner Mongolia, indicating a strong seasonal variation. 3.3.5. PAHs Measurements on PAHs using personal portable samplers were found in 5 papers, with 2 in north China (Shanxi, Chen et al., 2016; and Hubei, Ding et al., 2012), and one in each for east China (Jiangsu, Shen et al., 2014), southwest China (Yunnan, Downward et al., 2014) and central China area (Hubei, Lin et al., 2016). The level of exposure to PAHs varied greatly by season and energy source. In a study conducted in rural Hebei, north China, Ding et al. (2012) reported that the exposure to BaP for the cooking and non-cooking population were 200 ± 160 and 59 ± 37 ng/m3 in the winter, which decreased sharply to only 2.0 ± 0.4 and 1.6 ± 0.3 ng/ m3 in the summer. Chen et al. (2016) reported that the daily exposure to BaP for the population using wood, coal briquettes, LPG, and electricity were very different, 288, 5.6, 2.7, and 2.1 ng/m3 respectively in the summer. Similar variations across different household energies were also reported by Downward et al.(2014),

whose measurement in rural Yunnan, southwest from 2008 to 2009 showed that the exposure to BaP for the population using smoky coal, smokeless coal and biomass fuels were 74, 152, and 67 ng/m3, respectively. To examine the impact of clean stove adoption on PAH exposure, Lin et al. (2016) compared inhalation exposure to PAHs among rural residents in Hubei, central China between the group burning wood in gasifier stoves and those burning coals. The total PAHs concentrations were similar but the composition profiles were different between the two comparison groups. The BaP exposure were 27 ± 33 and 10 ± 11 ng/m3 respectively for the coal stove and wood gasifier stove users, which did not show a significant decline after adopting biomass gasifier stoves. Because of the simultaneous change in fuel and stoves, their results did not indicate a failure in improved biomass gasifier stoves. Indeed, the exposure reported among the wood users using gasifier stoves did decrease notably compared the exposure among wood users in traditional or typical stoves (Lin et al., 2016). The research by Shen et al. (2014) in rural Jiangsu, east China reported relatively lower exposure to PAHs in fall, 2012. The daily exposure to BaP were 3.3 ± 1.8 and 1.7 ± 0.5 ng/m3, for the cooking and non-cooking group, respectively. In this study, Shen et al.(2014) found that the calculated time-weighted average exposure to PAHs was lower than the directly measured exposure, and there was a significant underestimation in gaseous PAHs. Inhalation exposure to PAH derivatives were only found in two studies by Chen et al. (2017) in rural Shanxi, north China in the summer of 2011, and Shen et al. (2016) in rural Hubei, central China in winter 2012. In rural Hubei, central China, the daily inhalation exposure to the total nitrated and oxygenated PAHs were 0.14 and 6.2 ng/m3 for the population burning coals, which was found to be lower than 0.21 and 22 ng/m3 among those burning wood in gasifier stoves. In rural Shanxi, north China, although the measurement was in a summer season, the daily inhalation exposure to the total nitrated and oxygenated PAHs were 0.47 ± 0.28 and 18 ± 22 ng/m3, respectively, which were apparently higher than the levels reported in rural Hubei during a winter season.

W. Du et al. / Environmental Pollution 237 (2018) 625e638

4. Discussion In summary, this review achieved two purposes: first, the study assessed the status of research on HAP and inhalation exposure in rural China; second, this review examined the ranges and temporal-spatial variations of HAP and inhalation exposure as well as their differences in groups burning solid fuels and non-solid household energies. This study reviewed peer-reviewed publications in both English and Chinese. After title and abstract screening, full-text review and quality assessment, 92 papers were eventually included in data extraction and synthesis. In most studies included in this review, the daily average concentrations of air pollutants were well above China's national air quality standards and WHO standard limits. As expected, the use of clean energies reduced HAP. However, due to severe ambient air pollution, HAP in those homes adopting clean energies were still higher than the air quality standard (MEP, 2012; WHO, 2010; GAQAIQ, 2002). Portable instruments were preferable to evaluate inhalation exposure to air pollutants. Fortunately, the technology had been applied in more recent field studies, although the number of published studies using portable instruments was still very limited. In the current literature review, only two studies have used personal portable samplers to measure inhalation exposure concentration to CO, and only one study has used them to evaluate SO2 exposure concentration. There were relatively more studies on PM inhalation exposure, but most of them were based on gravimetric measurement of filters, instead of real-time instruments. In general, PM exposure was lower in summer than in winter, and the exposure in the southwest region was lower than in north and northeast area of China. Both the use of clean energies and use of a ventilation system could reduce PM inhalation exposure, although the level was still high above the WHO guideline level after fuels switching and ventilation system installation. This suggested that changes in household energy in pilot homes or villages alone were not enough and a concerted effort to simultaneously control air pollution in the village or regional scale is needed. Reducing widespread use of coals and woody materials for heating demand in cold winter, especially those low quality raw coals in low efficiency heating stoves, was one aspect that should be studied and controlled, which could improve both indoor and outdoor air quality. Researches had indicated that adverse health outcomes associated with PM exposure were not solely determined by the mass concentration but also by other characteristics like particle size and chemical compositions. Previous studies showed that for rural China, the mass fraction of PM1.0 in PM2.5 in inhaled air could be as high as 90% (Huang et al., 2017). Such high fractions of smaller particles would lead to more severe health impacts. Chemical compositions of PM were also studied by a few studies. The source apportionment result showed that biomass and coal combustions were major sources of exposed PM with considerable contributions from other sources like dust (Secrest et al., 2016). The source profile could also vary by region and study period. The information is crucial for conducting effective controls and deserves more extensive investigations. For PAH inhalation exposure, there was a similar trend that the exposure was lower in summer and among the population using clean energies like gas and electricity. In Northern China, during the summer season, except a very high exposure reported for wood users in rural Shanxi (due to use of unvented wood stoves indoor, and specific household structure like one shared room used for cooking, rest and sleep), the daily inhalation exposure to BaP was generally around several ng/m3. Applying a unit risk for lung cancer at 8.7  105 per ng/m3 BaP (WHO, 2010), the level of BaP exposure

635

corresponds to a lifetime cancer risk in the magnitude of 1.0  104. Therefore, even just considering the exposure at summer time when clean energies were used, the risk was still above the acceptable level and could be even higher than the serious level of 1  104. Besides BaP, many other isomers, particularly some nonpriority PAHs like dibenzopyrenes (Zhuo et al., 2017), have been documented to be carcinogenic and mutagenic, and are more toxic. When these PAHs were taken into account, the health risk associated with PAHs inhalation exposure would be larger. Even worse, a few studies also reported high occurrences of PAHs derivatives like nitrated PAHs and quinones that could be more toxic than the parent PAHs. As such, it's very important to control PAHs and the derivatives. Since PAHs are usually from incomplete combustion process, the control of PAHs could be often achieved as a co-benefit of air pollution controls on criteria pollutants like primary PM and SO2 on combustion sources (e.g. coal combustions). Current studies on HAP in rural China were limited, particularly compared to studies on other pollution sources. In the review, it revealed that HAP was affected by various influencing factors, such as room structure, season, fuel/energy type, region, meteorological conditions etc. However, quantitative information for these influencing factors were seldom reported, which made it hard to analyze the individual and interacting effects of these influencing factors. More studies in large scale that will provide more quantitative data are urgently needed. In addition, although clean fuels and improved stove intervention programs had been initiated by the central and local governments, and/or non-profit organizations, results from these pilot projects were very scarce. Due to the limited resources, health outcomes from solid fuel use in rural China, the health impacts associated with HAP, and improvements in health conditions after clean fuel and stove intervention were not included in this study. For a better understanding of HAP in rural China and filling the data and knowledge gap in this important field, more researches were needed in the future, especially in the areas of:  Characterization of household air pollution in rural area, and the temporal-spatial variations. There were large variations in fuels and stoves used across the country. And fast urbanization and industrialization has changed household energy structure for not only cooking but also heating. More field surveys and campaigns were required to characterize household air pollution of CO, SO2, PM and toxic components like heavy metals and PAHs. For PM, not only the total mass concentration but also size distribution and fractions of ultrafine particles should be studied.  Identification of key factors influencing indoor and outdoor air quality. HAP was affected by many factors such as household structure, fuel and stove type, and meteorological conditions. The number of factors in play and their influence might vary by region, and study period. Identification of the key factors and understanding how they alter pollution status is necessary to design effective solutions to improve air quality.  Evaluation of clean fuel and stove intervention programs and enablers promoting sustainable use. As mentioned above, while some intervention programs had been initiated, evaluation on their impacts were still scarce. Field evaluation is necessary for these programs, as the performance of new stoves and fuels might be different from that in laboratory test, as pointed out by some previous studies.  Assessment of personal exposure to air pollutants. Though daily inhalation exposure could be estimated as time-weighted average concentration from area concentration and activity pattern, personal portable samplers were preferable for a more accurate estimation of daily inhalation exposure. The

636

W. Du et al. / Environmental Pollution 237 (2018) 625e638

technology had been adopted in some past studies, but the number of studies were still limited, and the sample size was relatively small. In addition, there was a need to assess exposure to more air pollutants besides CO, SO2 and PM.  Apportionment of sources of household air pollution. Source apportionment was important for pollution control measures. Most current HAP studies focused only on contamination status and temporal-spatial variations of targeted pollutants and the study on the source of household air pollution were still lacking. Different source apportionment methods were required for different pollutants. For PM, the most widely used technology nowadays was based on unique source tracers and receptor models and it usually requires a large sample size to lower the uncertainty. Seasonal variations should also be considered in examining the pollution sources.  Health outcomes of household air pollution through biomarker analysis and epidemiological studies. Though not covered by this study, it is interesting and necessary to look into health outcomes of HAP through the analysis of unique biomarkers and epidemiological studies. Such study would enrich our understanding on the adverse impact of HAP, and help convince the public and policy makers to take more strident actions on this important issue.

5. Conclusions Owing to low efficiency and large residential consumption from solid fuel combustion with various traditional/typical stoves, HAP and induced inhalation exposures were severe in rural China. Residential combustion issues were widely concerned around the world and many peer-reviewed papers have been published, however, seldom had been conducted in China. Thus, this review was addressed by analyzing 92 of the publications both in English and Chinese, which were all conducted in rural China. A general conclusion was that for most air pollutants, severer HAP in kitchen than living room and outdoor, and higher in winter than summer. It was demonstrated that replacing traditional solid fuel/stove can reduce HAP for most air pollutants. however, large variations were still found from the different household structures, various stove-fuel types, meteorological conditions, and temporal and spatial influencing factors. Though some clean fuels or energies (including LPG, biogas, electricity) or stove intervention programs (from typical stoves to new ones) had been introduced, the HAP levels for these air pollutants were still exceeded the WHO and Chinese National Standards. In the assessment of inhalation exposure levels, portable personal samplers were more preferred compared to the estimation based on time-weighted average concentration. Many recent studies for inhalation exposure had conducted using portable personal samplers. Similar with the HAP, the results of inhalation exposure to most of the pollutants in rural China show higher inhalation exposure levels for the residents using traditional solid fuels and typical stoves compared with cleaner ones. The use of cleaner energies, such as LPG, biogas, electricity etc., could also mitigate the inhalation exposure level and human health, but still high above the standard levels set by WHO, since most intervention programs were promoted in small scale and last for a relatively short period. This suggested further programs should be promoted in the future in national wide scale and last for longer time. Fuel is one important factor affecting HAP and exposure, however, other factors like ventilation conditions also influence air quality notably. Unfortunately, quantitative information on other influencing factors was scarce, and the interacting effects of these factors were poorly understood. Future studies are expected to look

into this. Also, based on the results and findings from this review, we suggested some data gaps and future researches that might be interesting for the research community and informative for policy makers to make effective pollution control strategies. Acknowledgements The study is partly supported by National Natural Science Foundation of China (41701584) and China Postdoctoral Science Foundation (2017M612025). Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2018.02.054. Competing interests The authors declare no competing interests. References Alnes, L.W., Mestl, H.E., Berger, J., Zhang, H., Wang, S., Dong, Z., Ma, L., Hu, Y., Zhang, W., Aunan, K., 2014. Indoor PM and CO concentrations in rural Guizhou, China. Energy Sustain. Dev. 21, 51e59. Ando, M., Tadano, M., Asanuma, S., Tamura, K., Matsushima, S., Watanabe, T., Kondo, T., Sakurai, S., Ji, R., Liang, C., Cao, S., 1998. Health effects of indoor fluoride pollution from coal burning in China. Environ. Health Perspect. 106, 239. Baumgartner, J., Schauer, J.J., Ezzati, M., Lu, L., Cheng, C., Patz, J., Bautista, L.E., 2011. Patterns and predictors of personal exposure to indoor air pollution from biomass combustion among women and children in rural China. Indoor Air 21, 479e488. Bonjour, S., Adair-Rohani, H., Wolf, J., Bruce, N.G., Mehta, S., Pruss-Ustun, A., et al., 2013. Solid fuel use for household cooking: country and regional estimates for 1980-2010. Environ. Health Perspect. 121, 784e790. Booth, A., O'Rourke, A., Ford, N., 2000. Structuring the pre-search reference interview: a useful technique for handling clinical questions. Bull. Med. Libr. Assoc. 88, 239e247. Chafe, Z.A., Brauer, M., Klimont, Z., Dingenen, R.V., Mehta, S., Rao, S., Riahi, K., Dentener, F., Smith, K., 2014. Household cooking with solid fuels contributes to ambient PM2.5 air pollution and the burden of disease. Environ. Health Perspect. 122, 1314e1320. Chen, Z., Zhou, H., Jia, C.J., 2003. Investigation on the impacts on fluoride reduction effectiveness through stove improvement in West Shaanxian county in Henan. Chin. J. Endemiol. 22, 93 (In Chinese). Chen, Y., Du, W., Shen, G., Zhuo, S., Zhu, X., Shen, H., Huang, Y., Su, S., Lin, N., Pei, L., Zheng, X., Wu, J., Duan, Y., Wang, X., Liu, W., Wong, M., Tao, S., 2017. Household air pollution and personal exposure to nitrated and oxygenated polycyclic aromatics (PAHs) in rural households: influence of household cooking energies. Indoor Air 27, 169e178. Chen, Y., Shen, G., Huang, Y., Zhang, Y., Han, Y., Wang, R., Shen, H., Su, S., Lin, N., Zhu, D., Pei, L., Zheng, X., Wu, J., Wang, X., Liu, W., Wong, M., Tao, S., 2016. Household air pollution and personal exposure risk of polycyclic aromatic hydrocarbons among rural residents in Shanxi, China. Indoor Air 26, 246e258. Cheng, G., 2004. A study of clinical questions posed by hospital clinicians. J. Med. Libr. Assoc. 92, 445e458. Clark, M., Peel, J., Balakrishnan, K., Breysse, P., Chillrud, S., Naeher, L., Rodes, C., Vette, A., Balbus, J., 2013. Health and household air pollution from solid fuel use: the need for improved exposure assessment. Environ. Health Perspect. 121, 1120e1128. Cohen, A., Brauer, M., Burnett, R., Anderson, H., Frostad, J., Estep, K., Balakrishnan, K., Brunekreef, B., Dandona, L., Dandona, R., Feigin, V., Freedman, G., Hubbell, B., Jobling, A., Kan, H., Knibbs, L., Liu, Y., Martin, R., Morawska, L., Pope III, C.A., Shin, H., Straif, K., Shaddick, G., Thomas, M., van Dingenen, R., van Donkelaar, A., Vos, T., Murray, C., Forouzanfar, M., 2017. Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Disease study 2015. Lancet 389, 1907e1918. Ding, J., Zhong, J., Yang, Y., Li, B., Shen, G., Su, Y., Wang, C., Li, W., Shen, H., Huang, Y., Zhang, Y., Cao, H., Zhu, Y., Simonich, S.L., Tao, S., 2012. Occurrence and exposure to polycyclic aromatic hydrocarbons and their derivatives in a rural Chinese home through biomass fuelled cooking. Environ. Pollut. 169, 160e166. Downward, G.S., Hu, W., Rothman, N., Reiss, B., Wu, G., Wei, F., Chapman, R.S., Portengen, L., Qing, L., Vermeulen, R., 2014. Polycyclic aromatic hydrocarbon exposure in household air pollution from solid fuel combustion among the female population of xuanwei and fuyuan counties, China. Environ. Sci. Technol. 48, 14632e14641. Du, J., Huang, X., He, L., Luan, S., Yu, G., 2010. Characteristics of particulate mercury

W. Du et al. / Environmental Pollution 237 (2018) 625e638 in typical rural household of Northern China. J. Anhui Agric. Sci. 38, 3690e3963 (In Chinese). Du, W., Shen, G., Chen, Y., Zhu, X., Zhuo, S., Zhong, Q., Qi, M., Xue, C., Liu, G., Zeng, Z., Xing, B., Tao, S., 2017a. Comparison of air pollutant emissions and household air quliaty in rural homes using improved wood and coal stoves. Atmos. Environ. 166, 215e223. Du, W., Shen, G., Chen, Y., Zhuo, S., Xu, Y., Li, X., Pan, X., Cheng, H., Wang, X., Tao, S., 2017b. Wintertime pollution level, size distribution and personal daily exposure to particulate matters in the northern and southern rural Chinese homes and variation in different household fuels. Environ. Pollut. 231 (1), 497e508. Duan, X., Wang, B., Zhao, X., Shen, G., Xia, Z., Huang, N., Jiang, Q., Lu, B., Xu, D., Fang, J., Tao, S., 2014. Personal inhalation exposure to polycyclic aromatic hydrocarbons in urban and rural residents in a typical northern city in China. Indoor Air 24, 464e473. Edwards, R.D., Liu, Y., He, G., Yin, Z., Sinton, J., Peabody, J., Smith, K.R., 2007. Household CO and PM measured as part of a review of China's national improved stove program. Indoor Air 17, 189e203. Ezzati, M., Kammen, D., 2002a. The health impacts of exposure to indoor air pollution from solid fuels in developing countries: knowledges, gaps and data needs. Environ. Health Perspect. 110, 1057e1068. Ezzati, M., Kammen, D., 2002b. Household energy, indoor air pollution and health in developing countries: knowledge base for effective interventions. Annu. Rev. Energy Environ. 27, 233e270. Fischer, S.L., Koshland, C.P., 2007. Daily and peak 1 h indoor air pollution and driving factors in a rural Chinese village. Environ. Sci. Technol. 41, 3121e3126. Gao, X., Yu, Q., Gu, Q., Chen, Y., Ding, K., Zhu, J., Chen, L., 2009. Indoor air pollution from solid biomass fuels combustion in rural agricultural area of Tibet, China. Indoor Air 19, 198. General Administration of Quality Supervision, Inspection and Quarantine (GAQSIQ), 2002. Ministry of Health (MIH), and Ministry of Environmental Protection (MEP) of the People's Republic of China. Indoor air quality standards. GB/T 18883e2002. Gong, W., Zhao, Y., Luan, S., 2014. Characteristics of pm 1.0 and co in rural indoor air of five Chinese provinces. Chin. Sci. Bull. 59, 1553 (In Chinese). Hu, W., Downward, G.S., Reiss, B., Xu, J., Bassig, B.A., Zhang, L., Seow, W.J., Wu, G., Chapman, R.S., Tian, L., Wei, F., Vermeulen, R., Lan, Q., 2014. Personal and indoor PM2.5 exposure from burning solid fuels in vented and unvented stoves in a rural region of China with a high incidence of lung cancer. Environ. Sci. Technol. 48, 8456. Hu, Y., Yang, H., 2000. Study on the prevention of fluorine poisoning from coal burning by stove improvement. Chin. J. Contr. Endem. Dis. 15, 356e357 (In Chinese). Huang, Y., Du, W., Chen, Y., Shen, G., Su, S., Lin, N., Shen, H., Zhu, D., Yuan, C., Duan, Y., Li, B., Liu, J., Tao, S., 2017. Household air pollution and personal inhalation exposure to particles (TSP/PM2.5/PM1.0/PM0.25) in rural Shanxi, North China. Environ. Pollut. 231, 635e643. Kim, K., Jahan, S., Kabir, E., 2011. A review of diseases associated with household air pollution due to the use of biomass fuels. J. Hazard Mater. 192 (2), 425e431. Lewis, J., Pattanayak, S., 2012. Who adopts improved fuels and cookstoves? A systematic review. Environ. Health Perspect. 120, 637e645. Li, Q., Jiang, J., Wang, S., et al., 2017. Impacts of household coal and biomass combustion on indoor and ambient air quality in China: current status and implication. Sci. Total Environ. 576, 347. Lim, S.S., Vos, T., Flaxman, A.D., Danaei, G., Shibuya, K., Adair-Rohani, H., et al., 2012. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the global burden of disease study 2010. Lancet 380, 2224e2260. Lin, N., Chen, Y., Du, W., Shen, G., Zhu, X., Huang, T., Wang, X., Cheng, H., Liu, J., Xue, C., Liu, G., Eddy, Z., Xing, B., Tao, S., 2016. Inhalation exposure and risk of polycyclic aromatic hydrocarbons (PAHs) among the rural population adopting wood gasifier stoves compared to different fuel-stove users. Atmos. Environ. 147, 485e491. Liu, Q., Liu, L., Wang, X.Y., 2011. The effects of daily fuels combustion on indoor air quality in rural area of northern Anhui province. J. Environ. Health 28, 702e705 (In Chinese). Lv, J., Xu, R., Wu, G., Zhang, Q., Li, Y., Wang, P., Liao, C., Liu, J., Jiang, G., Wei, F., 2009. Indoor and outdoor air pollution of polycyclic aromatic hydrocarbons (PAHs) in Xuanwei and Fuyuan, China. J. Environ. Monit. 11, 1368e1374. Lv, J., Zhang, L., Xu, R., Liao, C., Zhu, Y., Jiang, G., Wei, F., 2010. Assessment of polycyclic aromatic hydrocarbons (PAHs) level in xuanwei and fuyuan, Yunnan province. Environ. Monit. Chin. 26, 1e6 (In Chinese). Liu, J., Mauzerall, D., Chen, Q., Zhang, Q., Song, Y., Peng, W., Klimont, Z., Qiu, X., Zhang, S., Hu, M., Lin, W., Smith, K., Zhu, T., 2016. Air pollutant emissions from Chinese households: a major and underappreciated ambient pollution source. Proc. Natl. Acad. Sci. Unit. States Am. 113, 7756e7761. Mehta, S., Shin, H., Burnett, R., North, T., Cohen, A., 2013. Ambient particulate air pollution and acute lower respiratory infections: a systematic review and implications for estimating the global burden of disease. Air Qual. Atmos. Health 6, 69e83. Ministry of Environmental Protection (MEP), 2012. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. Ambient air quality standards. GB 3095e2012. Pan, X., Dong, Z., Jin, X., Wang, B., Wang, L., Xu, X., 2001. Study on assessment for exposure to air pollution in rural areas. J. Environ. Health 18, 323e325 (In Chinese).

637

Pope, D., Bruce, N., Dherani, M., Jagoe, K., Rehfuess, E., 2017. Real-life effectiveness of improved stoves and clean fuels in reducing PM2.5 and CO: systematic review and meta-analysis. Environ. Int. 101, 7e18. Puzzolo, E., Pope, D., Stanistreet, D., Rehfuess, E., Bruce, D., 2016. Clean fuels for resource-poor settings: a systematic review of barriers and enablers to adoption and sustain use. Environ. Res. 146, 218e234. Quansah, R., Semple, S., Qchieng, C., Juvekar, S., Armah, F., Luginaah, I., Emina, J., 2017. Effectiveness of interventions to reduce household air pollution and/or improve health in homes using solid fuel in low-and-meddle income countries: a systematic review and meta-analysis. Environ. Int. 103, 73e90. Rehfuess, E., Puzzolo, E., Stanistreet, D., Pope, D., Bruce, N., 2014. Enablers and barriers to large-scale uptake of improved solid fuel stoves: a systematic review. Environ. Health Perspect. 122, 120e130. Secrest, M.H., Schauer, J.J., Carter, E.M., Lai, A.M., Wang, Y., Shan, M., Yang, X., Zhang, Y., Baumgartner, J., 2016. The oxidative potential of PM2.5 exposures from indoor and outdoor sources in rural China. Sci. Total Environ. 571, 1477e1489. Seow, W.J., Downward, G.S., Wei, H., Rothman, N., Reiss, B., Xu, J., Bassig, B.A., Li, J., He, J., Hosgood, H.D., Wu, G., Chapman, R.S., Tian, L., Wei, F., Caporaso, N.E., Vermeulen, R., Lan, Q., 2016. Indoor concentrations of nitrogen dioxide and sulfur dioxide from burning solid fuels for cooking and heating in Yunnan Province, China. Indoor Air 26, 776e783. Schardt, C., Adams, M., Owens, T., Keitz, S., Fontelo, P., 2007. Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC Med. Inf. Decis. Making 7, 16. Shan, M., Yang, X., Ezzati, M., Chaturvedi, N., Coady, E., Hughes, A., Shi, Y., Yang, M., Zhang, Y., Baumgartner, J., 2014. A feasibility study of the association of exposure to biomass smoke with vascular function, inflammation, and cellular aging. Environ. Res. 135, 165. Shen, G., Tao, S., Wei, S., Zhang, Y., Wang, R., Wang, B., Li, W., Shen, H., Huang, Y., Chen, Y., Chen, H., Yang, Y., Wang, W., Wei, W., Wang, X., Liu, W., Wang, X., Simonich, S.L.M., 2012. Reductions in emissions of carbonaceous particulate matter and polycyclic aromatic hydrocarbons from combustion of biomass pellets in comparison with raw fuel burning. Environ. Sci. Technol. 46, 6409e6416. Shen, G., Zhang, Y., Wei, S., Chen, Y., Yang, C., Lin, P., Xie, H., Xue, M., Wang, X., Tao, S., 2014. Indoor/outdoor pollution level and personal inhalation exposure of polycyclic aromatic hydrocarbons through biomass fuelled cooking. Air Qual. Atmos. Health 7, 449e458. Shen, G., Lin, W., Chen, Y., et al., 2015. Factors influencing the adoption and sustainable use of clean fuels and cookstoves in China -a Chinese literature review. Renew. Sustain. Energy Rev. 51, 741e750. Shen, G., Chen, Y., Du, W., Lin, N., Wang, X., Cheng, H., Liu, J., Xue, C., Liu, G., Xing, B., Tao, S., 2016. Exposure and size distribution of nitrated and oxygenated polycyclic aromatic hydrocarbons among the population using different household fuels. Environ. Pollut. 216, 935e942. Sinton, J.E., Smith, K.R., Peabody, J.W., Yaping, L., Xiliang, Z., Edwards, R., Quan, G., 2004. An assessment of programs to promote improved household stoves in China. Energy Sustain. Dev. 8, 33e52. Smith, K.R., 1993. Fuel combustion, air pollution exposure, and health: the situation in developing countries. Annu. Rev. Energy Environ. 18, 529e566. Smith, K., Samet, J., Romieu, I., Bruce, N., 2000. Indoor air pollution in developing countries and acute lower respiratory infections in children. Thorax 55, 518e532. Wang, S., Wei, W., Li, D., Aunan, K., Hao, J., 2010. Air pollutants in rural homes in Guizhou, China e concentrations, speciation, and size distribution. Atmos. Environ. 44, 4575e4581. Wei, Q., Dong, R., Shi, Z., Hou, J., Xiao, J., 2004. The investigation and analysis of indoor air quality in Sichuan province. Renew. Energy 118, 37e38 (In Chinese). Wei, X., Wei, Q., Wei, X., Xiao, J., Cheng, X., Dong, R., 2006. Influence on indoor air quality by using different type of stove and fuel. Trans. Chin. Soc. Agric. Eng. 22 (In Chinese). World Health Organization (WHO), 2010. WHO Guidelines for Indoor Air Quality: Selected Pollutants. ISBN 978 92 890 0213 4. Wu, F., Wang, W., Man, Y.B., Chan, C.Y., Liu, W., Tao, S., Wong, M.H., 2015. Levels of PM2.5/PM10 and associated metal (loid) s in rural households of Henan Province, China. Sci. Total Environ. 512, 194e200. Xiang, R.B., Song, J.B., Zhu, Y., Guang, W., 2016. Indoor air quality in kitchens in rural China. Environ. Eng. Sci. 33, 699e704. Yu, A., Chan, W., Xian, Y., 1986. The concentration characteristics of nitro Polycyclic aromatic hydrocarbons in particles in households using different fuels in Xuanwei. J. Hyg. Res. 15, 13e15 (In Chinese). Zhang, H., Luo, Z., Nie, J., Wen, T., Ping, B., 2006. The effect of fluoride pollution control in the last 20 years in Longli, Guizhou. Chin. J. Endemiology 25, 713e715 (In Chinese). Zhang, J., Smith, K.R., 2007. Household air pollution from coal and biomass fuels in China: measurements, health impacts, and interventions. Environ. Health Perspect. 2007 (115), 848e855. Zhang, J., Pa, S., Liu, L., Zeng, Y., 2007. Indoor air quality and its adverse health effects in rural area of China. J. Environ. Occup. Med 24, 412e416 (In Chinese). Zhao, Y., Gong, W., Luan, S., 2012. Pollution characteristics and health assessment of particulate mercury in hebei rural households. Environ. Sci. 33 (33), 2960e2966 (In Chinese). Zhong, J., Ding, J., Su, Y., Shen, G., Yang, Y., Wang, C., Simonich, S., Cao, H., Zhu, Y., Tao, S., 2012. Carbonaceous particulate matter air pollution and human

638

W. Du et al. / Environmental Pollution 237 (2018) 625e638

exposure from indoor biomass burning practices. Environ. Eng. Sci. 29, 1038e1045. Zhou, Z., Liu, K., Sun, Y., 2006. Characteristics of elements in PM and sources analysis of poor rural areas of southern Jiangsu province. Res. Environ. Sci. 19, 24e28 (In Chinese). Zhu, C.S., Cao, J.J., Shen, Z.X., Liu, S.X., Zhang, T., Zhao, Z.Z., Xu, H.M., Zhang, E.K., 2012. Indoor and outdoor chemical components of PM2.5 in the rural areas of

Northwestern China. Aerosol Air Qual. Res. 12, 1157e1165. Zhuo, S., Shen, G., Zhu, Y., Du, W., Pan, X., Li, T., Han, Y., Li, B., Liu, J., Cheng, H., Xing, B., Tao, S., 2017. Source-oriented risk assessment of inhalation exposure to ambient polycyclic aromatic hydrocarbons and contributions of non-priority isomers in urban Nanjing, a megacity located in Yangtze River Delta, China. Environ. Pollut. 224, 796e809.