Article pubs.acs.org/est
Pollutant Emissions and Energy Efficiency of Chinese Gasifier Cooking Stoves and Implications for Future Intervention Studies Ellison M. Carter,*,†,‡ Ming Shan,§ Xudong Yang,§ Jiarong Li,§ and Jill Baumgartner‡,∥ †
Department of Civil, Architectural and Environmental and Engineering, University of Texas at Austin, Cockrell School of Engineering, Austin 78712, Texas, United States ‡ Institute on the Environment, University of Minnesota, Minneapolis 55108, Minnesota, United States § Department of Building Science, Tsinghua University, Beijing, China ∥ Institute for Health and Social Policy and Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal QC H3A 0G4, Canada S Supporting Information *
ABSTRACT: Household air pollution from solid fuel combustion is the leading environmental health risk factor globally. In China, almost half of all homes use solid fuel to meet their household energy demands. Gasifier cookstoves offer a potentially affordable, efficient, and low-polluting alternative to current solid fuel combustion technology, but pollutant emissions and energy efficiency performance of this class of stoves are poorly characterized. In this study, four Chinese gasifier cookstoves were evaluated for their pollutant emissions and efficiency using the internationally recognized water boiling test (WBT), version 4.1.2. WBT performance indicators included PM2.5, CO, and CO2 emissions and overall thermal efficiency. Laboratory investigation also included evaluation of pollutant emissions (PM2.5 and CO) under stove operating conditions designed to simulate common Chinese cooking practices. High power average overall thermal efficiencies ranged from 22 to 33%. High power average PM2.5 emissions ranged from 120 to 430 mg/MJ of useful energy, and CO emissions ranged from 1 to 30 g/MJ of useful energy. Compared with several widely disseminated “improved” cookstoves selected from the literature, on average, the four Chinese gasifier cookstoves had lower PM2.5 emissions and higher CO emissions. The recent International Organization for Standardization (ISO) International Workshop Agreement on tiered cookstove ranking was developed to help classify stove performance and identify the bestperforming stoves. The results from this study highlight potential ways to further improve this approach. Medium power stove operation emitted nearly twice as much PM2.5 as was emitted during high power stove operation, and the lighting phase of a cooking event contributed 45% and 34% of total PM2.5 emissions (combined lighting and cooking). Future approaches to laboratory-based testing of advanced cookstoves could improve to include greater differentiation between different modes of stove operation, beyond those evaluated with the WBT.
■
INTRODUCTION AND BACKGROUND
important to successful stove dissemination and adoption. The focus of this study is on the identification of effective stove interventions. Gasifier stoves represent a growing class of household energy interventions that are considered advanced and capable of very clean combustion.7,9 These stoves use a two-stage combustion process, known as microgasification, that converts solid biomass to hot gas and then regulates combustion of those hot gases through controlled provision of a secondary air supply.10 This process should result in both increased efficiency and reduced air pollutant emissions relative to solid fuel combustion in open fires or traditional stoves. Gasifier stoves have been promoted as a potentially more affordable and readily accessible alternative to efficient, clean-burning liquefied petroleum gas
Almost half of all households in China and other developing countries burn biomass and coal in a wide variety of stoves for cooking.1 Solid fuel combustion usually emits high concentrations of gaseous pollutants and air particulates. The resulting household air pollution is the fourth largest contributor to the global burden of disease2 and a potentially important contributor to global and regional climate change.3,4 Despite its importance, few stove intervention programs have successfully reduced household air pollution exposure at a population level. Very low-polluting stoves could achieve climate and health benefits, particularly in light of new epidemiologic evidence suggesting that few benefits accrue for some health outcomes unless the very lowest polluting stoves are used.5,6 Several programs aim to reduce household air pollution by facilitating and encouraging implementation of over 100 million clean and efficient stoves.7,8 In China, the government explicitly promotes the adoption of advanced stoves and clean fuels. There are many components that are © 2014 American Chemical Society
Received: Revised: Accepted: Published: 6461
December 24, 2013 April 30, 2014 May 2, 2014 May 2, 2014 dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
Article
(LPG).11 In China, where improved stove dissemination programs have been successfully pursued in the past,12 Chinese ministries and stakeholders are planning new national stove initiatives that will rely heavily on gasifiers and other advanced combustion stoves,13 many of which are manufactured in China. Relatively little is known about gasifier stove performance and robustness. The few systematic laboratory investigations of gasifier stoves show that they are among the most efficient and least polluting of available “improved” stoves.14−17 However, no studies in China have systematically assessed their performance across different operating conditions commonly used during cooking. Field studies indicate that solid fuel stoves can perform differently under conditions of actual use compared with controlled laboratory testing,14,16 and behavioral and technological explanations for these discrepancies are hypothesized but not well studied. Two potential factors that may partially explain this discrepancy include stove operating conditions,15 such as power level during cooking, and the lighting phase of a cooking event.18 Carefully controlled laboratory tests do not capture the full range of stove operating conditions during cooking.19 For example, stove lighting and operation at different power levels during a single cooking event are two important aspects of cookstove use that are not always given the attention in the stove testing process, even though their measurement is included in recent versions of the WBT and readily achievable in a laboratory setting. In this study, we undertook laboratory investigations of four Chinese-manufactured gasifier cookstoves to assess their energy efficiency and pollutant emission performance, and to identify a high quality stove for future household energy intervention programs. The four stoves were evaluated using internationally recognized stove testing procedures, namely the WBT, version 4.1.2, which was the most up-to-date version at the time of testing. We also evaluated pollutant emissions from these stoves during additional stove operating conditions that were intended to simulate common Chinese cooking practices. This study contributes new knowledge to the field through assessment of gasifier stoves under operating conditions common to daily use in China, a country where nearly 700 million people cook with solid fuel.1
Primary and secondary air supplies are necessary for gasifier stove operation, but each of the stoves selected for this study achieved conditions necessary for gasification through different designs and user controls. For stoves A, B, and C, primary air was forcedly supplied through an opening at the stove base. Primary air supply to stoves A and B could be regulated by sliding a metal plate across this opening, thereby decreasing or increasing the total primary air supply. Users have no control over primary air flow into stove C, which enters through an opening at the stove base. Secondary air was a fan-driven, forced draft for stoves A, B, and C. The fans operated at a single speed, and the secondary air supply could be manually changed from low to high by turning a dial (stoves A and B) or a leverlike handle (stove C), which altered the position of a damper. Manufacturers of stoves B and C provided users with the recommended secondary air dial or lever positions for achieving low, medium, and high power. Stove A is a new stove under development and manufacturer guidelines are in progress, but it is also designed to achieve low, medium, and high power by changing fan settings. The ideal primary and secondary air supply conditions during use of stove D were less transparent. The stove had a single air supply dial on its front, and the settings to which the dial could be turned were preset by the manufacturer. As such, the user does not have separate control over the primary and secondary air supplies. Water Boiling Tests. Water Boiling Test Protocol. We measured each stove’s efficiency and pollutant emissions using the modified Water Boiling Test version 4.1.2 at the Beijing University of Chemical Technology Combustion Lab in July, 2012 and April, 2013. The experimental setup for these tests is provided in the SI (Figure S1), and the WBT protocol, which has since been updated to version 4.2.2, is described in detail elsewhere.19 Briefly, this test measures the energy transfer and pollutant emissions associated with bringing a pot of water to boil and maintaining it simmering under controlled conditions. A complete WBT includes emissions and energy efficiency measurements during three phases: (1) high power, cold start; (2) high power, hot start; and (3) low power, simmer for 45 min. Results for cold start and hot start phases are averaged when a single value representative of high power performance is desired, as is the case for the new International Organization for Standardization (ISO) International Workshop Agreement (IWA) tiered stove rating framework,20 which is discussed in greater detail below. We excluded the simmer phase of the WBT because Chinese flash fry cooking does not typically include a low power phase, and we aimed to identify high quality stoves for future intervention studies in China. Trained laboratory staff ignited and operated the stoves in accordance with manufacturer recommendations. The fuel chamber is completely filled with pelletized biomass fuel prior to the start of a given cooking event. Each combustion event is initiated during the lighting phase using a small, measured amount of dry, red pine hardwood starter fuel. We used 20−40 g of starter fuel (dry hardwood) to initially start a fire in the stove, with the exact starter fuel amount varying by stove type. A portion of the starter fuel was placed directly on the pellet fuel inside the fuel canister, and a gas lighter was used to light the starter fuel. For stoves without primary air control (stoves C and D), we assumed that the primary air was supplied to its greatest extent during the lighting phase. For stoves A and B, the primary air supply was adjusted to supply the maximum amount of air possible by opening the damper to its fullest extent without stifling the flame.
■
MATERIALS AND METHODS Cookstove and Fuel Characteristics. In preparation for a future intervention study in rural China, four gasifier stoves were selected from a number of available gasifier cookstoves and evaluated because they are manufactured and commercially available in China and have not been previously studied. We also selected these stoves because they were designed to burn processed (pelletized) biomass fuel and were capable of forced draft gasification, although they vary in their dimensions and the optimal loading mass of fuel (Supporting Information (SI) Table S1). Pelletized fuel is common for gasifier cooking stoves, and the biomass-based fuel used in this investigation was produced from processed pine tree waste (SI Table S2). Forced draft gasification refers to fan-driven supply of air to the stove combustion chamber. All four gasifier stoves were top-loading, batch-fed stoves. The selected stoves were designed for use with traditional Chinese stove design, which incorporates the chimney into the wall adjacent to the stove enclosure as opposed to directly attached to the stove. Although stoves C and D are also equipped to use their own chimney, we operated all stoves without chimneys. 6462
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
Article
Once a flame was started, additional starter fuel was gradually added until the flame was stable. Any remaining dry wood starter fuel was weighed and subtracted from the original amount. We then adjusted the secondary air supply to reflect the manufacturer guidelines for high power cooking and initiated the cooking event by placing a pot of water on the stove. We used two flat-bottomed aluminum pots for stove testing because the diameter of the opening for pot placement varied among our study stoves. A smaller pot was used with stoves A and B (500g) compared with stoves C and D (675g), but both sizes could hold at least 5 L of water. To consistently boil 5 L of water for each test, the mass of the dry pot was recorded before each WBT phase, and 5000 g of water was then added to the pot. The time between the end of the high power cold start phase and the beginning of the high power hot start phase was controlled by the time it took to separate spent/ burned fuel from unburned fuel for weighing. We quickly separated the fuel (10−15 min) so as to minimize the time between the cold and hot start phases but maintain appropriate safety of staff. Start time, end time, mass of water before and after, mass of fuel before and after, and mass of char remaining after were recorded for each WBT phase. We recorded pollutant emissions starting from the 5−10 min prior to stove lighting and ending once the boiling point of water was reached. The Portable Emissions Monitoring System designed by the Aprovecho Research Center (Cottage Grove, Oregon) was used to collect and measure cookstove emissions. Carbon monoxide (CO) was measured using an electrochemical cell (CO-AF; Alphasense Ltd.; United Kingdom) with a resolution of 1 ppm and a 30 s response time. Carbon dioxide (CO2) was measured using a nondispersive infrared analyzer (CO2S-PPM; SSTSensing; United Kingdom) with a resolution of 2 ppm and a 2 min response time. Particulate matter (PM) measurements were made using a red laser scattering photometer (Aprovecho Research Center; Cottage Grove, Oregon) with a resolution of 15 μg/m3 and a 1 s response time. All sensors were calibrated prior to conducting measurements. We also measured integrated gravimetric fine PM (aerodynamic diameter ≤2.5 μm; PM2.5) for each stove during the WBT. A PTFE membrane filter (47 mm, 2 μm pore size; HI-Q Environmental Products; San Diego, CA) was placed over a filter screen inside the filter cassette (HI-Q Environmental Products; LPH-102; San Diego, CA) that was positioned downstream of a PM2.5 cyclone (1783; URG Corporation; Chapel Hill, NC). Filters were conditioned in a humiditycontrolled environment for a 24-h period and then weighed preand postsampling using a microbalance with a 10 μg readability (AL204-IC; Mettler Toledo; Columbus, OH). Relative humidity levels did not exceed 40% in the controlled environment. The gravimetric measurement reflected PM2.5 emitted during both the cold start and hot start high power phases. While the real-time PM measurements made with the Aprovecho Testing System were representative of total suspended particles (rather than PM2.5, specifically) due to the lack of a particle size cut stage in the sampling train upstream from the PM photometer, it has been shown that particle size distributions of aerosol emissions from biomass combustion are unimodal with peaks in the range of 0.260.38 μm21 and 0.10.2 μm.22 Therefore, real-time trends in PM measurements collected with this system can be expected to approximate real-time trends in PM2.5. Analysis of Water Boiling Test Results. The key performance indicators evaluated for each stove address pollutant
generation, energy efficiency, and fuel use. Pollutant generation was evaluated by focusing on the following four indicators: • PM2.5: mass emitted per megajoule (MJ) delivered to the cooking pot • carbon monoxide (CO): mass emitted per MJ delivered to the cooking pot • PM2.5 emission rate • CO emission rate Energy efficiency and fuel use were evaluated based on four performance indicators: • overall thermal efficiency: the ratio of the energy delivered to the cooking pot to the total net energy in the fuel • specific fuel consumption: the fuel mass consumed during the WBT per volume of water remaining in the pot • cooking power: useful energy delivered to the cooking pot per time calculated by multiplying the stove’s firepower (fuel energy used by stove per time) by its overall thermal efficiency • time-to-boil: elapsed time from the start of the cooking event to when water reaches its local boiling point We made these calculations for both the cold start and hot start high power phases of the WBT, in accordance with the WBT protocol.19 Cookstove Performance and Rating. We compared performance of the four Chinese gasifier stoves evaluated in the present study with several stoves well characterized by laboratory testing by Jetter et al.17 The Phillips HD4012 and Sampada stoves were selected for comparison because they represent the range of gasifier cookstove performance. We also included the StoveTec TLUD (top-loading updraft) stove, the StoveTec Greenfire rocket stove, and the three-stone fire because a TLUD stove has a similar design to gasifier stoves and uses pelletized fuel, rocket stoves are widely disseminated improved cooking stoves,15 and the three-stone fire is representative of traditional open fire cooking for much of the world, including parts of China. Although Jetter et al.17 conducted WBTs with both low-moisture and high-moisture fuel, only stove performance data for dry fuel conditions were considered for comparison. With the exception of the StoveTec TLUD, which is a natural draft stove that burns pellet fuel, stoves selected for comparison burn wood-based, nonpelletized fuel, compared to pelletized fuel burned by the Chinese gasifier stoves. Differences in test performance, if observed, may be attributable to differences in fuel type, as well as differences in stove type.15,17 A primary goal of this study was to identify an appropriate stove for a future intervention study that will be aimed at evaluating the potential health benefits associated with the intervention. With this in mind, we focused on comparing the following four stove performance indicators: PM2.5 emissions per MJ of useful energy (defined as the energy that is delivered to the cooking pot), CO emissions per MJ of useful energy, PM2.5 indoor emission rate, and CO indoor emission rate. These four indicators of performance were selected because they address pollutant total and indoor emissions. In particular, the indoor emissions rates are most relevant to health because they are based on the World Health Organization guideline concentration values for those two pollutants.20 Overall, this stove rating system considers four categories of stove performance: fuel use, emissions, indoor emissions, and safety. 6463
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
Article
Within all categories except safety, several performance indicators are divided among high power and low power subcategories. A subtier score on a scale of 0−4, with 0 being the worst score and 4 being the best score, is assigned to prescribed ranges of performance for each indicator. Then, an overall score is assigned to the category based on the lowest subtier assigned. For example, a stove performing in Tier 1 with respect to high power CO emissions and Tier 3 with respect to high power PM2.5 emissions would be assigned Tier 1 for the emissions category. The performance indicator value range associated with the appropriate tier can be found elsewhere.20 Focusing on emissions, we ranked the stoves according to their lowest IWA emissions subtier ratings for high power performance. Controlled Stove Operating Conditions. We devised two testing protocols to simulate the stove operating conditions known to be common among cooks in rural China and elsewhere. We chose to carry out these tests with stove B because manufacturer stove directions facilitated our ability to reproducibly achieve medium and high power stove operation. Each additional stove operating condition was designed to be a discrete event that included the stove lighting phase, rather than a subset of 10 min selected from the high power phase of a water boiling test. Also, the small size of stove B made it more manageable to accurately and safely perform the desired tests. High power stove operation for 10 min simulated meat-searing or pan-frying of a sturdy vegetable (e. g., carrot or potato). Medium power stove operation for 10 min simulated pan-frying a mixture of both sturdy and delicate (e. g., leafy greens) vegetables together. Real-time measurements of pollutant concentrations (CO and PM2.5) were obtained in triplicate during both the high and medium power stove operation. We also measured real-time measurements of CO and PM2.5 concentrations during the lighting phase of each high power and medium power test so that emissions associated with the stove lighting could be compared relative to those associated with cooking. Real-time PM2.5 measurements were collected using side-by-side measurements with the Aprovecho Testing System’s red laser scattering photometer mentioned above and a DustTrak monitor (8520; TSI Inc.; Minneapolis, MN). Because colocated gravimetric samples were not possible at the time of stove testing, we collected side-by-side gravimetric PM2.5 samples and DustTrak measurements at a different time, using the same DustTrak monitor that was used in this study. The regression of gravimetric and DustTrak PM2.5 concentrations provided in the SI (Figure S2) was used to correct the DustTrak PM2.5 measurements. Using this approach, it is assumed that optical properties of particulate matter do not vary, which may lead to uncertainties in quantitative estimates of PM2.5 emitted. We measured CO using the same instrument described above.
Figure 1. PM2.5 emissions per MJ of useful energy versus CO emissions per MJ of useful energy for stoves A, B, C, and D (cold start, CS: open circles; hot start, HS: closed circles) compared with two gasifier stoves (Phillips, Sampada), a top-loading updraft stove (StoveTec TLUD), a rocket stove (StoveTec), and a minimally tended 3-stone fire tested by Jetter et al. (2012) (cold start: open squares; hot start: closed squares). Points represent mean values and error bars represent standard deviation.
hot start phase than the cold start phase for stove B (170 ± 69 mg/MJ), stove C (549 ± 135 mg/MJ), and stove D (436 ± 441 mg/MJ), while they were lower for stove A (124 ± 27 mg/ MJ). PM2.5 emissions from stoves A and B were not lower than those of the Phillips stove (cold start: 70 ± 10 mg/MJ; hot start: 50 ± 20 mg/MJ). With the exception of the cold start for stove B, PM2.5 emissions from stoves A and B were also not lower than those from the StoveTec TLUD stove (cold start: 90 ± 30 mg/MJ; hot start: 90 ± 30 mg/MJ), but they were lower than those of the StoveTec rocket stove. Stoves C and D also generated PM2.5 emissions lower than the StoveTec rocket stove during the cold start phase (380 ± 40 mg/MJ), but during the hot start phase, PM2.5 emissions for stoves C and D were slightly higher than the StoveTec rocket stove (370 ± 60 mg/MJ). All four Chinese gasifier stoves considered in this study performed much better than the three-stone fire (cold start: 1380 ± 430 mg/MJ; hot start: 1320 ± 240 mg/MJ). Mean CO emissions per MJ of useful energy during the cold start phase were lowest for stove A (1.5 ± 0.3 g/MJ) among the four Chinese gasifier stoves tested in this study. For the other three stoves, mean CO emissions per MJ of useful energy during the cold start phase were 9.3 ± 6.2 g/MJ, 21.2 ± 4.4 g/ MJ, and 16.0 ± 9.8 g/MJ for stoves B, C, and D, respectively. For stoves B and D, mean CO emissions more than doubled during the hot start phase relative to the cold start phase to 20.4 ± 2.1 g/MJ and 45.0 ± 33.8 g/MJ, respectively, whereas for stoves A and C, mean CO emissions fell slightly to 1.1 ± 0.3 g/ MJ and 18.6 ± 6.1 g/MJ, respectively. Stoves A and B performed within the range of performance of the two gasifier stoves and the TLUD stove selected from the literature for comparison. For example, during the cold start phase, the Phillips stove emitted 1.6 ± 0.1 g/MJ, and the Sampada stove emitted 8.3 ± 1.0 g/MJ, whereas the StoveTec TLUD stove emitted 5.6 ± 0.8 g/MJ. CO emissions per MJ from stoves C and D were greater than CO emissions from the three “improved” stoves selected from the literature and similar to CO emissions from the three-stone fire (cold start: 16.4 ± 1.3 g/MJ; hot start: 13.8 ± 0.6 g/MJ).
■
RESULTS AND DISCUSSION Chinese Gasifier Stove Performance. Pollutant Generation. The four gasifier cookstoves varied with respect to PM2.5 and CO emissions (Figure 1). Among the four stoves, mean (±SD) PM2.5 emissions per megajoule (MJ) of useful energy during the cold start phase were lowest for stove B (71 ± 98 mg/MJ). For the other three stoves, mean PM2.5 emissions per MJ of useful energy during the cold start phase were, in increasing order 188 ± 69 mg/MJ for stove A, 225 ± 247 mg/MJ for stove D, and 311 ± 222 mg/MJ for stove C. PM2.5 emissions per MJ of useful energy were higher during the 6464
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
Article
Figure 2. Overall thermal efficiency for stoves A, B, C, and D compared with two gasifier stoves (Phillips, Sampada), a rocket stove (StoveTec), and a three-stone fire tested by Jetter et al. (2012). Box plots represent mean values and standard errors. Error bars extend to 1.5 times the standard error.
Energy Efficiency and Fuel Use. The overall thermal efficiencies (OTEs) varied for the four Chinese gasifier stoves (Figure 2). Mean OTEs for stoves A and B were 27.5 ± 1.7% and 33.3 ± 0.8%, respectively, during the cold start phase of the WBT and 26.4 ± 1.5% and 32.9 ± 0.6%, respectively, during the hot start phase. The OTE performance of these two stoves was slightly lower than the Phillips (cold start: 36.2 ± 0.2%; hot start: 40.5 ± 0.5%) and StoveTec stoves (cold start: 35.4 ± 2.0%; hot start: 33.1 ± 2.1%), but similar to the Sampada (cold start: 26.7 ± 0.2%; hot start: 28.0 ± 1.5%) and higher than the three-stone fire (cold start: 13.7 ± 0.5%; hot start: 13.6 ± 0.3%). Mean OTEs for stoves C and D were 17.9 ± 4.8% and 23.7 ± 1.6%, respectively, during the cold start phase and 26.1 ± 1.5% and 20.6 ± 2.5%, respectively, during the hot start. Stove A boiled water the fastest (cold start: 9.7 ± 0.8 min; hot start: 9.8 ± 0.8 min) of the four Chinese gasifier stoves, while stove D was the slowest (cold start: 17.3 ± 1.7 min; hot start: 15.9 ± 1.6 min). The complete WBT results for specific fuel consumption, cooking power, and time-to-boil are available for each stove in the SI (Table S3). Cookstove Performance and Rating. Applying the ISO IWA tiered stove rating for emissions only to the four Chinese gasifier stoves considered in this study, stove A ranked the highest, followed by stoves B, C, and D (Table 1). Relative to
stoves selected from the literature for comparison, stove A ranked just below the Phillips and the StoveTec TLUD stoves on account of its Tier 2 indoor emissions rate. The results for stove B and stove D highlight the need to closely inspect subtier scores within the IWA rating system, which currently assigns a single tier value for total emissions and a single tier value for indoor emissions rates based on the lowest tier achieved for CO and PM2.5. Overall, stove B is ranked lower than the StoveTec rocket stove because of its Tier 1 CO total emissions and Tier 0 CO emission rate. However, PM2.5 total emissions and the indoor PM2.5 emission rate from stove B place it within Tier 3 and Tier 1, respectively, while the StoveTec rocket stove (Tier 2) emits twice as much PM2.5 as stove B and is rated Tier 0 with respect to indoor PM2.5 emission rate. The IWA tiered cookstove rating system is meant to serve as a conservative tool for stove program implementers and policy-makers, but under the current system, the potential for gasifier cookstoves to reduce emissions of a very important air pollutant, PM2.5, is undervalued. Overall, our results illustrate a range of possible performance from gasifier stoves. Although it does not yet appear that gasifier stoves can achieve PM2.5 emissions competitive with natural gas (0.004 ± 0.002) or LPG (0.025 ± 0.025 g/MJ),23 some gasifier cookstoves did achieve Tier 3 performance ratings with respect to total emissions. However, the current IWA stove rating system is not yet able to reward these important pollutant reductions when the CO emissions are not reduced to a similar degree. Alternatively, our results suggest that three of the four Chinese gasifier stoves tested could be redesigned with specific attention to reducing CO emissions. Furthermore, Chinese gasifier stoves that are equipped with a chimney are likely to achieve lower indoor emissions. Careful, iterative performance testing is vital to assessing the quality of emerging gasifier cookstoves before widely promoting or disseminating them to households. Based on our evaluation of cookstove rankings, it is very important for those who would use the IWA framework as a decision-making tool to closely evaluate subtiers, which are a standard part of IWA performance reporting. Controlled Stove Operating Conditions. Medium power, controlled operation of stove B had a higher PM2.5 emission rate than high power operation [medium power mean ± SE: 1.14 ± 0.16 mg/min; high power: 0.48 ± 0.061 mg/min; p = 0.026]. PM2.5 mass emissions during the cooking phase for
Table 1. ISO International Workshop Agreement Tiered Stove Rating for Four Chinese Gasifier Cookstoves Compared with Two Gasifier Stoves, a TLUD Stove, A Rocket Stove, And a Three-Stone Fire
stove name Phillips StoveTec Stove A StoveTec Stove B Sampada Stove D Stove C 3 stone fire
stove type gasifier TLUD gasifier rocket gasifier gasifier gasifier gasifier open fire
PM2.5 emission per useful energy [mg/MJ]
PM2.5 emission rate [mg/min]
CO emission per useful energy [g/ MJ]
CO emission rate [g/min]
3 3 3 2 3 1 2 1 0
3 3 2 0 1 0 0 0 0
4 4 4 4 1 3 0 0 1
4 4 4 2 0 1 0 0 0
6465
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
Article
medium power stove operation were approximately twice as high as PM2.5 mass emissions for high power stove operation [medium power: 11.55 ± 1.66 mg; high power: 5.38 ± 0.61 mg; p = 0.009] (Figure 3). At lower power, combustion
Figure 4. Real-time PM measurements during lighting and cooking phases of the water boiling tests (WBT) using stove B (n = 3).
total emissions and potentially identify techniques that could reduce the impact of lighting. Experimental designs such as these would provide valuable insight into inform stove selection for future intervention studies without requiring a potentially larger investment of resources to conduct additional tests like the controlled cooking tests24 or kitchen performance tests25 in the field. Evaluating stove performance over a wider range of real-world stove operating conditions during the laboratory stage of stove testing also might resolve some of the discrepancies observed between laboratory and field stove testing. As cookstove implementation programs grow to meet global clean cooking targets, cookstove testing should continue to be performed across a range of different operating modes deemed necessary by such programs to accurately identify cookstoves capable of meeting those targets and safeguarding human health. The conclusions drawn from the present study and others26,27,28 indicate that laboratory cookstove performance testing could improve, and potentially match cookstove evaluation in the field more closely, if pollutant emissions that occur during the lighting phase of a cooking event or test were further scrutinized. Although laboratory testing cannot completely substitute for field testing, to close the gap between laboratory and field testing results, different modes of operation for laboratory cookstove performance tests could be employed to better reflect local and regional cooking practices. This may become increasingly feasible as regional cookstove testing centers expand their role in decision-making processes that surround cookstove development and dissemination.
Figure 3. PM2.5 mass emitted during lighting (open bars) and cooking (shaded bars) for medium power (n = 5) and high power (n = 4) controlled stove operating conditions with stove B. Error bars represent ± SD.
efficiency might be less than at higher power and could potentially explain the increased PM2.5 mass emissions. We also found that for medium and high power stove operation, the lighting phase contributed 44.9 ± 5.9% and 33.9 ± 3.1%, respectively, to total PM2.5 mass emitted during the lighting and cooking phases combined. This fractional contribution should not be overlooked, particularly since the cook is present and therefore directly exposed during lighting.18 Since this study was conducted, the WBT protocol has been updated (version 4.2.2),19 and the instructions more explicitly include lighting in the procedure. Future studies could specifically evaluate the lighting phase of cooking events, taking into consideration nonstove factors, such as lighting material and user technique, in addition to differences between stove and fuel types. For three replicate WBTs with stove B, we observed that the PM2.5 peaks during the lighting phase are as high or higher than those during the WBT (Figure 4). Although total PM2.5 mass emissions cannot be derived from these data because side-byside gravimetric PM2.5 measurement was not collected during the lighting phase, the relative relationship between the peaks observed during the lighting phase and those observed throughout the cooking phase reveals a similar, non-negligible contribution made by the lighting phase to PM2.5 emissions during a cooking event or test. The difference in PM2.5 emissions between high power and medium power controlled stove operation and the contribution from lighting to overall PM2.5 emissions during a cooking event suggest that, prior to large-scale dissemination, cookstove implementation decision-makers could benefit from expanding lab-based stove testing. While the WBT already incorporates two power levels, it would be valuable to evaluate and differentiate additional modes of operation that might be specific to a region or nation and that are not currently addressed through the WBT. Lab-based stove testing could also evaluate the contribution of emissions from lighting relative to
■
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org/.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: (612) 625-7693; fax: (612) 626-5555; e-mail:
[email protected]. Notes
The authors declare no competing financial interest. 6466
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
Environmental Science & Technology
■
Article
(14) Roden, C. A.; Bond, T. C.; Conway, S.; Osorto Pinel, A. B.; MacCarty, N.; Still, D. Laboratory and field investigations of particulate and carbon monoxide emissions from traditional and improved cookstoves. Atmos. Environ. 2009, 43 (6), 1170−1181. (15) Jetter, J. J.; Kariher, P. Solid-fuel household cook stoves: Characterization of performance and emissions. Biomass Bioenergy 2009, 33 (2), 294−305. (16) MacCarty, N.; Still, D.; Ogle, D. Fuel use and emissions performance of fifty cooking stoves in the laboratory and related benchmarks of performance. Energy Sustainable Dev. 2010, 14 (3), 161−171. (17) Jetter, J.; Zhao, Y.; Smith, K. R.; Khan, B.; Yelverton, T.; DeCarlo, P.; Hays, M. D. Pollutant emissions and energy efficiency under controlled conditions for household biomass cookstoves and implications for indicators useful in setting international test standards. Environ. Sci. Technol. 2012, 46 (19), 10827−10834. (18) Roden, C. A.; Bond, T. C.; Conway, S.; Pinel, A. B. O. Emission factors and real-time optical properties of particles emitted from traditional wood burning cookstoves. Environ. Sci. Technol. 2006, 40 (21), 6750−6757. (19) Water Boiling Test (WBT) Global Alliance for Clean Cookstoves 2013. Water Boiling Test (WBT) Version 4.2.2. http:// www.cleancookstoves.org/our-work/standards-and-testing/learnabout-testing-protocols/. (20) International Workshop Agreement 11. Guidelines for Evaluating Cookstove Performance; International Organization for Standardization: Geneva, Switzerland, 2012. (21) Li, X.; Wang, S.; Duan, L.; Hao, J.; Nie, Y. Carbonaceous aerosol emissions from household biofuel combustion in China. Environ. Sci. Technol. 2009, 43 (15), 6076−6081. (22) Kleeman, M. J.; Schauer, J. J.; Cass, G. R. Size and composition distribution of fine particulate matter emitted from wood burning, meat charbroiling, and cigarettes. Environ. Sci. Technol. 1999, 33 (20), 3516−3523. (23) Zhang, J.; Smith, K. R.; Ma, Y.; Ye, S.; Jiang, F.; Qi, W.; Thorneloe, S. A. Greenhouse gases and other airborne pollutants from household stoves in China: A database for emission factors. Atmos. Environ. 2000, 34 (26), 4537−4549. (24) Controlled Cooking Test. Global Alliance for Clean Cookstoves. 2010. Controlled Cooking Test (CCT) Version 2.0. Washington, DC. http://www.cleancookstoves.org/our-work/ standards-and-testing/learn-about-testing-protocols/. (25) Kitchen Performance Test. Global Alliance for Clean Cookstoves. 2010. Kitchen Performance Test (KPT) Version 3.0. Washington, DC. http://www.cleancookstoves.org/our-work/ standards-and-testing/learn-about-testing-protocols/.
ACKNOWLEDGMENTS E.C. was supported by National Science Foundation (NSF) grant number DGE 1210108. We kindly acknowledge the facilities support from researchers at the Biomass Energy and Environmental Engineering Research Center at the Beijing University of Chemical Technology. The Tsinghua team acknowledges the financial support from the National Natural Science Foundation of China (Project Number 51038005) and Tsinghua University Internal Research Project Support (Project Number 20111081081). The manuscript’s contents are solely the responsibility of the grantee and do not necessarily represent the official views of the NSF.
■
REFERENCES
(1) Bonjour, S.; Adair-Rohani, H.; Wolf, J.; Bruce, N.; Mehta, S.; Prüss-Ustün, A.; Smith, K. R. Solid fuel use for household cooking: Country and regional estimates for 1980−2010. Environ. Health Perspect. 2013, 121 (7), 784−790. (2) Lim, S. S.; Vos, T.; Flaxman, A. D.; Danaei, G.; Shibuya, K.; Adair-Rohani, H.; Degenhardt, L. 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 2012, 380 (9859), 2224−2260. (3) Wilkinson, P.; Smith, K. R.; Davies, M.; Adair, H.; Armstrong, B. G.; Barrett, M.; Chalabi, Z. Public health benefits of strategies to reduce greenhouse-gas emissions: Household energy. Lancet 2009, 374 (9705), 1917−1929. (4) Bond, T. C.; Doherty, S. J.; Fahey, D. W.; Forster, P. M.; Berntsen, T.; et al. Bounding the role of black carbon in the climate system: A scientific assessment. J. Geophys. Res. 2013, 118 (11), 5380− 5552. (5) Baumgartner, J.; Schauer, J. J.; Ezzati, M.; Lu, L.; Cheng, C.; Patz, J. A.; Bautista, L. E. Indoor air pollution and blood pressure in adult women living in rural China. Environ. Health Perspect. 2011, 119 (10), 1390−1395. (6) Smith, K. R.; McCracken, J. P.; Weber, M. W.; Hubbard, A.; Jenny, A.; Thompson, L. M.; Bruce, N. Effect of reduction in household air pollution on childhood pneumonia in Guatemala (RESPIRE): A randomised controlled trial. Lancet 2011, 378 (9804), 1717−1726. (7) Venkataraman, C.; Sagar, A. D.; Habib, G.; Lam, N.; Smith, K. R. The Indian national initiative for advanced biomass cookstoves: The benefits of clean combustion. Energy Sustainable Dev. 2010, 14 (2), 63−72. (8) GACC (Global Alliance for Clean Cookstoves). Igniting Change: A Strategy for Universal Adoption of Clean Cookstoves and Fuels; GACC: Washington, DC, 2011. (9) Zhang, J. J.; Smith, K. R. Household air pollution from coal and biomass fuels in China: Measurements, health impacts, and interventions. Environ. Health Perspect. 2007, 115 (6), 848−855. (10) Roth, C. Micro Gasification: Cooking with Gas from Biomass, 2nd ed.; GIZ-HERA Poverty-oriented basic energy services. Federal Ministry for Economic Cooperation and Development, Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH: Germany, 2014. (11) Bruce, N. G., Rehfuess, E. A., Smith, K. R. Household energy solutions in developing countries. In Encyclopedia of Environmental Health; Nriagu, J. O., Ed.; Elsevier, Burlington, 2011; Vol. 3, p 6275. (12) CERS (China Energy Research Society) and CAREI (China Association of Rural Energy Industry). Rural Energy Development in China 1979−1999; China Energy Research Society and China Association of Rural Energy Industry: Beijing, China, 2000. (13) Smith, K. R.; Keyun, D. A Chinese national improved stove program for the 21st century to promote rural social and economic development. English translation of article in Chinese in. Energy Policy Res. 2010, 1, 24−25. 6467
dx.doi.org/10.1021/es405723w | Environ. Sci. Technol. 2014, 48, 6461−6467
