Emission Characteristics and Concentrations of Gaseous Pollutants in Environmental Moxa Smoke

The burning of moxa floss in moxibustion constitutes a major anthropogenic source of many gaseous pollutants, which has been associated with many different negative environmental health effects. The aim of the present study is to systemically study the concentration of gaseous pollutants emitted from different types of moxa floss combustion and present key information in abbreviated tabular form to assist in the assessment of air quality in moxibustion clinics and contribute to the safety evaluation of moxibustion. Sampling was divided into pre-combustion, combustion and postcombustion phases. The pollutants determined were carbon monoxide (CO), carbon dioxide (CO2), sulphur dioxide (SO2) and nitrogen dioxide (NO2) produced by burning three types of moxa floss samples. The average post-combustion concentrations for CO, CO2 and NO2 in moxibustion clinics were 9.333 ppm, 0.138% and 10.556 μg m, respectively. SO2 was below detectable limit. NO2 concentration decreased during post-combustion, possibly as a result of reactions from moxa floss combustion. The levels of target gaseous pollutants from 4 g of moxa floss combustion were not produced in quantities that exceeded present international air quality standards and occupational exposure limits. Data from our study is important for the recognition and control of occupational and non-occupational gaseous exposure and for the assessment of air quality in moxibustion clinics by professional authorities.


INTRODUCTION
Moxibustion is a therapeutic procedure using ignited moxa floss to apply heat to certain points or areas of the body surface for treating disease (World Health Organization, 2007). Smoke from the burning of moxa floss products, or environmental moxa smoke (EMS), has raised the possibility of its association with possible health problems and sensitization due to occupational and non-occupational exposure among practitioners, workers and patients in moxibustion clinics.
It has been recognized that combustion emissions constitute a complex mixture of pollution components and combustion processes are major anthropogenic contributors of many classical air pollutants, for example, carbon monoxide (CO), nitrogen oxides (NO x ), sulfur dioxide (SO 2 ) and particulate matter (PM), which have been associated with many different negative environmental health effects. There is also mounting evidence from epidemiologic studies regarding the associations between SO 2 , NO 2 and CO exposure and increases in cardiopulmonary hospital admissions and mortality, and cardiovascular emergency admissions caused by stroke (NO 2 ) and myocardial infarction (NO 2 and CO) (Tarlo et al., 2001;Koken et al., 2003;Tsai et al., 2003).
Given the lack of information regarding the concentration of gaseous pollutants in moxa smoke emitted from moxa floss combustion, the present study is an extension of our previous work on the PM 10 mass concentration and oxidative capacity of moxa smoke. (Huang et al., 2015) The purpose of this study was to systemically study the concentration of gaseous pollutants in EMS released from burning different types of moxa floss. The pollutants determined were carbon monoxide (CO), carbon dioxide (CO 2 ), sulphur dioxide (SO 2 ) and nitrogen dioxide (NO 2 ). Our study aims to present key information and data of gaseous pollutants in moxibustion clinics in abbreviated tabular form to provide a basis for the assessment of air quality in moxibustion clinics and contribute to the safety evaluation of moxibustion. We decided to compare the emission tests results in our study against four air quality standards so as to properly analyze the significance of the results. The four standards chosen were the European Union directive on ambient air quality, the National Ambient Air Quality Standards (NAAQS) established by the United States Environmental Protection Agency, the World Health Organization air quality guidelines for gaseous pollutants and China's Ambient Air Quality Standards (CAAQS). Comparisons were also made against existing occupational exposure guidelines such as the US National Institute for Occupational Safety and Health (NIOSH), the US Occupational Health and Safety Administration's (OSHA) for permissible exposure limits (PEL) and the Standard 1910.1000 table Z-1 'Limits for Air Contaminants'.

Experimental Material
Moxa floss of different storage years and ratios of the species Artemisia argyi gathered from Tongbai mountain in Henan province were obtained from a moxa floss production factory in Henan, Nanyang, China. Storage year refers to the number of years the dried mugwort leaves were stored before processing into moxa floss. Ratio refers to the weight of the starting material (dried mugwort leaves) to the weight of the finished product (moxa floss) in kilogram after processing, for example the ratio 3:1 means 3 kg of dried mugwort leaves was processed into 1 kg of moxa floss. Three moxa floss samples were selected in this study, specifically 3 year and 3:1 ratio (sample A), 10 year and 3:1 ratio (sample B) and 3 year and 15:1 ratio (sample C).

Sampling Procedure of CO, CO 2 , NO 2 and SO 2
Sampling was carried out in three empty simulation clinics located on the third floor in a building of Beijing University of Chinese Medicine in 2013. One simulation clinic was used as the sampling site for each moxa floss sample. For each simulation clinic, the walls were painted with emulsion paint, had cement flooring and there was no recent history of renovation. The doors and windows were usually kept closed. The floor area and volume of each simulation clinic was approximately 14.8 m 2 and 50.32 m 3 . The university was erected in 1956 and is located in a densely populated residential area on a side road about 10 m away from a busy road. No specific industrial sources for emissions are known of in this area. During the measurements, information on temperature, relative humidity and atmospheric pressure was recorded.
Before each emission test, the doors and windows of the simulation clinic were tightly sealed by a researcher. Sampling was divided into three phases: 1. Pre-combustion phase: the background concentrations of the target gaseous pollutants in the clinic were determined for 1 h prior to each moxa floss sample combustion. 2. Combustion phase: 4 g of moxa floss sample was ignited and each sample took approximately 20 to 25 minutes to burn out to reflect a typical duration of moxibustion administration in a normal treatment setting. 3. Post-combustion phase: the concentrations of the target gaseous pollutants were monitored in the clinic for another hour. Moxa floss samples were burned on a stainless steel tray located in the middle of the room at a height of approximately 1 m, similar to the sampling devices in all three simulation clinics. Each experiment was repeated thrice to obtain an average. The actual average concentrations of the target gaseous pollutants emitted from moxa floss combustion were found by deduction of the background concentrations of the respective gaseous pollutant. The same researcher performed all the aforementioned sampling and emission tests.
A portable non-dispersive infrared CO/CO 2 gas analyzer (HAD-GXH-3010/3011AE, Beijing Heng Odd Instrument Co. Ltd) was used to measure the CO and CO 2 concentrations and sampling points were set at 15-minute intervals for both the pre-and post-combustion phases to monitor the variations in CO and CO 2 concentrations . Gaseous specific chemiadsorption as NO 2 − , SO 4 2− , and SO 3 2− on a triethanolamine (TEA) coated microporous polyethylene tube (5.8 mm diameter, 60 mm height) allowed for NO 2 and SO 2 sampling using an air sampler (QT-2A, Beijing Pu Sheng Yang Trade Co. Ltd.) at a rate of 0.5 L min -1 . The sampling devices were located in the center of the clinic at a height of 1.2 m. The minimum detectable value using this method for SO 2 was 0.007 mg m -3 . The detection methods for the gaseous pollutants were based on the Chinese national ambient air quality standard.

CO Concentration
The background concentration of CO in simulation clinics B and C was relatively stable, but there was obvious fluctuation in the background CO concentration in clinic A during pre-combustion monitoring ( Fig. 1(a)). The fluctuation observed might have been due to possible photocatalytic surface reactions of wall paint constituents on exposure to light leading to degradations or secondary emissions (Salthammer and Fuhrmann, 2007). The concentration of CO in simulation clinics A, B and C varied from 0.2 ppm to 2.2 ppm, with an average of 0.96 ppm (0.95 mg m -3 ), 0.9 ppm (0.85 mg m -3 ) and 1.23 ppm (1.14 mg m -3 ) in the three clinics, respectively. The maximum post-combustion CO concentration for moxa floss samples A, B and C was 10.1 ppm, 10.6 ppm and 11 ppm, respectively, averaged 9.333 ppm (8.7 mg m -3 ), which was lower than the hourly limit of 10 mg m -3 set by CAAQS. Both the pre-and postcombustion concentrations of CO were below the CO hourly limit set by NAAQS (35 ppm) (40 CFR Part 50) and the WHO (26 ppm) (Raub and World Health Organization, 1999). Both results were also less than the occupational limit recommended by NIOSH (35 ppm) and OSHA (50 ppm) (National Institute for Occupational Safety and Health (NIOSH), 2012), as shown in Table 1. The post-combustion variation pattern for CO exhibited an interesting trend for all samples, in which the CO concentration continued to increase for 15 minutes after combustion before it remained in a relatively steady concentration ( Fig. 1(b)), implying that there was still smouldering after combustion.
The actual CO concentration produced from each type of moxa floss combustion was calculated after deduction from its background concentration. The actual CO concentration emitted was on average 8.18 ppm (7.78 mg m -3 ), 8.10 ppm (7.65 mg m -3 ) and 8.68 ppm (7.74 mg m -3 ) for samples A, B and C respectively. There was no observed difference in the actual average CO concentration produced from the three types of moxa floss. Furthermore, the average emission rate was 0.13 mg min -1 per cubic meter, which was lower than the level of 0.16 g min -1 recommended by the WHO for household fuel combustion under unvented conditions (World Health Organization, 2014).

CO 2 Concentration
The variation pattern of CO 2 concentration during precombustion phase was different from that of CO due to CO 2 expired by the researcher performing the sampling and emission tests ( Fig. 2(a)). The average background concentration of CO 2 in each simulation clinic was 0.0756%, 0.0759% and 0.0715%, respectively.
The variation pattern of CO 2 for both pre-and postcombustion phases in each simulation clinic was fairly synchronous (Fig. 2(b)). The CO 2 concentration produced by sample A during post-combustion varied from 0.118% to 0.164%, averaged 0.139%, while the CO 2 concentration of sample B, which was of the same ratio but longer storage year, ranged from 0.107% to 0.168%, averaged 0.136%. When comparing sample A and sample C, which were of the same storage year but different ratio, the CO 2 concentration for the latter ranged from 0.106% to 0.176%, averaged 0.140%. The post-combustion concentration of CO 2 for all the samples were lower than the TWA recommended by NIOSH and OSHA (Table 1).
To calculate the actual CO 2 concentration produced by combustion of moxa floss, the following Eqs. (1) to (4) were used to deduct the background concentration and contribution from human respiration.
where C exp is the CO 2 expiratory concentration of the researcher (%), C b60 is the background concentration of CO 2 at the 60-minute sampling point (%) and C b0 is the background concentration at the 0-minute sampling point (%) during the 1 h pre-combustion phase. ST: A short-term exposure limit designated by "ST" preceding the value. c C: A ceiling recommended exposure limit designated by "C" preceding the value; unless noted otherwise, the ceiling value should not be exceeded at any time. C emi+exp = C 0 -C b60 (3) where C emi+exp is the emission and expiratory concentration of CO 2 (%) during the combustion phase, C 0 is the concentration of CO 2 (%) at the 0-minute post-combustion sampling point.
where C a is the actual CO 2 concentration produced from moxa floss combustion (%). The actual CO 2 concentration was on average 0.03%, 0.028% and 0.031% for samples A, B and C, respectively. Similar to the results for CO emission, no significant difference could be found in the actual CO 2 concentration produced from the burning of the three moxa floss samples.

SO 2 and NO 2 Concentrations
Concentrations of SO 2 for both pre-and post-combustion phases in the simulation clinics were below detectable limit. For NO 2, the average background concentration in the three simulation clinics was 0.04 mg m -3 , 0.02 mg m -3 and 0.01 mg m -3 , respectively. The maximum NO 2 concentration during post-combustion was 0.018 mg m -3 , 0.009 mg m -3 and 0.008 mg m -3 for the three moxa floss samples, respectively. NO 2 concentrations for both pre-and post-combustion phases were lower than the level of 0.2 mg m -3 stated in the ambient air quality standards of Europe, United States, China and WHO (40 CFR Part 50;European Commission;World Health Organization, 2010), and also the short term (ST) limit set by the NIOSH and the ceiling value (C) established by OSHA (National Institute for Occupational Safety and Health (NIOSH), 2012) (Table 1).
NO 2 is a highly toxic oxidizing gas generally produced as a byproduct of combustion processes (Shu et al., 2010). It was thus worth noting that the NO 2 concentration decreased during post-combustion as compared to precombustion (Fig. 3). Besides producing thermal stimulation during moxibustion, moxa floss also emits visible light and infrared radiation during combustion that may play a role in the therapeutic actions of moxibustion (Deng and Shen, 2013). It is therefore hypothesized that the decrease in NO 2 may be due to photolysis of a portion of NO 2 in the  simulation clinics to produce NO and other oxidized nitrogen species (NO z ). Furthermore, atmospheric NO 2 could undergo chemical transformations to NO z , which may also co-vary with NO 2 , especially during air stagnation events (Brook et al., 2007). In our study, the windows and doors were sealed throughout the emission tests, which could highly lead to air stagnation. Moxibustion room has also been found to contain polycyclic aromatic hydrocarbons (PAHs) (Mo et al., 2014). These PAHs could adsorb to air particles and be nitrated by ppm levels of NO 2 containing traces of nitric acid (Pitts et al., 1978). It might also therefore be possible that the combustion of moxa floss led to thermal activation of adsorbed PAHs, resulting in reaction with NO 2 to form nitro-PAHs (NPAHs) (Pitts et al., 1978;Pitts, 1979;Kristovich and Dutta, 2005).

CONCLUSIONS
To the best of the authors' knowledge, this is the first report that systematically studied the gaseous pollutants in environmental moxa smoke during the combustion of moxa floss samples of different storage years and ratios. The concentrations of CO, CO 2 , SO 2 and NO 2 from burning 4 g of moxa floss did not exceed the hourly limit stated in available international standards for ambient air quality and were far lower than the occupational exposure limits set by NIOSH and OSHA.
In summary, the levels of target gaseous pollutants from 4 g of moxa floss combustion were not produced in quantities to cause concerns about exceeding present international standards and occupational exposure limits. However, practitioners and patients have different levels of exposure to moxa smoke. It is thus important to recognize and control occupational and non-occupational exposure to gaseous pollutants. To reduce exposure, possible measures such as good ventilation in moxibustion clinics should be undertaken. Data from our present study is an important source of information when assessing air quality in moxibustion clinics by professional authorities. Future studies could be conducted in an indoor mass balance model (box model) or test chamber to remove the influence of air exchange such as air change rate (ACH) and other possible factors on the emission of the gaseous pollutants.