Environmental Science
Processes & Impacts View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
PAPER
Cite this: Environ. Sci.: Processes Impacts, 2014, 16, 291
View Journal | View Issue
Urban atmospheric formaldehyde concentrations measured by a differential optical absorption spectroscopy method Xiang Li,a Shangshang Wang,ab Rui Zhoua and Bin Zhou*a In this study a differential optical absorption spectroscopy (DOAS) method was used to monitor formaldehyde (HCHO) concentrations in Shanghai ambient air at a research station in Fudan University. The measurements were carried out during April 2010–April 2011 and a total of 120 940 recorded data points were obtained. The average HCHO concentration was found to be the highest (10.0 ppbv) during August 2010 and the lowest (2.0 ppbv) during April 2010. The diurnal variation of HCHO and O3 followed very similar trends in all the seasons. This was evident from the fact that HCHO had a strong positive correlation with O3. Both peaked once in the morning (07:00–09:00 local time), and once in the night (16:00–19:00 local time). The peak concentrations varied from season to season, which could be attributed to the seasonal variation in anthropogenic activity, traffic movement and atmospheric boundary layer conditions. The background HCHO concentration in 2011 winter (similar to 12.0 ppbv) was an order of magnitude higher than that observed in 2010 spring (similar to 2.0 ppbv); corresponding
Received 21st October 2013 Accepted 28th November 2013
with the results of several pollution controls adopted by the Shanghai administrative government before and after the EXPO 2010 period (May 1, 2010–Oct. 31 2010). This study contributed the basic
DOI: 10.1039/c3em00545c
information for understanding the concentration level and the chemical processes of atmospheric
rsc.li/process-impacts
HCHO in a major metropolitan area.
Environmental impact As predicted environmental concentrations of atmospheric HCHO are in the ppbv range, extremely sensitive analytical techniques are required for the detection of HCHO in urban air. The research presented in this manuscript examined the use of a DOAS system for the detection of atmospheric HCHO and its relation to SO2, NO2 and O3. This work addressed the long-term variability of atmospheric HCHO in the Shanghai metropolitan area, and provided direct evidence that the HCHO levels in urban air were signicantly affected by the atmospheric photooxidation processes. This study contributes towards the understanding of the distribution, seasonal variation, the source, the fate and the atmospheric process of ambient HCHO in urban air.
Introduction HCHO is the simplest and most abundant carbonyl in urban air. It is formed as a result of photochemical oxidation and it is also released directly into the atmosphere. Photochemical oxidation is the largest source of HCHO in the atmosphere and it accounts for about 70 to 90 percent of HCHO found in the atmosphere depending on the location.1 HCHO is also released from vegetables, animal wastes, and cars when incomplete combustion occurs and from fuel combustion at oil reneries.2 HCHO is easily photolyzed, forming HO2 and thereby peroxides. Reactions with OH radicals and photolysis are the main HCHO loss a
Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science and Engineering, Fudan University, 220 Handan Road, Shanghai 200433, China. E-mail: [email protected]; lixiang@ fudan.edu.cn; Fax: +86 21-65642080; Tel: +86 21-65642521-804
b
School of Environment and Architecture, University of Shanghai for Science & Technology, 516 Jungong Road, Shanghai, 200093, China
This journal is © The Royal Society of Chemistry 2014
processes.3 Since HCHO plays an important intermediate role in the oxidation process of methane as well as of other nonmethane VOCs, studies on the formation and change of HCHO in the atmosphere have always been an important subject in the atmospheric environment. In remote and rural areas, the natural background concentration of HCHO in the air is typically lower than 1.0 ppbv.4 In urban environments, outdoor HCHO concentrations are more variable and depend on local conditions; annual averages are usually between 1.0 and 30.0 ppbv. Short-term peaks, e.g. in heavy traffic or during severe inversions, can range up to 54.3 ppbv.5 For example, the annual average concentrations of HCHO were 2.0 ppb in Chicago, 7.2 ppb in Los Angeles, 4.4 ppb in St. Louis, and 7.9 ppb in Houston in 2005.6 Very high atmospheric HCHO levels, above 30.0 ppbv, were recently reported for Mexico City and the Greater Houston area.6,7 Elsewhere in Europe and North America research showed that the concentration of HCHO in the air was generally lower than
Environ. Sci.: Processes Impacts, 2014, 16, 291–297 | 291
View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Environmental Science: Processes & Impacts
17 mg m3 (12.8 ppbv), but signicantly higher than that of acetaldehyde and other carbonyl species.8 Some Chinese cities also had a high atmospheric HCHO concentration, such as in Beijing, with 5.1 mg m3 (3.9 ppbv) in winter and 19.5 mg m3 (14.6 ppbv) in summer.9 Even in Guangzhou and Taiyuan in recent years, the atmospheric HCHO concentrations were also running up to 17.8 mg m3 (13.4 ppbv) and 32.0 mg m3 (24.0 ppbv), respectively.10 The high ambient levels of HCHO in the Chinese cities are partially attributed to the direct emissions from mobile and industrial sources.11 The problems of HCHO pollution in the air also existed in Hong Kong, during the summer and winter measurement periods, where the ambient HCHO concentrations ranged between 5.6 (4.2) and 9.1 mg m3 (6.8 ppbv).10 Accurate measurements of HCHO concentrations are important fundamental subjects for a complete understanding of the basic chemistry that occurs in the atmosphere. Highly sensitive techniques are required to measure HCHO concentrations under different circumstances. HCHO concentration measurements have long been performed using chemical derivatization methods.12–14 However, these methods are easy to be interfered by ambient temperature and oxidizing substances (such as O3), and have poor temporal resolution and poor selectivity for measuring HCHO in the air. The required integration time for sample collection has also largely limited the use of chemical methods because they cannot describe the rapid uctuations in HCHO concentrations. The scale of these uctuations is important because HCHO is an intermediate product during the photooxidation of hydrocarbons that undergoes rapid photolysis to produce HOx radicals. Despite the advantages of the HCHO derivatization process such as simplicity and relatively low cost, spectroscopic techniques provide a reliable and reproducible means of timeresolved collection of atmospheric HCHO concentrations. DOAS,15 FTIR,16 and Tunable Diode Laser Absorption Spectroscopy17 are spectroscopic methods oen used for in situ measurements, where the absorption by HCHO in the UV or IR regions is detected with a long-path setup such as a White cell system. Among these, DOAS is a relatively new technique, where scattered solar radiation is collected by a telescope from different directions in order to derive the column densities of absorbing species. It can give the mean value of the measured gas concentration along the light path from several hundred meters to several kilometers, which can eliminate the effect of pollution source discharge on the measured results.16,18,19 Thus, the detection result from DOAS is a more regional representative. This paper studies the application of the DOAS technique for the measurement of HCHO in the atmospheric environment. The measurements presented here were carried out in Shanghai from April 2010 to April 2011. The investigation aimed to measure a wide range of trace gases and, for an extended period of time, to assess the concentration changes on seasonal timescales. Of relevance here was that the observation suite included measurements of surface O3 and NO2. A full description of the site location and investigation details is also provided below.
292 | Environ. Sci.: Processes Impacts, 2014, 16, 291–297
Paper
Experimental methods Measurements and experimental setup The study area pertains to Shanghai (Fig. 1), which is the largest city in China. It has 17 county-level divisions with its suburbs extending up to 6340 km2, most of its land area is at, apart from a few hills in the southwest corner, with an average elevation of 4 m. Shanghai city is situated in 30 400 to 31 530 N latitude and 120 510 to 122 120 E longitude. The municipality is located at the mouth of the Yangtze River in the middle portion of the Chinese coast, and is bounded to the east by the East China Sea. Population of the city according to 2010 census is over 23 million, which is purely urbanized. With a subtropical maritime monsoon climate, Shanghai is generally mild and moist. Although situated in the southern part of China, Shanghai still has four distinct seasons – the warm spring (from March to May), the hot rainy summer (from June to September), the cool autumn (October and November) and the overcast cold winter (December to the next February). The city has an average temparature of 4.2 C in January and 27.9 C in July, and has an annual mean temperature of 16.1 C. The annual rainfall is about 1200 mm occurring mostly during the monsoon season corresponding to May– September. The air masses during monsoon months originate predominantly from the southeast direction (summer) and during the other seasons from the northwest direction (winter). The active DOAS measurements for HCHO were carried out on the campus of Fudan University (31 180 N, 121 300 E) from April 2010 to April 2011, which is located well within the urban northeastern part. By observing the light change from an articial light source, the active DOAS system measures the integrated concentration of atmospheric trace gases along the optical path, and yields the average concentration of trace gases by dividing the integrated concentration by the absorption path length.20,21 Because the light source is located together with the receiver, the telescope can both emit the light beam out and receive the returned beam. A retro-reector array was placed at a certain distance to fold the beam back to the telescope. Consequently, the light travels a double distance between the telescope and the retro-reector array. Including the light
Fig. 1 The overall view of the measurement site (from http:// map.baidu.com) and the light path of the DOAS system.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Paper
Environmental Science: Processes & Impacts
source (150 W xenon lamp), the transmitting/receiving telescope with a diameter of 210 mm of the DOAS system was set on the roof of 4th Teaching Building at an altitude of about 20 m above the ground (Fig. 1).22 The retro-reector was mounted at a height of approximately 44 m at the building of Yangpu Hightech Base, which is located south-east to the 4th Teaching Building at a distance of 0.7 km. The DOAS light beam travels above the campus and across Handan Road from north-west to south-east, totally yielding a round-trip optical path of about 1360 meters. About two-thirds of the DOAS light beam was on the campus. Handan Road, a trunk road with heavy traffic and an expressway tunnel beneath, runs 300 m south of the 4th Teaching building. Furthermore, there are several branch roads around the campus. The inuences of open agricultural res are negligible at all seasons and pollution from domestic fuels is quite nonexistent as people use natural gas for cooking. Therefore, vehicle emission is the only major pollutant source near the measurement site. Spectra evaluation and data analysis The DOAS technique is based on detecting relatively narrow absorption features of specied molecules that absorb light according to the Lambert–Beer law.15 In our experiments, several steps were taken to guarantee high quality in the determination of absorber concentrations. The measured spectra were corrected for electronic offset and stray light from sources other than the Xe arc lamp (i.e., scattered sun light). High pass ltering was used to separate narrow absorption structures from the broadband structures including Rayleigh and Mie scattering functions, and other instrument effects. A low pass ltering was applied to reduce high frequency noise. The concentrations of HCHO, O3 and NO2 were determined by simultaneously tting the reference spectra of these trace gases in different tting windows to the treated atmospheric spectrum using a nonlinear least-squares method.23 For HCHO, the tting window was restricted to 313– 340 nm in order to avoid the xenon lamp emission peak interference. In this tting window, for reducing the interference caused by other gases, SO2, NO2, and O3 were also considered besides HCHO, Fig. 2 is an example of this tting window. To ensure the lowest detection limits, the concentrations of NO2 and O3 were evaluated in other spectral regions (NO2 and HONO: 337.6–371.9 nm; O3 and SO2: 278.4–286.3 nm). In all tting windows, the lamp structure was considered. The detection limits (3s) have been estimated to be 1.0 ppb for HCHO, 1.0 ppb for NO2 and 1.5 ppb for O3 according to instrument noise in a light path of 1.36 km and an integration time of 3 min. Meteorological parameters like temperature, solar radiation, humidity, surface pressure, and wind speed and direction, and visibility were obtained from the meteorological station of Shanghai Pudong Meteorological Bureau.
Results and discussion Frequency distribution of the HCHO concentration Fig. 3 shows the relative frequency distribution of ambient HCHO concentrations (ppbv) in different ranges for the study
This journal is © The Royal Society of Chemistry 2014
Fitting example, from (a) to (d), they are the differential absorption spectra of HCHO (10.6 ppbv), O3 (26.3 ppbv), SO2 (1.5 ppbv), NO2 (7.5 ppbv) respectively, (e) is the lamp structure, (f) is the atmospheric differential absorption spectrum (black line) including the fitting spectrum (red line), (g) is the residual spectrum. Fig. 2
period (April 2010–April 2011) in Shanghai. There were a total of 120 940 recorded datasets during the observation period. It showed that 85% of all HCHO concentrations lie in the range of 0–14.0 ppbv and 11% lie in the range of 14.0–38.0 ppbv. The remaining 4% of the data were in the concentration range below 0 ppbv due to the noise of the instrument.
Monthly variations of the HCHO concentration Fig. 4 shows the monthly variation of HCHO concentrations during April 2010–April 2011 over the study area. Each box represents each month, where lower and higher boundaries of box limits are the 25 and 75 percentiles. The median value and the mean concentrations are shown as respectively the solid lines and hollow squares in the boxes. The low and high external lines represent the 90 and 10 percentiles while the maximum and minimum values are marked with a solid triangle outside the boxes. This graph gives the statistical
Environ. Sci.: Processes Impacts, 2014, 16, 291–297 | 293
View Article Online
Environmental Science: Processes & Impacts
Paper
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Diurnal variations of HCHO associated with O3 and NO2
Relative frequency distribution of ambient HCHO levels for 120 940 data points measured in Shanghai during April 2010–April 2011.
Fig. 3
Monthly variations of formaldehyde concentrations in the air of Shanghai during 2010–2011. Symbolic representation: hollow squares inside box (mean value), solid line inside box (median value), box limits (25 and 75 percentiles), low and high external lines (10 and 90 percentiles), and solid triangle outside the box (maximum and minimum values).
Fig. 4
information on the general evolution of the concentration for a given month. It indicates not only the dynamics of the concentrations for every month, but also high values of the concentration happening on a monthly basis. Monthly variations suggested high concentrations of HCHO (10.0 ppbv) during the hottest two months (July–August) and low HCHO concentrations (
Processes & Impacts View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
PAPER
Cite this: Environ. Sci.: Processes Impacts, 2014, 16, 291
View Journal | View Issue
Urban atmospheric formaldehyde concentrations measured by a differential optical absorption spectroscopy method Xiang Li,a Shangshang Wang,ab Rui Zhoua and Bin Zhou*a In this study a differential optical absorption spectroscopy (DOAS) method was used to monitor formaldehyde (HCHO) concentrations in Shanghai ambient air at a research station in Fudan University. The measurements were carried out during April 2010–April 2011 and a total of 120 940 recorded data points were obtained. The average HCHO concentration was found to be the highest (10.0 ppbv) during August 2010 and the lowest (2.0 ppbv) during April 2010. The diurnal variation of HCHO and O3 followed very similar trends in all the seasons. This was evident from the fact that HCHO had a strong positive correlation with O3. Both peaked once in the morning (07:00–09:00 local time), and once in the night (16:00–19:00 local time). The peak concentrations varied from season to season, which could be attributed to the seasonal variation in anthropogenic activity, traffic movement and atmospheric boundary layer conditions. The background HCHO concentration in 2011 winter (similar to 12.0 ppbv) was an order of magnitude higher than that observed in 2010 spring (similar to 2.0 ppbv); corresponding
Received 21st October 2013 Accepted 28th November 2013
with the results of several pollution controls adopted by the Shanghai administrative government before and after the EXPO 2010 period (May 1, 2010–Oct. 31 2010). This study contributed the basic
DOI: 10.1039/c3em00545c
information for understanding the concentration level and the chemical processes of atmospheric
rsc.li/process-impacts
HCHO in a major metropolitan area.
Environmental impact As predicted environmental concentrations of atmospheric HCHO are in the ppbv range, extremely sensitive analytical techniques are required for the detection of HCHO in urban air. The research presented in this manuscript examined the use of a DOAS system for the detection of atmospheric HCHO and its relation to SO2, NO2 and O3. This work addressed the long-term variability of atmospheric HCHO in the Shanghai metropolitan area, and provided direct evidence that the HCHO levels in urban air were signicantly affected by the atmospheric photooxidation processes. This study contributes towards the understanding of the distribution, seasonal variation, the source, the fate and the atmospheric process of ambient HCHO in urban air.
Introduction HCHO is the simplest and most abundant carbonyl in urban air. It is formed as a result of photochemical oxidation and it is also released directly into the atmosphere. Photochemical oxidation is the largest source of HCHO in the atmosphere and it accounts for about 70 to 90 percent of HCHO found in the atmosphere depending on the location.1 HCHO is also released from vegetables, animal wastes, and cars when incomplete combustion occurs and from fuel combustion at oil reneries.2 HCHO is easily photolyzed, forming HO2 and thereby peroxides. Reactions with OH radicals and photolysis are the main HCHO loss a
Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science and Engineering, Fudan University, 220 Handan Road, Shanghai 200433, China. E-mail: [email protected]; lixiang@ fudan.edu.cn; Fax: +86 21-65642080; Tel: +86 21-65642521-804
b
School of Environment and Architecture, University of Shanghai for Science & Technology, 516 Jungong Road, Shanghai, 200093, China
This journal is © The Royal Society of Chemistry 2014
processes.3 Since HCHO plays an important intermediate role in the oxidation process of methane as well as of other nonmethane VOCs, studies on the formation and change of HCHO in the atmosphere have always been an important subject in the atmospheric environment. In remote and rural areas, the natural background concentration of HCHO in the air is typically lower than 1.0 ppbv.4 In urban environments, outdoor HCHO concentrations are more variable and depend on local conditions; annual averages are usually between 1.0 and 30.0 ppbv. Short-term peaks, e.g. in heavy traffic or during severe inversions, can range up to 54.3 ppbv.5 For example, the annual average concentrations of HCHO were 2.0 ppb in Chicago, 7.2 ppb in Los Angeles, 4.4 ppb in St. Louis, and 7.9 ppb in Houston in 2005.6 Very high atmospheric HCHO levels, above 30.0 ppbv, were recently reported for Mexico City and the Greater Houston area.6,7 Elsewhere in Europe and North America research showed that the concentration of HCHO in the air was generally lower than
Environ. Sci.: Processes Impacts, 2014, 16, 291–297 | 291
View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Environmental Science: Processes & Impacts
17 mg m3 (12.8 ppbv), but signicantly higher than that of acetaldehyde and other carbonyl species.8 Some Chinese cities also had a high atmospheric HCHO concentration, such as in Beijing, with 5.1 mg m3 (3.9 ppbv) in winter and 19.5 mg m3 (14.6 ppbv) in summer.9 Even in Guangzhou and Taiyuan in recent years, the atmospheric HCHO concentrations were also running up to 17.8 mg m3 (13.4 ppbv) and 32.0 mg m3 (24.0 ppbv), respectively.10 The high ambient levels of HCHO in the Chinese cities are partially attributed to the direct emissions from mobile and industrial sources.11 The problems of HCHO pollution in the air also existed in Hong Kong, during the summer and winter measurement periods, where the ambient HCHO concentrations ranged between 5.6 (4.2) and 9.1 mg m3 (6.8 ppbv).10 Accurate measurements of HCHO concentrations are important fundamental subjects for a complete understanding of the basic chemistry that occurs in the atmosphere. Highly sensitive techniques are required to measure HCHO concentrations under different circumstances. HCHO concentration measurements have long been performed using chemical derivatization methods.12–14 However, these methods are easy to be interfered by ambient temperature and oxidizing substances (such as O3), and have poor temporal resolution and poor selectivity for measuring HCHO in the air. The required integration time for sample collection has also largely limited the use of chemical methods because they cannot describe the rapid uctuations in HCHO concentrations. The scale of these uctuations is important because HCHO is an intermediate product during the photooxidation of hydrocarbons that undergoes rapid photolysis to produce HOx radicals. Despite the advantages of the HCHO derivatization process such as simplicity and relatively low cost, spectroscopic techniques provide a reliable and reproducible means of timeresolved collection of atmospheric HCHO concentrations. DOAS,15 FTIR,16 and Tunable Diode Laser Absorption Spectroscopy17 are spectroscopic methods oen used for in situ measurements, where the absorption by HCHO in the UV or IR regions is detected with a long-path setup such as a White cell system. Among these, DOAS is a relatively new technique, where scattered solar radiation is collected by a telescope from different directions in order to derive the column densities of absorbing species. It can give the mean value of the measured gas concentration along the light path from several hundred meters to several kilometers, which can eliminate the effect of pollution source discharge on the measured results.16,18,19 Thus, the detection result from DOAS is a more regional representative. This paper studies the application of the DOAS technique for the measurement of HCHO in the atmospheric environment. The measurements presented here were carried out in Shanghai from April 2010 to April 2011. The investigation aimed to measure a wide range of trace gases and, for an extended period of time, to assess the concentration changes on seasonal timescales. Of relevance here was that the observation suite included measurements of surface O3 and NO2. A full description of the site location and investigation details is also provided below.
292 | Environ. Sci.: Processes Impacts, 2014, 16, 291–297
Paper
Experimental methods Measurements and experimental setup The study area pertains to Shanghai (Fig. 1), which is the largest city in China. It has 17 county-level divisions with its suburbs extending up to 6340 km2, most of its land area is at, apart from a few hills in the southwest corner, with an average elevation of 4 m. Shanghai city is situated in 30 400 to 31 530 N latitude and 120 510 to 122 120 E longitude. The municipality is located at the mouth of the Yangtze River in the middle portion of the Chinese coast, and is bounded to the east by the East China Sea. Population of the city according to 2010 census is over 23 million, which is purely urbanized. With a subtropical maritime monsoon climate, Shanghai is generally mild and moist. Although situated in the southern part of China, Shanghai still has four distinct seasons – the warm spring (from March to May), the hot rainy summer (from June to September), the cool autumn (October and November) and the overcast cold winter (December to the next February). The city has an average temparature of 4.2 C in January and 27.9 C in July, and has an annual mean temperature of 16.1 C. The annual rainfall is about 1200 mm occurring mostly during the monsoon season corresponding to May– September. The air masses during monsoon months originate predominantly from the southeast direction (summer) and during the other seasons from the northwest direction (winter). The active DOAS measurements for HCHO were carried out on the campus of Fudan University (31 180 N, 121 300 E) from April 2010 to April 2011, which is located well within the urban northeastern part. By observing the light change from an articial light source, the active DOAS system measures the integrated concentration of atmospheric trace gases along the optical path, and yields the average concentration of trace gases by dividing the integrated concentration by the absorption path length.20,21 Because the light source is located together with the receiver, the telescope can both emit the light beam out and receive the returned beam. A retro-reector array was placed at a certain distance to fold the beam back to the telescope. Consequently, the light travels a double distance between the telescope and the retro-reector array. Including the light
Fig. 1 The overall view of the measurement site (from http:// map.baidu.com) and the light path of the DOAS system.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Paper
Environmental Science: Processes & Impacts
source (150 W xenon lamp), the transmitting/receiving telescope with a diameter of 210 mm of the DOAS system was set on the roof of 4th Teaching Building at an altitude of about 20 m above the ground (Fig. 1).22 The retro-reector was mounted at a height of approximately 44 m at the building of Yangpu Hightech Base, which is located south-east to the 4th Teaching Building at a distance of 0.7 km. The DOAS light beam travels above the campus and across Handan Road from north-west to south-east, totally yielding a round-trip optical path of about 1360 meters. About two-thirds of the DOAS light beam was on the campus. Handan Road, a trunk road with heavy traffic and an expressway tunnel beneath, runs 300 m south of the 4th Teaching building. Furthermore, there are several branch roads around the campus. The inuences of open agricultural res are negligible at all seasons and pollution from domestic fuels is quite nonexistent as people use natural gas for cooking. Therefore, vehicle emission is the only major pollutant source near the measurement site. Spectra evaluation and data analysis The DOAS technique is based on detecting relatively narrow absorption features of specied molecules that absorb light according to the Lambert–Beer law.15 In our experiments, several steps were taken to guarantee high quality in the determination of absorber concentrations. The measured spectra were corrected for electronic offset and stray light from sources other than the Xe arc lamp (i.e., scattered sun light). High pass ltering was used to separate narrow absorption structures from the broadband structures including Rayleigh and Mie scattering functions, and other instrument effects. A low pass ltering was applied to reduce high frequency noise. The concentrations of HCHO, O3 and NO2 were determined by simultaneously tting the reference spectra of these trace gases in different tting windows to the treated atmospheric spectrum using a nonlinear least-squares method.23 For HCHO, the tting window was restricted to 313– 340 nm in order to avoid the xenon lamp emission peak interference. In this tting window, for reducing the interference caused by other gases, SO2, NO2, and O3 were also considered besides HCHO, Fig. 2 is an example of this tting window. To ensure the lowest detection limits, the concentrations of NO2 and O3 were evaluated in other spectral regions (NO2 and HONO: 337.6–371.9 nm; O3 and SO2: 278.4–286.3 nm). In all tting windows, the lamp structure was considered. The detection limits (3s) have been estimated to be 1.0 ppb for HCHO, 1.0 ppb for NO2 and 1.5 ppb for O3 according to instrument noise in a light path of 1.36 km and an integration time of 3 min. Meteorological parameters like temperature, solar radiation, humidity, surface pressure, and wind speed and direction, and visibility were obtained from the meteorological station of Shanghai Pudong Meteorological Bureau.
Results and discussion Frequency distribution of the HCHO concentration Fig. 3 shows the relative frequency distribution of ambient HCHO concentrations (ppbv) in different ranges for the study
This journal is © The Royal Society of Chemistry 2014
Fitting example, from (a) to (d), they are the differential absorption spectra of HCHO (10.6 ppbv), O3 (26.3 ppbv), SO2 (1.5 ppbv), NO2 (7.5 ppbv) respectively, (e) is the lamp structure, (f) is the atmospheric differential absorption spectrum (black line) including the fitting spectrum (red line), (g) is the residual spectrum. Fig. 2
period (April 2010–April 2011) in Shanghai. There were a total of 120 940 recorded datasets during the observation period. It showed that 85% of all HCHO concentrations lie in the range of 0–14.0 ppbv and 11% lie in the range of 14.0–38.0 ppbv. The remaining 4% of the data were in the concentration range below 0 ppbv due to the noise of the instrument.
Monthly variations of the HCHO concentration Fig. 4 shows the monthly variation of HCHO concentrations during April 2010–April 2011 over the study area. Each box represents each month, where lower and higher boundaries of box limits are the 25 and 75 percentiles. The median value and the mean concentrations are shown as respectively the solid lines and hollow squares in the boxes. The low and high external lines represent the 90 and 10 percentiles while the maximum and minimum values are marked with a solid triangle outside the boxes. This graph gives the statistical
Environ. Sci.: Processes Impacts, 2014, 16, 291–297 | 293
View Article Online
Environmental Science: Processes & Impacts
Paper
Published on 28 November 2013. Downloaded by McMaster University on 27/10/2014 14:33:24.
Diurnal variations of HCHO associated with O3 and NO2
Relative frequency distribution of ambient HCHO levels for 120 940 data points measured in Shanghai during April 2010–April 2011.
Fig. 3
Monthly variations of formaldehyde concentrations in the air of Shanghai during 2010–2011. Symbolic representation: hollow squares inside box (mean value), solid line inside box (median value), box limits (25 and 75 percentiles), low and high external lines (10 and 90 percentiles), and solid triangle outside the box (maximum and minimum values).
Fig. 4
information on the general evolution of the concentration for a given month. It indicates not only the dynamics of the concentrations for every month, but also high values of the concentration happening on a monthly basis. Monthly variations suggested high concentrations of HCHO (10.0 ppbv) during the hottest two months (July–August) and low HCHO concentrations (
