- Email: [email protected]
Ambient concentrations of 1,3-butadiene in the UK
Chemico-Biological Interactions 135– 136 (2001) 177– 206
www.elsevier.com/locate/chembiont
Ambient concentrations of 1,3-butadiene in the UK G.J. Dollard *, C.J. Dore, M.E. Jenkin National En6ironmental Technology Centre, AEA Technology En6ironment, Culham, Abingdon, Oxfordshire OX14 3ED, UK
Abstract This paper assesses the current knowledge of 1,3-butadiene as an atmospheric pollutant, considers measurement techniques and reviews available data on 1,3-butadiene monitoring and emissions estimates. Atmospheric chemistry, sources of emission, current legislation, measurement techniques and monitoring programmes for 1,3-butadiene are reviewed. There have been comparatively few studies of the products of oxidation of 1,3-butadiene in the atmosphere. However, on the basis of the available information, and by analogy with the oxidation mechanism for the widely-studied and structurally similar natural hydrocarbon isoprene (2-methyl-1,3-butadiene), it is possible to define some features of the likely oxidation pathways for 1,3-butadiene. The total UK 1,3-butadiene emission to the atmosphere for 1996 has been estimated at 10.60 kTonnes. 1,3-Butadiene is a product of petrol and diesel combustion; consequently this total is dominated by road transport exhaust emissions (accounting for some 68% of the total). Off-road vehicles and machinery are responsible for 14% of the total UK emission. 1,3-Butadiene is used in the manufacture of numerous rubber compounds, and consequently emissions arise from both the manufacture and use of 1,3-butadiene in industrial processes. Emissions from the chemical industry account for 18% of the UK total emission- 8% from 1,3-butadiene manufacture and 10% from 1,3-butadiene use. The United Kingdom Expert Panel on Air Quality Standards (EPAQS) has published a report on 1,3-butadiene, and recommended a national air quality standard of 1.0 ppb (expressed as an annual rolling mean). This was adopted by the Government as part of the National Air Quality Strategy (NAQS) in 1997, and a target of compliance by 2005 was set. Work conducted for the review of the NAQS (1999) indicated that it was likely that all locations would be compliant with the national standard by the end of 2003. As a result, the review updated the air quality objective for 1,3-butadiene, with the deadline for compliance
* Corresponding author. Tel.: + 44-1235-463040; fax: +44-1235-463001. E-mail address: [email protected] (G.J. Dollard). 0009-2797/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 9 - 2797 ( 0 1 ) 0 0 1 9 0 - 9
178
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
being brought forward to 31/12/2003. The UK Hydrocarbon Monitoring Network provides continuous hourly measurements of 1,3-butadiene at 13 sites, and has been operational since 1993. The dataset that is available allows spatial and temporal trends to be evaluated, and has proved to be invaluable in characterising the current ambient levels of 1,3-butadiene in the UK. Hourly maximum concentrations of 1,3-butadiene of up to 10 ppb (1 ppb =1 ppb, i.e. 1 vol. of 1,3-butadiene in 1 000 000 000 vol. of air. 1 ppb of 1,3-butadiene is ca. equal to 2.25 mg m − 3 at 20°C) may be measured for several hours at the sites. Monthly mean concentrations are typically 0.1 –0.4 ppbv. At most sites, these levels are driven by emissions from motor vehicles. Occasionally emissions of 1,3-butadiene from industrial sources may elevate 1,3-butadiene concentrations to several tens of ppb. Trend analysis of the data suggests that ambient concentrations of 1,3-butadiene in the UK are declining at about 10% per year. © 2001 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction This paper gives an overview of the current knowledge of 1,3-butadiene as an atmospheric pollutant in the UK. The atmospheric chemistry, sources of emission, current legislation, measurement techniques and monitoring programmes in the UK are considered. 1,3-Butadiene is a hydrocarbon of empirical formula C4H6. It is a conjugated diene, consisting of a C4 chain incorporating two carbon-carbon double bonds separated by a carbon-carbon single bond, as indicated below in Fig. 1 This structure distinguishes it from the other butadiene isomer 1,2-butadiene, in which the double bonds are adjacent (i.e. not conjugated). It is a colourless, slightly aromatic gas at STP, with an odour threshold of 1.6 ppm [1]. 1,3-butadiene has melting and boiling points of −108.9 and − 4.41°C, respectively. The Expert Panel on Air Quality Standards (1,3-butadiene, EPAQS 1994) accepted 1,3-butadiene as a genotoxic carcinogen (see also [2]), and the USEPA have classified 1,3-butadiene as a Group B2 carcinogen [1]. The name ‘1,3-butadiene’ does not comply with the precise IUPAC nomenclature for the compound, which should be buta-1,3-diene. However, the name 1,3-butadiene is used throughout this report as it is the most commonly used name, which identifies the compound structure precisely. Other common names include butadiene, vinylethylene and bivinyl, however, of these, butadiene does not distinguish between the two isomers identified above. As a conjugated dialkene the molecule is highly reactive. Typical atmospheric lifetimes vary from several hours to several days depending on conditions.
Fig. 1. The structure of 1,3-butadiene.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
179
1,3-Butadiene is emitted into the atmosphere primarily from vehicle exhaust emissions, but significant emissions also arise from both its manufacture, and its use in industry. Vehicle emissions are a significant source of numerous pollutants, and consequently have been studied in detail. However, speciation of volatile organic compounds, to identify emissions of individual species, is a relatively recent development. Consequently, although it is possible to generate estimates for current and future emissions of 1,3-butadiene, which are considered to be reliable, there are still refinements being made to the estimates. The most significant industrial use of 1,3-butadiene is associated with the manufacture of synthetic rubbers; more specifically with the manufacture of polystyrene-butadiene-styrene (or SBS rubber). As a result, in the UK, the manufacture of tyres is an important source of atmospheric emissions. In addition there are emissions associated with the manufacture of 1,3-butadiene in the UK. The industrial sources of 1,3-butadiene are considered in more detail in Section 3. There have been numerous studies conducted to investigate the carcinogenic properties of 1,3-butadiene [18]. These studies have either been concerned with accidental human exposure, or animal exposure under laboratory conditions. However, combining the results from these studies does not produce a conclusive picture, and it is clear that there is a high degree of uncertainty in any conclusions drawn from these studies. The Expert Panel on Air Quality Standards (EPAQS) conducted an assessment of the health impact of 1,3-butadiene and recommended an air quality standard of 1 ppb expressed as an annual running mean.
2. The atmospheric chemistry of 1,3-butadiene
2.1. Atmospheric lifetime 1,3-butadiene (CH2 =CHCH = CH2) is a conjugated dialkene. Similarly to many other unsaturated organic compounds (i.e. those which are not fully hydrogenated), 1,3-butadiene is highly reactive under atmospheric conditions: it is potentially removed by gas-phase reactions with the hydroxyl (OH) radical, the nitrate (NO3) radical and ozone (O3), and information is also available for its reaction with nitrogen dioxide (NO2). The corresponding rate coefficients are presented in Table 1, together with lifetimes of 1,3-butadiene with respect to these reactions under a variety of representative atmospheric conditions. The reaction with NO2 is clearly an insignificant removal process, although corresponding reactions may be important for some larger conjugated dienes [3,4]. The relative importance of the other removal routes depends on ambient conditions. The OH radical is predominantly generated in the atmosphere by photochemical processes, most notably the photolysis of ozone in the presence of water vapour O3 +sunlight (UV band)O(1D) +O2
(1)
O(1D) +H2O OH + OH
(2)
180
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 1 Rate coefficients for the reactions of 1,3-butadiene with OH, O3, NO3 and NO2, and lifetimes for its removal by these reactions under a variety of representative conditions Rate coefficient (T= 298 K)
Representative lifetimesb
cm3 molecule−1 s−1a
Conditionsc
Lifetime
OH
6.7×10−11
O3
6.3×10−18
NO3 NO2
1.0×10−13 3.0×10−20
Winter average Summer average photochemical episode Background photochemical episode Average Background urban episode
21 h 4.2 h 25 min 2.4 days 18 h 11 h 4.2 years 150 days
a
Rate coefficients taken from the reviews of [6,7]. Lifetimes with respect to removal by reaction with OH, O3, NO3 and NO2 are calculated for a variety of representative conditions, as summarised in footnote ‘c’. The lifetimes are the time taken for 1,3-butadiene to decay to 1/e of its starting concentration under the given conditions. c Representative OH concentrations are 0.008, 0.04 and 0.4 pptv for winter, summer and photochemical episode conditions, respectively. O3 concentrations are 30 and 100 ppbv for background and photochemical episode, respectively. NO3 average concentration is 10 pptv. NO2 concentrations are 10 and 100 ppbv for background and urban episode, respectively [5]. b
Consequently, OH is present at significant concentrations during daylight hours, and is more abundant during the summer than the winter. As discussed in detail elsewhere (e.g. [5]), reaction with the OH radical is generally the major fate of most trace atmospheric species. The information presented in Table 1 shows that it is important for 1,3-butadiene, with the 1/e lifetime with respect to reaction with OH typically varying from ca. 1 day in the winter to ca. 30 min during a summertime photochemical episode. The NO3 radical is formed from the reaction of NO2 with O3, as follows: NO2 +O3 NO3 +O2
(3)
During daylight, NO3 is photolysed extremely efficiently (on the timescale of a few seconds), and also reacts rapidly with NO, such that it is unable to accumulate to a significant concentration. During the night-time, however, the chemistry of both NOX and NO3 differs from the daytime behaviour, as described fully in [5]. In the absence of sunlight, NOX tends to be in the form of NO2 with the concentration of NO only being significant in the vicinity of direct emission sources. Provided the NO concentration is sufficiently suppressed, NO3 may be present at significant concentrations, although its concentration is clearly very dependent on the ambient conditions. On the basis of the available rate coefficient (Table 1), and a representative average NO3 concentration, the average 1/e lifetime for 1,3-butadiene with respect to removal by reaction with NO3 is ca. 11 h. This suggests that removal by reaction with NO3 may be of comparable importance to the OH reaction, particularly under night-time and winter-time conditions.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
181
Unlike the reactions with OH and NO3, reaction with O3 can potentially occur throughout the diurnal cycle. However, as shown in Table 1, the typical lifetimes of 1,3-butadiene with respect to reaction with O3 at representative ambient levels are significantly longer than those for the reactions with OH and NO3 under comparable conditions. Reaction with O3 is, therefore, only a minor removal process for 1,3-butadiene.
2.2. Degradation chemistry Available information suggests that the degradation chemistry of 1,3-butadiene follows the pattern observed for unsaturated hydrocarbons in general [6–9]. During daylight, the chemistry initiated by reaction with OH radicals predominantly converts 1,3-butadiene into a series of first generation oxygenated products by the following generic sequence, also involving the organic peroxy radicals (RO2 = HOC4H6O2), oxy radicals (RO= HOC4H6O), the hydroperoxy radical (HO2) and NOX : OH +1,3-butadiene (+ O2) RO2
(4)
RO2 +NO RO + NO2
(5)
RO oxygenated products+ HO2
(6)
HO2 +NO OH +NO2
(7)
The addition of OH to the double bonds in 1,3-butadiene initiates the sequence. However, OH is regenerated (reaction (7)), such that this mechanism is a catalytic cycle converting 1,3-butadiene into its first generation oxygenated products, and also oxidising NO to NO2. As described in detail in [5], the subsequent photolysis of NO2 leads to the generation of O3. Consequently, in common with a large number of other volatile organic compounds, the photo-oxidation of 1,3-butadiene leads to the generation of O3 as a by-product. Laboratory studies have identified acrolein (propenal) as a major first generation product [10,11] and furan as a minor product [11] as shown in the schematic in Fig. 2. In addition, numerous other oxidised products have very recently been detected [12], although not quantified, and some of these are also shown in the figure. By analogy with the oxidation mechanism for the widely-studied and structurally-similar natural hydrocarbon, isoprene (2-methyl-1,3-butadiene), formaldehyde and 4-hydroxy-2-butenal are also predicted to be major first generation products [6,7,9]. The further oxidation of these oxygenated products under atmospheric conditions generates glyoxal and glycolaldehyde as additional intermediate compounds during the eventual breakdown to CO and CO2 (see Fig. 2). In some respects, the night-time chemistry initiated by reaction with NO3 radicals follows a similar pattern (although it cannot lead directly to O3 formation). Acrolein and formaldehyde are formed as minor first generation products, but the major products are carbonyl compounds which retain ‘nitrate’ (ONO2) groups following the initial attack of NO3 [13]. These have been identified as
182
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
O2NOCH2CH =CHCHO and O2NOCH2C(O)CH = CH2 [14]. There is no information on the further oxidation of complex oxidised organic nitrogen compounds of this type. It is probable that they will be removed predominantly by reaction with OH or direct photolysis. This may ultimately generate CO2 and release NOX, if the oxidation occurs in the gas phase. However, as the further oxidation is likely to generate intermediate products containing a large number of polar substituent groups (e.g. ONO2, OH, CHO), uptake into aqueous droplets may also be important. Dry deposition of pollutants from the atmosphere i.e. the direct absorption of the pollutant at the earth’s surface — vegetation, soil, water etc. is often another important removal pathway. Deposition studies of 1,3-butadiene conducted by Dore (reported in [5]) concluded that dry deposition was unlikely to be a significant removal pathway for 1,3-butadiene. The study investigated the deposition over a variety of surface types. Soil and sea water were chosen as they represent important surface cover types. Spinach and grass were chosen to indicate the impact of differing leaf area indices on the deposition velocity. A summary of the findings is given in the Table 2. The removal of 1,3-butadiene through wet deposition (i.e. in rainfall) is expected to be insignificant, due to the low solubility of 1,3-butadiene in water.
Fig. 2. Schematic representation of the major intermediate oxidised products formed during the complete atmospheric degradation of 1,3-butadiene to CO2 (see text).
Fig. 3. UK emissions of 1,3-butadiene, 1970 –1996.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
183
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
184
Table 2 Dry deposition velocities of 1,3-butadiene over differing surface types Surface type
Deposition velocity (mm s−1)
Spinach Grass Soil Sea water
78 82 63 1.5
3. UK sources of 1,3-butadiene The majority of 1,3-butadiene emissions arise from road transport exhaust emissions, however there are also significant contributions to the UK total from other types of transportation and activities in the chemical industry. Total UK emissions of 1,3-butadiene have been summarised in the following table (Table 3), and are plotted in Figs. 3 and 4. This data has been extracted from the National Atmospheric Emissions Inventory [15], and is given in UK Emissions of Air Pollutants 1970– 1996 [16]. To date, it has not been possible to include emission estimates of 1,3-butadiene from stationary combustion in the NAEI, as no suitable emission factors are known. This will be addressed when suitable ‘specia-
Fig. 4. UK emissions estimates and projections of 1,3-butadiene from road transport.
Source sector
UK emissions of 1,3-butadiene by source sector (kTonnes)
Chemical industry Use Manufacture Road transport Petrol vehicles DERV vehicles Off-road transport and machinerya Total UK emission a
1970
1980
1990
1991
1992
1993
1994
1995
1996
1.91 1.04 0.87 5.57 5.07 0.50 1.85 9.33
2.12 1.16 0.96 7.52 6.93 0.59 1.61 11.25
2.06 1.12 0.94 10.66 9.85 0.81 1.45 14.17
1.90 1.03 0.86 10.71 9.84 0.87 1.49 14.10
1.86 1.01 0.84 10.27 9.41 0.86 1.51 13.64
1.76 0.96 0.80 9.57 8.68 0.89 1.47 12.80
1.90 1.04 0.86 8.90 7.97 0.93 1.44 12.25
1.97 1.07 0.89 8.13 7.17 0.96 1.41 11.51
1.94 1.06 0.88 7.21 6.31 0.90 1.45 10.60
Includes emissions from Landfill. Data extracted from the NAEI [16].
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 3 UK emissions of 1,3-butadiene by source sector
185
186
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 4 UK 1996 emissions of 1,3-butadiene for specific industrial processes Industrial process
UK 1,3-butadiene emission (Tonnes)
Styrene–butadiene rubber manufacture Nitrile rubber production Polybutadiene manufacture ABS and SAN resinsa
751 38 170 100
a
‘ABS’ is acrylonitrile-butadiene-styrene; ‘SAN’ is styrene and acrylonitrile copolymer. Data extracted from the NAEI [16]
tion profiles’ are available to assign the emissions of VOC into emissions of individual components. The emission of 1,3-butadiene from tyre burning is not considered here, as the emissions are small (the UK emission is estimated at ca. 17 tonnes). 1,3-Butadiene is present in natural gas, and consequently there is an emission from gas pipe leakage. This has been omitted, as there is considerable uncertainty associated with current estimates.
3.1. Use in the chemical industry The primary use of 1,3-butadiene in the chemical industry is associated with the manufacture of polymers and rubber products. The presence of a double bond at either end of the C4 carbon chain (see Fig. 1) make it ideal for polymerisation reactions. The 1996 emissions of 1,3-butadiene from the different industrial processes are given in Table 4. Of the four different processes given in Table 4, styrene-butadiene rubber is produced in the largest quantities, and consequently generates the largest emission of 1,3-butadiene. Styrene– butadiene rubber (or SBS rubber, styrene– butadiene– styrene rubber) is a hard wearing synthetic rubber that is used in: tyres, hoses, industrial goods, shoes etc. [17]. It is manufactured by polymerising styrene and 1,3-butadiene. Atmospheric emissions arise from the storage of the feedstock, as well as the manufacture processes. The remaining three processes, production of: polybutadiene, nitrile rubbers, ABS and SAN resins, are similar in that 1,3-butadiene is used either in the production of an end product polymer, or is used to generate an intermediate compound. Emissions to the atmosphere arise from fugitive sources, as well as specific point locations.
3.2. Road transport Exhaust emissions of 1,3-butadiene from road vehicles account for ca. 70% of the total emissions in the UK as quoted by the NAEI. Earlier national emission estimates from vehicles were quoted as VOC, however, speciation into individual components has been developed by the NAEI over recent years making it possible to estimate emissions of 1,3-butadiene. The methodology for estimating emissions
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
187
from road transport vehicles is continually under review, but has been well established for VOC. As a result, it is expected that the speciation profiles give the largest potential for reducing uncertainties in estimates of 1,3-butadiene emissions in the short term. 1,3-Butadiene is not a major component in petrol or diesel, but is produced during the combustion process from higher olefins, hence evaporative emissions are not significant. The current emission estimates for 1,3-butadiene are given in Fig. 3. This indicates increasing emissions from petrol vehicles to 1990, followed by a steady decrease from 1990 onwards. Two factors are thought to have caused this feature. First, the annual petrol consumption in the UK steadily increased from in the late 1980s through until 1990. For the first several years of the 1990s, the petrol consumption decreased. This factors was also combined with the introduction of legislation in 1992, which required new petrol driven cars to be equipped with catalytic converters. As the penetration of catalytic equipped vehicles into the national vehicle fleet increased, the 1,3-butadiene emissions decreased. This increase in penetration is expected to continue for several years. To illustrate the impact that catalytic converters have on the 1,3-butadiene emissions from petrol driven road vehicles, several emission factors have been included below in Table 5. In addition, diesel vehicles have been included. Data from National Road Transport Emissions Model (pers. comm. Murrells 1999). It is evident from Table 5 that a significant reduction in the emissions from petrol driven cars has been achieved. Emissions from urban driving have been reduced by a factor of ca. 70, and emissions from cold starts have been reduced by a factor of ca. 10. Fig. 4 gives the current emission estimates and projections for 1,3-butadiene from the NAEI Road Transport Emission Model (pers. comm. Murrells 1999). It is evident that emissions are expected to fall steadily until ca. 2004. After which, Table 5 Selected emission factors (for different vehicle types and abatement technologies) 1,3-Butadiene emission factors (mg/km) Vehicle type
Petrol cars
Diesel cars
HGV
Abatement type
ECE 15.04 91/441/EEC Stage II Pre-Stage I Stage I Stage II Pre-Stage I Stage I Stage II
Road classification Urban
Rural single
Motorway
Cold start
37.74 1.14 0.5 4.95 3.75 2.62 32.27 16.14 14.52
16.50 0.00 0.00 4.38 3.35 2.35 41.32 26.86 24.79
8.70 0.08 0.03 2.47 1.44 1.01 12.73 9.55 8.27
35.80 3.07 3.07 11.99 11.99 11.99 53.25 53.25 53.25
188
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 6 Emissions from off-road transport and machinery Off-road transport and machinery
UK 1,3-butadiene emission (kTonnes)
Aircraft — civil and military Railways Marine, including coastal, fishing and naval Domestic machinery Agriculture machinery and vehicles Industrial- including airport support vehicles Landfill Total
0.08 0.15 0.28 0.45 0.08 0.37 0.03 1.45
increasing use of road transportation and attainment of the maximum market penetration for catalytic converters then results in a steady, but slow rise of 1,3-butadiene emissions from the road transport sector. Petrol cars remain the most significant vehicle type.
3.3. Off road transport and machinery Emissions of 1,3-butadiene in this sector arise primarily from off-road transportation and machinery in the industrial, domestic and agricultural sectors. Table 6 gives an indication of the contributions to the source sector total. Results have been taken from the NAEI for 1996.
4. Air quality guidelines in the UK In 1994, the UK Expert Panel on Air Quality Standards (EPAQS) published a report on 1,3-butadiene, considering not only current concentrations, but studies of human and animal exposure to elevated levels of 1,3-butadiene (1,3-butadiene, [18]). The study accepted that 1,3-butadiene is a genotoxic carcinogen, and consequently no safe level can be defined. However, the panel agreed on a recommended concentration that would ensure ‘the risks are exceedingly small and unlikely to be detectable by any practicable method.’ From the epidemiological studies that had been conducted, it was noted that there was considerable uncertainty, making the extrapolation of risk impossible. However, the panel recommended that ‘an Air Quality Standard for 1,3-butadiene in UK of 1 ppb measured as a running annual average’ be introduced. The panel further recommended that this standard be reviewed after a period of, at most, 5 years. The United Kingdom National Air Quality Strategy (NAQS) was published in 1997 and reviewed the standard and ambient concentrations of 1,3-butadiene. It was noted that the increased use of road vehicles equipped with catalytic converters was causing a reduction in the ambient 1,3-butadiene concentrations. It was considered ‘likely’ that by the year 2000, there would be no exceedances of the
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
189
EPAQS recommendation in urban background locations and ‘most roadside locations next to heavily trafficked roads’. In addition, it was noted that the current foreseen emission reductions would result in the EPAQS recommended 1 ppb level would be attained in ‘virtually all locations’. The government officially accepted the EPAQS recommendation, as a standard for 1,3-butadiene, and specified an objective of attaining this standard by 2005. The NAQS also noted that the EPAQS recommended standard would be reviewed by 1999, and further assessment would be conducted as appropriate. A review of the [19] was conducted in 1998 ([20] Review), and it was proposed that the deadline for attaining the 1 ppb standard adopted in the NAQS be brought forward to the 31 December 2003. This decision was based on modelling work conducted by [21,22] utilising the most current data available from the national monitoring networks (see below). 5. Monitoring in the UK
5.1. 1,3 -Butadiene automatic monitors Automatic measurement of 1,3-butadiene in the ambient atmosphere in the UK is generally conducted by gas chromatography. Typically, hydrocarbons in air are sampled and measured in the following manner: the air sample is passed through a bed of adsorbent material, which is cooled to increase the trapping efficiency and enable the removal of the more volatile species, such as 1,3-butadiene. The cold trap is flash heated to release the trapped components into a stream of inert carrier gas, in this case helium. The sample is carried onto a capillary column. The compounds pass along this column at different speeds (broadly depending on volatility). The temperature of the column is increased across the separation phase to ensure that the different components are suitably separated. This stage of the analysis will last typically for 40–50 min, however, 1,3-butadiene ‘elutes’ i.e. reaches the end of the column after ca. 10–15 min. Each component passes from the column to the detector. The detector typically used is a flame ionisation detector (FID), which gives an increased signal as each component passes. As a result, the output from the detector is a chromatogram consisting of a series of peaks, eluting at different times, and of different size. For a particular column type, temperature cycle and carrier gas flow, the elution time of each compound (and hence the order in which the components arrive at the detector) is well known typically from test samples consisting of single components. As a result, each peak in the chromatogram may be identified. The peak areas indicate the amount of each component in the sample. These peak areas are converted into mass values by applying ‘response factors’ i.e. the effective sensitivity of the detector to each of the components. These response factors are calculated from calibration samples, where a known amount of each component is introduced to the detector. The conversion from a mass to a concentration value is achieved from sample volume information.
190
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
The measurement of 1,3-butadiene is dominated by measurements conducted with automatic point monitors, giving compounds in the C2 to C8 or C9 range. A national network of these instruments has been established, however the expense of the equipment and running costs has meant that there are limited measurements with this type of detector from other sources. There have been relatively few measurements using the passive sampler technique, even though studies have indicated a good performance. It is thought that the low use of this technique has primarily arisen from a lack of end user awareness.
5.2. The UK hydrocarbon monitoring network Following the publication of the UK Government White Paper on the environment [23] the Department of the Environment set up its Enhanced Urban Monitoring Initiative. This was to establish comprehensive monitoring sites in city locations. The second stage of the initiative was the UK Hydrocarbon Monitoring Network. Two pilot studies were conducted at Exhibition Road, London (roadside location) and Middlesborough (urban background location). The first network site was commissioned in 1991, and by the end of 1995, there were 12 operational sites. In 1997, the Marylebone Road site was added to the network. Site information is indicated in Table 7. More detailed site information may be found in the ‘UK National Air Quality Archive’ (see reference list).
Table 7 The UK hydrocarbon monitoring network sites Site name
Site type
Inception date
Middlesbrough Belfast Birmingham Cardiff Edinburgh London, Eltham London, UCL Bristol Harwell Leeds Liverpool Southampton Marylebone road
Urban industrial Urban background Urban background Urban background Urban background Suburban Roadside Urban background Rural Urban background Urban background Urban centre Roadside
Pre 1993 1993 1993 1993 1993 1993 1993a 1994 1994b 1994 1995 1995 1997
a
This site was operational prior to 1993 as part of the pilot study. Although established in 1994, the site was used as a development platform resulting in an incomplete dataset until 1995. b
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
191
5.2.1. Data handling, ratification and quality control With numerous network sites continuously producing hourly measurements of more than 20 compounds, robust data ratification procedures are essential. Data processing and ratification procedures were developed to automate the reliable identification and quantification of the compounds. Pattern recognition software was used to ensure reliable compound identification. Particular attention was paid to the reliability of the 1,3-butadiene and benzene data, as national guideline values and standards exist for both compounds. Some of the procedures are on-line, however, some quality assurance and quality control procedures require detailed checking. Consequently ‘provisional’ data is made available on-line, but ‘ratified’ data are usually made available several months following the measurement. 5.2.2. Dissemination of information Provisional data are used to generate values of the rolling annual mean for 1,3-butadiene (and benzene) which are passed to television text services for on-line display. These pages are updated hourly, and the rolling annual mean value is accompanied by a ‘level’ indicator to aid interpretation of the concentration values. In addition, provisional hourly values of 1,3-butadiene (and benzene) are passed to the National Air Quality Archive http://www.aeat.co.uk/netcen/aqarchive/archome/ html. This archive is updated daily. All 1,3-butadiene concentration measurements are reported in ppb from the network and may be converted into m gm − 3 U by multiplying by ca. 2.25. 6. Ambient concentrations The UK Hydrocarbon Network has been operational for several years and has generated data of high quality. Thus, the large data set available from the network can be used to evaluate numerous different trends in the ambient 1,3-butadiene concentration.
6.1. Seasonal trends It is difficult to interpret seasonal trends from plots of hourly 1,3-butadiene concentration, due to the large variations from hour to hour. Consequently, to illustrate points more clearly the data presented in Figs. 5– 7 have been plotted as a rolling mean of 90 days. This averaging period retains general trends, but reduces the impact of short-term variations. The seasonal cycling of 1,3-butadiene concentration is caused by several factors. Evidently, the source strength is important when looking at variations over shorter time periods, but here the impact is reduced as the points are expressed as averages across a relatively long time period. If variations in the road traffic activity over long periods are considered to be small, then there are two factors, which affect the concentration- atmospheric mixing and chemical removal. The cycling of atmospheric mixing during a 24 h period is discussed in the following section. Evidently,
192 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 5. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means).
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 6. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means). 193
194 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 7. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means).
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
195
Table 8 Annual decreases in concentration by site Site name
Belfast Birmingham Bristol Cardiff Edinburgh Eltham (London) Leeds Middlesbroughb UCL (London) Liverpoolc Southamptonc Marylebone Roadc
Concentration decrease (ppb per year)
0.025 0.027 0.018 0.018 0.0042 0.0025 0.0081 0.025 0.025 – – –
Mean concentration (ppb)
0.23 0.28 0.25 0.27 0.12 0.21 0.22 0.28 0.43 – – –
Concentration decrease (ppb per year) Daytimea
Night-timea
0.024 0.022 0.026 0.021 0.00057 0.0019 0.00032 0.023 0.0031 – – –
0.025 0.032 0.013 0.013 0.0012 0.0053 0.011 0.021 0.020 – – –
a Results calculated from deseasonalised time series. Daytime and night-time are defined as the hourly measurements from 07:00 to 18:00 h and 19:00 to 06:00 h, respectively. b Episodes at Middlesbrough have not been removed from the data, and consequently affect the calculation of the annual concentration decrease. c Not enough data exists from these sites to enable the trend analysis to be conducted.
the length of daytime periods varies with the seasons, and hence the mixing also varies. In the summer, effective vertical mixing is present for considerably longer than during winter days and the mixing depth is greater. As a result, there is a greater dilution effect during the summer. Table 1 indicates different rates of the chemical removal of 1,3-butadiene observed in the summer and the winter. Fig. 5 gives the rolling 90 day mean for the Eltham site. The Eltham site is situated in London away from localised sources, such as roads, and is classified as a suburban site (Tables 8 and 9). An annual pattern of concentration for this site is clearly present from mid 1994. Maximum concentrations are attained in October or November, and are ca. 0.37 ppb. Concentrations then decrease through December – March, reaching the annual minimum values, typically in April or May. These annual minima (expressed as 90 day averages) are ca. 0.13 ppb for the Eltham site. Concentrations then steadily increase through to the annual maxima observed in October or November. By comparison, the UCL and Marylebone Road sites are roadside sites, and consequently concentrations are dominated by the emissions from one particular section of road. The road transport activity on this single road section may be affected for long periods of time by factors such as roadworks, rail strikes etc. Hence, even when concentrations are expressed as 90 day averages, it is possible that there are periods which show significant deviation from a regular annual pattern of concentration cycling.
196
1,3-Butadiene
All road types
Urban roads
Rural and motorways
Year
Number of road links
Maximum concentration (ppb)
Number of road links
Maximum concentration Number of road (ppb) links
Maximum concentration (ppb)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Total
1616 1113 645 327 140 35 12 1 0 0 15 226
2.1 1.9 1.6 1.4 1.2 1.1 0.9 0.8 0.6 0.5
1288 876 493 249 104 30 12 1 0 0 7508
2.1 1.9 1.6 1.4 1.2 1.1 0.9 0.8 0.6 0.5
1.5 1.3 1.2 1.0 0.9 0.7 0.6 0.5 0.4 0.4
328 237 147 78 36 5 0 0 0 0 7718
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 9 Estimated exceedances of NAQS 1,3-butadiene standard [22]
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
197
The urban background sites shown in Fig. 8 (all sites except Harwell — the rural site) display similar trends to those noted for Eltham, except that the absolute concentrations are higher, ranging from ca. 0.1 to 0.4 ppb.
6.2. Diurnal trends Figs. 8 and 9 give plots for each site averaged in such as way as to represent the average diurnal cycle for each calendar year. These plots show several features of interest. First, they indicate that the concentrations are generally decreasing with time — this is considered in more detail below. UK Long-Term Trends from Measurements. One of the other features, which these plots are designed to show, are variations in concentration across a diurnal period. Years with a significantly reduced data capture have been removed. There are two major facts affecting the 1,3-butadiene concentration throughout the day. The first is the source strength, in most cases this will simply be the level of road traffic activity in the vicinity of the monitoring site. The second will be meteorological conditions, and in particular the mixing conditions. In general, vertical mixing increases throughout the morning reaching a maximum around midday or early afternoon. Vertical mixing then decreases, and during night-time hours is small. Thus the effective dilution of the 1,3-butadiene varies throughout a 24 h period. The diurnal trends for the Eltham site can be seen in Fig. 8 Eltham is a background site in London, and displays a typical diurnal variation. 05:00 – 09:00 h; concentrations of 1,3-butadiene starts to increase from 05:00 h, and rise sharply from 07:00 h as the traffic activity increases. Morning peak concentrations are reached at ca. 09:00 h, which corresponds to the peak activity on the roads (Transport Statistics Report). 10:00 – 14:00 h; from 10:00 h, the concentrations falls until 14:00 h, this correlates with decreasing traffic activity. In addition the vertical mixing increases from mid-morning (through until ca. 15:00–16:00 h). 14:00 – 21:00 h; the afternoon rush hour is spread over a longer period than the morning [24]. Vertical mixing increases until 15:00–16:00 h, and then decreases throughout the afternoon and evening. As a result, 1,3-butadiene concentrations steadily increase from 14:00 until ca. 21:00 h. 21:00 – 05:00 h; concentrations fall from 21:00 h through to 05:00 h due to the decreased traffic activity. Meteorological variations, particularly variations in the vertical mixing, have a clear impact on the 1,3-butadiene concentrations. This is illustrated by comparing typical concentrations at 03:00 and 15:00 h on Fig. 8. The concentrations are not dissimilar, but the expected traffic flows for these times will differ enormously. Other sites can be seen to display similar trends across an annually averaged 24 h period (Figs. 8 and 9), however, there are some particular features of note. The London UCL site is located at the roadside, and consequently concentrations are higher than those observed an urban background sites (Fig. 8). In addition, it is evident that the decrease of concentrations during the middle part of
198 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 8. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206 199
Fig. 9. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
200
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
the day are less pronounced. This may be attributed to the different patterns of traffic activity observed for Central London (UCL), and Outer London (Eltham). It is also evident that there is a substantial decrease in concentration from the 1997 data to the 1998 data. The Marylebone Road site (Fig. 8) is also located in Central London, but its traffic flows do not show good correlation with the general levels of the traffic activities in central London. This is because the traffic patterns in Marylebone Road are not typical of those in central London; here, there is no large decrease in traffic activities between the morning and afternoon rush hour periods (suggesting a congested road for much of the daytime). This is reflected in the concentration variations evident in Fig. 8. The Harwell monitoring site is a rural location, and consequently concentrations are low. In addition, as the site is some distance from a major road, the diurnal trend observed at e.g. Eltham is not present. The general trend of increased concentrations at night is caused by the changes in atmospheric mixing as earlier mentioned. In 1995, there was a significant ‘industrial’ release of 1,3-butadiene in the Middlesbrough area. After much publicity it was discovered that a sea tanker vented residual 1,3-butadiene to air (see below). The impact on the 90 day mean (Fig. 7) and the diurnal 1993 plot (Fig. 9) are evident. From hourly data, not shown here, there also appears to be an episode that occurred in 1998 (an hourly concentration value reaching ca. 60 ppb). This explains the behaviour of the 1998 trace in the diurnal plot (Fig. 9).
6.3. UK long-term trends from measurements The figures presented above give an indication that concentrations of 1,3-butadiene have been decreasing during recent years. It is possible to quantify the annual decrease in concentration by ‘deseasonalising’ the hourly data and then evaluating the decrease with time. The results from such an analysis are summarised in the following table for the various measurement sites. The annual decrease in concentration for many of the sites is comparable to ca. 10% of the annual mean concentration. Given that 1,3-butadiene emissions from road transport comprise the major source at most of these sites, the most likely cause of the reductions is the introduction of catalytic converters. It is evident from the figures that there is a particularly large decrease in concentration from 1997 to 1998 at several sites. If UCL is taken as an example location, then the percentage decreases in concentration observed from 1993 to 1997 are generally consistent with the percentage decreases observed in the national 1,3-butadiene emissions from urban areas up to 1997 (extracted from the road transport sector of the UK National Atmospheric Emissions Inventory (NAEI) — the order of 5– 10% per year. This is also in general agreement with modelled decreases of maximum concentration given below; this comparison is not yet possible with data from 1998 onwards. The observed decrease from 1997 to 1998 for UCL is a remarkable 40%. 1998 emission estimates for 1,3-butadiene have not yet been finalised by the NAEI.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206 201
Fig. 10. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
202
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
6.4. UK long-term trends from modelling The work by [22], in support of the NAQS, estimated future roadside concentrations of 1,3-butadiene. A relationship between 1,3-butadiene emissions and concentrations was indirectly established by using the well-characterised relationship between NOX roadside concentrations and NOX road transport emissions. The use of future emission estimates then enabled future 1,3-butadiene concentrations to be estimated. The following table gives estimates for the maximum roadside concentrations of 1,3-butadiene (expressed as fixed annual means) and the number of roadlinks exceeding the NAQS standard of 1 ppb (expressed as an annual running mean). It should be noted that due to the uncertainties associated with the method, and the different way in which the averages are calculated, values of maximum concentration (fixed means) which exceed 0.7 ppb were regarded as exceeding the NAQS target value of 1 ppb (annual running mean).
6.5. Spatial trends Figs. 10 and 11 give mapped 1,3-butadiene concentrations modelled by [22], for 1996 and 2005. The areas of high concentration are easy to identify, being London, Birmingham, Manchester, Liverpool and Leeds. Evidently, it is London that dominates both 1996 and 2005. Clearly this is in agreement with observations made from the data supplied by the UK Hydrocarbon Monitoring Network.
Fig. 11. Background concentrations of 1,3-butadiene (ppb)-1996; (b) background concentrations of 1,3 butadiene (ppb)-2005.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
203
Table 10 Exceedences of the national air quality standard Maximum running annual mean Site
1999
1998
1997
1996
1995
1994
Belfast Birmingham Bristol Cardiff Edinburgh Harwell (rural site) Leeds Liverpool London Eltham London Marylebone London UCL Middlesbrough Southampton
0.3 0.4 0.4 0.5 0.2 0.1 0.4 0.4 0.4 2.4 0.6 0.3 0.6
0.5 0.5 0.5 0.5 0.2 0.2 0.5 0.5 0.5 2.5 0.9 0.5 0.7
0.5 0.5 0.5 0.7 0.2 0.2 0.5 0.5 0.5
0.5 0.7 0.7 0.5 0.2 0.2 0.5 0.5 0.5
0.5 0.5 0.5 0.7 0.2
0.7 0.7
0.9 0.5 0.9
0.9 0.7 0.9
0.9 0.7
0.7 0.2
0.5 0.7 1.1
6.6. Middlesbrough tanker episode To date the most significant episode was measured at Middlesbrough in 1995. In the afternoon of 31 July 1995, 1,3-butadiene hourly concentrations rose to exceptionally high levels, peaking at over 80 ppb. Concentrations then remained high until the early hours of the following morning. High concentrations were also noted overnight from 1 to 2 August and in the afternoon of 2 August. There then followed consultation with the local authority (and a degree of press coverage) in an effort to locate the source. It came to light that a marine tanker had vented residual 1,3-butadiene after making a delivery. Venting of the tanks to atmosphere is a standard operation to empty the tanks of any residual contents, prior to loading of a different compound. Venting is usually conducted at sea, however on this particular occasion it was conducted at harbour during a period of unfavourable wind direction. This type of chemical release was, at the time, not covered by any legislation. The impact of the release on the rolling annual mean, of ca. 0.18 ppb, was to cause an increase the order of 50%. Such episodes are an issue concerning local populations. Fig. 9 indicates that another episode occurred at Middlesbrough during 1998. An individual hourly concentration exceeded 60 ppb on 14 May 1998, but the origin of this elevated concentration is not known.
6.7. Exceedences of the national air quality standard Table 10 summarises the maximum annual running mean for all of the network monitoring sites in the UK hydrocarbon network from the point 1 year of data became available.
204
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
This summary shows that the majority of network sites are below the national standard for 1,3-butadiene (1.0 ppb). Exceedences have been recorded in the two most recent years at the Kerbside monitoring site on London’s Marylebone Road (very heavy traffic) and at the London UCL site in 1994. The London UCL site samples air about 6 m from the side of a moderately busy roadway. At both of these sites, the main emission source is motor vehicle exhaust.
7. Conclusions 1,3-butadiene has been accepted as a genotoxic carcinogen (1,3-butadiene, [18]), and consequently much work has been conducted in establishing 1,3-butadiene emissions, concentrations, and the behaviour of the compound in the atmosphere. 1,3-butadiene is highly reactive under atmospheric conditions, with an estimated atmospheric lifetime varying from several minutes to months. The UK NAEI have quantified emissions of 1,3-butadiene, and report a total emission of 10.60 kTonnes for 1996. The total UK emission is noted to be decreasing. This is primarily caused by the reduction in emissions from road vehicles as the penetration of catalytic converters into the vehicle fleet increases. For 1996, 1,3-butadiene emissions are dominated by road vehicle exhaust emissions (comprising some 68% of the total emission); whilst emissions from off-road vehicles and machinery account for 14%. The remaining emissions arise from the chemical industry, and are from 1,3-butadiene manufacture and use in the production of various rubber compounds. These two processes account for 8 and 10% of the UK 1996 total emission, respectively. Work conducted by EPAQS (‘1,3-butadiene’, 1994) recommended a National Air Quality Standard of 1.0 ppb (expressed as an annual rolling mean). This was adopted by the government in 1997 as part of the NAQS. More specifically, the air quality objective that was set for 1,3-butadiene was to achieve the 1.0 ppb standard by 2005. Of the monitoring sites on the UK Hydrocarbon Monitoring Network, all have indicated compliance with this standard for some time except the Marylebone Road site. Modelling studies conducted by [22] as part of the NAQS review estimated all roadside locations to be compliant with the 1,3-butadiene standard by the end of 2003. As a result, the official air quality objective for 1,3-butadiene was amended. The deadline for compliance now stands at achieving the standard by 31 December 2003, as specified in the NAQS review. Occupational exposure limits are currently quoted as a LTEL of 10 ppm, although this is currently under review. The UK Hydrocarbon Monitoring Network has generated a comprehensive time series of 1,3-butadiene concentrations across the UK since 1993. This dataset is unparalleled and has given an insight into temporal and spatial trends of 1,3-butadiene concentrations across the UK. The data has also allowed comparisons to be made with modelled concentrations, and emission estimates. The UK Hydrocarbon Monitoring Network provides continuous hourly measurements of 1,3-butadiene at 13 sites, and has been operational since 1993. The
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
205
dataset that is available allows spatial and temporal trends to be evaluated, and has proved to be invaluable in characterising the current ambient levels of 1,3-butadiene in the UK. Hourly maximum concentrations of 1,3-butadiene of up to 10 ppb may be measured for several hours at the sites. Monthly mean concentrations are typically 0.1– 0.4 ppb. At most sites, these levels are driven by emissions from motor vehicles. Occasionally emissions of 1,3-butadiene from industrial sources may elevate 1,3-butadiene concentrations to several tens of ppb. Trend analysis of the data suggests that ambient concentrations of 1,3-butadiene in the UK are declining at about 10% per year.
Acknowledgements The authors would like to thank the following colleagues at AEA Technology Environment for their valuable contributions and input into this report: Brian Jones, Trevor Davies, Tim Murrells and Justin Goodwin. The work was funded as part of a programme of research managed by the Air and Environmental Quality Division of the UK Department of the Environment, Transport and the Regions.
References [1] UATW — Unified Air Toxics Website published by the USEPA (1998). http://www.epa.gov/ ttnuatw1/hlthef/butadien.html. [2] EH40/97 Occupational Exposure Limits, 1997. Health and Safety Executive Publications, Sheffield. ISBN 0-7176-1315-1. [3] J.P. Shi, R.M. Harrison, Rapid NO2 formation in diluted petrol-fuelled engine exhaust: a source of NO2 in winter smog episodes, Atmos. Environ. 31 (1997) 3857 – 3866. [4] R.M. Harrison, J.P. Shi, J.L. Grenfell, Novel nighttime free radical chemistry in severe nitrogen dioxide pollution episodes, Atmos. Environ. 32 (1998) 2769 – 2774. [5] PORG (1997). Ozone in the United Kingdom: Fourth report of the Photochemical Oxidants Review Group. Department of the Environment, Transport and the Regions, London. [6] R. Atkinson R. Gas-phase tropospheric chemistry of organic compounds. J.Phys.Chem.Ref. Data (1994) Monograph 2. [7] R. Atkinson, Gas-phase tropospheric chemistry of volatile organic compounds: 1 alkanes and alkenes, J. Phys. Chem. Ref. Data 26 (1997) 215 – 290. [8] M.E. Jenkin, S.M. Saunders, M.J. Pilling, The tropospheric degradation of volatile organic compounds: a protocol for mechanism development, Atmos. Environ. 31 (1997) 81 – 104. [9] M.E. Jenkin, A.A. Boyd, R. Lesclaux, Peroxy radical kinetics resulting from the OH-initiated oxidation of 1,3-butadiene, 2,3-dimethyl-1,3-butadiene and isoprene, J. Atmos. Chem. 29 (1998) 267–298. [10] A. Maldotti, C. Chiorboli, C.A. Bignozzi, C. Bartocci, V. Carassiti, Photooxidation of 1,3-butadiene containing systems: rate constant determination for the reaction of acrolein with OH radicals, Int. J. Chem. Kinet. 12 (1980) 905 – 913. [11] T. Ohta, Furan ring formation in OH-initiated photooxidation of 1,3-butadiene and cis-1,3-pentadiene, Bull. Chem. Soc. Jpn. 57 (1984) 960 – 966. [12] X. Liu, H.E. Jeffries, K.G. Sexton, Hydroxyl radical and ozone initiated photochemical reactions of 1,3-butadiene, Atmos. Environ. 33 (1999) 3005 – 3022.
206
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
[13] I. Barnes, V. Bastian, K.H. Becker, Z. Tong, Kinetics and products of the reactions of NO3 with monoalkenes, dialkenes and monoterpenes, J. Phys. Chem. 94 (1990) 2413 – 2419. [14] H. Skov, J. Hjorth, C. Lohse, N.R. Jensen, G. Restelli, Products and mechanisms of the reactions of the nitrate radical with isoprene, 1,3 butadiene and 2,3 dimethyl-1,3 butadiene in air, Atmos. Environ. 26A (1992) 2771 –2783. [15] NAEI– UK Emissions of 1,3-Butadiene pers. comm. Goodwin J, (1999). [16] A.G. Salway, H.S. Eggleston, J.W.L. Goodwin, J.E. Berry, T.P. Murrells, UK Emissions of Air Pollutants 170-1996. Published by the NAEI for the DETR (1999). [17] Chem-Facts United Kingdom, Edition. Chemical Intelligence Services. Chrompack VOC-AIR Instrument Manuals (1991). Available from Chrompack. [18] 1,3-Butadiene (EPAQS) 1994. HMSO, London. Also available at http:// www.environment.detr.gov.uk/airq/aqs/13butad/but01.htm.. [19] UK NAQS Review: a Consultation Document, 1998. Published by Department of the Environment, Transport and the Regions. [20] UK NAQS- The United Kingdom National Air Quality Strategy, 1997. Published by The Stationary Office Limited, London. [21] J.R. Stedman, Revised High Resolution Maps of Background Air Pollutant Concentrations in the UK: 1996. AEA Technology AEAT-3133, published for the DETR (1998). [22] J.R. Stedman, C.J. Dore, An Empirical Model for Quantifying Current and Future Benzene and 1,3-Butadiene Concentrations at the Roadside. AEA Technology AEAT-4301, published for the DETR (1999). [23] This Common Inheritance, Government White Paper, HMSO, London, 1990. [24] Transport Statistics Report: London Traffic Monitoring Report: 1992. Department of Transport, HMSO 0-11-551137-7, 1992.
.
www.elsevier.com/locate/chembiont
Ambient concentrations of 1,3-butadiene in the UK G.J. Dollard *, C.J. Dore, M.E. Jenkin National En6ironmental Technology Centre, AEA Technology En6ironment, Culham, Abingdon, Oxfordshire OX14 3ED, UK
Abstract This paper assesses the current knowledge of 1,3-butadiene as an atmospheric pollutant, considers measurement techniques and reviews available data on 1,3-butadiene monitoring and emissions estimates. Atmospheric chemistry, sources of emission, current legislation, measurement techniques and monitoring programmes for 1,3-butadiene are reviewed. There have been comparatively few studies of the products of oxidation of 1,3-butadiene in the atmosphere. However, on the basis of the available information, and by analogy with the oxidation mechanism for the widely-studied and structurally similar natural hydrocarbon isoprene (2-methyl-1,3-butadiene), it is possible to define some features of the likely oxidation pathways for 1,3-butadiene. The total UK 1,3-butadiene emission to the atmosphere for 1996 has been estimated at 10.60 kTonnes. 1,3-Butadiene is a product of petrol and diesel combustion; consequently this total is dominated by road transport exhaust emissions (accounting for some 68% of the total). Off-road vehicles and machinery are responsible for 14% of the total UK emission. 1,3-Butadiene is used in the manufacture of numerous rubber compounds, and consequently emissions arise from both the manufacture and use of 1,3-butadiene in industrial processes. Emissions from the chemical industry account for 18% of the UK total emission- 8% from 1,3-butadiene manufacture and 10% from 1,3-butadiene use. The United Kingdom Expert Panel on Air Quality Standards (EPAQS) has published a report on 1,3-butadiene, and recommended a national air quality standard of 1.0 ppb (expressed as an annual rolling mean). This was adopted by the Government as part of the National Air Quality Strategy (NAQS) in 1997, and a target of compliance by 2005 was set. Work conducted for the review of the NAQS (1999) indicated that it was likely that all locations would be compliant with the national standard by the end of 2003. As a result, the review updated the air quality objective for 1,3-butadiene, with the deadline for compliance
* Corresponding author. Tel.: + 44-1235-463040; fax: +44-1235-463001. E-mail address: [email protected] (G.J. Dollard). 0009-2797/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 9 - 2797 ( 0 1 ) 0 0 1 9 0 - 9
178
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
being brought forward to 31/12/2003. The UK Hydrocarbon Monitoring Network provides continuous hourly measurements of 1,3-butadiene at 13 sites, and has been operational since 1993. The dataset that is available allows spatial and temporal trends to be evaluated, and has proved to be invaluable in characterising the current ambient levels of 1,3-butadiene in the UK. Hourly maximum concentrations of 1,3-butadiene of up to 10 ppb (1 ppb =1 ppb, i.e. 1 vol. of 1,3-butadiene in 1 000 000 000 vol. of air. 1 ppb of 1,3-butadiene is ca. equal to 2.25 mg m − 3 at 20°C) may be measured for several hours at the sites. Monthly mean concentrations are typically 0.1 –0.4 ppbv. At most sites, these levels are driven by emissions from motor vehicles. Occasionally emissions of 1,3-butadiene from industrial sources may elevate 1,3-butadiene concentrations to several tens of ppb. Trend analysis of the data suggests that ambient concentrations of 1,3-butadiene in the UK are declining at about 10% per year. © 2001 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction This paper gives an overview of the current knowledge of 1,3-butadiene as an atmospheric pollutant in the UK. The atmospheric chemistry, sources of emission, current legislation, measurement techniques and monitoring programmes in the UK are considered. 1,3-Butadiene is a hydrocarbon of empirical formula C4H6. It is a conjugated diene, consisting of a C4 chain incorporating two carbon-carbon double bonds separated by a carbon-carbon single bond, as indicated below in Fig. 1 This structure distinguishes it from the other butadiene isomer 1,2-butadiene, in which the double bonds are adjacent (i.e. not conjugated). It is a colourless, slightly aromatic gas at STP, with an odour threshold of 1.6 ppm [1]. 1,3-butadiene has melting and boiling points of −108.9 and − 4.41°C, respectively. The Expert Panel on Air Quality Standards (1,3-butadiene, EPAQS 1994) accepted 1,3-butadiene as a genotoxic carcinogen (see also [2]), and the USEPA have classified 1,3-butadiene as a Group B2 carcinogen [1]. The name ‘1,3-butadiene’ does not comply with the precise IUPAC nomenclature for the compound, which should be buta-1,3-diene. However, the name 1,3-butadiene is used throughout this report as it is the most commonly used name, which identifies the compound structure precisely. Other common names include butadiene, vinylethylene and bivinyl, however, of these, butadiene does not distinguish between the two isomers identified above. As a conjugated dialkene the molecule is highly reactive. Typical atmospheric lifetimes vary from several hours to several days depending on conditions.
Fig. 1. The structure of 1,3-butadiene.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
179
1,3-Butadiene is emitted into the atmosphere primarily from vehicle exhaust emissions, but significant emissions also arise from both its manufacture, and its use in industry. Vehicle emissions are a significant source of numerous pollutants, and consequently have been studied in detail. However, speciation of volatile organic compounds, to identify emissions of individual species, is a relatively recent development. Consequently, although it is possible to generate estimates for current and future emissions of 1,3-butadiene, which are considered to be reliable, there are still refinements being made to the estimates. The most significant industrial use of 1,3-butadiene is associated with the manufacture of synthetic rubbers; more specifically with the manufacture of polystyrene-butadiene-styrene (or SBS rubber). As a result, in the UK, the manufacture of tyres is an important source of atmospheric emissions. In addition there are emissions associated with the manufacture of 1,3-butadiene in the UK. The industrial sources of 1,3-butadiene are considered in more detail in Section 3. There have been numerous studies conducted to investigate the carcinogenic properties of 1,3-butadiene [18]. These studies have either been concerned with accidental human exposure, or animal exposure under laboratory conditions. However, combining the results from these studies does not produce a conclusive picture, and it is clear that there is a high degree of uncertainty in any conclusions drawn from these studies. The Expert Panel on Air Quality Standards (EPAQS) conducted an assessment of the health impact of 1,3-butadiene and recommended an air quality standard of 1 ppb expressed as an annual running mean.
2. The atmospheric chemistry of 1,3-butadiene
2.1. Atmospheric lifetime 1,3-butadiene (CH2 =CHCH = CH2) is a conjugated dialkene. Similarly to many other unsaturated organic compounds (i.e. those which are not fully hydrogenated), 1,3-butadiene is highly reactive under atmospheric conditions: it is potentially removed by gas-phase reactions with the hydroxyl (OH) radical, the nitrate (NO3) radical and ozone (O3), and information is also available for its reaction with nitrogen dioxide (NO2). The corresponding rate coefficients are presented in Table 1, together with lifetimes of 1,3-butadiene with respect to these reactions under a variety of representative atmospheric conditions. The reaction with NO2 is clearly an insignificant removal process, although corresponding reactions may be important for some larger conjugated dienes [3,4]. The relative importance of the other removal routes depends on ambient conditions. The OH radical is predominantly generated in the atmosphere by photochemical processes, most notably the photolysis of ozone in the presence of water vapour O3 +sunlight (UV band)O(1D) +O2
(1)
O(1D) +H2O OH + OH
(2)
180
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 1 Rate coefficients for the reactions of 1,3-butadiene with OH, O3, NO3 and NO2, and lifetimes for its removal by these reactions under a variety of representative conditions Rate coefficient (T= 298 K)
Representative lifetimesb
cm3 molecule−1 s−1a
Conditionsc
Lifetime
OH
6.7×10−11
O3
6.3×10−18
NO3 NO2
1.0×10−13 3.0×10−20
Winter average Summer average photochemical episode Background photochemical episode Average Background urban episode
21 h 4.2 h 25 min 2.4 days 18 h 11 h 4.2 years 150 days
a
Rate coefficients taken from the reviews of [6,7]. Lifetimes with respect to removal by reaction with OH, O3, NO3 and NO2 are calculated for a variety of representative conditions, as summarised in footnote ‘c’. The lifetimes are the time taken for 1,3-butadiene to decay to 1/e of its starting concentration under the given conditions. c Representative OH concentrations are 0.008, 0.04 and 0.4 pptv for winter, summer and photochemical episode conditions, respectively. O3 concentrations are 30 and 100 ppbv for background and photochemical episode, respectively. NO3 average concentration is 10 pptv. NO2 concentrations are 10 and 100 ppbv for background and urban episode, respectively [5]. b
Consequently, OH is present at significant concentrations during daylight hours, and is more abundant during the summer than the winter. As discussed in detail elsewhere (e.g. [5]), reaction with the OH radical is generally the major fate of most trace atmospheric species. The information presented in Table 1 shows that it is important for 1,3-butadiene, with the 1/e lifetime with respect to reaction with OH typically varying from ca. 1 day in the winter to ca. 30 min during a summertime photochemical episode. The NO3 radical is formed from the reaction of NO2 with O3, as follows: NO2 +O3 NO3 +O2
(3)
During daylight, NO3 is photolysed extremely efficiently (on the timescale of a few seconds), and also reacts rapidly with NO, such that it is unable to accumulate to a significant concentration. During the night-time, however, the chemistry of both NOX and NO3 differs from the daytime behaviour, as described fully in [5]. In the absence of sunlight, NOX tends to be in the form of NO2 with the concentration of NO only being significant in the vicinity of direct emission sources. Provided the NO concentration is sufficiently suppressed, NO3 may be present at significant concentrations, although its concentration is clearly very dependent on the ambient conditions. On the basis of the available rate coefficient (Table 1), and a representative average NO3 concentration, the average 1/e lifetime for 1,3-butadiene with respect to removal by reaction with NO3 is ca. 11 h. This suggests that removal by reaction with NO3 may be of comparable importance to the OH reaction, particularly under night-time and winter-time conditions.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
181
Unlike the reactions with OH and NO3, reaction with O3 can potentially occur throughout the diurnal cycle. However, as shown in Table 1, the typical lifetimes of 1,3-butadiene with respect to reaction with O3 at representative ambient levels are significantly longer than those for the reactions with OH and NO3 under comparable conditions. Reaction with O3 is, therefore, only a minor removal process for 1,3-butadiene.
2.2. Degradation chemistry Available information suggests that the degradation chemistry of 1,3-butadiene follows the pattern observed for unsaturated hydrocarbons in general [6–9]. During daylight, the chemistry initiated by reaction with OH radicals predominantly converts 1,3-butadiene into a series of first generation oxygenated products by the following generic sequence, also involving the organic peroxy radicals (RO2 = HOC4H6O2), oxy radicals (RO= HOC4H6O), the hydroperoxy radical (HO2) and NOX : OH +1,3-butadiene (+ O2) RO2
(4)
RO2 +NO RO + NO2
(5)
RO oxygenated products+ HO2
(6)
HO2 +NO OH +NO2
(7)
The addition of OH to the double bonds in 1,3-butadiene initiates the sequence. However, OH is regenerated (reaction (7)), such that this mechanism is a catalytic cycle converting 1,3-butadiene into its first generation oxygenated products, and also oxidising NO to NO2. As described in detail in [5], the subsequent photolysis of NO2 leads to the generation of O3. Consequently, in common with a large number of other volatile organic compounds, the photo-oxidation of 1,3-butadiene leads to the generation of O3 as a by-product. Laboratory studies have identified acrolein (propenal) as a major first generation product [10,11] and furan as a minor product [11] as shown in the schematic in Fig. 2. In addition, numerous other oxidised products have very recently been detected [12], although not quantified, and some of these are also shown in the figure. By analogy with the oxidation mechanism for the widely-studied and structurally-similar natural hydrocarbon, isoprene (2-methyl-1,3-butadiene), formaldehyde and 4-hydroxy-2-butenal are also predicted to be major first generation products [6,7,9]. The further oxidation of these oxygenated products under atmospheric conditions generates glyoxal and glycolaldehyde as additional intermediate compounds during the eventual breakdown to CO and CO2 (see Fig. 2). In some respects, the night-time chemistry initiated by reaction with NO3 radicals follows a similar pattern (although it cannot lead directly to O3 formation). Acrolein and formaldehyde are formed as minor first generation products, but the major products are carbonyl compounds which retain ‘nitrate’ (ONO2) groups following the initial attack of NO3 [13]. These have been identified as
182
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
O2NOCH2CH =CHCHO and O2NOCH2C(O)CH = CH2 [14]. There is no information on the further oxidation of complex oxidised organic nitrogen compounds of this type. It is probable that they will be removed predominantly by reaction with OH or direct photolysis. This may ultimately generate CO2 and release NOX, if the oxidation occurs in the gas phase. However, as the further oxidation is likely to generate intermediate products containing a large number of polar substituent groups (e.g. ONO2, OH, CHO), uptake into aqueous droplets may also be important. Dry deposition of pollutants from the atmosphere i.e. the direct absorption of the pollutant at the earth’s surface — vegetation, soil, water etc. is often another important removal pathway. Deposition studies of 1,3-butadiene conducted by Dore (reported in [5]) concluded that dry deposition was unlikely to be a significant removal pathway for 1,3-butadiene. The study investigated the deposition over a variety of surface types. Soil and sea water were chosen as they represent important surface cover types. Spinach and grass were chosen to indicate the impact of differing leaf area indices on the deposition velocity. A summary of the findings is given in the Table 2. The removal of 1,3-butadiene through wet deposition (i.e. in rainfall) is expected to be insignificant, due to the low solubility of 1,3-butadiene in water.
Fig. 2. Schematic representation of the major intermediate oxidised products formed during the complete atmospheric degradation of 1,3-butadiene to CO2 (see text).
Fig. 3. UK emissions of 1,3-butadiene, 1970 –1996.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
183
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
184
Table 2 Dry deposition velocities of 1,3-butadiene over differing surface types Surface type
Deposition velocity (mm s−1)
Spinach Grass Soil Sea water
78 82 63 1.5
3. UK sources of 1,3-butadiene The majority of 1,3-butadiene emissions arise from road transport exhaust emissions, however there are also significant contributions to the UK total from other types of transportation and activities in the chemical industry. Total UK emissions of 1,3-butadiene have been summarised in the following table (Table 3), and are plotted in Figs. 3 and 4. This data has been extracted from the National Atmospheric Emissions Inventory [15], and is given in UK Emissions of Air Pollutants 1970– 1996 [16]. To date, it has not been possible to include emission estimates of 1,3-butadiene from stationary combustion in the NAEI, as no suitable emission factors are known. This will be addressed when suitable ‘specia-
Fig. 4. UK emissions estimates and projections of 1,3-butadiene from road transport.
Source sector
UK emissions of 1,3-butadiene by source sector (kTonnes)
Chemical industry Use Manufacture Road transport Petrol vehicles DERV vehicles Off-road transport and machinerya Total UK emission a
1970
1980
1990
1991
1992
1993
1994
1995
1996
1.91 1.04 0.87 5.57 5.07 0.50 1.85 9.33
2.12 1.16 0.96 7.52 6.93 0.59 1.61 11.25
2.06 1.12 0.94 10.66 9.85 0.81 1.45 14.17
1.90 1.03 0.86 10.71 9.84 0.87 1.49 14.10
1.86 1.01 0.84 10.27 9.41 0.86 1.51 13.64
1.76 0.96 0.80 9.57 8.68 0.89 1.47 12.80
1.90 1.04 0.86 8.90 7.97 0.93 1.44 12.25
1.97 1.07 0.89 8.13 7.17 0.96 1.41 11.51
1.94 1.06 0.88 7.21 6.31 0.90 1.45 10.60
Includes emissions from Landfill. Data extracted from the NAEI [16].
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 3 UK emissions of 1,3-butadiene by source sector
185
186
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 4 UK 1996 emissions of 1,3-butadiene for specific industrial processes Industrial process
UK 1,3-butadiene emission (Tonnes)
Styrene–butadiene rubber manufacture Nitrile rubber production Polybutadiene manufacture ABS and SAN resinsa
751 38 170 100
a
‘ABS’ is acrylonitrile-butadiene-styrene; ‘SAN’ is styrene and acrylonitrile copolymer. Data extracted from the NAEI [16]
tion profiles’ are available to assign the emissions of VOC into emissions of individual components. The emission of 1,3-butadiene from tyre burning is not considered here, as the emissions are small (the UK emission is estimated at ca. 17 tonnes). 1,3-Butadiene is present in natural gas, and consequently there is an emission from gas pipe leakage. This has been omitted, as there is considerable uncertainty associated with current estimates.
3.1. Use in the chemical industry The primary use of 1,3-butadiene in the chemical industry is associated with the manufacture of polymers and rubber products. The presence of a double bond at either end of the C4 carbon chain (see Fig. 1) make it ideal for polymerisation reactions. The 1996 emissions of 1,3-butadiene from the different industrial processes are given in Table 4. Of the four different processes given in Table 4, styrene-butadiene rubber is produced in the largest quantities, and consequently generates the largest emission of 1,3-butadiene. Styrene– butadiene rubber (or SBS rubber, styrene– butadiene– styrene rubber) is a hard wearing synthetic rubber that is used in: tyres, hoses, industrial goods, shoes etc. [17]. It is manufactured by polymerising styrene and 1,3-butadiene. Atmospheric emissions arise from the storage of the feedstock, as well as the manufacture processes. The remaining three processes, production of: polybutadiene, nitrile rubbers, ABS and SAN resins, are similar in that 1,3-butadiene is used either in the production of an end product polymer, or is used to generate an intermediate compound. Emissions to the atmosphere arise from fugitive sources, as well as specific point locations.
3.2. Road transport Exhaust emissions of 1,3-butadiene from road vehicles account for ca. 70% of the total emissions in the UK as quoted by the NAEI. Earlier national emission estimates from vehicles were quoted as VOC, however, speciation into individual components has been developed by the NAEI over recent years making it possible to estimate emissions of 1,3-butadiene. The methodology for estimating emissions
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
187
from road transport vehicles is continually under review, but has been well established for VOC. As a result, it is expected that the speciation profiles give the largest potential for reducing uncertainties in estimates of 1,3-butadiene emissions in the short term. 1,3-Butadiene is not a major component in petrol or diesel, but is produced during the combustion process from higher olefins, hence evaporative emissions are not significant. The current emission estimates for 1,3-butadiene are given in Fig. 3. This indicates increasing emissions from petrol vehicles to 1990, followed by a steady decrease from 1990 onwards. Two factors are thought to have caused this feature. First, the annual petrol consumption in the UK steadily increased from in the late 1980s through until 1990. For the first several years of the 1990s, the petrol consumption decreased. This factors was also combined with the introduction of legislation in 1992, which required new petrol driven cars to be equipped with catalytic converters. As the penetration of catalytic equipped vehicles into the national vehicle fleet increased, the 1,3-butadiene emissions decreased. This increase in penetration is expected to continue for several years. To illustrate the impact that catalytic converters have on the 1,3-butadiene emissions from petrol driven road vehicles, several emission factors have been included below in Table 5. In addition, diesel vehicles have been included. Data from National Road Transport Emissions Model (pers. comm. Murrells 1999). It is evident from Table 5 that a significant reduction in the emissions from petrol driven cars has been achieved. Emissions from urban driving have been reduced by a factor of ca. 70, and emissions from cold starts have been reduced by a factor of ca. 10. Fig. 4 gives the current emission estimates and projections for 1,3-butadiene from the NAEI Road Transport Emission Model (pers. comm. Murrells 1999). It is evident that emissions are expected to fall steadily until ca. 2004. After which, Table 5 Selected emission factors (for different vehicle types and abatement technologies) 1,3-Butadiene emission factors (mg/km) Vehicle type
Petrol cars
Diesel cars
HGV
Abatement type
ECE 15.04 91/441/EEC Stage II Pre-Stage I Stage I Stage II Pre-Stage I Stage I Stage II
Road classification Urban
Rural single
Motorway
Cold start
37.74 1.14 0.5 4.95 3.75 2.62 32.27 16.14 14.52
16.50 0.00 0.00 4.38 3.35 2.35 41.32 26.86 24.79
8.70 0.08 0.03 2.47 1.44 1.01 12.73 9.55 8.27
35.80 3.07 3.07 11.99 11.99 11.99 53.25 53.25 53.25
188
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 6 Emissions from off-road transport and machinery Off-road transport and machinery
UK 1,3-butadiene emission (kTonnes)
Aircraft — civil and military Railways Marine, including coastal, fishing and naval Domestic machinery Agriculture machinery and vehicles Industrial- including airport support vehicles Landfill Total
0.08 0.15 0.28 0.45 0.08 0.37 0.03 1.45
increasing use of road transportation and attainment of the maximum market penetration for catalytic converters then results in a steady, but slow rise of 1,3-butadiene emissions from the road transport sector. Petrol cars remain the most significant vehicle type.
3.3. Off road transport and machinery Emissions of 1,3-butadiene in this sector arise primarily from off-road transportation and machinery in the industrial, domestic and agricultural sectors. Table 6 gives an indication of the contributions to the source sector total. Results have been taken from the NAEI for 1996.
4. Air quality guidelines in the UK In 1994, the UK Expert Panel on Air Quality Standards (EPAQS) published a report on 1,3-butadiene, considering not only current concentrations, but studies of human and animal exposure to elevated levels of 1,3-butadiene (1,3-butadiene, [18]). The study accepted that 1,3-butadiene is a genotoxic carcinogen, and consequently no safe level can be defined. However, the panel agreed on a recommended concentration that would ensure ‘the risks are exceedingly small and unlikely to be detectable by any practicable method.’ From the epidemiological studies that had been conducted, it was noted that there was considerable uncertainty, making the extrapolation of risk impossible. However, the panel recommended that ‘an Air Quality Standard for 1,3-butadiene in UK of 1 ppb measured as a running annual average’ be introduced. The panel further recommended that this standard be reviewed after a period of, at most, 5 years. The United Kingdom National Air Quality Strategy (NAQS) was published in 1997 and reviewed the standard and ambient concentrations of 1,3-butadiene. It was noted that the increased use of road vehicles equipped with catalytic converters was causing a reduction in the ambient 1,3-butadiene concentrations. It was considered ‘likely’ that by the year 2000, there would be no exceedances of the
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
189
EPAQS recommendation in urban background locations and ‘most roadside locations next to heavily trafficked roads’. In addition, it was noted that the current foreseen emission reductions would result in the EPAQS recommended 1 ppb level would be attained in ‘virtually all locations’. The government officially accepted the EPAQS recommendation, as a standard for 1,3-butadiene, and specified an objective of attaining this standard by 2005. The NAQS also noted that the EPAQS recommended standard would be reviewed by 1999, and further assessment would be conducted as appropriate. A review of the [19] was conducted in 1998 ([20] Review), and it was proposed that the deadline for attaining the 1 ppb standard adopted in the NAQS be brought forward to the 31 December 2003. This decision was based on modelling work conducted by [21,22] utilising the most current data available from the national monitoring networks (see below). 5. Monitoring in the UK
5.1. 1,3 -Butadiene automatic monitors Automatic measurement of 1,3-butadiene in the ambient atmosphere in the UK is generally conducted by gas chromatography. Typically, hydrocarbons in air are sampled and measured in the following manner: the air sample is passed through a bed of adsorbent material, which is cooled to increase the trapping efficiency and enable the removal of the more volatile species, such as 1,3-butadiene. The cold trap is flash heated to release the trapped components into a stream of inert carrier gas, in this case helium. The sample is carried onto a capillary column. The compounds pass along this column at different speeds (broadly depending on volatility). The temperature of the column is increased across the separation phase to ensure that the different components are suitably separated. This stage of the analysis will last typically for 40–50 min, however, 1,3-butadiene ‘elutes’ i.e. reaches the end of the column after ca. 10–15 min. Each component passes from the column to the detector. The detector typically used is a flame ionisation detector (FID), which gives an increased signal as each component passes. As a result, the output from the detector is a chromatogram consisting of a series of peaks, eluting at different times, and of different size. For a particular column type, temperature cycle and carrier gas flow, the elution time of each compound (and hence the order in which the components arrive at the detector) is well known typically from test samples consisting of single components. As a result, each peak in the chromatogram may be identified. The peak areas indicate the amount of each component in the sample. These peak areas are converted into mass values by applying ‘response factors’ i.e. the effective sensitivity of the detector to each of the components. These response factors are calculated from calibration samples, where a known amount of each component is introduced to the detector. The conversion from a mass to a concentration value is achieved from sample volume information.
190
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
The measurement of 1,3-butadiene is dominated by measurements conducted with automatic point monitors, giving compounds in the C2 to C8 or C9 range. A national network of these instruments has been established, however the expense of the equipment and running costs has meant that there are limited measurements with this type of detector from other sources. There have been relatively few measurements using the passive sampler technique, even though studies have indicated a good performance. It is thought that the low use of this technique has primarily arisen from a lack of end user awareness.
5.2. The UK hydrocarbon monitoring network Following the publication of the UK Government White Paper on the environment [23] the Department of the Environment set up its Enhanced Urban Monitoring Initiative. This was to establish comprehensive monitoring sites in city locations. The second stage of the initiative was the UK Hydrocarbon Monitoring Network. Two pilot studies were conducted at Exhibition Road, London (roadside location) and Middlesborough (urban background location). The first network site was commissioned in 1991, and by the end of 1995, there were 12 operational sites. In 1997, the Marylebone Road site was added to the network. Site information is indicated in Table 7. More detailed site information may be found in the ‘UK National Air Quality Archive’ (see reference list).
Table 7 The UK hydrocarbon monitoring network sites Site name
Site type
Inception date
Middlesbrough Belfast Birmingham Cardiff Edinburgh London, Eltham London, UCL Bristol Harwell Leeds Liverpool Southampton Marylebone road
Urban industrial Urban background Urban background Urban background Urban background Suburban Roadside Urban background Rural Urban background Urban background Urban centre Roadside
Pre 1993 1993 1993 1993 1993 1993 1993a 1994 1994b 1994 1995 1995 1997
a
This site was operational prior to 1993 as part of the pilot study. Although established in 1994, the site was used as a development platform resulting in an incomplete dataset until 1995. b
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
191
5.2.1. Data handling, ratification and quality control With numerous network sites continuously producing hourly measurements of more than 20 compounds, robust data ratification procedures are essential. Data processing and ratification procedures were developed to automate the reliable identification and quantification of the compounds. Pattern recognition software was used to ensure reliable compound identification. Particular attention was paid to the reliability of the 1,3-butadiene and benzene data, as national guideline values and standards exist for both compounds. Some of the procedures are on-line, however, some quality assurance and quality control procedures require detailed checking. Consequently ‘provisional’ data is made available on-line, but ‘ratified’ data are usually made available several months following the measurement. 5.2.2. Dissemination of information Provisional data are used to generate values of the rolling annual mean for 1,3-butadiene (and benzene) which are passed to television text services for on-line display. These pages are updated hourly, and the rolling annual mean value is accompanied by a ‘level’ indicator to aid interpretation of the concentration values. In addition, provisional hourly values of 1,3-butadiene (and benzene) are passed to the National Air Quality Archive http://www.aeat.co.uk/netcen/aqarchive/archome/ html. This archive is updated daily. All 1,3-butadiene concentration measurements are reported in ppb from the network and may be converted into m gm − 3 U by multiplying by ca. 2.25. 6. Ambient concentrations The UK Hydrocarbon Network has been operational for several years and has generated data of high quality. Thus, the large data set available from the network can be used to evaluate numerous different trends in the ambient 1,3-butadiene concentration.
6.1. Seasonal trends It is difficult to interpret seasonal trends from plots of hourly 1,3-butadiene concentration, due to the large variations from hour to hour. Consequently, to illustrate points more clearly the data presented in Figs. 5– 7 have been plotted as a rolling mean of 90 days. This averaging period retains general trends, but reduces the impact of short-term variations. The seasonal cycling of 1,3-butadiene concentration is caused by several factors. Evidently, the source strength is important when looking at variations over shorter time periods, but here the impact is reduced as the points are expressed as averages across a relatively long time period. If variations in the road traffic activity over long periods are considered to be small, then there are two factors, which affect the concentration- atmospheric mixing and chemical removal. The cycling of atmospheric mixing during a 24 h period is discussed in the following section. Evidently,
192 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 5. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means).
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 6. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means). 193
194 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 7. Seasonal trends in 1,3-butadiene concentration (90 day running mean of daily means).
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
195
Table 8 Annual decreases in concentration by site Site name
Belfast Birmingham Bristol Cardiff Edinburgh Eltham (London) Leeds Middlesbroughb UCL (London) Liverpoolc Southamptonc Marylebone Roadc
Concentration decrease (ppb per year)
0.025 0.027 0.018 0.018 0.0042 0.0025 0.0081 0.025 0.025 – – –
Mean concentration (ppb)
0.23 0.28 0.25 0.27 0.12 0.21 0.22 0.28 0.43 – – –
Concentration decrease (ppb per year) Daytimea
Night-timea
0.024 0.022 0.026 0.021 0.00057 0.0019 0.00032 0.023 0.0031 – – –
0.025 0.032 0.013 0.013 0.0012 0.0053 0.011 0.021 0.020 – – –
a Results calculated from deseasonalised time series. Daytime and night-time are defined as the hourly measurements from 07:00 to 18:00 h and 19:00 to 06:00 h, respectively. b Episodes at Middlesbrough have not been removed from the data, and consequently affect the calculation of the annual concentration decrease. c Not enough data exists from these sites to enable the trend analysis to be conducted.
the length of daytime periods varies with the seasons, and hence the mixing also varies. In the summer, effective vertical mixing is present for considerably longer than during winter days and the mixing depth is greater. As a result, there is a greater dilution effect during the summer. Table 1 indicates different rates of the chemical removal of 1,3-butadiene observed in the summer and the winter. Fig. 5 gives the rolling 90 day mean for the Eltham site. The Eltham site is situated in London away from localised sources, such as roads, and is classified as a suburban site (Tables 8 and 9). An annual pattern of concentration for this site is clearly present from mid 1994. Maximum concentrations are attained in October or November, and are ca. 0.37 ppb. Concentrations then decrease through December – March, reaching the annual minimum values, typically in April or May. These annual minima (expressed as 90 day averages) are ca. 0.13 ppb for the Eltham site. Concentrations then steadily increase through to the annual maxima observed in October or November. By comparison, the UCL and Marylebone Road sites are roadside sites, and consequently concentrations are dominated by the emissions from one particular section of road. The road transport activity on this single road section may be affected for long periods of time by factors such as roadworks, rail strikes etc. Hence, even when concentrations are expressed as 90 day averages, it is possible that there are periods which show significant deviation from a regular annual pattern of concentration cycling.
196
1,3-Butadiene
All road types
Urban roads
Rural and motorways
Year
Number of road links
Maximum concentration (ppb)
Number of road links
Maximum concentration Number of road (ppb) links
Maximum concentration (ppb)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Total
1616 1113 645 327 140 35 12 1 0 0 15 226
2.1 1.9 1.6 1.4 1.2 1.1 0.9 0.8 0.6 0.5
1288 876 493 249 104 30 12 1 0 0 7508
2.1 1.9 1.6 1.4 1.2 1.1 0.9 0.8 0.6 0.5
1.5 1.3 1.2 1.0 0.9 0.7 0.6 0.5 0.4 0.4
328 237 147 78 36 5 0 0 0 0 7718
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Table 9 Estimated exceedances of NAQS 1,3-butadiene standard [22]
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
197
The urban background sites shown in Fig. 8 (all sites except Harwell — the rural site) display similar trends to those noted for Eltham, except that the absolute concentrations are higher, ranging from ca. 0.1 to 0.4 ppb.
6.2. Diurnal trends Figs. 8 and 9 give plots for each site averaged in such as way as to represent the average diurnal cycle for each calendar year. These plots show several features of interest. First, they indicate that the concentrations are generally decreasing with time — this is considered in more detail below. UK Long-Term Trends from Measurements. One of the other features, which these plots are designed to show, are variations in concentration across a diurnal period. Years with a significantly reduced data capture have been removed. There are two major facts affecting the 1,3-butadiene concentration throughout the day. The first is the source strength, in most cases this will simply be the level of road traffic activity in the vicinity of the monitoring site. The second will be meteorological conditions, and in particular the mixing conditions. In general, vertical mixing increases throughout the morning reaching a maximum around midday or early afternoon. Vertical mixing then decreases, and during night-time hours is small. Thus the effective dilution of the 1,3-butadiene varies throughout a 24 h period. The diurnal trends for the Eltham site can be seen in Fig. 8 Eltham is a background site in London, and displays a typical diurnal variation. 05:00 – 09:00 h; concentrations of 1,3-butadiene starts to increase from 05:00 h, and rise sharply from 07:00 h as the traffic activity increases. Morning peak concentrations are reached at ca. 09:00 h, which corresponds to the peak activity on the roads (Transport Statistics Report). 10:00 – 14:00 h; from 10:00 h, the concentrations falls until 14:00 h, this correlates with decreasing traffic activity. In addition the vertical mixing increases from mid-morning (through until ca. 15:00–16:00 h). 14:00 – 21:00 h; the afternoon rush hour is spread over a longer period than the morning [24]. Vertical mixing increases until 15:00–16:00 h, and then decreases throughout the afternoon and evening. As a result, 1,3-butadiene concentrations steadily increase from 14:00 until ca. 21:00 h. 21:00 – 05:00 h; concentrations fall from 21:00 h through to 05:00 h due to the decreased traffic activity. Meteorological variations, particularly variations in the vertical mixing, have a clear impact on the 1,3-butadiene concentrations. This is illustrated by comparing typical concentrations at 03:00 and 15:00 h on Fig. 8. The concentrations are not dissimilar, but the expected traffic flows for these times will differ enormously. Other sites can be seen to display similar trends across an annually averaged 24 h period (Figs. 8 and 9), however, there are some particular features of note. The London UCL site is located at the roadside, and consequently concentrations are higher than those observed an urban background sites (Fig. 8). In addition, it is evident that the decrease of concentrations during the middle part of
198 G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
Fig. 8. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206 199
Fig. 9. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
200
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
the day are less pronounced. This may be attributed to the different patterns of traffic activity observed for Central London (UCL), and Outer London (Eltham). It is also evident that there is a substantial decrease in concentration from the 1997 data to the 1998 data. The Marylebone Road site (Fig. 8) is also located in Central London, but its traffic flows do not show good correlation with the general levels of the traffic activities in central London. This is because the traffic patterns in Marylebone Road are not typical of those in central London; here, there is no large decrease in traffic activities between the morning and afternoon rush hour periods (suggesting a congested road for much of the daytime). This is reflected in the concentration variations evident in Fig. 8. The Harwell monitoring site is a rural location, and consequently concentrations are low. In addition, as the site is some distance from a major road, the diurnal trend observed at e.g. Eltham is not present. The general trend of increased concentrations at night is caused by the changes in atmospheric mixing as earlier mentioned. In 1995, there was a significant ‘industrial’ release of 1,3-butadiene in the Middlesbrough area. After much publicity it was discovered that a sea tanker vented residual 1,3-butadiene to air (see below). The impact on the 90 day mean (Fig. 7) and the diurnal 1993 plot (Fig. 9) are evident. From hourly data, not shown here, there also appears to be an episode that occurred in 1998 (an hourly concentration value reaching ca. 60 ppb). This explains the behaviour of the 1998 trace in the diurnal plot (Fig. 9).
6.3. UK long-term trends from measurements The figures presented above give an indication that concentrations of 1,3-butadiene have been decreasing during recent years. It is possible to quantify the annual decrease in concentration by ‘deseasonalising’ the hourly data and then evaluating the decrease with time. The results from such an analysis are summarised in the following table for the various measurement sites. The annual decrease in concentration for many of the sites is comparable to ca. 10% of the annual mean concentration. Given that 1,3-butadiene emissions from road transport comprise the major source at most of these sites, the most likely cause of the reductions is the introduction of catalytic converters. It is evident from the figures that there is a particularly large decrease in concentration from 1997 to 1998 at several sites. If UCL is taken as an example location, then the percentage decreases in concentration observed from 1993 to 1997 are generally consistent with the percentage decreases observed in the national 1,3-butadiene emissions from urban areas up to 1997 (extracted from the road transport sector of the UK National Atmospheric Emissions Inventory (NAEI) — the order of 5– 10% per year. This is also in general agreement with modelled decreases of maximum concentration given below; this comparison is not yet possible with data from 1998 onwards. The observed decrease from 1997 to 1998 for UCL is a remarkable 40%. 1998 emission estimates for 1,3-butadiene have not yet been finalised by the NAEI.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206 201
Fig. 10. Annual diurnal mean concentrations for 1,3-butadiene at the sites indicated.
202
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
6.4. UK long-term trends from modelling The work by [22], in support of the NAQS, estimated future roadside concentrations of 1,3-butadiene. A relationship between 1,3-butadiene emissions and concentrations was indirectly established by using the well-characterised relationship between NOX roadside concentrations and NOX road transport emissions. The use of future emission estimates then enabled future 1,3-butadiene concentrations to be estimated. The following table gives estimates for the maximum roadside concentrations of 1,3-butadiene (expressed as fixed annual means) and the number of roadlinks exceeding the NAQS standard of 1 ppb (expressed as an annual running mean). It should be noted that due to the uncertainties associated with the method, and the different way in which the averages are calculated, values of maximum concentration (fixed means) which exceed 0.7 ppb were regarded as exceeding the NAQS target value of 1 ppb (annual running mean).
6.5. Spatial trends Figs. 10 and 11 give mapped 1,3-butadiene concentrations modelled by [22], for 1996 and 2005. The areas of high concentration are easy to identify, being London, Birmingham, Manchester, Liverpool and Leeds. Evidently, it is London that dominates both 1996 and 2005. Clearly this is in agreement with observations made from the data supplied by the UK Hydrocarbon Monitoring Network.
Fig. 11. Background concentrations of 1,3-butadiene (ppb)-1996; (b) background concentrations of 1,3 butadiene (ppb)-2005.
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
203
Table 10 Exceedences of the national air quality standard Maximum running annual mean Site
1999
1998
1997
1996
1995
1994
Belfast Birmingham Bristol Cardiff Edinburgh Harwell (rural site) Leeds Liverpool London Eltham London Marylebone London UCL Middlesbrough Southampton
0.3 0.4 0.4 0.5 0.2 0.1 0.4 0.4 0.4 2.4 0.6 0.3 0.6
0.5 0.5 0.5 0.5 0.2 0.2 0.5 0.5 0.5 2.5 0.9 0.5 0.7
0.5 0.5 0.5 0.7 0.2 0.2 0.5 0.5 0.5
0.5 0.7 0.7 0.5 0.2 0.2 0.5 0.5 0.5
0.5 0.5 0.5 0.7 0.2
0.7 0.7
0.9 0.5 0.9
0.9 0.7 0.9
0.9 0.7
0.7 0.2
0.5 0.7 1.1
6.6. Middlesbrough tanker episode To date the most significant episode was measured at Middlesbrough in 1995. In the afternoon of 31 July 1995, 1,3-butadiene hourly concentrations rose to exceptionally high levels, peaking at over 80 ppb. Concentrations then remained high until the early hours of the following morning. High concentrations were also noted overnight from 1 to 2 August and in the afternoon of 2 August. There then followed consultation with the local authority (and a degree of press coverage) in an effort to locate the source. It came to light that a marine tanker had vented residual 1,3-butadiene after making a delivery. Venting of the tanks to atmosphere is a standard operation to empty the tanks of any residual contents, prior to loading of a different compound. Venting is usually conducted at sea, however on this particular occasion it was conducted at harbour during a period of unfavourable wind direction. This type of chemical release was, at the time, not covered by any legislation. The impact of the release on the rolling annual mean, of ca. 0.18 ppb, was to cause an increase the order of 50%. Such episodes are an issue concerning local populations. Fig. 9 indicates that another episode occurred at Middlesbrough during 1998. An individual hourly concentration exceeded 60 ppb on 14 May 1998, but the origin of this elevated concentration is not known.
6.7. Exceedences of the national air quality standard Table 10 summarises the maximum annual running mean for all of the network monitoring sites in the UK hydrocarbon network from the point 1 year of data became available.
204
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
This summary shows that the majority of network sites are below the national standard for 1,3-butadiene (1.0 ppb). Exceedences have been recorded in the two most recent years at the Kerbside monitoring site on London’s Marylebone Road (very heavy traffic) and at the London UCL site in 1994. The London UCL site samples air about 6 m from the side of a moderately busy roadway. At both of these sites, the main emission source is motor vehicle exhaust.
7. Conclusions 1,3-butadiene has been accepted as a genotoxic carcinogen (1,3-butadiene, [18]), and consequently much work has been conducted in establishing 1,3-butadiene emissions, concentrations, and the behaviour of the compound in the atmosphere. 1,3-butadiene is highly reactive under atmospheric conditions, with an estimated atmospheric lifetime varying from several minutes to months. The UK NAEI have quantified emissions of 1,3-butadiene, and report a total emission of 10.60 kTonnes for 1996. The total UK emission is noted to be decreasing. This is primarily caused by the reduction in emissions from road vehicles as the penetration of catalytic converters into the vehicle fleet increases. For 1996, 1,3-butadiene emissions are dominated by road vehicle exhaust emissions (comprising some 68% of the total emission); whilst emissions from off-road vehicles and machinery account for 14%. The remaining emissions arise from the chemical industry, and are from 1,3-butadiene manufacture and use in the production of various rubber compounds. These two processes account for 8 and 10% of the UK 1996 total emission, respectively. Work conducted by EPAQS (‘1,3-butadiene’, 1994) recommended a National Air Quality Standard of 1.0 ppb (expressed as an annual rolling mean). This was adopted by the government in 1997 as part of the NAQS. More specifically, the air quality objective that was set for 1,3-butadiene was to achieve the 1.0 ppb standard by 2005. Of the monitoring sites on the UK Hydrocarbon Monitoring Network, all have indicated compliance with this standard for some time except the Marylebone Road site. Modelling studies conducted by [22] as part of the NAQS review estimated all roadside locations to be compliant with the 1,3-butadiene standard by the end of 2003. As a result, the official air quality objective for 1,3-butadiene was amended. The deadline for compliance now stands at achieving the standard by 31 December 2003, as specified in the NAQS review. Occupational exposure limits are currently quoted as a LTEL of 10 ppm, although this is currently under review. The UK Hydrocarbon Monitoring Network has generated a comprehensive time series of 1,3-butadiene concentrations across the UK since 1993. This dataset is unparalleled and has given an insight into temporal and spatial trends of 1,3-butadiene concentrations across the UK. The data has also allowed comparisons to be made with modelled concentrations, and emission estimates. The UK Hydrocarbon Monitoring Network provides continuous hourly measurements of 1,3-butadiene at 13 sites, and has been operational since 1993. The
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
205
dataset that is available allows spatial and temporal trends to be evaluated, and has proved to be invaluable in characterising the current ambient levels of 1,3-butadiene in the UK. Hourly maximum concentrations of 1,3-butadiene of up to 10 ppb may be measured for several hours at the sites. Monthly mean concentrations are typically 0.1– 0.4 ppb. At most sites, these levels are driven by emissions from motor vehicles. Occasionally emissions of 1,3-butadiene from industrial sources may elevate 1,3-butadiene concentrations to several tens of ppb. Trend analysis of the data suggests that ambient concentrations of 1,3-butadiene in the UK are declining at about 10% per year.
Acknowledgements The authors would like to thank the following colleagues at AEA Technology Environment for their valuable contributions and input into this report: Brian Jones, Trevor Davies, Tim Murrells and Justin Goodwin. The work was funded as part of a programme of research managed by the Air and Environmental Quality Division of the UK Department of the Environment, Transport and the Regions.
References [1] UATW — Unified Air Toxics Website published by the USEPA (1998). http://www.epa.gov/ ttnuatw1/hlthef/butadien.html. [2] EH40/97 Occupational Exposure Limits, 1997. Health and Safety Executive Publications, Sheffield. ISBN 0-7176-1315-1. [3] J.P. Shi, R.M. Harrison, Rapid NO2 formation in diluted petrol-fuelled engine exhaust: a source of NO2 in winter smog episodes, Atmos. Environ. 31 (1997) 3857 – 3866. [4] R.M. Harrison, J.P. Shi, J.L. Grenfell, Novel nighttime free radical chemistry in severe nitrogen dioxide pollution episodes, Atmos. Environ. 32 (1998) 2769 – 2774. [5] PORG (1997). Ozone in the United Kingdom: Fourth report of the Photochemical Oxidants Review Group. Department of the Environment, Transport and the Regions, London. [6] R. Atkinson R. Gas-phase tropospheric chemistry of organic compounds. J.Phys.Chem.Ref. Data (1994) Monograph 2. [7] R. Atkinson, Gas-phase tropospheric chemistry of volatile organic compounds: 1 alkanes and alkenes, J. Phys. Chem. Ref. Data 26 (1997) 215 – 290. [8] M.E. Jenkin, S.M. Saunders, M.J. Pilling, The tropospheric degradation of volatile organic compounds: a protocol for mechanism development, Atmos. Environ. 31 (1997) 81 – 104. [9] M.E. Jenkin, A.A. Boyd, R. Lesclaux, Peroxy radical kinetics resulting from the OH-initiated oxidation of 1,3-butadiene, 2,3-dimethyl-1,3-butadiene and isoprene, J. Atmos. Chem. 29 (1998) 267–298. [10] A. Maldotti, C. Chiorboli, C.A. Bignozzi, C. Bartocci, V. Carassiti, Photooxidation of 1,3-butadiene containing systems: rate constant determination for the reaction of acrolein with OH radicals, Int. J. Chem. Kinet. 12 (1980) 905 – 913. [11] T. Ohta, Furan ring formation in OH-initiated photooxidation of 1,3-butadiene and cis-1,3-pentadiene, Bull. Chem. Soc. Jpn. 57 (1984) 960 – 966. [12] X. Liu, H.E. Jeffries, K.G. Sexton, Hydroxyl radical and ozone initiated photochemical reactions of 1,3-butadiene, Atmos. Environ. 33 (1999) 3005 – 3022.
206
G.J. Dollard et al. / Chemico-Biological Interactions 135–136 (2001) 177–206
[13] I. Barnes, V. Bastian, K.H. Becker, Z. Tong, Kinetics and products of the reactions of NO3 with monoalkenes, dialkenes and monoterpenes, J. Phys. Chem. 94 (1990) 2413 – 2419. [14] H. Skov, J. Hjorth, C. Lohse, N.R. Jensen, G. Restelli, Products and mechanisms of the reactions of the nitrate radical with isoprene, 1,3 butadiene and 2,3 dimethyl-1,3 butadiene in air, Atmos. Environ. 26A (1992) 2771 –2783. [15] NAEI– UK Emissions of 1,3-Butadiene pers. comm. Goodwin J, (1999). [16] A.G. Salway, H.S. Eggleston, J.W.L. Goodwin, J.E. Berry, T.P. Murrells, UK Emissions of Air Pollutants 170-1996. Published by the NAEI for the DETR (1999). [17] Chem-Facts United Kingdom, Edition. Chemical Intelligence Services. Chrompack VOC-AIR Instrument Manuals (1991). Available from Chrompack. [18] 1,3-Butadiene (EPAQS) 1994. HMSO, London. Also available at http:// www.environment.detr.gov.uk/airq/aqs/13butad/but01.htm.. [19] UK NAQS Review: a Consultation Document, 1998. Published by Department of the Environment, Transport and the Regions. [20] UK NAQS- The United Kingdom National Air Quality Strategy, 1997. Published by The Stationary Office Limited, London. [21] J.R. Stedman, Revised High Resolution Maps of Background Air Pollutant Concentrations in the UK: 1996. AEA Technology AEAT-3133, published for the DETR (1998). [22] J.R. Stedman, C.J. Dore, An Empirical Model for Quantifying Current and Future Benzene and 1,3-Butadiene Concentrations at the Roadside. AEA Technology AEAT-4301, published for the DETR (1999). [23] This Common Inheritance, Government White Paper, HMSO, London, 1990. [24] Transport Statistics Report: London Traffic Monitoring Report: 1992. Department of Transport, HMSO 0-11-551137-7, 1992.
.