emissions of polycyclic aromatic hydrocarbons and their oxy-and nitro-derivative compounds measured in road tunnel environments

• Polycyclic aromatic compounds including PAH, oxyand nitro-derivatives • Includes several compounds not previously measured in road tunnels • Measured in road tunnels and urban


25
• Polycyclic aromatic compounds including PAH, oxy-and nitro-derivatives 26 • Includes several compounds not previously measured in road tunnels 27 • Measured in road tunnels and urban background 28 • Large decline in PAH, but not nitro-PAH since 1992-96 29 • 1-Nitropyrene promising specific marker of diesel exhaust  The World Health Organisation (WHO) has recommended guidelines for concentrations of BaP 75 producing excess lifetime cancer risks of 1/10 000, 1/100 000 and 1/1 000 000 of 1.2, 0.12 and 76 0.012 ng m -3 , respectively (WHO, 2000). It is estimated that 20% of the population of the EU is 77 exposed to BaP levels higher than the EU target of 1 ng m -3 and 88% is exposed to levels higher 78 than the reference level of 0.12 ng m -3 (EEA, 2015).  These studies therefore suggest that NPAH and OPAH may pose more toxic hazard in the urban 120 environment than PAH.  There is therefore a need accurately to assess the relative and overall contribution of road traffic  However, accurately measuring on-road vehicular emissions is complicated by the mixture of 135 engine and fuel types and emission control technologies present. Furthermore, experimental studies 136 in a laboratory will only yield data on specific vehicles, engine characteristics and/or fuel 137 formulations (e.g. Zielinska et al., 2004a,b).

139
Sampling in road tunnels is expected to provide a more realistic traffic profile than using 140 dynamometer tests (Oda et al., 2001;Wingfors et al., 2001). Road tunnels provide the advantage of 141 a realistic distribution of on-road vehicles ; relatively low rates of dispersion and chemical reactivity 142 ; and a lack of inputs from other primary sources. Additionally, repeated monitoring in a tunnel 143 environment can assess historic changes in PAH emission profiles in response to changes in fuel 144 usage or emission control measures (Ravindra et al., 2008).  In the present investigation, the total (gas + particle phase) concentrations of selected PAHs, 156 OPAHs and NPAHs were measured in a road tunnel environment in central Birmingham, U.K and 157 simultaneously at an urban background site in southwest Birmingham. Comparative measurements 158 (only particle phase) were performed in a tunnel of the Paris ring road (France). The specific aims 159 were to i) obtain realistic 'traffic signatures' for these compounds; ii) compare investigated tunnel 160 concentrations and chemical profiles in relation to the national vehicle fleet compositions; iii) 161 compare these tunnel concentrations with those observed in the ambient urban environment ; and 162 iv) assess the temporal trend of these compounds in the tunnel and relate this to observed changes in 163 traffic characteristics during this period. and external ring roads ( Figure S5). Mean vehicle speed was slightly higher for the external ring 207 road than for the internal one (69 and 57 km h -1 , respectively) and inversely for tunnel fill rate (12 208 and 18%, respectively). The latter is a measure of congestion and indicates freely flowing traffic.

312
Quantification of OPAHs and NPAHs was performed by internal standard calibration using native 313 standards and deuterated surrogate compounds added before extraction (Table S3 and S4). 9-

314
Fluorenone-d9 and 1-nitropyrene-d9 were used as surrogate standards. proportion of each compound present in the particle-phase for QT, are presented in Table 1.

332
Concentrations of additional NPAHs and OPAHs analysed in PdPT samples are presented in Table   333 S5. Compounds below quantification limit are not reported in the table.

436
The phase partitioning behaviour of PAH, OPAH and NPAH in QT and ambient measurements are 437 different (Table 1).

439
For most LMW 3-ring compounds and HMW 5+ ring compounds, the proportion of compounds in

469
Comparing levels of compounds measured in the QT with those observed at the background site 470 EROS (tunnel/ambient ratios) will allow the assessment of other influencing factors (e.g. non-traffic 471 sources, relative rates of atmospheric loss processes) to their overall and relative concentrations in 472 the urban atmosphere. The tunnel/ambient ratios measured in the present study are shown in Figure   473 3. It can be seen that these ratios vary considerably for different individual compounds.  The tunnel/ambient ratio of Acy is considerably higher than those of other PAHs. This is consistent     The principal atmospheric loss process for NPAHs is expected to be photoreactivity including direct  The results of the present study suggest 1NPyr is degraded more rapidly than other NPAHs. While  This previous study was conducted in central Birmingham during winter. The QT/EROS ratio in the 583 present study is shown to be a factor ~10 and ~4.5 higher than the previous study for 1NPyr and 584 9NAnt respectively. This may partly be attributed to higher input of pollutants in the city centre 585 compared to the background EROS site and the fact that sampling in the present study was 586 conducted in the late summer leading to potentially faster rates of photolytic degradation in the 587 ambient atmosphere.

589
The tunnel/ambient ratio of 1NNap is a factor ~2.2 higher than that of 2NNap. Experimental studies    Figure S3).

657
The trend in the number of gasoline and diesel vehicles on the road is reflected in the volume of 658 gasoline and diesel fuel supplied to the U.K market (see Figure S4) and national fuel sales. In the The results of the present study therefore suggest that the temporal variation in PAH and NPAH