Particulate organic nitrates

PII:

Atmospheric Environment Vol. 32, No. 14/15, pp. 2601—2608, 1998 ( 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain S1352–2310(97)00483–4 1352—2310/98 $19.00#0.00

PARTICULATE ORGANIC NITRATES: SAMPLING AND NIGHT/DAY VARIATION TORBEN NIELSEN,*,- JESPER PLATZ,- KIT GRANBY,‡ ASGER B. HANSEN,‡ HENRIK SKOV‡ and AXEL H. EGEL"V‡ -Chemical Reactivity, Department of Environmental Science and Technology, Building 313, Ris+l National Laboratory, P.O. Box 49, DK-4000 Roskilde, Denmark; and ‡National Environmental Research Institute, P.O. Box 358, DK-4000 Roskilde, Denmark (First received 3 March 1997 and in final form 15 October 1997. Published June 1998) Abstract—Atmospheric day and night concentrations of particulate organic nitrates (PON) and several other air pollutants were measured in the summer 1995 over an open-land area in Denmark. The sampling of PON was evaluated comparing 24 h samples with two sets of 12 h samples. These results indicate that the observed low contribution of PON to NO is real and not the result of an extensive loss during the y sampling. Empirical relationships between the vapour pressure and chemical formula of organic compounds were established in order to evaluate the gas/particle distribution of organic nitrates. A positive correlation between the PON to NO ratio and the inverse temperature may indicate that the organic ; nitrates associated with particles also may be present in the gas phase. The observed day and nighttime levels of PON were of the same magnitude, although the day to night ratio varied from 0.4 to 4.4. ( 1998 Elsevier Science Ltd. All rights reserved Key word index: Organic nitrates, atmospheric chemistry, daytime, nighttime, sampling, particles, vapour pressure.

1. INTRODUCTION

The atmospheric formation of ozone over the open land is controlled by the presence of NO at least x during the summertime (Finlayson-Pitts and Pitts, 1996; Nielsen et al., 1996). Bifunctional organic nitrates have been found in the gas phase as well as unidentified bi- and/or multifunctional organic nitrates associated with particles in smog-chamber experiments and are formed by reactions of alkenes with OH radicals at day-time (Palen et al., 1993; Tuazon and Atkinson, 1990) and with NO3 radicals at nighttime (Atkinson, 1991; Hoffmann et al., 1997; Skov et al., 1992). These compounds are considered to be a possible source of NO2 in remote areas (Wangberg et al., 1996) and they may thus affect the formation of ozone from the photooxidation of CO and CH4 (Ayers et al., 1992). Recently, some gas phase C2—C4 hydroxynitrates and dinitrates have been identified. Daytime processes were estimated to be the major source of these compounds (O’Brien et al., 1995, 1997). Organic nitrates are suspected of making up a significant part of the so-called missing NOy. Concurrent

* Author to whom correspondence should be addressed.

measurements of NO and the individual nitrogen y species (NO, NO , HNO , PAN, PPN and inorganic 2 3 nitrate) over the open land and background areas indicates that the unidentified fraction is 6—44% of the total NO (Crosley, 1996; Fahey et al., 1986; Nielsen et y al., 1995; Parrish et al., 1993). The atmospheric levels of the particulate organic nitrates (PON), i.e. organic nitrates associated with particles, have earlier been determined to be 14$5 ng (N) m~3 (0.25% of total NO ) in over an open area in Denmark (Nielsen et al., y 1995) and 45 ng(N) m~3 in Los Angeles air (Mylonas et al., 1991). Both investigations applied the IR absorption features of the nitrate groups for the determination. In order to determine whether the low contribution of PON to NO is real or caused by extensive y loss during the sampling, the possible influence of sampling artefacts has been studied. The day and night levels are compared. The measurements were performed in the summer 1995. The sampling site was over open land at a location relatively close to NO sources. The results are also compared with x the previous results from the spring and summer 1993 (Nielsen et al., 1995). Also, the relation between vapour pressure and structure and the temperature effect on the gas/particle distribution is discussed.

2601

T. NIELSEN et al.

2602 2. EXPERIMENTAL

2.1. Site description and meteorological conditions The measuring station is located in a horizontally homogeneous agricultural field in a typical Danish farming area on Zealand. The geographical co-ordinates are 12°07@07AE, 55°41@42AN and the altitude above sea level is 15 m. The nearest town is Roskilde (40,000 inhabitants), located 6 km SSW of the site. The Danish capital, Copenhagen (850,000 inhabitants), is located 35 km to the east. Located 2 km west of the site is Roskilde inlet. A highway with traffic load of appr. 6000 cars each day runs in the north-south direction along the coast. The terrain altitude variation is less than 5 m in a radius of 1 km from the site. Maps of the area have been presented by Nielsen et al. (1981 and 1994). The vegetation of the surroundings includes grass, barley, rye, wheat, rape, potato plants, bushes and different deciduous and coniferous trees, although the number of coniferous trees is small. The particulate organic nitrates (PON) samples were collected mainly in warm and sunny periods in May—July 1995. The winds originated from the west sector half of the time and otherwise from east and southeast in the other cases. The wind speed was typical for Danish conditions at about 4 m s~1. 2.2. Particulate organic nitrates Sampling: 12- and 24 h samples were collected on glass fiber filters with Hi-Vol samplers and stored in a freezer (!18°C) until analysis. The sampling volumes were typically 750 m3 for 12-h samples and 1500 m3 for 24 h samples. The 24 h samples were collected in the period 0700—0700, and the 12 h samples in the periods 0700—1900 and 1900—0700. The 0700—1900 samples are called the day-samples and the 1900—0700 samples are called the night-samples in Section 3, although the period from sunset to sunrise was not more than 61—71 h during this investigation. However, the 2 reasonable, as the global radiation (see appellations 2appear Table 2) during the 12 h night period was only 7% of that during the 12 h day period. 33 samples were collected in the period 23 May—21 July 1995. Two of the samples were collected in May, 11 were collected in June and 20 were collected in July. 23 of the 33 samples were 12 h and 10 were 24 h. Results from nine sets of two 12h samples and one 24 h were collected. Extraction and analysis: The filters were extracted three times, 30 min each time, with 100 ml chloroform (Merck p.a.) in an ultrasonic bath. 7 ml tetrachlorethylene (Merck, Uvasol) were added to the concentrated extracts, and the mixture was concentrated further to 5 ml. The C Cl solu2 infrared 4 tions were analysed using Fourier transformation spectroscopy (FTIR, Perkin-Elmer Spectrometer 1760X). Each sample was analysed with 1000 scans. The peak area between 1617 and 1663 cm~1 was determined representing the total amount of organic nitrates in the sample. A typical spectrum of a sample has been published elsewhere (Nielsen et al., 1995). C Cl solutions of different organic ni2 4 (Fluka, 95%), 3-nitrooxy-2-butanone trates, i-propyl nitrate and 3-nitrooxy-2-butanol, were used for calibration. By comparing the measured sample areas to the standard areas the concentrations of PON in the samples were calculated. 2.3. Other components Nitric acid and particulate inorganic nitrate were collected on NaCl coated denuders (0.4 mm i.d.* 50 cm, 12 h sampling time, flow 1 l min~1) followed by NaF coated filters, and determined by ion chromatography with UV detection (Nielsen et al., 1995; Skov et al., 1997). PAN was measured by gas chromatography with electron capture detection (Nielsen et al., 1982). O was measured by a UV-absorption 3 monitor, NO and gas NO by a two chamber chemiluminesy

cence monitor, SO by UV-fluorescence (Nielsen et al., 1994). 2 twelve hour samples of formic acid, acetic acid and methane sulphonic acid were collected by filterpacks. The particles including MSA were collected on a Teflon filter in the front and the gas phase fraction on two back-up cellulose filters impregnated with Na CO . The analyses were performed by 2 3 means of ion-chromatography (Granby et al., 1997). The sampling of VOC (C —C ) was performed with three 4 h 2 8 (07—11, 11—15 and 15—19) and one 12 h samples (19—07) in the campaigns. The samples were analysed by trapping on activated carbon, thermal desorption, cryofocussing and GCFID analysis (Hansen and Palmgren, 1996). The meteorological station determined wind speed, wind direction, temperature, temperature difference (1.5 and 9 m), humidity, global radiation, pressure and precipitation every 30 min with conventional micro-meteorological instruments (Nielsen et al., 1994).

3. RESULTS AND DISCUSSIONS

3.1. Possible identity, vapour pressure and gas/particle distribution Organic nitrates may be formed by three pathways in the atmosphere (Flocke et al., 1991; Tuazon and Atkinson, 1990; Wayne et al., 1991): (1) Reactions of peroxy radicals with NO. The reaction is a source of monoalkyl nitrates. (2) Reactions of alkenes with HO radicals, formation of a hydroxyperoxy radical and subsequent reaction with NO. The reaction is a source of hydroxynitrates and smog-chamber experiments with isoprene, b-pinene and 1-octene have shown formation of PON (Palen et al., 1993). (3) Reactions of alkenes with NO , formation of 3 a nitrooxyperoxy radical and subsequent oxidation to an organic hydroxy nitrate or oxonitrate compound. The gas-phase derivatives of, e.g. isoprene have been identified in smog-chamber experiments (Skov et al., 1992). In ambient air a number of different C —C alkyl 1 17 nitrates have been identified in gas phase (Flocke et al., 1991; Luxenhofer et al., 1996). Furthermore, O’Brien et al. (1995, 1997) have identified 6 C —C 2 4 hydroxynitrates and 1,2-dinitrooxybutane in the gas phase. Luxenhofer et al. (1996) were unable to identify alkyl nitrates associated with particles. The gas/particle distribution of organic compounds depends on the vapour pressure of the compound. In general, there is a strong need for physical/chemical parameters of organic pollutants in order to estimate their fate in the environment and a number of papers have shown that it may be possible to provide relationships between these parameters and chemical structure of the compound. These parameters include, e.g. water solubility, octanol/water partition coefficients and sorption coefficients (Dickhut et al., 1994; Helweg et al., 1997; Nielsen et al., 1997a; Schwarzenbach et al., 1993). The important recent work of Hoffmann et al. (1997) illustrates, that there also is a strong need for relations between the vapour pressure and the chemical structure.

Particulate organic nitrates: variations

The following relationship describes the influence of the number of carbon atoms, nitrate, hydroxy and carbonyl group on the vapour pressure of the compound in liquid phase at ambient temperature: log P"!(0.4069$0.0057)]no. of C-atoms!(2.144 $0.070)]no. of nitrate groups!(1.961$0.057)]no. of hydroxy groups!(1.130$0.071)]no. of carbonyl groups #(4.466$0.077), n"183, r"0.98, p(0.001 (see Fig. 1). The applied data set for the correlation includes C —C hydrocarbons, C —C alcohols, 7 29 1 18 C —C diols, C —C carbonyls, C —C alkyl ni2 10 5 18 1 20 trates and C —C hydroxynitrates and dinitrates 2 3 (CRC, 1982; Hallquist et al., 1996; Kames and Schurath, 1992; Luxenhofer et al., 1996; Saxena and Hildemann, 1996). Most of the experimental data were achieved by extrapolating to 20°C. The data from Luxenhofer et al. (1996) and Saxena and Hildemann (1996) were at 25°C. The calculated P values of pinone- and caronealdehyde and C -C 8 20 alkyl nitrates agreed with 56$20% with those recently determined experimentally by Hallquist et al. (1996) and Luxenhofer et al. (1996). The relation can also be applied to compounds being in solid phase in their pure state at ambient temperature, because the gas-particle partitioning in ambient air is determined by the vapour pressure of the sub-cooled liquid and not by the sublimation pressure of the solid compound (Pankow and Bidleman, 1992). The correlation includes 183 compounds having P varying 10 orders of magnitude. Most of the P values have been attained

2603

by applying the Clausius—Clapeyron equation (Schwarzenbach et al., 1993) predicting an inverse relation between ln P and ¹: ln P"!*H](R¹)~1#A where *H is the heat of vapourisation and A is a constant. *H appeared to be independent of the temperature, as the mean r$2p was 0.9996$0.0003 for the ln P!¹~1 correlations. The relation between the vapour pressure at ambient temperature and the chemical structure has been applied to calculate the vapour pressure of different organic nitrates (see Table 1). Organic nitrates are a group of compounds which may have great variations in their vapour pressure varying 12 orders of magnitude from C alkyl 5 nitrates to C dihydroxynitrates. Thus, the class of 25 organic nitrates will contain compounds, which are only present in ambient air in the gas phase, some which is associated with particles and some will be partly gas phase and particle associated. At equilibrium, the gas-to-particle distribution is dependent on the vapour pressure, temperature and particle concentration (Pankow, 1987, 1994; Pankow and Bidleman, 1992), but the nature of the particles probably also affects the distribution. As a matter of fact, the distribution of polar organic compounds between water and suspended solid matter is affected by polar interactions between functional groups in the substrate and the compound itself (Nielsen et al., 1997a). A strong inverse relationship between vapour pressure and particle proportion values exists for nonpolar compounds in typical ambient conditions using field data including alkanes, PAH and organochlorine compounds (mainly PCB). The following relation have been derived (Finizio et al., 1997; Pankow and Bidleman, 1992): F]A~1"10~b 3]TSP]P~1

Fig. 1. The figure shows the relation between the vapour pressure of organic compounds (x-axis) and the vapour pressure calculated from their chemical structure (y-axis) applying the equation: log P"!0.4069]no. of C-atoms !2.144]no. of nitrate groups !1.961]no. of hydroxy groups !1.130]no. of carbonyl groups #4.466.

where F is amount of the compound associated with particles, A the amount in gas phase, TSP the concentration of particles (kg m~3) and b a constant depen3 dent on the temperature. At 20°C an appropriate value of b is !8 (Finizio et al., 1997; Pankow and 3 Bidleman, 1992). Thus, the F/A ratio is smaller than 0.01 if P'2]10~5 mm Hg and TSP"20 kg m~3, i.e. for P'2]10~5 mm Hg one can consider the compound to be exclusive in the gas phase. Correspondingly,

Table 1. Estimated vapour pressure (P, mm Hg) of different organic nitrates at 20°Ca No. of carbon atoms 5 10 15 20 25

Dihydroxynitrates

Hydroxynitrates

Carbonylnitrates

Alkylnitrates

2]10~4 1]10~6 1]10~8 1]10~10 1]10~12

0.02 2]10~4 2]10~6 2]10~8 2]10~10

0.2 0.002 2]10~5 2]10~7 2]10~9

3 0.03 3]10~4 2]10~6 2]10~8

! The expression for the calculation of P is shown in the legend to Fig. 1.

2604

T. NIELSEN et al.

one can consider the compound to be exclusively associated with particles, if P(2]10~9 mm Hg. Although the extrapolation of this data set to organic nitrates should be done cautiously, the set and Table 1 confirm, that C alkyl nitrates (P"3] 17 10~5 mm Hg), C hydroxynitrates (0.05 mm Hg) and 4 C dinitrates (0.04 mm Hg) should be present in the 4 gas phase in ambient air. From the data set above and Table 1, it can be estimated that higher alkyl nitrates ('C ) may contribute to the presence of PON com18 pounds. However, if they are present in ambient air, it should be possible also to identify these compounds in the gas phase, as P"1]10~8 mm Hg for C alkyl 25 nitrates. The PON compounds may also be bi- and multifunctional organic nitrates. For bifunctional organic nitrates the number of carbon has to be at least 12. For nitrates with three functional groups the number of carbon atoms has to be at least 7. Therefore, it will be very difficult, if not impossible, to determine the exact chemical structure of bi- and multifunctional organic nitrates present as PON in ambient air considering the low amounts of PON, the large number of possible isomers and difficulties in preparation and handling of bi- and multifunctional organic nitrates. A major VOC source for PON formation may be biogenic emissions, terpenes (C ) and sesquiterpenes 10 (C ) (Calogirou et al., 1996; Ko¨nig et al., 1995). Diesel 15 exhaust emissions of alkenes may also be a source for VOC precursors. The content of alkenes in diesel fuel is in the range 0.2—2.2% (Siegl and Wallington, 1996). Table 2 shows the night and day mean concentrations of C —C alkenes, C —C alkanes and the traffic 2 5 2 8 VOC, toluene. Nielsen et al. (1997b) have shown that most of the alkenes had originated from biogenic sources. This also indicates that the emissions of terpenes and sesquiterpenes are important (Ko¨nig et al., 1995). The measured levels of ethylene and the other alkenes were 6—20 times higher than the modelled values, while the modelling of the traffic VOC, toluene, agreed with the measured values. Also, the C —C alkanes had a much lower ratio of the meas2 8 ured to the modelled values (2.8) than the alkenes. The conclusion of Nielsen et al. (1997b) is that the emission inventories underestimated the biogenic emissions of alkenes and this implied that biogenic emissions were the major source of alkenes. 3.2. Sampling validation As discussed in the previous section some organic nitrates may be present both in the gas phase and associated with particles (Nielsen et al., 1995). Physical as well as chemical processes may interact in the sampling of these compounds. Physical losses by evaporation of the organic nitrates from sampled particles may affect the results (Finlayson-Pitts and Pitts, 1986). Another possible error is chemical decomposition of PON during sampling. In order to validate the sampling technique the results from 12 h and 24 h samples were compared. However, first the two

Hi-Vol samplers were intercompared by collecting seven sets of identical 24 h samples and comparing the amounts of collected sulphate. This test was satisfactory as the amounts of sulphate agreed to within 1—2%. The comparison of PON in 12 h and 24 h samples is shown in Fig. 2. On average the content in the 24 h samples was 85$18% ($2p) of the content of PON in the set of two sequential 12 h samples covering the same time interval. The variation in the data set was investigated most thoroughly with respect to variations in the meteorological parameters and the air pollution parameters, but no systematic correlation was found. Thus, it seems to be accidental that the early sample set is much closer to equivalence between 12#12 and 24 h than the last four. The results indicate that the low PON to NO ratio found y previously (Nielsen et al., 1995) and in this investigation (see Table 2) is real and not caused by an extensive loss during sampling. Otherwise, the ratio should be much lower than 85%. Also, the samples were collected in connection with warm and sunny weather. It is believed that the risk for blow-off increases with the temperature and the risk for chemical degradation increases with the presence of photochemical oxidants. Therefore, the comparison should represent a worst case situation for Danish conditions. 3.3. PON occurrence and day/night variation The measured PON made up, as mentioned earlier, a minor proportion of total NO (0.17—0.28%, see y Table 2). The mean PON levels in the day (11$3 ng (N) m~3) and night samples (10$4 ng (N) m~3) in 1995 and in the 24 h samples from 1993 (14$5 ng (N) m~3) were not significantly different (t-test, p"0.16—0.26). As discussed in Section 3.1 the organic nitrates contributing to PON is believed to consist of semi-volatile species. The particle to gas ratios of these species are expected to increase with decreasing temperature. The correlation between the ratio PON (kg (N) m~3) to NO (kg (N) m~3) (NO " ; z NO —NO ) and the inverse temperature applying y x both the 1993 and the 1995 data set support this: PON/NO "(60$26)](1/¹(K))!(0.19$0.10), z r"0.33, p(0.05. In the 1993 data set a remarkable correlation between the 24 h levels of PON and the night NO dose was 3 observed (Nielsen et al., 1995). However, it was not evident, if this implied that NO chemistry had been 3 the dominant PON source. From Fig. 3, it can be deduced that the ratio between the day and subsequent night PON concentrations varies strongly. Thus, the PON concentration of the 9 July day sample was 4.4 times higher than the following night sample, while the PON concentration of the 11—12 July night sample was 2.8 times higher than that of the 11 July day sample. In 6 of the 10 sets of samples the day concentration was higher than the night concentration,

38$2 1.7$0.6 2.3$0.8 0.09$0.03 2.0$0.8 5.1$0.7 10$2 203$25 0.25 21

14$5 0.9$0.3 0.19$0.06 4.9$1.3 0.6$0.3 0.32$0.04"

Concentration 24 h 1993

! Day defined as 0700—1900 and night defined as 1900—0700. " Only 7 d. # Mean from 700—1300.

PON (ng (N) m~3) NO (kg (N) m~3) HNO (kg (N) m~3) 3 Gas NO (kg (N) m~3) y Inorganic nitrate (kg (N) m~3) PAN (kg (N) m~3) C -C alkanes (ppb C) 2 8 C -C alkenes (ppbv) 2 5 Toluene (ppbv) Ozone (ppb) Formic acid (kg m~3) Acetic acid (kg m~3) Methane sulphonic acid (kg m~3) SO (ppb) 2 Wind speed (m s~1) Temperature (°C) Radiation (W m~2) PON proportion (%) of total NO y No. of samples

Compound 10$4 1.0$1.1 0.036$0.012 5$4 0.9$0.7 0.12$0.07 45$13 3.9$0.9 0.17$0.05 26$7 1.3$0.3 1.7$0.5 0.13$0.05 0.49$0.13 3.4$0.7 16$2 38$8 0.23 10

Concentration night 1995 11$3 0.7$0.4 0.13$0.05 3.2$0.9 0.9$0.5 0.25$0.07 28$6 7.6$1.6 0.26$0.09 40$5 2.3$0.6 3.0$0.9 0.12$0.04 0.9$0.4 4.5$0.7 21$2 522$46 0.34 13

Concentration Day 1995

14 1.5

1.2 0.7 3.5 0.7 1.0 2.1 0.6 1.9 1.5 1.5 1.8 1.7 0.9 1.9 1.3

Day to night ratio

0.021 5.2 0.43 0.06 64 5.3 0.15 6 1.2 1.7 0.05 0.44 2.5 13 48 0.37 1

20.7

Night conc. 11 July 1995

21.5 1.8# 0.26 5.5# 1.60 0.45 40 9.2 0.34 38# 3.9 4.4 0.07 1.1 2.1 24 377 0.30 1

Day conc. 13 July 1995

Table 2. Average concentration ($2p) of particulate organic nitrates (PON) and other pollutants and average meteorological conditions in the night and day samplesa in the period 23/05—21/07 1995, in 24 h samples! in the period 23/03—25/06 1993 and the night of 11—12 July and the day of 13 July 1995

Particulate organic nitrates: variations 2605

2606

T. NIELSEN et al.

Fig. 2. Comparison of nine sets of one 24 h sample and two 12 h samples. The figure shows the ratio between the amount of particulate organic nitrates (PON) in the 24 h sample and the sum of the PON amounts in the two sequential 12 h samples.

Fig. 3. The figure shows the day and night concentrations of particulate organic nitrates (PON). The day concentrations cover the period 0700—1900. The night concentrations cover the period 1900—0700. The x-axis shows the start date for the sampling, i.e. the night-sample for the 23 May covers the period 23 May 1900—24 May 0700.

in 2 sets the night concentrations were highest, and in the remaining 2 sets the day and night concentrations were similar. The 10 sets were carefully investigated by means of air mass trajectory plots and weather maps from each day at midnight and noon from the Deutscher Wetterdienst (1995). This analysis showed that in 4 sets (23 May, 26 June, 10 and 11 July) the day and night samples represent air pollution having the same origin of the air masses. The trajectories in these four sets had not been affected by the passage of frontal systems. Despite the similarity in source, the day/night PON ratios for these sets were also highly

variable (0.4—2). A closer look at the meteorology and air chemistry, however, can provide an explanation for this variability. On 9 July a high-pressure system moved from the North Sea to Scandinavia forcing a front above Denmark towards the south. The high-pressure system gave very stable weather in Denmark during the following days with winds from the east. The air mass trajectories passed over Southern Sweden. On the 12 July, the high--pressure system moved further towards southeast. At noon on the 12 July the wind direction changed from east to southeast. The front passed Denmark from the south in the early morning hours of the 13 July. Just after the front passage the SO concentration increased from (0.25 to 5.5 ppb 2 indicating long-range transport of air pollutants from Central Europe. The air mass trajectory arriving at Lille Valby on 13 July 1200 had passed Poland 24 h earlier. Thus, the air mass trajectories also confirmed that the air masses giving the night PON on 11—12 July and those giving the day episode of 13 July had a different origin. The night of 11—12 July was characterised by clear sky and very low wind speeds (see Table 2). The concentrations of most of the VOC were elevated. C —C alkanes and C —C alkenes were 1.4 2 8 2 5 times higher than their average night concentration, and acetylene (11—12 July: 0.73 ppb, night average: 0.61$0.20 ppb) and benzene (11—12 July 0.73 ppb, night average: 0.59$0.17 ppb) were also slightly elevated. The poor mixing of air pollutants (low wind speeds during the night) may also imply that significant concentrations of terpenes may have been present, although the emissions are much lower at nighttime than at daytime. The so-called photochemical age of NO appeared to be young, as the ratio of y inorganic nitrate, PAN and nitric acid to NO was y a factor of two lower than that on the other nights. NO was not present besides in the period 500—700, which was more than 1 h after sunrise, and therefore not had caused depletion of NO during the night. 3 Considering the young age of NO it is difficult to y imagine that most of the PON had been formed at day—time and transported to the measuring site during the night. The 13 July day sample was characterised by a long-range transport episode as indicated by the morning SO peak in addition to photochemical 2 activity. PAN was 0.55 kg N m~3 at 1300 and had its maximum (0.75 kg N m~3) at 1700. Correspondingly, the H O maximum (1.1 ppb) also was late in the 2 2 afternoon (1800—1830). Ozone was 59 ppb at 1230—1300, but unfortunately a technical failure caused the ozone measurements to stop at 1300. The levels of formic acid, acetic acid, nitric acid and particulate inorganic nitrate was also elevated (see Table 2). From the discussion above, PON production induced both by OH at daytime and by NO at nighttime 3 appears possible. The 1993 correlation between the 24-h levels of PON and the night NO dose (Nielsen 3 et al., 1995) apparently due to a combination of reasons, i.e. that the NO chemistry is an important 3

Particulate organic nitrates: variations

PON source and episodes with night NO doses usual 3 are associated with elevated photochemical activity in the daytime hours.

4. CONCLUSIONS

The low contribution of particulate organic nitrates (PON) to NO appears to be real and not a sampling y artefact. The major VOC sources for PON formation are estimated to be biogenic alkenes, terpenes and sesquiterpenes. The PON compounds are probably bi- and multifunctional organic nitrates. The organic compounds associated with particles may also be present in the gas phase. Empirical relationships between the vapour pressure and the chemical formula were established for organic compounds containing nitrate, hydroxy and/or oxo groups. PON may be formed both during the daytime and at night. A condition for the latter is that the night concentrations of higher alkenes is sufficient. The observed day and night concentrations of PON were in the same magnitude, although the day to night ratio varied from 0.4 to 4.4. Acknowledgements—Financial support of the Danish Centre for Air Pollution Processes and Models under the Danish Environmental Research Programme and the Danish Science Research Council is grateful acknowledged. Dr S+ren Hvilsted, Ris+, is acknowledged for his permission to apply the Perkin-Elmer FTIR spectrophotometer for this investigation. Ole Hertel, National Environmental Research Institute, is acknowledged for providing air mass trajectories for this investigation.

REFERENCES

Atkinson, R. (1991) Kinetics and mechanisms of the gasphase reactions of the NO radical with organic com3 pounds. Journal of Physical Chemical Reference Data 20, 459—507. Ayers, G. P., Penkett, S. A., Gillett, R. W., Bandy, B., Galbally, I. E., Meyer, C. P., Elsworth, C. M., Bentley, S. T. and Forgan, B. W. (1992) Evidence for photochemical control of ozone concentrations in unpolluted marine air. Nature 360, 446—449. Calogirou, A., Larsen, B. R., Brussol, C., Duane, M. and Kotzias, D. (1996) Decomposition of terpenes by ozone during sampling on Tenax. Analytical Chemistry 68, 1499—1506. CRC (1982) Handbook of Chemistry and Physics, 62nd edn. CRC Press, Boca Raton, FL. Crosley, D. R. (1996) NO blue ribbon panel. Journal of y Geophysical Research 101D, 2049—2052. Delmas, R., Lacaux, J. P., Menaut, J. C., Abbadie, L., Roux, X. le, Helas, G. and Lobert, J. (1995) Nitrogen compound emission from biomass burning in tropical African savannah FOS/DECAFE 1991 experiment (Lamto, Ivory Coast). Journal of Atmospheric Chemistry 22, 175—193. Deutscher, Wetterdienst (1995) European Meteorological Bulletin, Amtsblatt des Deutschen ¼etterdienst, Offenbach, Germany. Dickhut, R. M., Miller, K. E. and Ardren, A. W. (1994) Evaluation of total molecular surface area for predicting air—water partitioning properties of hydrophobic aromatic chemicals. Chemosphere 29, 283—297.

2607

Fahey, D. W., Hu¨bler, G., Parrish, D. D., Williams, E. J., Norton, R. B., Ridley, B. A., Singh, H. B., Liu, S. C. and Fehsenfeld, F. C. (1986) Reactive nitrogen species in the troposphere: Measurements of NO, NO , HNO , partic2 3 ulate nitrate, peroxyacetyl nitrate (PAN), O , and total 3 reactive nitrogen (NO ) at Niwot Ridge, Colorado. Joury nal of Geophysical Research 91D, 9781—9793. Finizio, A., Mackay, D., Bidleman, T. and Harne, T. (1997) Octanol-air partition coefficients as a predictor of partitioning of semi-volatile organic chemicals to aerosols. Atmospheric Environment 31, 2289—2296. Finlayson-Pitts, B. J. and Pitts, J. N. Jr. (1986) Atmospheric Chemistry: Fundamentals and Experimental techniques. Wiley, New York. Flocke F., Volz-Thomas, A. and Kley, D. (1991) Measurements of alkyl nitrates in rural and polluted air masses. Atmospheric Environment 25, 1951—1960. Granby, K., Egel+v, A. H., Nielsen, T. and Lohse, C. (1997) Carboxylic acids: Seasonal variation and relation to chemical and meteorological parameters. Journal of Atmospheric Chemistry (in press). Hallquist, M., Wa¨ngberg, I. and Ljungstro¨m, E. (1997) Atmospheric behaviour of dicarbonyl products from the atmospheric oxidation of selected terpenes. In ¹he Oxidizing Capacity of the ¹roposphere. ed. B. Larsen, B. Versino, and G. Angeletti. EUR 17482, pp. 345—349. European Commission, Brussels, Belgium. Hansen, A. B. and Palmgren, F. (1996) VOC air pollutants in Copenhagen. Science of ¹otal Environment 189/190, 451—457. Helweg, C., Nielsen, T. and Hansen, P. E. (1997) QSPRs compared with HPLC experiments for the estimation of K for N-PAC’s. In Procedings of the seventh Interna08 tional ¼orkshop on QSARs in Environmental Sciences. ed. G. Schu¨u¨rmann, Soc. Environmental Toxicology and Chemistry (accepted for publication). Hoffmann, T., Odum, J. R., Bowman, F., Collins, D., Klockow, D., Flagan, R. C. and Seinfeld, J. H. (1997) Formation of organic aerosols from the oxidation of biogenic hydrocarbons. Journal of Atmospheric Chemistry 26, 189—222. Kames, J. and Schurath, U. (1992) Alkyl nitrates and bifunctional nitrates of atmospheric interest: Henry’s law constants and their temperature dependencies. Journal of Atmosheric Chemistry 15, 79—95. Ko¨nig, G., Brunda, M., Puxbaum, H., Hewitt, C. N., Duckham, C. S. and Rudolph, J. (1995) Relative contribution of oxygenated hydrocarbons to the total biogenic VOC emissions of selected mid-European agricultural and natural plant species. Atmospheric Environment 29, 861—874. Luxenhofer, O., Schneider, M., Dambach, M. and Ballschmiter, K. (1996) Semivolatile long chain C -C alkyl 6 17 nitrates as trace compounds in air. Chemosphere 33, 393—404. Mylonas, D. T., Allen, D. T., Ehrmann, S. H. and Pratsinis, S. E. (1991) The sources and size distribution of organonitrates in Los Angeles aerosol. Atmospheric Environment 25, 2855—2861. Nielsen, T., Samuelsson, U., Grennfelt, P. and Thomsen, E. L. (1981) Peroxyacetyl nitrate in long-range transported polluted air. Nature 293, 553—555. Nielsen, T., Hansen, A. M. and Thomsen, E. L. (1982) A convenient method for preparation of pure standards of peroxyacetyl nitrate for atmospheric analyses. Atmospheric Environment 16, 2447—2450. Nielsen, T., Granby, K., Egel+v, A. H., Hummelsh+j, P., Jensen, F. P. and Skov, H. (1994) Lille Valby, Denmark. In ¹he ¹OR Network. A Description of ¹OR Measurements Stations, eds T. Cvitas and D. Kley, pp. 36—45. EUROTRAC, International Scientific Secretariat, GarmischPartenkirchen, Germany. Nielsen, T., Egel+v, A. H., Granby, K. and Skov, H. (1995) Observations on particulate organic nitrates and unidentified components of NO , Atmospheric Environment 29, y 1757—1769.

2608

T. NIELSEN et al.

Nielsen, T., Pilegaard, K., Egel+v, A. H., Granby, K., Hummelsh+j, P., Jensen, N. O. and Skov, H. (1996) Atmospheric nitrogen compounds: Occurrence, composition and deposition. Science of ¹otal Environment 189/190, 459—465. Nielsen, T., Siigur, K., Helweg, C., J+rgensen, O., Hansen, P. E. and Kirso, U. (1997a) Sorption of polycyclic aromatic compounds to humic acid as studied by high performance liquid chromatography. Environmental Science and ¹echnology 31, 1102—1108. Nielsen, T., Hertel, O., Christensen, C. S., Egel+v, A. H., Granby, K., Hansen, A. B., Platz, J. and Skov, H. (1997b) Evaluation of a trajectory model to simulate formation of tropospheric ozone, other photochemical oxidants and atmospheric reaction products. In ¹he Oxidizing Capacity of the ¹roposphere. eds B. Larsen, B. Versino, and G. Angeletti, EUR 17,482, pp. 426—430. European Commission, Brussels, Belgium. O’Brien, J. M., Shepson, P. B., Muthuramu, K., Hao, C., Niki, H., Hastie, D. R., Taylor, R. and Roussel, P. B. (1995) Measurements of alkyl and multifunctional organic nitrates at a rural site in Ontario, Journal of Geophysical Research 100D, (22) 795—22,804. O’Brien, J. M., Shepson, P. B., Wu, Q., Biesenthal, T., Bottenheim, J. W., Wiebe, H. A., Anlauf, K. G. and Brickell, P. (1997) Production and distribution of organic nitrates, and their relationship to carbonyl compounds in an urban environment. Atmospheric Environment 31, 2059—2069. Palen E. J., Allen D. T., Pandis S. N., Paulson S., Seinfeld, J. H. and Flagon R. C. (1993) Fourier transform infrared analysis of aerosol formed in the photooxidation of 1octene. Atmospheric Environment 27, 1471—1477. Pankow, J. F. (1987) Review and comparative analysis of the theories on partitioning between the gas and aerosol particulate phases in the atmosphere. Atmospheric Environment. 21, 2275—2283. Pankow, J.F. (1994) An absorption model of gas/particle partitioning of organic compounds in the atmosphere. Atmospheric Environment 28, 185—188. Pankow, J. F. and Bidleman, T. F. (1992) Interdependence of the slopes and intercepts from log-log correlations of measured gas-particle partitioning and vapour pressure—I.

Theory and analysis of available data. Atmospheric Environment 26, 1071—1080. Parrish, D. D., Buhr, M. P., Trainer, M., Norton, R. B., Shimshock, J. P., Fehsenfeld, F. C., Anlauf, K. G., Bottenheim, J. W., Tang, Y. Z., Wiebe, H. A., Roberts, J. M., Tanner, R. L., Newmann, L., Bowersox, V. C., Olszyna, K. J., Bailey, E. M., Rodgers, M. O., Wang, T., Berresheim, H., Roychowdhury, U. K. and Demerjian, K. L. (1993) The total reactive oxidized nitrogen levels and the partitioning between the individual species at six rural sites in Eastern North America. Journal of Geophysical Research 98D, 2927—2939. Saxena, P. and Hildemann, L. M. (1996) Water-soluble organics in atmospheric particles: a critical review of the literature and application of thermodynamics to identify candidate compounds. Journal of Atmospheric Chemistry 24, 57—109. Schwarzenbach, R. P., Gschwend, P. M. and Imboden, D. M. (1993) Environmental Organic Chemistry. Wiley, New York. Siegl, W. O. and Wallington, T., Ford Motor Company, 1996, private communication. Skov, H., Hjorth, J., Jensen, N. R. and Restelli, G. (1992) Products and mechanisms of the reaction of the nitrate radical (NO ) with isoprene, 1,3-butadiene and 2,33 dimethyl-1,3-butadiene in air. Atmospheric Environment 26, 2771—2783. Skov, H., Egel+v, A. H., Granby, K. and Nielsen, T. (1997) Relationships between ozone and other photochemical products at Ll. Valby, Denmark. Atmospheric Environment 31, 685—691. Tuazon, E. C. and Atkinson, R. (1990) A product study of the gas-phase reaction of isoprene with the OH radical in the presence of NO . International Chemistry Kinetics 22, x 1221—1236. Wangberg, I., Barnes, I. and Becker, K. H. (1996) Atmospheric chemistry of bifunctional cycloalkyl nitrates. Chemical Physical ¸etters 261, 138—144. Wayne, R. P., Barnes, I., Biggs, P., Burrows, J. P., CanosaMas, C. E., Hjorth, J., Bras, G. le, Moortgat, G. K., Perner, D., Poulet, G., Restelli, G. and Sidebottom, H., 1991, The nitrate radical: Physics, chemistry, and the atmosphere, Atmospheric Environment 25, 1—203.