Bo Larsen

Gas-phase terpene oxidation products: a review

Abstract Terpenes are emitted in large quantities from vegetation into the troposphere, where they react readily with ozone, OH and NO3 radicals leading to a number of oxidation products. The current knowledge about gas-phase terpene oxidation products is reviewed. Their formation and decomposition pathways, their products and their relevance for the troposphere, and their chemical analysis are discussed. Data on oxidation kinetics, and product yields is presented for 23 terpenes and 65 oxidation products. A total of 84 references are quoted.

Decomposition of Terpenes by Ozone during Sampling on Tenax

The ozonolytic decomposition of terpenes and terpenoids during sampling on Tenax is investigated in an artificial air system at ambient concentration levels. The detrimental effect of 8-150 ppbv ozone depends on the chemical structure of the compounds:u2009 saturated terpenoids such as 1,8-cineole, camphor, and bornyl acetate are unaffected by ozone. The terpenes and terpenoids which containing one C-C double bond are slightly decomposed in the order camphene < β-pinene ≈ myrtenal < α-pinene < sabinene ≈ citronellal. The compounds containing two or more double bonds are significantly decomposed in the order d-limonene ≈ citral < linalool < β-ocimene < terpinolene << α-terpinene ≈ β-caryophyllene. For α-pinene, sabinene and d-limonene, their ozonolysis products are found on the tubes:u2009 pinonaldehyde, 5-(1-methylethyl)bicyclo[3.1.0]hexan-2-one, and 3-(1-methylethenyl)-6-oxoheptanal. The analytical recoveries are significantly enhanced for many compounds when the sampling duration is reduced from 10 min to 30 s, explained by the time available for ozonolysis. A miniature ozone scrubber with multiple layers of MnO(2)-coated copper nets was developed and thoroughly tested. The optimal number of plies is found to be 8, which ensures quantitative recoveries for all test compounds except α-terpinene, β-caryophyllene, citral, and citronellal. The results that are reported here call into question previous data on terpenes and terpenoids and/or their oxidation products where measurements have been carried out without the prior removal of ozone.

Ozonolysis at vegetation surfaces : A source of acetone, 4-oxopentanal, 6-methyl-5-hepten-2-one, and geranyl acetone in the troposphere

Abstract The present study gives a possible explanation for the ubiquitous occurrence of 6-methyl-5-hepten-2-one and acetone in ambient air and reports for the first time on a widespread occurrence of geranyl acetone and 4-oxopentanal. We have conducted a series of laboratory experiments in which it is demonstrated that significant amounts of geranyl acetone, 6-methyl-5-hepten-2-one (6-MHO), 4-oxopentanal (4-OPA), and acetone are formed by the reaction of ozone with foliage of common vegetation in the Mediterranean area (Quercus ilex>Citrus sinensis>Quercus suber>Quercus freinetto>Pinus pinea). In order to rule out biological formation, epicuticular waxes were extracted from the leaves, dispersed on glass wool and allowed to react with a flow of artificial air. Significant amounts of 6-MHO and 4-OPA were formed at ozone concentrations of 50–100xa0ppbv, but not at zero ozone. A number of terpenoids common in vegetation contain the structural element necessary for ozonolytic formation of 6-MHO. Two sesquiterpenes (nerolidol; farnesol), and a triterpene (squalene) selected as representative test compounds were demonstrated to be strong precursors for acetone, 4-OPA, and 6-MHO. Squalene was also a strong precursor for geranyl acetone. The atmospheric lifetime of geranyl acetone and 6-MHO is less than 1xa0h under typical conditions. For the present study, we have synthesized 4-OPA and investigated the kinetics of its gas-phase reaction with OH, NO3, and O3. A tropospheric lifetime longer than 17xa0h under typical conditions was calculated from the measured reaction rate constants, which explains the tropospheric occurrence of 4-OPA. It is concluded that future atmospheric chemistry investigations should included geranyl acetone, 6-MHO, and 4-OPA. In a separate experiment it was demonstrated that human skin lipid which contains squalene as a major component is a strong precursor for the four above-mentioned compounds plus nonanal and decanal. The accidental touching of material which later comes into contact with ozone can lead to strong artifact formation of these carbonyl compounds. Previously published results on these compounds must be seen in this new light.

Determination of polar terpene oxidation products in aerosols by liquid chromatography-ion trap mass spectrometry

Abstract A new method for the determination of polar terpene oxidation products in secondary aerosols is described. It is based on collection of particles on PTFE filters and extraction with dichloromethane followed by analysis with liquid chromatography–ion-trap mass spectrometry (MSn) using pneumatically assisted electrospray ionisation (ESI) and atmospheric pressure chemical ionisation (APCI) with the ion-trap operated in the product scan mode. Separations were achieved on a C18 reversed-phase column with methanol–water (0.1% acetic acid) as eluent. The method has a high sensitivity (instrument detection limit 0.7–7 pg/μl at S/N=3) and precision (5–10%) and a good linearity. Acidic oxidation products produce strong signals with ESI. They appear as negative quasimolecular ions [M−H]−, acetate adducts [M+CH3COO]− and molecular clusters [2M−H]−. In the MS2 mode these acids show strong signals from neutral loss of CO2: [M−H−CO2]−, and/or weaker signals from loss of H2O: [M−H−H2O]−, [M−H−H2O−CO2]−. Mass spectra were recorded by APCI for a number of oxygenated terpenoid standards containing keto groups, hydroxy groups, aldehyde groups, or epoxy groups. These compounds give intense signals as their positive quasimolecular ions [M+H]+, methanol adducts [M+H+CH3OH]+ and fragments from loss of water, such as [M+H−H2O]+, [M+H+CH3OH−H2O]+ and [M+H−2H2O]+. By MS2 and MS3 neutral loss of H2O and 2H2O is observed. The method has been tested in analysis of aerosol from O3–α-pinene and O3–myrtenol and has been proved to compare well with classical methods based on derivatisation and gas chromatography mass spectrometry. Three new compounds, tentatively identified in aerosol are reported here for the first time: 10-hydroxypinonic acid, 9-hydroxynorpinonic acid and pinalic 4-acid.

Analysis of chloro- and nitroanilines and -benzenes in soils by headspace solid-phase microextraction

Abstract Quantitative analysis by headspace solid-phase microextraction (HS-SPME) of twenty chloro- and nitrobenzenes and -anilines spiked into soil samples was possible when the calibration was performed with the same matrix. The SPME response increased with addition of water to the air dried soil samples, and by optimisation of extraction conditions such as temperature, extraction time and sample agitation. Distribution constants between the fibre coating materials and water were determined. They showed that polyacrylate, rather than polydimethylsiloxane, should be used in the analysis of polar compounds. Matrix effects, mainly depending on the organic carbon content of the soil, were so large that quantitative analyses of real soil samples would not be reliable with model matrix calibration. This problem would be overcome if a nearly exhaustive extraction could be achieved. Recovery studies showed that this was possible only in the case of extraction of the lightest non-polar analytes from a soil with a very low content of organic carbon. Instead quantification could be performed with calibration by standard addition. By this approach the non-linear calibration curves at concentrations near the detection limits would cause inaccurate results at trace level, whereas it could be applied successfully within the linear ranges. Hence, the potential of HS-SPME in soil analysis is primarily as a rapid screening technique.

Isoprene and its degradation products as strong ozone precursors in Insubria, Northern Italy

Abstract Frequent smog episodes occur during spring, summer, and autumn in Insubria, Northern Italy. On a test site in this area the atmospheric concentration of the photo-oxidants ozone and peroxyacetyl nitrate has been monitored over a year (2000) together with ozone precursors listed in the European Union Air Quality Directive 2002/3/EC, such as nitrous oxides (NOX) and volatile organic compounds (VOC) including hydrocarbons and carbonyls. The results of this study revealed a strong impact of biogenic isoprene on the air quality. In winter isoprene was detected at the ppt level and correlated with anthropogenic VOC. However, during the growing season isoprene exhibited a distinct diurnal variation with maximum concentrations late in the afternoon reaching up 70xa0ppbC attributed to strong emissions from the abundant vegetation of broad-leaf deciduous trees in this area. A new HPLC-MS method was developed for the determination of isoprenes primary atmospheric oxidation products methacrolein as its 2,4-dinitrophenylhydrazone and methyl vinyl ketone as an unusual double derivative with 2,4-dinitrophenylhydrazine. Methacrolein and methyl vinyl ketone followed the same diurnal and annual trends as isoprene. The average monthly concentration of isoprene and these products ranged from around 10xa0ppbC in June, July and September to 20xa0ppbC in August, which constitutes 15–30% of C3–C9 VOCs. The contribution from isoprene photo-oxidation to the ambient air formaldehyde concentrations was also found to be high during this period ranging from 30% to 60% in May, June, July and August. From the atmospheric VOC and NOX concentrations the local photochemical ozone formation was estimated by the incremental reactivity approach. The calculations showed that in summer isoprenes contribution to the local ozone formation was as high as 50–75%.

Quantifying the Impact of Residential Heating on the Urban Air Quality in a Typical European Coal Combustion Region

The present investigation, carried out as a case study in a typical major city situated in a European coal combustion region (Krakow, Poland), aims at quantifying the impact on the urban air quality of residential heating by coal combustion in comparison with other potential pollution sources such as power plants, industry, and traffic. Emissions were measured for 20 major sources, including small stoves and boilers, and the particulate matter (PM) was analyzed for 52 individual compounds together with outdoor and indoor PM10 collected during typical winter pollution episodes. The data were analyzed using chemical mass balance modeling (CMB) and constrained positive matrix factorization (CMF) yielding source apportionments for PM10, B(a)P, and other regulated air pollutants namely Cd, Ni, As, and Pb. The results are potentially very useful for planning abatement strategies in all areas of the world, where coal combustion in small appliances is significant. During the studied pollution episodes in Krakow, European air quality limits were exceeded with up to a factor 8 for PM10 and up to a factor 200 for B(a)P. The levels of these air pollutants were accompanied by high concentrations of azaarenes, known markers for inefficient coal combustion. The major culprit for the extreme pollution levels was demonstrated to be residential heating by coal combustion in small stoves and boilers (>50% for PM10 and >90% B(a)P), whereas road transport (<10% for PM10 and <3% for B(a)P), and industry (4-15% for PM10 and <6% for B(a)P) played a lesser role. The indoor PM10 and B(a)P concentrations were at high levels similar to those of outdoor concentrations and were found to have the same sources as outdoors. The inorganic secondary aerosol component of PM10 amounted to around 30%, which for a large part may be attributed to the industrial emission of the precursors SO2 and NOx.

Sampling and analysis of terpenes in air. An interlaboratory comparison

Abstract An interlaboratory comparison on the sampling and analysis of terpenes in air was held within the framework of the BEMA (Biogenic Emissions in the Mediterranean Area) project in May 1995. Samples were drawn and analysed by 10 European laboratories from a dynamic artificial air generator in which five terpenes were present at low ng l −1 levels and ozone varied between 8 and 125 ppbv. Significant improvements over previous inter-comparison exercises in the quality of results were observed. At the ozone mixing ratio of 8 ppbv a good agreement among laboratories was obtained for all test compounds with mean values close to the target concentration. At higher mixing ratios, ozone reduced terpene recoveries and decreased the precision of the measurements due to ozonolysis during sampling. For β-pinene this effect was negligible but for the more reactive compounds significant losses were observed in some laboratories ( cis -β-ocimene = trans -β-ocimene > linalool > d-limonene). The detrimental effect of ozone was significantly lower for the laboratories which removed ozone prior to sampling by scrubbers. Parallel sampling was carried out with a standardised sampler and each individual laboratorys own device. A good agreement between the two sets of results was obtained, clearly showing that the majority of laboratories used efficient sampling systems. Two different standard solutions were analysed by each laboratory. Only in a few cases did interference in the GC separation cause problems for the quantification of the terpenes (nonanal/linalool). However, making up of standards for the calibration of the analytical equipment (GC-MS or GC-FID) was pointed out as a source of error in some laboratories.

Atmospheric pressure ionisation multiple mass spectrometric analysis of pesticides

Liquid chromatography-multiple mass spectrometry (LC-MS) has been investigated for analysis of polar pesticides in water using an ion-trap instrument and atmospheric pressure ionisation. Carbamate, triazine and phenylurea pesticides were best ionised as positive ions with atmospheric pressure chemical ionisation, while phenoxy acid herbicides, nitrophenols and bentazone yielded stronger signals as negative ions with pneumatically assisted electrospray. The ion fragmentation processes and pathways were studied in detail by MS, MS2, MS3 and MS4. All compounds were observed as their protonised or deprotonised molecular ions by MS and in the successive fragmentation by MS the structures of typical (diagnostic) product ions were tentatively identified for each class of pesticide. Phenylureas yield an ion at m/z 72 by MS2, corresponding to O=C=N+(CH3)2. Carbamates produce [M+H-CONCH3]+ fragments by MS2 from neutral loss of methylisocyanate. Characteristic fragmentation pathways for triazine pesticides are [M+H]+-->m/z 174-->m/z 146-->m/z 110 and [M+H]+-->m/z 174-->m/z 132-->m/z 104 by MS-MS2-MS3-MS4 from cleavage of lateral chains in the triazine ring followed by ring opening. Phenoxy acid herbicides produce peculiar fragments by MS2 from loss of the acidic group possibly as the corresponding lactone. Nitrophenols are subject to loss of both *OH radical and NO groups thereby forming the correspondent phenols and quinones. The performance of the method with respect to quantitation compares favourably with traditional methods. With the ion-trap run in a time scheduled single ion monitoring mode, typical limits of detection (LODs) are in the low pg range and the repeatability standard deviations are between 3 and 15%. Assuming extraction of 1-l water samples and 1 ml final volumes the injection of 50-microl aliquots corresponds to LODs well below the requirement for the European Union water directive (EC/80/778).

Enhanced selectivity in the analysis of chlorobiphenyls on a carborane phenylmethylsiloxane copolymer gas chromatography phase (HT-8)

Abstract The congener specific analysis of polychlorinated biphenyls (PCBs) on a new chemically bonded stationary GC phase, HT-8 (1,7-dicarba-closo-dodecarborane phenylmethyl siloxane) is investigated. The incorporation of carborane in the polysiloxane backbone enhances selectivity and temperature stability of the phase. On a 60 m × 0.25 mm I.D. fused-silica column coated with 0.25 μm HT-8, a total of 106 congeners in Aroclor mixtures elute as separated peaks. By the use of mass spectrometric detection 138 congeners can be analysed without interference from coeluting PCBs. In literature, no other GC phase has been reported with such a high PCB selectivity. An interesting feature of HT-8 is its near-baseline separation of the critical pair CB138/CB163. Separations for these congeners on other GC phases are reviewed. The performance of a thin film (0.15 μm) version of the HT-8 column with fast temperature programming is tested. It is concluded that congener-specific, rapid screening of seven indicator PCBs (CB28, CB52, CB101, CB118, CB138, CB153 and CB180) with a total GC run-time of 13 min is feasible with an accuracy sufficient for most purposes.