Kevin C. Clemitshaw

A Review of Instrumentation and Measurement Techniques for Ground-Based and Airborne Field Studies of Gas-Phase Tropospheric Chemistry

ABSTRACT The development, applications and intercomparisons of instrumentation and measurement techniques for ground-based and airborne field studies of gas-phase tropospheric chemistry are reviewed. Filter radiometry, chemical actinometry and scanning spectroradiometry for j-NO2 and j-(O1D) are described. Detection of OH using L-POAS/DOAS, MOAS, LIF/FAGE and CIMS is discussed. Observations of NO3 using DOAS, MIESR and CRDS, and of HO2 and RO2 with CA, CIMS/IMR-MS, MIESR, and LIF/FAGE are also reviewed. GC-FID, GC-ECD and GC-MS analyses of NMHCs and alkyl, peroxyacyl and bi-functional organic nitrates are described, together with applications of CIMS/PTR-MS, DOAS and LIF. CO measurements utilising GFC, GC-HgO/UV, VUV-RF and TDLAS are presented. Measurements of O3 using UV photometry, chemiluminescence, electrochemistry, LIDAR/DIAL and DOAS are discussed. Chemiluminescence and LIF detection methods for NO with photochemical and thermal convertors for NO2 and NOy are also discussed, as are TDLAS, MIESR, DOAS, CRDS and other approaches for NO2. HONO measurements using DOAS, CIMS, CRDS, TDLAS, denuder systems, chemiluminescence and on-line analyses of NO2 − are described. For HONO2, filter packs, denuder systems, chemiluminescence methods, mist chambers, TDLAS, CIMS and LIF are presented. Direct and indirect fluorimetric, chromatographic and spectroscopic detection techniques are discussed for CH2O and higher carbonyls. Observations of H2O2 and ROOH utilising colorimetry, chemiluminescence, fluorescence, HPLC and TDLAS are also described.

Simultaneous observations of nitrate and peroxy radicals in the marine boundary layer

This paper describes the most extensive set of simultaneous measurements of the concentrations of nitrate (NO3) and peroxy (sum of HO2 + RO2, R = alkyl and acyl) radicals to date. The measurements were made in the coastal marine boundary layer over the North Sea, at the Weybourne Atmospheric Observatory on the North Norfolk coast during the spring and autumn of 1994. In spring the average nighttime concentration of NO3 measured by differential optical absorption spectroscopy, was about 10 parts per trillion (ppt) (maximum 25 ppt). The corresponding peroxy radical concentration, measured by the chemical amplifier technique, averaged about 2 ppt (maximum 6 ppt), although this is likely to be an underestimate of the total radical concentration. There is a significant positive correlation between the two sets of radicals, which has not been reported previously. A box model of the marine boundary layer is used to show that this correlation arises from the processing of reactive organic species by NO3. During spring the relatively long lifetime of NO3 (up to 18 min) at night is controlled by reaction with dimethyl sulfide (DMS), and the model predicts significant production of HNO3, methyl tiomethylen (CH3SCH2O2) and other peroxy radicals, HCHO, and eventually sulfate. A nighttime production rate for the hydroxyl (OH) of about 2 x 10(4) molecules cm(-3) s(-1) is estimated. During one night in autumn the NO3 lifetime of about 3 min is too short to be explained by reaction with unsaturated hydrocarbons, but is satisfactorily accounted for by the heterogeneous loss of N2O5 on deliquesced aerosols in relatively polluted conditions.

Estimates of atmospheric hydroxyl radical concentrations from the observed decay of many reactive hydrocarbons in well-defined urban plumes

An analytical system has been developed which allows the confident identification and measurement of 35 hydrocarbons of different reactivities in air samples collected in many locations. This paper describes the application of the technique to follow the differential decay of hydrocarbons in urban plumes spreading from London during the summer. The data have been used to determine atmospheric hydroxyl radical concentrations, averaged over several hours, on the assumption that the decay of the hydrocarbons is entirely due to reaction with these radicals. Hydroxyl radical concentrations were derived from the measured decay of several alkenes which agreed with theoretical estimates. There are strong indications, however, that substituted aromatic molecules decay much faster than could be accounted for solely by reaction with hydroxyl radicals; this may indicate the presence of a further chemical removal mechanism. 18 refs., 9 figs., 5 tabs.

Gas-phase ultraviolet absorption cross-sections and atmospheric lifetimes of several C2C5 alkyl nitrates

Abstract Gas-phase ultraviolet absorption cross-sections of ethyl nitrate, 1-propyl nitrate, 2-propyl nitrate, 2-methyl, 1-propyl nitrate, 1-butyl nitrate and 1-pentyl nitrate have been measured over the wavelength range 220–340 nm using a dual-beam, diode array spectrometer. Each alkyl nitrate spectrum appears to be the sum of at least two Gaussian-shaped absorptions with an intense π−π ∗ band extending from 190–240 nm having a shoulder between 250–340 nm due to a π−π ∗ system. The absorption cross-sections recorded for ethyl nitrate, 1-propyl nitrate, 2-propyl nitrate and 1-butyl nitrate are within 10% of previous data: those of 2-methyl, 1-propyl nitrate and 1-pentyl nitrate have been measured for the first time. For ethyl nitrate, absorption cross-sections between 280–340 nm in the tail of the near-ultraviolet band declined with decreasing temperature from 298-233 K. At two-dimensional numerical model of tropospheric chemistry was used to calculate atmospheric lifetimes with respect to photodissociation and OH radical reaction that are markedly dependent on season, latitude and altitude. Relatively long, surface level atmospheric lifetimes of several days to weeks confirm that the C2C5 alkyl nitrates may act as temporary reservoirs for NOx and suggest that they may also constitute a significant fraction of total reactive odd-nitrogen, NOy, particularly during winter at northern hemisphere high latitudes.

Relationships between ozone photolysis rates and peroxy radical concentrations in clean marine air over the Southern Ocean

Measurements of the sum of inorganic and organic peroxy radicals (RO2) and photolysis rate coefficients J(NO2) and J(O1D) have been made at Cape Grim, Tasmania in the course of a comprehensive experiment which studied photochemistry in the unpolluted marine boundary layer. The SOAPEX (Southern Ocean Atmospheric Photochemistry Experiment) campaign included measurements of ozone, peroxides, nitrogen oxides, water vapor, and many other parameters. This first full length paper concerned with the experiment focuses on the types of relationships observed between peroxy radicals and J(NO2), J(O1D) and √[J(O1D)] in different air masses in which ozone is either produced or destroyed by photochemistry. It was found that in baseline air with ozone loss, RO2 was proportional to √[J(O1D)], whereas in more polluted air RO2 was proportional to J(O1D). Simple algorithms were derived to explain these relationships and also to calculate the concentrations of OH radicals in baseline air from the instantaneous RO2 concentrations. The signal to noise ratio of the peroxy radical measurements was up to 10 for 1-min values and much higher than in other previous deployments of the instrument in the northern hemisphere, leading to the confident determination of the relationships between RO2 and J(O1D) in different conditions. The absolute concentration Of RO2 determined in these experiments is in some doubt, but this does not affect our conclusions concerned either with the behavior of peroxy radicals with changing light levels or with the concentrations of OH calculated from RO2. The results provide confidence that the level of understanding of the photochemistry of ozone leading to the production of peroxide via recombination of peroxy radicals in clean air environments is well advanced.

Observations of new particle production in the atmosphere of a moderately polluted site in eastern England.

Measurements of particle number density and Fuchs surface area, together with a range of gaseous pollutant concentrations, have been made in June 1995 at a coastal site in eastern England which receives air from a range of polluted and less polluted origins. Periods of enhanced local particle production were identified and found to be associated predominantly with relatively polluted air sectors. An examination of the factors contributing to homogeneous nucleation and hence new particle production suggests that those most important at this location are probably the production of hydroxyl radicals and the availability of ammonia. A numerical modeling study calculating characteristic timescales for new particle production and for condensation onto existing aerosol surfaces is able to predict periods of new particle production. The model suggests that oxidation of dimethyl sulphide and sulphur dioxide and homogeneous nucleation of sulphuric acid and water, probably in combination with ammonia, are the source of new particles at this site.

Temperature-dependent absorption cross-sections of gaseous nitric acid and methyl nitrate

Abstract Absorption cross-sections for HNO3 and CH3ONO2 were measured in the wavelength region 220–340 nm, using a dual-beam diode array spectrometer, with a spectral resoltuion of 0.3 nm. The results at room temperature were in good agreement with earlier measurements. Absorption over most of the range showed a distinct temperature dependence, with a similar decline in cross-section with decreasing temperature in the range 295-239 K for both molecules. The results have quite large effects on the calculated photodissociation rate of HNO3 in the lower stratosphere, especially at the low temperatures and high solar zenith angles (SZA) characteristics of the polar winter and spring. For example, at 20 km altitude, with an SZA of 80°, the photolysis rate at a temperature of 200 K is approximately a factor of 5 smaller than at 298 K.

Reactive Nitrogen Compounds at Spitsbergen in the Norwegian Arctic

Simultaneousindependent measurements of NOy and NOx(NOx= NO + NO2) by high-sensitivitychemiluminescence systems and of PAN (peroxyacetylnitrate) and PPN (peroxypropionyl nitrate) by GC-ECDwere made at Spitsbergen in the Norwegian Arcticduring the first half year of 1994. The average mixingratio of the sum of PAN and PPN (denoted PANs)increased from around 150 pptv in early winter to amaximum of around 500 pptv in late March, whereasepisodic peak values reached 800 pptv. This occurredsimultaneously with a maximum in ozone which increasedto 45–50 ppbv in March–April. The average NOxmixing ratio was 27 pptv and did not show any cyclethrough the period. The NOy mixing ratio showeda maximum in late March, while the difference betweenNOy and PAN decreased during spring. This is anindication of the dominance of PAN in the NOybudget in the Arctic, but possible changes in theefficiency of the NOy converter could alsocontribute to this. Although most PAN in theArctic is believed to be due to long range transport,the observations indicate local loss and formationrates of up to 1–2 pptv h-1 in April–May.Measurements of carbonyl compounds suggest thatacetaldehyde was the dominant, local precursor ofPAN.

Budget of NOy species measured at a coastal site

Abstract Concentrations of total NO y as well as the major individual species comprising NO y , i.e. NO, NO2, PAN, HNO3, HONO and aerosol nitrate, have been measured in three campaigns at the Weybourne Atmospheric Observatory on the north Norfolk coast in Eastern England. The sum of the individual NO y species ( ∑NO yi ) accounted for a high proportion (typically 96%) of total NO y in the two Winter campaigns carried out, but considerably less (about 79%) in a Summer campaign. It was found that the apparent deficit in ∑NO yi in the summer campaign samples correlated closely with the concentration of NH x (sum of gaseous ammonia and particulate ammonium) and it appears that excessive cleaning of the converter led to a sensitivity to these species. Subtraction of the NH x interference leads to a high degree of closure, close to 100% in morning and nighttime samples and 95% in afternoon samples. Thus, unlike many published studies, there is no substantial “missing” NO y and the small deficit in afternoon samples is probably explicable in terms of unmeasured photochemically generated organic nitrates. The magnitude of the ratio ∑NO yi / NO y appeared largely independent of the NO y concentration. The NO x / NO y ratio tended to increase with increased NO y in the full dataset consistent with lesser oxidation of NO x in the more polluted airmasses. This facet was not obvious, however, in a subset of marine airmass samples. On a short timescale, however, the NO x / NO y ratio can exhibit a strong diurnal cycle. An attempt to simulate this cycle of NO x / NO y by means of a numerical model was fairly successful, though there was some lowering of the midpoint of the oscillation towards the end of the run. It was also demonstrated that if the rate of diurnally averaged nitric acid deposition were to match its rate of formation from NO2, a relatively constant NO x / NO y ratio in time-averaged samples could be explained without invoking atmospheric conversion of HNO3 to NO2. Nevertheless, the latter remains a possible explanation for the relative constancy of the NO x / NO y ratio.

Laboratory Studies of the Response of a Peroxy Radical Chemical Amplifier to HO2 and a Series of Organic Peroxy Radicals

The response of a peroxy radical chemical amplifier, supplied by the University of East Anglia (the UEA-PERCA), to HO2 and seven organic peroxy radicals (CH3O2, C2H5O2, neo-C5H11O2, HOCH2CH2O2, CH3CH(OH)CH(O2)CH3, (CH3)2C(OH)C(O2)(CH3)2 and CH3C(=O)O2) has been investigated. The peroxy radicals were produced in air at typical ambient levels (ca. 20–30 pptv) by reaction of CO or an appropriate organic precursor with OH radicals, generated from the near UV photolysis of nitrous acid (HONO) in a flow reactor. Experiments carried out at room temperature and atmospheric pressure in dry air, allowed measurement of the response of the UEA-PERCA to the organic peroxy radicals relative to the response to HO2 (denoted Ψobs), for reagent NO concentrations in the range 1–8 ppmv. The results indicate that HO2, CH3O2 and larger peroxy radicals containing polar functional groups are removed to a certain extent on the pyrex surfaces of the inlet zone of the UEA-PERCA, prior to reaching the reaction zone where the amplification chemistry occurs. For C2H5O2 and larger alkyl peroxy radicals, heterogeneous removal in the inlet zone appears to be minor. With the assumption that neo-C5H11O2 is not removed heterogeneously, the results are used to derive the following fractional responses (denoted Ψ) of the UEA-PERCA to the peroxy radicals ([NO] = 3 ppmv): HO2, (69 ± 5)%; CH3O2, (78 ± 5)%; C2H5O2, (95 ± 7)%; neo-C5H11O2, (74 ± 5)%; HOCH2CH2O2, (73 ± 5)%; CH3CH(OH)CH(O2)CH3, (77 ± 6)%; (CH3)2C(OH)C(O2)(CH3)2, (81 ± 6)%; and CH3C(=O)O2, (76 ± 5)%. Further experiments established that the response of the UEA-PERCA to HO2 was increased by ca. 10% when the air is moist (20–30% relative humidity). This is interpreted in terms of competitive adsorption of H2O and HO2 on the pyrex surfaces. A similar influence was observed when the inlet zone was coated with teflon. For the majority of peroxy radicals studied, the reduction of Ψ to less than 100% is almost entirely due to heterogeneous removal in the inlet zone. In the case of neo-C5H11O2, however, gas-phase reactions in the reaction zone have a significant influence on the response, due in part to the formation of t-C4H9O2 as an intermediate in its conversion to HO2. The influence of reactions of the RO2 and RO intermediates which limit the yield of HO2 in the reaction zone are discussed, and it is shown that calculations of the fractional conversion of neo-C5H11O2 to HO2 (i.e. Ψ) under idealised, well mixed conditions, do not give a good description of the observed dependence of Ψ on the variation of the reagent NO concentration. The results are discussed in terms of the interpretation of field measurements of peroxy radicals made using the chemical amplification technique.