HNO3 Vapor Research Articles

Concentration gradients and dry deposition of nitric acid vapor at the Mauna Loa Observatory, Hawaii

From August 14 to September 5, 1991, we measured the mixing ratio of nitric acid vapor and several aerosol species (nitrate, sulfate, ammonium and sodium) at three heights (2 m, 10 m and 30 m) at the Mauna Loa Observatory. Well‐defined logarithmic gradients of temperature and HNO3 vapor were measured throughout the experiment, indicating rapid HNO3 vapor deposition to the ground. Measured dry deposition velocities for HNO3 vapor ranged from 0.27 to 4 cm s−1. The collection efficiency for HNO3 vapor varied with sampler location and orientation on the sampling tower, due to concentration depletion by the tower or the sampler itself. We found that our long‐term samplers at 8 m may underestimate free tropospheric HNO3 mixing ratios by about 20% due to dry deposition.

Atmosphere × Canopy Interactions of Nitric Acid Vapor in Loblolly Pine Grown in Open‐Top Chambers

AbstractMany studies that address the impact of tropospheric O3 on agricultural and forested ecosystems utilize the open‐top chamber. During the production of O3 using electrical discharge generators fed with dry air, there is an inadvertent addition of HNO3 vapor, a highly reactive trace gas. While several studies have proposed that HNO3 vapor introduces artifacts, none has measured concentrations of the odd‐N2 trace gas in the chamber or investigated the fate of the N in the context of whole‐plant physiology and growth. These questions were investigated using open‐top chambers containing seedlings of loblolly pine (Pinus taeda L.) during the 1988 growing season in Oak Ridge, TN. The O3 treatments consisted of charcoal‐filtered or subambient (0.96 µmol m−3, 24‐h mean), ambient (1.62 µmol m−3, 24‐h mean), and elevated (2.36 µmol m−3, 24‐h mean) concentrations, the last being accomplished by proportional O3 addition over the diurnal period. Measurements of the HNO3 vapor concentration during dry periods only (no rainfall or ground‐level fog) averaged 28.6 nmol m−3 (subambient), 55.4 nmol m−3 (ambient air), and 240.0 nmol m−3 (elevated O3), an 8.4‐fold range. For every 100 mol of O3 added to the chamber, 28 mol of HNO3 vapor were inadvertently added; this ratio is several times higher than that previously reported. This result, taken with published estimates of leaf conductance to HNO3 vapor, indicates a maximum N deposition in the form of HNO3 vapor ranging from 19.5 pmol N cm−2 leaf area h−1 (subambient O3) to 171.9 pmol N cm−2 h−1 (elevated O3). Given the nutrient content of the seedlings and knowledge of the fate of HNO3 vapor on the leaf surface and leaf interior, the degree to which N deposition via HNO3 vapor met the N requirements of the loblolly pine seedlings was estimated. Seedlings in the elevated O3 treatment had an upper‐limit estimate of 3.5% for the needles and 1.8% for the whole plant of N derived from HNO3 vapor. The concentration of HNO3 vapor in the chambers, site of HNO3 vapor deposition, N requirements of the loblolly pine seedlings, and estimated threshold for phytotoxic effects of HNO3 vapor indicate that the inadvertent production of this odd‐N2 trace gas is important in understanding the atmospheric chemistry within the chambers, but that the level of N loading in this study is unlikely to affect physiology or growth. However, we recommend that studies that employ higher O3‐exposure scenarios recognize the potential for inadvertent N deposition, particularly in trees grown in N‐deficient substrate.

The uptake of HNO3 onto ice, NAT, and frozen sulfuric acid

The uptake of HNO3 onto ice, HNO3‐H2O layers, and solid H2SO4/H2O solutions were studied in a cylindrical flow tube in conjunction with a chemical ionization mass spectrometer detection system. An efficient loss of HNO3 from the gas‐phase to these surfaces was observed and the uptake coefficient was ≥0.3. The HNO3‐H2O layers formed, however, had HNO3 vapor pressures significantly in excess of the expected values for NAT (nitric acid trihydrate). Supersaturations in HNO3 of at least a factor of three and up to ten were observed suggesting a hindrance to the formation of NAT. These results imply that the condensation temperatures of Type I PSCs are 2–3 K colder than that expected for NAT.

Nitric Acid Trihydrate Particle Formation in the Polar Stratosphere and Its Effect on Nitric Acid Transport to the Troposphere.

Growth and sedimentation of type-I polar stratospheric clouds (type-I PSCs) particles including nitric acid are examined using a numerical model. The PSCs particles which grow in the upper particle layer have enough large size (-10μm) to descend rapidly to the tropopause within several days. In the lower part of PSCs layer, evaporation of HNO3 from particulate matter occurs and acts as source of HNO3 vapor. From view point of stratospheric HNO3 distribution, the particle sedimentation effectively functions as vertical transport mechanism of HNO3 vapor from the upper PSCs layer to the lower layer and/or to the troposphere.

Balloon observations of nitric acid aerosol formation in the Arctic stratosphere: II. Aerosol

Balloonborne aerosol measurements over the size range r = 0.01 to 10 μm on 31 January 1990 are used to determine the HNO3 vapor mass deposited in Arctic nitric acid trihydrate (NAT) clouds in the 21–24 km region. The clouds consisted mainly of small (r≈ 0.2 μm) particles and appeared to have been formed during recent cooling because the r ≥ 0.15 μm aerosol concentration in the coldest part of the cloud was nearly equal to the condensation nuclei concentration, suggesting a large vapor supersaturation. The maximum inferred HNO3 vapor mixing ratio which condensed in the cloud was about 4 ppbv, in good agreement with the observed depletion of HNO3 vapor in the same altitude region under similar temperature conditions 23 hours earlier.

Nucleation and Growth of HNO3·3H2O Particles in the Polar Stratosphere

Growth of nitric acid trihydrate (NAT) particles on background stratospheric aerosols is examined for an isolated air parcel cooled at a uniform rate. The free energy barrier against nucleation and rates of cooling are varied over a range of probable values. During the process of nucleation, the saturation ratio of HNO3 vapor reaches a maximum value between 2 and 15, corresponding to supercooling by 1–4 K. Significant supersaturation may be maintained after nucleation due to the small surface area of NAT available for vapor deposition. If cooling rates exceed 0.5–1 K day−1, small (<2 μm radius) particles of NAT are produced. A major fraction of the available condensation nuclei is activated and removal of HNO3 by gravitational settling is slow. If cooling rates are less than 0.5–1 K day−1, the number of aerosols that nucleate is reduced, leading to differential growth of large (>2 μm radius) NAT particles. Gravitational settling of NAT particles could result in removal of HNO3 on time scales close to one week. Observations of 5 μm radius particles in clouds at temperatures above the water frost point may reflect condensation of NAT on ice particles that fall through a column of air as it is cooled. Rapid condensation of HNO3 on ice particles is promoted by the high supersaturation attained during nucleation and maintained during subsequent cooling. This process provides a mechanism for irreversible removal of HNO3.

Measuring tropospheric HNO3: Problems and prospects for nylon filter and mist chamber techniques

We performed a series of laboratory and field measurements to evaluate the mist chamber technique for determining tropospheric HNO3 concentrations. Both the mist chamber and standard Nylon filter techniques exhibit high collection efficiency and excellent agreement measuring HNO3 vapors from a permeation source. When simultaneously sampling ambient air in eastern Virginia the Nylon filter measured an average of 70% higher HNO3 concentration than the mist chamber technique. Our results indicate that O3 causes a low‐level positive artifact interference in HNO3 measurements performed with the filter technique. This O3‐induced error is small, however, compared to the large difference between atmospheric HNO3 concentrations determined with the two techniques. We hypothesize that unidentified (organic?) nitrogen species in the atmosphere react for form NO3− on the filter and this phenomenon may interfere with Nylon filter measurements of HNO3 vapor. These potential interferences did not appear to affect measurements of HNO3 with the mist chamber method, but we are continuing studies to evaluate potential interference from various atmospheric species. Because the mist chamber technique employs an aerosol prefilter, it is probably best suited for measuring HNO3 concentrations when particle‐NO3− levels are low. In light of our preliminary results, we caution users of the Nylon filter method to continue to critically evaluate this technique for specificity in determining HNO3 vapor phase concentrations in the atmosphere, particularly at low concentrations typical of the remote troposphere.

Stratospheric cloud micro‐layers and small‐scale temperature variations in the Arctic in 1989

Thin layers of large (r ≈ 2–3 μm) particles were observed in the Arctic between 14 and 27 km on 23 and 30 January 1989. Although the particle mass is consistent with the available HNO3 vapor, it is not always clear whether the morphology of the layers is the result of recent condensation or evaporation as the layers appear under conditions of both supersaluration and undersaturation with respect to nitric acid trihydrate (NAT). Large amplitude, small vertical scale temperature variations appear to be associated with the layers but not in a simple manner. In regions supersaturated with respect to NAT, condensation on only a few of the available condensation nuclei takes place. Such preferential condensation into thin large particle layers suggests that this class of nuclei may be different than the previously assumed sulfate aerosol nuclei. The possibility that they may be derived from solid fuel rocket effluent and spacecraft ablation products is examined and cannot be excluded. Such a source may explain the appearance of these cloud layers at high altitude.

The vapor pressures of supercooled HNO3/H2O solutions

A procedure utilizing the Gibbs‐Duhem relation is used to extrapolate vapor pressures of supercooled HNO3 mixtures to 190 K. Values of A and B from the equation logP = A − B/T are presented for nitric acid solutions between 0.20 and 0.25 mole fraction HNO3. In the stratosphere, if sufficient HNO3 vapor is present because it has not come into equilibrium with the nitric acid trihydrate, supercooled nitric acid solutions could condense at temperatures up to 1.5 ±0.8 K above the ice point.

Dual‐polarization airborne lidar observations of polar stratospheric cloud evolution

Dual‐polarization λ=0.532 μm lidar data show systematic polar stratospheric cloud (PSC) evolution along a portion of the Airborne Arctic Stratospheric Expedition DC‐8 flight of January 31, 1989. This flight leg was roughly aligned with air parcel motion on isentropic surfaces from 400–500K, where the local adiabatic cooling rate was ≈20 K/day. Type 1 PSCs show low depolarization ratios and scattering ratios which approach intermediate limiting values as ambient temperature decreases. These data suggest that Type 1 particles formed by rapid cooling may be nearly spherical and are restricted in size by partitioning of a limited HNO3 vapor supply among many competing growth sites. Type 2 PSCs appear at temperatures below estimated local frost points with increases in depolarization and scattering typical of larger ice crystals.

Comparison of stratospheric clouds in the Antarctic and the Arctic

Balloonborne measurements of stratospheric clouds using optical particle counters in the Arctic and the Antarctic are compared in terms of particle size distribution. There appear to be two distinct classes of particles, a small mode (r ≥ 0.2 µm) in which at least half of the available condensation nuclei have grown and a large mode (r ≈ 2–3 µm) in which fewer than 1% of the available condensation nuclei have grown. The latter particles appear in thin layers (0.1 to 0.3 km) while the former appear in relatively thicker layers (2 to 5 km). Temperatures dictate a nitric acid trihydrate composition for most of the layers and the inferred HNO3 vapor in the large particles is comparable to that available. While the small particle mode is probably related to fast cooling events such as those associated with mountain lee waves, the origin of the thin layers of large particles is not apparent.

Induction of nitrate reductase activity in red spruce needles by NO2 and HNO3 vapor

The induction of the enzyme nitrate reductase in needles may be a prerequisite for the assimilation of foliar-absorbed nitrogen oxide pollutants by red spruce (Picearubens Sarg.) trees. To test for induction of nitrate reductase, 1-year-old red spruce seedlings were exposed to NO2, HNO3 vapor, or acid mist containing nitrate, and the activity of nitrate reductase in needles was measured. One day after exposure to NO2 (75 nL•L−1) began, nitrate reductase activity was three times greater than that of unexposed control plants. One day after exposure ended, the nitrate reductase activity returned to the control level. Older red spruce seedlings that had been excavated from a spruce–fir stand exhibited a similar pattern of response, but the level of nitrate reductase activity was much lower than that of the 1-year-old seedlings. Nitric acid vapor (75 nL•L−1) also induced nitrate reductase in red spruce needles, and the pattern of response was similar to that with NO2, except that the nitrate reductase activity did not return to control levels until 2 days after exposure ended. Exposure of seedlings to acid mist containing nitrate (pH 3.5 and 5.0) did not result in a change in nitrate reductase activity. These results indicate that red spruce is capable of assimilating NO2 and HNO3 vapor and that hypotheses of forest decline based on foliar assimilation of pollutant nitrogen oxides are tenable.

Polar stratospheric clouds and the Antarctic ozone hole

A theoretical model is presented for the formation and growth of polar stratospheric clouds (PSCs). The process is assumed to occur in three stages: (1) a “precursor” stage of supercooled H2SO4‐H2O stratospheric aerosol droplets at temperatures well above the frost point; (2) an intermediate (type I) stage of PSC particle formation by codeposition of HNO3 · 3H2O at temperatures near, but above the frost point; (3) a final (type II) stage of deposition of pure water ice and HNO3 · 3H2O at temperatures below the frost point. The calculated temperature dependence of optical backscattering agrees well with values observed during two Arctic airborne lidar experiments except for small systematic errors at the 30‐mbar pressure level which may be due to poor characterization of the temperature field. Additional theoretical calculations address the issue of PSC formation in Antarctica. These results show that at the 70‐mbar level (near 17 km), about 80% and 30% of the respective HNO3 and H2O vapor available may be sequestered in relatively large (4‐μm radius) PSC particles at a temperature near 189 K, some 2 K below the estimated local frost point. Such large particles would fall at a rate of about 2 km wk−1, suggesting that PSCs may act to redistribute these trace gases in the stratosphere or perhaps remove them altogether by sedimentation to the troposphere.

Airborne lidar observations of Arctic polar stratospheric clouds: Indications of two distinct growth stages

Airborne lidar observations of Arctic polar stratospheric clouds (PSCs) during January 1984 and January 1986 show constrast suggestive of two distinct PSC growth stages delineated by the frost‐point temperature. Results obtained at temperatures 2–6K above the frost point indicate a stage of significant, but limited, particle growth such as proposed in recent models of PSC formation by co‐deposition of HNO3 and H2O vapors. Results obtained at a temperature near the frost point indicate the formation of somewhat larger crystalline particles.

Nitric acid cloud formation in the cold Antarctic stratosphere: a major cause for the springtime ‘ozone hole’

Large depletions in stratospheric ozone were first reported by Farman et al. [1] at Halley Bay (76 °S), and confirmed by satellite observations [2]. Chubachi [3] gives a detailed account of ozone decreases and temperatures in the lower stratosphere during the spring of 1982 at 69 °S. There is now evidence [2] for annual declines in total ozone by ~6 and 3 % in regions of total ozone minima and maxima, respectively, from September to mid-October since the late 1970s. We propose here a chemical mechanism for the formation of the ozone hole. It involves removal of gaseous odd nitrogen by ion- and/or aerosol-catalysed conversion of N2O5 and ClONO2 to HNO3 vapour, followed by heteromolecular HNO3–H2O condensation, leading to HNO3–H2O aerosols. At an altitude of 17 km, these processes start at temperatures below 205 ± 5 K, well above the condensation temperature of pure water vapour. We propose that the absence of gaseous odd nitrogen and catalytic methane oxidation reactions driven by sunlight in early spring lead to large OH concentrations which rapidly convert HCl to ClOX. Catalytic reactions of ClOX and BrOX cause drastic ozone destructions and can account for the springtime ‘ozone hole’ first observed by Farman et al. [1]. By our model the depletion would be mainly due to emissions of industrial organic chlorine compounds. Arctic regions may also become affected. The depletion lasts while HNO3, but not HCl, is incorporated in the particles in the temperature range 205 ± 5 to 192 K.

Spectroscopy and transmittances for the LIMS Experiment

The Limb Infrared Monitor of the Stratosphere (LIMS) instrument on the NIMBUS 7 spacecraft sensed limb emission profiles for 7 months starting in October 1979. Vertical temperature profiles are inferred from measurements in two CO2 channels, and constituent profiles are obtained from the O3, HNO3, NO2, and water vapor channels. The line parameters used to calculate transmittances in the retrieval algorithm are presented. Their adequacy for LIMS and for other applications are assessed, and recommendations for future investigations are outlined.

Tropospheric gas‐phase and particulate nitrate measurements

Measurements of HNO3 vapor and particulate NO3− were made in 1977 and in 1978 over western North America and the Pacific Ocean, both in the boundary layer and in the free troposphere. The boundary layer HNO3 ranged from &lt;0.01 to 0.86 ppbv, with a median of 0.06–0.08 in remote areas. Free tropospheric HNO3 was more uniform, ranging from &lt;0.03 to 0.54 ppbv with a median of 0.12 ppbv. The lack of an average gradient from 70°N to 55°S at 5–6 km suggests that a dispersed natural NOx source is responsible for this free tropospheric HNO3, rather than an anthropogenic one, since most urban NOx is emitted in the northern hemisphere. This and the large variability of boundary layer HNO3 with proximity to populated areas suggest that much anthropogenic NOx is removed fairly near its source. Particulate nitrate varied from &lt;0.04 to 0.8 ppbm (median 0.11) in the free troposphere and from &lt;0.04 to 3.5 ppbm (median 0.15) in the boundary layer. The molar ratio of vapor to particulate nitrate is generally much larger than 1 in the free troposphere.

Global tropospheric measurements of mitric acid vapor and particulate nitrate

Measurements of HNO3 vapor and particulate NO3− were made over continental and marine areas, at altitudes from .2 to 7 km, as a part of Project Gametag. HNO3 vapor was found to be as common over marine areas, far from anthropogenic sources of NOx, as over remote continental regions, ranging from &lt; .02 to .3 ppbv. Urban influenced continental values ranged from .3 to .86 ppbv. Comparable HNO3 was found both above and within the boundary layer, in contrast to theoretical predictions of a negative HNO3 gradient with altitude. In many regions of the troposphere there was more HNO3 vapor than participate NO3−, although the reverse is true in the marine boundary layer, where it appears that heterogeneous chemistry converts HNO3 and Cl− to HCl and particulate NO3−. This rather limited data set suggests an HNO3 lifetime on the order of a few days.

Metallic contamination of food during preparation and storage: Development of methods and some preliminary results

A chemical procedure for studying trace metals leached from metallic cooking utensils and preserving cans used in the preparation and storage of food has been developed. The method consists in the destruction of the major part of organic matter with HNO 3-vapour followed by a complete mineralization of residues with small amounts of HNO 3 in Teflon bombs at 150–160°C under a pressure of 3–12 kg/cm 2, depending on the amount and composition of the samples. Subsequently, an ion-exchange step removes major components and concentrates the trace elements in a dilute HNO 3-solution, suitable for analysis. The ion-exchange separation, which is performed with an automatic ion-exchange separator, is practically free from blank level problems, e.g., typically a mean of less than 2 per cent of the sample levels of the elements being determined. Preliminary results show that large amounts of aluminium are released from vessels to the water during boiling at the same pH-range which exist for most drinking water in Sweden.

Statistical-band-model analysis and integrated intensity for the 218 μm bands of HNO_3 vapor*

The 21.8 μm absorption bands of HNO3 vapor were measured at 40°C. Statistical-band-model analysis of the data resulted in spectral-band-model parameters and yielded an integrated intensity of 393 ± 15% (cm−2 atm−1) at 40°C between 390 and 502 cm−1.