Aerosol Chamber Experiments on the Chemical Composition of ... In this paper we summarise briefly the experimental procedures and discuss preliminary ... samples, particle size distributions by differential mobility particle sizing (DMPS), ... (ABMS), which yielded HNO3/H2O molar ratios with a time resolution of about 15 s.
Aerosol Chamber Experiments on the Chemical Composition of Polar Stratospheric Cloud Particles R. Tiede1, O. Möhler1, A. Nink1, M. Schnaiter1, J. Schreiner2, P. Zink2, U. Schurath1 and K. Mauersberger2 1. Forschungszentrum Karlsruhe (FZK), IMK3, Karlsruhe, Germany 2. Max-Planck-Institut für Kernphysik (MPI-K), Atmosphärenphysik, Heidelberg, Germany 1. Introduction The impact of polar stratospheric clouds (PSC) on stratospheric ozone depletion depends on their chemical composition and phase. In the large AIDA aerosol chamber of FZK, aerosol processes can be investigated at stratospheric temperatures (-60°C to –90°C), pressures (100 to 140 hPa), and trace gas concentrations. During four intensive measurement campaigns we studied supercooled binary H2SO4/H2O and ternary H2SO4/H2O/HNO3 aerosol systems. In this paper we summarise briefly the experimental procedures and discuss preliminary results of a ternary aerosol experiment which was part of the most recent campaign in the AIDA aerosol chamber. 2. Experimental The AIDA aerosol chamber has a volume of 84.3 m3 (height ca. 7 m, diameter 4 m). Due to its large volume the residence time of sub-µm particles exceeds 10 h under simulated stratospheric conditions. The chamber can be thermostated between 183 K and 330 K with a homogeneity (vertical and horizontal) significantly better than ∆T = 0.5 K. Temperature homogeneity in the gas phase is monitored by 20 temperature sensors. By pumping the chamber (e.g. from 140 to 120 hPa, like shown in Fig. 1 at t=4 h), it is possible to induce adiabatic cooling of the aerosol without changing the wall temperature. Adiabatic cooling can be maintained over time periods of several minutes, yielding water vapour supersaturations ≤1.6 with respect to water ice. Similarly, adiabatic warming is induced when the initial pressure is re-established by adding synthetic air to the chamber. Water vapour mixing ratios are measured with a frost point hygrometer. During the most recent AIDA campaign intercalibrations were carried out with the FISH Lyman-α hygrometer of FZJ (cf. [1] and [2]). Preliminary analysis of the data showed reasonable agreement. Sulphuric acid particles in the AIDA chamber are log-normally distributed with a mean diameter of 0.2 µm, σ=1.5, initial number concentration ca. 4000 cm-3. Due to the relatively low number concentration coagulation of particles does not play an important role during the course of an experiment. Total sulphate concentrations were determined by analysing filter samples, particle size distributions by differential mobility particle sizing (DMPS), and particle number concentrations by condensation particle counting (CNC). Chemical composition of the particles was analysed by Aerosol Beam Mass Spectrometry (ABMS), which yielded HNO3/H2O molar ratios with a time resolution of about 15 s. The instrument is the AIDA version of another ABMS system which has been described elsewhere [3]. HNO3 vapour was measured by Chemical Ionisation Mass Spectrometry (CIMS). HNO3 is very efficiently adsorbed on the cold aerosol chamber surface and in cold sampling lines. To eliminate HNO3 losses in the Teflon sampling line of the CIMS, it was necessary to heat the line to about 20°C. Since HNO3 is rapidly evaporated by the aerosol particles at this temperature, [HNO3]total = [HNO3]gas + [HNO3]particle is determined by the CIMS system. Due to adsorption at the chamber walls the partial pressure of HNO3 rapidly drops below 1 ppb when a small amount of HNO3 is added to the cold AIDA chamber. To keep [HNO3] in
the stratospherically relevant range of a few ppb it was necessary to add HNO3 vapour continuously to the gas phase throughout the duration of the experiment. In the beginning, i.e. before sulphuric acid aerosol was added, a dynamic equilibrium of fluxes was thus established, i.e. Fwall = Fin
(eqn. 1)
With Fwall = flux to the walls, Fin = constant influx of HNO3 vapour into the chamber, Fwall = kwall [HNO3],
(eqn. 2)
thus leading to constant [HNO3]gas if kwall = constant, which appeared to be the case. When this steady-state was established, sulphuric acid aerosol was added (from t=-0.4h to t=0h in Figs. 1 and 2), thus shifting the steady-state: [ HNO3 ] gas =
k wall
Fin + k particle
(eqn. 3)
Because the particles eventually saturate with HNO3, kparticle and thus [HNO3]gas are functions of time. Temperature changes induced by adiabatic cooling or warming also lead to uptake or release of HNO3 by the particles, and thus to changes of [HNO3]gas. However, if (at least after some time) kwall » kparticle ,
(eqn. 4)
[HNO3]gas will relax to the initial steady-state again. If this assumption holds while sulphuric acid aerosol is added, or at least while the particle composition changes due to temperature changes, it is possible to analyse the kinetics of HNO3 uptake and release by the particles. The preliminary results suggest so far that eqn. 4 holds at least during the adiabatic cooling and warming events. However, this has to be verified in the future. The experimental procedure was then as follows: • The chamber was filled with 140 hPa synthetic air at 200 K, yielding ca. 100% r.h. with respect to traces of ice on the chamber walls. • The constant influx of HNO3 vapour into the chamber was started. • When a constant level of [HNO3]gas had established sulphuric acid aerosol was added. • When after 4 hours a constant particle composition had established, the steady-state conditions were shifted by changing the gas phase temperature adiabatically for several minutes, as described above, and as depicted in Fig. 1. Prior to and after these perturbations the aerosol composition and [HNO3]total relaxed back to their initial steady-state values.
Figure 1. Pressure and average gas temperature during the experiment. At t = 0 h sulphuric acid particles were added into the chamber. Between 4 and 6.5 h the temperature was disturbed by first adiabatic cooling and then adiabatic warming.
3. Results By constantly adding HNO3 vapour to the chamber a steady-state gaseous concentration of 1.5 ppbv was achieved. After introduction of the sulphuric acid aerosol the sum concentration of nitric acid in the particles and in the gas phase [HNO3]total increased to ca. 15 ppbv, indicating that at 200 K 90% of the nitric acid resides in the particles, Fig. 2. [HNO3]total then decreases with approximately the same rate as the particle number concentration. Assuming that the size distribution does not change significantly, which is reasonable for particle concentrations of a few 1000 cm-3, it is possible to calculate the specific nitric acid content of the condensed phase, [HNO3]particle, which is plotted in Fig. 3.
Figure 2. Total nitric acid and particle number concentrations.
Figure 3. Specific nitric acid concentration and molar ratio of nitric acid to water
This figure also depicts the molar ratio HNO3/H2O as measured with the ABMS: After reaching steady-state around t=4 h the adiabatic cooling event (↓) leads to rapid uptake of nitric acid by the particles. Water vapour uptake is faster than the uptake of nitric acid, therefore the molar ratio first decreases, then increases again when the temperature returns to its initial value. The opposite behaviour is observed during the warming event (↑), i.e. nitric acid is released from the particles. Note that after the short periods of cooling (↓) and warming (↑), the relaxation of the molar ratio HNO3/H2O and of [HNO3]particle back to their equilibrium values occurs on a very long time scale! 4. Conclusions and further work The preliminary analysis of our data shows that it is possible to study the kinetics of nitric acid exchange between gas and particle phases in the AIDA chamber under simulated stratospheric conditions. Further experiments on binary (HNO3/H2O) and ternary (H2SO4/H2O/HNO3) supercooled aerosol systems, which were made at different temperatures and relative humidities during our recent AIDA campaign, are awaiting analysis. Preliminary calculations with the equilibrium model of Luo et al. [4] were in good agreement with the steady-state concentrations observed in this work. However, model calculations of mass transfer rates under non-equilibrium conditions must also be carried out and compared with the uptake and evaporation rates triggered by the adiabatic cooling and warming events. Further experimental work in AIDA will focus on the formation of solid phases and on ice nucleation processes. References [1] Zöger M., et al. 1999, Fast in situ stratospheric hygrometers: A new family of balloon-borne and airborne Lyman a photofragment fluorescence hygrometers, J. Geophys. Res., 104, 1807-1816 [2] Schiller C., this issue [3] Schreiner J., et al., 1999, Chemical analysis of polar stratospheric cloud particles, Science, 283, 968-970 [4] Luo B., et al., 1999, Vapour pressures of H2SO4/HNO3/HCl/HBr/H2O solutions to low stratospheric temperatures, Geophys. Res. Let., 22, 247-250
