Research Staff: Jin-Sheng Lin and Azadeh Tabazadeh

Up to now laboratory experiments under upper tropospheric conditions have only measured the deliquescence and efflorescence relative humidities of binary ammoniated salts in the absence of gas phase nitric acid. In this research we use a thermodynamic electrolyte model to examine how HNO3, which is abundant in the atmosphere, can affect the deliquescence and efflorescence properties of binary ammoniated salts in the upper troposphere.

Traditionally the deliquescence relative humidity (DRH) for a binary salt is defined as the relative humidity at which a dry salt particle instantaneously turns into an aqueous solution droplet. In fact many laboratory observations at or near room temperature show that binary salts do abruptly change phase at a fixed RH known as the DRH. However, in the atmosphere such a definition is too simple because a dried up salt particle is exposed to other gaseous components (besides just H2O vapor which determines the RH) such as gas phase HNO3. For example, in previous research we have shown that HNO3 is not only highly soluble in an aqueous solution of (NH4)2SO4 at cold upper tropospheric temperatures, but it can interact with ammonium and sulfate ions in the aqueous salt solution to form (NH4)3H(SO4)2 (1ectivicite). Letovicite is a new salt that cannot form in a pure aqueous (NH4)2SO4 solution, which is only exposed to gas phase H2O, because this salt formation requires the presence of H+ in solution.

In Figure 1a the variation of the DRH is shown for the (NH4)2SO4 system both in the presence and in the absence of gas phase HNO3. The HNO3 concentration ranges from zero to background (100pptv) and polluted (up to 2ppbv) levels in the upper troposphere. The green line shows the standard DRH where the gas phase HNO3 is set to zero in the calculations, which is the condition used in all the laboratory experiments to date. The black lines show the effect on the DRH once HNO3 is introduced into the gas phase. It is clear from the results shown in Figure 1a that HNO3 can strongly affect the DRH behavior of the (NH4)2SO4 system. Similar results are shown in Figure 1b for the NH4HSO4 system. Overall, the more HNO3 gas phase in the atmosphere, the more it can affect the DRH behavior of the binary ammoniated aerosol system. In addition, as shown in Figure 1 there is no abrupt change in the physical state of the thermodynamic system (instantaneous change of a dry salt particle into an aqueous solution) at the deliquescent point, which is typical for such a phase transition in the presence of only water vapor.

Exposure of ammoniated salts to nitric acid can also affect the efflorescence relative humidity (ERH) of binary ammoniated salts in the upper troposphere. Efflorescence is the RH at which a binary aqueous solution of a salt instantaneously turns into the crystalline state. Figure 2 illustrates how saturation ratios (crystalline) for (NH4)2SO4 and NH4HSO4, both at background mixing ratio levels, change with relative humidity and HNO3 concentration at 230 K. Crystalline states of a salt in solution can only form when the saturation ratio (SR) of the salt exceeds unity. As shown in Figure 2, SR of letovicite is higher than that of either of the binary salts in the aqueous solution by a factor of over 10 for RHs above 50 %, indicating that the likelihood for letovicite to crystallize from these solutions is favored over the binary salts.

The results of this work show that both the DRH and ERH of binary ammoniated salts in the upper troposphere are significantly impacted by gas phase HNO3. The DRH cannot occur via the standard abrupt change in the physical state of the system (a salt particle quickly turning into an aqueous solution droplet). In addition, it will be unlikely for either (NH4)2SO4 or NH4HSO4 to crystallize in upper tropospheric aerosol solutions. Instead, the crystalline state of both salts in the upper troposphere is letovicite which can then co-exist with an aqueous solution. Thus, the state of aerosols (containing both ammonium and sulfate ions) is often mixed in the upper troposphere. Both the ice nucleating properties and the heterogeneous reactivity of upper tropospheric particles are dependent upon the aerosol solution phase and composition and therefore the effect of HNO3 should be considered to accurately predict the physical state of ammoniated aerosol particles in the upper troposphere.

Figure 1. The effect of HNO3 uptake on the deliquescence of (NH4)2SO4 (a) and NH4HSO4 (b) in the upper troposphere. The ice nucleation and ice saturation lines are marked on the plot. The green lines show the standard DRH behavior in the absence of HNO3. The black lines show the effect of HNO3 uptake for a gas phase mixing ratio of ~ 100 pptv and 2 ppbv. Lines marked as S show the point where the aqueous solutions first appears. Line L (in panels a and b) marks the point where letovicite will fully dissolve in the aqueous solution. Between the lines S and L the equilibrium system is mixed phase. Below the line labeled as S the dry salt will be composed of letovicite plus NH4NO3 and NH4HSO4 in panels a and b, respectively. See the text for more detail. Model calculations were performed using a fixed binary salt concentration of 10-9 mole m-3 (~ 100 ppt at 200 mb) for all panels. The HNO3 concentration was set to zero for the standard calculations and was varied from 1 (~ 100 ppt at 200 mb) to 20 (~ 2 ppbv at 200 mb) x 10-9 moles m-3 in the HNO3 simulations. Note that in order to obtain the results shown neither H2SO4 nor HNO3 hydrates were allowed to from during the model simulations.

Figure 2. The effect of HNO3 uptake on the efflorescence of (NH4)2SO4 (a) and NH4HSO4 (b) solution in the upper troposphere at 230 K. The variations in the salt saturation ratio as a function of RH are also shown. The room temperature (RT) ERH for both (NH4)2SO4 and letovicite is shown as a dashed line. The ice saturation line at 230 K is also marked. The shaded area marks the region where both ice and ammoniated salts are saturated in the aqueous solution. See text for more detail. The model conditions used for binary salts and various HNO3 concentrations are described in the caption for Figure 1. In order to obtain salt saturation ratios in solution we have assumed that the solution was supercooled and therefore neither salts nor ice were allowed to form during the simulations.