- Experimental studies on the adsorption of SO2 on volcanic ashes (2007)
- The adsorption of SO2 on synthetic and on natural volcanic glasses was studied. The synthetic glasses were of rhyolitic, dacitic and andesitic composition and were synthesized in a high-temperature furnace at 1600°C. The natural volcanic glasses were Lipari obsidian and Puu Waawaa obsidian. Before the adsorption experiments the glasses were ground up to powder with a planetary mill under dry conditions. The surface area of the powder then was determined with a surface area analyzer. Adsorption experiments were conducted at -80 °C, -20 °C, 0 °C, 25 °C and 150 °C. The experiments covered a pressure range from 0.1 to 984 mbar. The experiments at 0 °C (p = 0.1 to 984 mbar) and at 25 °C (p = 38 to 938 mbar) were performed with andesitic, dacitic and rhyolitic glass. Additional experiments with rhyolitic glass were carried out at 150 °C (p = 118 to 538 mbar), -20 °C (p = 75 mbar) and -80 °C (p = 46 mbar). Two experiments were performed with Lipari obsidian and Puu Waawaa obsidian respectively, each at 0 °C (p = 31 to 949 mbar). During the experiments the amount that adsorbes on the surface of the respective glass powder was determined volumetrically. For this purpose a device was designed, consisting of several glass containers, each of known volume. The glass powder was stored in the device, which then was purged with pure SO2. The amount of adsorbed SO2 then was determined from the pressure drop in the device, that occured due to adsorption. Equilibrium pressure was reached within a few hours (4-5 h). During all experiments SO2 adsorbed readily on the surface of the glasses. The adsorption isotherms from the experiments at room temperature could be classified as type II isotherms, suggesting the formation of multilayers of SO2 on the glass surface. The adsorption-desorption isotherms showed a hysteresis-like behaviour, suggesting that remarkable amounts of SO2 remain on the surface of the glass even after desorption. During the experiments at room temperature about 30 wt% of the originally adsorbed SO2 remained on the surface after desorption. XRF measurements confirmed this. Moreover, the values for the monolayer capacity Vm for SO2, that were derived from the BET isotherms, suggest that the binding of nearly the whole first monolayer was irreversible. The amount of adsorbed gas strongly depended on the temperature. It was shown that low temperatures promote the amount of adsorbed gas. An universal expression of the temperature dependence of adsorption was derived by developing a regression model for each of the synthetic glasses. The regression model is given as ln (c) = A(1/T)+ B ln p + C; where p is the pressure in mbar; c is the amount of adsorbed SO2 in mg/m2 and T is the temperature in Kelvin. According to the regression model, the amount of adsorbed SO2 varies with exp (1/T ). The precoefficients A, B and C depend on the composition of the glass, indicating that the amount of adsorbed SO2 also depends on the glass composition: Andesite: A=1644.28; B=0.29; C=-7.43; Dacite: A=2139.52; B=0.29; C=-9.32; Rhyolite: A=909.75; B=0.21; C=-4.48. The heats of adsorption Delta(Ha) for the synthetic glasses were inferred from the regression model: For rhyolite Delta (Ha) was approx. 7.6 kJ/mol, for dacite Delta (Ha) was approx. 17.8 kJ/mol and for andesite Delta(Ha) was approx. 13.7 kJ/mol. The heats of adsorption Delta(Ha) for the natural glasses were inferred from their BET isotherms: For the Lipari obsidian and for the Puu Waawaa obsidian Delta(Ha) was approx. 15 kJ/mol. The experimental adsorption data fitted both the BET equation, describing multilayeradsorption, and the Freundlich equation, desribing monolayer adsorption, quite well. For adsorption at 0 °C the following BET constants were derived: Rhyolite: C=93.4; Vm=0.32 cm3/m2; Dacite: C=5.86; Vm=0.34cm3/m2; Andesite: C=72.02; Vm=0.29cm2/m2; Lipari obsidian: C=16.00; Vm=0.33cm3/m2; Pu Waawaa obsidian: C=20.10;Vm=0.47cm3/m2. The following Freundlich Constants were derived: 1/a (in ln (mg/m2)/ ln(mbar))/ ln (k) (in ln(mg/m2)). Rhyolite: 0.25/-1.01; Dacite: 0.28/-1.47; Andesite: 0.27/-1.37; Lipari obsidian: 0.62/-3.59; Pu Waawaa obsidian: 0.64/-3.32; Experimental results provided evidence for both chemical (a) and physical (b) adsorption mechanisms: (a) the amount of adsorbed SO2 appears to depend on the glass composition and the adsorption is partially irreversible; (b) adsorption relationships, like the isotherm type and the enthalpies of adsorption are more characteristic for physical adsorption. Geological implications: Adsorption of SO2 on volcanic ash during a natural volcanic eruption mainly occurs in the umbrella region of the volcanic plume. According to the regression model, adsorption is controlled by the partial pressure of SO2 at the maximum ascent height of the plume and by the ambient stratospheric temperature prevailing at the maximum ascent height. The total amount of adsorbed SO2 depends on the total surface area of the ash suspended in the plume, which again results from the starting gas mass-fraction sigma = (mGas/mash) in the eruption column and the grain size distribution of the ejected material. A decrease in sigma = (mGas/mash) results in an increase of ash mass in the plume and thus in an increase of the total surface area of the ash suspended in the plume. The relative amount of adsorbed SO2 (e. g. the adsorbed amount, compared to the total amount in the plume) depends on the initial SO2 content x(SO2) = n(SO2)/n(Gas) in the volcanic gas: The higher the molar fraction of SO2 in the starting gas, the less is the percentage fraction of adsorbed SO2 relative to the totally available amount in the eruption column. Therefore, that if the SO2 is strongly diluted for example by water vapour it will be adsorbed by the volcanic ash very strongly, so that the impact of such eruptions on the environment is likely to be small. On the other hand, if the SO2 concentration in the volcanic gas is high, only part of it will be adsorbed and a much stronger impact of the eruption on climate is expected. For a plume model with a starting gas mass-fraction sigma = 0.03, an initial SO2 content of 1.17 Mole%, an ascent height of the eruption column of 9 km and a grain size distribution typical for Plinian eruptions, the ejected SO2 is completly adsorbed by the volcanic ash. It is likely, that the adsorbed SO2 desorbs again, except for the first monolayer, when the partial pressure of SO2 drops due to dilution. This effect probably accounts for the apparent increase in stratospheric SO2 concentration 1–2 days after an eruption, which is often observed in satellite measurements.