Iron dissolution kinetics of mineral dust at low pH during simulated atmospheric processing

Abstract

Abstract. We investigated the iron (Fe) dissolution kinetics of African (Tibesti) and Asian (Beijing) dust samples at acidic pH with the aim of reproducing the low pH conditions in atmospheric aerosols. The Beijing dust and three size fractions of the Tibesti dust (<20 μm: PM20; <10 μm: PM10; and <2.5 μm: PM2.5) were dissolved at pH 1, 2 and/or 3 for up to 1000 h. In the first 10 min, all dust samples underwent an extremely fast Fe solubilisation. Subsequently, the Fe dissolution proceeded at a much slower rate before reaching a stable dissolution plateau. The time-dependant Fe dissolution datasets were best described by a model comprising three acid-extractable Fe pools each dissolving according to first-order kinetics. The dissolution rate constant k (h−1) of each pool was independent of the source (Saharan or Asian) and the size (PM20, PM10 or PM2.5) of the dust but highly dependent on pH. The "fast" Fe pool had a k (25 h−1 at pH = 1) of a similar magnitude to "dry" ferrihydrite nanoparticles and/or poorly crystalline Fe(III) oxyhydroxide, while the "intermediate" and "slow" Fe pools had k values respectively 50–60 times and 3000–4000 times smaller than the "fast" pool. The "slow" Fe pool was likely to consist of both crystalline Fe oxide phases (i.e., goethite and/or hematite) and Fe contained in the clay minerals. The initial mass of the "fast", "intermediate" and "slow" Fe pools represented respectively about 0.5–2%, 1–3% and 15–40% of the total Fe in the dust samples. Furthermore, we showed that in systems with low dust/liquid ratios, Fe can be dissolved from all three pools, whereas at high dust/liquid ratios (e.g., in aerosols), sufficient Fe may be solubilised from the "fast" phase to dominate the Fe dissolved and to suppress the dissolution of Fe from the other Fe pools. These data demonstrated that dust/liquid ratio and pH are fundamental parameters controlling Fe dissolution kinetics in the dust. In order to reduce errors in atmospheric and climate models, these fundamental controlling factors need to be included.