Simultaneous removal of dissolved sulphide and dissolved methane from anaerobic effluents with hollow fibre membrane contactors.

Dissolved gases in the effluent of anaerobic reactors, specifically dissolved methane (D-CH4) and sulphide (S2-), are a drawback for anaerobic-based sewage treatment plants (STPs). This article studied the simultaneous desorption/removal of both gases from anaerobic effluents with hollow fibre membrane contactors (HFMCs), evaluating two types of membrane materials (e.g. microporous and dense) at different operating conditions (atmospheric air as sweeping gas or vacuum, and different gas/liquid flows and vacuum pressures). The transfer of other gases, such as O2 and CO2, was studied as well. Desorption/removal efficiencies up to 99% for D-CH4 and 100% for S2- were obtained, with the higher efficiencies reported for the dense HFMC and with air as sweeping gas. It was found that the removal mechanism for S2- was oxidation with O2 from the air. In addition, the use of air as sweeping gas allowed the obtention of a nearly O2 saturated effluent, with more elevated dissolved oxygen concentrations in the microporous HFMC. Finally, it was found that the higher mass-transfer resistance in the dense membrane was compensated by a better performance in the liquid phase (lower mass-transfer resistance) in this unit, which allowed better D-CH4 desorption efficiencies.

Textural and geochemical constraints on andesitic plug emplacement prior to the 2004-2010 vulcanian explosions at Galeras volcano, Colombia.

Hazardous sequences of vulcanian explosions are thought to result from the repeated emplacement and destruction of degassed, highly crystalline magma plugs in the shallow conduit of arc volcanoes. The processes governing the timing and magnitude of these explosions are thought to be related to magma ascent rate and efficiency of degassing and crystallisation. We study a rare suite of time-constrained ballistic bombs from the 2004-2010 period of activity of Galeras volcano to reconstruct magma plug architecture prior to six individual vulcanian explosions. We find that each plug was vertically stratified with respect to crystallinity, vesicularity and melt volatile content, melt composition and viscosity. We interpret this structure as resulting from multiple bubble nucleation events and degassing-driven crystallisation during multi-step ascent of the magma forming the plug, followed by spatially variable crystallisation within the plug under contrasting conditions of effective undercooling created by degassing. We propose that the shallow conduit evolved from more open degassing conditions during 2004-2008 to more closed conditions during 2009-2010. This resulted in explosions becoming smaller and less frequent over time during 2004-2008, then larger and more frequent over time during 2009-2010. This evolution was controlled by changing average ascent rates and is recorded by systematic changes in plagioclase microlite textures. Our results suggest that small volume vulcanian explosions (~ 105 m3) should generally be associated with longer repose times (hundreds of days) and produce ballistics characterised by small numbers of large, prismatic plagioclase microlites. Larger volume vulcanian explosions (~ 106 m3) should be associated with shorter repose times (tens of days) and produce ballistics characterised by high numbers of small, more tabular plagioclase microlites.