Occurrence and behaviour of colloidal silica and silica-rich nanoparticles through stages of reverse osmosis treating coal seam gas associated water

Reverse Osmosis (RO) membrane filtration is a very common process for treating a wide range of groundwater types including produced water from coal seam gas (coalbed methane) wells. Mineral scaling limits water recovery for RO membranes and costs money in terms of treatment and downtime. Silica scaling can be particularly troublesome as it is often irreversible. Mitigating silica scaling requires an understanding of its occurrence, speciation mechanism and its interdependency with other operation factors. This study uses a range of techniques to show that silica colloids form during later stages of an RO process with very high recovery. This happens at silica concentrations above the solubility that would normally indicate high risk of silica scale. However, instead of scale, colloids preferentially formed which means the process can operate at high recoveries with RO performance maintained by regular cleaning cycles. The concentration of the colloidal silica through the RO stages was measured through the difference in total and dissolved silica. Once the existence was established with this technique, the particles were trapped and their size, morphology and composition were investigated with Scanning Electron Microscopy (SEM) in conjunction with Energy Dispersive X-Ray Spectroscopy (EDS). This revealed the particles to be predominantly silica with limited other elements involved.

Silica fouling during groundwater RO treatment: The effect of colloids' radius of curvature on dissolution and polymerisation

Silica fouling during groundwater reverse osmosis (RO) treatment can have a significant impact on filtration performance. To better understand this phenomenon, the equilibrium kinetics of amorphous colloidal silica were studied at conditions relevant to RO of silica-rich alkaline groundwater. The impact of particle size was investigated using synthetic monodisperse silica nanoparticles. Bench scale experiments were conducted by monitoring dissolved silica concentration of aqueous suspensions of colloids of 100 and 300 nm diameter and pH 8.5 to 9.5. The equilibrium data was determined from existing established rate law equations. This study concluded that surface energy has a major impact on silica dissolution rate constant, particularly for colloidal silica. Observations of Ostwald ripening in bidisperse silica dispersions further confirmed these results, which indicate that dissolution and redeposition is responsible for the problematic silica fouling behaviour during RO treatment. 2D modelling based on inferred equilibrium data allows visualization of scale layer growth in agreement with cross-sectional scanning electron micrographs of autopsied membranes.

Capacitance of Nanoporous Carbon-Based Supercapacitors Is a Trade-Off between the Concentration and the Separability of the Ions

Nanoporous carbon-based supercapacitors store electricity through adsorption of ions from the electrolyte at the surface of the electrodes. Room temperature ionic liquids, which show the largest ion concentrations among organic liquid electrolytes, should in principle yield larger capacitances. Here, we show by using electrochemical measurements that the capacitance is not significantly affected when switching from a pure ionic liquid to a conventional organic electrolyte using the same ionic species. By performing additional molecular dynamics simulations, we interpret this result as an increasing difficulty of separating ions of opposite charges when they are more concentrated, that is, in the absence of a solvent that screens the Coulombic interactions. The charging mechanism consistently changes with ion concentration, switching from counterion adsorption in the diluted organic electrolyte to ion exchange in the pure ionic liquid. Contrarily to the capacitance, in-pore diffusion coefficients largely depend on the composition, with a noticeable slowing of the dynamics in the pure ionic liquid.

Mixed Matrix Carbon Molecular Sieve and Alumina (CMS-Al2O3) Membranes

This work shows mixed matrix inorganic membranes prepared by the vacuum-assisted impregnation method, where phenolic resin precursors filled the pore of α-alumina substrates. Upon carbonisation, the phenolic resin decomposed into several fragments derived from the backbone of the resin matrix. The final stages of decomposition (>650 °C) led to a formation of carbon molecular sieve (CMS) structures, reaching the lowest average pore sizes of ~5 Å at carbonisation temperatures of 700 °C. The combination of vacuum-assisted impregnation and carbonisation led to the formation of mixed matrix of CMS and α-alumina particles (CMS-Al2O3) in a single membrane. These membranes were tested for pervaporative desalination and gave very high water fluxes of up to 25 kg m(-2) h(-1) for seawater (NaCl 3.5 wt%) at 75 °C. Salt rejection was also very high varying between 93-99% depending on temperature and feed salt concentration. Interestingly, the water fluxes remained almost constant and were not affected as feed salt concentration increased from 0.3, 1 and 3.5 wt%.

Downstream processing of reverse osmosis brine: Characterisation of potential scaling compounds

Reverse osmosis (RO) brine produced at a full-scale coal seam gas (CSG) water treatment facility was characterized with spectroscopic and other analytical techniques. A number of potential scalants including silica, calcium, magnesium, sulphates and carbonates, all of which were present in dissolved and non-dissolved forms, were characterized. The presence of spherical particles with a size range of 10-1000 nm and aggregates of 1-10 microns was confirmed by transmission electron microscopy (TEM). Those particulates contained the following metals in decreasing order: K, Si, Sr, Ca, B, Ba, Mg, P, and S. Characterization showed that nearly one-third of the total silicon in the brine was present in the particulates. Further, analysis of the RO brine suggested supersaturation and precipitation of metal carbonates and sulphates during the RO process should take place and could be responsible for subsequently capturing silica in the solid phase. However, the precipitation of crystalline carbonates and sulphates are complex. X-ray diffraction analysis did not confirm the presence of common calcium carbonates or sulphates but instead showed the presence of a suite of complex minerals, to which amorphous silica and/or silica rich compounds could have adhered. A filtration study showed that majority of the siliceous particles were less than 220 nm in size, but could still be potentially captured using a low molecular weight ultrafiltration membrane.

New method to determine PSD using supercritical adsorption: Applied to methane adsorption in activated carbon

Adsorption of supercritical fluids is increasingly carried out to determine the micropore size distribution. This is largely motivated by the advances in the use of supercritical adsorption in high energy applications, such as hydrogen and methane storage in porous media. Experimental data are reported as mass excess versus pressure, and when these data are matched against the theoretical mass excess, significant errors could occur if the void volume used in the calculation of the experimental mass excess is incorrectly determined [Malbrunot, P.; Vidal, D.; Vermesse, J.; Chahine, R.; Bose, T. K. Langmuir 1997, 13, 539]. The incorrect value for the void volume leads to a wrong description of the maximum in the plot of mass excess versus pressure as well as the part of the isotherm over the pressure region where the isotherm is decreasing. Because of this uncertainty in the maximum and the decreasing part of the isotherm, we propose a new method in which the problems associated with this are completely avoided. Our method involves only the relationship between the amount that is introduced into the adsorption cell and the equilibrium pressure. This information of "direct" experimental data has two distinct advantages. The first is that the data is the "raw" data without any manipulation (i.e., involving further calculations), and the second one is that this relationship always monotonically increases with pressure. We will illustrate this new method with the adsorption data of methane in a commercial sample of activated carbon.