Simon Smart

Ceramic membranes for gas processing in coal gasification

Coal is the most abundant fossil fuel in the world and is likely to outlast gas and oil for centuries. However, with global issues like climate change at the forefront of public attention there is a trend towards the development of a carbon constrained economy. As a result, research has intensified in the last decade on modes of operating coal fired power plants with carbon capture and storage (CCS). In particular, pre-combustion options via coal gasification, especially integrated gasification combined cycle (IGCC) processes, are attracting the attention of governments, industry and the research community as an attractive alternative to conventional power generation. It is possible to build an IGCC plant with CCS with conventional technologies however; these processes are energy intensive and likely to reduce power plant efficiencies. Novel ceramic membrane technologies, in particular molecular sieving silica (MSS) and pervoskite membranes, offer the opportunity to reduce efficiency losses by separating gases at high temperatures and pressures. MSS membranes can be made preferentially selective for H2, enabling both enhanced production, via a water–gas shift membrane reactor, and recovery of H2 from the syngas stream at high temperatures. They also allow CO2 to be concentrated at high pressures, reducing the compression loads for transportation and enabling simple integration with CO2 storage or sequestration operations. Perovskite membranes provide a viable alternative to cryogenic distillation for air separation by delivering the tonnage of oxygen required for coal gasification at a reduced cost. In this review we examine ceramic membrane technologies for high temperature gas separation and discuss the operational, mechanical, design and process considerations necessary for their successful integration into IGCC with CCS systems.

Cobalt oxide silica membranes for desalination

This work shows for the first time the potential of cobalt oxide silica (CoO(x)Si) membranes for desalination of brackish (1 wt.% NaCl), seawater (3.5 wt.% NaCl) and brine (7.5-15 wt.% NaCl) concentrations at feed temperatures between 25 and 75 °C. CoO(x)Si xerogels were synthesised via a sol-gel method including TEOS, cobalt nitrate hydrate and peroxide. Initial hydrothermal exposure (<2 days) of xerogels prepared with various pH (3-6) resulted in densification of the xerogel via condensation reactions within the silica matrix, with the xerogel synthesised at pH 5 the most resistant. Subsequent exposure was not found to significantly alter the pore structure of the xerogels, suggesting they were hydrostable and that the pore sizes remained at molecular sieving dimensions. Membranes were then synthesised using identical sol-gel conditions to the xerogel samples and testing showed that elevated feed temperatures resulted in increased water fluxes, whilst increasing the saline feed concentration resulted in decreased water fluxes. The maximum flux observed was 1.8 kg m(-2) h(-1) at 75 °C for a 1 wt.% NaCl feed concentration. The salt rejection was consistently in excess of 99%, independent of either the testing temperature or salt feed concentration.

Nanoporous organosilica membrane for water desalination

Nanoporous organosilica membranes are successfully coated on porous alumina tubes and tested for desalination via membrane distillation. The membranes produced pure water (up to 13 kg m(-2) h(-1)) across an extreme range of salt concentrations (10-150 g L(-1) NaCl) at moderate temperatures (≤60 °C) without exhibiting the characteristic flux decay of competing materials.

Reversible Redox Effect on Gas Permeation of Cobalt Doped Ethoxy Polysiloxane (ES40) Membranes

This work reports the remarkable effect of reversible gas molecular sieving for high temperature gas separation from cobalt doped ethoxy polysiloxane (CoES40) membranes. This effect stemmed from alternating the reducing and oxidising (redox) state of the cobalt particles embedded in the ES40 matrix. The reduced membranes gave the best H2 permeances of 1 × 10−6 mol m−2 s−1 Pa−1 and H2/N2 permselectivities of 65. The reduction process tailored a molecular gap attributed to changes in the specific volume between the reduced cobalt (Co(OH)2 and CoO) particles in the ES40 structure, thus allowing for the increased diffusion of gases. Upon re-oxidation, the tailored molecular gap became constricted as the particles reversed to Co3O4 resulting a lower gas diffusion, particularly for the larger gases ie. CO2 and N2. The ES40 matrix proved to be structurally rigid enough to withstand the reversible redox effect of cobalt particles across multiple cycles.

Combined investigation of bulk diffusion and surface exchange parameters of silver catalyst coated yttrium-doped BSCF membranes

The combined effect of minor yttrium doping and silver catalyst deposition on the surface kinetics (k(chem)) and bulk diffusion (D(chem)) of BSCF (Ba(0.5)Sr(0.5)Co(0.8)Fe(0.2)O(3-δ)) perovskite membranes was explored using electrical conductivity relaxation (ECR) and validated using oxygen permeation measurements. Yttrium doping of BSCF to form Ba(0.5)Sr(0.5)Co(0.8)Fe(0.175)Y(0.025)O(3-δ) (BSCFY) improved both the surface exchange kinetics and the bulk diffusion by an average of 44% and 177% respectively, supporting improved oxygen permeation measurements. The deposition of a silver catalyst on BSCFY further improved the surface kinetics by 63-450% at intermediate operating temperatures (600-750 °C), and reduced the activation energy from 163 to 90 kJ mol(-1). Interestingly, these improvements did not translate into enhanced oxygen fluxes for the silver coated thicker 0.5 and 1 mm membranes, indicating that the oxygen ion transport was limited by bulk diffusion. However, oxygen permeation measurements on catalyst-coated 0.3 mm-thick membranes yielded improvements of 20-35% in the range 600-900 °C. The silver catalyst was beneficial in overcoming surface kinetic limitations for the thinner 0.3 mm BSCFY membranes, thus suggesting that the critical thickness of BSCFY membranes lies around ∼0.4 mm and validating the ECR measurements.

Shortened double-walled carbon nanotubes by high-energy ball milling

Despite significant scientific interest, there are currently no widely accepted methods for the production or shortening of CNT that offer fine control over CNT length distribution. This paper reports the production of shortened double walled carbon nanotubes (DWNT) by high-energy ball milling and their characterisation via TEM, SEM, Raman, TGA and XPS techniques. Image analysis showed that ball milling was effective at shortening DWNT; however, fine control of the tube length distribution was not possible. The high-energy milling was found to lead to DWNT destruction if samples were processed for longer than 4 min. Ball milling was also found to qualitatively increase amorphous carbon content. A slight increase in side wall oxidation with increased ball milling time was observed via XPS. These well characterised DWNT samples will be employed in polymer nanocomposite and CNT toxicology studies.

Preparation, characterization and performance of templated silica membranes in non-osmotic desalination

In this work we investigate the potential of a polyethylene glycol-polypropylene glycol-polyethylene glycol, tri-block copolymer as a template for a hybrid carbon/silica membrane for use in the non-osmotic desalination of seawater. Silica samples were loaded with varying amounts of tri-block copolymer and calcined in a vacuum to carbonize the template and trap it within the silica matrix. The resultant xerogels were analyzed with FTIR, Thermogravimetric analysis (TGA) and N2 sorption techniques, wherein it was determined that template loadings of 10 and 20% produced silica networks with enhanced pore volumes and appropriately sized pores for desalination. Membranes were created via two different routes and tested with feed concentrations of 3, 10 and 35 ppk of NaCl at room temperature employing a transmembrane pressure drop of <1 atm. All membranes demonstrated a salt rejection capacity of >85% (in most cases >95%) and fluxes higher than 1.6 kg m−2 h−1. Furthermore, the carbonized templated membranes displayed equal or improved performance compared to similarly prepared non-templated silica membranes, with the best results of a flux of 3.7 kg m−2 h−1 with 98.5% salt rejection capacity, exceeding previous literature reports. In addition, the templated silica membranes exhibited superior hydrostability demonstrating their potential for long-term operation.

Iron Oxide Silica Derived from Sol-Gel Synthesis

In this work we investigate the effect of iron oxide embedded in silica matrices as a function of Fe/Si molar ratio and sol pH. To achieve homogeneous dispersion of iron oxide particles, iron nitrate nonahydrate was dissolved in hydrogen peroxide and was mixed with tetraethyl orthosilicate and ethanol in a sol-gel synthesis method. Increasing the calcination temperature led to a reduction in surface area, although the average pore radius remained almost constant at about 10 Å, independent of the Fe/Si molar ratio or sol pH. Hence, the densification of the matrix was accompanied by similar reduction in pore volume. However, calcination at 700 °C resulted in samples with similar surface area though the iron oxide content increased from 5% to 50% Fe/Si molar ratio. As metal oxide particles have lower surface area than polymeric silica structures, these results strongly suggest that the iron oxides opposed the silica structure collapse. The effect of sol pH was found to be less significant than the Fe/Si molar ratio in the formation of molecular sieve structures derived from iron oxide silica.

Synthesis of mesoporous carbon–silica nanocomposite water-treatment membranes using a triconstituent co-assembly method

A direct synthesis method is introduced to prepare mesoporous carbon–silica nanocomposite (CSN) membranes for water-treatment applications. Unlike the intricate and expensive nanocasting method, this triconstituent co-assembly method is a one-pot synthesis method using Pluronic F127 as the templating agent with a hybrid organic–inorganic matrix formed by tetraethylorthosilicate (TEOS), resorcinol and formaldehyde. The silica content is varied in the polymer solution to investigate the material properties, stability of the nanocomposite mesostructure and membrane performance in vacuum membrane distillation (VMD). The CSN materials are carbonised under nitrogen at temperatures of 600–900 °C without any significant lattice shrinkage, demonstrating excellent stability. They possess a highly ordered porous structure with moderate BET surface area (430–550 m2 g−1) and narrow pore size distribution at around 5.5–7.6 nm. Based on the FTIR and NMR analyses, there is no covalent bond between the carbon and silica networks, but the carbon compound was found to affect the condensation degree of the silica. Raising the temperature from 700 to 900 °C leads to further condensation of the carbon network, which in turn releases hydroxyl or water groups that can attack adjacent siloxane bonds. The CSN membranes performed well in VMD with water permeation flux up to 12 L m−2 h−1 and salt rejection >99%. This work shows that a different strategy of modifying silica-based membrane can be successfully applied for the desalination of saline waters through VMD.

Porous ceramic membranes for membrane reactors

This chapter discusses the research and development of porous ceramic membranes and their application as membrane reactors (MRs) for both gas and liquid phase reaction and separation. The most commonly used preparation techniques for the synthesis of porous ceramic membranes are introduced first followed by a discussion of the various techniques used to characterise the membrane microstructure, pore network, permeation and separation behaviour. To further understand the structure-property relationships involved, an overview of the relevant gas transport mechanisms is presented here. Studies involving porous ceramic MRs are then reviewed. Of importance here is that while the general mesoporous nature of these membranes does not allow excellent separation, they are still more than capable of enhancing reaction conversion and selectivity by acting as either a product separator or reactant distributor. The chapter closes by presenting the future research directions and considerations of porous ceramic MRs.