Self-healing ceramic membranes with increased lifetime for CO₂ capture in industrial processes and power production

Goal of the Project:
The SEALEM project aims at a radical improvement of lifetime and performance of ceramic gas separation membranes by developing ground-breaking self-healing mechanisms, better understanding of degradation phenomena based on chemical stresses and cation diffusion, and improving bulk transport and kinetics. The project educates one PhD candidate and trains one post-doctoral researcher.
Technical content:
The project contains five R&D sub-projects (SP) with the following technical content: SP1: Self-healing membranes: Manufacturing of model membranes and tubular membranes and investigation of two proprietary mechanisms for self-healing. SP2: Oxygen and hydrogen transport: Bulk transport and surface kinetics. SP3: Lifetime analysis: Measurement of cations diffusion, kinetic decomposition study, and evaluation of chemical stresses and effects on membrane integrity to perform lifetime analysis. SP4: Development of innovative prototype module: Design and components selection of new rig enabling in-situ heating, building and commissioning of single tube rig and assessment of single tube rig. SP5: Management, IPRs and dissemination: Overall management of the project, dissemination, networking and IPR handling.

R&D challenges:
Membrane efficiency relates to a high flux density, which in turn is given by a high mixed (ambipolar) transport of ions and electrons. The membrane should be as thin as possible, but in order to utilise this, its surface kinetics must be fast. Additionally, the demands on robustness of the membrane increase as the membranes get thinner. This is firstly because the chemical potential differences over the membrane get distributed over a shorter length giving rise to higher chemical potential gradients and hence faster destructive processes related to cation diffusion. The latter are related to the so-called chemical creep, or walkout of the membranes, which sets the ultimate limitation of an otherwise well-behaving membrane. Difference in diffusion between cations may also lead to demixing and possibly decomposition of the compound when exposed to a chemical gradient, thereby leading to accelerated dysfunction and possibly breakdown of the membrane.
Another important hurdle for oxygen and hydrogen transport membranes arises from chemical stresses when the membrane material is exposed to two gases with different compositions. The oxygen transport membranes (OTMs) tend to expand on the side with lower oxygen content, and the hydrogen transport membranes tend to expand on the side with higher water vapour pressure. In both cases, this leads to mechanical stresses, which either relax due to creep or may lead to sub-critical crack growth and finally complete mechanical failure of the membrane.
Finally, a thin membrane has larger chance of pinholes, caused for instance by foreign particles, during manufacturing steps as well as by cation diffusion due to enhanced chemical gradients locally.
In order to have more robust membranes the project tests novel self-healing strategies and the first stages towards modules of tubes that tolerate high pressure gradients, have cold seals, can be heated internally, and replaced individually for prolonged module lifetime.

Results to date:
The project started medio 2013. To date, Internal ohmic heating has been successfully applied to asymmetric tubular membranes for both hydrogen and oxygen transport, and in situ sealing of the tubes has been achieved. The tubes can be internally heated up to 800-900°C, depending on the current and voltage applied. Furthermore, gas permeation experiments have been conducted during internal heating, indicating that the membrane can be operated under these conditions. A paper on this topic is under preparation. We have finished the design and prototype construction of a single tube sample holder specialized for internal ohmic heating and simple replacement of faulty tubes.

For self-healing of ceramic membranes, a number of candidate materials have been tested and the proof of concept has been established for a few of the materials systems. When testing and screening the candidate systems we have simultaneously focused on preparation of disc assemblies with thin films - serving as model systems - and manufacturing of real tubular multi-layered membranes. The model disc assemblies are prepared by pulsed laser deposition (PLD) or tape casting whereas the tubular membranes are produced using green methods combining water based extrusion and dip-coating. The self-healing process is characterized and quantified ex situ by analysis of the layered cross section before and after high temperature annealing, as well as by inspection of artificial pinholes which are introduced in the membranes by ion milling. In situ verification of the self-healing process by using leakage tests of sealed samples in a chemical gradient is furthermore an important part of the experimental work. One of the self-healing systems investigated the past 1,5 years is particularly promising as it forms a dense sealing cap on one side of a faulty membrane at a temperature and within a timeframe that is operationally realistic. By optimizing the layered architecture and its microstructure, this system is furthermore able to self-assemble a dense, dual-phase OTM. A patent application based on the self-healing concepts was filed 18. March 2016.

In addition to the paper published two years ago on the thermo-chemical expansion of BSCF, the surface diffusion study entailing BSCF and La2NiO4 has been finalized this past year and an article was recently accepted for publication in J. Am. Cer. Soc. The surface diffusion of these materials is studied by grain boundary grooving where the variation in grain boundary groove width with annealing temperature and time is used to determine the materials’ surface diffusion rate. The study of the interaction of proton conducting oxides with hydrogen and water – focusing on the correlation between bulk and surface properties – is furthermore advancing and the results were presented at the 18th International Conference on Solid State Protonic Conductors in September 2016.