Abstract
Solid oxide fuel cells (SOFCs) are electrochemical energy conversion devices that generate bulk electricity and heat by converting the chemical energy of the fuel directly into electrical energy without combustion. The product of the electrochemical reaction is pure water vapour which makes the conversion process clean in the sense that it does not contribute to increasing the amount of greenhouse gases released into the atmosphere as is the case of the products of combustion reactions. If the fuel used for the SOFC is pure hydrogen that was produced by an electrolysis process utilising renewable energy sources, such as solar and/or wind energies, then the SOFC would be a considered renewable and sustainable energy source.Inkjet printing is a non-standard SOFC fabrication technique that can address high material costs by decreasing the amount of the required raw materials whilst providing high process reproducibility and speed compared to other SOFC fabrication techniques, such as powder dry pressing, nanolithography and atomic layer deposition techniques. Inkjet printing can also be up scaled for mass production.
Tape casting is a simple process for casting ceramic-based slurries to produce thick films. This process has been used to fabricate various components of SOFCs, such as anodes, electrolytes, interlayers, and cathodes.
This study was conducted with the aim to demonstrate proof-of-concept research into inkjet printing electrolyte and anode functional layer of a ceriabased SOFC on an anode support substrate fabricated using tape casting. This class of materials are important for high-efficiency intermediate-temperature SOFC fabrication.
In this research, the possibility of printing 20 mol% gadolinium-doped ceria (20GDC) as the electrolyte of a solid oxide fuel cell using a piezoelectric inkjet materials printer was investigated. A 20GDC nanopowder was used as the raw material to formulate and prepare a stable and printable ink.
A blank solid-free ink containing polyvinylpyrrolidone (PVP) as a dispersant, and a mixture of deionised water and neat (99.9%) ethanol as solvents, was first prepared to assess the printability of such a carrier.
Various concentrations of the blank ink were prepared, and the density, viscosity, and surface tension of each concentration were measured to calculate the Z-number of each concentration. The Z-number was taken as the printability indicator, and the concentration with an optimal Z-number was selected as the carrier for the solid-loaded ink. Various concentrations of the solid-loaded ink were prepared, and an ultrasonic probe was used to reduce and limit the electrolyte particle size in the ink to prevent clogging of the printer nozzles. The Z-number of the solid-loaded ink was calculated to ensure printability, again by measuring its density, viscosity, and surface tension. The particle size distribution was determined using a nanoparticle tracking analyser utilising dynamic laser scattering.
Both the blank ink and the electrolyte ink were successfully printed on an alumina substrate. Printer control parameters, such as jetting waveform, jetting frequency, resolution, nozzle voltages, cartridge temperature, and platen temperature, were optimised to allow for a constant speed satellite-free droplet ejection, resulting in uniformly printed layers. Stroboscopic imaging was used to measure the speed of the ejected droplets. The printed layers were characterised using scanning electron microscopy (SEM), optical microscopy, and surface profilometry.
To build up on the success in printing the electrolyte, an anode support consisting of nickel oxide micropowder, graphite, and 20GDC micropowder was fabricated using tape casting. An anode support slurry was prepared, and its viscosity was optimised to allow for the tape casting of a uniform 1-mm thick film. Also, an anode functional layer (AFL) ink consisting of nickel oxide micropowder and 20GDC nanopowder was prepared using the same carrier used for printing the electrolyte.
The tape cast AFL thick film was cut into 20-mm diameter buttons using ultrashort pulsed laser at 1033-nm, and then bisqued at 950 °C. The printability of the AFL ink was checked in the same manner that was used for the electrolyte ink. The AFL ink was then printed on top of the anode support and the resulting bilayer was bisqued at 950 °C. The electrolyte was then printed on top of the anode support and AFL forming a half-cell.
Sintering experiments were successfully conducted to obtain the final halfcell. Cracking and curling of the samples were observed. Layer-by-layer sintering and co-sintering in air and argon atmospheres were carried out. Deformation of the samples was observed. Laser sintering using 355-nm and 1064-nm nanosecond lasers was applied on the anode support. Sintering was observed on the lased samples and verified using Raman spectroscopy, albeit the samples were cracked.
Dilatometry studies of the anode support and the electrolyte were then undertaken in air and argon to investigate the cause of deformation of the samples after sintering. The results of the dilatometry experiments showed that the anode support cannot be sintered in argon, and that co-sintering in air is not achievable due to a difference of about 7% in the final shrinkage of the anode support and the electrolyte.
This proof-of-concept research into inkjet printing of the electrolyte and the anode functional layer of a ceria-based solid oxide fuel cell recommends conducting further research in two directions: printing of ceria-based high entropy oxide electrolytes, since these proved to be very promising in earlier computational research; and optimising sintering temperature programs and procedures for solid oxide fuel cells based on ceramic materials, including laser sintering.
Date of Award | 2024 |
---|---|
Original language | English |
Awarding Institution |
|
Supervisor | Amin Abdolvand (Supervisor) & Thomas Jones (Supervisor) |