To enable transition to a low-carbon economy of clean and versatile power conversion devices, we are optimising the designs of combustors and energy convertors for future engines, gas turbines and micro-power systems.
Our research addresses issues of energy security and environmental pollution – two of the most significant social challenges of the 21st century.
We develop and employ advanced experimental and numerical methodologies to improve the understanding of the fundamentals leading to the clean and efficient use of diverse fuel formulations including existing fossil fuels, future fuels and renewables.
We aim to optimise power conversion technologies and contribute to the development of future energy systems including near-zero emission devices, micro-reactors and fuel cells.
Other important focus areas of our research include combustion safety, fire suppression, and reduction of fire and explosion risks.
Our ultimate goal is to develop novel, clean and sustainable energy solutions that ensure transition to a low emissions economy. Specific topics of research include:
An advanced laser diagnostics facility exists within the Clean Combustion Laboratory in parallel with advanced computational fluid dynamic and experimental capabilities for flow and fire simulation, burner development and engine testing.
Consulting work may be carried on a range of industrial problems including air conditioning, heat transfer and fire safety.
This project provides a novel understanding of a turbulent combustion environment that spans the entire regime from non-premixed to stratified and then premixed. This is highly relevant to lean combustion gas turbines and direct-injection engines where mixed-mode combustion prevails. The burner provides a challenging platform for combustion modellers where the requirement is for a single-model to be able to compute all modes of combustion.
The project is highly relevant to combustion in direct-injection engines and gas turbines where compositional inhomogeneity and high rates of shear dominate the flow within the combustion chamber. Developing a database for model validation in well-controlled burners is important.
This is made possible with the piloted inhomogeneous burner designed and built at the University of Sydney. In the burner, the reacting mixture is introduced through two annular tubes such that the inner tubes can slide upstream of the jet exit plane. Both shear and compositional inhomogeneity can be introduced independently by varying the composition, the velocity ratio and the recess distances of the inner tube with respect to the annulus. A parallel experimental and numerical program is continuing to explore various research aspects of mixed-mode flames and to enable their computations.
Standard backlight shadowgraph imaging is performed to dual angle and double-pulse to enable the direct visualisation of the shape, volume and velocity of liquid fragments of irregular shapes similar to those shown here. This dual-angle particle tracking velocimetry (PTV) approach provides extremely useful information on spray morphology including quantification of volume flux in atomising sprays. The experimental layout consists of two time-shifted lasers, each split into two beams, two long distance microscope lenses and two cameras oriented 90-degrees to each other, operating in PIV mode. The accuracy of the joint volume velocity measurements has been carefully assessed using mono-dispersed droplets and micro spheres of known size.
The technique has also been examined for different air-assisted sprays covering regimes from Rayleigh to multi-mode breakup. The correlation between exit Weber number and fragment velocity indicates that this technique is capable of identifying the transitions from one well-established breakup-regime to another. By introducing terminologies such as fragment residence time and fragment volume flow rate, the overall volume flow rate of both the mono-dispersed drops as well as low Weber number air-assisted sprays may be recovered. This PIV method is a powerful diagnostic tool to simultaneously track and size arbitrarily shaped liquid-fragments. It also provides a viable technique to measure liquid mass flux in the near-field of sprays. Such information is valuable for the development of modelling capabilities for spray jets and flames.
Auto-ignition is the key mechanism that initiates combustion in compression ignition engines. This project is aimed at resolving the complexities of the process with respect to a range of fuels and coflow conditions. A jet-in-hot-coflow burner has been developed for this purpose and been used to study a range of gaseous and liquid fuels over a range of coflow conditions. The figure shows flames of dimethyl ether igniting at different downstream locations over a range of temperatures in the coflow.
Detailed imaging of reactive species, heat release rates and flowfields are important to resolve the details of the ignition mechanism and the possible transition of the flame of modes of autoignition to premixed flame propagation. Imaging of chemiluminescence and LIF of selected species is employed for this purpose and for a range of fuels.
Resolving the complexities associated with the initiation, growth, and emission of ultra-fine particles that pollute our atmosphere is one of the critical unanswered questions at the frontier of combustion and aerosol science. Our team of researchers is applying complex experimental and numerical methods to bring about a more detailed understanding of the mechanisms of soot inception. A particular focus here is the transition from molecules such as polycyclic aromatic hydrocarbons (PAH) to nanoparticles and hence more mature soot such as seen in the image. Studies span a range of fuels as well as a range of flames from laminar to turbulent.
A useful aspect of particle formation in flames is the scalable synthesis of nanomaterials into a range of products from catalysts to smart sensors, biomaterials and biomedical devices. Flame spray pyrolysis (FSP) has emerged as a preferred manufacturing route, yet difficulties remain in the ability to control the product quality and formation rates. The study will monitor the evolution of solid particles and liquid fragments in ‘early’ regions of spray flows and build capability for advanced diagnostics of particle dynamics and synthesis in particle-laden flows.
Reacting, turbulent flows are extremely challenging to model. Any computational fluid dynamics (CFD) tool must be able to resolve or model the full range of turbulence length scales, while accurately transporting multiple species and accounting for the interaction of turbulence on the reaction zone in a physically correct way.
The Clean Combustion group has pioneered the modelling of turbulence-chemistry interactions through the conditional moment closure method, and more recently has been instrumental in developing the multiple-mapping conditioning approach which aims to provide a unified representation of premixed and non-premixed combustion under a single modelling approach.
This has led to the development of sparse-Lagrangian models for turbulent reacting flows which offer high accuracy and very low computational cost. The group leads an international collaboration from Australia, Germany, India and China developing the open-source combustion code known as mmcFoam that is being applied to combustion of gaseous, liquid and solid fuels and to the formation of solid nanoparticles. The prediction of temperature in a gas jet by mmcFoam is shown in the figure.
Compressible combustion models require a sophisticated coupling of the models for reaction rates with the underlying CFD solver. Particularly challenging is the requirement to preserve accurately the thermodynamic state of the underlying gas mixture when the flow is severely under-resolved, as is the norm in industrial applications. Here the aim is to permit the use of very high order accurate numerical schemes with sophisticated reaction models which preserve the beneficial properties of both, with applications in industrial safety, gas turbine combustion chambers and very high speed propulsion systems.
In addition to model development, the group is advancing the state of the art in underlying numerical algorithms for mixing and reacting flows, developing new governing equations and solution methods which improve the accuracy of representation of both turbulence and the mixing zone. Methods developed within the group have been implemented in the open-source CFD algorithm OpenFOAM, and in-house codes, both of which are parallelised and thus can be used to explore the fundamentals of turbulence and reactions through large-scale computations on national high performance computing facilities.
Hydrogen fuel cell technology is considered the most promising path to increase the flight time of small drones while retaining the benefits of electric propulsion. Recent progress in fuel cell technology and small-scale hydrogen storage has resulted in several demonstrator platforms with formidable endurance. However, the power density of fuel cell technology is low compared to other power sources, and fuel cells are therefor often hybridised with batteries or ultra-capacitors.
The Sydney propulsion lab investigates ways to effectively balance the load between the multiple power sources to extend the durability and operational robustness while keeping the implications on the complexity of the balance-of-plant to a minimum. Our research also focuses on water management for non-humidified operation and the use of machine learning techniques for health monitoring and prognostics of hybrid fuel cell-based systems.