Rock Cores Go Digital
Analysing rock cores takes more than fancy X-ray equipment
Three years ago a group of scientists and major petroleum companies formed a research consortium called Digital Core to explore new ways of analysing oil bearing rock samples. Today, Digital Core leads the world in measuring and modelling porous rock, and the backbone of the consortium is the Computed Tomography Facility that researchers at the Department of Applied Mathematics built from the ground up.
When it comes to tapping an oil or gas reservoir, the fine line between commercial success and failure often comes down to how well you understand the properties of the rocky matrix that contains the oil and gas. How much oil or gas is stored in the rock? How easily can that oil and gas be extracted? Is it possible to flush the oil and gas out by injecting some fluid into the porous rock (and if so, at what rate)? These are all important questions that can take enormous sums of money and large amounts of time to answer by studying rock cores extracted from potential reservoir sites.
Professor Mark Knackstedt, one of Digital Core's founders and currently head of Applied Maths, has spent many years attempting to mathematically model complex materials and specifically oil-bearing rock. The work began in the early 1990's by building three dimensional models of rock structures using information from two dimensional thin sections, a laborious and painstaking job. In 1998, while visiting the United States, he was shown three-dimensional tomographic images of road samples and became excited by the opportunity to directly image 3D structure in detail. Tomography is the process of taking many X-ray projections of a sample at different angles and stitching them together with software to create a three dimensional image of the sample. It's often referred to as CT or computed tomography.
"This was pretty impressive stuff," says Professor Knackstedt. "I returned to Australia proposing that we do some tomographic imaging of rock ourselves; so, a group of us at Applied Maths including Stephen Hyde, Tim Senden and many others, put together a linkage proposal to build our own facility. "The proposal got up and we then had to decide whether to buy an off-the-shelf system or build our own. Having several experimentalists (such as Arthur Sakellariou, Tim Sawkins and Tim Senden) in the department with experience at developing a variety of equipment we felt we could design something that was much better than anything we could purchase, so we set about building our own CT unit. Of course, there were many problems to solve in the process and it ended up taking around two and half years to build; much of this time was spent at a white board discussing the design of the facility.
But what Applied Maths came up with was one of the most powerful and flexible micro X-ray CT facilities in the world; a unit so impressive that similar units have been made and sold to companies around the world. It's described as ‘micro' because it's primarily designed to scan objects with length scales ranging from microns through to millimetres.
However, the researchers quickly realised that powerful hardware was only half the story. When you're scanning complex rock structures (over multiple length scales simultaneously) you're generating vast data sets; data sets so huge that you literally need a supercomputer to work with, manipulate and analyse them.
"In many ways it was fortunate that it took two and half years to build the physical equipment," comments Professor Knackstedt. "That's because it took a long time to write software and to build the computational infrastructure to handle the sort of data that was being generated. Indeed, handling the enormous datasets is one of the biggest problems with tomography.
"In a sense, the jewel in crown of Digital Core is the work done by the computational people, particularly Adrian Sheppard, Rob Sok, Holger Averdunk who initiated the work. This group has expanded considerably and now includes over 12 postdoctoral and more senior fellows building a broad range of software tools to handle and work with this data in a timely fashion. And that's something that we believe is missing from almost every other CT facility in the world working in this area.
"About three years ago we knew that we had something that was quite unique globally, and we were getting extremely positive feedback from the petroleum sector. The oil and gas industry spends vast quantities on extracting rock cores from reservoirs and traditional methods of analysis take months to years. Our tomographic scanning and digital analysis only takes days and provides much more detailed information than was previously available.
"And so we formed Digital Core, a consortium consisting of a group from here at Applied Maths, a group of petroleum engineers from the University of NSW, and money and resources from 14 of the world's major oil and gas companies."
And Digital Core has proved an enormous success with oil companies obtaining precious intelligence on their cores while the university has received funding to refine and extend the tomographic analysis of rock and a range of other complex materials.
"Our research has extended way beyond just oil-bearing rock," observes Professor Knackstedt. "For example, some of the methods we have developed on analysing porous rock has extended to research on bones and osteoporosis. We're also developing applications that will be valuable to tissue engineering, foamed materials, fossils and a range of manufacturing materials."
Digital Core partners are now considering how the consortium might develop in the coming years.
"The value of Digital Core's expertise has now been conclusively demonstrated and the demand for our expertise is only growing," says Professor Knackstedt. "It'll be interesting to see where things go in the years to come.