Pilot Plant SimulationCore Metallurgy needed to be sure that the applied sampling schedule and control philosophy would not compromise results.
Core Metallurgy was commissioned to design, construct and operate a pilot-scale Toowong Process™ plant in 2012. The process removes arsenic from copper concentrates, increasing the concentrate’s value as a source of copper metal and reducing its environmental impact during subsequent processing.
The small scale of the plant necessitated the use of batch processing in several key operations that at full scale would likely be operated continuously, such as leach residue filtration. Furthermore, a relatively large minimum sample size was required by the analytical laboratory and the effect of periodic removal of these sample on the process equilibrium was not well understood.
The flowsheet included two significant recycles that would play the main role in determining how quickly the process would reach steady-state. Both recycles included batch processes that would further disturb process conditions. A number of different feed types were intended to be run through the pilot plant, and the plant was required to reach equilibrium on each before being changed to the next.
Core Metallurgy needed to answer the following questions in order to finalise plant design and operations scheduling:
How long would it take to reach approximate steady-state once operation began on any new feed-type?
Would the intended sample volume and sampling schedule perturb the process unacceptably?
Would the discrete nature of batch process outputs allow downstream continuous processes to come to equilibrium?
A model of the plant including all significant chemistry and process control logic was built and 600-hour simulations carried out.
Outcomes and recommendations were:
Steady state equilibrium
The process takes 400 hours, or a little over two weeks, to reach approximately 99% of steady-state as measured on the leach discharge liquor arsenic concentration after starting from zero arsenic. The process should not be considered to be at steady-state before then
The nominated sampling schedule and sample volumes improve overall arsenic extraction by 0.001% due to increased residence time in the reactors during periods of no flow after sampling. This was considered negligible.
A buffer tank would be needed between the Precipitation Feed Makeup Tank and the precipitation reactors in order to avoid significant oscillation of process conditions in those reactors. Other batch processes had negligible effect on continuous process equilibriums.
Hours Cycle Time
The Pregnant Liquor Tank receives solution batch-wise from the upstream filters, and discharges batch-wise to two separate destination processes, each with their own requirements and timings. As a result the solution volume in this tank cycles on two time-scales: approximately 11 hours and approximately 100 hours. Two recommendations follow:
- The minimum volume of this tank should be 20 L in order to allow for the longer-term volume cycle.
- The planned method of controlling solution advance from this tank to destination processes is too rigid and may result in unexpected solution inventory growth or depletion over time. Solution volume advanced to the Autoclave Feed Tank should be calculated based on level in the PLT rather than a predetermined value.
Download Demo Software
Test-drive a FREE demonstration version of ITHACA on your projects.
Let the makers of ITHACA show you why it is the software of choice for your application.
Discuss your Project
Tell us what you are looking to achieve through simulation. We would love to discuss how ITHACA can help you.