Separation of Rare Earths – Art vs. Science (IX)
This 9th, and last, article is part of a series of articles that flow from one to another and are almost prerequisite reading to understand the progression of my discussion on REO separation. The previous articles can be accessed:
In article VIII, we discussed the differences in scale of rare earth separation plants. The size and hence CAPEX of a plant is proportionate to the tonnes per year throughput of REO, the type of REO, the number of products separated, the quality of the final REO and the form of the REO produced. Other differences can be attributed to reagent supply and storage, waste disposal, local regulatory requirements and other environmental factors. For these reasons, comparing CAPEX between different businesses in different countries is a difficult task.
A key difference in CAPEX can be materials of construction. It would be expected that a US separation facility would be constructed primarily from a specialty stainless steel. The main reason for this is fire protection. It must be stressed that the organic solvent is essentially kerosene and is therefore quite flammable. The mixer / settlers are most probably purchased through a reputable manufacturer and are produced under strict quality control. There would be a sophisticated fire management system that would include:
- Flame or smoke detection in individual vessels
- Automatic fire suppression foam system
- Vessel dumping and contents transfer away from other solvent containing vessels
- Water cannons to cool vessels from adjacent fires
In contrast to the US example, a Chinese plant will be made from high density plastic, purchased in sheet form and cut on-site, constructed using semi-skilled workers with heat welding equipment. It is unlikely that there will be any sophisticated fire management provisions. The difference in CAPEX is therefore extreme. A US plant could cost $200 million. A Chinese plant would be less than $50 million.
On the issue of reagents, it may be necessary to produce the acids on site. This could add $50 million to the CAPEX. An environmentally acceptable waste management system could also cost $50 million, especially if recycling of water or reagents is involved.
Get our daily investorintel update
Process control systems add significant cost. In an earlier article, I mentioned the almost complete lack of control systems in my plant visits in China. What I did not realize at that time was how critical the feed REO spectrum was controlled. There was even recycling of product so as to maintain the feed REO spectrum to within design levels. This allows very simple control systems to be used. This significantly reduces CAPEX and OPEX.
Similarly, the OPEX for different separations are not necessarily directly comparable. Reagents for LREO separation are different to those for HREO separation. Manning levels and labour costs are different around the globe. As an indication, it should be possible to process LREO for approximately $5/kg, but HREO separation would be more like $20/kg. Again, between business comparisons are very difficult.
As this is the last planned article on separation, I will try to answer the science versus art question.
Science would include:
- Circuit modeling allowing rapid process development time
- Use of solvent additives to assist phase disengagement, to raise the flash point for fire prevention
- Sophisticated process control
Art would include:
- All the lessons learned by the Chinese in the last decade of processing to simplify the process, to improve recovery and to reduce costs
In my next series of articles, I will discuss the minerals utilized in REO production. This will allow you to understand some of the differences in the whole production process and to see why there are differences in CAPEX, OPEX and NPV.
Mr Mackowski is a qualified engineer in mineral processing with over 30 years technical and operational experience in rare earths, uranium, industrial minerals, nickel, kaolin ... <Read more about Steve Mackowski>