EDITOR: | June 10th, 2014 | 7 Comments

Separation of Rare Earths – Art vs Science (VI)

| June 10, 2014 | 7 Comments

Mackowski-Steve-3In this, the sixth article on separation of Rare Earths, I want to discuss the processes that produce the final product. Again, as per previous articles have suggested, there is no across industry term for this part of the flow sheet. I am going to use the term “finishing” as this is the finish of your flow sheet (for each product) and where the products depart for the customer. This term is used in the titanium dioxide pigment industry as this is where the crude, but pure titanium dioxide pigment has it’s size distribution adjusted and surface properties tailored to the paint requirements of the customer. The use of the term “finishing” in REO plants matches the pigment industry processes since this is where those somewhat esoteric specifications of the customer need to be met, if required.

I am going to need to simplify this discussion as there are many, many different specifications that are adjusted and met in the finishing section. The distinction I am going to make is by splitting the products into two groups. Those destined for the metals market, where elemental composition is the driving specification and those destined for uses where the surface properties are additionally very important. So firstly, the metal REO products.

Once a REO separation is complete the valuable rare earth element (REE) is in aqueous solution and needs to be precipitated as a solid for transfer to the customer. This is typically as an oxide. Note. Carbonates are upstream to this where further upgrade is still needed before the REO is pure enough for metals manufacture. So typically the REE is precipitated as an oxalate, calcined at high temperature to produce the oxide and then packaged. The next step (at the customers premises) is smelting where the rare earth oxide is put into a molten bath of REE metal and a reducing agent. I won’t go into the chemistry here. It’s outside of scope and better described by more qualified others. So the REO in essence melts and converts from an oxide to it’s elemental metal form. The size distribution and surface properties of the feed REO material are not critical in this reduction step, so the finishing of the REO is pretty simple. Precipitate, calcine (if required)and package. As stated this process applies to those REO destined for the metals route. So the neodymium, praseodymium, samarium, dysprosium, terbium oxalates or oxides are finished this way.

The other group of REO products are finished in a somewhat similar manner using precipitation, but have very different controls. These are the REO whose surface properties are critical to the downstream use. For example:

  • Lanthanum as a crude oil cracking catalyst
  • Cerium as an automobile exhaust catalyst
  • Cerium as a polishing powder
  • Europium as a phosphor
  • Yttrium as a phosphor

All of these products have very specific size distributions, surface area requirements and other more physical related parameters. This makes finishing that much more problematic as you have to control the finishing process to meet these particle surface related requirements. As an example, the European Union alters it’s automobile emissions standards (lower, always lower!). This changes the demand on the exhaust system catalyst (as well as other things outside of scope). It is generally not the chemical properties of the catalyst (cerium oxide) that change, it’s the particle surface related properties. The REO separation plant (finishing section) has to meet these new, more onerous specifications to sell cerium oxide into the new market. This process is ever evolving for those REO in the particle performance sphere. So catalysts and most certainly phosphors are part of this space. In fact, the performance of some phosphors is so critical (and confidential) that the phosphor company will only perform the very last stages of finishing themselves. Needless to say these applications are coupled with very onerous product quality specifications.

So when people say that separation is difficult, it is not just the solvent extraction! The finishing section is also very important. This is why during process development (around PEA time) you must have a view of your final customer. You are also tailoring your finishing section to meet their requirements. Go for an average specification to cover all or more bases, and the catalyst requirements move a little; trouble! So a clear view of the requirements of the customer now and into the future is very important.

So to complete the separation flow sheet, we must add the finishing sections for cerium and lanthanum, as well as for all those separated REOs that we discussed in earlier articles. Now as these are so customer specific and the surface parameters so critical, I’ll leave this section to each REO proponent to explain in their PEA or equivalent.

In the next article, I’ll present an overview of how this “generic” LREO circuit approach can be viewed when thinking about developing a HREO circuit, or a tolling circuit, or whatever separation processing blend you may be thinking about.

Steve Mackowski


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>

Copyright © 2018 InvestorIntel Corp. All rights reserved. More & Disclaimer »


  • Lid

    I enjoy reading your article and learning from you, Thanks Mr. Mackowski.

    June 10, 2014 - 10:38 AM

  • Tim Ainsworth

    Great detail at the investor level Steve. I think most of us have been conditioned to think of CoP as a single, static number and the same for selling prices per element.
    In reality, each product, as distinct from element, will have it’s own CoP determined by customer requirement, presuming you actually have a customer at the start of the process. CoP will be a moving feast as a plant moves thru various product streams, across various elements, and can only really ever be expressed as an average.
    By the same token actual sales prices relate to the product produced rather than simple published oxide prices by element.
    Obviously “profit” will be determined by the margin between CoP & sales price and vary product to product, effectively the value added in the process from a customer’s perspective.
    I.e., relatively basic Ce & La carbonate processed and blended to spec that then enters further up the customers own supply chain creating an efficiency dividend for catalyst manufacturers.
    Also begs the question, is there an “IP security” premium for ROW production to spec? Solvay’s recent experience in China might support that thought:
    Ditto China withdrawing supply from WR Grace 2007.

    June 10, 2014 - 1:11 PM

    • Steve Mackowski

      Tim. In a later section I will deal with Cost of Production and effect on revenue by section and try to explain why the many apparently disjointed business models occur. Getting the ground work done first.
      Thx for the comments and look forward to that lunch.

      June 10, 2014 - 10:10 PM

  • slava Ukraina

    Great discussion of a complex topic. Thanks for taking the time and effort.

    June 10, 2014 - 7:44 PM

  • Bob Richardson

    Thanks Steve for a most readable and interesting introduction to thus subject, which as a LT Lynas investor I’ve greatly appreciated.

    June 13, 2014 - 8:50 PM

  • Separation of Rare Earths – Art vs. Science (Part IX) | InvestorIntel

    […] of Rare Earths – Art vs. Science (I), (II), (III), (IV), (V), (VI), (VII) and […]

    July 2, 2014 - 8:27 AM

  • Mackowski Success Factors for Rare Earths Development (Part 1: Funding) | InvestorIntel

    […] my previous series of articles of Rare Earth Oxides (REO) Art vs. Science series, I discussed the issue of separation. How different types of […]

    September 9, 2014 - 9:55 AM

Leave a Reply

Your email address will not be published. Required fields are marked *