Off the top of my head I don’t think this has been done. Google says a couple of micro-floats have been done. It is an interesting problem, never heard of this before.

· I would separate Ni from copper first, assuming millerite and pentlandite behave similarly by flotation.

· Then we need to test. Google says millerite floats a bit better than pentlandite at pH 10. Many stages of flotation could be used. Cyanide probably won’t help much. We can change the redox conditions, millerite should still float between -100mv and 300mV SHE where pentlandite should not. Although I have floated pentlandite with nitrogen with no problem. We will need NaHS or something similar. We should check the magnetic properties. I suspect static properties could be used by I don’t have the equipment for that.

· Assaying will be difficult. You can only go by the nickel and sulphur and iron grades. Need some mineralogy to prove it out on the best result.

· Pentlandite tends to be ‘flame’ locked with pyrite or pyhotite. Like a tiger striped particle. We need to know the minerlogy of millerite in your sample. If millerite is liberated we can exploit this as well. If it is miserly locked we we need to go below 10 um then there is not many options.

http://www.sciencedirect.com/science/article/pii/S0301751611000251

Here is what I know about copper (chalcopyrite) and nickel (pentlandite) flotation:

1) Bulk copper-nickel float using PAX at around pH 9.0 to reject pyrrhotite and other gangue. Use CMC if/as needed to reject MgO gangue (such as talc, etc.). Use an iron depressor if/as needed to help depress pyrite / pyrrhotite if there is a lot of iron in the feed.

2) Re-float the Cu-Ni bulk con using PAX at pH 11 to 12 to float off the chalcopyrite into a copper con and depress the pentlandite into a nickel con.

Not sure about the flotation kinetics of millerite and how it compares to those of pyrrhotite, chalcopyrite and pentlandite. But Google says millerite floats a bit better than pentlandite at pH 10. This needs to be confirmed.

Without knowing nothing else, my guess would be that its behaviour would be somewhere between that of pentlandite and chalcopyrite. If this is the case, it should be possible to tweak pH parameters (perhaps, floating closer to pH 12 in the Cu-Ni separation) to ensure it reports to the nickel con. Then what I would do is investigate conditions that would allow separating the millerite from the pentlandite (perhaps using pulp redox potential?).

You may also want to read these 3 research papers https://www.911metallurgist.com/blog/millerite-nis-serapation-pentlandite-chalcopyrite

Millerite does not float as well as pentlandite, but I have never been asked to separate them.

I need to ask, why do you want to separate Millerite from pentlandite? They are both Nickels...

I hope this helps a little.

Anyone else can pitch in...