Is the thickness of the froth increasing as a result of adding more frother? If so, drainage has likely improved and this reduced the amount of entrainment of the gangue.

There is usually a strong correlation between rates of flotation and frother dosage - and to achieve an equivalent recovery, longer flotation times are needed at lower frother dosages.

I have a mistake in my question, the copper recovery didn´t increase. With more frother I obtained a low copper recovery. But the Zn and Pb recovery was much lower. So, I improve the selectivity of the process, but I lost some copper. And I was expecting an increase in the copper recovery as result of adding more frother.

Because I have made a mistake in my question!

So, with the addition of more frother I obtain a higher grade of Cu, but a lower recovery. And I improve the selectivity. I need to find a justification to that.

Looking for the granulometric distribution, I have a d80=0,030 mm, so I have a lot of small particles. Blend has a high weakness, so I assumed that the most part of the small particles are gangue, and they are entrained between the frother bubbles. But I'm not sure about this.

Your answer was a good help to me.

I have a lot of small size particles, but the liberation size of the particle is very small too. So, maybe, the use of more frother was a good way to improve my flotation results.

I developed a mineral liberation analyse and I concluded about the existence of many grains with 100% of pyrite. Can I say that the biggest parts of the particles that are retained in the froth are pyrite, liberated pyrite?

How can I get better results, i.e. how can I avoid the froth retention?

So, you have a primary P80 of 30 um and the sulphides are essentially liberated, including the pyrite.

Managing mechanical entrainment requires many cleaning stages and, as a first estimate, one column flotation stage is equivalent to three mechanical stages with respect to obtaining selectivity against entrainment.

It is not mentioned whether what you are doing is only a laboratory test program or a laboratory screening in preparation of plant test. If it is only a laboratory program, then applying multiple cleaning stages is the most practical solution.

However, before going in this direction, it may be worthwhile re-examining the collector dosage(s) and flotation time(s) used for these tests. With the increased selectivity obtained via improved froth drainage, you now have the head-room to increase collector dosages and/or flotation times for higher copper recoveries.

Of the two, collector dosage or flotation time, collector dosage would be the first choice as lengthening the flotationtime will mostly get you back to the point you were when you asked your first question. Fast flotation kinetics and short flotation times are the most effective ways to manage recovery of gangue by mechanical entrainment at the rougherand scavenger stages.

Such a fine P80 always come with lower efficiencies on collecting all the fine particles. So are you dealing only with a rougher-scavenger recovery? Or are you also dealing with cleaning stages? I assume that you are running laboratory scale tests. Can you provide more info about your ore? For example ore head grade?

You could try increasing collector dosage; liberated pyrite should not be a problem if you are using a collector with a good selectivity on the Cu/Pb circuit, and an adequate level of pH on Zn circuit.

Which is the first head of mineral and depressants are using in bulk if you have a circuit copper recovery is low because not doing a good separation copper or lead have a high copper tailings.

I'll give more information about my flotation tests:

Gridding stage: 90 minutes; add CAL (20 g/ton); P80=30um

Flotation: aerating (5 minutes) - conditioning (add collector (30g/ton),MIBC, and CAL until pH=10.5 (5 minutes) - 1st flotation (3 minutes) - Conditioning (add CAL until pH=10.5) (1 minute) - 2nd flotation (2 minutes) - Conditioning (add CAL until pH=10.5) (1 minute) - 3rd flotation (6 minutes).

The ore was a complex sulphide of Cu/Zn/Pb (1.55% Cu; 1.56% Zn; 0.55% Pb; 35% Fe)

I did this test with 3 different dosage of collector, and I had a better recovery with 527E (30g/ton) and a better selectivity with 20 g/ton. Unfortunately after I increasing the quantity of frother I have no chance to test different collector dosages.

If your goal is to separate a copper/lead concentrate and a zinc concentrate from the tailings, the pH of 10.5 during first flotation stage is not the best option for you. Copper and Lead float better under a pH between 7-9. Maybe you should consider adding a dosage of 600-800 g/t of ZnSO4, to depress Zinc. Once you have finished the lead/copper circuit, you can reactivate the zinc using 200 g/t of CuSO4, pH around 10 or 10.5, and your collector dosage. The main problem is not the frother, is the pH of your tests.

Try bulk flotation of all three. Then try separating Zn, then Pb.

The problem is that their evidence should not use lime in bulk circuit that activates the zinc as you have high iron sulfide + you to use 200-300 gr / ton of sodium cyanide and 50-100 gr / ton of zinc sulfate at neutral pH, in the zinc circuit 50-100 g / t copper sulphate 5 minutes conditioning with lime to pH 10.5 and 5 minutes + filling with copper and another 5 minutes of conditioning with the collector and the foaming and cleaning a pH of 11.5.