Why do you feed high dose of xanthates to discard as residue. We never come across such problems in 25 years of flotation for lead, zinc, copper flotation.

When measured, the residual xanthate concentrate can indicate if the addition was insufficient (zero residual), at starvation (barely detectable residual), or excessive (high residual). In the latter case, one could determine how much of the xanthate has been absorbed by difference between the added concentration and the residual one.

However, given the commodity (PGM), it is likely that high dosages are used in the first place. As such, I am wondering of the reasoning leading to the desire to measure xanthate residuals.

There are four reasons that residual measurements might be useful. First, you can save money by not overdosing. Second, you can save money by not under dosing. If your incoming ore has a significant float reagent demand variation over time, you might consider either cranking it up and leaving it there, or measuring the residuals regularly. You are trading increased reagent costs for better recovery. Usually a profitable trade-off! But as mentioned above, high dosing can be forgotten. I was at a concentrator that was overdosing to the tune of 300k/year. That was a small output mine too. And third, you can save money by knowing the excess residuals in your makeup water under a reuse circumstance (I have measured large residual xanthates in makeup water). Finally, you reduce toxic CS2 gas emissions in your tailings. If I was a reagent manufacturer, I would recommend to dose and forget it. If you are interested in residual measurement, we can rent you sensor that will allow you to measure xanthate residuals in your concentrator should you find a need in your operation.

Not to jump in with a sales pitch, but there is so much truth to what you are saying. We implement our KSX Expert System with image analysis on flotation cells all the time. By doing this, you not only optimize recovery, but you save big time on reagent doses. There really is so much potential in automation, even the ability to compensate for the nonlinear dart valves.

I can only agree with you! I think we are spoilt for choice these days compared to metallurgists 50 years ago. There is however still a number of old plants that is far behind in terms of automation. Do any of you think there is a need for the measurement of residual xanthate in real-time to control reagent dosage?

Yes but I'm still looking for a machine that can match the 75yr old metallurgist and can tell when the sampler has snuck into the blower room for a quick smoke, from the froth appearance. We comment on hiring adequate skills but it seems that despite higher salaries we do not recognize ability.

I don't think there is a need to measure residual xanthate. By implementing a few simple rules that only rely on data from existing measurements that are readily available and accurate, you can actually maximize your xanthate efficiency. Recovery, tailing limits, pulp levels, etc.

Of course, to implement these rules effectively, you would want a good Expert System, and a good froth analysis system like PlantVision. When combined, you should be able to increase your recovery but also cut reagent use by 15-30% (those are not only high percentages, they typical).

As for the 75 year old metallurgist, I don't view these systems as replacements for people, and I don't think I ever will. Most of the time, I try to sit down with those guys and learn as much as I can. I then take their knowledge and incorporate into the strategies that the software implements at their plant.

The average age of a miner is dropping at an insane rate, and in my opinion we are seeing the last generation of great miners in the US. There is no way that the "millennials" will work at a single mine site for 20 years. I think you'll be lucky to get 3-4 from this next round of graduates. It’s my goal to preserve the knowledge of these truly experienced miners, and then leave a tool behind that can continue to teach young inexperience miners long after the greats have retired. By watching the strategies of the Expert System, a young metallurgist will get insight into the processes that were the product of years of experience. You preserve the existing knowledge, and then maximize the implementation of that knowledge with an expert system. It’s a great solution for everyone.

Re: residuals. Perhaps consider these points:

By measuring residuals, you are measuring the collector demand of the ore directly. No guessing or estimating or averaging. This enables you to modulate your dosing to accommodate input ore grade variation.

Certainly, xanthates break down in water at all pHs. But this breakdown has little to no effect on the residuals under typical flotation conditions. The time for xanthate breakdown in water at high pH is much longer (hours) compared to the time to react with the sulfide ore, which is on the order of minutes. You will not always see residuals. Residuals more often than not indicate you are overdosing.

Overdosing is a way to prevent lost recovery caused by temporary excursions from optimal grade recovery. As you note in your first post, you arrived at a concentrator, and they were severely overdosing. I surmise the origin of this is, at one time, the ore grade was higher and they needed more to get optimal recovery. Then they left it there because why risk lower recovery? This is where overdosing gets costly.

The optimal dose can and does change within a single ore body as we all know. Measuring residuals will tell you when you have added too much reagent to give optimal recovery. I surmise a direct chemical measurement like this is highly complementary to cameras that give purely phenomenological results and may be used in an overall control strategy.

Still yet, no one can argue that just measuring recovery will get you a long way. That’s what all the old guys told me. 

Can you clarify what do you mean by xanthates being only 50-60% active?

I could understand reaching equilibrium between the species in solution and those absorbed on the mineral surfaces with the equilibrium point being 50% reacted. My understanding of "collector activity" is equivalent to "purity" - and a 50-60% active means a purity of 50-60%.

The measurement of residual xanthates is difficult to do accurately and reproducibly. Despite this, such measurements have values in diagnostics of process issues.

A more general technical solution to control of the dosage of collectors is a feedback loop on the g/t rate from the tailings assays. Other process variables also impact tailings assays. However, in the absence of changes to these variables during a given time period, collector dosages and tailings assays tend to be highly correlated. This requires a good on-stream analysis system and this is another story altogether.

Many years ago the MPRU (now Centre for Minerals) at UCT used to run several training courses at the various platinum mines. One of the courses was on measuring residual xanthate levels. What they found was that it was four times higher than it needed to be and that the xanthate addition rate had been increased gradually over the years. The driver was the fear of loss in PGM recovery no one wants to be responsible for even 1% loss (not that this can be reliably measured) and driven from the top down. So a great question but your industry and the management structure probably won't let you change the addition rate to more appropriate and cost effective xanthate additions.

The possibility of a 1% PGM loss means a significant revenue loss and could be potentially career limiting.