• To participate in the 911Metallurgist Forums, be sure to JOINLOGIN
  • Use Add New Topic to ask a New Question/Discussion about Leaching.
  • OR Select a Topic that Interests you.
  • Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
    Using Add Reply allows you to Attach Images or PDF files and provide a more complete input.
  • Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.

Assess and Evaluate Gold Recovery Improvements (11 replies)

Sugar Watkins
9 months ago
Sugar Watkins 9 months ago

What is the best method to determine gold recovery improvements in a cyanide leach circuit?

Depending on the head grade and through-put of a gold leach circuit, Just a 1% increase in gold recovery can mean millions of dollars of additional revenue per year to the bottom line of a mining company. Bottle rolls seem to be the most common laboratory method to determine recovery differences due to ore changes, process changes or reagent additions. However, control bottle roll results, from the largest and most respected metallurgy and research labs, can vary as much as 2+%. With this poor accuracy, using this method could mean missing some important process improvements that could add up to a significant amount of gold in a year.

Are bottle rolls the only method to determine gold recovery or are there better methods out there that can accurately detect small improvements in recovery?

Paul Morrow
9 months ago
Paul Morrow 9 months ago

Bottle roll tests are only the most preliminary test. Once one has progressed to pilot plant testing then there are stirred tank tests that permit one to add oxygen or other reagents.

Given that there are almost as many SGS laboratories as there are corner coffee shops I suggest you contact the nearest to you and ask for a demonstration.

However, as you said a 1% increase in gold is significant when one is looking at 1 part per million of gold in many ores then the accuracy in measuring that is sometimes only =/-5%.

Even doing a daily metal balance on the plant Feed gold minus recovery is equal to lost gold, is not 100% accurate as most people compare feed and tailings assays on the same day, whereas current tailings is normally gold that entered the plant perhaps 48hours earlier.

Having a full time process metallurgist to see why gold is being lost, and making process changes constantly to improve recovery, will enable you to recover more gold, but not every plant is looking to maximise recovery, some look to the maximum extraction in a given feed. 94% recovery of the gold in 5,500tph, may be better than 95% recovery in 5,000tph.

If you do not have process metallurgical staff, there are several companies willing and able to provide services BBA would certainly be a willing participant, but I've found from experience that while the test is on-going the host company staff are encouraged and also become more productive, but this falls of when the test ends.

Stirred tests will likely help you; they are slower than bottle rolling, but more representatives, if there is sufficient accuracy in your laboratory. 

9 months ago
Sturmbann 9 months ago

Bottle rolls are indeed the standard and from a process improvement perspective they can give a useful indication of benefits - particularly if the baseline correlation between lab and plant is understood. In my experience this is most effective if the nature of gold losses are understood and quantified by the use of diagnostic leach tests. I don’t think you will find a lab program that can perfectly replicate what you see in the plant, so plant trials plus before and after analysis of leach feed and tailings samples in detail is the best information. Again diagnostic leaching and size by size assay analysis provide very useful information to quantify opportunities and improvements on both laboratory and plant samples. I'd agree from recent experience that small scale stirred vessel tests can also give a reliable estimate of plant scale performance improvements.

Tarun Karakoti
9 months ago
Tarun Karakoti 9 months ago

I believe in addition to the two submissions, you probably will add diagnostic leach test as well so as to give you all indications on what is going on in the plant as well as what is happening to the ore in the circuit. Do not forget that you will need the deportment of metals in drawing up your budget as well. Also the bottle roll has been accepted and practiced since cyanidation became the routine process strategy for gold or other mineral processing. I think also that the integrity of the results from bottle roll tests is a factor. How well you trust those doing the test for you and i know SGS like above said are a reputable company to give results with a trusted confidence level. But of course if you yourself are involved in the test work then it makes it easier for you to trust it.

Bill Rico
9 months ago
Bill Rico 9 months ago

Before a bottle roll test the heat cyanide test to determine which will be the maximum gold and silver extraction and will compare the process extraction efficiency.

David Kano
9 months ago
David Kano 9 months ago

You hit the head of the nail. Sampling and analytical error can be 5% on extraction. Analytical error for gold is 0.01 gpt using a 1 assay ton sample. This is 1% on a 1 gpt feed.

9 months ago
Subhash-Kumar-Roy 9 months ago

Diagnostic test (which encompasses bottle roll) is a baseline to understand what you are dealing with, rudimentary parameters are important i.e. Degree of liberation, density, PH, Cyanide addition Oxygen availability and so forth. When you understand the outcome of your diagnostic test then obviously you will choose the right approach to enhance your gold dissolution (including adding lead nitrate if it is necessary, if the sulphides are available in your ore). Free cyanide also is important in ensuring that the exposed gold particle can be dissolved during your circuit retention time. Diagnostic test can be tricky if you do not have equipment so Performance laboratories are available.

9 months ago
Unterstarm 9 months ago

In a process plant this is much simpler. CIL means mill product P80 and this means that sample size is representative (100g suffice but normally about 300 to 400 g is used). I have set up a series of bottle roll tests which determine the current recovery, recovery if extended residence time was available, the maximum recovery and the gold lost. Diagnostic tests are done on selected samples once quarter. Pretty good measure of performance - does not mean every day is perfect.

Marshal Dienes
9 months ago
Marshal Dienes 9 months ago

As suggests, the main source of variability in bottle rolls tests is sampling.
It occurs when taking analytical samples for measurement of the test feed grade, tailings content and solution content, involving variability in the calculated recovery. But the sampling uncertainty appears also between replicate tests as they are conducted on various samples.

The only way to decrease the sampling uncertainty is to increase the sample size. Tests in stirred tanks are using much more material than bottle rolls. It is why less variability can be expected.

The size of the primary sample before grinding is a key factor in the overall sampling error. That is to say if 100 g is sufficient for the test sample (only for finely ground material), it has to be produced from a much larger primary sample.

Sugar Watkins
9 months ago
Sugar Watkins 9 months ago

It looks like the agitated laboratory leach test might be a better way for us to obtain statistically significant results. Since we have reagents that can improve gold and silver recovery and increase throughput, our dilemma has been finding a laboratory test to evaluate smaller gains in leach recovery, the 0.5 – 2% range. Bottle rolls can work when we see large improvements and can do a lot of tests to give statistical significance. However, in most cases there are limits to the number of rolls you can do per day. When doing the rolls over several days other variables can be introduced. Depending on who is doing the bottle rolls, they can be a bit sloppy as well.

Using a laboratory agitated leach test using an analytical balance should improve accuracy. Does anyone have a method they are willing to share?

Kumar Choudhry
9 months ago
Kumar Choudhry 9 months ago

Many decades earlier, I had improvised a bottle roll (BR) leach procedure termed "Mini-CIL" to save time, costs & improve turn-around for the geologist & mine/mill daily planners; up-to 300-samples per day (2-lab shifts) can be leached as compared to maybe a dozen standard BR's. Data generated from 300-mini-CIL's were compared to CIL BR leach results & a correlation of >0.96 was obtained; a procedural summary is provided...details can be sent if there is interest: 150-mL plastic bottle, 30-g (% P80 of feed to leach), 25-wt% solids, 80C, 1-wt% CN solution in NaOH, pH ~11.2; 20-gpL attrition virgin C; place stacked samples on shaker unit (to & fro motion), leach for 1-hr; vent bottles every 15-min; screen C in slurry thru' appropriate Tyler mesh; you can wash the C & fire-assay or filter & wash the tails & assay solids or AAS the pregnant undiluted filtrate. Head values should already be available from the lab.

Bill Fraser
9 months ago
Bill Fraser 9 months ago

Bottle rolls are an amenability test. Agitated leach is better for accurate work with standard agitation and O2 bubbling if required. Sample at 0.5, 1, 2, 4, 6, 8 and 24 hours with pH adjustment, CN titration and top up. All washings to be preserved for mass balance. If you want to demonstrate an improvement of only 1 or 2 % you will need very good lab technicians, but it could be possible.

Please join and login to participate and leave a comment.