- To participate in the 911Metallurgist Forums, be sure to JOIN & LOGIN
- Use Add New Topic to ask a New Question/Discussion about Flotation.
- 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.
MgO Depression in Ci-Ni Flotation (53 replies)
There are various ways- chemical and physical. What is your flowsheet? Are you on a single mill circuit or dual? Talc is fun in any circuit (Cu/Ni/Au)
Have removed from rougher tails and depressed in cleaners in bot Cu/Ni (Pt flot) and in Au Sulphide flotation.
What mineralogical form is the MgO in the feed and concentrate? Talc or other mineral such as serpentine of course makes a huge difference to the adopted strategy. Guar gum of course is well known talc depressant.
CMC, guar and quebracho are all known talc depressants (if it is a talc issue you have). I have had success in depressing naturally floating talc with high doses of CMC. We were able to get down from 20-25% MgO in the feed to <6% in the final Cu/Ni concentrate when the right CMC was presented at the right dosage and was "cured" for the correct amount of time.
I would have though Guar, or Sesbania Gum as the depressant. The latter having less price volatility and being slightly cheaper! From what I can see Nickel mines depressing talc in the form of MgO are using guar.
Guar will be the first choice to test. Try 200g/t. CMC would be the second choice, anyway you should try MgO depressant (ask you supplier).
Remember, that use of depressants will drop your recovery, so you should balance between MgO grade, con grade and recovery. Also there is usually mechanical MgO recovery (with water) and depressant will have less effect in that case. You should understand what amount out your 20% is naturally floatable.
Keep in mind that you need the right Guar. As Guar use has been expanding rapidly over the past 5+ years for use as a gelling agent in fracturing fluids. Production in India (where most guar is produced) has changed. These days many plants have been set up to produce Fast Hydrating Guars - that build viscosity as quickly as possible - this is the main market for Guar at present. From testing I have seen FHG does not perform as well in MgO depressant application - although I cannot give you a conclusion on why that is.
Still, if you go with Guar, be aware that it has a high chance of spiking in price due to the demand for fracturing. Prices are relatively low this year, compared to previous. However, if the monsoon in India fails, you can expect prices to skyrocket (triple or more) - if you are dosing at 200g/t that could soon become prohibitively expensive. That is why I would suggest trialling alternatives at the same time - Sesbania/CMC.
Tell me where you can find guar? Where is it used? Tell us what you are interested in detail and I postorayus you a detailed answer.
This could be a topic of a very long discussion. Firstly I did not say the CMC was better than guar. We have successfully changed a lot of users. Same grade and recovery at reduced $/ton treated. As per your statement on the cost of guar! In a lot of cases the CMC's have performed better. The other factor is that there are a lot of CMC's out there with different characteristics. These all need to be evaluated. We have done a lot of work on the different CMC's and as result the mines are using different ones. The other factor is that CMC works more on repulsion and you have a dispersion effect. The CMC's slime cleans the sulfides. This is a different mechanism to the guar where you have unchanged attraction. The frother also plays a major part in this. Not necessarily just a straight replacement. The other positive is the CMC does not break down in the tank like the guar. That also affects the efficiency of the guar. The mines don't always check the pH or clean the tanks.
Sorry but I am now bored with following this discussion. The relationship between THEORY and what can be do practically have to be realised. CAPEX is KING, OPEX rules!
Which Cytec reagents are you using now? CMC and SHMP can be used as depressants for MgO. SHMP may have a better performance.
Remove MgO in slightly acidic pH somewhere between 3 to 5.5 using Sodium Oleate both as frother and collector. The conditioning time may be tested for different interval. Dolomite is easier to float than Calcite.
The problem splices, they floated. Try soapstone flotation. MgO high.
If your desired mineral is in form of sulfides, did you try inverse flotation? I mean using Sodium Hydrosulfide to depress Sulfide minerals and using some Collectors to float MgO? This procedure is well known in the last stage of Iron Beatification, when removing Phosphate from Iron Oxides.
I think that you need a number of strategies to successfully remove talc from nickel and copper sulphide concentrates, since 22-24% MgO is a lot of talc.
The first step is classification - which usually means some base metal losses - but a significant amount of talc can be removed this way.
Certainly the various talc depressants (CMC, guar, etc.) are reasonably effective, and as noted, managing entrainment at a low percent solids (<20%) is a useful tool.
Above comment about reverse flotation is very interesting - certainly routinely employed in copper sulphide-molybdenite separation, where the molybdenite is a natural 'floater'. If chalcopyrite is present, high levels of NaSH will be required to displace the collector. Not sure about the suggestion - talc is naturally floating and doesn't require collectors to aid flotation.
The final option is to leach the sulphide concentrates to remove any remaining talc not the most cost-effective option and technically more challenging. Besides being technically feasible, each separation stage needs to be economically attractive. What is the MgO content i.e. rejection level as well as the penalty schedule, for the two sulphide concentrates?
Some trade-off studies, comparing capex and opex of each separation stage and overall flowsheet versus base metal recoveries and product revenues based on MgO content, would settle the issue.
Unfortunately we have 70% of serpentine grade. Well as it has Aero 7260. How to be a serpentine?
If you have serpentine gangue minerals and your valuables are in sulphide minerals, you will need a dispersant. I include some links to relevant technical papers below. One plant I worked on we did very well with soda ash and hexa-meta-phosphate to control serpentine in roughing and scavenging. We then used guar in the cleaners because we also had talc.
See http://is.gd/ZgHtmV http://is.gd/OZGUWx
and
It could be prospective to check the process used for Ca and Mg silicate depression in Tyrnyauz (NaSiO3 in a good portion + steaming and time). This mode is relatively cheap and well known. Actually all sulphides go there to concentrate and silicates dropping. Guar, quebrakho, agar and etc. are too expensive here.
We did not the mineralogy to start with (i.e. MgO was present vas serpentine); the initial comments did give the impression that the MgO recovery was caused by flotation i.e. talc) rather than entrainment and probably slime attachment (i.e. serpentine).
This lack of detail does 'blow out' a discussion as well as lots of 'rat holes'.
Mg3Si4O10(OH)2 or MgO. That is the question. Some different animals here!
Yes, MgO is typically reported as assayed by XRF but it doesn't say much about the mineral present.
I had applied research experience with high MgO ore over 9% and I found that not only talc fine powder specie was present but also a rich variety included Chrysotile o fibres of asbestos. So the process flowsheet start with washing crushed ore, natural inverse MgO fine stage flotation, and mechanical arrangement for fibre caption in flotation pulp before to CuPb-Ag-Zn flotation and final Pb Oxide suphidization recovery.
You may want to consider chasing down exactly what form that Mg is in (talc, serpentine etc) as mentioned early in this blog. You should then determine if it is reporting to the concentrate due to liberation/associations with sulphides or because of its natural floatability (liberated silicates). As you can see many different options and opinions above can be taken but understanding exactly what the problem is would be the key here. I would recommend at least a quick look by a mineralogist to determine the context of these Mg-silicates if not a complete size by size analysis.
It is great if you want to take the easy way and have other people solve your problem for you despite being a process engineer. Not sure if applying your knowledge and taking pride in solving your issue is important these days. High MgO is challenging, but not unusual - not in this neck of the woods anyway. Any lab work or research done before giving up! Cytec and other suppliers will do the test work for free - they want you for a customer of course, but they will do all the hard work for you.
Well, flotation can be a tricky animal. Simply put: If you don't test it you won't know. I know experts that have been in the flotation testing business for over 40 years and they still grab a paddle and test the ore before drawing any conclusion. There is no way around!
It’s because XRF results are usually reported as oxides for metallurgical assay reasons. They should all add up to -100.
MgO is part of the major element oxides that geologists use to determine rock compositions. The major elements are divided up with oxygen so that the total should add to 100 as stated earlier. Confusion starts as people who see the value may think that it is one mineral i.e. SiO2 which is actually all the silicate in the rock so a rock like a Dunite which is almost entirely olivine will have up to 35%SiO2 and an MgO of 20% as the formula is (Mg,Fe)2(SiO4). That’s why it's so important to not assume that MgO content in the Conc. is just talc. There is a good chance that it is as it tends to flot so readily but without confirmation by a mineralogist you are assuming that it's liberated talc but I have seen slimed serpentine behave in a similar manner or even just sulphides exposed on serpentine that cause the MgO to increase.
Is this reporting technique correct? Does it confuse people?
A very experienced friend of mine in XRF told me that X-Ray Fluorescence measures element concentrations which can be reported as element oxides or just elements. A normative calculation allocates elements to minerals in a given sequence.
For example, simply taking S% and Fe % and calculating the wt% of pyrite (as long as S is only present as sulphide and pyrite is the only sulphide mineral).
"This can work for well-known systems; otherwise calculating mineralogy from chemistry can be sometimes misleading" In conclusion informing Mg as MgO from XRF may be highly misleading.
If you ask chemist they will tell you that they measure Mg, and that they calculate MgO as a request from metallurgist. XRF measures elements too, it doesn't measure MgO.
There are so many minerals containing Mg in the same ore that it is convenient to identify them rather than trying to bulk them as "MgO"
Of course, we process minerals, not elements. XRF results are good for metallurgical accounting and as guidelines for mineralogy. XRD or QEMscan will give you the mineral composition but these techniques have their limitations too.
I concur with your comments. We worked at the same Western Australian plant with a high MgO problem. The combination of soda ash for pH control, CMC and a dispersant such as tetra sodium pyrophosphate worked well with all but the worst ores. A dispersant was essential.
Nickel losses were too high when reverse flotation was tested. %MgO was the basis for penalty charges or rejection of the nickel concentrates by the smelter, hence the focus on this assay rather than the mineralogy.
I am aware of suggested "last resort" solution of leaching the concentrate being applied at plant scale to zinc, but not to nickel.As always in our industry, there is no universal solution. The ore much be characterised and then the options tested.
To improve the quality of the copper-nickel bulk concentrate need to find an effective treatment for depression reagent and flotation of pyrrhotite (activation) of pentlandite. Roughing collective concentrate presumably includes the following minerals, chalcopyrite, pentlandite, pyrrhotite, serpentine, talc. You must pick a reagent regime for perechistnogo cycle.
There are options for what some skheh reagent regime?
Reverse flotation at low pH would certainly put Mg in solution allowing a solid/liquid separation subject to the amount of Cu and Ni that goes into solution and if the acid consumption can be justified. Once the mg is in solution, it can be precipitated by raising the pH which will turn it into a hydroxide or by preferential carbonisation as is done when separating magnesium carbonate from limestone in Dolomite.
It is easy to rid of MgO in Cu/Ni.
The first thing is to use floatation as separation's way and for that you have to grind good your ore. You must know that MgO is floating easy than other material and it can float without to put any other reagent, but it is floating better with any frother you can use, especially Dow-froth 250.If you want to maximize you rid recovery, put also a collector (PAX) at the second stage of this talc cleaning. This operation must be done in two stages in continuous. The tailing of the second stage can be float normally for getting copper and cobalt and you must respect the pH and the potential if you are using NaHS.
For high contents of MgO bearing species, typically serpentine, that is not related to a talc issue where known talc depressants, i.e. CMC, guar and quebracho do not give success, then consider the use of tri-block copolymers at considerably lower dose. Also look at some of the literature relating to treatment of ultramafic nickel ores containing high MgO (e.g. Thompson Mine, Canada) for further guidance.
Characteristics of the Cu/Ni mine you mentioned sound likes the Jinchuan Cu/Ni mine in China. MgO is very easy sliming and natural floatable.
Flotation fails to respond after 25 micron and below 5 micron. Few minerals are very soft and ground very fast before valuable minerals are ground. Many times chemicals do not respond as per literature given in text books or CYTECH advice. Now what to do? For all such problems simple solutions are cloning and separating through Spirals, GMS,
If you are a mineral engineer and know panning, test it on pan and see the result.
The MgO can be present as a range of minerals. If present as natural floating gangue like Talc or Pyroxene, then a depressant can help. We can measure this with the Bikerman test method. If the MgO is present as minerals like Olivine then it might be recovered due to non-liberation or entrapment. If present as mica and ultafines, then entrainment might be the issue. We have a MLA (Mineral Liberation Analyser) at our Johannesburg lab and can run scans to answer. From here we can do Bikerman work and then possibly recommend a way forward. http://is.gd/TDnro4
Agree with the folks that say CMC or guar. Really need to do some lab tests and cost comparison to see which is most cost effective for your operation. CMC used to be more expensive than guar, now often it is the other way around. If your system is set up to use one, you can easily switch it to use the other if the costs drive it that way.
You can use the frother agent and AM2a at the preflot to eliminate the talc and graphite. Then you can make the flotation of copper.
Talc and serpentine having Mn are very soft ores and Sp. Gr is less compared to other metals. MGS is good for such concentrates. We have used for Lead and graphite carbon separation. Only care need to be taken is to use it after Trash trammel screen to remove trash from slurry or else it will not work.
Old post but I'll add. We had the same problems to the degree that we were going to be unable to operate due to Fe/Mgo issues. For us it was mostly talc. We solved the issue by running Guar to our flashes and coarse circuit. The fine circuit already had it. Separate size flotation also works wonders.
Also be careful not to overdose your depressant or it can also (depending on what it is) suppress your sulphides. If it is Talc, only a depressant will work with any surety. Use a pan in the coarse part of your plant (can't be done in the fines if you have them) to see if it's excessive (it'll look like the pans full of snow down to a grey streak of particles in the pan on top of the minerals) so you can have a go at increasing guar etc. Don't aim to suppress ALL of it if your feeds fairly variable or you will overdose and that stuffs expensive nowadays. That tiny 'smidge' (technical term) in your pan tells you whether it's still there, decreasing or increasing. If it's slimes you'll see that too (hard to wash away from the pan) and in that case you'll need a bubble cleaner like polysil or calgon.
Anything we want to know we should 1st try practically feasibility then go for discussion.
Tell me how to get rid or depress and reduce MgO in the Cu / Ni ores flotation? Using reagents company Cytec, but MgO 22-24%