An interesting question! Using a gas different to air is not something that is probably considered as the air is a composite of all the above gases. The chances of using a specific gas for flotation is not likely as the cost to manufacturer that gas would eventually be prohibitive ( I stand corrected by any one more in the know) Why would you use something else when the gas needed is right there around you. Bubble size is more about the reagent being used and how it is used as much as it is about air /gas type used.

I'd see gas selection as more driven by other factors when not using air, e.g.

•Nitrogen in copper-moly separation

•Possibly CO2 in copper-moly separation to assist pH control

I agree, you can check the gas pressure charts on Google images which will help determine the differences. "Bubble size distribution" will not be related to the gas but rather the equipment and frothing chemicals used.

Just my five cents worth!

Variations in bubble size is mainly a function of frother concentration and it can be detected and used to determine critical coalescence concentration, any greater concentrations than CCC will give you the smaller bubble sizes.

The size of bubble is not mentioned. Today technologies are available to produce 1mm and below micro bubbles.

CASE STUDY: We are using MULTO TECH GAS SAPRGER in COLUMN FLOTATION IN HZL INDIA since 1999 and working fine. Till date no problems. Today technology has further gone one step ahead we call pico bubbles-smaller than 1mm. Any specific advantage of gas injection may be spelt out. Oxygen is used in some special application of injecting Oxygen in fine bubble form, but not in flotation.

Is there a reason you're targeting a non-air gas? To meet a flotation application for a specialty gas and make it cost effective you'd need specialized flotation chambers which vent to a combined vapor recovery and recompression system. You'd also need to have your launders feeding to an isolated piping leg to maintain the gas. It's not impossible, I don't even think it would be all that costly to do so, but I simply cannot come up with a reason you'd need to do it when reagents aren't all that expensive.

Interesting question as noted but the only meaningful variable with gas type that would affect bubble size, under the same pressure, temperature and shearing device, would be solubility. Nitrogen and oxygen are relatively insoluble in water at room temperatures and I would expect them to form similar sized bubbles under the same conditions. However, carbon dioxide, being quite soluble, upon bubble formation, would up end producing a smaller range of bubbles due to dissolution, depending upon the initial bubble sizes (=surface area).

Nitrogen is the cheapest gas, after air, then followed by oxygen - all readily made on site using PSAs or if serious volumes are required, ASUs. Carbon dioxide is not a cheap gas and is recovered from wells.

As is well known, nitrogen is used in separation of molydbenite from chalcopyrite (minimize oxidation of NaSH), and higher dissolved oxygen levels does improve a number of sulphide flotation rates and separations either directly (i.e. chemically) or due to ORP.

Thank you all for your answers.

In fact, the reason I'm asking this question is that I'm investigating the effect of various parameters (i.e. frother type and concentration, aeration rate, temperature, gas type, pH, impeller speed) on bubble size and stability in a two-phase system.

I know about the cost of non-air gases, but it's a laboratory study and cost doesn't matter.

Any recommendation in this field!

You're right. Oxygen and Nitrogen have similar behavior in solubility and they make bubbles with similar sizes.

So I shouldn't use Nitrogen and Oxygen at room temperature? What temperature is suitable?

As you know, CO2 reacts with water to yield H2CO3. This makes pH to reduce.

Is it true that CO2 bubbles are less stable than O2 and N bubbles because CO2 makes the solution to be acidic?

As far as CO2 is concerned, its most readily used to change your mineral to a carbonate. So as a pre-treatment for a specific separation it can be considered worthwhile for a chemical separation given a large enough affinity for reaction so long as your resonance time is appropriate. Though calcium would become a prime concern because you'll not re-dissolve it!

The stability of the bubbles will be dependent on the pH as a saturated solution will not as readily dissolve the gas bubbles. Effectively it could not be safely used for Gold Recovery, but mildly acidic solutions would not be necessarily affected. However, neither would oxygen or nitrogen.

Interesting topic, control of bubble size in a flotation machine!

As has already been mentioned the physical parameters such as gas speed through an orifice, orifice opening size, impeller velocity, pulp density (or pressure on bubbles) affect bubble size at point of generation, maintaining that size after generation is going to be difficult without some form of pressure control.

The vapour pressure of the gas in use will also affect bubble size, a gas with high vapour pressure will expand more rapidly at atmospheric pressure leading to bigger bubbles, but as the bubbles expand they then tend to "explode" and hence form a swarm of smaller bubbles until some sort of equilibrium is achieved.

PV=nRT is applicable in this case.

Chemical stabilizers for bubbles may be possible as well.

I presume that because you are undertaking an investigation that would have relevance in practice, so the slurry temperatures would need to be similar to that found in practice, say 15 to 45 C.

However, if it is just pure research then any temperature that you like below 100C. Gases become more soluble at lower solution temperatures and so some small differences in bubble size may be observed with temperature for oxygen and nitrogen.

With regard to CO2, any 'instability' of CO2 bubbles would be related to solubility and contact time, not the induced pH caused by dissolution.

There are two aspects to bubble formation and ultimately gas dissolution : one is forming the bubbles (which is generally about shearing - lots of ways of doing this, although can nucleate directly using pressure - smaller bubbles, greater surface area, higher dissolution rate) and the other is minimizing coalescence of the newly formed bubbles - which means dispersion. Unless kept apart, small bubbles quickly coalesce into larger ones, undoing all that hard work in forming them, and for people who are after dissolution (e.g. DO level in gold leaching), very inefficient use of the gas.

Interesting discussions! PSA devices can provide whatever O2/N2 mixture that you want. We add NaHS as a way of introducing a sulphidizing agent to the solution, but we could also add H2S to the gas phase if we choose. Covered WEMCO cells in molybdenum processing are common and by oxygen reduction, we have effectively done the flotation in nitrogen. If it was my PhD topic, I would be looking at the O2/N2 ratio effects. Go and have a drink of Guinness and then a glass of Coopers Extra Stout, with different gas mixtures, and watch the effects in a glass as you drink them.

If cost is not a factor, have you thought about introducing noble gases within your study using Helium and Argon? This research could be useful for atmospherically sensitive compounds which readily decompose/oxidize under normal atmospheric conditions. Obviously this does not fit in with the current industry practices, but may be useful for a future application. There could be a range of applications in the future when froth flotation could benefit from oxygen free environments, and yes space mining could fall into one of those applications.

Your proposed Guinness vs. Coopers Extra Stout is most intriguing. How many repeated tests would you anticipate are required to achieve a statistically meaningful data set?

Reagents (frothers) are the only way I've used to stabilize bubbles up to now. I know about influence of pH on bubble stability but I haven't seen it myself in my experiments yet. I want to find a suitable condition that leads to smaller and more stable bubbles in a two-phase system.

Would you explain what you mean by contact time? Is that contact time of bubbles and solid particles? If it is, there’s no mineral in my system. So I should study on gas dispersion to gain an acceptable stability. Gases mixtures would be an interesting topic and I don't think it's been addressed much.

Actually yes I've thought about using noble gases, but my priorities are O2, N and CO2. I might apply these gases too if the time isn't problem.

'Contact time' means the time that the gas bubbles spend in contact with the solution. In general, the longer the contact time, the greater the gas dissolution and for soluble gases, the smaller the bubble sizes!

It's still ambiguous to me which gas causes more stability. I'm going to figure it out.

The reduction of bubble diameter has the added benefit of increasing the available bubble surface area for the same amount of injected air. So producing small bubbles is good.

But it also causes entrainment issues.

When you say two phase system do you mean gas and water or gas and slurry? In a three phase system (gas-water-minerals) mineral attachment has a stabilizing effect on gas bubbles.

In a two phase system (gas and water) the gas with the highest density, lowest solubility coefficient, and lowest vapour pressure should produce more stable bubbles. You can test this using gas like chlorine, nitrogen, oxygen, carbon dioxide, helium and even hydrogen (WOW I wonder what that would do to flotation kinetics?)

Imagination supported by basic scientific principles is the way to go!

I am with you on this. Minerals flotation with froth is a 3-phase system. All froths are stabilized by the third phase, solid particles. This is complicated in mineral flotation by collectors providing a mineral surface modified by chemicals and frothers dissolved in the liquid phase modifying the interfacial surface tension of both the liquid-solid interfaces and the liquid-gas interfaces. Lots of things to consider!

For more stable bubble it is advisable to add frother.

N2 will generate smaller bubbles and more stable froth. Because N2 is insoluble in water and can reduce the gas diffusion process between bubbles which will increase bubble size and destabilize the froth phase.

This is a very interesting study that you are busy with. Stone Three Mining manufactures and sells the Anglo Platinum Bubble Sizer that is used globally at concentrator plants to measure the bubble size distribution in flotation cells. This product could be of great value to you. More information on this device can be found athttp://is.gd/kbXdLj

Bubble size is determined by surface tension and shear forces, with this I mean that it does not matter the gas you are using, you will be able to achieve any bubble size distribution if you know how much shear you need. There is a particular type of flotation machines that use a plunging jet to shear gas into virtually any bubble size you want simply by modifying the air flow rate and jet velocity. A good example is the Jameson Cell that uses a jet of slurry of about 17 m/s and air to pulp ratio of about 0.6 which is able to generate bubble sizes of 0.3-0.6 mm. If you increase the jet velocity and reduce the air flow rate you will be able to produce even smaller bubble sizes but they will not have sufficient buoyancy to carry particles therefore flotation process will not be very efficient.

Also, as mentioned above it is crucial that you flotation system has frother in excess of its critical concentration. Frother is a reagent that avoids coalescence, in other words it preserves the bubble size produced by the gas dispersion mechanism. If your system does not have frothed or if it is lower than the CCC you will not be able to take advantage of the minimum bubble size produced by any given flotation technology.

In summary if you research is looking at the effect of bubble size that can be easily achieved by varying shear forces. I would suggest not using different gases unless you are looking into pulp chemistry effects.

Nitrogen is useful mainly for Molybdenum process and for mineral with more molecular weight

As you are investigating the effect of different parameters on Bubble Size, don't forget the effect of Slurry Density. This is a common variable in large scale plants; while the feeding rate or cyclone performance get fluctuation, so we will have change of density. In this case the bubble size will be disturbed.

On the other hand, the size of particles (Slurry Granulometry) is another effective parameter which I saw you didn't mention in your discussion. I have experienced the effects of both above parameters myself.

Alkaline cleaner flotation stage at any mechanical or pneumatic device, requires this kind of gas in control redox environment as the Nitrogen inert gas to reduce hydrosulphide reagent consumption, or Oxygen soluble to avoid iron sulphides in bubbles of final concentrate and CO2 to disperse the flocculated pulp by Ca(OH)2 making unselective concentrate, so the last one should be the smaller froth size and the cleanest product with frother like MIBC to keep the carrying capacity.

Gases can play quite a significant role in flotation in a number of ways as you point out. We did a lot of research as well as plant trials with mainly oxygen and nitrogen starting back in 1992. Of course the Russians had done quite a bit of work in this area during the 1950s and 1960s great thinkers in this classical period of flotation research.

We did patent most of our findings (now lapsed) as well as publishing quite a few papers. Just as a comment, it is generally thought that dissolved oxygen directly consumes hydrosulphide and sulphide ions; this is not the case. The reaction is catalyzed by base metal ions in solution, nonetheless nitrogen sparging during conditioning does significantly lower NaSH consumption rates. We also used oxygen conditioning to overcome the oxygen demand associated with fine grinding, steel media and iron sulphides (pyrrhotite the most demanding) - really improves galena flotation. It also accelerates the flotation of chalcopyrite (ORP effect) - handy when there are residence time limitations or improving selectivity against other sulphides.