Laboratory Testing & General Mineral Processing Engineering

Laboratory Testing & General Mineral Processing Engineering

  • To participate in the 911Metallurgist Forums, be sure to JOINLOGIN
  • Use Add New Topic to ask a New Question/Discussion about Mineral Processing or Laboratory Work.
  • 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.

Guidelines on flux calculation gold smelting (2 replies and 1 comment)

C
Collin Amonde
8 years ago
Collin Amonde 8 years ago

I am currently using Borax,  Silica flour, Soda ash and Potassium nitrate,  the sludge contains also copper, I need a guideline on flux calculation to use in order to move the copper or any other Base metal to the slag. 

I would like to know more about flux calculation for gold smelting. I use Borax, Silica flour, Soda ash and Potassium nitrate. Most of my sludge contains Copper, what is the best flux calculation to use say if one have a dried calcine of 95 kg.

Please advice on a formula to use on fluxes for dried gold calcine to smelt in an induction furnace. 

David
8 years ago
David 8 years ago
1 like by shukeri

The flux is calculated to contain enough lead Oxide (litharge) so that if it is completely reduced, it will produce a prill weighing (usually) 25-28 grams. How do I know how much that is? I’m going to tell you. Ain’t gonna be no more igerunt prospectors out there. Litharge is composed of one atom of lead and one atom of oxygen. We represent it as PbO. Pb being lead and O being oxygen. Lead has a molecular weight of 207 and oxygen 16. Since they are combined the molecular weight of PbO is 223 (see link for Periodic chart of elements). In other words pbO Is 98.2% lead. If we reduce it to metallic lead then for every 100 gr of pbO we will be left with 98.2 gr of pure lead. Does that help? 

Well, it helps me because I know something that you don’t. I know that we want a prill of around 25-28 gr. So I am going to put 26 gr of PbO in my flux. When this PbO is all reduced by the flour/ore it will produce a lead prill of 25.79 gr. That is plenty close enough for government work. If we get a prill that only weighs 18 grams, we know that some of the flour must have been used up in reducing, not the litharge, but the sample. There wasn’t enough flour to reduce both the sample and the PbO. So, what can be done to fix this problem? Simple, we run the sample again and this time we put in a little more flour. How much more? I’m going to get to that but right now my ESP tells me that some brains out there are beginning to overheat and some folks are thinking bad thoughts about me and why they ever started reading this to begin with. I think there are two things that we must accomplish. We must cool down the brain cells and we must adjust the attitude of the reader. I feel this can best be accomplished by a single action. Take a trip to the fridge and recover another (or two) of those cool, soothing, foamy, unguents that I know you have squirreled away. Hey guys, we ain’t playing now. This is "Jet Airline" stuff.

Now that things are back to normal (do I hear muttering out there?), I’m going to give you a recipe for a flux that will, with minor adjustments, work most of the time. I know you aren’t going to set up an assay lab but this is what happens when you send in a sample to be assayed.

This is a good starting flux for quartz or "neutral" ore samples.

  • Ground sample ---------30 gr
  • Litharge (PbO) --------- 30 gr
  • Soda (Na2CO3) -------- 30 gr
  • Flour ------------------------ 2 gr
  • Silica Sand -------------- 10 gr

The amounts used are not correct for all ores. If your prill is too small, add more flour. I happen to know that 1 gr of flour will produce about 12 gr of lead from the litharge. So, if your prill is 10 gr too light you should add another gram of flour to the next assay. It’ll be close enough. Another thing that can go wrong is that sometimes the ore produces a melt that is too thick and viscous to pour properly. Could probably add more sand or, more effective, would be to add a couple of grams of Borax. Borax produces a thinner, more liquid melt. If borax is used you should be aware that it attacks the clay crucible so don’t use so much that you get a hole in it.

I should also mention that molten Litharge will dissolve the crucible and the firebricks that line the furnace. That is another reason for the quartz sand in the flux. It protects the crucible.

http://is.gd/E4grKy

Use the Social Share Bar on the Left. Tell everyone you can about https://www.911metallurgist.com/metallurgy/ It's FREE & GOOD.

Jean Rasczak
6 years ago
Jean Rasczak 6 years ago

SMELTING GOLD FROM STEEL WOOL CATHODES

This discussion comprises an early draft for my contribution to:

Trainor D & D Menne [1990]. The pyrometallurgy of gold refining.
Pyrometallurgy Seminar. 9 November, Murdoch University, Perth,
Western Australia, pages 155 to 183.

THE DIRECT SMELTING OF DORe-LOADED STEEL WOOL CATHODES

D Menne, Consulting Metallurgist, 14 June 1998

1.0 Summary

2.0 Introduction

3.0 Forming Low Melting-Point Iron Eutectics

4.0 Forming Borax Glasses

5.0 Mixed Fluxes

6.0 Oxidative Attack on Crucibles

7.0 Basic Attack on Crucibles

8.0 Pouring the Bar

9.0 Slag/Dore Separation

10.0 Troubleshooting

11.0 The Preferred Kiln, Crucible and Heat Source

12.0 Preferred Insulating Refractories

13.0 References

1.0 SUMMARY

Dore containing substantial levels of iron has a high melting point and is hard and inhomogenous.

The former makes smelting difficult while the latter causes problems in drilling samples for analysis and also in stamping identifying marks onto the bars.

Mints have in general not been happy to accept dore containing much over 20/1000 parts iron (assessed by attraction to a magnet), as this causes segregation which prevents good sampling.

Furthermore such dore tends to spit in an induction furnace.

The direct smelting of dore loaded steel wool cathodes avoids the steps of dissolution, decantation and filtration of a corrosive slurry which at smaller operations lacking engineered facilities, can be messy, noxious and labour-intensive.

Furthermore, the removal of iron by acid washing has often proved ineffectual, particularly where low plating densities have led to the formation of coherent gold plating.

To date, the two major problems of direct smelting have been the need for high flux additions and an inability to balance the fluidity and aggressiveness of the slag.

Minor problems have included a tendency to foam (allowing only a small fraction of the crucible volume to be used), stabilising an emulsion of dore in the slag phase (by surface charge as well as viscous drag effects), poor dore/slag separation (due to enamelling), and failure to sufficiently remove iron (probably due to insufficient oxidation or mildly reducing conditions).

This monograph provides the background to the development of a flux, which forms a slag that behaves well during melting, pouring and separation from dore.

The flux comprises 47% sodium nitrate, 42% silica and 11% fused borax.

Where anhydrous boric acid is not available, the recipe must be modified to 43% sodium nitrate, 39% silica and 18% boric acid.

The major innovations with respect to conventional wisdom is the avoidance of soda ash, fluorspar and pyrolusite, and the use of fused borax (anhydrous boric acid B203, also known as borax glass) rather than borax (sodium tetraborate Na2B407.10H20).

Furthermore use of sodium rather than potassium nitrate is advised, and the presence of moisture in the crucible charge should be avoided.

At an addition of merely 1 to 3kg flux per kg steel wool, the flux will generally produce dore buttons containing around 3/1000 parts iron from dore loaded steel wool cathodes containing even up to 70% iron, while providing a clay-graphite
crucible life of around 40 smelt campaigns.

Notes on bullion pouring, troubleshooting and preferred kiln, crucible, heating and insulating hardware are also included.

The key conclusions are that the melt should be kept at 950C and occasionally paddled until the bulk of the iron has been oxidised and taken up into slag.

Thereafter the temperature should be increased to 1100C, and poured as soon as the melt is the consistency of honey and is homogenous.

2.0 INTRODUCTION

The large cathode surface area provided by steel wool has led to its general adoption in the recovery of dissolved gold by electrowinning.

However the iron substrate has proved undesirable because the resulting iron contamination of the dore makes the dore less manageable and refineable.

Dore containing substantial levels of iron has a high melting point and is hard.

The former makes smelting difficult, while the hardness causes problems in drilling samples for analysis and also in stamping identifying marks onto the bars.

Mints have in general not been happy to accept dore containing much over 20 to 30/1000 parts iron (generally assessed by attraction to a magnet).

Only about 50/1000 parts iron dissolves in typical bullion, the excess forming a separate layer which segregates and prevents good sampling.

This is due to the low solubility of iron in copper and silver generally found in typical bullion - where the amount of these diluents is low, the iron causes less significant or negligible problems.

Furthermore if the mint applies the modern practice of induction melting, the melt tends to spit due to atmospheric oxidation of hot iron-rich modules being carried to the surface by convection.

This has led to additional refining costs being levied to suppliers, and delays in payment for bullion delivered as payments cannot be made on initial sampling, but have to await final refining.

Furthermore mints are keen to avoid the disputes inherent in poor sampling.

The general approach used to reduce the levels of iron in dore produced from steel wool cathodes to satisfactory levels has been to dissolve out as much of the iron as possible with acid prior to smelting.

A particular advantage of this approach is that if done well, the flux requirements can be reduced to 0.03kg borax per kg bullion (about 0.12 kg borax/initial kg irons).

However, gold metallurgists at smaller operations have been attempting to move away from the acid washing of steel wool cathodes prior to smelting in order to avoid the messy, noxious and time-consuming intermediate steps of dissolution,
decantation, and filtration of a corrosive slurry.

At larger operations more sophisticated provisions have been warranted, which ensure that the operation can be rapidly and easily carried out.

The trend to direct smelting at most operations applying Zadra elution has been further accelerated because of the poor efficiency of acid washing where coherent gold filming protects the steel wool from effective acid attack.

Such coherent gold films have proved common both at smaller operations where the electroplating current density is generally low e.g. under 10amp per square metre, and where Zadra elution ensures a highly favourable electrolyte resistivity e.g. under

0.2 Siemens.

Operations treating ores high in silver generally also favour a move away from acid treatment because of the significant risks of dissolving and losing silver in the presence of chlorides and oxidants.

The attractiveness of direct smelting of cathodes has been greatly improved by the development of electrowinning cells that allow gold loadings in excess of 4.5kg Au/kg Fe.

These loadings are achieved by using thin cathodes and parallel electron and electrolyte flux.

Less than 20% steel needs to be slagged off loaded cathodes produced in such cells.

Another advantage of direct smelting is that the amphoteric nature of iron can be exploited, to form slags less aggressive to crucible walls.

According to Laatsch (1925), the pyrometallurgical refining of gold from steel is generally conceded to be so onerous that hydrometallurgy is advised when there is more than 80% steel.

Goldroom smelting practice indicates that pyrometallurgy might even be impossible if the gold concentration is too low.

This is in contrast to Barcza (1985) who reports that methods of phase concentration of gold by pyrometallurgy which can provide 99.5% recovery from feedstock containing 20 parts per million gold have been developed.

However, attempts to direct smelt loaded steel wool cathodes have met with limited success, principally because earlier flux recipes which were drawn on were developed to treat oxidised iron in the form of heavy iron minerals, or rust flakes captured by gravity equipment.

In contrast the iron contained in gold-loaded cathodes needs to be converted to the ferrous or ferric state before it can be slagged.

The requirement to provide for the oxidation of iron if it is to be successfully slagged forms the basis for adding oxidants or providing for in-situ oxidation e.g. by air injection.

It is convenient to add sodium required for fluxing as nitre to promote oxidation.

In fact some Rand gold operations where the existing goldroom contains a calcine furnace, the steel wool is calcined prior to smelting as an alternative to in-situ oxidation during smelting.

This has allowed the operators to apply the recipes originally developed for smelting concentrates containing iron minerals and rust.

It is this same need to oxidise iron that mitigates against the use of soda ash which is a mild reductant, as a flux.

(The tendency for soda ash to cause foaming due to rapid gas generation is also undesirable).

The development of a suitable flux for loaded steel wool cathodes can be achieved by providing for the abovementioned iron oxidation, as well as for the other features desired of a flux, namely that it forms a slag which:

[a] Partitions the bulk of the non-precious metals into the slag phase,

[b] Has a low melting point,

[c] Does not attack the crucible at an excessive rate,

[d] Has a low viscosity,

[e] Limits volatilisation of molten metals,

[f] Separates easily from the dore button after solidification.

This monograph discusses the major features and development of a good flux to direct smelt dore loaded steel wool cathodes.

3.0 FORMING LOW MELTING-POINT IRON EUTECTICS

Fluxes are aimed to take up non-noble metals from dore as oxides, in some cases oxidizing these metals if necessary.

The resulting mix is a slag.

Where the metals are not oxidized, they report to and stay in the melt - and are removed only with great difficulty eg passing Cl2 to the melt.

Sufficiently oxidizing conditions are necessary to adequately pre-oxidize the material to be smelted.

Copper in particular, being "near-noble" can prove difficult to slag out without providing sufficiently oxidizing conditions.

Figure 2 shows that the noble behaviour of copper becomes marked at higher temperatures [and is promoted if carbon in the crucible material reacts to reduce the copper to metal - common for too hot and too lengthy smelting campaigns] By contrast, a highly oxidizing flux can lose silver to the slag.

Metals which are not partioned substantially either to melt or slag can bind through the transition layer In electrowon gold, the steel wool generally constitutes the major impurity to be removed.

The removal of iron thus forms the basis for developing a flux recipe.

A ternary diagram developed by Bowen, Shairer and Willens (1930) indicates that a ferrous alkali slag, 2[Na2O.SiO2].[Fe2O3.2SiO2] which melts around 900C (see Figure 1) provides a low melting point along the aegirine/waterglass eutectics tieline, and can thus be applied to slag off the steel wool.

The choice of the above point on the ternary diagram has been chosen in order to achieve a maximum uptake of Fe2O3, while ensuring a slag melting point around 200C below that of dore (a factor known from practice to be necessary to provide a good slag), and also to maintain a maximum SiO2 content (in order to limit basic attack on the crucible as discussed further on).

Table 1 below gives the flux recipe which may be derived from the above, applying the minimum quantity of Na as NaNO3 to achieve the desired oxidation of iron.

Table 1 - Alkali Flux Recipe

NaNO3: 60%

SiO2: 40%

In a melt, NaNO3 provides 1.5 O2- (the rest escaping as NO).

Thus 5.2kg of this flux, which is stoichiometric with kg steel wool is forming the aegirine/waterglass eutectic, provides 2 times the stoichiometric quantity of oxygen needed to oxidise iron to the ferric state.

In practice this slag has proved fairly nonaggressive to the crucible.

However it is viscous, and if much the gold swept into the slag phase by turbulence during the pour [too fast, from a height too great], the shot formed in the slag is not released back to the button before solidification.

Borosilicate glasses discussed in Section 4 below provide greater fluidity and flexibility for slagging and dore droplet disengagement.

A gold-iron phase diagram derived from the data of Seigle (1956) is shown in figure 5, and illustrates a number of features important to removing iron from gold.

Firstly, no more than 27% Fe can be tolerated when the gold melts to avoid inclusion of refractory high-iron solid phase components in the bar.

Secondly, no more than 3% Fe must be aimed for if the risk of a second iron-rich phase within the bar is to be totally avoided.

These requirements dictate that the gold must not be melted (and thus be able to take up metallic iron) until the bulk of the iron has been axidised and drawn into the slag phase.

This is conveniently done by keeping the melt at 950C and occasionally paddling until the melt is uniform.

Thereafter the temperature can be taken to 1100C and the melt poured as soon as homogeneity is achieved and the slag has a consistency of honey.

Although prior oxidation (calcination) of the steel wool is applied, this step is not essential.

The prior oxidation of iron reduces the risk that insufficient oxidant has been added, or is lost prior to use, or isused in attack of the crucible carbon.

However, sufficient time must still be taken to take the iron oxide up into the slag phase, and care must be taken not to use excessive temperatures or carry the smelt on too long with consequent risk of reduction of the iron to metal by crucible carbon.

4.0 FLUXES TO FORM FORMING BORAX GLASSES

Borosilicate glasses have considerable advantages, and provide greater processing flexibity.

They have very low melting points; a high metal oxide carrying capacity and retain fluidity well which makes them very easy to slag off many metals.

The amphoteric nature of boron makes it relatively non- aggressive in its own right to crucible material [but it does enhance erosion, due to the great fluidity it can impart to a melt].

A major feature of borax glasses is their contribution to redox control, the basic mechanism which dictates the distribution coefficient between bullion and slag.

Soda-boric melts have significantlu lower oxygen ion activity than soda-silicate melts.

Such lower oxygen ion activity improves removal of copper from bullion to slag [albeit at the risk of doing the same for silver, particularly if silver levels increase - see Atmore et al [1971]. This would occur when treating electrum, common in primary ores and particularly high in epithermal deposits].

Bugbee (1940) records conclusions of T K Rose that an optimised borosilicate flux suitable for slagging many metals comprises: 2[Na2O.2SIO2].2[2B2O3.SiO2].3SiO2

Traditionally when slags of this type have been aimed for, boron has been introduced as borax [Na2O.2B2O3.1OH2O].

However to provide maximum oxidising potential, the flux may be made from sodium nitrate and fused borax according to the recipe given in Table 2.

The application of fused borax (B2O3) as a boron source has proved to be the key for the development of the most successful mixed flux revealed in the next section.

According to Weast (1979/80) fused borax has a low melting point, and Singer and Singer (1963) record that it has a particularly low surface tension.

Consequently it will wet the steel wool and engulfs undissolved flux reagents during the initial stages of smelting, serving as a medium for the slag reactions to proceed in.

More importantly, all the boron added as fused borax is available to interact with metal oxides in the melt, unlike borax which has already to some extent satiated the oxygen bonds with the associated sodium.

Sodium introduced with borax would not only be an unnecessary diluent which does not contribute to the reactions required (e.g. iron oxidation), but serves to take up some of the bridging capacity of the boron which could otherwise be expended on iron.

Sodium, being a strong base would also increase the reactivity of the slag towards the crucible.

A further advantage of fused borax is that it does not introduce water (even as water of crystallisation as in borax, Na2O.2B2O3.1OH2O or as bound water in boric acid, B2O3.3H2O which contain 47 and 44% water respectively) into the melt.

Such water can promote the loss of the oxidising power of NaNO3 by allowing volatile HNO3 to form and be driven off while the crucible charge is heating up.

A final reason for preferring the fused borax is that it does not contribute to loss of heat through volatilisation of contained water.

Table 2 - Mk I Non-ferrous Metal Flux

SiO2: 46%

B2O3: 24%

NaNO3: 30%

150g of this flux is used per g-mole of the RO metal oxides, and 230g per g-mole of the R2O3 metal oxides.

This corresponds to 1.8kg flux per kg Fe in the ferric state, and 2.3kg flux per kg Fe in magnetite [the usual calcined product], and 2.7 kg flux per kg Fe in the ferrous state [haematite]

Further quantities of this flux are added to remove base metals such as Cu and Ni from dore. This flux has proved to be of acceptable fluidity, but is unfortunately reasonably aggressive towards crucibles.

This flux has a melting point of 850C, which is well above the minimum which may be derived from Levin, Robins and MC Murdie (1964).

On their Na2O-B2O3-SiO2 phase diagram, a 'sweet spot' flux melting at under 600C may be derived.

The composition is given in Table 2b below, and corresponds to 6[Na2O.2SiO2].7[2B2O3.SiO2].

It should be noted that the stoichiometric excess of 2B2O3.SiO2 over Na2O.SiO2 is very important to achieve a low melting point.

This slag would be very aggressive towards crucibles, until it has taken up a substantial quantity of iron.

Table 2b - Mk II Non-ferrous Metal Flux

SiO2: 40%

B2O3: 34%

NaNO3: 26%

5.0 MIXED FLUXES

Sodium borosilicate fluxes comprising mixtures of the simple flux as well as the mixtures discussed above allow one to combine their best features and avoid the worst.

However the use of a greater number of slag components affects both the melting point and the vitrification of the slag.

In general it is desirable to aim for the highest and sharpest melt transition temperature range, consistent with adequate fluidity for the pour, and experience to date indicates that incorporating an excessive number of slag components prevents achieving this aim.

The determination of a preferred flux composition comprising both the aegirine/waterglass eutectic shown on Figure 1 and a borax glass as proposed by T K Rose has however provided a reasonably satisfactory compromise.

In this regard a number of plant-scale experimentation has led to the development of a flux of composition first published in a paper by the author of "Heap Leaching", given at the AUS I M M Regional Conferences on Gold Mining, Metallurgy and Geology at Kalgoorlie October 9-11, 1984.

An improved composition is given in Table 3.

Table 3 - Practical Flux for Loaded Steel Wool Cathodes (a) If B203 (b) If H2B03 is available is available NaNO3: 47% 43% SiO2 : 42% 39% B2O3 : 11% 18% Added at the stoichiometric rate of 3.1kg flux as per

Table 3(a) per kg of steel wool (to slag 25% of the iron as Fe2O3 with flux of composition as given in Table 1, and 75% of the iron as FeO with a flux of composition as given in Table 2) to produce dore containing as little as 2/1000 parts Fe.

Even flux additions as low as 1kg flux per kg steel wool have provided acceptable dore (viz. having around 3/1000 parts Fe, however requiring increased temperature, and creating a relatively viscous and slow-congealing slag.

A further advantage of this flux is that it provides around 40 melt campaigns per clay/graphite crucible.

(The operating life of silicon-bonded carbide crucibles has not yet been determined using this flux).

These rather surprising findings are believed to be due to the amphoteric nature of Fe, which can be applied to advantage if the iron is maintained in the Fe2O3 state in the slag.Fe2O3 not only acts as a bridge for neutralising oxygen bonds between acid components such as SiO2 and alkali components such as Na2O, but as a buffer to reduce slag attack on refractories, and as a structure breaker which increases slag fluidity.

Calculations indicate that the slag comprises about [B2O3.Fe2O3].2[Na2O3].4[SiO2.Fe2O3]

There is of course always room for improvements and fine tuning for particular operations.

Thus it is probably useful to review some of the problems and features of fluxes to provide further perspectives on this technology.

The flux 3b can be used but is prone to loss of HNO3 while heating, particularly if temperatures in excess of 1100C are rapidly built up. Preferably firing should be adjusted to achieve a melt the consistency of honey within 1 to 2 hours and to keep the melt there for one hour, paddling the charge gently a few times during the procedure.

6.0 OXIDATIVE ATTACK ON CRUCIBLES

The carbonaceous components of ordinary clay/graphite as well as of the silicon-carbon bonded clay/graphite crucibles generally applied for smelting dore are subject to oxidative attack.

It stands to reason that the addition of the minimum quantity of oxidant, to the extent merely of achieving the oxidation required in the melt, will minimise oxidative attack on the crucible.

Thus the quantity of initial alkali flux can be fixed by providing for the minimum quantity of flux to achieve stoichiometric in-situ oxidation.

However because of thermal decomposition of the nitrate, some nitrous oxides inevitably escape from the slag before giving up all their oxygen.

The extent of this inefficiency is determined inter alia by melt temperature, slag fluidity and crucible geometry; lower melt temperatures, less fluid slags and deeper crucibles (see Bugbee (1940)), predictably leading to a closer approach to stoichiometry.

Any oxygen deficiency can be further made up by introducing atmospheric oxygen which may be adsorbed at the slag surface or injected into the melt with a lance, avoiding impingement on the crucible walls.

Merely 1m3 air provides sufficient oxygen to oxidise 1kg Fe. Note however that dry air (e.g. from an instrument air line) should be used to avoid explosive steam generation in the crucible. Air injection stirs the melt, promoting slag reactions (and crucible attack). However adequate stirring can occur without air injection (by thermal convection as well as by in-situ gas generation) particularly if the flux viscosity is kept reasonably low.

(A recipe modification to achieve this is discussed in a later section).

On balance it is probably most practicable to avoid much reliance on atmospheric oxygen as surface adsorption is slow while air injection requires some dexterity and complicates the process.

Note that lower oxygen solubility at higher temperatures could reduce the effective oxidation of the steel wool.

Should further nitrate additions be desired to improve in-situ oxidation it is desirable to add this with sufficient silica to form the low melting-point (under 800C) and non-aggressive acid eutectic Na2O.3SiO2 (see Figure 1).

The relative proportions of the oxidant powder should be as given in Table 2.

Figure 4 from Levin, Robbins and McMurdie clearly demonstrates the dangers of deviating substantially from the suggested eutectics (and corresponding timelines) for example by adding either NaNO3 or SiO2 in excess : higher melting and viscous slags can easily result.

Table 2 - Composition of Non-aggressive Oxidising Powder NaNO3 50%SiO2 50%0.7kg of this flux will provide stoichiometrically for the oxidation of 1kg Fe to Fe2O3.

The need for more in-situ oxidation is most likely to arise if significant quantities of non-ferrous metals are recovered with the bullion during electrowinning.

In such cases it may be convenient to compound a single flux to both oxidise and slag these metals.

The recipe given in Table 2, will achieve this.

Paul (1986) warns against the use of excessive quantities of nitre, which could lead to the oxidation of silver and its loss to the slag phase.

Trainor (1986) warns to add the nitre slowly if used to supplement a hot charge due to the risk of excessive foaming, following rapid gas release (noting though that a small amount of dry salt sprinkled over the slag surface will tend to settle the reaction down).

D
Dave Tahija
6 years ago

Thank you so much for posting this. It is most informative.We send a gravity concentrate offsite for high-intensity CN leaching and the offsite refiners sometimes have issues with high-iron pours as you describe. This will be helpful to them and us.

Please join and login to participate and leave a comment.