EP0274187A2

EP0274187A2 - Improvements in or relating to the fluidised-bed roasting of sulphide minerals

Info

Publication number: EP0274187A2
Application number: EP87309351A
Authority: EP
Inventors: Alan Johnson
Original assignee: Commonwealth Scientific and Industrial Research Organization CSIRO; Electrolytic Zinc Company of Australasia Ltd
Current assignee: Commonwealth Scientific and Industrial Research Organization CSIRO; Electrolytic Zinc Company of Australasia Ltd
Priority date: 1986-12-24
Filing date: 1987-10-22
Publication date: 1988-07-13
Also published as: NO875386L; EP0274187A3; NO875386D0; KR880007774A; ES2008399A6; KR930005070B1

Abstract

A process for improving the fluidising properties of material in the bed of a fluidised-bed roaster, said material including particles in fine, intermediate, and coarse size ranges, is characterised by increasing the proportion of material of intermediate size range therein. The advantages that can be gained by increasing the proportion of material in the intermediate particle size range include better gas and solid contacting, and an improvement in the ability of the bed to carry a greater proportion of coarse lumps as well as rejecting the fine sticky particles.

Description

This invention relates to improvements in the process for fluidised-bed roasting of sulphide minerals such as ores or concentrates of zinc, copper, lead, iron, nickel, and molybdenum, or their mixtures.
The process for fluidised-bed roasting of sulphide minerals is well known. Essentially, the raw materials, hereinafter referred to as concentrates, are fed into a roaster where they form a fluidised bed, maintained by air passing upwards through a grate. The bed depth is controlled by withdrawing roasted concentrate either as bed overflow or as underflow. The finer particles of concentrate which are elutriated from the roaster are separated from the exit gases in a waste heat boiler and in following dust collection equipment. The exit gases contain sulphur dioxide and are normally passed to a sulphuric acid plant. The product from the bed and dust collection equipment is generally referred to as calcine. The oxidation of sulphide minerals during roasting generates heat and the bed temperature has to be controlled by, for example, cooling coils or water sprays.
It is well known that the particle size of concentrates is very important in the fluidised bed roasting process. A broad range of particle sizes gives better fluidisation than a narrow range, but certain problems are experienced with both the fine and the coarse fractions. Particles finer than about 0.1 mm in diameter tend to produce channelling in the bed and this results in poor contact between gas and solids. An excess of coarse solids produces a partly defluidised bed unless the gas velocity is raised, but this tends to make the bed surge and again there is poor contact between gas and solids.
The applicants have found that particles finer than about 0.1 mm in diameter tend to be highly reactive, which causes them to overheat and, depending upon their composition, to form molten phases which make the bed sticky and difficult to discharge. In extreme cases, such fine particles can lead to sintering and defluidisation of the bed.
Most sulphide concentrates are largely composed of particles which are finer than 0.1 mm in size and tend to produce sticky beds in fluidised-bed roasting. Some sulphide concentrates tend to oxidise and partially agglomerate to form very coarse lumps before they are roasted. Such lumps can be removed by screening but, since the concentrate is normally fed in a moist state to the roaster, it is difficult to perform this operation with a small screen aperture.
The fluidised-bed roaster is usually designed for a fairly narrow range of air flowrates due to constraints on the concentrate feed rate and downstream treatment of the roaster gases. These constraints arise because of the demand for a steady supply of calcine for the metal recovery process, and the need to maintain the flowrate of exit gas and its sulphur dioxide tenor to within the limits imposed by the acid plant. Consequently, it may not be possible to raise the air flowrate to an existing roaster to maintain fluidisation with a concentrate feed containing coarse lumps. Even if the air flowrate can be raised to maintain fluidisation this has the undesirable effect of increasing the amount of material which is elutriated from the roaster to the dust collection equipment.
One method which is sometimes used to remove coarse lumps from the bed is to discharge by bed underflow withdrawal. However, some roasters are designed to discharge by overflow only and have inadequate or no facilities for underflow discharge. Even if discharge by underflow withdrawal is practised the bed may still be poorly fluidised because it contains an excess of coarse particles. Furthermore, underflow discharge does not prevent an excess of fine particles from producing a sticky bed.
The applicants have conducted fluidising tests on different size fractions of calcines. From these tests the applicants have discovered that the behaviour of material in the bed of a fluidised-bed roaster can be improved by increasing the proportion of material in the intermediate particle size range in the said bed. The required intermediate particle size range will depend upon the particle size distribution of concentrates and of bed material, but is generally between 0.1 and 3 mm, preferably between 0.1 and 1 mm, and most preferably between 0.1 and 0.5 mm. The advantages that can be gained by increasing the proportion of material in the intermediate particle size range include better gas and solid contacting, and an improvement in the ability of the bed to carry a greater proportion of coarse lumps as well as rejecting the fine sticky particles.
According to the present invention, the proportion of material in the intermediate particle size range in the roaster bed can be increased by any one of the following three procedures, or any combination of these procedures.
In the first procedure of the present invention, bed overflow or underflow calcine from the roaster is subjected to size separation to produce the desired intermediate size material which is then fed back into the roaster, either separately or with fresh concentrate. Size separation of calcine is achieved by screening, air classification, or other methods known to those skilled in the art.
In a further embodiment of the first procedure of the present invention, the oversize particles of calcine are separated by screening, air classification, or other methods known to those skilled in the art. The remaining calcine is fed back into the roaster at a point which is high above the bed so that the fine particles less than 0.1 mm in size are carried away by roaster gases and cannot enter the bed and become sticky, Accordingly, only the desired intermediate size material is returned to the bed.
The amount of intermediate size material that is recycled to the bed according to the first procedure of the present invention is preferably between 10 to 100% by weight of the charge of fresch concentrate.
The second procedure of the present invention involves the physical and chemical conditioning of a lumpy or fine concentrate before roasting to form intermediate size particles in the previously specified size range.
The third procedure applies when material discharged from the bed is found to be of the desired size range, in which case a proportion of the said material can be recycled to the said bed without needing to subject it to a size separation operation.
In the abovementioned second procedure, the applicants have found that a certain type of mixer can be used to perform the required conditioning. The said mixer is characterised by a high shearing action and, although the applicants do not wish to be restricted to a particular make of such a mixer, a V-Blender of a Schugi mixer have been found to be particularly suitable for this duty. In the present invention, the moist concentrate to be conditioned is fed into the said mixer together with a suitable reagent. Although the said mixer can produce intermediate size particles from moist concentrate alone, the applicants have found it better to add a reagent to prevent the intermediate size particles which are formed from breaking down excessively during roasting. Bentonite, waste sulphite liquor from paper pulping operations, calcine dust together with sulphuric acid, and solutions of molasses or zinc sulphate have all been found to be suitable reagents. The moisture content of the concentrate should desirably be adjusted to promote conditioning and this may require the addition of water or a sufficient volume of reagent solution.
Although the said mixer has the ability to breakdown any coarse lumps in the concentrate the applicants have found in some cases that the best results are obtained when the concentrate is in a finely divided form before the conditioning operation. To achieve this situation the applicants have found it to be advantageous to subject the concentrate to a size reduction operation so that the concentrate so treated and passing to the conditioning step is substantially less than 0.1 mm in size. Suitable equipment for size reduction, such as crushers and disintegrators, are well known to those skilled in the art, and the applicants do not wish to be restricted to any one particular type of crusher.
It would be expected by those skillled in the art that the conditioned concentrate would have to be dried before it is fed to the roaster to prevent excessive breakdown and dust losses occurring during the roasting process. It would also be expected by those skilled in the art that the conditioned concentrate would have to be subjected to size separation to remove particles which were too coarse or too fine, or both, before roasting. For example, British Patent 809,765 discloses a process for the roasting of zinc sulphide concentrates in which the concentrate is pelletised in the size range of 4 to 65 Tyler mesh (4.76 to 0.21 mm). In the Example given in British patent 809,765 the pellets were dried and screened to give a size range of 6 to 20 Tyler mesh (3.36 to 0.841 mm) before being roasted. Surprisingly, the applicants have found that it is not necessary to dry the conditioned concentrate and therefore the cost of drying is avoided. Furthermore, the applicants have found it possible to control the conditioning process so that very little, if any, coarse or undersize particles are formed or remain, and therefore it is not necessary to subject the conditioned concentrate to a size separation operation.
In an embodiment of the second procedure of the present invention, intermediate size particles, of the previously specified size range, are formed by conditioning calcine dust with any of the said reagents in the said mixer, with or without the addition of concentrate. This has the added advantage of maximising sulphur removal.
The first two procedures of the present invention are illustrated by the following Examples.

Example 1

This Example illustrates the first procedure of the present invention. Samples of zinc sulphide concentrates A and B containing:
were screened at 10 mm. The screen analysis of the minus 10 mm screened concentrates was:
The screen analysis was determined by placing 100 g of oven dried concentrate on a sieve shaker for 5 minutes and is not necessarily representative of the sizing of the concentrates in the moist state. Concentrate A was originally a very fine concentrate which had oxidised and partly formed coarse lumps. Concentrate B was a fine unoxidised concentrate.
The concentrates were roasted in a pilot fluidised-bed roaster. The roaster had an internal diameter of 457 mm but this expanded to a diameter of 660 mm above the bed. The bed overflowed through a pipe, 1000 mm above the grate. The solid feed entered the roaster at a point 500 mm above the bed overflow pipe. The bed temperature was controlled at 875°C by means of a cooling coil and water sprays. Air was used to fluidise the bed, and the flowrate was controlled to give a velocity, based on the empty cross sectional area of the bed, of 40 m/min at 875°C. This was equivalent to the maximum operation velocity in a certain full-scale roaster. The roaster gases left the top of the roaster and the entrained calcine dust was collected in a cyclone and scrubber.
Table 1 shows the screen analysis of calcine taken from bed overflow or from the bed at the end of a particular test.
In Test 1.1, the moist concentrates A and B which had been screened at 10 mm were blended together in the proportions of 0.65 parts of A to 0.35 parts of B and fed at the rate of about 50 kg/hr to the pilot roaster. Within 11 hours of operation the bed defluidised. Table 1 shows that the bed at the end of Test 1.1 was very coarse.
In Test 1.2, concentrate A was screened at 6 mm and fed to the pilot roaster alone at the rate of about 50 kg/hr. Other conditions were the same as those in Test 1. Within 8 hours of operation the bed showed signs of partial defluidisation and after 13 hours of operation it was completely defluidised. Hot samples of bed overflow calcine appeared to be sticky and did not flow freely. The screen analysis for Test 1.2 in Table 1 shows that the bed overflow from a poorly fluidised bed can contain a lot of fine particles. Table 1 also shows that the bed at the end of Test 1.2 was very coarse and it is evident that screening concentrate A at 6 mm did not prevent the bed from defluidising.
Test 1.3 was a repeat of Test 1.1 except that 0.2 parts of classified calcine with a particle size range of 0.1 to 1 mm were added to 1 part of moist blended concentrates screened at 10 mm. The feed rate to the pilot roaster was about 50 kg/hr of moist blended concentrates and 10 kg/hr of classified calcine. The classified calcine was prepared by screening calcine with 0.1 and 1 mm mesh sieves. Initially, bed calcines from previous tests were used to prepare the classified material. However, as Test 1.3 proceeded, only freshly generated bed overflow calcine was used to prepare the classified material. Within 7 hours the bed showed signs of defluidisation and the test was abandoned. Table 1 shows that the proportion of 0.1 to 1 mm material in the bed had been raised to 65.6% by recycle but this was evidently insufficient to maintain fluidisation. The minimum fluidising velocity of the bed at the end of Test 1.3 was measured at ambient temperature and found to be 50 m/min.
Test 1.4 was a repeat of Test 1.3 except that the recycle of classified calcine was raised to 0.33 parts for 1 part of moist blended concentrates. The feed rate to the pilot roaster was about 49 kg/hr of moist blended concentrates and 16 kg/hr of classified calcine. Throughout the test there was no evidence of defluidisation, operating conditions were steady and 27.5 kg/hr of bed overflow was produced. Table 1 shows that the proportion of 0.1 to 1 mm material in the bed overflow was 78.7%. The minimum fluidising velocity of the bed overflow was measured at ambient temperature and found to be 23 m/min. The result of Test 1.4 show that with sufficient recycle of classified calcine, fluidisation can be maintained.
Test 1.5 was a repeat of Test 1.2 except that 0.33 parts of classified calcine with a size range of 0.1 to 1 mm were added to 1 part of moist concentrate A which was screened at 10 mm, not 6 mm. The feed rate to the pilot roaster was about 49 kg/hr of concentrate A and 16 kg/hr of classified calcine. Throughout the test there was no evidence of defluidisation or sticky bed overflows. Operating conditions were steady and 37 kg/hr of bed overflow was produced. Table 1 shows that the proportion of 0.1 to 1 mm material in the bed overflow was 70.8% and only 3.7% of material finer than 0.1 mm was present. The minimum fluidising velocity of the bed overflow was 40 m/min at ambient temperature. These results illustrate that with sufficient recycle of classified calcine, fluidisation can be maintained and that bed overflows that are substantially free of fine sticky particles can be produced. The amount of recycle chosen for concentrate A in Test 1.5 was probably the bare minimum.

Example 2

This Example illustrates the second procedure of the present invention.
A moist sample of concentrate A was treated in a crusher to give a product with about 80% passing 0.1 mm.
The crushing concentrate A was conditioned in a V-blender by adding to 10 kg batches between 120 and 200 mLs of a solution containing 130 g of Zn/L as ZnSO₄, together with sufficient water to give a moisture content of 11% and mixing for 3 minutes. The screen analysis of the conditioned concentrate was:
This screen analysis was determined by placing 100 g of oven-dried conditioned concentrate on a sieve shaker for 5 minutes and is not necessarily representative of the sizing in the moist state.
Tests were conducted in the pilot roaster using the same operating conditions as those described in Example 1. The screen analysis of samples of calcine taken from bed overflow are shown in Table 2.
In Test 2.1, 0.65 parts of moist conditioned concentrate A were blended with 0.35 parts of moist concentrate B, which had been screened at 10 mm. The moist blend was fed to the pilot roaster at the rate of 55 kg/hr. Throughout the test there was no evidence of defluidisation. The bed overflow had a similar sizing to that in Tests 1.4 and 1.5 of Example 1 and had a minimum fluidising velocity of 34 m/min at ambient temperature. The amount of bed overflow was 15 kg/hr which is 27.3% of the feed rate and corresponds to the typical performance of a certain full-scale fluidised-bed roaster. Test 2.1 illustrates that moist conditioned concentrate can be added to another unconditioned concentrate as a supply of intermediate size material to form a well fluidised bed without generating an excessive amount of dust.
In Test 2.2, the moist conditioned concentrate A was fed to the pilot roaster, without any concentrate B, at the rate of about 53 kg/hr. There was no evidence of defluidisation during the test. The bed overflow had a similar sizing to that in Tests 1.4 and 1.5 of Example 1 and the minimum fluidising velocity was 32 m/min at ambient temperature. The amount of bed overflow was 20.1 kg/hr which is 37.9% of the feed rate. Test 2.2 confirms that moist conditioned concentrate can be roasted alone without generating an excessive amount of dust.
It will be clearly understood that the invention in its general aspects is not limited to the specific details referred to hereinabove.

Claims

1. Process for improving the fluidising properties of material in the bed of a fluidised-bed roaster, said material including particles in fine, intermediate, and coarse size ranges, characterised by increasing the proportion of material of intermediate size range in the said bed by at least one of the following procedures (a), (b) and (c):

(a) subjecting the material discharged from the said bed to size separation thereby providing an intermediate size fraction and recycling the said fraction to the said bed;

(b) conditioning a proportion of roaster feed material in a mixer to produce material of intermediate size range and feeding this into the roaster without drying or size separation;

(c) in the case in which material discharged from the said bed already lies within the desired intermediate size range, recycling this material to the said bed without subjecting it to a size separation operation.

2. Process according to claim 1 in which the material of intermediate size range is between 0.1 and 3 mm, preferably between 0.1 and 1 mm, and most preferably between 0.1 and 0.5 mm.

3. Process according to claim 1 in which the percentage of material of intermediate size range recycled is between 10 and 100% of the fresh material fed to the roaster.

4. Process according to claim 1 in which the proportion of roaster feed material which is conditioned is between 10 and 100%.

5. Process according to claim 1 in which a reagent is used to assist conditioning, and said reagent is selected from bentonite, sulphite liquor, calcine dust together with sulphuric acid, a solution of molasses, and a solution of zinc sulphate.

6. Process according to claim 1 in which a proportion of roaster feed material is conditioned in a mixer with a high shearing action.

7. Process according to claim 1 for improving the fluidising properties of material in the bed of a fluidised-bed roaster, said material including particles in an intermediate size range between 0.1 and 3 mm, and in fine and coarse size ranges respectively smaller and larger than the said intermediate size range; characterised by subjecting material discharged from the said bed to size separation thereby providing a fraction in the said intermediate size range and recycling the said fraction to the said bed in a proportion between 10 and 100% of the fresh material fed to the roaster.

8. Process according to claim 1 for improving the fluidising properties of material in the bed of a fluidised-bed roaster, said material including particles in an intermediate size range between 0.1 and 3 mm, and in fine and coarse size ranges respectively smaller and larger than the said intermediate size range; characterised by conditioning a proportion of roaster feed material in a mixer to produce material of the said intermediate size range and feeding this into the roaster without drying or size separation, the amount of roaster feed material which is so conditioned being between 10 and 100% of the total roaster feed material.

9. Process according to claim 7 or claim 8 in which the intermediate size range is between 0.1 and 1 mm and more preferably between 0.1 and 0.5 mm.

10. Process according to claim 8 in which a reagent is used to assist conditioning, and said reagent is selected from bentonite, sulphite liquor, calcine dust together with sulphuric acid, a solution of molasses, and a solution of zinc sulphate.

11. Process according to claim 10 in which the conditioning is carried out in a mixer with a high shearing action.