Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK First edition 2010 Copyright © 2010 Elsevier B.V. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-444-53082-0
For information on all Elsevier publications visit our website at books.elsevier.com
Printed and bound in The Netherlands 11 12 13 10 9 8 7 6 5 4 3 2 1
Working together to grow libraries in developing countries www.elsevier.com | www.bookaid.org | www.sabre.org
Introduction
Volume 2 of the ‘Flotation Reagents Handbook’ is a continuation of Volume 1, and presents fundamental and practical knowledge on flotation of gold, platinum group minerals and the major oxide minerals, as well as rare earths. Rather than reiterating what is well known about flotation of gold, PGMs and oxide minerals, emphasis has been placed on the separation methods which are not so effective when using conventional treatment processes. These difficult separation methods are largely attributed to problems with selectivity between valuable minerals and gangue minerals, especially in the flotation of oxide ores and base metal oxides, such as copper, lead and zinc oxide ores. Literature on flotation of gold, PGMs, rare earths and various oxides is rather limited, compared to literature on treatment of sulphide-bearing ores. As mentioned earlier, the main problem arises from the presence of gangue minerals in the ore, which have flotation properties similar to those of valuable minerals. These minerals have a greater floatability than that of pyrochlore or columbite. In the beneficiation of oxide minerals, finding a selectivity solution is a major task. This volume of the Handbook is devoted to the beneficiation of gold, platinum group minerals and, most important, oxide minerals. The book contains details on flotation properties of the major minerals. The fundamental research carried out by a number of research organizations over the past several decades is also contained in this book. Commercial plant practices for most oxide minerals are also presented. The major objective of this volume of the Handbook is to provide practical mineral processors that are faced with the problem of beneficiation of difficult-to-treat ores, with a comprehensive digest of information available, thus enabling them to carry out their development testwork in a more systematic manner and to assist in the control of operating plants. This book will also provide valuable background information for researchers, university students and professors. The book contains comprehensive references of worldwide literature on the subject. New technologies for most of the oxide minerals included in this volume were developed by the author.
ix
– 17 –
Flotation of Gold Ores
17.1
INTRODUCTION
The recovery of gold from gold-bearing ores depends largely on the nature of the deposit, the mineralogy of the ore and the distribution of gold in the ore. The methods used for the recovery of gold consist of the following unit operations: 1.
2.
3.
4.
The gravity preconcentration method, which is used mainly for recovery of gold from placer deposits that contain coarse native gold. Gravity is often used in combination with flotation and/or cyanidation. Hydrometallurgical methods are normally employed for recovery of gold from oxidized deposits (heap leach), low-grade sulphide ores (cyanidation, CIP, CIL) and refractory gold ores (autoclave, biological decomposition followed by cyanidation). A combination of pyrometallurgical (roasting) and hydrometallurgical route is used for highly refractory gold ores (carbonaceous sulphides, arsenical gold ores) and the ores that contain impurities that result in high consumption of cyanide, which have to be removed before cyanidation. The flotation method is a technique widely used for the recovery of gold from goldcontaining copper ores, base metal ores, copper nickel ores, platinum group ores and many other ores where other processes are not applicable. Flotation is also used for the removal of interfering impurities before hydrometallurgical treatment (i.e. carbon prefloat), for upgrading of low-sulphide and refractory ores for further treatment. Flotation is considered to be the most cost-effective method for concentrating gold.
Significant progress has been made over the past several decades in recovery of gold using hydrometallurgical methods, including cyanidation (CIL, resin-in-pulp), bio-oxidation, etc. All of these processes are well documented in the literature [1,2] and abundantly described. However, very little is known about the flotation properties of gold contained in various ores and the sulphides that carry gold. The sparse distribution of discrete gold minerals, as well as their exceedingly low concentrations in the ore, is one of the principal reasons for the lack of fundamental work on the flotation of gold-bearing ores. In spite of the lack of basic research on flotation of gold-bearing ores, the flotation technique is used not only for upgrading of low-grade gold ore for further treatment, but
1
2
17.
Flotation of Gold Ores
also for beneficiation and separation of difficult-to-treat (refractory) gold ores. Flotation is also the best method for recovery of gold from base metal ores and gold-containing PGM ores. Excluding gravity preconcentration, flotation remains the most cost-effective bene ficiation method. Gold itself is a rare metal and the average grades for low-grade deposits vary between 3 and 6 ppm. Gold occurs predominantly in native form in silicate veins, alluvial and placer deposits or encapsulated in sulphides. Other common occurrences of gold are alloys with copper, tellurium, antimony, selenium, platinum group metals and silver. In massive sulphide ores, gold may occur in several of the above forms, which affects flotation recovery. During flotation of gold-bearing massive sulphide ores, the emphasis is generally placed on the production of base metal concentrates and gold recovery becomes a secondary consideration. In some cases, where significant quantities of gold are contained in base metal ores, the gold is floated from the base metal tailings. The flotation of gold-bearing ores is classified according to ore type (i.e. gold ore, gold copper ore, gold antimony ores, etc.), because the flotation methods used for the recovery of gold from different ores is vastly different.
17.2
GEOLOGY AND GENERAL MINERALOGY OF GOLD-BEARING ORES
The geology of the deposit and the mineralogy of the ore play a decisive role in the selection of the best treatment method for a particular gold ore. Geology of the gold deposits [3] varies considerably not only from deposit to deposit, but also within the deposit. Table 17.1 shows major genetic types of gold ores and their mineral composition. More than 50% of the total world gold production comes from clastic sedimentary deposits.
Table 17.1 Common genetic types of gold deposits Ore type
Description
Magmatic
Gold occurs as an alloy with copper, nickel and platinum group metals. Typically contains low amount of gold Placer deposits, in general conglomerates, which contain quartz, sericite, chlorite, tourmaline and sometimes rutile and graphite. Gold can be coarse. Some deposits contain up to 3% pyrite. Size of the gold contained in pyrite ranges from 0.01to 0.07 μm This type contains a variety of ores, including(a) gold-pyrite ores, (b) goldcopper ores, (c) gold-polymetallic ores and (d) gold oxide ore, usually upper zone of sulphide zones. The pyrite content of the ore varies from 3% to 90%. Other common waste minerals are quartz, aluminosilicates, dolomite etc. Sometimes are very complex and refractory gold ores. Normally the ores are composed of quartz, sericite, chlorites, calcite and magnetite. Sometimes the ore contains wolframite and scheelite
Ores in clastic sedimentary rock Hydrothermal
Metasomatic or scarn ores
17.3
Flotation Properties of Gold Minerals and Factors Affecting Floatability
3
Table 17.2 Major gold minerals Group
Mineral
Chemical formula
Impurity content
Native gold and its alloys
Native gold Electrum Cuproauride Amalgam Bismuthauride
Au Au/Ag Au/Cu Hg/Au Au/Bi
0–15% Ag 15–50% Ag 5–10% Cu 10–34% Au 2–4% Bi
Tellurides
Calaverite Sylvanite Petzite Magyazite
AuTe3 (Au,Ag)Te2 (Au,Ag)Te Au(Pb,Sb,Fe)(S,Te11)
Unstable
Krennerite Platinum gold Rhodite Rhodian gold Aurosmiride
AuTe2(Pt,Pl) AuPt AuRh AuRh Au,Ir,Os
Up to 10% Pt 30–40% Rh 5–11% Rh 5% Os + 5–7% Ir
Gold associated with platinum group metals
In many geological ore types, several sub-types can be found including primary ores, secondary ores and oxide ores. Some of the secondary ores belong to a group of highly refractory ores, such as those from Nevada (USA) and Chile (El Indio). The number of old minerals and their associations are relatively small and can be divided into the following three groups: (a) native gold and its alloys, (b) tellurides and (c) gold associated with platinum group metals. Table 17.2 lists the major gold minerals and their associations.
17.3 FLOTATION PROPERTIES OF GOLD MINERALS AND FACTORS AFFECTING FLOATABILITY Native gold and its alloys, which are free from surface contaminants, are readily floatable with xanthate collectors. Very often however, gold surfaces are contaminated or covered with varieties of impurities [4]. The impurities present on gold surfaces may be argentite, iron oxides, galena, arsenopyrite or copper oxides. The thickness of the layer may be of the order of 1–5 µm. Because of this, the flotation properties of native gold and its alloys vary widely. Gold covered with iron oxides or oxide copper is very difficult to float and requires special treatment to remove the contaminants. Tellurides, on the other hand, are readily floatable in the presence of small quantities of collector, and it is believed that tellurides are naturally hydrophobic. Tellurides from Minnesota (USA) were floated using dithiophosphate collectors, with over 9% gold recovery.
4
17.
Flotation of Gold Ores
30
Adsorption of xanthate (%)
3 25 20
2
15 10 1 5 0 0
10
20
30
40
50
60
70
80
Conditioning time with xanthate (minutes)
Figure 17.1 Relationship between adsorption of xanthate on gold and conditioning time in the presence of various concentrations of xanthate.
Flotation behaviour of gold associated in the platinum group metals is apparently the same as that for the platinum group minerals (PGMs) or other minerals associated with the PGMs (i.e. nickel, pyrrhotite, copper and pyrite). Therefore, the reagent scheme developed for PGMs also recovers gold. Normally, for the flotation of PGMs and associated gold, a combination of xanthate and dithiophosphate is used, along with gangue depressants guar gum, dextrin or modified cellulose. In the South African PGM operations, gold recovery into the PGM concentrate ranges from 75% to 80%. Perhaps the most difficult problem in flotation of native gold and its alloys is the tendency of gold to plate, vein, flake and assume many shapes during grinding. Particles with sharp edges tend to detach from the air bubbles, resulting in gold losses. This shape factor also affects gold recovery using a gravity method. In flotation of gold-containing base metal ores, a number of modifiers normally used for selective flotation of copper lead, lead zinc and copper lead zinc have a negative effect on the floatability of gold. Such modifiers include ZnSO4·7H2O, SO2, Na2S2O5 and cyanide when added in excessive amounts. The adsorption of collector on gold and its floatability is considerably improved by the presence of oxygen. Figure 17.1 shows the relationship between collector adsorption, oxygen concentration in the pulp and conditioning time [4]. The type of modifier and the pH are also important parameters in flotation of gold.
17.4
FLOTATION OF LOW-SULPHIDE-CONTAINING GOLD ORES
The beneficiation of this ore type usually involves a combination of gravity concentra tion, cyanidation and flotation. For an ore with coarse gold, gold is often recovered by gravity and flotation, followed by cyanidation of the reground flotation concentrate. In
17.6
Flotation of Carbonaceous Clay-Containing Gold Ores
5
some cases, flotation is also conducted on the cyanidation tailing. The reagent combina tion used in flotation depends on the nature of gangue present in the ore. The usual collectors are xanthates, dithiophosphates and mercaptans. In the scavenging section of the flotation circuit, two types of collector are used as secondary collectors. In the case of a partially oxidized ore, auxiliary collectors, such as hydrocarbon oils with sulphidi zer, often yield improved results. The preferred pH regulator is soda ash, which acts as a dispersant and also as a complexing reagent for some heavy metal cations that have a negative effect on gold flotation. Use of lime often results in the depression of native gold and gold-bearing sulphides. The optimum flotation pH ranges between 8.5 and 10.0. The type of frother also plays an important role in the flotation of native gold and gold-bearing sulphides. Glycol esters and cyclic alcohols (pine oil) can improve gold recovery significantly. Amongst the modifying reagents (depressant), sodium silicate starch dextrins and low molecular-weight polyacrylamides are often selected as gangue depressants. Fluorosilicic acid and its salts can also have a positive effect on the floatability of gold. The presence of soluble iron in a pulp is highly detrimental for gold flotation. The use of small quantities of iron-complexing agents, such as polyphosphates and organic acids, can eliminate the harmful effect of iron.
17.5
FLOTATION OF GOLD-CONTAINING MERCURY/ANTIMONY ORES
In general, these ores belong to a group of difficult-to-treat ores, where cyanidation usually produces poor extraction. Mercury is partially soluble in cyanide, which increases consumption and reduces extraction. A successful flotation method [5] has been developed using the flowsheet shown in Figure 17.2, where the best metallurgical results were obtained using a three-stage grinding and flotation approach. The metallurgical results obtained with different grinding configurations are shown in Table 17.3. Flotation was carried out at an alkaline pH, controlled by lime. A xanthate collector with cyclic alcohol frother (pine oil, cresylic acid) was shown to be the most effective. The use of small quantities of a dithiophosphate-type collector, together with xanthate was beneficial.
17.6
FLOTATION OF CARBONACEOUS CLAY-CONTAINING GOLD ORES
These ores belong to a group of refractory gold ores, where flotation techniques can be used to (a) remove interfering impurities before the hydrometallurgical treatment process of the ore for gold recovery, and (b) to preconcentrate the ore for further pyrometallur gical or hydrometallurgical treatment. There are several flotation methods used for beneficiation of this ore type. Some of the most important methods are described below.
6
17.
Flotation of Gold Ores
Feed Grind 1 Classification 1 Classification 2
Grind 2
Scalp Float
Classification Flotation 1 Cleaner
Classification Grind 3 Flotation 2
Cleaner 1 Cleaner 2 Cleaner 3
Final tailing
Concentrate to smelter
Figure 17.2 Flotation flowsheet developed for the treatment of gold-containing mercury–antimony ore.
Table 17.3 Gold recovery obtained using different flowsheets [5] Product
Single-stage grind-flotation Two-stage grind-flotation Three-stage grind-flotation
% Recovery in concentrate
Tailing assays (%, g/t)
Au
Ag
Sb
As
S
Au
Ag
Sb
As
S
88.1 92.2 95.3
89.2 91.8 95.2
72.9 93.4 95.7
68.4 78.7 81.2
70.1 81.2 85.7
1.7 1.0 0.7
5.0 4.1 2.2
0.04 0.015 0.005
0.035 0.022 0.015
0.38 0.27 0.19
17.6
Flotation of Carbonaceous Clay-Containing Gold Ores
17.6.1
7
Preflotation of carbonaceous gangue and carbon
In this technique, only carbonaceous gangue and carbon are recovered by flotation, in preparation for further hydrometallurgical treatment of the float tails for gold recovery. Carbonaceous gangue and carbon are naturally floatable using only a frother, or a combi nation of a frother and a light hydrocarbon oil (fuel oil, kerosene, etc.). When the ore contains clay, regulators for clay dispersion are used. Some of the more effective regulating reagents include sodium silicates and oxidized starch.
17.6.2
Two-stage flotation method
In this technique, carbonaceous gangue is prefloated using the above-described method, followed by flotation of gold-containing sulphides using activator–collector combinations. In extensive studies [6] conducted on carbonaceous gold-containing ores, it was established that primary amine-treated copper sulphate improved gold recovery considerably. Ammonium salts and sodium sulphide (Na2S · 9H2O) also have a positive effect on gold-bearing sulphide flotation, at a pH between 7.5 and 9.0. The metallurgical results obtained with and without modified copper sulphate are shown in Table 17.4.
17.6.3
Nitrogen atmosphere flotation method
This technique uses a nitrogen atmosphere in grinding and flotation to retard oxidation of reactive sulphides, and has been successfully applied on carbonaceous ores from Nevada (USA). The effectiveness of the method depends on (a) the amount of carbo naceous gangue present in the ore, and (b) the amount and type of clay. Ores that are high in carbon or contain high clay content (or both) are not amenable for nitrogen atmosphere flotation.
Table 17.4 Effect of amine-modified CuSO4 on gold-bearing sulphide flotation from carbonaceous refractory ore Reagent used
CuSO4 + xanthate Amine modified CuSO4 + xanthate
Product
Gold sulphide concentrate Gold sulphide tail Head Gold sulphide concentrate Gold sulphide tail Head
Weight (%)
30.11 69.89 100.00 26.30 73.70 100.00
Assays (%, g/t) Au
S
9.63 1.86 4.20 13.2 0.85 4.10
4.50 0.49 1.70 5.80 0.21 1.68
% Distribution Au 69.1 30.9 100.0 84.7 15.3 100.0
S 79.7 20.3 100.0 90.8 9.2 100.0
8
17.
17.7
Flotation of Gold Ores
FLOTATION OF GOLD-CONTAINING COPPER ORES
The floatability of gold from gold-containing copper gold ores depends on the nature and occurrence of gold in these ores, and its association with iron sulphides. Gold in the porphyry copper ore may appear as native gold, electrum, cuproaurid and sulphosalts associated with silver. During the flotation of porphyry copper-gold ores, emphasis is usually placed on the production of a marketable copper-gold concentrate and optimization of gold recovery is usually constrained by the marketability of its concentrate. The minerals that influence gold recovery in these ores are iron sulphides (i.e. pyrite, marcasite, etc.), in which gold is usually associated as minute inclusions. Thus, the iron sulphide content of the ore determines gold recovery in the final concentrate. Figure 17.3 shows the relationship between pyrite content of the ore and gold recovery in the copper concentrate for two different ore types. Most of the gold losses occur in the pyrite. The reagent schemes used in commercial operations treating porphyry copper–gold ores vary considerably. Some operations, where pyrite rejection is a problem, use a dithiopho sphate collector at an alkaline pH between 9.0 and 11.8 (e.g. OK Tedi/PNG Grasberg/ Indonesia). When the pyrite content in the ore is low, xanthate and dithiophosphates are used in a lime or soda ash environment. In more recent years, in the development of commercial processes for the recovery of gold from porphyry copper–gold ores, bulk flotation of all the sulphides has been empha sized, followed by regrinding of the bulk concentrate and sequential flotation of copper– gold from pyrite. Such a flowsheet (Figure 17.4) can also incorporate high-intensity conditioning in the cleaner–scavenger stage. Comparison of metallurgical results using the standard sequential flotation flowsheet and the bulk flotation flowsheet are shown in Table 17.5. A considerable improvement in gold recovery was achieved using the bulk flotation flowsheet.
Gold recovery in Cu cleaner conc. (%)
100
80
60 1 40 2
20
0 0
1
2
3 4 5 6 7 8 Pyrite content of ore (%)
9
10
Figure 17.3 Effect of pyrite content of the ore on gold recovery in the copper–gold concentrate at 30% Cu concentrate grade (1: ore from Peru; 2: ore from Indonesia).
17.7
Flotation of Gold-Containing Copper Ores
Flotation feed
9
Bulk scavenger
Bulk rougher
Regrind High-intensity conditioning
Cu-Au rougher
Cu-Au cleaner 1
Cu-Au scavenger
Combined tailing
Cu-Au cleaner 2 Cu-Au cleaner 3 Cu-Au cleaner concentrate
Figure 17.4
Bulk flowsheet used in the treatment of pyritic copper–gold ores [8]. Table 17.5
Comparison of metallurgical results using conventional and bulk flotation flowsheets on ore from peru Flowsheet used
Conventional (sequential Cu/Au) Bulk (Figure 17.4)
Product
Cu/Au concentrate Cu/Au tail Head Cu/Au concentrate Cu/Au ail Head
Weight (%)
2.28 97.72 100.00 2.32 97.68 100.00
Assays (%, g/t)
% Distribution
Au
S
Au
S
27.6 0.031 0.66 27.1 0.032 0.66
32.97 0.23 0.98 36.94 0.14 0.96
95.4 4.6 100.0 95.2 4.8 100.0
76.7 23.3 100.0 85.8 14.2 100.0
During beneficiation of clay-containing copper-gold ores, the use of small quantities of Na2S (at natural pH) improves both copper and gold metallurgy considerably. In the presence of soluble cations (e.g. Fe, Cu), additions of small quantities of organic acid (e.g. oxalic, tartaric) improve gold recovery in the copper concentrate. Some porphyry copper ores contain naturally floatable gangue minerals, such as chlor ites and aluminosilicates, as well as preactivated quartz. Sodium silicate, carboxy methylcellulose and dextrins are common depressants used to control gangue flotation. Gold recovery from massive sulphide copper–gold ores is usually much lower than that of porphyry copper–gold ores, because very often a large portion of the gold is associated with pyrite. Normally, gold recovery from these ores does not exceed 60%. During the treatment of copper–gold ores containing pyrrhotite and marcasite, the use of Na2H2PO4 at alkaline pH values depresses pyrrhotite and marcasite, and also improves copper and gold metallurgy.
10
17.
17.8
Flotation of Gold Ores
FLOTATION OF OXIDE COPPER–GOLD ORES
Oxide copper–gold ores are usually accompanied by iron hydroxide slimes and various clay minerals. There are several deposits of this ore type around the world, some of which are located in Australia (Red Dome), Brazil (Igarape Bahia) and the Soviet Union (Kalima). Treatment of these ores is difficult, and even more complicated in the presence of clay minerals. Recently, a new class of collectors, based on ester-modified xanthates, have been successfully used to treat gold-containing oxide copper ores, using a sulphidization method. Table 17.6 compares the metallurgical results obtained on the Igarape Bahia ore using xanthate and a new collector (PM230, supplied by Senmin in South Africa). The modifier used in the flotation of these ores included a mixture of sodium silicate and Calgon. Good selectivity was also achieved using boiled starch.
17.9
FLOTATION OF GOLD–ANTIMONY ORES
Gold–antimony ores usually contain stibnite (1.5–4.0% Sb), pyrite, arsenopyrite, gold (1.5–3.0 g/t) and silver (40–150 g/t). Several plants in the United States (i.e. Stibnite/ Minnesota and Bradly) and Russia have been in operation for some time. There are two commercial processes available for treatment of these ores: 1.
2.
Selective flotation of gold-containing sulphides followed by flotation of stibnite with pH change. Stibnite floats well in neutral and weak acid pH, whereas in an alkaline pH (i.e. >8) it is reduced. Utilizing this phenomenon, gold-bearing sulphides are floated with xanthate and alcohol frother in alkaline medium (i.e. pH > 9.3) followed by stibnite flotation at about pH 6.0, after activation with lead nitrate. Typical metallurgical results using this method are shown in Table 17.7. Bulk flotation followed by sequential flotation of gold-bearing sulphides, and depression of stibnite. This method was practiced at the Bradly concentrator (USA) Table 17.6
Effect of collector PM230 on copper/gold recovery from Igarape Bahia oxide copper/gold ore [8] Reagent used
Product
Weight (%)
Assays (%, g/t) Au
Na2S = 2500 g/t PAXa = 200 g/t Na2S = 2500 g/t PAXa/PM230 (1:1) = 200 g/t a
PAX = potassium amyl xanthate.
Copper Cl concentrate Copper tail Feed Copper Cl concentrate Copper tail Feed
9.36 90.64 100.00 10.20 89.80 100.00
% Distribution S
33.3 14.15 1.61 1.46 4.65 2.65 39.5 21.79 0.61 0.42 0.61 0.42
Au
S
67.0 50.0 33.0 50.0 100.0 100.0 88.0 85.5 12.0 14.5 12.0 14.5
17.10
Flotation of Arsenical Gold Ores
11 Table 17.7
Metallurgical results obtained using a sequential flotation method Product
Weight (%)
Gold concentrate Stibnite concentrate Tailing Feed
2.34 4.04 93.62 100.00
Assays (%, g/t)
% Distribution
Au
Ag
Sb
Au
Ag
Sb
42.3 6.2 0.65 1.86
269.3 559.8 18.7 46.4
20.0 51.0 0.7 3.2
53 13 34 100.0
13 51 36 100.0
15 64 21 10.0
Courtesy of stibnite plant (Minnesota, 1976).
Table 17.8 Plant metallurgical results obtained using a bulk flotation method Product
Weight (%)
Gold concentrate Antimony concentrate Middlings Bulk concentrate Tailing Feed
1.80 1.80 0.50 4.10 95.90 100.00
Assays (%, g/t)
% Distribution
Au
Ag
Sb
Au
Ag
Sb
91.1 13.0 46.6 51.7 0.6 2.7
248.8 684.2 248.8 440.0 3.1 21.0
1.5 51.3 20.0 29.0 0.2 1.3
61.0 9.0 8.6 78.6 21.4 100.0
31.3 58.6 6.0 85.9 14.1 100.0
2.0 75.0 8.0 85.0 15.0 100.0
Courtesy of the Bradly concentrator (USA).
and consisted of the following steps: (a) bulk flotation of stibnite and gold-bearing sulphides at pH 6.5 using lead nitrate (i.e. Sb activator) and xanthate, (b) the bulk concentrate is reground in the presence of NaOH (pH 10.5) and CuSO4, and the goldbearing sulphides are refloated with additions of small quantities of xanthate, (c) cleaning of the gold concentrate in the presence of NaOH and NaHS. The plant metallurgical results employing this method are shown in Table 17.8. Recent studies conducted on ore from Kazakhstan have shown that sequential flotation using thionocarbamate collector gave better metallurgical results than those obtained with xanthate.
17.10
FLOTATION OF ARSENICAL GOLD ORES
There are two major groups of arsenical gold ores of economical value. These are the massive base metal sulphides with arsenical gold (i.e. the lead–zinc Olympias deposit, Greece) and arsenical gold ores without the presence of base metals. Massive, base metal
12
17.
Flotation of Gold Ores
arsenical gold ores are rare, and there are only a few deposits in the world. A typical arsenical gold ore contains arsenopyrite as the major arsenic mineral. However, some arsenical gold ores, such as those from Nevada in the USA (Getchel deposit), contain realgar and orpiment as the major arsenic-bearing minerals. Pyrite, if present in an arsenical gold ore, may contain some gold as minute inclusions. Flotation of arsenical gold ores associated with base metals is accomplished using a sequential flotation technique, with flotation of base metals followed by flotation of goldcontaining pyrite/arsenopyrite. The pyrite/arsenopyrite is floated at a weakly acid pH with a xanthate collector. Arsenical gold ores that do not contain significant base metals are treated using a bulk flotation method, where all the sulphides are first recovered into a bulk concentrate. In case the gold is contained either in pyrite or arsenopyrite, separation of pyrite and arsenopyrite is practiced. There are two commercial methods available. The first method utilizes arseno pyrite depression and pyrite flotation, and consists of the following steps: 1.
2.
Heat the bulk concentrate to 75°C at a pH of 4.5 (controlled by H2SO4) in the presence of small quantities of potassium permaganate or disodium phosphate. The temperature is maintained for about 10 min. Flotation of pyrite using either ethyl xanthate or potassium butyl xanthate as collector. Glycol frother is also usually employed in this separation.
This method is highly sensitive to temperature. Figure 17.5 shows the effect of tempera ture on pyrite/arsenopyrite separation. In this particular case, most of the gold was associated with pyrite. Successful pyrite/arsenopyrite separation can also be achieved with the use of potassium peroxy disulphide as the arsenopyrite depressant. The second method involves depression of pyrite and flotation of arsenopyrite. In this method, the bulk concentrate is treated with high dosages of lime (i.e. pH > 12), followed
Pyrite/arsenopyrite recovery (%)
100 Pyrite 80 60 40 Arsenopyrite 20 0 0
20 40 60 Heating temperature (°C)
80
Figure 17.5 Effect of temperature on separation of pyrite and arsenopyrite from a bulk pyrite/ arsenopyrite concentrate.
17.11
Flotation of Gold From Base Metal Sulphide Ores
13
by a conditioning step with CuSO4 to activate arsenopyrite. The arsenopyrite is then floated using a thionocarbamate collector. Separation of arsenopyrite and pyrite is important from the point of view of reducing downstream processing costs. Normally, roasting or pressure oxidation followed by cya nidation is used to recover gold.
17.11
FLOTATION OF GOLD FROM BASE METAL SULPHIDE ORES
Very often lead-zinc, copper-zinc, copper-lead-zinc and copper-nickel ores contain signifi cant quantities of gold (i.e. between 1 and 9 g/t). The gold in these ore types is usually found as elemental gold. A large portion of the gold in these ores is finely disseminated in pyrite, which is considered non-recoverable. Because of the importance of producing commercial-grade copper, lead and zinc concentrates, little or no consideration is given to improvement in gold recovery, although the possibility exists to optimize gold recovery in many cases. Normally, gold recovery from base metal ores ranged from 30% to 75%. In the case of a copper-zinc and copper-lead-zinc ore, gold collects in the copper concentrate. During the treatment of lead-zinc ores, the gold tends to report to the lead concentrate. Information regarding gold recovery from base metal ores is sparse. The most recent studies [9] conducted on various base metal ores revealed some important features of flotation behaviour of gold from these ores. It has been demonstrated that gold recovery to the base metal concentrate can be substantially improved with the proper selection of reagent schemes. Some of these studies are discussed below. 17.11.1
Gold-containing lead-zinc ores
Some of these ores contain significant quantities of gold, ranging from 0.9 to 6.0 g/t (i.e. Grum/Yukon, Canada; Greens Creek, Alaska; and Milpo, Peru). The gold recovery from these ores ranged from 35% to 75%. Laboratory studies have shown that the use of high dosages of zinc sulphate, which is a common zinc depressant used in lead flotation, reduces gold floatability significantly. The effect of ZnSO4 · 7H2O addition on gold recovery in the lead concentrate is illustrated in Figure 17.6. In order to improve gold recovery in the lead concentrate, an alternative depressant to ZnSO4 · 7H2O can be used. Depressant combinations such as Na2S + NaCN, or Na2SO3 + NaCN, may be used. The type of collector also plays an important role in gold flotation of lead-zinc ores. A phosphine-based collector, in combination with xanthate, gave better gold recovery than dithiophosphates. 17.11.2
Copper-zinc gold-containing ores
Gold recovery from copper-zinc ores is usually higher than that obtained from either a leadzinc or copper lead-zinc ore. This is attributed to two main factors: when selecting a reagent
17.
Gold recovery in Lead concentrate (%)
14
Flotation of Gold Ores
70 Greens Creek ore (Alaska)
60 50 40
Grum ore Yukon (Canada)
30 20 10 0 0
100
200
300
400
500
ZnSO4 · 7 H2O (g/t)
Figure 17.6
Effect of ZnSO4 additions on gold recovery from lead–zinc ores.
scheme for treatment of Cu-Zn ores, there are more choices than for the other ore types, which can lead to the selection of a reagent scheme more favourable for gold flotation. In addition, a non-cyanide depressant system can be used for the treatment of these ores, which in turn results in improved gold recovery. This option is not available during treatment of lead-zinc ores. Table 17.9 shows the effect of different depressant combina tions on gold recovery from a copper-zinc ore. The use of a non-cyanide depressant system resulted in a substantial improvement in gold recovery in the copper concentrate.
Table 17.9 Effect of different depressant combinations on gold recovery to the copper concentrate from lower zone Kutcho Creek ore Depressant system
Product
Weight (%) Assays (%, g/t) Au
ZnSO4, NaCN, CaO Cu concentrate 3.10 pH 8.5 Cu, 10.5 Zn Zn concentrate 5.34 Tailings 91.56 Feed 100.00 3.05 Na2SO3, NaHS, CaO Cu concentrate pH 8.5 Cu, 10.5 Zn Zn concentrate 5.65 Tailings 91.30 Feed 100.00 Courtesy of Esso Canada Resources.
Ag
% Distribution Sb
Au
Ag
Sb
20.4 26.2 330 45.1 85.6 2.8 1.20 0.61 55.4 4.6 3.4 82.2 0.77 0.11 0.58 50.3 11.0 15.0 1.4 0.95 3.60 100.0 100.0 100.0 32.5 28.1 2.80 68.3 87.4 2.3 1.20 0.55 54.8 4.7 3.2 84.6 0.43 0.10 0.52 27.0 9.4 13.1 1.45 0.98 3.66 100.0 100.0 100.0
17.12
Conclusions
17.11.3
15
Gold-containing copper-lead-zinc ores
Because of the complex nature of these ores, and the requirement for a relatively complex reagent scheme for treatment of this ore, the gold recovery is generally lower than that achieved from a lead-zinc or copper-zinc ore. One of the major problems associated with the flotation of gold from these ores is related to gold mineralogy. A large portion of the gold is usually contained in pyrite, at sub-micron size. If coarse elemental gold and electrum are present, the gold surfaces are often coated with iron or lead, which can result in a substantial reduction in floatability. The type of collector and flowsheet configuration play an important role in gold recovery from these ores. With a flowsheet that uses bulk Cu–Pb flotation followed by Cu–Pb separation, the gold recovery is higher than that achieved with a sequential Cu–Pb flotation flowsheet. In laboratory tests, and Aerophine collector type, in combination with xanthate, had a positive effect on gold recovery as compared to either dithiophosphate or thionocarbamate collectors. Table 17.10 compares the metal lurgical results obtained with an Aerophine collector to those obtained with a dithio phosphate collector. Because of the complex nature of gold-containing Cu–Pb–Zn ores, the reagent schemes used are also complex. Reagent modifiers such as ZnSO4, NaCN and lime have to be used, all of which have a negative effect on gold flotation.
17.12
CONCLUSIONS
The flotation of gold-bearing ores, whether for production of bulk concentrates for further gold recovery processes (i.e. pyrometallurgy, hydrometallurgy) or for recovery of gold to base metal concentrates, is a very important method for concentrating the gold and reducing downstream costs. The flotation of elemental gold, electrum and tellurides is usually very efficient, except when these minerals are floated from base metal, massive sulphides. Flotation of gold-bearing sulphides from ores containing base metal sulphides present many challenges and should be viewed as flotation of the particular mineral that contains gold (i.e. pyrite, arsenopyrite, copper, etc.), because gold is usually associated with these minerals at micron size. Selection of a flotation technique for gold preconcentration depends very much on the ore mineralogy, gangue composition and gold particle size. There is no universal method for flotation of the gold-bearing minerals, and the process is tailored to the ore character istics. A specific reagent scheme and flowsheet are required for each ore. There are opportunities in most operating plants for improving gold metallurgy. Most of these improvements come from selection of more effective reagent schemes, including collectors and modifiers. Perhaps the most difficult ores to treat are the clay-containing carbonaceous sulphides. Significant progress has been made in treatment options for these ores. New sulphide activators (i.e. amine-treated CuSO4, ammonium salts) and nitrogen gas flotation are amongst the new methods available.
16
Table 17.10 Effect of collector type on Cu–Pb–Zn–Au metallurgical results from a high-lead ore, Crandon (USA) Collector
30 g/t xanthate 20 g/t dithiophosphate 3477
30 g/t xanthate 20 g/t aerophine 3418A
Cu concentrate Pb concentrate Zn concentrate Tailing Feed Cu concentrate Pb concentrate Zn concentrate Tailing Feed
Weight (%)
2.47 1.80 13.94 81.79 100.00 2.52 1.92 13.91 81.65 100.00
Assays (%, g/t)
% Distribution
Au
Cu
Pb
Zn
Au
Cu
Pb
Zn
22.4 2.50 1.10 0.71 1.33 31.3 2.80 0.90 0.41 1.30
25.5 0.80 0.60 0.089 0.80 26.1 0.90 0.50 0.093 0.82
1.20 51.5 0.80 0.28 1.30 1.10 51.1 0.72 0.30 1.35
4.50 8.30 58.2 0.52 8.80 5.00 9.20 58.5 0.44 8.80
41.6 3.4 11.5 43.5 100.0 60.6 4.1 9.6 25.7 100.0
78.6 1.8 10.4 9.1 100.0 80.1 2.1 8.5 9.3 100.0
2.3 71.3 8.6 17.8 100.0 2.1 72.5 7.4 18.0 100.0
1.3 1.7 92.2 4.8 100.0 1.4 2.0 92.5 4.1 100.0 17.
Courtesy of Exxon coal.
Product
Flotation of Gold Ores
References
17
REFERENCES 1. Kudryk, V., Carigan, D.A., and Liang, W.W., Precious Metals, Mining Extraction and Processing. AIME, 1982. 2. Martins, V., Dunne, R.C., and Gelfi, P., Treatment of Partially Refractory Gold Ores, Randol Gold Forum, Australia, 1991. 3. Baum, W., Mineralogy as a Metallurgical Tool in Refractory Ore, Progress Selection and Optimization, Randol Gold Forum, Squaw Valley, 1990. 4. Fishman, M.A., and Zelenov, B.I., Practice in Treatment of Sulphides and Precious Metal Ores, Izdatelstro Nedra (Russian), Moscow, Vol. 5, pp. 22–101, 1967. 5. Sristinov, N.B., The Effect of the Use of Stage Grinding in Processing of Refractory ClayContaining Gold Ore, Kolima, No. 1, pp. 34–40, 1964. 6. Bulatovic, S.M., and Wyslouzil, D.M., Proceedings of the 2nd International Gold Symposium, Flotation Behaviour of Gold During Processing of Porphyry Copper-Gold Ores and Refractory Gold-Bearing Sulphides, Lima, Peru, 1996. 7. Bulatovic, S.M., Evaluation of New HD Collectors in Flotation of Pyretic Copper-Gold Ores from B.C. Canada, Internal R&D Report LR029, 1993. 8. Bulatovic, S.M., An Investigation of the Recover of Copper and Gold from Igarape Bahia Oxide Copper-Gold Ores, Report of Investigation LR4533, 1997. 9. Bulatovic, S.M., An Investigation of Gold Flotation from Base Metal Lead-Zinc and Copper-Zinc Ores, Interim Report LR049, 1996.
– 18 –
Flotation of Platinum Group Metal Ores
18.1
INTRODUCTION
In chemical terms the six main platinum group elements (PGE), ruthenium, rhodium, palladium, osmium, iridium and platinum, belong to the group VIII transition metals, to which also belong iron, nickel and cobalt. These elements have long been considered, when grouped with gold and silver, as ‘precious metals’. This, in fact, is misleading because the mineralogy and geochemistry of silver and gold do not correlate with that of PGE. Also, in literature, there are two terms of reference, including PGE and platinum group minerals (PGM). From a flotation point of view, PGM is the more common term. There fore, the term PGM will be used in this text. The chemical similarity between the six PGE and iron, nickel and cobalt accounts for the fact that they tend to concentrate together as a result of geological processes. This is quite important not only for the formation of PGM ores, but also for beneficiation.
18.2
MINERALS AND CLASSIFICATION OF PGM ORES
There are over 100 different platinum group minerals. Some of the most common PGM are shown in Table 18.1. The stoichiometry of most of the PGM named [1] is known, but because these minerals are subject to a wide range of element substitution, as indicated in Table 18.1, there is little consistency between an ideal formula for the individual minerals and compositions of the given minerals from various locations. In general, PGM are concentrates in the crust found in two different ways: (a) by leaching the metal-rich lava (mantle) deposited into the crust, which is known as chemical weathering, especially in a hot climate where silica and magnesia are leached away. This leaves a residue enriched in iron and nickel, which contains the PGM elements; and (b) melting a portion of the mantle may give rise to ultramafic or basalic lava, which is then squeezed upwards as a result of pressure within the earth to intrude the crust or extrude lava on the surface. This magma is not particularly rich in nickel or PGM; however, because of their siderophile nature [2], the group VIII metals are also chalcophile in nature, that is they prefer to form bonds with sulphur than oxygen.
19
20
18.
Flotation of Platinum Group Metal Ores
Table 18.1 List of platinum group minerals and their compositions PGM
Ideal formula
Other elements present
Anduoite Arsenopalladinite Atheneite Atokite Borovskite Braggite Cooperite Daomanite Erlichamanite Froodite Genkinite Geversite Guanglinite Hollingworthite Hongshlite Iravsite Iridium Isoferroplatinum Kotulskite Majakite Monochelite Nigglite Omelite Osmium Palarstanide Palladium Platiniridium Rhodium Ruthenium Ruthenosmiridium Sperrilite Temagamite Uvantserite Vysotaskite Xingzhongite Zvyagintsevite
RuAs2 Pd8As2.5Sb0.5 (PdHg)3As PdSn Pd8SbTe4 (PtPd)S PtS PtCuAsS2 OsS2 PdBi2 (PtPd)4Sb3 PtSb2 Pd3As RhAsS Pt(Cu) IrAsS Ir Pt3Fe PtTe PtNiAs PtTe3 PtSn OsAs2 Os Pd8(SnAs)3 Pd (IrPt) Rh Ru (IrOsRu) PtAs2 PdHgTe3 Pd(BiPb)2 PdS (IrCuRh)S Pd3Pb
(RuOsIr)As (PdCu)AsSb (PdHgAuCu)AsSb (PdPt)Sn (PdPtNiFe)SbBiTe (PtNiPd)S (PtNiPd)S (PtCuAs)S (OsRhIrPdRu)S (PdPt)Bi (PtPdRhNiCu)SbAsBi Pt(SbBi) (Pd)As (RhPdPtIr)AsS (Pt)Cu (IrRuRhPt)AsS (IrPtFeOsRhPdNi) (PtFeCuNi) (PdPt)(TeBiSb) (PdNiAs) PtPd(TeBi) (PtBiSb)Sn (OsRuFeNiIrCo)As (OsIrRuPt) (PdPtAuCu)(AsSnSb) PdHg (IrPtFeOsCuNi) RhPt RuIrRhOsPdFe (IrRuOsPtRhFeNiPd) (Pt)(AsSb) (Pt)HgTe)Bi Pd(BiPb) (PdFePt) (IrCuRhFePbPtOs)S (PdPtFeNiCu)Pb
These sulphide deposits are able to concentrate these metals by a factor of 100–1000 ppm and form PGM deposits, together with precious metals, nickel and copper. Almost always the PGM deposits contain nickel minerals. The PGM deposits can be classified into the following two groups: (a) PGM-dominated deposits and (b) nickel–copper-dominated deposits. Of major interest concerning this chapter will be the PGM-dominated deposits. The flotation of copper–nickel-containing PGM was discussed in Volume I of this book.
18.3
Description of PGM-Dominated Deposits
18.3
21
DESCRIPTION OF PGM-DOMINATED DEPOSITS
According to the processing characteristics of PGM-dominated deposits, they can be divided into the following three groups: (a) Morensky type, (b) hydrothermal deposits and (c) placer deposits. Each type of deposit is briefly described below. 18.3.1
Morensky-type deposits
The Morensky-type deposits can be found in very large bodies of basaltic magma, which were intruded into stable continental rock. An example includes the Busheld Complex in South Africa and the Great Dyke of Zimbabwe. Mineralization similar to the above is also found in the Stillwater Complex in Montana, USA. The Busheld Complex consists of varieties of ore types, including high-chromium ores, ore with floatable gangue minerals and small but significant quantities of ultrafine slimes that are important from a processing point of view. The Stillwater Complex consists of a sequence of differential layers of mafic and ultramafic rocks, which extend for a strike length of up to 40 km and has a maximum exposed thickness of about 7.4 m [3]. There are several mineralization zones at the Stillwater Complex, including a PGM-rich zone and a low-grade zone. The Stillwater ore that is processed nowadays contains olivine, plagioclase, as well as plagioclase-brauzite, all of which are naturally hydrophobic gangue minerals. Another similar origin deposit is Lac des Illes in Canada. This complex is apparently contrary to a somewhat general rule in that of intrusion and is regarded as Archean age and may be therefore intruded prior to the Kenora origin into a technically unstable environment. 18.3.2
Hydrothermal deposits
An example of a hydrothermal deposit is the New Rambler deposit, described by McCal lum et al. [4] in the Medicine Bow Mountains in south-western Wyoming, USA, which contains a significant amount of PGM. The ore occurs in irregular pods that are hydrothermally decomposed into metadiorite and metagabbro zones. Pyroxenite and peridotite are reported to be intersected at a depth beneath the ore zone. All have been affected by supergene alteration. The main sulphides in the ore include pyrite, chalcopyrite, pyrrhotite, covellite and marcasite with associations of electrum, pentlandite and PGM. There is no evidence that the depth may be a result of an alteration in the original concentration of magmatic sulphides. It may be a result of concentration of hydrothermal solutions. 18.3.3
Placer deposits
The eluvial and alluvial PGM deposits have been processed in the Soviet Union, Canada, Columbia and the United States. Most of these deposits are associated with Alaskan-type ultrafamic rocks, which are, themselves, enriched in PGM, in particular, in the vicinity of
22
18.
Flotation of Platinum Group Metal Ores
concentration of chromite and with alpine ultrafamic bodies. As a process of weathering, there is a marked change in Pt/(Pt–Pd) ratio as compared to the source becoming greatly increased in the former due to the greater ease with which Pd dissolves and is removed in a weathering enrichment. Examples of this include the placer related to the Norilsk sulphide deposits and deposits found in Ural region USSR.
18.4 EFFECT OF MINERALOGY ON RECOVERY OF PLATINUM GROUP MINERALS The recovery of PGM minerals is a subject which has received very little attention in published literature. This is mainly due to the fact that major PGM producers are sur rounded by secrecy, therefore, neither commercial processes nor research work on recovery of PGM is publically available. Long-term research work conducted by a number of research organizations and data collected from a number of operating plants are summarized in this chapter. From a processing point of view, PGM-containing ores can be divided into three general groups as follows: 1. 2. 3.
ores amenable to gravity preconcentration, ores amenable to flotation and ores that can only be treated using a hydrometallurgical route.
18.4.1
Ores amenable to gravity preconcentration
The most important features of these ores are (a) the valuable constituents occur as minerals of high density, (b) they do not have middlings and (c) the grain-size distribution falls in a region where a gravity technique can be adopted successfully. Ore types where gravity preconcentration is used include Alaskan-type deposits, alluvial and fossil placer deposits. In the Alaskan-type deposits, the principal PGM minerals include Pt–Fe alloys, isoferro platinum (Pt2Fe) and platiniridium (Ir,Pt). There are several producing plants that process these ores, mainly in rural mountain areas (USSR). The alluvial deposits were treated in the early 20th century. The PGM in these deposits occur as alloys, usually as Pt rich in the form of loose grains and nuggets. These deposits have been mined in a number of countries, including Russia, Columbia and South Africa. Although there is a comprehensive review of the placer deposits [5], very little is known about PGM recovery using a gravity preconcentration method. Some of these deposits contain clay minerals, which require pretreatment before preconcentration. It should be mentioned that the PGM ores from Alaska contain magnetite, which is removed before gravity preconcentration. The fossil placer deposits are in fact gold-bearing conglomerates that carry small amounts of PGM, together with gold, uranium and other heavy minerals. However, studies conducted revealed that some of the fossil placer deposits contain about 22 PGM species, including Ir–Os–Ru alloys, sperrylite and isoferroplatinum.
18.5
Copper-Nickel and Nickel Sulphide Deposits with PGM as a By-Product
23
There are several operating mines that recover PGM and gold from fossil placer deposits, some of which include Witwatersrand and Geduld mines in South Africa. 18.4.2
Ores amenable to flotation
Classification of the ores amenable to flotation Based on flotation processing characteristics, these ores can be divided into the following major groups: (a) PGM sulphide-dominated deposits. In these deposits, PGM are in general associated with base metal sulphides, as grain boundaries between sulphides and silicates. In some cases, the PGM may be present in solid solution with sulphides. From these deposits, PGM are recovered in a bulk Cu/Ni/Co/PGM concentrate that is further processed using pyrometallurgical techniques. In many cases these ore types contain floatable non-opaque gangue minerals, including talc, chlorites, etc. (b) PGE-dominated deposits. This in fact is a term for stratiform deposits containing sparse sulphides and PGM concentration in a range between 5 and 30 g/t. These ores are typified by the Morensky Reef of the Bushveld Complex in South Africa. Mineralization of a similar type is found in the Stillwater Complex in Montana, USA. These deposits are characterized by a variety of different gangue minerals and high content of PGM sulphide minerals, such as cooperate (PtS), braggite [(PtPd)S] and vysotskite (PdS). Note that these minerals are rare and non-existent in most PGM-bearing copper-nickel sulphide deposits. Typical deposits that belong to this group include the Morensky Reef (South Africa), the Stillwater Complex (USA) and Lac des Illes (Canada).
18.5
COPPER-NICKEL AND NICKEL SULPHIDE DEPOSITS WITH PGM AS A BY-PRODUCT
Prior to discovery of the PGM Morensky Reef deposit, copper-nickel deposits in Ontario, Canada, and the Norilsk (USSR) were the principal sources of PGM production. However, about 40% of the world’s production of PGM comes from different Cu–Ni deposits. The major deposits from this group are discussed in the following sections. 18.5.1
The Sudbury area in Ontario, Canada
Mineralogical examination of these ores [8] revealed a variety of PGM and their associa tions. The michenerite (PdBiTe) and sperrylite (PtAs2) are the most common platinum/ palladium minerals for many deposits in the Sudbury region. Other minerals of economic value found in these deposits are moncheite (PtTe2), froodite (PdBi2), inszwaite (PtBi2), iravsite (IrAsS), niggliite (PtSn) and mertiate (PdSb3). Most of these minerals are liberated at a relatively coarse size (40–200 μm).
24
18.5.2
18.
Flotation of Platinum Group Metal Ores
The Norilsk Talnakh ore in Russia
In this area, the PGM are distributed in (a) disseminated sulphides, mostly in pyrrhotite, chalcopyrite and pentlandite. The predominant platinum minerals are Pt–Fe alloys, coop erate (PtS) and sperrilite (PtAs2); (b) massive sulphide ores where the predominant PGM are Pt–Fe alloys, rustenburgite (Pt3Sn) and sperrilite (PtAs2), occurring in fine inclusions in chalcopyrite and pyrrhotite; and finally (c) disseminated veins and brecia ores that may consist of mainly chalcopyrite or pyrrhotite. The PGM in these ores is present as Pt-(cooperate) and Pd-(rysotkite) sulphides. 18.5.3
Pechenga Cala Peninsula (USSR)
The ores from this region are of tholeiitic intrusions hosting Cu–Ni sulphides with relatively low PGM content. In these ores, most of the palladium is associated with pentlandite, where the platinum and rhodium are mainly associated with pyrrhotite. Only sperrilite and Pt–Fe alloys have, so far, been found in these ores. 18.5.4
Other deposits
Other deposits of significant value as PGM carriers include the Kambalda district in Western Australia, the Pipe Mine in Thomson Manitoba, Canada, and the Hitura deposit in Finland. The mineralization of PGM in these ores is similar to that of the Sudbury region ores.
18.6
CHROMIUM DEPOSITS WITH PGM
There are a number of deposits of this type with different origins. The geological environ ments are well described in the literature [8,10]. Most economical PGM chromite deposits are described as follows: (a) Podiform chromite deposits occur in ultrafamic bodies referred to as alpine types and are located in Tibet and North-western China. (b) Stratiform chromite deposits occur in different layered intrusions, such was Bushveld (South Africa) and the Great Dyke (Zimbabwe). The best known chromite deposit, with a number of operating plants, is the UG2 Complex located below the Morensky Reef. It ranges in thickness from 15 to 255 cm and dips at an angle of 5–70º towards the centre of the Bushveld Complex. Mineralogically, it consists mainly of chromite (60–90%) or thopyroxene (5–25%) and plagioclase (5–15%) with only trace amounts of base metal sulphides. PGM are usually closely associated with sulphides, such as laurite (RuS2), cooperate (PtS), braggite [(PtPd)S], Pt–Fe alloys, sperrilite (PtAs2) and vysotskite (PdS). The average chemical analyses of the PGM from various areas are shown in Table 18.2.
18.7
Flotation of PGM-Containing Ores
25 Table 18.2
Average chemical analyses of PGM from various areas of the UG2 deposits Area
Group
Marikoma Brits Hoekfontein South-western region Bushveld complex Moandagchoek North-eastern region Bushveld complex
18.7 18.7.1
A1 A2 A3 A4 A5 – A B C D E F
Assays (g/t) Pt
Pd
Rh
Ru
Ir
Au
Total
1.58 3.09 2.91 2.85 2.55 2.61 2.67 3.04 4.33 5.25 3.14 4.31
1.29 0.77 0.99 1.34 0.23 1.87 1.53 2.50 3.92 3.53 3.09 2.43
0.49 0.51 0.28 0.49 0.40 0.49 0.51 0.56 0.95 0.73 0.81 0.91
0.72 0.90 1.17 1.06 0.86 0.99 0.93 1.00 1.22 1.40 0.97 1.51
<1.0 <0.5 <1.0 <0.5 <0.5 0.05 <0.5 <0.5 0.16 <0.1 0.45 0.09
<0.2 <0.2 0.06 0.03 <0.1 0.17 0.03 0.07 0.07 <0.1 0.09 0.02
4.08 5.27 5.41 5.77 4.04 6.18 5.68 7.17 10.65 10.91 8.55 9.30
FLOTATION OF PGM-CONTAINING ORES
Introduction
There is little published data on the flotation of PGM-containing ores. Development work on beneficiation of PGM ores has been conducted by mining companies themselves and by a few research organizations close to the mining companies, which produce PGM. Many operating plants treating PGM ores use conventional flotation techniques and the metallurgical results are below optimum in a number of these plants. Each ore type described in Section 18.6.2 require different flowsheets and reagent schemes, which is dictated by the mineral composition of the ore and the geological setting, as well as the type of PGM carrier minerals. During the past 10 years of research work, a new technology has been developed to cope with difficult-to-treat ores, such as chromium-containing PGM ore and PGE-dominated ores. The following sections discuss the flotation properties and practices of the different ore types. 18.7.2
Flotation properties of PGM from sulphide-dominated deposits
Most of the current commercial operations that treat PGM from sulphide-dominated deposits are located in South Africa (Morensky Reef), Stillwater mines (Montana, USA) and Lac des Illes (Ontario, Canada). From a processing point of view, most of these ore types contain hydrophobic gangue minerals, including talc, which has a negative effect on PGM recoveries. Other major factor that affects flotation recovery of PGM is the presence of a variety of sulphide minerals, including pyrrhotite, pentlandite, chalcopyrite, violarite and pyrite, where
26
18.
Flotation of Platinum Group Metal Ores
the PGM are associated with all sulphides. In addition, in some operating plants, a portion of the PGM is represented by braggite, vysotkite, monchelite and Pt–Fe alloys. In general, the flotation properties of PGM from sulphide-dominated deposits are very dependent on the ratio of the individual sulphide minerals present in the ore and the nature and occurrence of hydrophobic gangue minerals present in the ore. Each of the sulphide minerals, which are PGM carriers (i.e. pyrrhotite, pyrite, pentlan dite, etc.) have different flotation properties under some flotation conditions. The selectiv ity between sulphide minerals and gangue minerals is relatively poor in principle, and in the majority of cases, a hydrophobic gangue depressant has to be used. The flotation behaviour of the individual sulphide minerals contained in PGM sulphide dominated ores can be described as follows: Pyrrhotite is a relatively slow floating mineral, especially monoclinic pyrrhotite, which is usually present in these ore types. The floatability of pyrrhotite also decreases when using certain hydrophobic mineral depressants, such as guars and dextrins. The flotation of pyrrhotite may improve with small additions of copper sulphate (CuSO4). Chalcopyrite and pentlandite float well using a xanthate collector and in certain opera tions, the recovery can reach greater than 90%. Violarite is the least floatable mineral of all the sulphides and represents a major loss of PGM in the flotation tailing from a number of operations. Figure 18.1 shows the rate of flotation of different sulphides from operation A (UG2 Complex). In these experiments, xanthate was used as the primary collector with dithiopho sphate as the secondary collector. 100 Chalcopyrite 80
Pentlandite
Recovery (%)
Pyrrhotite 60 Valleriite 40
20
0 0
3
6
9
12
15
Flotation time (minutes)
Figure 18.1 feed ore.
Rate of flotation of different sulphides from the Morensky Reef operation on a mill
18.7
Flotation of PGM-Containing Ores
27
One of the major problems associated with beneficiation of PGM from sulphide-dominated deposits is the presence of hydrophobic gangues, such as talc, chlorites, carbonates and aluminosilicates. The concentrates produced in most of the Morensky Reef operations (South Africa) varies from 80 to 150 g/t of combined PGM, where most of the contaminants are silicates, aluminosilicates and talc (i.e. up to 60%). The major hydrophobic gangue depressants used are carboxymethyl cellulose (CMC) and different modifications of guar gums. In recent years, a new line of hydrophobic gangue depressants were developed, based on a mixture of guar gums and low-molecular-weight polyacrylates modified with organic acid, which are extremely effective. With the use of these depressants, the grade of the PGM concentrate could increase from 100 up to 40 g/t without any loss in recovery. 18.7.3
Reagent practice in flotation of PGM sulphide-dominated ores
There is very little published information available on flotation of PGM ores in general, especially for the operating plants in the Morensky Reef and the UG2 operations. Most operations treating PGM sulphide-dominated ores have similar reagent schemes, with maybe a different choice of hydrophobic gangue depressants. Most of these plants use CuSO4 as the principal sulphide activator. In the past 10 years, extensive research was carried out by a number of research organizations with the objective of developing new technology for the beneficiation of these ore types. The main research work was directed towards finding better gangue depressants. Reagent schemes – Collectors and activators The principal sulphide activator used in most operating plants is small additions of CuSO4, normally added to the secondary rind and scavenger flotation stages. Although CuSO4 improves PGM recovery, it may also reduce the concentrate grade because an excess of CuSO4 will activate the gangue minerals. Figure 18.2 shows the effect of level of CuSO4 on the PGM grade–recovery relationship from the Morensky Reef Plant A ore. In these experiments, carboxymethyl cellulose (CMC) was used as the main gangue depressant. In recent years, a number of alternative activators were examined. It was found that organic acids along with a mixture of organic acid and thiourea can replace CuSO4 with significant improvement in PGM recovery and selectivity. The results obtained using different activators on the Morensky Operation B ore are compared in Table 18.3. The highest concentrate grade and PGM recoveries were achieved using a mixture of oxalic acid and thiourea. The use of CuSO4 as an activator was examined in relation to the point of addition and type of depressant used [11]. It was concluded that the point of reagent addition played an important role in PGM recovery. The primary collector used in PGM flotation is xanthate. As a choice of secondary collectors, dithiophosphates and mercaptans are used in some operating plants. The type of xanthate has a significant effect on PGM recoveries. Studies conducted on the Stillwater Complex by the US Bureau of Mines [12] indicated that the type of xanthate had a significant effect on PGM recovery (Table 18.4).
28
18.
Flotation of Platinum Group Metal Ores
100 200 g/t CuSO4
90
100 g/t CuSO4
PGM recovery (%)
80
0 g/t CuSO4
70 60 50 40 30 20 0
20
40
60
80
100
120
Total PGM grade (g/t)
Figure 18.2
Effect of level of CuSO4 on the PGM grade–recovery relationship.
Table 18.3 Effect of different activators on PGM flotation and upgrading Activator
Product
CuSO4 = 220 g/t PGM cleaner concentrate PGM rougher concentrate PGM rougher tail Feed (calc) PGM cleaner concentrate Oxalic acid/ DETA (80:20 ratio) PGM rougher concentrate = 350 g/t PGM rougher tail Feed (calc) Oxalic acid/ PGM cleaner concentrate thiourea (60:40 ratio) PGM rougher concentrate = 350 g/t PGM rougher tail Feed (calc)
Weight (%) Assays (g/t)
% Distribution
Pt
Pd
Au
Pt
Pd
Au
1.67 6.90 93.10 100.00 1.10
120 35.5 0.45 2.87 198
61.8 8.26 70.0 67.0 60.0 18.2 2.20 85.3 81.4 66.3 0.31 0.08 14.7 18.6 33.7 1.54 0.23 100.0 100.0 100.0 101 13.2 74.2 70.5 60.6
5.70 94.3 100.00 0.87
44.4 0.35 2.87 250
23.8 2.77 88.5 88.4 67.8 0.19 0.08 11.5 11.6 32.2 1.54 0.23 100.0 100.0 100.0 132 19.9 75.0 74.0 63.0
4.02 95.98 100.00
66.6 35.9 0.23 0.11 2.90 1.55
3.78 92.3 93.3 69.1 0.07 7.7 6.7 30.9 0.22 100.0 100.0 100.0
The highest PGM recovery was achieved using sodium amyl and sodium isobutyl xanthate. Using a mercaptan collector alone gave poor PGM recovery. However, when using xanthate with mercaptan, substantial improvement in PGM recoveries was achieved.
18.7
Effect of type of xanthate on PGM recovery from the Stillwater ore (USA) Concentrate
Tailing
Assays (g/t)
K-amyl xanthate Na-amyl xanthate Na-isobutyl xanthate Mercaptan Na-isobutyl xanthate + mercaptan
% Distribution
Assays (g/t)
% Distribution
Pt
Pd
Pt
Pd
Pt
Pd
Pt
Pd
34.1 34.2 31.0 55.8 31.0
89.9 80.6 77.5 114.7 83.7
64 83 81 53 90
54 63 65 35 80
1.24 0.62 0.61 1.55 –
4.96 3.72 3.70 6.51 –
36 17 19 47 10
46 37 35 65 20
Flotation of PGM-Containing Ores
Table 18.4
29
30
18.
Flotation of Platinum Group Metal Ores
Table 18.5 Effect of collectors from the PM series on PGM recovery from the Morensky operation A ore PGM cleaner concentrate
Na-isobutyl xanthate + R3477a Na-isobutyl xanthate + PM301 Na-isobutyl xanthate + PM305 Na-isobutyl xanthate + PM306 Na-isobutyl xanthate + PM308 a
PGM rougher concentrate
Assays (g/t)
% Distribution
Assays (g/t)
% Distribution
Pt
Pt
Pd
Pt
Pd
Pt
Pd
Pd
110
60.5
71
65
36.2
17.8
84.3
82.2
160
98.5
82.3
80.6
65.2
36.1
94.4
94
180
100.3
76.6
74
45.3
24.1
87.4
86.8
244
128
73.3
71.8
67.2
37.7
86.6
84.3
73.1
70.0
37.2
19.6
85.5
84.0
120.5
62.3
Cytec dithiophosphate.
In recent studies, a new line of PGM collectors had been developed [13] known as the PM series. These collectors are an ester-modified mixture of xanthate + mercaptan. The reaction product forms an oily greenish-coloured liquid. The results obtained using the PM series of collectors are shown in Table 18.5. High PGM recovery was obtained using a combination of sodium amyl xanthate plus collector PM301. Collector PM306 was the most selective collector from the PM300 series.
Choice of hydrophobic gangue depressants Choosing a depressant for hydrophobic gangue depression is dependent on the type of gangue present in the ore. During treatment of ores that contain talc, carboxymethyl cellulose (CMC) is normally used as the gangue depressant, or in some operations, guar gum + CMC. Typical examples of talc-containing ores are the Stillwater Complex (USA) and Lac des Illes (Canada). Both operations use CMC for talc depression. In the Stillwater operation, the additions of CMC are relatively high (i.e. up to 600 g/t) and are added to the ball mill, the PGM roughers and cleaners. Laboratory and pilot plant studies [14] on the Stillwater ore showed that the molecular weight of the CMC affected both PGM grade and recovery. Figure 18.3 shows the effect of molecular weight of CMC on PGM grade–recovery relationship. The best results were obtained using CMC with an average 300,000 molecular weight, corresponding to a viscosity of over 3000 cps. Studies conducted by the University of Cape Town (South Africa) researchers indicated that the point of CMC addition [15] had a significant effect on sulphides (PGM carriers) grade and recovery.
18.7
Flotation of PGM-Containing Ores
31
100
CMC MW = 300,000
PGM recovery (%)
80
60
40 CMC MW = 200,000
CMC MW = 150,000 20
0 0
40
80
120
160
200
240
PGM concentrate grade (g/t)
Figure 18.3
Effect of CMC molecular weight on PGM grade–recovery relationship.
It should be noted that in several operating plants from the Morensky Reef and Stillwater Complex, from which plant metallurgical results are available, the total PGM recoveries ranges from 82% to 85% PGM. The grade of concentrate from the Morensky operations ranges from 80 to about 120 g/t (Plants A and B). Most of the contaminants are silicates and talc. 18.7.4 Reagent practice in flotation of Cu–Ni and Ni ores with PGM as the by-product The flotation of Cu–Ni and Ni ores is discussed in Chapter 16 (Volume 1). In most operating plants, the emphasis is usually placed on Cu–Ni and Ni recovery and concentrate grade, and most of the research on these ores was directed towards improvement in Cu–Ni recovery and pentlandite–pyrrhotite separation, whereas little or no attention was paid to improvement in recovery of PGM. In operations from the Sudbury Region (Canada), PGM are recovered as by-products of Cu–Ni concentrates. The idealized flowsheet of the Inco Metal PGM recovery flowsheet is shown in Figure 18.4. Laboratory studies conducted on Falconbridge ores, also from the Sudbury Region, during 1980 [16] showed that PGM recovery can be improved with the use of a secondary collector. Figure 18.5 shows the effect of level of secondary collector on PGM recovery in a Cu–Ni bulk concentrate. The highest PGM recoveries were achieved using isobutyl dithiophosphate (Minerec 2087) as the secondary collector. Plant data from the Copper Cliff Mine showed that about 85% of the platinum was recovered in a Cu–Ni concentrate, most of which was from the nickel concentrate. The
32
18.
South range offset mines
Flotation of Platinum Group Metal Ores
South B north range mine
Frood-stobie primary mill
Tails PGM
Clarabelle primary mill
Bulk concentrate
Tails PGM
Rough pyrrhotite concentrate
Copper cliff secondary mill
Ni concentrate
Cu concentrate
Ni reverberatory furnaces Ni converters
Cu Flash furnace
Ni Matte slow cooling
Copper cliff Cu2S + Au/Ag
Ni Matte separation
Ni metal + PGM
Smelter complex
Cu converters
Ni3S2 product
Electrorefining
Cu metal
Further treatment
Anode Slime treatment
Au, Ag Te, Se
Pressure carbonylation
Ni Metal
Residue + PGM Leaching
Figure 18.4 of PGM.
PGM residue
Idealized flowsheet used at the Inco metals operation (Sudbury, Canada) for recovery
plant metallurgical results are shown in Table 18.6. Similar plant results were obtained at other Inco operations. In the Norilsk Region, research work [17] was carried out on Oktyabrski disseminated Cu/ Ni–PGM ore. This ore contains high-grade PGM, most of which is represented by palladium. The results using different collectors are shown in Table 18.7.
18.7
Flotation of PGM-Containing Ores
33
PGM recovery in teh Cu/Ni bulk concentrate (%)
100
xanthate + dithiophosphate (Minerec 2087)
80
xanthate + mercaptan 60
xanthate only 40
20
0 0
3
6
9
12
15
Flotation time (minutes)
Figure 18.5
Effect of secondary collectors on PGM recovery in a bulk Cu–Ni concentrate.
Table 18.6 Platinum recovery in the Copper Cliff plant Product
Weight (%)
Copper concentrate Nickel concentrate Tails Feed
13.0 29.0 58.0 100.00
Assays (%, g/t)
% Distribution
Cu
Ni
Pt
Cu
Ni
Pt
29.2 2.28 0.93 4.58
0.91 12.8 0.22 4.42
1.80 3.04 0.41 1.39
83.0 14.0 3.0 100.0
3.0 85.0 12.0 100.0
17.0 65.0 18.0 100.0
Improvement in overall PGM recoveries was obtained using xanthate as the primary collector and dithiophosphate as the secondary collector. A slight improvement in metal lurgical results was achieved when using mercaptan as the secondary collector. 18.7.5
Reagent practice in flotation of PGM from chromium-containing ores
The major problem associated with processing of high-chromium ores includes the following:
34
Table 18.7 Effect of secondary collectors on PGM from the Norilsk (Russia) disseminated Cu/Ni-PGM ore Collector
180 g/t Xanthate
30 g/t Mercaptan
10.25 5.58 15.83 84.17 100.00 10.60 6.45 17.05 82.95 100.00 11.49 6.29 17.78 82.22 100.00
Assays (%)
% Distribution
Cu
Ni
Pt
Pd
Cu
Ni
Pt
Pd
29.6 2.0 19.88 0.18 3.3 30.3 1.32 19.34 0.13 3.4 27.5 1.16 18.19 0.31 3.5
0.8 12.8 5.03 0.12 0.9 0.7 11.41 4.81 0.12 0.92 1.2 9.83 4.26 0.18 0.91
6.5 55.0 26.95 0.63 4.8 5.8 58.98 25.92 0.40 4.75 6.1 52.4 22.47 0.86 4.7
55.0 188 101.9 1.62 17.5 49.5 180.9 98.97 0.27 17.1 52.0 165.5 93.05 0.92 17.3
92.0 3.4 95.4 4.6 100.0 94.5 2.5 97.0 3.0 100.0 90.3 2.1 92.4 7.6 100.0
9.1 79.4 88.5 11.5 100.0 9.2 80.0 89.2 10.8 100.0 15.1 68.0 83.1 16.9 100.0
13.9 75.0 88.9 11.1 100.0 12.9 80.1 93.0 7.0 100.0 14.9 70.1 85.0 15.0 100.0
32.2 60.0 92.2 7.8 100.0 30.7 68.0 98.7 1.3 100.0 35.4 60.2 95.6 4.4 100.0
Flotation of Platinum Group Metal Ores
30 g/t Dithiophosphate 150 g/t Xanthate
CuCl concentrate Ni/PGM concentrate Bulk concentrate Bulk flot tail Head (calc) CuCl concentrate Ni/PGM concentrate Bulk concentrate Bulk flot tail Head (calc) CuCl concentrate Ni/PGM concentrate Bulk concentrate Bulk flot tail Head (calc)
Weight (%)
18.
30 g/t Xanthate
Product
18.7
Flotation of PGM-Containing Ores
35 Table 18.8
Chemical analyses of UG2 high-chromium ore Element
Assays
Platinum (Pt) Palladium (Pd) Nickel (Ni) Sulphur (S) Copper (Cu) Chromium (Cr) Iron (Fe) Gold (Au)
2.06 g/t 1.29 g/t 0.10% 0.04% 0.011% 20.0% 18.5% <0.02 g/t
• High chromium content in PGM concentrates has a negative effect on pyro- and hydrometallurgical processing. • The major carriers of PGM are a variety of minerals and alloys, where the flotation properties of the PGM minerals and alloys are not well defined. These ores have very little to no sulphides present that are PGM carriers. In recent years, extensive research [18] has been conducted on these ore types with the objective of finding a more effective PGM collector and chromium depressant. Research work was conducted on UG2 high-chromium ore. Detailed chemical analyses of the highchromium ore used in this research are presented in Table 18.8. The PGM carriers in this ore include a variety of PGM minerals (sperrilite) and its alloys. The main problems identified associated with processing this ore type were (a) poor concentrate grade, (b) low rate of PGM flotation, (c) excessive chromium reporting to the PGM concentrate and (d) high collector consumption. It was established that the reason for high collector consumptions was the presence of small, but significant, quantities of clay-like slimes. The high collector consumption was the principal reason for the excessive amount of chromium reporting to the PGM concen trate (mainly as fines). Types of secondary collectors were extensively examined in research work. Figure 18.6 shows the effect of secondary collectors on the PGM grade–recovery relationship. The highest PGM recovery was achieved using collector PM443, which is an amine + ester-modified xanthate. Among the chromium slime depressants evaluated, modified mixtures of organic acids, RQ depressants and a low-molecular-weight polyacrylic acid + pyrophosphate mixture were there. The effect of different chromium depressants on chromium assays of the PGM concentrate are illustrated in Figure 18.7. Significant improvement in chromium depression has been achieved using depressants from the KM series, representing mixtures of organic acid and low-molecular-weight acrylic acid mixtures. It is, therefore, possible to depress chromium during PGM flotation and at the same time reduce collector consumption. The relationship between the level of collector and level of KM3 depressant is shown in Table 18.9.
36
18.
Flotation of Platinum Group Metal Ores
100 95 90
PGM recovery (%)
85 80
xanthate + PM 443
75 70 65
xanthate + PM 280
60
xanthate only 55 50 0
40
80
120
160
200
240
280
PGM concentrate grade (g/t)
Figure 18.6
Effect of different secondary collectors on PGM grade–recovery relationship.
4.0
Cr2O3 in the PGM concentrate recovery (%)
none 3.5 300 g/t RQ1
3.0 2.5
300 g/t KM1
2.0 1.5 1.0
300 g/t KM2
0.5
Depressant Dosage
0.0 Rougher
Cleaner 1
Cleaner 2
Cleaner 3
Cleaner 4
Flotation stage
Figure 18.7 Effect of different depressants on chromium assays of the PGM concentrate.
18.7
Flotation of PGM-Containing Ores
37 Table 18.9
Effect of depressant KM3 on collector consumption during PGM flotation from UG2 high-chromium ore Reagent (g/t)
PGM cleaner concentrate
PGM rougher concentrate
Assays (%, g/t)
% Distribution
Assays (%, g/t)
% Distribution
Collector
Depressant
Pt
Pd
Cr
Pt
Pd
Cr
Pt
Pd
Cr
Pt
Pd
Cr
330 330 200 200 160
0 200 200 200 400
80.1 90 110 120 135.1
54.3 60.9 71 77.6 86.2
3.8 2.5 2.2 2.0 0.9
57 60 70 75 77
61 64 71 76 77
0.4 0.2 0.2 0.15 0.02
21.0 28.2 33.2 38.3 40.2
13.0 17.4 20.5 23.9 25.2
7.3 6.0 4.2 3.8 2.5
82 84 86 88 90
80 82 84 87 89
4.1 2.6 1.5 1.3 0.14
The data shown in this table demonstrate that overall collector consumption can be reduced by 50% with the use of slime/chromium depressant, KM3. At the same time, the chromium assays in the PGM concentrate reduced from 3.8% to 0.9% Cr. It is obvious that high collector consumption is responsible for high chromium content in the cleaner concentrate. Comparative continuous locked cycle tests were conducted using the reagent scheme currently used in an operating plant and the new reagent scheme developed during the research on ore from the Waterval plant (South Africa). These results are compared in Table 18.10. A substantial improvement in metallurgical results was achieved using the new reagent scheme. This new reagent scheme included collector PM443 and depressant KM3. The collector type plays a significant role in PGM recovery from high-chromium ores. Collectors were examined in detail [19] on several high-chromium ores, where new collectors from the PM series were included in the evaluation. These collectors are Table 18.10 Comparison of results using the new and standard plant reagent scheme from Waterval Plant (South Africa) Reagent scheme
Product
Weight (%) Assays (%, g/t) Pt
Pd
% Distribution Cr
Pt
Pd
2.08 Newly developed scheme PGM Cl concentrate PGM comb tail 97.92 100.00 Feed (calc)
89.54 55.54 1.02 89.0 86.1 0.24 0.19 – 11.0 13.9 2.10 1.34 – 100.0 100.0
Standard pant scheme
86.01 49.08 2.72 79.8 76.7 0.45 0.31 – 20.2 23.3 2.16 1.29 – 100.0 100.0
PGM Cl concentrate 2.01 97.99 PGM comb tail 100.00 Feed (calc)
38
18.
Flotation of Platinum Group Metal Ores
Table 18.11 Effect of type of collector on PGM rougher–scavenger flotation from high-chromium ores Collector type
PAXa PAXa, R3477b PAXa, R404b PAXa, PM301 PAXa, PM305 SIBXa, PM303 a b
PGE rougher concentrate
PGE rougher + Scavenger concentrate
Assays (g/t)
% Distribution
Assays (g/t)
% Distribution
Pt
Pd
Pt
Pd
Pt
Pd
Pt
Pd
110.7 120.4 110.1 116.6 113.8 122.4
96.8 98.5 97.0 94.5 96.3 97.9
55.1 66.3 64.3 70.2 80.2 82.2
54.3 64.2 62.1 70.0 80.0 81.0
45.5 44.3 46.3 42.3 43.3 44.6
40.4 39.8 41.1 38.0 39.6 40.1
81.2 84.8 85.2 88.5 92.5 92.3
80.3 83.5 83.6 86.2 91.1 92.1
Xanthates. Dithiophosphates.
ester-modified mixtures of xanthate and dithiophosphates. The results are presented in Table 18.11. The highest PGM recovery was achieved using a combination of isobutyl xanthate and collector PM303. 18.7.6
Flotation of oxide PGM ores
There are only a few known oxidized PGM deposits in which the ore is in the development stage. These deposits can be found in Brazil and Australia. The PGM in these ores is usually represented by different PGM minerals and alloys, finely disseminated in a gangue matrix. Using a flotation method with conventional reagent schemes, results in low PGM recoveries, ranging from 65% to 70% PGM. Recent studies conducted on an ore from Brazil [20] indicated that a mixture of organic acid and thiourea has a positive effect on PGM recovery from oxidized ores. Figure 18.8 shows the effect of organic-acid-modified thiourea on PGM flotation from oxidized PGM ore. This data show that substantial improvement in PGM grade and recovery was achieved using organic-acid-modified thiourea.
18.8
PLANT PRACTICE IN TREATMENT OF PGM ORES
In contrast to other sulphide-treatment flowsheets and reagent schemes, which are rela tively simple, the flowsheet and reagent schemes for treatment of PGM ores can be highly complex, and varies from one ore type to the next. In general, the type of flowsheet used to treat PGM ores largely depends on the type of ore. For example, ores that are sulphide dominated have the simplest flowsheet but
18.8
Plant Practice in Treatment of PGM Ores
39
80 Depressant Dosage
70 PGM recovery (%)
450 g/t 60 300 g/t 50
40
150 g/t none
30 0
20
40
60
80
100
PGM concentrate grade (g/t)
Figure 18.8 Effect of level of oxalic acid/thiourea mixture on the PGM grade–recovery relationship (oxide PGM ore from Brazil).
relatively complex reagent scheme. Chromium-containing PGM ores have a complex flowsheet but relatively simple reagent scheme.
18.8.1
Flowsheets for treatment of sulphide-dominated PGM ores
A generalized flowsheet for treatment of sulphide-dominated PMG ores is presented in Figure 18.9. There can be some variation in this flowsheet, such as (a) retreatment of the cleaner tailings, (b) regrinding the scavenger concentrate and (c) the number of cleaning stages. This flowsheet is used in several operations from the Bushveld Complex (South Africa), Stillwater Complex (USA) and Lac des Illes (Canada).
18.8.2
Flowsheets for treatment of Cu–Ni-containing PGM ores
These flowsheets are usually designed for treatment of Cu–Ni ores with the PGM being recovered as a by-product. A typical flowsheet used to treat Cu–Ni–containing PGM ores is shown in Figure 18.10. The configuration of these flowsheets may vary considerably, depending on the amount and type of pyrrhotite present in the ore.
40
18.
Flotation of Platinum Group Metal Ores
Ore Grinding
Conditioning Regrind
Primary rougher
Secondary rougher
Scavenger
Final tailing
Primary cleaner Secondary cleaner
Recleaners
PGM concentrate
Figure 18.9
Generalized flowsheet for treatment of sulphide-dominated PGM ores.
In some cases, where the ore has a high PGM value contained in the pyrrhotite, an additional PGM recovery stage is required.
18.8.3
Flowsheet used for treatment of high-chromium PGM-containing ores
These flowsheets are specifically designed to maintain the chromium content in the PGM concentrate as low as possible, since the chromium is an unwanted impurity. The generalized flowsheet for treatment of high-chromium PGM-containing ores is shown in Figure 18.11. Usually, these flowsheets include a two-stage PGM flotation. In stage 1, a high-grade PGM concentrate is recovered after coarse grinding. The rougher tailing is reground followed by the stage 2 of PGM flotation and upgrading, where a low-grade concentrate is recovered.
18.9
Reagent Schemes Used to Treat PGM-Containing Ores
41
Feed Washing
Desliming
Slime
To Iron
circuit
Crushing
Grinding
Classification
Magnetic separation
Nickel rougher
Primary rougher
Scavenger
Nickel cleaner
Low-grade nickel concentrate
Magnetic tailing
First cleaner
Second cleaner
Scavenger cleaner 1
Scavenger cleaner 2 Cu/Ni concentrate
Figure 18.10
18.9
Regrinding Scalp flotation
Scalp cleaner
Flowsheet for treatment of semi-massive Cu–Ni PGM-containing ores.
REAGENT SCHEMES USED TO TREAT PGM-CONTAINING ORES
The reagent schemes used for treatment of PGM-containing ores varies considerably and depend largely on the type of ore being treated. In some operations, emphasis is placed on maximizing the PGM recovery, while a low-grade concentrate is maintained. Table 18.12 lists the ore type and reagent scheme, along with metallurgical results achieved in some PGM operations. In the majority of operations, collector consumption is relatively high, especially in plants treating high-chromium ores. It appears that PGM concentrates with high chromium contents are in fact related to a high collector consumption, which usually results from entrapment of fine chromium in the concentrate.
42
18.
Flotation of Platinum Group Metal Ores
Ore Primary grind
PGM flotation
Secondary grind
PGM rougher
PGM scavenger
PGM recleaner
PGM secondary cleaner 1
Regrind
PGM primary cleaner concentrate
PGM secondary cleaner 2
Primary cleaner
Final tailing
PGM secondary cleaner 3
PGM secondary cleaner concentrate
Figure 18.11 Flowsheet used to treat high-chromium PGM-containing ores. Table 18.12 Ore type, reagent scheme and metallurgical results from major operating plants Name of operation
Ore type/reagent scheme/metallurgical results
Amplats – Mine #1 South Africa, Morensky Reef
Ore: Sulphide-dominated PGM ore composed of Cu, Ni, pyrrhotite and some pyrite. This ore contains a fair amount of floatable gangue minerals Grind: To a K80 of about 105 μm Reagents: CuSO4 = 200–300 g/t, CMCa = 200–400 g/t, amyl xanthate = 100–250 g/t, dithiophosphate = 40–80 g/t Metallurgy: Total PGM concentrate grade = 70–85 g/t, PGM recovery = 82–85% Ore: Sulphide-dominated PGM ore containing nickel, pyrrhotite and a little copper. Floatable gangue was dominated by talc and chlorites
Amplats – Mine #2 South Africa, Morensky Reef
(Continued )
18.9
Reagent Schemes Used to Treat PGM-Containing Ores
Table 18.12
43
(Continued )
Name of operation
Amplats – Mine #3 South Africa, Morensky Reef
Stillwater complex Montana, USA
Norilsk complex Siberia, Russia
UG2 Morensky Reef Plant A
UG2 Morensky Reef Plant B
Barrier Reef Plant WF1
Ore type/reagent scheme/metallurgical results Grind: To a K80 of 87 μm Reagents: CuSO4 = 100–200 g/t, dibutyl xanthate = 320 g/t. Modified guar gum = 200–250 g/t Metallurgy: 90–100 g/t total PGM in concentrate, PGM recovery = 80–82% Ore: Sulphide-dominated PGM deposit containing Cu/Ni and mixed pyrite–pyrrhotite. The main floatable gangues are calcite, chlorites with lesser talc Grind: K80 = 95 μm Reagents: CuSO4 = 100–150 g/t, isopropyl xanthate = 150 g/t, guar = 150–200 g/t, mercaptan = 30–40 g/t Metallurgy: Total PGM in concentrate = 110–120 g/t, PGM recovery = 84–86% Ore: Sulphide-dominated PGM-containing Cu, Ni associated with PGM. Principal gangue floatable mineral is talc Grind: K80 = 115 μm Reagents: CMC = 400–600 g/t; sodium amyl xanthate = 80–150 g/t, dithiophosphate = 20–40 g/t Metallurgy: Total PGM in concentrate = 300–600 g/t, PGM recovery = 86–88% Ore: Massive sulphide Cu/Ni ore with high PGM content. Main gangue minerals are serpentine and pyrrhotite. The bulk of the PGM is contained in pentlandite and monoclinic pyrrhotite Grind: K80 = 74 μm Reagents: Lime = 200–300 g/t; CuSO4 = 0–300 g/t; xanthate. Mixture = 40–60 g/t; aeroflot = 20–30 g/t Metallurgy: Grade is variable, PGM recovery = 70–85% Ore: PGM dominated with some chromium. Main gangue. Minerals are calcite, silicate and some aluminosilicate. The ore contains a moderate amount of clay-like slimes Grind: K80 = 85–100 μm Regrind: Regrind the middlings Reagents: Potassium amyl xanthate = 300–400 g/t, dithiophosphate= 30–50 g/t, guar gum = 50–100 g/t Metallurgy: Total PGM in concentrate = 300–400 g/t, PGM recovery = 80–84% Ore: PGM-dominated ores – with very little sulphides and the main gangue minerals include silicate, mica, aluminosilicate and some chromium Grind: K80 = 95 μm Regrind: Cleaner tailings Reagents: Sodium isobutyl xanthate = 280–350 g/t; dithiophosphate = 50–60 g/t; CuSO4 = 50–100 g/t; guar gum = 50–100 g/t Metallurgy: Total PGM in concentrate = 180–200 g/t, PGM recovery = 80–82% Ore: High-chromium PGM ore. Main gangue minerals are chromite with some silicates, calcite and clay-like slimes Grind: K80 = 150 μm for stage 1; K80 = 90 μm for stage 2 (Continued )
44 Table 18.12
18.
(Continued )
Name of operation
Amplats Barrier Reef Plant WF2
a
Flotation of Platinum Group Metal Ores
Ore type/reagent scheme/metallurgical results Regrind: Cleaner tailings Reagents: Xanthate mixture = 250–300 g/t; mercaptan = 30–50 g/t; modified guar = 50–100 g/t; CuSO4 = 50–100 g/t Metallurgy: Total PGM in concentrate = 80–110 g/t at 2.75% Cr2O3, PGM recovery = 83% Ore: High-chromium PGM ore. Dominant gangue minerals are chromite with some non-opaque gangue. Ore contains moderate amount of clay-like slimes Grind: K80 = 150 μm for stage 1; K80 = 95 μm for stage 2 Reagents: Isobutyl xanthate = 200–300 g/t; dithiophosphate =20–35 g/t; modified guar = 50–100 g/t Metallurgy: Total PGM in concentrate = 75–90 g/t at t95% Cr2O3, PGM recovery = 83%
CMC, carboxymethyl cellulose.
In fact, high collector consumptions are related to the presence of clay-like slimes, which are known to consume collectors. Recent studies conducted on high-chromium ores [20] indicated that collector consumption can be substantially reduced (i.e. up to 60%) by using a suitable slime depressant/dispersant. REFERENCES 1. Cabri, L.I., The Platinum Group Minerals. In (Cabri, L.J. ed) The Platinum Group Elements: Mineralogy, Geology, Recovery. Johannesburg, South Africa, CIM Special Volume 23, Chapter 11, 1981. pp. 234–250. 2. Naldrett, A.J., and Duke J.M., Platinum Metals in Magnetic Sulphide Ores, Science, Vol. 208, pp. 1417–1424, 1980. 3. Jones, W.R., Peoples, J.W., and Howland, A.L., Igneous and Tectonic Structures of Stillwater Complex, Montana, US Geological Survey Bulletin, Vol. 10714, pp. 281–335, 1960. 4. McCallum, I.S., Loucks, R.R., Carlson, R.R., Cooley, E.F., and Doerge, T.A., Platinum Metals Associated with Hydrothermal Copper Ores of the New Rambler Mine, Medicine Bow Moun tain, Economic Geology, Vol. 71, pp. 1429–1459, 1976. 5. Mertie, J.B., Economic Geology of the Platinum Group Minerals, US Geological Survey, Professor Paper #630, 1969. 6. Feather, C.E., Mineralogy of Platinum Group Minerals in the Withwatersrand, South Africa, Economic Geology, Vol. 71, pp. 1399–1428, 1976. 7. Naldrett, A.J., Platinum Group Elements, In (Cabri, L.J. ed) Deposits of Platinum Group Elements; Mineralogy, Geology and Recovery. Johannesburg, South Africa, CIM Special Volume 23, Chapter 10, pp. 198–230, 1981. 8. Crocket, J.H., Platinum Group Elements in Mafic and Ultrafamic Rocks; A Survey, Canadian Minerals, Vol. 17, pp. 391–403, 1979. 9. Razin, I.V., Begizov, V.D., and Meshonkina, V.I., Data on Mineralogy of Platinum Metals in Talnakh Deposit, International Geological Review, Vol. 17, pp. 6–56, 1975. 10. Gerhard, V.G., The Mineral Resources of the Bushveld Complex, Institute of Geological Research of the Bushveld Complex, University of Pretoria, S.A., 2002
References
45
11. Wise, J., Harris, P., and Bradshaw, D., The Role of Reagent Suite on Optimizing Pentlandite Recoveries from the Morensky Reef, Minerals Engineering, Vol. 19, No. 12, pp. 1290–1300, 2006. 12. Morrice, E., Valkiewicz, J.W., and Casale, G., Pilot Plant Flotation of Serpentinized PlatinumPalladium Ore from Stillwater Complex, Report of Investigation 8885, 1976. 13. Bulatovic, S., Evaluation of Alternative Reagent Schemes for the Flotation of Platinum Group Minerals from Various Ores, Minerals Engineering, Vol. 17, No. 16, pp. 931–939, 2003. 14. Bulatovic, S., and Bigg, A.C.T., An Investigation of the Recovery of PGM from Stillwater Ore, Report of Investigation LR3946, 1978. 15. Ekmekci, Z., Bradshaw, D.J., Harris, P.J., and Buswel, A.M., Interactive Effect of Milling Media and CuSO4 Additions on the Flotation Performance of Sulphide Minerals from the Morensky ore, Part II Froth Stability, International Journal of Mineral Processing, Vol. T8, pp. 164–174, 2006. 16. Bulatovic, S., and Newman, D., The recovery of Copper/Nickel and PGM from Strathcona Plant, Report of Investigation LR3958, 1984. 17. Yasenko, A.A., Alekceva, L.I., Salaikin, Y.A., and Zakkharov, B.A., Improvement of the Technology of Concentrating Disseminated Platinum Containing Copper-Nickel Ores, Tsvetnie Metaly, No. 2, pp. 11–13, 1999. 18. Bulatovic, S., An Investigation of the Effectiveness of New PGM Collector on Amplats PGM Ore, Report of Investigation LR5670, 1995. 19. Bulatovic, S., and Jessup, T., An Investigation of Alternative Reagent Schemes in Treatment of High Chromium PGM Ores, Report of Investigation LR00099-274, 2006. 20. Bulatovic, S., and Jessup, T., Recovery of PGM from Oxidized PGM ore from Brazil, Report of Investigation LR00099-312, 2006.
– 19 –
Flotation of Oxide Copper and Copper Cobalt
Ores
19.1
INTRODUCTION
Flotation practice of oxide copper minerals dates back to almost 60 years ago, and has been applied in Central Africa (Congo) by Union Miniere (Belgium). The process involves two basic flotation methods: (a) fatty acid flotation of oxide copper minerals from siliceous ore, and (b) sulphidization of oxide copper minerals followed by flotation using sulfhydryl collectors, such as xanthate [1] from carbonate ores. In the past 50 years, extensive research has been carried out on a variety of oxide copper minerals, and only a few of the many innovative processes have been introduced into operating plants. It was not until recently that new technology has been developed and introduced into some operating plants around the world. One of the major problems with flotation of oxide copper minerals, at industrial scale, is that the floatability of oxide copper minerals from natural ores depends largely on the mineralogy of the ore and the gangue composition. The floatability of oxide copper minerals that are present in the ore containing carbonaceous and dolomitic gangue is significantly different from the flotation properties of oxide copper containing siliceous gangue minerals. The presence of various types of clay slimes also has a significant effect on flotation properties of oxide copper minerals [2].
19.2
OXIDE COPPER ORES AND MINERALS
More than 120 oxide-containing minerals have been identified, mainly from the Central and South African regions, but only a few of these minerals have any economic value. Some of the most important copper oxide minerals are listed in Table 19.1. In most cases, oxide copper ores contain more than one copper oxide mineral, and also contain mixtures of sulphide and oxide copper minerals. From a processing point of view, the oxide copper ores can be divided into the following five groups: Oxide copper ores. In oxide ores copper is predominantly malachite with significant quantities of cobalt oxides. According to the mineral composition, these ores can be
47
48
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Table 19.1 List of economically valuable copper oxide minerals Mineral
Chemical formula
Cu content (% Cu)
Specific gravity (SG)
Colour
Cuprite Tenorite Malachite Azurite Brochantite Atacamite Antlerite Chrysocolla Chaecantite
Cu2O CuO Cu2(OH)CO3 Cu3(OH)2(CO3)2 Cu4(OH)6SO4 Cu2(OH)2Cl Cu3(OH)2SO4 CuO · SiO2 CuSO4 · 5H2O
88.8 80.0 57.4 55.3 56.6 44.6 54.0 10–36 25.5
5.9 6.5 3.9 3.7 3.9 3.8 3.9 2–2.4 2.2
Brick red Black Green Blue Emerald green Green, blue Emerald green Blue Deep blue
sub-divided into two main groups: (a) oxide ore that contains carbonaceous gangue minerals (carbonate, dolomite) with little or no silica; and (b) oxide ore, where silica is the predominant gangue mineral. The gangue composition of the ore plays a decisive role in selection of reagent scheme for beneficiation of the ore. These ores also contain cobalt minerals, mainly carrollite (CoCuSO4) and cobaltite (CoAsS). Copper oxide mixed ore – Type 1. The main copper minerals found in these ores include malachite, pseudo-malachite, chrysocolla and some tenorite. These ores also may contain mainly siliceous gangue minerals, including spherocobaltite as the main cobalt minerals. The carbonaceous types also contain an appreciable amount of clay slime minerals. Copper oxide mixed ore – Type 2. In contrast to Type 1, this ore type contains cuprite, malachite and azurite as the main copper oxide minerals. This ore type predominantly contains carbonaceous gangue, and usually, significant amounts of clay-like slimes. Mixed copper sulphide oxide ores. These contain varieties of both sulphide and oxide minerals, and are the most complex copper-bearing ores from a beneficiation point of view. The major copper minerals present in this ore type include bornite, chalcocite, covellite, malachite, cuprite and chrysocolla. In some cases, significant amounts of cobalt minerals are also present in this ore. Copper oxide gold ores. Although this ore type is not abundant, they are of significant value because they contain gold. Only a few deposits in Brazil and Australia are known. The copper in these ores is represented by cuprite, native copper, antlerite and tenorite. The gold is associated with cuprite, as an auricupride and several sulphosalts. The major problem associated with treatment of this ore type is the presence of large amounts of clay slimes in the form of iron hydroxide and illite. Most of the oxide copper deposits are located in the former Republic of Zaire (Katanga) and Zambia. Only a few deposits are located in Chile, Peru, Canada and the United States. From most of the south and North American deposits, oxide copper is recovered using a hydrometallurgical method.
19.3
19.3
Flotation Properties of the Individual Copper Minerals and Mixtures
49
FLOTATION PROPERTIES OF THE INDIVIDUAL COPPER MINERALS AND MIXTURES
The flotation characteristics of the oxide copper minerals from natural ore are dependent on several main factors, some of which include the following: • Chemical composition and physical structure of the oxide copper minerals and the ionic composition of the slurry phase play important roles in the floatability of various oxide minerals. The oxide copper minerals are often porous, and in some cases, water soluble. Some of the oxide minerals tend to slime during grinding, and flotation of fine oxide minerals is rather difficult. • The gangue constituents and their nature are sometimes determining factors in selection of a treatment process for beneficiation of oxide copper ores. Highly weathered ores usually contain a fairly large amount of slimes, which has a negative effect on the floatability of oxide copper minerals. Also, there is an appreciable difference in floatability between oxide minerals from carbonaceous and siliceous ores. • The mechanical strength of the surface layers of many of the oxide copper minerals is weak. Therefore, flotation of oxide copper ores using sulphidization method, can improve by reducing turbulence and attrition within the flotation cell [3]. Floatability of malachite is one of the most important oxide copper minerals for production of copper from oxide ores using flotation. Extensive research has been carried out by a number of researchers [4–7] in which various flotation methods were examined. Hydroxamaic acid flotation has been established from laboratory research work, which has the chemical formula as shown in Figure 19.1.
R1
Figure 19.1
N
C
R2O
OH
Hydroxamaic acid formula.
where R1 is organic ligand (alkyl benzyl, etc.) and R2 may be organic or inorganic, and is a suitable malachite collector. It was found that the effectiveness of hydroxamaic acid was dependent on flotation pH and collector concentration. Figure 19.2 shows the relationship between malachite recovery and flotation pH. It has been proposed that chelation mechan ism involves CuOH+, where hydroxamate has a high chemsorption specifically for copper. Although good metallurgical results have been obtained in the laboratory, it has not found any plant application to date. The sulphidization process, which was first successfully applied on a commercial scale on lead carbonate ores, is currently the most popular method used during treatment of oxide copper ores that contain malachite and carbonaceous gangue. The commonly used sulphi dizers are Na2S · 9H2O and NaHS, with xanthate or xanthate ester.
50
19.
Flotation of Oxide Copper and Copper Cobalt Ores
100
Copper recovery (%)
80
60
40
20
0 4
5
6
7
8
9
10
11
12
Flotation pH
Figure 19.2 Effect of pH on malachite recovery using hydroxamic acid as collector.
Carboxilic acid flotation of malachite has been commercially used for over 70 years. This collector is prepared by heating a mixture of hydrolysed palm oil (or oleic acid) and fuel oil in a 3:1 ratio. This mixture is manly used for recovery of malachite from siliceous ores. The use of carboxylic acid for malachite flotation from carbonaceous ores resulted in both reduced concentrate grade and recovery. Cationic flotation of malachite, using mono- and diamines in alkaline pulp, was also examined. Malachite floats readily using mono-amines under laboratory conditions. Figure 19.3 illustrates the floatability of pure malachite with different amines. It should be pointed out that there are several varieties of malachite, Cu4(PO4)2(OH)4·OH. Pseudomalachite is difficult to float, and it is well known that pseudo-malachite can be floated with anionic collectors, but responds poorly to the sulphidization method. In a number of oxide ores, cuprite (Cu2O, Cu = 88.8%, SG = 5.9) is present as secondary minerals together with sulphides, malachite and tenorite. Cuprite can be floated using either sulphidization or anionic flotation methods. The flotation properties of cuprite are some what different from that of malachite. For example, using a sulphidization method for flotation of cuprite requires higher dosages of sulphidizer. Some ore deposits contain cuprite as the principal mineral. Typically, these deposits contain appreciable amounts of slimes and clay minerals. The laboratory studies conducted on these types of ore indicated that improved metallurgical results can be achieved using the sulphidization method with ester-modified xanthate [8]. Tenorite (CuO; Cu = 80%, SG = 6.5) is usually present in mixed copper oxide and sulphide ore. The flotation properties of tenorite are similar to that of cuprite.
19.4
Cobalt and Copper Cobalt Oxide Ores
51
100
monoamine
Malachite recovery (%)
90
diamine 7 EtO
80
monoamine 2 EtO
monoamine 7 EtO
monoamine 11 EtO 70
60 1
4
7
10
13
16
Collector concentration (mg/L)
Figure 19.3
Floatability of malachite with C-18 mono- and dialkylamines at pH 8.5–9.0.
Azurite (Cu3(OH2)(CO2)2, Cu = 55.3%, SG = 3.7) usually appears in small quantities together with malachite in a number of deposits in Zambia and Congo. From plant and laboratory data, azurite has similar flotation properties as malachite. Atacamate (Cu2(OH2)Cl; Cu = 44.6%, SG = 3.8) is common to the Atacoma desert in Chile, for which this mineral was named. As an individual mineral, it does not have any significant economic value. No data on the floatability of this mineral are known. Chrysocolla (CuOxSiO2; Cu = 10–36%, SG = 2–2.4) is the most studied mineral of all the oxide minerals. Extensive laboratory studies have been conducted by numerous researchers [9–11]. The laboratory research work indicates that chrysocolla can be floated using the sulphidization method, as shown in Figure 19.4, or by hydroxamate collectors. However, none of these processes have been applied at an industrial scale. In a number of operations, chrysocolla has been recovered using a hydrometallurgical technique. The flotation properties of bronchantite, antlerite and chalcantite were not examined. These oxide minerals are contained in an altered sulphide ore in some deposits in South America and Zambia.
19.4
COBALT AND COPPER COBALT OXIDE ORES
In the deposits where oxide cobalt is present, it is common to have oxide copper minerals . The cobalt is, therefore, recovered in a bulk copper–cobalt concentrate that is processed using a hydrometallurgical technique to produce separate copper and cobalt metals. Oxide
52
19.
Flotation of Oxide Copper and Copper Cobalt Ores
100
Recovery (%)
80
dixanthogen emulsion + 10 mg/L AmX
60
40
20 aqueous dixanthogen emulsion
aqueous benzene emulsion
0 0
100
200
300
400
500
Concentration of Na2S•9H2O (mg/L)
Figure 19.4
Effect of Na2S concentration on the flotation of chrysocolla.
cobalt minerals belong to the heterogenite group, which consists of complex hydrated cobalt oxides of various compositions and degrees of crystallization. In view of the complex mineralization, oxide cobalt minerals are known to be rather difficult to float. The floatability of oxide cobalt minerals is strongly influenced by the presence of small amounts of copper in its crystalline structure. From the point of view of flotation properties, the cobalt minerals can be classified in two main groups: (a) crystalline varieties with compositions closely responding to the formula CoO·OH. Cobalt is trivalent and only minor amounts of impurities enter the structure. This is the most difficult variety of cobalt to float using conventional reagent schemes; and (b) crypto crystalline or amorphous varieties that contain various amounts of copper–nickel–iron and also bivalent cobalt. Their formula is of the form (χCo2O3·γCoO·ZCuO)H2O + n% hygroscopic H2O. These varieties of cobalt can be floated using reagent schemes used to float oxide copper minerals.
19.5
FLOTATION PRACTICE IN BENEFICIATION OF OXIDE COPPER MINERALS
Selection of a reagent scheme for beneficiation of oxide copper ores depends on many factors; some of the more important ones being • Type of oxide copper minerals present in the ore. • Type of gangue minerals – some ore types contain silicate gangue free of slimes, which are the most amenable to flotation. Ores with dolomitic gangue can be beneficiated
19.5
Flotation Practice in Beneficiation of Oxide Copper Minerals
53
using sulphidization only. These ores usually contain an appreciable amount of clay slimes that have a detrimental effect on flotation. Some oxide ores contain talc, iron hydroxides and iron oxides. In general, each ore type requires the selection of different reagent schemes. • Degree of liberation – the relatively fine-grained ores are more amenable to flotation than the disseminated ores, which require finer grinding. • Chemical composition and physical structure of the copper minerals play an important role in the floatability of oxide copper minerals [12]. Oxide copper minerals are often porous and aqueous soluble. Because of that, they tend to slime during grinding. During the past two decades, there has been an appreciable amount of research work conducted mainly on the application of hydroxamates for oxide copper flotation. These reagents have yet to find industrial application. In recent years, a new class of collectors, consisting of xanthated fatty acids (TY collector), and monoester-modified xanthate (PM230) have found industrial applications with improved metallurgical results. From plant practice, treating oxide copper and copper cobalt ores, two basic flotation methods are practiced: (a) sulphidization flotation method, and (b) anionic flotation method. 19.5.1
Sulphidization flotation method
This method is the most commonly used in beneficiation if oxide copper-bearing ore. The reagent schemes used to treat oxide copper ores, mixed copper sulphide oxide ores and oxide copper cobalt ores varies from one ore type to the next, mainly by type of collector and sulphidizer used. The choice of reagent scheme depends largely on the type of natural ore to be treated. The three main groups of reagents used in beneficiation of oxide copper and copper cobalt ores include (a) sulphidizers, (b) collectors and (c) modifiers and depressants. Choice of sulphidizer and effect on flotation The most preferred sulphidizer used in flotation of oxide copper minerals is Na2S · 9H2O. Other sulphidizers used in operating plants include NaHS and (NH4)2S. Actually, the selection of a sulphidizer is based on the consumption required for flotation of oxide copper from particular ore types. For example, in some cases the consumption requirement of NaHS is much higher than for Na2S. Figure 19.5 shows the effect of different levels of sulphidizer on the recovery of malachite using xanthate collector. From the data generated, higher dosages of NaHS are required to achieve activation of malachite. From plant and laboratory experience [13], the sulphidization method using xanthate collector is sensitive to the following, major factors: • Rate of sulphidizer additions must be carefully controlled to obtain optimum sulphidization and prevent excess SH– ions that may cause depression. • Sometimes higher additions of sulphidizer are required, especially if the ore contains excessive amounts of slimes.
54
19.
Flotation of Oxide Copper and Copper Cobalt Ores
100
Na2S•9H2O
Copper recovery (%)
80
NaHS 60
40
(NH4)2S
20
0 0
500
1000
1500
2000
2500
Activator additions (g/t)
Figure 19.5 Effect of levels of different sulphidizers on copper flotation from the Kolwezi open pit ore (Congo, Africa).
The consumption rate of sulphidizer also depends on the type of collector used. When using xanthate only, the sulphidizer rate is much higher than when using certain secondary collectors, such as dithiophosphates. Choice of depressants In a large number of oxide flotation plants, sodium silicate (Na2SiO3) is used as a gangue depressant. In the past two decades, a new line of depressants has been developed and introduced into a number of operating plants. Some of these depressants include (a) a mixture of sodium phosphate and lignin sulphonate (i.e. depressant 3XD), (b) a mixture of a low-molecular-weight acrylic acid and sodium silicate (depressant 2D) and (c) hydrosol based on the reaction of sodium silicate with alumina sulphate (depressant SD). These depressants were extensively examined on copper oxide ores from the Nchanga mine in Zambia. Figure 19.6 shows the grade–recovery relationship using different depressant combina tions. Depressants 3XD and 2MD have shown excellent gangue depression. The presence of clay in the ore has a detrimental effect on copper oxide flotation. Results from experimental development test conducted on various clay containing ore types using AQ depressants showed that in the presence of these depressants, the results improved markedly using the sulphidization flotation method.
19.5
Flotation Practice in Beneficiation of Oxide Copper Minerals
55
100 Depressant Type
Copper recovery (%)
80
60
2MD
40
3XD
800 g/t sodium silicate
20
none 0 0
5
10
15
20
25
30
35
Copper concentrate grade (%)
Figure 19.6 Effect of levels of various depressants on copper grade–recovery relationship from Nchanga open pit ore.
Table 19.2 shows the effect of these depressants on oxide copper metallurgical results. The ore used in these studies was from Dima (Shaba Province, Congo) underground oxide ore. The results obtained demonstrated that the type of gangue depressant plays a sig nificant role in achieving good results. Research on gangue depressants has been conducted by numerous researchers [16,17]. It has been proven that sodium silicate does not depress calcite or dolomite using the sulphidization flotation method. It requires much higher additions (i.e. up to 2000 g/t) to depress dolomite. Using Cataflot P39 (a modifying agent developed by Pierrefitte-Abibu) showed excellent calcite depression at lower addi tion rates. The pilot plant results obtained on a Morrocan oxide copper ore (Table 19.3) showed significant increase in copper recovery with the use of Cataflot 39 over that of sodium silicate. Other depressants examined included polysaccharides, polyacrylamides, polyphosphates and carboxymethylcellulose. None of these depressants are found in industrial application. Choice of collectors As mentioned earlier in this chapter, the choice of collector is very much dependent on the type of copper minerals, as well as the type of gangue minerals present in the natural ore. If the ore contains siliceous gangue minerals, then various fatty acid modifications can be used as the principal collector in plant practice. Ores containing carbonaceous and dolo mitic gangue minerals, where sulphidization method is used, xanthate collector is used as
56
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Table 19.2 Effect of AQ depressants on oxide copper results Depressant
No depressant 650 g/t Na2SiO3 650 g/t AQ2 650 g/t AQ3
Product
Weight (%)
Copper cleaner concentrate Copper combined tail Head (calc) Copper cleaner concentrate Copper combined tail Head (calc) Copper cleaner concentrate Copper combined tail Head (calc) Copper cleaner concentrate Copper combined tail Head (calc)
13.92 86.08 100.00 12.49 87.51 100.00 14.50 85.50 100.00 17.58 82.42 100.00
Assays (%)
% Distribution
CuT
CoT
CuT
CoT
23.98 1.73 4.67 28.30 1.47 4.82 27.02 0.81 4.61 26.42 0.71 5.23
– – – 1.74 0.21 0.40 1.55 0.18 0.37 1.96 0.20 0.51
71.5 28.5 100.0 73.3 26.7 100.0 85.0 15.0 100.0 88.8 11.2 100.0
– – – 54.2 45.8 100.0 59.4 40.6 100.0 67.6 32.4 100.0
Table 19.3 Effect of Cataflot 29 on oxide copper flotation Depressant
Head (% Cu)
Tail (% Cu)
Concentrate (% Cu)
Recovery (% Cu)
Na2SiO3 Cataflot P39
2.16 2.15
0.38 0.29
34.5 35.0
82.5 87.8
the primary collector. Mercaptan and/or dithiophosphates are used as secondary collectors, when the ore contains cobalt minerals. The fatty-acid-based collectors have been employed for the past 60 years for flotation of oxide copper/cobalt minerals from Congo, a company formerly owned by Union Minera (Belgium). The fatty acid modification was used in operating plants at Kolwezi, Koumbore and Kakanda. The fatty acid used was hydrolysed palm oil prepared as a mixture consisting of 75% hydrolysed palm oil/21% gas oil/4% Unitol. This mixture is passed through a colloidal mill in the presence of 0.5% soda ash solution. The fatty acid prepared in this manner does not produce voluminous froth and is more selective than ordinary fatty acid mixtures. Experimental laboratory testwork conducted on the Kolwezi siliceous ore [18] with the above-mentioned mixture, with different degrees of dispersion showed substantial differences in metallurgical results (Table 19.4). Poor results were achieved when there was no dispersion of the mixture. The best results were obtained when the mixture was treated for 10 min in an ultrasonic mixer. In each case, the mixture was dissolved in a 0.5% soda ash solution.
19.5
Flotation Practice in Beneficiation of Oxide Copper Minerals
57
Table 19.4 Effect of degree of fatty acid mixture dispersion on Cu/Co flotation from siliceous Kolwezi open pit ore – laboratory-locked cycle tests Ultrasonic mixture pretreatment time (min)
0 5 10
Head assays (%)
Tailing assays (%)
Concentrate assays (%)
% Recovery
Cu
Co
Cu
Co
Cu
Co
Cu
Co
4.50 4.60 4.61
0.60 0.59 0.60
1.50 1.35 0.85
0.47 0.31 0.19
20.1 23.3 25.2
1.28 2.20 2.65
72.0 75.1 84.3
34.5 55.2 68.1
Xanthated fatty acid mixture is a new line of collectors, specifically designed for beneficiation of oxide copper ores that contain dolomitic and carbonaceous gangue miner als [19]. This collector was developed after extensive laboratory development testwork. The effectiveness of this collector was compared to a standard xanthate collector in a series of continuous locked cycle tests (Table 19.5). Using TY3 collector improved the copper recovery by 10%, while the cobalt recovery remained unchanged. The consumption of sulphidizer in the presence of TY3 was sig nificantly reduced. Based on the encouraging results obtained from laboratory testing, a plant test was conducted at the Kolwezi concentrator, during which a number of important factors in preparation of the fatty acid xanthate emulsion were discovered, some of which included: • High-power mixture for emulsion preparation was required to obtain a stable emulsion of fatty acid and xanthate mixture. • The collector emulsion was more stable when using potassium xanthate instead of sodium xanthate. Table 19.5 Comparison of results using xanthate and TY3 collector using dolomitic oxide Cu/Co ores Test conditions
Plant SNBX 3300 g/t NaHS Laboratory-locked cycle SNBX 2600 g/t NaHS Laboratory-locked cycle TY3 collector 1990 g/t NaHS
Product
Copper concentrate Copper combined tail Head (calc) Copper concentrate Copper combined tail Head (calc) Copper concentrate Copper combined tail Head (calc)
Weight (%)
13.01 86.97 100.00 11.96 88.04 100.00 16.76 83.24 100.00
Assays (%)
% Distribution
Cu
Co
Cu
Co
31.84 1.68 5.61 32.57 2.13 5.77 28.96 1.03 5.71
– – – 1.34 0.12 0.27 0.97 0.12 0.26
74.0 26.0 100.0 67.5 32.5 100.0 85.0 15.0 100.0
– – – 60.3 39.7 100.0 60.0 40.0 100.0
58
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Table 19.6 Average plant results obtained with TY collector – 30-day plant test Collector
TY3 SNBX
Product
Weight (%)
Copper concentrate Final tailing Head (calc) Copper concentrate Final tailing Head (calc)
Assays (%)
14.27 85.73 100.00 13.80 86.20 100.00
% Distribution
Cu
Co
Cu
Co
23.73 0.82 4.09 22.94 1.14 4.15
2.23 0.21 0.50 2.04 0.20 0.45
83.0 17.0 100.0 76.3 23.7 100.0
64.0 36.0 100.0 62.5 37.5 100.0
The plant trial emulsion consisted of 70% amyl xanthate, 20% hydrolysed palm oil and 10% fuel oil. The average results obtained in the plant are compared in Table 19.6. A significant improvement in copper recovery was realized with the use of the TY3 collector in the Kolwezi plant. Nowadays, TY3 is used in a number of operating plants in Africa. Collectors from the PM series are a mixture of xanthate/mercaptans, modified with esters or surfactants, specifically developed for flotation of mixed copper oxide ores and copper cobalt oxide ores. There are several collectors from the PM series, including PM230, PM250 and PM270. Experimental laboratory work was conducted on the mixed copper oxide sulphide ores from the Komoto plant in Chile using collectors from the PM series [20]. The effect of these collectors on copper grade–recovery relationship is illustrated in Figure 19.7. The results indicated that both grade and recovery can be significantly improved, compared to the results obtained with xanthate. 100 MM 290 210 g/t PM 250 200 g/t Na amyl xanthate 250 g/t
Copper recovery (%)
90
80
Collector Type 70
60
50 5
10
15
20
25
30
35
40
Copper concentrate grade (%)
Figure 19.7
Effect of collectors from the PM series on the copper grade–recovery relationship.
19.6
Industrial Practice in Flotation of Oxide Copper and Copper-Cobalt Ores
19.6
59
INDUSTRIAL PRACTICE IN FLOTATION OF OXIDE COPPER AND COPPER-COBALT ORES
Most operating plants that treat oxide copper and copper-cobalt ores are found in Central Africa and Southern Africa regions. A few operations exist in Chile, Brazil and Peru, where they treat mixed oxide sulphide ores or oxide copper gold ore. In general, the reagent schemes used in these plants depend largely on the type of ore being treated. The following sections describe the operating practices of the major plants that treat oxide and mixed oxide sulphide copper ores. 19.6.1
Kolwezi concentrator (Kongo) – Oxide siliceous ore
For many years this plant has treated an oxide siliceous ore using the hydrolysed palm oil mixture. The palm oil–fuel oil mixture is heated to about 60°C in the presence of soda ash and then passed through a colloidal mill before it is added to the copper conditioner. A typical reagent scheme used to treat the Kolwezi siliceous ore is shown in Table 19.7. The soda ash and sodium silicate are added to the grinding mills and the palm oil emulsion to the copper conditioner. The addition of soda ash is quite important, as the water used in the plant contains an appreciable amount of calcium and magnesium, where the soda ash acts as a water softener. The flowsheet used in this plant (Figure 19.8) consists of a rod mill–ball mill grinding system and a copper rougher–scavenger flotation circuit, followed by two cleaning stages. Initially, the plant used a rake classifier, but now the rake classifiers have been replaced by cyclones. One of the main problems associated with beneficiation of the Kolwezi siliceous ore is the production of malachite and pseudomalachte slimes that have a relatively low flotation rate. Most of the copper losses occurring in the plant are in the –15 µm fraction. Experi mental testwork conducted with a different palm oil emulsifier indicated that copper recovery from the fine fraction can be significantly improved with the use of petroleum sulphonate (Petrosol 845) as the emulsifier [21] for palm oil. Significant improvement in copper recovery was realized in the fine fractions with the use of palm oil emulsified with Petrosol 845. Table 19.7 Reagent scheme used to treat the Kolwezi siliceous ore Reagent
Depressants and modifiers Soda ash Sodium silicate Collectors and frothers Palm oil emulsion Pine oil
Additions (g/t) Grinding
Copper flotation
1500–2000 800–1200 – – –
– – – 2000–2500 30–40
60
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Ore feed Rod mill
Ball mill
u/f
Cyclones
o/f Conditioning
Copper rougher 1
Slimes
Copper rougher 2
Cyclones Sand Copper scalper
Tail
Copper cleaner Combined tailings Copper recleaner
Copper cleaner concentrate
Figure 19.8
Typical flowsheet used in treatment of dolomitic oxide ores.
Other plants that treat siliceous copper oxide ores include Panda and Kabolela plants from the same area. The gangue in this ore is composed of argillaceous and siliceous schist. Both plants essentially use the same flowsheet and reagent scheme, as that described for the Kolwezi plant. Typical plant results during treatment of a siliceous ore are presented in Table 19.8. These are average results achieved from 1980 to 1982.
19.6
Industrial Practice in Flotation of Oxide Copper and Copper-Cobalt Ores
61
Table 19.8 Kolwezi plant results during treatment of the siliceous ore Product
Weight (%)
Copper concentrate Copper tailing Head (calc)
19.6.2
13.26 86.72 100.00
Assays (%)
% Distribution
Cu
Co
Cu
Co
26.65 1.00 4.40
2.51 0.25 0.55
80.3 19.7 100.0
60.3 39.7 100.0
Industrial practice in beneficiation of dolomitic oxide ores
Industrial plants that treat dolomitic ores include the Kolwezi concentrator (Kongo) and the Nchanga concentrator (Zambia), along with several smaller plants in the Kolwezi district. The reagent scheme generally used in these concentrators is presented in Table 19.9. Sodium silicate is used as the common depressant, and also acts as a dispersant together with the soda ash. In the majority of operating plants, Na2S · 9H2O is used as the principal activator. Some operating plants in Zambia use NaHS as a sulphidizer. Sodium or potassium xanthates are the principal collectors used, where mercaptans are used as secondary collectors. In the 1980s, a new collector (i.e. fatty acid-modified xanthate) was introduced into the Kolwezi concentrator with significant improvement in copper recovery. In 1995, collectors from the PM series were tested in the Nchanga concentrate improving results. The plant results obtained in the Kolwezi concentrate using xanthate and TY3 are compared in Table 19.10. Collector TY3 also had a positive effect on cobalt recovery. The flowsheet used to treat dolomitic oxide copper ores is somewhat different from that used in the beneficiation of siliceous oxide copper ores. This is due to the fact that dolomitic ore usually contains elevated amounts of slimes, in which case a split circuit flowsheet has been adopted in a number of operations. The typical flowsheet used for treatment of dolomitic ores is shown in Figure 19.8. Usually, the scavenger tailings are deslimed and the sand fraction is retreated in a scalp copper flotation stage. When the ore is deslimed before flotation, a large amount of fine copper is lost in the slime fraction.
Table 19.9 Reagent scheme used to treat dolomitic ores Addition rate (g/t) Depressants and modifiers
Activators
Collectors
500–1000 g/t Na2SiO3 Na2CO3 to pH 8–9.5 100–200 Guar
1500–300 Na2S 1200–3000 NaHS
150–300 xanthate 50–80 mercaptan 150–200 TY3 100–200 PM290
62
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Table 19.10 Kolwezi plant results using xanthate and TY3 (1980–1982) Collector
320 g/t SIPX 280 g/t TY3
Product
Copper concentrate Copper tailing Head (calc) Copper concentrate Copper tailing Head (calc)
Weight (%)
15.57 84.43 100.00 12.56 87.44 100.00
Assays (%)
% Distribution
Cu
Co
Cu
Co
24.6 1.03 3.70 25.1 0.76 3.82
1.4 0.24 0.42 2.4 0.20 0.48
76.5 23.5 100.0 82.5 17.5 100.0
51.9 48.1 100.0 62.8 37.2 100.0
SIPX, sodium isopropyl xanthate.
19.7 INDUSTRIAL PRACTICE IN BENEFICIATION OF MIXED SULPHIDE OXIDE ORES The mixed sulphide oxide ores usually contain two or more oxide minerals, including cuprite, malachite and tenanntite. The sulphide copper minerals are represented by covellite and bornite. Examples of this type of operation are located in the former Republic of Zaire (Komoto, Dima 1 and 2 plants), and the Nchanga open pit plant in Zambia. In general, this ore type is treated using two distinct circuits: sulphide copper flotation followed by oxide copper flotation, manly using the sulphidization method. The basic reagent scheme used in the concentrators varies and is dependent on the type of copper minerals present in the ore, and the mineral composition of the gangue. The reagent scheme used in the three main concentrators treating this ore type is presented in Table 19.11. There are only slight differences in the reagent schemes used to treat the mixed sulphide oxide ores. The type of sulphidizier and collectors are the main variance in the reagent
Table 19.11 Reagent schemes used in beneficiation of mixed sulphide oxide ores Concentrator
Reagent consumption (g/t) Sulphide circuit
Komoto, Kongo Na2SiO3 = 300, CaO = 460 Ethyl xanthate = 100, frother 41G = 15 Na2SiO3 = 200, Na2CO3 = 300 Dima, Kongo K-amyl xanthate = 60, frother 41G = 10 Nchanga, PM290 = 15, Na-amyl xanthate = 30 Zambia Pine oil = 10
Oxide circuit Na2CO3 = 200, NaHS = 1350, fuel oil = 50 K-amyl xanthate = 210, frother 41G = 20 Na2SiO3 = 600, NaHS = 3000 Mineral oil = 90, frother 41G = 15 Na2CO3 = 300, Na2S = 1200, PM290 = 40 Na-amyl xanthate = 100, kerosene = 100
19.7
Industrial Practice in Beneficiation of Mixed Sulphide Oxide Ores
63
Ore feed Rod mill
Ball mill
u/f
Cyclones o/f
Copper sulphide rougher
Copper sulphide cleaner
Conditioning
Copper oxide rougher
Copper sulphide re-cleaner
Copper oxide 1st cleaner
Copper sulphide cleaner concentrate
Copper oxide 2nd cleaner
Copper oxide scavenger
Tail
Copper oxide cleaner concentrate
Figure 19.9
Typical flowsheet used in treatment of mixed sulphide oxide ores.
schemes used. The generalized flowsheet used for treatment of mixed sulphide ores is shown in Figure 19.9. Some operations use semi-autogenous mills for primary grinding (Komoto, Dima) with grind finenesses ranging from 55% to 60% minus 200 mesh. The plant metallurgical results achieved in these concentrators are presented in Table 19.12. In most cases, the results obtained on mixed copper sulphide oxide ores are better than those obtained on oxide ores. The floatability of oxide copper from mixed ore is usually better than the floatability of copper from oxide ores. An appreciable amount of the cobalt in these ores is represented by sulphide cobalt minerals, mainly carrolite.
64
19.
Flotation of Oxide Copper and Copper Cobalt Ores
Table 19.12 Plant results from mixed copper sulphide oxide ores Concentrator
Komoto (1,2) Congo, Africa Dima Kongo, Africa Nchanga ZCCM, Zambia
Product
Weight (%)
Cu sulphide concentrate Cu oxide concentrate Cu flotation tail Head (calc) Cu sulphide concentrate Cu oxide concentrate Cu flotation tail Head (calc) Cu sulphide concentrate Cu oxide concentrate Cu flotation tail Head (calc)
7.11 2.57 90.32 100.00 2.78 8.59 88.63 100.00 1.82 3.53 94.65 100.00
Assays (%)
% Distribution
Cu
Co
Cu
Co
43.6 22.0 0.93 4.50 55.31 21.24 0.82 4.09 47.5 37.0 0.72 2.80
3.61 1.52 0.06 0.35 2.23 1.58 0.16 0.34 – – – –
68.83 12.55 18.62 100.0 37.65 44.64 17.71 100.0 30.3 45.8 23.9 100.0
74.26 11.30 14.44 100.0 18.24 39.84 41.92 100.0 – – – –
REFERENCES 1. Saquet, J.J., and Mining, K., Metallurgical Operation of Union Miniere du Haut Katanga, Mining Engineering, Vol. 14, Nr 12, pp. 71–81, 1962. 2. Bulatovic, S., Bigg, A.C.T., and Yen, T., Laboratory Development Testwork on Kolwezi and Komoto Oxide and Mixed Copper Cobalt Ores, Report of Investigation No. 3, 1979. 3. Gaudin, A.M., Flotation, McGraw-Hill Book Company Inc., New York, 1957, pp. 431–450. 4. Normand, J., The Adsorption of Potassium Octyl Hydroxamate on Malachite, Thesis for Masters in Metallurgical Engineering, McGill University, Montreal, Canada, 1974. 5. Cuyper, D.J. Flotation of Oxide Copper Ores; Paper Presented on the GDMB General Meeting; Hague Netherlands, 1976. 6. Saquet, J.J., Kolwezi Mining and Metallurgical Operation of Union Miniere du Haut Katanga. Mining Engineering, Vol. 14, Nr 12, p. 71/81, 1975. 7. Ser, F., Sulphydric Flotation of Previous Sulphidized Oxide Copper Minerals of Nachanga Consolidated Copper Mines Limited (Zambia), Rudy, Vol. 5, pp. 169–174, 1970. 8. Bulatovic, S.. The Investigation into Recovery of Gold Containing Cuprite Ores using Sulphidi zation with Ester Modified Xanthate, Report of Investigation LR3894, 1996. 9. Castro, S., Gaytan, H., and Goldfarb, S., The Stabilizing Effect of Na2S on the Collector Coating of Chrysocolla, International Journal of Mineral Processing, Vol. 10, No. 3, pp. 71–82, 1976. 10. Gonzales, G., The Recovery of Chrysocolla with Different Long Chain Surface-active Agents as Flotation Collectors, Journal of Applied Chemistry Biotechnology, Vol. 16, No. 28, pp. 31–38, 1978. 11. Parks, G.A., and Kovac, S., Thermal Activation of Chrysocolla for Xanthate Flotation, Society of Mining Engineers, No. 346, pp. 28–34, 1966. 12. Aplan, F.F., and Fuerstenau, D.W., Froth Flotation, AIME, New York, Chapter 7, pp. 170–214, 1962. 13. Bulatovic, S., Bigg, A.C.T., and Yen, T., Improvement in Plant Performance of Seven Concen trators from Shaba Province (Zaire), Report of Investigation LR2550 Vol. 2, pp. 1–250, 1982.
References
65
14. Bulatovic, S., Development of Reagent Scheme for Beneficiation of Nachanga (Zambia) Open Pit Oxide Ore, Report of Investigation LR5635, pp. 20–156, 1993. 15. Yen, T., and Bulatovic, S., Investigation for the Improvement of Concentrate Grades and Metal Recoveries from Western Group of Concentrators, Report of Investigation No. 4 LR2550, 1985. 16. Ealy, G.K., Concentration of Copper and Copper Oxides by Flotation at Nacimiento, Mining Congress Journal, Nr 33, pp. 63–66, 1973. 17. French Patent No. 1,519,540, May 11, 1966 18. Bulatovic, S., Bigg, A.C.T., and Yen, T., In Plant Studies of Different Collectors at Dima Concentrator, Gecamines, Zaire, Report of Investigation LR2530, 1981. 19. Bulatovic, S., Bigg, A.C.T., and Yen, T., Development and Plant Application of New Collector for Oxide Copper/Cobalt Flotation for Gecamines, Zaire Flotation Plant; Copper 91, Aug 18–21, 1991, Ottawa Canada, pp. 300–338. 20. Bulatovic, S., and Jessup, T., Evaluation of New Line of Collector on Zambian Sulphide/Oxide Ore and the Chilean Oxide Ore, R&D Report 87, 1993. 21. Bulatovic, S., Bigg, A.C.T., and Yen, T., Investigation for the Improvement in Concentrator Performance of the Western Group of Concentrates, Report 5, 1982.
– 20 –
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
20.1 SOME GEOLOGICAL AND MINERALOGICAL FEATURES OF MIXED SULPHIDE OXIDE AND OXIDE LEAD ZINC ORES There are a variety of mixed sulphide oxide ores and oxide lead zinc ores, and these ores cannot be classified into any specific group due to the vast differences in geology and mineral compositions. Only limited literature is available on the geological and miner alogical characteristics of these ores. From a processing point of view, these ores can be classified into the following groups: • • •
Mixed sulphide and oxide lead zinc ores Oxide lead zinc ores Oxide zinc ore with little or no lead present.
20.1.1
Mixed sulphide oxide lead zinc ores
In general, the calceous-dolomitic rocks from the Cambrian age are affected by their upper beds, by sulphide mineralization of lead, zinc and iron contemporaneous with sedimentation. The oxide lead and zinc minerals are disseminated through dolomitic limestone. As a consequence of the action of the descending process, these formations may assume different types of mineralization. According to the intensity of the oxidation process, which is associated with the different characteristics of the country rock, this country rock may be (a) principally calceous, (b) calceous with dolomitized zones and (c) primarily dolomitized. In the case of calceous rock formation, the mineralization is well defined in parallel veins with the width ranging from 1 to 6 m. In these ores, the lead can vary from 3% to 12% zinc. With calceous rock types, the oxide mineralization extends beyond the veins forming concretion and small masses. The lead content of these mineralization types range from 1.5% to 2.5%, of which �50% is oxidized. The zinc content ranges from 5% to 12%, of which 70% is oxide zinc. In the case where dolomitization is extensive, the oxidized mineralization is present everywhere, without regular outline. These are relatively low-grade ores. In all of the
67
68
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
above-described cases, clay is often present as a filling in the mineral cavities, which are created as a result of dissolution of limestone and by dolomitization. These ore types are found in Sicily, Ireland, Morocco and Canada. 20.1.2
Oxide lead ores
There are only a few deposits of oxide lead ores, some of which are processing plants. These include Mibladen Mine in Morocco, several deposits in Sardinia and Tynagh deposit in Ireland. In general, the mineral composition varies significantly. The limestone deposits contain barite and dolomite. In these ores, lead is represented by cerussite with little to no galena present. The oxide lead ore in a dolomitic matrix may contain clay slimes and a fairly large amount of pyrite and barite; both of these minerals have a negative effect on oxide lead flotation. These ores may contain one or several lead minerals, including cerussite, anglesite and pyromorphite. Some lead oxide ore varieties may contain oxide copper minerals in the range of 0.2–1% Cu. 20.1.3
Zinc oxide ores
As for the lead oxide ore types, there are a variety of zinc oxide ores. There are three basic ore types of economic value: (a) ores with hemimorphite as the predominant zinc oxide minerals, (b) ores with smithsonite as dominant zinc oxide minerals and (c) ores with a mixture of smithsonite and willemite. Typical mineral compositions of this ore are shown in Table 20.1. The smithsonite ore type is usually composed of a mixed assemblage consisting of dolomite, Fe-oxyhydroxide, quartz clay (kaoline and montmorillonite), minor amount of
Table 20.1 Mineral composition of the major zinc oxide ores Mineral
Smithsonite Calcite Cerussite Geothite Quartz Kaolinite Feldspar Jerusite Dolomite Hemimorphite Willemite Barite
Formula
ZnCO3 CaCO3 PbCO3 FeO(OH) SiO2 Al2SiO6(OH)4 KAlSi3O8 KFe3(SO4)2(OH)6 MgO Zn2H2SiO5 Zn2SiO4 BaSO4
Weight (%) Smithsonite
Hemimorphite
Willemite and smithsonite
38 5 <5 28 15 5 <5 <5 – – – –
2 – Trace 20 10 10 3 – 30 18 – –
7 4 1 5 20 5 8 – 5 – 21 10
20.2
Flotation Properties of Individual Oxide Lead Zinc Minerals of Economic Importance
69
feldspar and Mn-oxyhydroxide. Hemimorphite is typically present as coarse, radiating prismatic crystals between 100 and 500 µm in length. In some ores the hemimorphite grains commonly exhibit rimming and inclusins of fine Fe-hydroxides and iron-stained clay minerals. The hemimorphite ore type is composed of a variety of gangue minerals. Some ores contain a fine-grained aolitic limestone with fossil fragments and random replacement veins of smithsonite, which is highly pigmented with hemimorphite. The goethite in these ores is present as pigmented crystalline masses. The ferruginous calcite in the ore is sometimes replaced with hemimorphite aggregates or crystals, 50–60 µm in size. The hemimorphite can also be found intergrown with goethite and limestone. This ore type belongs to the refractory ore type, found in Egypt and North Africa. The mixed willemite–smithsonite ore has the simplest mineral composition of the three basic ore types. The silicate, goethite and barite are the principal gangue minerals. Will emite is a major zinc oxide mineral present as free crystals ranging from 50 to 500 µm in size. Smithsonite is usually stained with Fe-hydroxides and sometimes is associated with silicate as inclusion and/or attachments. The barite content of the ore may vary from several percent up to 12%. A few deposits of this ore type are found in Mexico and South America.
20.2
20.2.1
FLOTATION PROPERTIES OF INDIVIDUAL OXIDE LEAD ZINC MINERALS OF ECONOMIC IMPORTANCE Oxide lead and zinc minerals of economic value
There is a fairly large number of oxide lead and zinc minerals, only a few of which have been studied. Table 20.2 shows a list of the oxide lead and zinc minerals of economic value. These minerals can occur separately (e.g. cerussite) or as mixtures of two or more oxide minerals. Depending on the formation of the ore body with oxide minerals, the ore may Table 20.2 Lead and zinc oxide minerals of economic value Mineral
Chemical formula
Content (%)
Specific gravity (g/cm3)
Cerusite Anglesite Pyromorphite Crocoite Wulfenite Mimettizite Plumbojarosite and mimetite Calamine (hemimorphite) Clinohedrite Smithsonite Willemite Zincite
PbCO3 PbSO4 Pb5(PO4)3Cl PbCrO4 Pb(MnO4) Pb3(AsO4)3Cl PbFe6(SO4)4(OH)12 Zn4(OH)2TSi2O7TOH Ca2Zn2(OH2)Si2O T 7H2O ZnCO3 Zn2SiO4 ZnO
77.0 68.3 75.8 64.6 55.8 69.5 18.3 67.3 60.0 35.0 74.0 20.0
6.5 6.3 6.5 5.9 7.1 7–7.2 4.5 3.4–3.5 3.4–3.7 4.3–4.4 3.9–4.2 5.5
70
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
contain only lead or zinc oxide minerals or a mixture of the lead and zinc oxide minerals. Very often there are ores that contain mixtures of sulphide and oxide lead minerals, which are processed in operating plants. There has been very little research data or literature for a number of the minerals listed in Table 20.2. Most recently, extensive laboratory research work was carried out on a variety of natural ores containing lead oxide, zinc oxide and mixed lead zinc oxide ores.
20.2.2
Flotation properties of oxide lead minerals
Flotation of the lead oxide minerals is a difficult problem not least because there are no known direct acting collectors. Normally, during oxide lead flotation, a sulphidization method is used with xanthate as a collector. In the majority of cases, the ore is pretreated using a desliming process, especially if the ore contains clay and Fe-hydroxides. Another method includes the preconcentration using heavy liquid. Of all the lead oxide minerals, cerussite and anglesite [1,2] have been studied the most. The flotation properties of cerussite and anglesite are different in such a way as the anglesite is less amenable to sulphidization than cerussite. The sulphidization process for both minerals is a delicate process and is sensitive to (a) level of additions of sulphidizer, (b) the number of sulphidization stages and (c) conditioning time with sulphidizer. Excess of sulphidizer has a negative effect on cerussite and anglesite recovery. Figure 20.1 shows the effect of the level of Na2S on cerussite and anglesite recovery. These results have demonstrated that too high an addition of sulphidizer results in mineral depression. Depending on the gangue composition of the ore, stage additions of sulphidizer have been proven useful [3]. Choice of sulphidizer Normally the Na2S and NaHS are used as sulphidizers. The choice between Na2S and NaHS depends on the gangue composition, as well as the ratio between cerussite and anglesite in the ore. The anglesite sulphidization process is more complete when using NaHS. When using Na2S as the sulphidizer, aeration with sulphidizer was proven to be beneficial. When cerussite is only slightly sulphidized, it is sensitive and, with the absence of aeration, rapidly loses floatability. Studies conducted on lead oxide flotation from mixed ore showed that without aeration, low lead recoveries were achieved. Table 20.3 shows the effect of aeration with sulphidizer on lead oxide metallurgical results. Prolonged aeration however reduces both lead grade and recovery. The use of barium sulphide as a secondary sulphidizer [4] was examined on oxidized lead ores from Sicily (BaS). The results obtained were encouraging. Sulphidization using Na2S can also be improved with the use of ammonium salts (chloride and sulphate). These reagents are used in cases where the ore contains clay minerals and calcium carbonate, which prevents suphidization due to the production of soluble calcium bicar bonate. The ammonium increases the solubility of calcium carbonate and improves sulphidization.
20.2
Flotation Properties of Individual Oxide Lead Zinc Minerals of Economic Importance
71
100 Cerusite
80
Recovery (%)
Anglesite
60
40
20
0 500
Figure 20.1 oxide ore.
1000 1500 2000 Na2S•9H2O addition (g/t)
2500
Effect of level of Na2S on cerussite and anglesite recovery from carbonate lead
Table 20.3 Effect of aeration on lead oxide recovery from mixed ore Aeration or conditioning time (min)
Conditioning 5 min Aeration 5 min Aeration 10 min Aeration 15 min Aeration 20 min
Na2S (g/t)
1200 1200 1200 1200 1200
Pb oxide cleaner concentrate
Pb oxide rougher concentrate
Grade
Recovery
Grade
Recovery
47.3 51.5 60 61 55
45.5 55.0 66.8 70 50.5
28 30 33 35 30
60 70 85.5 90 68.6
Choice of collector During oxide lead flotation, the choice of collector is rather limited to xanthates, which are used in operating plants. Dithiophosphates and mercaptans are used as secondary collectors. This is due to the fact that natural ores contain a variety of floatable gangues, for which the anionic flotation process is not applicable. The use of chelating agents as flotation collectors for oxide lead flotation have been extensively examined [5,6]. Oximes/fuel oil
72
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
and modified mercaptans have been proven to be good collectors for cerussite and anglesite in laboratory testing. However, these collectors have not been used at an industrial scale. Wulfenite and minetizit are mainly associated with oxide lead silver ores. Very little to no research data are available on these two minerals. Most recently, research work was carried out on these minerals using natural ore. It has been demonstrated that wulfenite and minetizit can be recovered using sulphidization and modified xanthate. 20.2.3
Flotation properties of oxide zinc minerals
The initial research work on oxide zinc flotation dates back to the 1930s at the University of Liege, Belgium [7]. This research work was based on the earned experience of flotation of oxide copper ores. Fatty acids, which are good collectors for copper carbonates, were not applicable because the gangue minerals in the zinc oxide ores were also floatable in the presence of fatty acids. Research work conducted at the University of Liege was carried out on several ore types that contained hemimorphite and smithsonite. The mercaptan-type collectors were effective with both minerals in the presence of sulphidizer. Although mercaptans floated oxide zinc minerals, the recoveries were not satisfactory. Later in the research work, amine-type collectors were introduced to flotation of oxide zinc minerals with significant improvement in metallurgy over that obtained with mercaptans. Extensive research work was performed on Gorno Calamine ore from Sicily, where a large number of amines were examined [8]. Table 20.4 shows the type and performance of the different amines evaluated. Table 20.4 List and performance of the amines examined on the Gorno zinc oxide ore Manufacturer
Commercial name
Results Unfavourable
Flot Ore Company (England) Cytec, USA Akzo Noble, USA
Hoechst, Germany Akzo Noble, USA
Flotbel 1 Flotbel CA Flotbel CH-A Aeroamine Armac T Armac S Armac CD Armac 12D Armac HTD Duomac S Duomac C Flotigan O Flotigan S Diamine DHPL Diamine acetate Kokoamine
Moderate
Good
Very good
× × × × × ×
×
× × × × × × × × × ×
20.2
Flotation Properties of Individual Oxide Lead Zinc Minerals of Economic Importance
73
The results demonstrated that the most satisfactory results were obtained using amines made from coconut oil, whereas the amines made from soya gave poor results. The most effective amines found during this research were Armac C (Akzo Nobel, USA) and Kokoamine KK (Akzo Nobel, USA). The combination of Na2S and amine was then introduced into the Gorno plant in 1954. The pH during flotation was found to be very important. For flotation of calamine, the optimum pH was 10.5. In the presence of willemite–hemimorphite, the pH with RNH2 was about 11.5. The effect of pH on zinc oxide flotation in the presence of amine is illustrated in Figure 20.2. Most recently, a new line of collectors [9] have been studied. These collectors are based on xanthated fatty acid (collectors from the DS series (Xanthate = 60%, Fatty Acid = 20% and Fuel Oil = 20%)) and xanthated mixtures of fatty acid and amine (DAS series (Xanthate = 50%, Fatty Acid = 20%, Amine Oxide = 20% and Fuel Oil = 10%)). These collectors were examined on a number of oxide lead zinc ores from Egypt, South America and Canada. Typical results obtained with the new collectors are presented in Figure 20.3. The best results were obtained using collector DAS-2, consisting of xanthated fatty acid + amine. The reagent was prepared by ultrasonic agitation of xanthate (50%), fatty acid (25%) and amine (25%). The mixture is a yellow-coloured solution. The choice of gangue depressants is closely related to the gangue type present in the ore. Usually the oxide zinc ores have a complex gangue composition and, therefore, selection of depressants is dictated by the gangue composition present in the ore. The most common depressants used include Sodium silicate Guar Starch 100
Zinc rougher recovery (%)
80
60
40
20
0 5
Figure 20.2
6
7
8 9 Flotation pH
10
11
12
Effect of pH on mixed oxide zinc flotation using amine (Armac C).
74
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
100
Collector DAS 1
Zinc rougher recovery (%)
80
DAS 2 60
40
Amine (Armac C) Xanthate
20
0 50
Figure 20.3
100
150 200 Collector additions (g/t)
250
300
Effect of type of collector on flotation of mixed oxide zinc ore.
Cellulose Different acrylic polymers Phosphates (Na2HPO4, HM phosphates). The application of these depressants was discussed in the preceding sections.
20.3 PRACTICES IN THE BENEFICIATION OF MIXED AND OXIDE LEAD ZINC ORES There are only a few operating plants that treat mixed and oxide lead zinc ores. Most of these plants are located in Sicily, Ireland, Morocco, Egypt and Canada, but there are operating plants found in the USA, Mexico and Argentina. In the past 20 years, extensive research has been conducted on a variety of oxide lead zinc ores with the objective of developing a commercial treatment process. Some of these processes are commercially applied in several operations. 20.3.1 Reagent scheme and plant practice for beneficiation of mixed sulphide oxide ores The beneficiation of mixed lead zinc sulphide oxide ores is a complex process and is dependent on gangue composition of the ore. There are two basic types of mixed sulphide oxide ores that have been extensively studied. These include (a) ores with dolomitic and
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
75
calcite gangue present in the ore with moderate clay content and (b) ores with the presence of borite and clay minerals. Treatment of mixed lead zinc sulphide oxide ores with dolomitic and calcite gangue These ore types are abundant in Northern Canada and Mexico. Extensive research work was carried out on these ores with the objective of developing a treatment process for commercial application. One of the major tasks in the development testwork was to use a non-cyanide flotation method in the sulphide flotation. Typical reagent schemes used in beneficiation of the mixed lead zinc oxide sulphide ores are shown in Table 20.5. The compositions of the reagent mixtures are as follows: Depressant MQ3
Depressant MKF
Collector PAX/Armeen C
70% 20% 10% 60% 20% 20% 44% 44% 12%
ZnSO4 T 7H2O Na2S2O3 Na2S2O5 Na2SiO3 Acumer 9000 Thiourea potassium amyl xanthate Armeen C Ethofat 242/12 (emulsifying agent)
The flowsheet used for beneficiation of mixed oxide sulphide ore is illustrated in Figure 20.4. The continuous plant simulation results obtained in the laboratory are shown in Table 20.6.
Table 20.5 Reagent schemes used for mixed lead zinc oxide sulphide ores [10] Reagent
Additions (g/t) Pb sulphide circuit Pb oxide circuit Zn sulphide circuit Zn oxide circuit
Na2CO3 Na2S T 9H2O MQ3 MKF Sodium silicate ‘N’ CuSO4 DV177 (acrylic acid) R241 (Cytec) SIBX 3894 (Cytec) Armeen C/PAX emuls MIBC
1500–1800 500 600 200
1500–2000 800
1700 1500
700 18 28
450 6 70
1200 100
1500
70 16 125
4
76
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
Ore Feed
Grinding
Lead sulphide scavenger
Lead sulphide rougher
Lead sulphide cleaner scavenger
Lead sulphide st 1 cleaner
Conditioner 1
Lead sulphide nd 2 cleaner
Conditioner 2 Conditioner 3 Zinc sulphide rougher
Zinc scavenger
Zinc sulphide st 1 cleaner
Zinc sulphide cleaner scavenger
Lead sulphide rd 3 cleaner
Lead sulphide cleaner concentrate
Zinc sulphide 2nd cleaner
Zinc sulphide rd 3 cleaner
Zinc sulphide cleaner concentrate
Air
Air
Aeration
Aeration
Lead oxide rougher
Aeration Lead oxide scavenger
Lead oxide st 1 cleaner
Lead oxide 2nd cleaner
Lead oxide cleaner concentrate
Desliming Conditioning Zinc oxide rougher
Zinc oxide st 1 cleaner
Slimes
Zinc oxide scavenger
Final tailings
Zinc oxide nd 2 cleaner
Zinc oxide rd 3 cleaner
Zinc oxide cleaner concentrate
Figure 20.4
Typical flowsheet for beneficiation of mixed sulphide oxide lead zinc ores.
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
77
Table 20.6 Continuous locked cycle test results from mixed lead zinc sulphide oxide ore Product
Pb sulphide concentrate Pb oxide concentrate Pb sulphide + oxide concentrate Zn sulphide concentrate Zn oxide concentrate Zn sulphide + oxide concentrate Zn final tail Feed
Weight (%)
16.77 10.11 26.88 23.28 13.67 36.95 36.27 100.00
Assays (%, g/t)
% Distribution
Pb
Zn
Ag
Pb
Zn
Ag
72.5 48.6 63.5 3.56 3.81 3.67 1.14 18.8
5.28 8.63 6.25 60.5 31.1 50.0 3.14 21.2
1002 312 757 163 138 14.3 16.9 263
64.5 26.1 90.6 4.4 2.8 7.2 2.2 100.0
4.2 4.1 8.3 66.3 20.0 86.7 5.4 100.0
69.0 8.5 77.5 13.5 6.7 20.2 2.3 100.0
Treatment of mixed lead zinc sulphide oxide ores with barite–calcite gangue minerals There are only few operations treating mixed lead zinc sulphide oxide ores that contain barite–calcite gangue minerals. A typical example of such an operation is the Tynagh oxide complex in Ireland [11]. In this deposit, the oxide ores are generally located at the bottom and at the ends of the sulphide mud ores. The major gangue mineral is barite (large quantities) and minor amounts of clay. This ore assays 8.5% Pb(total), 6% Pb(oxide), 6.8% Zn(total) and 5% ZnO. This ore is treated using a bulk lead zinc sulphide flotation method followed by oxide lead flotation. The reagent scheme used in this circuit is shown in Table 20.7. Cyanide is used to depress pyrite, since some of the ore types contain an appreciable amount of pyrite. The results obtained in the plant are presented in Table 20.8. Table 20.7 Reagent scheme used for treatment of the Tynagh mixed sulphide oxide lead zinc ore Reagent
Additions (g/t) Bulk circuit
Cytec AC633 Lime Sodium cyanide Na-isopropyl xanthate Na-hydrogen sulphide K-amyl xanthate Copper sulphate Sodium silicate
150 560 30 90
Lead oxide circuit
100 1300 160
150 330
78
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores Table 20.8 Plant metallurgical results – Tynagh mixed ore Assays (%)
Product
Bulk Pb/Zn concentrate Oxide Pb concentrate Tailings Feed
% Distribution
Pb (total)
Pb (oxide)
Zn (total)
Zn (oxide)
Pb (total)
Zn (total)
25.70 45.10 3.39 8.46
– – 2.70 6.20
19.0 – 4.51 6.83
– – 3.80 5.10
47.5 20.1 32.4 100.0
43.5 – 56.5 100.0
Practices in beneficiation of oxide lead zinc ores There are several operations that treat oxide lead zinc ores, including several operations in Sicily, Morocco and Mexico. The beneficiation flowsheets, in general, are similar to those found in most operating plants. The generalized flowsheet is presented in Figure 20.5. A possible difference in the flowsheet might be the addition of a gravity preconcentration stage for lead. In some cases, the heavy liquid separation is performed prior to lead zinc oxide flotation. The reagent scheme selection is very dependent on (a) type of gangue minerals present in the ore and (b) type of lead and zinc oxide minerals. In plant practice, lead oxide minerals are recovered using a sulphidization method with xanthate as the primary collector and mercaptans as the secondary collector. The zinc oxide flotation is performed after desliming the oxide lead tailings. In the case of zinc oxide, a sulphidization method is used with amine as the primary zinc collector. The choice of depressant is dependent on the type of gangue minerals present in the ore. A typical reagent scheme used for flotation of oxide lead zinc ores is shown in Table 20.9. Collector sterylamine acetate works well on smithsonite, but not as well on calamine. When smithsonite is present in the ore, better results are achieved using a tallow amine emulsion with elevated additions of fuel oil emulsion. Metallurgical results obtained in an operating plant in Morocco are presented in Table 20.10. An appreciable amount of zinc reported to the slime fraction. Attempts were made by some researchers to float oxide zinc from the slimes, but with little success. 20.3.2
Practices in beneficiation of oxide zinc ores
The zinc oxide ores of any economic value are represented by smithsonite and calamine. Willemite, franklinite and other zinc oxide minerals are quite rare. The gangue minerals are usually represented by calcite ferooxides, dolomite and hemimorphite. The composition of gangue minerals, however, varies considerably and may also contain clay, talk, Fe-hydro xide and other minerals. Over the past 30 years, considerable work has been conducted on zinc oxides, for which a number of different methods were developed and examined in a number of different plants. The highlights of the most important methods are presented below.
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
79
Ore Feed
Grinding
u/f
Classification
Conditioning
Lead oxide rougher
Lead oxide scavenger
Lead oxide 1st cleaner
Slimes Desliming
Conditioning 1 Conditioning 2
Lead oxide 2nd cleaner
Zinc oxide rougher
Zinc oxide scavenger
Zinc oxide 1st cleaner
Zinc oxide cleaner scavenger
Lead oxide 3rd cleaner
Lead oxide cleaner concentrate
Zinc oxide 2nd cleaner
Zinc oxide 3rd cleaner
Zinc oxide tails
Zinc oxide
cleaner concentrate
Figure 20.5
1.
Generalized flowsheet used in beneficiation of oxide lead zinc ores.
Fatty acid flotation method. In this method, the smithsonite was floated using a short carbon chain fatty acid from calcitic and dolomitic gangue minerals. The calcite and
80
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores Table 20.9 Reagent scheme used in beneficiation of oxide lead zinc ore Additions (g/t)
Reagent
Depressants and modifiers Soda ash Sodium silicate ‘N’ Carboxymethyl cellulose Low-molecular-weight acrylic acid Na2S T 9H2O Collectors Potassium amyl xanthate Mercaptan (Cytec’s R407) Sterylamine acetate Fuel oil emulsion
Lead oxide
Zinc oxide
700–1000 (pH 10) 900–1100 – 300–500 800–2000
– 500–800 200–600 – 1500–3000
80–150 10–30 10–30 –
– – – 300–600
Table 20.10 Plant results from a Morocco lead zinc oxide ore Product
Lead concentrate Zinc concentrate Zinc tailing Slimes Feed
2.
3.
Weight (%)
7.78 15.18 64.34 12.70 100.00
Assays (%)
% Distribution
Pb
Zn
Pb
Zn
55.3 3.6 0.67 3.30 5.70
4.60 38.9 1.25 11.66 8.77
75.5 9.6 7.6 7.3 100.0
4.1 70.1 9.0 16.8 100.0
dolomite were depressed with citric acid. This method has not found application in any plant practice. The reverse flotation method. It was examined by several Russian scientists. This method involves depression of smithsonite using sodium phosphate and dextrin. Calcite and dolomite are floated using oleic acid in stage additions. The tailings become the zinc concentrate. Oxide zinc minerals activation using CuSO4. The use of CuSO4 for zinc oxide activation and flotation is based on two different processes: (i) Pretreatment of the pulp with sulphidizer followed by reconditioning with CuSO4 and zinc oxide flotation with xanthate or xanthate + mercaptan. (ii) The second method involves sulphidization of the smithsonite during heating of the pulp to 50–60°C followed by activation with CuSO4 and flotation of the activated oxide zinc using xanthate and/or an Aeroflot type collectors.
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
81
The CuSO4 activation method produces good zinc oxide metallurgical results when the ore contains dolomitic/calcitic gangue minerals. If the ore contains Fe-hydroxides, this method, however, produces poor results. 4.
Sulphidization – amine flotation. This method was extensively examined dating back to the 1940s and early 1950s [11], and currently is being used in almost all industrial plants treating oxidized zinc ores. Amines suitable for flotation of oxidized zinc minerals include salts of aliphatic amines, octadecyl amine and hydrooxyinolin.
Summary of collectors used for oxide zinc flotation A list of collectors used in the oxide zinc flotation is presented in Table 20.11. In every method, sulphidizer Na2S is used. The consumption of Na2S largely depends on the type of gangue mineral present in the ore, and ranged from 2000 to 5000 g/t. Ores containing clay slimes or Fe-hydroxides require relatively high levels of additions. Xanthate consumption ranges from 200 to 700 g/t, and mercaptan consumption, when used as a secondary collector with xanthate, ranges from 50 to 100 g/t. In general, mercaptans are usually used with xanthates. In the cases where CuSO4 is used as an activator, mercaptan is used as a primary collector. Typical amine consumptions ranged from 80 to 150 g/t. Amine is a fast-acting collector and does not require conditioning time. Summary of modifiers and depressants used for oxidized zinc flotation There are a fairly large number of modifiers and depressants that are used for flotation of oxidized zinc minerals. Table 20.12 lists some of the most commonly used depressants. In actual practice, sodium silicate is most commonly used because of its depressing action on most of the gangue minerals present in the ore. Sodium tripolyphosphate acts as a general depressant and dispersant in concentrations up to 250 g/t. At higher dosages, it acts as an activator. Guar of methyl carboxylic sodium salt is an extremely effective depressant for alumina, silica and calcite in concentrations even below 500 g/t. This depressant together with sodium silicate is used in the treatment of oxidized zinc ores that contain aluminosilicates, dolomite and silicates, as the principal gangue minerals.
Table 20.11 Collectors used in oxide zinc flotation Reagent
Commercial name
Zinc mineral
Xanthate
Potassium amyl xanthate Sodium isobutyl xanthate Cytec’s R407, R404 Clariant’s M91 Salt of aliphatic amines Primary octdecyl amine Hydro oxyxinolin
Smithsonite Smithsonite, ancite Smithsonite Smithsonite, willemite Calamine Calamine, clinohedrite Calamine, willemite
Mercaptan Amine
82
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores Table 20.12
Summary of the most common depressants used for oxide zinc flotation Commercial name
Chemical name
Molecular weight
Na2SiO3 Na5P3O10 Suspendol PKK Suspendol PC CMC Acrol TR30 Acrol J25 250 Cataflot P40 DV177 Starch Accumer 9000
Sodium silicate Tripolyphosphate Phenol condensation, anionic Polycarboxylic acid Carboxymethyl cellulose Anionic modified guar Anionic modified guar gan Acrylate Acrylic acid Starch of methylcarboxilic salt Modified acrylate
– – High Medium Medium Low Low Low Low High Low
Acrylate and low-molecular-weight acrylic acids are effective slime depressants/disper sants. Suspendol PKK and PC improved separation efficiency between zinc and iron oxides. Starches, mainly hydrolysed starch and cooked starch, are effective depressants in the treatment of oxidized zinc ores that contain iron oxides and Fe-hydroxides. Reagent schemes used to treat different oxidized zinc ores As mentioned earlier, the selection of reagent scheme for treatment of oxidized zinc ores depends very much on the type of oxide zinc mineral present in the ore, as well as the type and consumption of gangue minerals. Table 20.13 shows some of the most commonly used Table 20.13 Reagent schemes used for beneficiation of oxide zinc ores Ore type and description
Reagents used
Predominant ZnO mineral = smithsonite (grade 9–15% Zn) Gangue minerals are composed of calcite, dolomite and silica and often contains clay minerals (i.e. kaoline, montmorillanite) Predominant ZnO mineral = calamine (grade 4–10% Zn) Ore gangue geothite, silica, dolomite with or without Fe-hydroxide Mixed calamine/smithsonite (grade 8–15% Zn) This ore type has a relatively simple gangue composition. Silicate, goethite and barite are the principal gangue minerals. Complex calamine/smithsonite ore (grade 6–12% Zn) This ore type is found in South America (Peru). The major gangue minerals are goethite, Fe-hydroxides, dolomite, siderite and clay mineral.
2000–2500 g/t sodium silicate; 3500–4500 g/t Na2S T 9H2O; 300–500 g/t DV177; 100–150a g/t amine; 10–20 g/t pine oil (frother)
a
1500–2500 g/t N2SiO3 (type N); 4000– 5000 g/t Na2S T 9H2O; 300–500 suspendol PC; 90–150 g/t amine/pine oil/fuel oil (80:10:10) emulsion 1000–1500 g/t Na2SiO3 (type N); 3000–5000 Na2S T 9H2O; 200–300 g/t Na5P3O10; 200– 300 g/t CMC; 150–180 g/t. Amine/pine oil/ fuel oil (70:10:20) emulsion 1000–1500 g/t Na2SiO3 (type N); 500–1500 caustic starch; 3000–6000 g/t Na2S T 9H2O; 200–300 g/t Na5P3O10; 250–300 g/t amine: xanthate (1:1) emulsion
Typical caly minerals, which determines the type of dispersant to be used.
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
83
Table 20.14 Continuous locked cycle test results from a refractory zinc oxide ore from Peru Weight (%)
Product
Zn cleaner concentrate Zn combined tail Slimes Feed (calc)
18.00 60.63 21.37 100.00
Assays (%)
% Distribution
Pb
Zn
CaO
Fe
Overall
Flotation
0.72 0.36 0.62 0.49
32.2 2.52 11.3 9.74
7.09 27.8 22.0 22.9
3.18 19.8 12.4 15.2
59.5 15.7 24.8 100.0
79.1 20.9 – 100.0
reagent schemes developed for treatment of various zinc oxide ore types. The most difficult to treat ores are those with high clay and high Fe-hydroxide contents. In these cases, a large portion of the zinc is lost to the slime fraction. The results obtained on this type of ore (i.e. Acha ore from Peru) are presented in Table 20.14. Good metallurgical results were usually achieved on the smithsonite and calamine ores. 20.3.3
Flotation of oxide lead silver ore
Oxide lead silver ores are very rare and there is only one operating plant in the world treating this type of ore. The process development and plant design was accomplished during 2005–2006 and has been in operation since December 2007 [12,13]. There are several ore types present in the ore. The ore The principal lead oxide minerals include pyromorphite, wulfenite, mimetite and plum bojerusite. Some galena is also present in this ore type. The principal gangue minerals include silicate, dolomite, siderite, ferohydrooxides and clay minerals. About 20% of the ore is represented by ultra-fine slime with an average size of K80 = 12 µm. These slimes are composed of Fe-hydroxides, kaolin and slimes of plumbojerusite, rich in silver. Processing characteristics of the ore Silver oxide lead ores have much different flotation processing characteristics. Although this ore responds to sulphidization–xanthate system, silver recovery in the lead concentrate was usually poor and amounted to about 30–40%. Floatability of lead minerals also was not satisfactory. Most of the lead and silver losses occurred in a finer fraction (i.e. −200 mesh). It should be noted that the ore contains an appreciable amount of gold. Electron probe microscopy indicated that finer particles of oxide lead and silver were coated with ultra-fine slimes of Fe-hydroxide.
84
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores
Research and development Research and development concentrated on developing an effective reagent scheme and flowsheet that would produce satisfactory metallurgical results. The problem encountered during the development testwork was that no literature was available for flotation of a mixture of pyromorphite, wulfenite, mamitizite and plumbojerusite. In addition, some of the silver minerals present in the ore included argentojarosite, diaphovite and fizelyte. During the development work, a fairly large number of reagent combinations were examined, mainly collector type and decoating reagents to improve lead–silver floatability. It was discovered that using CuSO4 in the sulphide prefloat and in the oxide lead flotation together with Na2S improved silver and lead metallurgical results significantly. Figure 20.6 shows the effect of CuSO4 on silver and lead oxide results. It appears that CuSO4 is responsible for decoating the lead and silver mineral surfaces from clay minerals. The conditioning time with sulphidizer was found to be critical in achieving good lead and silver metallurgical results. Figure 20.7 shows the effect of conditioning time on lead–silver rougher recoveries. Prolonged conditioning times had a negative effect on lead and silver recoveries. Plant reagent scheme The development testwork yielded a reagent scheme that is currently being used in the plant. This reagent scheme is shown in Table 20.15.
80 Lead
Recovery (%)
70
60 Silver 50
40
30 0
Figure 20.6
50
100 150 CuSO4 addition (g/t)
200
250
Effect of CuSO4 additions on oxide lead silver recovery.
20.3
Practices in the Beneficiation of Mixed and Oxide Lead Zinc Ores
85
80
Lead
Recovery (%)
70
Silver
60
50
40
30 0
5
10
15
20
25
Conditioning time (min)
Figure 20.7
Effect of conditioning time with Na2S on lead and silver recoveries.
Table 20.15 Reagent scheme used for beneficiation of oxide lead silver ore (Peru operation) Reagent
Sodium silicate ‘N’ Polyacrylic acid (DV177) Na2S T 9H2O Modified xanthate Kerosene CuSO4
Additions (g/t)
Ag prefloat
Pb–Ag circuit
600 200 – 100 – 200
1100 300 5000–6000 800–1100 300–400 –
The use of CuSO4 was extremely beneficial for improvement in silver recovery con tained in the fine fractions. Plant metallurgical results The plant metallurgical results obtained during 6 months of operation are shown in Table 20.16.
86
20.
Flotation of Mixed Lead Zinc Sulphide Oxide and Oxide Lead and Zinc Ores Table 20.16 Plant results – January to June 2008
Product
Pb/Ag concentrate Pb/Ag tailings Feed (calc)
Weight (%)
16.34 83.66 100.00
Assays (%, g/t)
% Distribution
Pb
Zn
Ag
Pb
Zn
Ag
48.5 3.07 10.5
1631 172 410
10.7 2.1 3.5
75.5 24.5 100.0
65.0 35.0 100.0
50.0 50.0 100.0
This ore type is unique and there are only two known deposits of this kind, both of which are located in South America. REFERENCES 1. Ray, M., Formanek R., and Chataignon, A., The Influence of Certain Inorganic Salts on the Flotation of Lead Carbonate, Transactions of the American Institute of Mining Engineers, p. 185 Nov 1950 (Mining Engineering). 2. Wark, I.W., Principals of flotation, Melbourne Institute of Mining and Metallurgy, pp. 334–342, 1938. 3. Ray, M., Flotation of oxide lead zinc and copper ore, Institute of Mining and Metallurgy, Vol. 4, pp. 541–548, Sept 1954. 4. Fleming, M.G., Effect of Alkalinity on the Flotation of Lead Minerals. Transactions of the American Institute of Mining Engineers Vol. 193, pp. 1231–1236, 1952. 5. Usoni, L., Rinelli, G., and Marabini, A.M., Chelating Agents and Fuel Oil: A New Way to Flotation, AIME General Annual Meeting. New York, USA;1971. 6. Marabini, A.M., and Rinelli, G., Flotation of Pitch Blende with Chelating Agents and Fuel Oil, Transactions of the Institution of Mining and Metallurgy, No. 82, pp. C225–C228, 1973 7. Ray, M., Fatty Acids and Soap Flotation Applied to Oxidized Ores, Engineering and Mining Journal, No. 136, pp. 221–223, May 1935. 8. Billi, M., and Quai, V., Developments and Results Obtained in Treatment of Zinc Oxide Ore at the AMMI Mines, Interim Report, May 1976. 9. Bulatovic, S., New Line of Zinc Oxide Collectors has been Development, Technical Note, No. 6, June 1992. 10. Bulatovic, S., Process Development for Treatment of Pb/Zn of Oxide Sulphide Ore from Prairie Creek (Canada), Report of Investigation, January 2006. 11. Morrissey, C.J., and Whitehead, D., Origin of the Tynah Residual Orebody, Ireland, Proceedings of the 9th Commonwealth Mining and Metallurgical Congress, London UK, May 1969. 12. Bulatovic, S., Process Development for Beneficiation of Oxide Lead, Silver, Gold Ore for Minera Corona, Peru, SBM Report of Investigation, May 2007. 13. Bulatovic, S., Plant Design and Start-up for Beneficiation of Oxide Lead, Silver, Gold Ore for Minera Corona, Peru, SBM Report of Investigation, November 2007.
– 21 –
Flotation of Tin Minerals
21.1
INTRODUCTION
Beneficiation of tin ores exclusively includes a combination of gravity preconcentration and flotation, where the flotation includes (a) flotation of sulphides from the gravity concentrate and (b) flotation of sulphides from the fines, followed by tin flotation from the sulphide tailing after desliming. Tin treatment processes are relatively complex and include a fairly large number of unit operations, such as (a) staged gravity concentration, (b) retreatment of gravity concentrates using regrinding and sulphide flotation from the gravity tailings and (c) desliming of the fines and staged sulphide tin flotation. This chapter discusses in detail the beneficiation processes for various tin ores.
21.2
MINERAL COMPOSITION OF VARIOUS TIN ORES
The most important tin mineral is cassiterite (SnO2). Theoretically, the tin content of cassiterite is 78%. However, in the majority of cases, cassiterite contains impurities and the tin content may vary from 65% to 78%. The major impurities of cassiterite include tantalum, niobium, titanium and other elements, usually in the form of solid solutions. The impurities in the cassiterite often have a pronounced effect on flotation properties of cassiterite. Based on studies [1] of a number of tin ore bodies, tin can be classified into three major groups: Group 1:
Group 2: Group 3:
Cassiterite contained in pegmatitic veins contain significant quantities of (Nb, Ta)2O5 with traces of wolframite and manganese. This type of cassiterite is fragile and tends to slime during grinding. Cassiterite from quartz veins. Cassiterite from this group contain about 1% (Nb,Ta)2O5 and about 0.3–0.4% wolframite. Cassiterite from sulphide veins usually contains vanadium, sulphur and wolframite.
87
88
21.
Flotation of Tin Minerals
Table 21.1 List of tin-containing minerals Mineral
Formula
Theoretical content (% Sn)
Specific gravity
Colusite Herzen bergite Sulphostannati Canfieldite Franckelite Teallite Cilindrite Silicates Arandesit Stokezit Boratin Nordenskioldine Thoreaulite
Cu3 (As, Sn, V, Fe, Te) S4SnS SnS
5.0–8.9 78.8
4.4–9.6 –
Ag8SnS6 PbS-SnS2 5PbS-2SnS2-SbS3 Pb3Sn4-Sb2-S14
10.0 9.4–17.3 30.5 26.6
6.3 3.5–5.5 6.4 5.4
3SnSiO4-2SnO2-4H2O H2CaSnSi3O11 SnO2-6B2O3-5H2O CaSn(BO2)3 Ta2O5-SnO2
48–55 26–43 5.5 42 15–17
4.0 3.2 4.3 4.2 7.6
The second most important tin mineral is stanin (CuS-FeS-SnS2). Theoretically, stanin contains 27.5% Sn, 29.9% S, 29.5% Cu and 13.1% Fe. Other minerals that contain tin (Table 21.1) do not represent any significant economic value.
21.3
BRIEF DESCRIPTION OF TIN DEPOSITS
The most important tin deposits are hydrothermal deposits (hypothermal and mesothermal). The magmatic deposits do not often contain tin mineralization. Tin may also be present in pegmatitic ore bodies. However, tin found in pegmatitic deposits can be classified into two basic types: (a) quartz–cassiterite lenses in granite, when cassiterite is associated with topaz, beryl and, to a lesser degree, sulphides; (b) sulphide deposits, where tin is mainly cassiterite associated with arsenopyrite, pyrite, chalcopyrite and pyrrhotite. Such deposits are common in South America (Peru, Bolivia). Because cassiterite is a stable mineral and does not tend to decompose, it forms sand deposits by decomposition of pegmatitic and quartz–cassiterite deposits. Such deposits are common in Asia. Based on the degree of dissemination, the tin-bearing deposits can be classified into three distinct groups: 1.
2.
Disseminated deposits. In these deposits, the cassiterite grains range from 0.2 to 0.001 mm. Cassiterite is mostly dispersed in gangue matrix of alumosilicates, tourmaline and quartz. Recovery of cassiterite from these deposits is quite difficult. Medium-coarse-grained, less-disseminated ores. The size of cassiterite particles in this ore type ranges from 1.0 to 0.2 mm in size. Typical ore bodies of this type are pegmatitic and tin-containing sulphide ores.
21.4
3.
Beneficiation of Tin Ores
89
Coarse-grained tin ores. The average grain size of cassiterite in this ore type ranges from 0.1 to 1 mm and higher.
Depending on the composition of disseminated and medium-coarse-grained ore, they can be divided into two basic groups: sulphides and chloritic tourmaline ores. In the sulphide ore, the minerals are represented by pyrite, pyrrhotite, arsenopyrite, chalcopyrite, galena and stannin. Less common are sphalerite and bismuth. The chloritic tourmaline ore type contains significant quantities of ferrosilicates, tourma line and chlorites. The tin content in this ore type is usually high compared to the other ore types and can reach over 2% Sn in the ore. The coarse-grained tin ores are usually represented by cassiterite–quartz and pegmatitic formations. These ores can be a complex formation containing varieties of gangue miner als. The pegmatitic ore type, in addition to tin, can contain significant amounts of tantalum and niobium. The world’s major tin deposits are elongated zones over 2000 km, extending from Indonesia to Malaysia, Thailand and Burma and into China. Commercial production of tin is almost exclusively from placer deposits, and these are major producers in the world. Other placer deposits worth mentioning are Brazil, Nigeria and the Congo. For beneficiation of tin from these deposits, the physical concentration method is used exclusively. Hard rock deposits are richer in tin than in the placer deposits, ranging from 0.6% up to 5% Sn. Such known deposits are located in Brazil, Canada, Bolivia, Peru and the USA. Because the tin from these ores is disseminated, beneficiation processes include a combi nation of gravity preconcentration and flotation.
21.4
BENEFICIATION OF TIN ORES
There are three main methods used for beneficiation of tin ores: (a) physical concentration including gravity concentration, magnetic separation and electrostatic separation; (b) flota tion and (c) a combination of gravity preconcentration and flotation. The physical concentration is primarily used in beneficiation of alluvial and some coarser grained vein deposits. The combination of gravity and flotation is normally used for beneficiation of hard rock ores. Flotation is only employed for beneficiation of disseminated tin ores. 21.4.1
Gravity beneficiation method
In the early 1950s and 1960s, a large portion of tin was produced from alluvials. The tinbearing alluvials are usually washed and sized, and the various size fractions are concen trated by gravity using jigs and tabling. Nowadays, significant progress has been made to improve gravity equipment and develop new equipment, such as Mozley drum separators designed to concentrate finer tin fractions. A generalized gravity concentration flowsheet is shown in Figure 21.1.
90
21.
Flotation of Tin Minerals
Dredged sand
Grizzly
Oversize + 8 inch
Undersize - 6 inch Trommel
Oversize + 1/2 inch
Undersize - 1/2 inch Scrubbing
Desliming
Primary jig
Slimes overflow Tail
concentrate Secondary jig
Tail
concentrate
Concentrate de-watering
Tabling
Final concentrate to dressing or drying plant
Figure 21.1
Middling
Tailings disposal
Generalized gravity concentration flowsheet.
The gravity concentrate may be finely upgraded by magnetic separation, electro static separation and flotation. Magnetite present in the gravity concentrate is removed by low-intensity magnetic separation (2000–3000 G magnetic field strength) and electrostatic upgrading if the concentrate contains zircon. Flotation is added to remove sulphides from the tin concentrate. The cassiterite concentrate produced by gravity after final ‘dressing’ is usually high grade, ranging from 50% to 70% Sn with relatively good recovery.
21.4
Beneficiation of Tin Ores
91
The cassiterite ore from underground, hard rock veins is finer grained than alluvial deposits. From this ore, good concentrate grade is produced but at relatively low tin recoveries. Fines from the gravity tailing assay between 0.4% and 0.8% Sn. The placer deposits are more efficiently mined by dredging. In some operating plants, scrubbing the ore is practiced before tin concentration. 21.4.2 Combination of gravity–flotation tin beneficiation method (lodge deposits) The so-called hard rock lodge deposits are much richer in tin than placer deposits, but nowadays, these deposits contain �0.4–1.5% Sn, as mined. The rich lodge deposits, which are treated exclusively by gravity, are mined out. Today’s low-grade deposits contain fine-grained, highly disseminated cassiterite that is much more difficult to process. Tin losses to the fines (<200 µm) are the major problem. Recovery from such an ore does not exceed 50% Sn when using a gravity method. By using a combina tion of gravity and flotation, tin recoveries improved up to 70% Sn in a number of operating plants. Tin concentration from these deposits is further complicated because these deposits contain sulphides, and occasionally tin-bearing sulphides. These sulphides have to be removed from the final tin concentrate. Some ores (New Brunswick, Canada) contain wolframite, which is removed by either flotation or high-intensity magnetic separation. The complex Lodge deposits have quite a complicated flowsheet. Typical examples include operating plants in Australia and South America. An example of a plant that uses both gravity and flotation is the Wheal Jane Concentrator in Cornwall, UK, shown in Figure 21.2. The Cornwall ore assays 1.26% Sn, 2% Zn and 0.4% Cu. Three concentrates are produced using this flowsheet, including (a) Zn/Cu concentrate assaying �30% Zn and 5% Cu, (b) tin gravity concentrate assaying 30–40% Sn and (c) tin flotation concentrate that assays �14% Sn. In general, the tin concentrate produced from this type of ore is lowgrade concentrate. Most recently, with the help of advanced technology, a high-grade tin concentrate can be produced. This is discussed in the following sections. 21.4.3 Flotation Introduction The chemistry of cassiterite flotation has been a subject of considerable research for many years. The findings that sulphosuccinamates, phosphonic acid and arsonic acid were selective collectors for cassiterite flotation lead to the introduction of flotation as a complementary recovery process to gravity concentration at most primary tin mill concentrators in the early 1970s. In spite of continued research, subsequent progress in development has been rather limited. Cassiterite flotation still remains a secondary tin recovery process in most plants, for beneficiation of cassiterite below 40 µm size.
92
21.
Flotation of Tin Minerals
Ore Feed Classifier
Cyclones Regrind
sand
Grind
Desliming 2
slimes
sand
Screen
Conditioner Desliming 1
AA
Copper/zinc scavenger
sand Thickener
Hydrosizer
Conditioner
Shaking tables
tail
to AA
con
Bulk sulphide flotation
Regrind
Low intensity magnetic separator
Conditioner
mag's
non-mag's
Regrind Copper/zinc flotation
Conditioner
Tin conditioner Tin rougher flotation
to Tin Gravity Concentrator
Copper/zinc flotation
Copper/zinc concentrate
Tin cleaner flotation FeS, FeAsS to tails
High intensity magnetic separator Tin flotation concentrate
mag's Final tail
Figure 21.2 Combination of gravity–flotation beneficiation flowsheet (simplified), Wheal Jane Concentrator, Cornwall, UK.
The lack of progress in cassiterite flotation can be attributed to a number of factors, some of which are described below: • The chemical complexity of the pulp, which affects cassiterite surface properties and, in turn, affects collector adsorption. • The type of gangue minerals present, i.e. the selective flotation of cassiterite depends on the proper selection of depressant for certain gangue minerals. • Understanding the interaction of tin collectors with chemical parameters associated with pulp chemistry and adsorbtion of cassiterite is quite difficult to reconcile with actual plant practice.
21.4
Beneficiation of Tin Ores
93
• It should be pointed out that the results obtained under laboratory conditions often do not work when implemented into operating plants. Tin collectors and chemistry There are only a few collectors suitable for tin flotation that have been introduced into operating plants in the 1970s, but today they have been replaced (i.e. arsonic acid, phospho nic acid) due to toxicity and high prices. Other collectors that have been extensively studied include oleic acid, sodium oleate, alkyl phosphoric acid and hydroxamates [2–4]. More recently, studies were undertaken to recover tin from gravity tailings at the San Rafael tin mine in Peru. During these studies, a highly effective collector mixture [5] consisting of succinamate phosphoric ester and sodiumhylenesulponate was established. This mixture was more effective than sulphosuccinamate alone, or arsonic acid. Arsonic acid collector This collector has the formula shown in Figure 21.3. The identification of p-tolyl arsonic acid as a selective collector for cassiterite flotation led to the introduction of this collector into many industrial plants. The first recorded industrial use of p-tolyl arsonic acid was at the Alterberg mine in Germany. By the early 1970s, this collector was introduced into a number of operations, including Rooiberg and Union Tin (South Africa), the Renison and Cleveland tin mines (Australia). This collector is toxic and because of that, Hoechst discontinued this collector in 1973, and later in 1976, Mitsubishi, the only supplier remaining, ceased production. Most recently, interest has been renewed in the use of methyl benzyl and a mixture of pand o-tolyl arsonic acid as a cassiterite collector [6,7]. The reagent under the trade name MTAA, which consists of approximately 50:50 of p- and o-tolyl arsonic acid, is now exclusively used in cassiterite flotation in the Peoples Republic of China. From a chemical point of view, arsonic acid is a weak dibasic acid. The arsonic acid in flotation is the most effective at a pH range of 3.5–5.0. Above or below this pH range, the effectiveness diminishes. Figure 21.4 shows the effect of pH on tin flotation using arsonic acid. An unusual feature of arsonic acid flotation of cassiterite is the immobility to recover cassiterite coarser than 40 µm in size. The results obtained at the Renison Mine (Australia) indicated that cassiterite recovery in fractions above 20 µm drops sharply (Table 21.2). Induction time measurement on cassiterite particles further explained the difficulties to establish particle/bubble contact with coarse cassiterite. This is not the case with other cassiterite collectors.
OH H3C
As
OH
O
Figure 21.3
Structural formula of p-tolyl arsonic acid.
94
21.
Flotation of Tin Minerals
100
Tin recovery (%)
80
60
40
20
0 0
2
4
6
8
10
12
Rougher flotation pH
Figure 21.4
Effect of pH on tin recovery using p-tolyl arsonic acid. Table 21.2
Size-by-size SnO2 recovery using p-tolyl arsonic acid (Renison mine) Size fraction (µm)
Recovery (% Sn)
2–5 5–15 15–20 20–50
85.0 87.2 50.0 5.5
Phosphonic acid With the development of arsonic acid as a specific cassiterite collector, research was initiated to develop a chemical compound as effective as arsonic acid, but less toxic. Also, arsonic acid production is quite an expensive process. The reason for such a behaviour of arsenic acid is that arsenic is a member of the group 5A elements in the periodic table. Phosphorus and antimony are also group 5 elements and are known to be chemically similar to arsenic. On this basis [8,9], the antimonic acids were found to be poor cassiterite collectors. The alkyl phosphonic acids were not selective collectors. The ethylphenylene phosphonic acid was found to produce similar or better results compared to p-tolyl arsonic acid. The structural formula for phosphonic acid (Figure 21.5) is similar to that of p-tolyl arsonic acid but arsenic was replaced with phosphorus. The styrene phosphonic acid radicals are C6H5–CH–CH and p-ethylphenylene CH3–CH2–C6H4.
21.4
Beneficiation of Tin Ores
95
OH R
CH
CH
P OH O
Figure 21.5
Chemical structure of styryl phosphonic acid.
The phosphonic acids, much like arsonic acid, are weak dibasic acids. In non-polar solvents, phosphonic acids associate with long chains through molecular hydrogen bonding. From the point of view of cassiterite flotation, the adsorption of phosphonic acid on cassiterite increases with decreasing pH and reaches a maximum at pH 2.0 and sharply decreases at a pH below 2.0. Using phosphonic acid as a collector if cations are present in the flotation pulps affects the cassiterite flotation negatively. High iron levels in particular have a strong depressing effect on flotation using phosphonic acid. The phosphonic acid flotation of cassiterite is similar in many ways to that of arsonic acid. Optimum plant flotation for both collectors is generally considered to be in the pH region 4.5–5.5. The phosphonic acid, like arsonic acid, also has the inability to effec tively recover cassiterite coarser than 20 µm. The reason for this is not known, because no research data is available on the flotation properties of plus 20 micron cassiterite particles. A phenomenon observed in both laboratory and pilot plant testing of ores with phos phonic acid collectors is complete cassiterite flotation at a pH below 4.0. In fundamental practice, it indicates that a pH region below 4 is the region of maximum flotation. However, in plant practice, at a low pH (below 4), loss of flotation occurred. The loss of flotation at a low pH has not been established. It is, however, postulated that loss of flotation is believed to be associated with complex solution chemical interaction between phosphonic acid collectors and cationic species, in particular, those of iron, which is always present in industrial flotation pulp. It should be noted that from data produced by several plants, it is indicated that phosphonic acid for many ores is less selective than the arsonic acid. Sulphosuccinamate collectors The sulphosuccinamate surfactants were first synthesized by American Cyanamid Com pany (currently Cytec, USA) in the late 1940s. Later in 1968, the alkyl sulphosuccinamate [10] was patented as a flotation collector. The general structure formula for alkyl sulphosuccinamate is shown in Figure 21.6. The sulphosuccinamate surfactants CA540 (Allied Chemical, USA) and 845 (Cytec) both contain four dissociable groups, three carboxylate and one sulphonate. Both collectors contain impurities such as dodecyl amine, maleic acid derivative and residual alcohols. This may explain the quite strong frothing properties of these collectors. The sulphosuccinamates are not completely soluble in aqueous solution at high con centration. Maximum concentration for Aerosol 22 (Cytec) was found to be about 30% by weight. The sulphosuccinamates are recently used in most operating plants as a
96
21.
Flotation of Tin Minerals
CH2COONa COONa
CH
CH2
HC
N
O
SO3Na
Figure 21.6
C
C18H37
Structural formula for sulphosuccinamate.
replacement for phosphonic and arsonic acids. This is due to the fact that these collectors are non-toxic and are more cost-effective. The only commercial use for arsonic acid is in the Republic of China. Dicarboxilic acids as cassiterite collector In 1980, in search for new collectors for cassiterite, a new collector [11] was synthesized at Freiberg Mining Academy. The investigations were carried out with alkane carboxylic acids with the general structure as shown in Figure 21.7, which was altered by controlled substitution with –COOH and other groups. Although these collectors were effective cassiterite collectors, their selectivity against topaz was not satisfactory. In order to overcome this problem, amino naphthol disulphonic acid was found to be a good topaz depressant in the presence of alkane carboxylic acid. This reagent scheme has not been tested in a commercial operation. Depressant choice during tin flotation The depressant of choice for cassiterite flotation depends very much on the type of gangue minerals present in the ore. Extensive research work has been carried out in which a number of depressants have been examined on tin ores containing different gangue miner als [12–14]. A number of these depressants have been introduced into various operating plants around the world. A list of reagents used for beneficiation of cassiterite ores is shown in Table 21.3. A large number of organic reagents, although considered good depressants for certain gangue minerals, also have a depressing effect on cassiterite, some of which include citric acid, tanic acid and tartaric acid. Depressants for certain gangue minerals present in cassiterite ore are summarized in Table 21.4. Some gangue minerals such as tourmaline and topaz often occur associated with cassiterite ore. The separation of topaz and tourmaline is particularly troublesome since
R
CH
(CH)2N
COOH
COOH
Figure 21.7
General formula of alkane carboxylic acid used for cassiterite flotation studies.
21.4
Beneficiation of Tin Ores
97 Table 21.3
List of depressants used for beneficiation of various cassiterite ores Inorganic reagents
Organic reagents
Na2SiO3 Na2SiF6 H2SiF6 H2SO4 HCl H3PO4 HF NH4Cl NaPO3 Na2HPO4 NaF – – – –
Oxalic acid Acetic acid 1,2-Ethanecarbonic acid Tartaric acid Citric acid Cupferon Haematoxylene Pyrogalol Recorcinole Hydrochinon Chininhydrochlorid Pheuolic compounds Alcaloides Tanic acid Amino naphthol disulphonic acid
Table 21.4 Depressants for certain gangue minerals present in cassiterite Quartz
Mica
Feldspar
Na2SiO3 H2SiF6
Na2SiO3 H2SiF6
Na2SiO3
Tourmaline
Topaz
Fe2O3
Tautalite
Na2SiF6
Na2SiF6
H2SiF6 Na2SiF6
NaPO3 H2SO4 NaOH Citric acid
Garnet
Oxalic acid
Oxalic acid Citric acid
Tanic acid
not only do these minerals possess similar floatability as cassiterite, but they also occur in significant quantities. Studies carried out on ore that contains topaz and tourmaline [15] indicate that tourmaline and topaz can be floated ahead of cassiterite using dodecil amine hydro chloride as collector. The citric acid can be used to activate both topaz and tourmaline in a pH range from 2.9 to 5.8. Topaz may be selectively activated by fluoride and phosphate ions. Ore that contains pyrophyllite similarly pyrophyllite can be selectively floated ahead of cassiterite using a combination of tall oil and alkyl sulphate. This method is practiced in the Galimovsky plant in Russia.
98
21.
21.5 21.5.1
Flotation of Tin Minerals
PRACTICES IN BENEFICIATION OF TIN-CONTAINING ORES
Factors effecting selection of treatment process
Research work has shown that cassiterite from various deposits and often even from parts of the same deposit differ in chemical composition, colour, flotation properties, chemical activity and electrophysical characteristics. Therefore, the mineralogical composition of tin ores and the physiological properties of the minerals, in particular cassiterite, determine to a great extent the quantity of tin lost during gravity processing and especially during flotation. According to their floatability, ores and products containing cassiterite can be subdivided into the following three groups: 1. 2.
3.
Easily floated ore – typically cassiterite/quartz ores belong to this group. Moderately difficult floating ores – cassiterite sulphide deposits with small quantities of chloritic tourmaline and Fe-oxides are considered to be moderately difficult. Difficult-to-float (or sometimes virtually unfloatable) ore – these ores contain large quantities of chlorite, tourmaline, topaz, fluorite and limonite.
Impurities of the cassiterite surface determine the beneficiation reagent scheme. Cassi terite is seldom, if ever, pure cassiterite. The main impurities are iron, followed by tantalum, columbium and tungsten. These impurities may enhance or adversely effect the flotation recovery of the mineral; the flotation responses largely depend on the flotation method/reagent scheme applied (i.e. phosphonate collectors work better with iron-rich cassiterite, while sulphosuccinamates may float cassiterite containing less iron better). The mineral composition of the given gangue from which cassiterite is to be recovered plays a major role in determining the flotation selectivity and, ultimately, the concentrate’s final grade and value. Complex silicates such as tourmaline, topaz and chlorites have similar flotation properties as cassiterite, which in many cases, represents a problem in obtaining high-grade tin flotation concentrate. The quality of the water in which the cassiterite flotation takes place is also highly important. Both ions found in process water supply and those generated by the minerals present in the pulp may affect the performance of the collectors as well as the surfaces of either cassiterite or gangue minerals by either depressing or activating them. Because there are no similarities in the ore or cassiterite flotation pulp in the world, it is the requirement that the flowsheet and reagent scheme usually be custom-made for each particular case. Great care should be exercised in order that laboratory testing duplicate the commercial plant conditions, which is quite a difficult task.
21.5.2
Development work and operation of cassiterite flotation plants
Between 1970 and 1980, a large number of tin-operating plants have introduced flotation of tin from gravity tailing fines. Nowadays, only a few of these plants are in operation, where
21.5
Practices in Beneficiation of Tin-Containing Ores
99
the reagent scheme has changed significantly due to the development of new collectors. Some of the flotation plants that were in production between 1970 and 1980 are described in the following sections. Renison (Australia) The Renison Bell tin mine is a large, but complex, oxide mineral deposit. Renison ore consists largely of pyrrhotite, quartz, dolomite, siderite and dorite. The chemical composi tion of the iron varies considerably. Some of the ore types are high in copper and silver. Table 21.5 shows the chemical analyses of various ore types. Between 1964 and 1970, Renison had only operated the gravity circuit with overall tin recovery of 45–50%. Extensive research work was carried out in the early 1970s by Goldfields to develop a flotation flowsheet. The work was conducted in Japan. As a result, a flotation flowsheet was developed (Figure 21.8) consisting of multiple-stage desliming and sulphide preflotation followed by tin flotation and upgrading. The desliming was conducted at 6 µm, although attempts were made to deslime at 2 µm, but unsuccessfully. The original Renison reagent scheme (Table 21.6) changed over the years. With respect to collectors, in the late 1970s arsonic acid was first replaced with styrene phosphonic acid, and in the early 1980s with sulphosuccinamate (CA540, Allied Chemi cals). Best results were achieved using styrene phosphonic acid. The NaF was replaced with Na2SiF6, which had improved selectivity. A summary of the plant metallurgical results obtained in 1982 are shown in Table 21.7.
Table 21.5 Chemical analyses of Renison’s various ore types (data from 1978) Ore type
SB/Flt
Flt
Flt
SB
Flt
SB
SB
Flt
MS
MS
Sample
0A1
0A2
0A3
0A4
0A5
0A6
0A7
0A8
0A9
0A10
%Sn %SnS %S %Fe %Cu %As %F %SiO2 %Al2O3 %MnO %MgO %TiO2 %CaO %K2O
1.49 0.05 21.0 33.0 0.11 0.60 0.50 27.0 4.0 0.2 3.0 0.3 0.9 0.5
2.31 0.10 15.0 35.0 1.00 2.40 1.10 5.0 3.9 1.6 0.8 0.6 1.8 0.2
0.65 0.04 11.0 21.0 0.33 0.95 0.80 37.0 7.7 0.8 2.4 1.2 1.9 0.7
2.02 0.02 16.0 27.0 0.15 0.28 1.80 25.0 3.8 0.2 9.9 0.2 1.8 2.0
2.06 0.02 19.0 27.0 0.46 4.60 0.20 30.0 2.4 1.6 1.1 0.07 0.60 0.10
2.10 0.03 11.0 22.0 0.13 0.20 2.10 26.0 2.5 0.05 12.8 0.08 1.70 2.60
2.94 0.02 15.0 26.0 0.20 1.20 1.20 25.0 1.9 0.10 9.3 0.07 1.40 1.50
0.96 0.02 21.0 30.0 0.71 2.00 0.66 32.0 2.4 0.05 2.40 0.10 0.80 0.60
2.69 0.05 38.0 50.0 0.23 0.31
0.79 0.03 40.0 48.0 0.21 0.28
<0.1
<0.10
100
21.
Table tailings
Flotation of Tin Minerals
Primary classifier slimes
Primary deslimer
Secondary deslimer
Sulphide rougher
Sulphide scavenger
Sulphide cleaner
Tertiary deslimer
Final slimes
Conditioner
Tin rougher
Tin scavenger
Sulphide recleaner
Tin 1st cleaner Sulphide cleaner concentrate
Tin tails
Tin 2nd cleaner
Tin 3rd cleaner
Tin 4th cleaner
Tin cleaner concentrate
Figure 21.8 Renison cassiterite flotation flowsheet.
Union and Rooiberg (South Africa) Prior to the introduction of cassiterite flotation recovery of tin by gravity processing at Union, it was about 53% Sn at a concentrate grade of 56%, and at the Rooiberg mill about 66% tin was recovered at a grade of 64%. At both mines there were large accumulations of old tailings that assayed 0.6% Sn, with about 70% of their tin content being in the –43 µm fraction. The flotation development testwork was conducted at Goldfields research facilities in Johannesburg. Before flotation testing, the coarser, low-grade fraction was removed by multiple-stage cycloning to yield a product assaying 1% Sn, which was 97% –37 µm and contained 65% of the tin in 46% of the weight of the total tailings samples. This product was then deslimed to give a flotation feed assaying 1.3% Sn.
21.5
Practices in Beneficiation of Tin-Containing Ores
101
Table 21.6 Original reagent scheme used at the Renison flotation plant Reagent
Depressants and modifiers Copper sulphate (CuSO4) Sodium silicate (Na2SiO3) Sodium fluoride (NaF) Sulphuric acid Collectors and frother Xanthate p-Tolyl arsonic acid (PTTA)a Cresylic acid a
Additions (g/t)
pH
Sulphide circuit
Tin circuit
Sulphide
Tin
200–300 – –
– 800–1500 400–600
– – – 6.0
– – – 4.5–5.0
50–70 – 20–30
– 400–700 –
– – –
– – –
Emulsified with NaOH.
Table 21.7 Plant results from early 1982 Product
Weight (%)
Assays (% Sn)
Distribution (% Sn)
Gravity concentrate Tin flotation concentrate HMS float Sulphide concentrate Table tailing Tn flotation tailing Slimes Feed
1.7 0.7 23.2 45.0 6.8 17.0 5.6 100.0
41.6 23.5 0.17 0.4 0.48 0.30 0.75 1.24
58.3 13.8 3.1 14.5 2.7 4.0 3.6 100.0
Initial flotation work was conducted with different collectors. The results obtained are summarized in Table 21.8. The highest tin recovery was achieved using collector p-tolyl arsonic acid. Based on laboratory data, a small plant was designed where ethyl phenyl phosphonic acid was used as collector at 450 g/t additions. Depressant used included 1000 g/t Na2SiO3 and 1000 g/t NaF. The plant results are shown in Table 21.9. Wheal Jane (UK) The Wheal Jane plant had incorporated flotation together with gravity since start-up of this plant. The flotation development testwork was carried out at the Warren Spring laboratory. The ore treated at the Wheal Jane plant is very complex and finely disseminated. About 86% is less than 50 µm, of which 50% is less than 20 µm.
102
21.
Flotation of Tin Minerals
Table 21.8 Effect of different collectors on tin flotation from the Union and Rooiberg old tailings Collector
Assay (% Sn) % Distribution (Sn) Addition (g/t) pH value Number of cleaning stages
Isohexil phosphonic acid Sodium oleate p-Tolyl arsonic acid Ethyl phenyl phosphonic acid Dialkyl sulphosuccinamate
100
5
2
11
72
250 350 330
4.5–5 4.5–5
2 2
25 25
80 62
400
2.5
2
10
70
Table 21.9 Plant results obtained from Union tin and Rooiberg old tailings Product
Weight (%)
Assays (% Sn)
% Distribution (Sn)
Tin cleaner concentrate Sulphide concentrate Tin flotation tailing Slimes Feed
1.5 2.0 83.2 13.0 100.0
19.6 0.68 0.12 0.4 0.49
61.6 3.3 21.3 13.8 100.0
Development work was carried out with different collectors. At the end, N-(1,2 dicar boxyethyl)-n octadecil sulphosuccinamate emulsified with fuel oil in a ratio of 8:1 was the final collector. The flotation flowsheet is shown in Figure 21.2. The final tin flotation reagent scheme included collector R845 (Cytec) emulsified with fuel oil as a tin collector (890 g/t); H2SO4 for pH control; citric acid (200 g/t) and Na2SiF6 (450 g/t). From the flotation feed assaying 0.87% Sn, a concentrate grade assaying 10.6% Sn at 81.5% Sn recovery was achieved. Further upgrading of the concentrate was achieved by magnetic separation. In the early 1980s, a mixture of sodium silicate and aluminum sulphate (65:35 ratio) was used. This mixture is acidified to a pH of 2.5. No data exist on the effectiveness of this mixture. Valkoomesky plant (Russia) The flotation feed at the Valkoomesky plant contains tourmaline (18%), biotite (13%), muscovite (17%), limonite (2%) and sulphides (5%). Tin assays in the flotation feed averaged about 0.5% Sn, of which the bulk was contained in the –48 to +12 µm fractions. Flotation of tin was carried out with sea water using oxidized petroleum solution in kerosene (1:2 ratio). The results obtained are summarized in Table 21.10. Similar results were achieved with the use of RV-2 collector (p-nitrobenzeneazosalicylic acid).
21.5
Practices in Beneficiation of Tin-Containing Ores
103
Table 21.10 Flotation results from the Valkoomesky plant using sea water Product
Weight (%)
Assays (% Sn)
% Distribution (Sn)
Tin cleaner concentrate Tin cleaner tailing Tin rougher concentrate Sulphide concentrate Slimes Flotation tailing Feed
2.21 5.14 7.35 6.54 1.42 84.69 100.00
20.57 1.08 6.94 0.74 0.67 0.04 0.60
75.54 9.25 84.78 8.03 1.58 5.61 100.0
Table 21.11 Huanuni plant gravity circuit results (2000) Product
Weight (%)
Assays (% Sn)
% Distribution (Sn)
Final gravity concentrate Coarse tailing Fine tailing Sulphide concentrate Feed
3.02 48.50 41.00 7.48 100.00
54.0 0.38 1.56 0.73 2.51
65.0 7.3 25.6 8.1 100.0
The development work on the Huanuni tin recovery from fines began in early 2002 [16]. In the initial testwork, different collector combinations were examined. The effect of these different collectors on tin flotation from the fines is shown in Table 21.11. The above collectors showed good selectivity towards tourmaline and muscovite. Huanuni concentrator (Bolivia) The Huanuni plant primarily used a gravity circuit until the current operation. The gravity circuit fines assayed 1.2–1.3% Sn, which is further treated using gravity tables. In general, the run-of-mine ore is composed of quartz and silicates, 40–50%, and sulphides (pyrite, marcasite, pyrrhotite and arsenopyrite). The principal tin mineral is cassiterite, with minor amounts of stannite. Based on liberation studies, a large portion of the tin is liberated at 300–400 µm size. A portion of the tin is liberated at–12 µm size. The generalized gravity concentration flowsheet is shown in Figure 21.9. The final metallurgical results obtained using gravity averages about 65% Sn recovery. The results obtained during the year 2000 are shown in Table 21.12. Depressants used in these tests included Na2SiF6, Na2SiO3 and DA663 (low-molecular weight polyacrylamide). Collector PL520 is composed of dialkyl sulphosuccinamate (R845) and a mixture of phosphoric acid ester (SM15) modified with fatty alcohol ester sulphate in a ratio of 60:20:20.
104
21.
4-mm screen
12-mm ore
35-mesh screen
–4 mm
+4 mm
Jig
con
–35 m
35-mesh screen +35 m
–35 m Conditioner
con
Jig
tail
tail
Jig con
tail
Rod Mill
tail 1-mm screen
Spirals
Sulphide rougher
Gravity table tail Gravity table
Sulphide scavenger
Sulphide 1st cleaner
con
middling
Final tail
con
tail
Sulphide 2nd cleaner
con
Gravity table
tail
con
–1 mm
+1 mm
Gravity table
Gravity table con tail
con
Jig
+35 m Rod mill
Rod mill
tail
tail
Flotation of Tin Minerals
Sulphide concentrate
middling
Gravity table
Final tin concentrate con tailGravity table
tail
Figure 21.9
Generalized Huanuni gravity concentrate flowsheet. Table 21.12
Effect of different collectors on tin flotation from the Huanuni fines Collector
Additions (g/t)
pH value
Number of cleaning stages
Concentration (% Sn)
Recovery (% Sn)
Dialkyl sulphosuccinamate Collector PL520 Collector LAC2 Collector TX26
350
3.5
3
12
65
250 250 250
3.5 3.5 3.5
3 3 3
30 28 25
88 86 90
Collector TX26 is composed of dialkyl sulphosuccinamate and a mixture of phosphoric acid ester modified with carbamic acid derivative (OMC123) in a ratio of 50:25:25. Collector LAC2 is similar in composition to TX26, except that the carbamic acid is replaced with oxidized sulphonate solution in kerosene (R825). Collector PL520 was selected as the final collector due to its low-frothing properties. After selection of the collector, a series of final locked cycle tests were conducted using the flowsheet shown in Figure 21.10. The final reagent scheme is shown in Table 21.13.
21.5
Practices in Beneficiation of Tin-Containing Ores
Fine tailing
Conditioner
105
<8-µm slime
Cyclones
Conditioner
Sulphide rougher
Sulphide 1st cleaner
Sulphide cleaner concentrate
Conditioner
Tin 1st cleaner
Tin 1st cleaner scavenger
Conditioner Tin 2nd cleaner
Tin rougher Tin 3rd cleaner
Conditioner Tin scavenger
Gravity separator Tin concentrate
Tin combined tails
Figure 21.10
Final flowsheet for tin recovery from fines at the Huanuni Concentrator.
The overall results obtained in the locked cycle tests are presented in Table 21.14. By adding a flotation circuit, the overall tin recovery increased from 65% to 87%. In 2005, the project was in the feasibility study stage. It is not known if the Huanuni flotation plant, which was under construction in 2001, has been completed.
106
21.
Flotation of Tin Minerals
Table 21.13 Final reagent scheme Reagent
Reagent additions (g/t)
Depressants and modifiers Na2SiO3 type ‘N’ Na2SiF6 H2SO4 DA666 (low-molecular-weight acrylic acid) CuSO4 Collectors PL520 Amyl xanthate Methyl isobutyl carbinol
Sulphide circuit
Condition
Tin rougher
Tin cleaners
– – 400 –
500 – – –
600 900 To pH 4.5 150
400 300 To pH 3.5 100
250
–
–
–
– 30 20
– – –
320 – –
50 – –
Table 21.14 Average metallurgical results from five continuous locked cycle tests Product
Weight (%)
Assays (% Sn)
% Distribution (Sn)
Flotation gravity concentrate Tin rougher tailing Sulphide concentrate Slimes Feed
2.14 85.66 4.40 7.80 100.00
56.2 0.16 0.30 0.50 1.38
86.5 9.8 0.9 2.8 100.0
San Rafael, Minsur (Peru) The San Rafael tin mine is one of the richest ore mine in the world, with a head grade of about 5% Sn. The mine actually started in the 1960s as a copper operation, and later tin was discovered and the operation started as a tin operation. The ore is coarse grained, and a portion of the tin was recovered at a relatively coarse-grind size (i.e. 8 mm size). The main gangue minerals found in this ore were silicates, pyrite, pyrrhotite, tourmaline and minor amounts of copper and silver. Initially, the San Rafael mine operated using a gravity circuit only. This circuit utilized jigs, spirals and tables. About 80% of the tin was recovered using the gravity circuit, at an average 54%. The gravity tailing assayed 1.3% Sn was found in a deposit near Valey. In the early 1980s, research work was conducted to recover tin from gravity tailings. Based on the data generated in this study, a flotation plant was designed and started operation in 1985. The San Rafael flotation flowsheet is shown in Figure 21.11. The tin third cleaner concentrate, on average, assayed 36% SnO2 and was upgraded to 54% in a Mozley drum separator. The reagent scheme used in this plant is shown in Table 21.15.
21.5
Practices in Beneficiation of Tin-Containing Ores
107
Fine gravity tail Thickener
Effluent
Final slimes
underflow
Cyclones
Conditioner Sulphide rougher
Sulphide scavenger
Conditioner
Tin rougher
Tin scavenger
Sulphide 1st cleaner
Sulphide 2 Cleaner
Conditioner
nd
Tin 1st Cleaner
Tin 1st Cleaner scavenger
Sulphide cleaner concentrate Tin 2nd cleaner
Tin 3rd cleaner
Gravity concentrator (Mozley drum separator) Tin concentrate
Tin final tails
Figure 21.11 San Rafael (Peru) flotation flowsheet.
For pH control, H2SO4 was used. Tin was floated at a pH 5.0–5.5 in the roughers and 4.5–5.0 in the cleaners. One of the major operating problems arose from the excessive frothing in the roughers and cleaners using collector R845. To solve this problem, a highintensity conditioner was installed in the cleaner feed, which effectively broke the froth and served the purpose of froth control.
108
21.
Flotation of Tin Minerals
Table 21.15 San Rafael’s reagent scheme Reagent additions (g/t)
Reagent
Depressants and modifiers Na2SiF6 Na2SiO3 CuSO4 Collectors R845 (Cytec) Amyl xanthate Dithiophosphate (R3477)
Sulphide circuit
Sn rougher + scavanger
Sn cleaners
– 200–300 250
300–500 800–1000 –
200–300 250–400 –
– 60–80 15–20
450 – –
50–100 – –
Table 21.16 Flotation metallurgical results over a 6-month period Product
Tin final concentrate Sulphide concentrate Slimes Tailings Feed
Weight (%)
2.0 9.2 6.6 82.2 100.0
Assays (%)
% Distribution (Sn)
Sn
ST
54.4 0.3 1.1 0.25 1.39
0.1 40.5 – – –
78.2 2.0 5.2 14.6 100.0
The flotation results obtained during a 6-month period are shown in Table 21.16. Currently, the final concentrate grade assays 54% Sn at a 94.5% tin recovery. It should be noted that, at present, the old tailings are being retreated and processed through gravity and flotation. REFERENCES 1. Pryor, E.J., and Vrobel, S.A., Studies in Cassiterite Flotation, Bulletin of the Institution of Mining and Metallurgy, Vol. 532, p. 201, 1951. 2. Polkin, S.I., Flotation of Rare Metal and Tin Ores, Gosgorte-khizdat, Vol. 637, pp. 77–82, 1971, (Russian text). 3. Krasnukhina, A.V., Lepetov, S.F., and Vakhromova, S.P., Concentration of Tin Ores, Report of Scientific Research Institute on Tin, Novosibirsk, p. 35, 1971. 4. Streltsin, V.G., and Ponovich, M.N., Proceedings of Symposium on Flotation of Cassiterite, Novosibirsk, p. 15, 1973 (Russian text). 5. Bulatovic, S.M., New Collector Mixture for Tin Flotation – The Recovery of Tin from San Rafael Gravity Tailings, Report of Investigation, 2005, p. 262.
References
109
6. Zhu, J., and Zhu, Y., The Effect of Ions in Water on the Benzyl Arsonic Acid Flotation of Cassiterite Slimes, Journal of Central-South Institute of Mining and Metallurgy, Vol. 1, No. 1, pp. 29–37, 1985. 7. He, J.Z., and Liu, M.X., Innovation in Separation Technology for Fine Gravity Semi-Products, Mineral Processing and Extractive Metallurgy, IMM, London, 1984. pp. 553–562. 8. Cuyper, J., and Sales, A., Flotation of Cassiterite, Proceedings of International Tin Symposium, La Paz, Bolivia, pp. 175–182, 1977. 9. Kirchberg, H., and Wottgen, L., The Effect of Phosphorus and Antimony Surfactants on Cassiterite Flotation, Chemistry, Physics and Application of Surface Active Substances, London. pp. 693–704, 1976. 10. Arbiter, N., Beneficiation of Cassiterite Ore by Froth Flotation, British Patent 1,110,643, 1968. 11. Baldauf, H., Scoen.Herr, J., and Schubert, H., Alkane Dicarboxilic Acids and Amino NaphtholSulphonic Acids – a New Reagent Regime for Cassiterite Flotation, International Journal of Mineral Processing, Vol. 15, pp. 117–133, 1985. 12. Polkin, S.J., and Korzova, R.V., The Flotation of Cassiterite and Tourmaline By Means Of DNS and High Molecular Tannins, Tsvetnie Metally, Vol. 13, No. 10, pp. 10–13. 13. Strebzyn, V.G., Selective Flotation of Cassiterite in the Presence of Iron-Bearing Minerals, Obogasthenie Rud, No. 13, pp. 3–6, 1968. 14. Topfer, G., Gruner, U., and Menzer, D., The behaviour of gangue minerals in the flotation of cassiterite, Symposium for Tin Beneficiation, Pudi, Praha, pp. 277–280, 1971. 15. Andrews, P.R.A., Flotation Characteristics of Cassiterite, Tourmaline and Topaz, MSC Thesis, University of Melbourne, Australia, 1971. 16. Bulatovic, S., Development Testwork on Tin Recovery for Huanuni Fines, Report of Investiga tion, 2002, p. 260.
– 22 –
Flotation of Niobium
22.1
INTRODUCTION
Niobium minerals, especially columbite, are also associated with other valuable minerals, such as tantalum, zircon and rare earth minerals. Pyrochlore and a mixture of pyrochlore and columbite have different origins, and therefore, beneficiation of pyrochlore and columbite are different from that of the mixed tantalum niobium ores. In actual plant practice, the treatment process is significantly different from that used for mixed niobium tantalum ores. This is due to the fact that the beneficiation process is largely determined by the nature of gangue minerals present in the ore. In most cases, the beneficiation process applicable for pyrochlore ore cannot be successfully applied for beneficiation of tantalum/ niobium ores.
22.2
GENERAL OVERVIEW OF PYROCHLORE-CONTAINING ORES
There are two major types of pyrochlore-containing ores: pegmatite ores and carbonatites. This classification is based on the mineral composition of these ore types. The main waste minerals contained in the pegmatite ores include quartz and nepheline. This ore type also includes granites, where pyrochlore is represented in a coarse crystalline form. Granites are composed of cryolite and topaz as the main gangue minerals. Carbonatite ores are mainly composed of calcite, dolomite and phosphates as the main gangue minerals. The beneficiation process for pegmatites containing pyrochlore mostly includes gravity preconcentration. Such deposits are common in Africa (Kongo, Madagascar). The major minerals contained in pyrochlore-containing ores are pyrochlore, columbite and sometimes ilmenorutile to a lesser extent. Table 22.1 shows pyrochlore minerals present in pegmatite and carbonatite ores. The gangue composition of the various carbonatite ores varies considerably. Calcite– dolomite content in some ores ranges from 30% (Niobec, Canada) up to 70% (Panda Hills, Africa). From a mineralogical point of view, pyrochlore usually occurs in crystallized form, as well as octahedron form. Pyrochlore occurs in considerable range of colours, varying from translucent white to opaque black appearance with glassy surfaces. The Nb2O5
111
112
22.
Flotation of Niobium
Table 22.1 Pyrochlore minerals contained in pegmatite and carbonatite ores Mineral
Formula
Pyrochlore Columbite Ilmenorutile
(NaCa)2Nb2O6F (FeMn)Nb2O6 (Ti,Nb,Fe)O2
Assays (%) Nb2O5
Ta2O5
38–65 23–77 0.3–6.6
0–5.8 1–40.0 60–72
Specific gravity
Hardness
4.1–5.4 5.3–6.6 5.94
5.0–5.5 6.0–6.5 7.0
content of pyrochlore crystals is dependent on the amount of Na/Ca content and can range from 38% to 65% Nb2O5. It has been established that the colour of pyrochlore plays an important role in the floatability of pyrochlore.
22.3
FLOTATION PROPERTIES OF PYROCHLORE
The treatment process and flotation properties of pyrochlore are very much dependent on the gangue composition of the ore. The selective flotation of pyrochlore from carbonatite ore is not possible since calcite and dolomite have similar flotation properties as pyro chlore. In addition, in the presence of carbonates, the stable pH required for flotation of pyrochlore (i.e. 5.0–5.5) cannot be maintained. In the case of carbonatite ores, a beneficiation process involves preflotation followed by reactivation and flotation of pyrochlore. In the case of pegmatitic ores that contain silicates, biotite, albite and limonite, as the gangue minerals, direct flotation of pyrochlore can be achieved with a variety of different collectors. 22.3.1
Flotation of pyrochlore from carbonatite ores
The successful flotation of pyrochlore from carbonatite ores depends on a number of factors: • • •
efficiency of removal of calcite and dolomite before pyrochlore flotation type of pretreatment of calcite–dolomite tailing before pyrochlore flotation type of pyrochlore collectors.
The calcite/dolomite in actual practice is recovered using fatty acid as a collector and starch or dextrin as the pyrochlore depressant during calcite–dolomite flotation. A number of studies have been conducted [1,2] in which different fatty acid modifica tions were examined. High selectivity and high calcite–dolomite recoveries were obtained with emulsified fatty acid with soda ash and sodium silicate. Table 22.2 shows the results from calcite/dolomite flotation using different fatty acid type collectors and various modifications.
22.3
Flotation Properties of Pyrochlore
113 Table 22.2
Effect of different fatty acids and various modifications on calcite–dolomite recovery from carbonatite ores Nb2O5 depressant used during % Nb2O5 in calcite Flotation CaO flotation concentrate pH
Calcite–dolomite Recovery (%) collectors
Oleic acid Tall oil fatty acid Emulsified tall oil EMF1a EMF2b a b
CaO 65 70 72
MgO 58 Caustic corn starch 62 Caustic corn starch 66 Caustic corn starch
88 89
80 85
Caustic corn starch Caustic corn starch
10.0 8.4 7.5
8.5 8.0 8.2
4.6 4.2
8.5 8.3
Tall oil / Na2CO3 / Na2SiO3 = 60:20:20. Tall oil / Na2CO3 / Na2SiO3 = 65:15:20.
The best results were achieved using fatty acid emulsified with soda ash and sodium silicate. The effectiveness of emulsified fatty acid EMF2 was dependent on flotation pH. Figure 22.1 illustrates the effect of pH on calcite–dolomite recovery using 400 g/t collector EMF2. In actual plant practice, by removing the calcite–dolomite, the pyrochlore in the flotation feed is significantly upgraded. In some ores, which assay 0.4% Nb2O5 in the feed after calcite–dolomite preflotation, the pyrochlore assays in the pyrochlore flotation feed is over 1.2% Nb2O5. 100 CaO 80
Recovery (%)
MgO 60
40
20
0 7
Figure 22.1
8
9 Flotation pH
10
11
Effect of pH on calcite–dolomite recovery from carbonatite ores.
114
22.
Flotation of Niobium
100
Niobium recovery (%)
80
60 HCl 40 HNO3
H2SO4
20
0 0
10 20 30 40 50 Niobium concentrate grade (%Nb2O5)
60
Figure 22.2 Effect of type of acid used in the pretreatment of pyrochlore flotation feed on the grade–recovery relationship.
The calcite flotation tailing in most cases is pretreated before niobium flotation. The pretreatment reagents used include acids, such as sulphuric acid and hydrochloric acid. Studies conducted on carbonate flotation tailing on the Orca (Canada) ore [3] showed that the use of hydrochloric acid in the pretreatment stage improved niobium metallurgy significantly. Sulphuric and nitric acids were less effective. Figure 22.2 shows the grade– recovery relationship using different acids in the pretreatment stage. In each experiment, about 1000 g/t of acid was used in the pretreatment stage. A conditioning time of 15 min was maintained. The use of acid in the pretreatment stage also improves removal of residual fatty acid. Choice of modifiers and depressants In the majority of cases, oxalic acid has been proven to be a selective gangue depressant during pyrochlore flotation. Fluorosilicic acid or hydrofluoric acid have been used in a number of operating plants as secondary gangue depressants. In fact, HF has been replaced with fluorosilicic acid. The effect of level of oxalic acid and fluorosilicic acid on pyrochlore flotation is presented in Table 22.3. The results showed that higher additions of oxalic acid improved both pyrochlore grade and recovery. Higher additions of fluorosilicic acid have a negative effect on pyrochlore recovery. The effect of sodium hexametaphosphate (Calgon, Canada and USA) and sodium pyrophosphate on pyrochlore flotation has been investigated [4]. Small addition of Calgon (50 g/t) was found to have a beneficial effect on depressing the gangue minerals, including aegirine, whilst nearly doubling the niobium content of the froth product.
22.3 Flotation Properties of Pyrochlore
Table 22.3 Effect of level of oxalic acid on pyrochlore metallurgical results Reagent additions (g/t) Oxalic acid
H2SiF6
300 500 700 700 700 700
200 200 200 300 400 500
Feed (% Nb2O5)
0.85 0.83 0.85 0.89 0.85 0.86
Rougher concentrate
Cleaner concentrate
Grade (% Nb2O5)
Recovery (% Nb2O5)
Grade (% Nb2O5)
Recovery (% Nb2O5)
5.8 6.7 8.5 9.3 8.3 8.5
70.2 75.4 78.8 75.6 70.3 66.6
50.2 52.3 55.6 57.4 57.2 58.1
61.3 66.5 69.6 65.4 60.0 55.2
115
116
22.
Flotation of Niobium
Sodium silicate has a strong depressing effect on pyrochlore, and it is sometimes used during calcite flotation. Sodium silicate hydrosol is prepared by reacting ferric chloride and silicate, followed by acidification of the mixture, which has a positive effect on selectivity. The addition of small quantities of hydrosol (100 g/t) resulted in significant improvement in concentrate grade. Extensive studies have been carried out using orthodihydroxybenzene, known as cate chol (commercial name). This reagent has improved the rate of fine pyrochlore flotation and also has a beneficial effect on selectivity. Research work with this reagent was conducted on carbonatite ore from Canada. Collector choice In the majority of cases, amines are used as pyrochlore collectors during treatment of carbonatite ores. Aliphatic mono amines, aliphatic diamines, condesates of capritic acid and partially neutralized diamines are the principal collectors for pyrochlore. Tallow diamine acetate (Duomac T, Akzo Nobel, USA and Canada) is also used as a pyrochlore collector. The effect of different amine collectors has been examined on Niobec ore from Canada through development testwork [5]. Table 22.4 shows the results from the laboratory continuous locked-cycle tests conducted using different amine collectors manufactured by Akzo Nobel, USA, Canada – Clariant, Germany. Depressants used in these experiments include oxalic acid and fluorosilicic acid. The use of quinolines [7] were examined with the addition of fuel oil as co-collector. According to the data provided (Table 22.5), quinolines are effective pyrochlore flotation collectors. The number of carbons in the quinoline structure determines the grade and recovery of pyrochlore. Quinolines have not found industrial application due to the cost of these reagents. 22.3.2
Flotation of pyrochlore from pegmatitic ores
Pegmatite-containing niobium ores can be relatively complex and may contain biotite, enargite, albite, feldspar and ziron as the main gangue minerals. Some pegmatite ores (Araxa, Brazil) have a simple gangue composition, consisting mainly of quartz. Table 22.4 Effect of different amines on pyrochlore flotation from St. Honore Niobec ore Collector
Duomac T Duomac T/Ethofat C25 CES 109 CES 109/Duomac T TAP 100
Manufacturer
Akzo Nobel Akzo Nobel Akzo Nobel Akzo Nobel Clariant
Head (% Nb2O5)
0.92 0.90 0.89 0.90 0.91
Final Nb2O5 concentrate Assays (%)
% Distribution
54.3 55.4 57.4 56.6 50.0
66.4 68.0 72.0 75.4 70.2
22.3
Flotation Properties of Pyrochlore
117 Table 22.5
Metallurgical results obtained with different quinolines Collector
Oil
8-Quinolinol 2-Methyl 8-quinolinol 4-Methyl 8-quinolinol 6-Methyl 8-quinolinol
Head (% Nb2O5)
Furnace oil Burner oil Burner oil Burner oil
1.3 1.3 1.3 1.3
Final Nb2O5 concentrate Assays (%)
% Distribution
10.36 9.21 19.3 5.59
86.0 85.5 80.4 90.2
Research work carried out at the Mechanobre Institute in Russia involved the evaluation of cationic and anionic collectors [8]. The anionic collectors examined included sodium oleate and sodium alkyl sulphate. The results obtained indicated that with the use of sodium oleate, both zircon and pyrochlore can be floated with good recoveries (Figure 22.3). Flotation of pyrochlore using sodium alkyl sulphate is dependent on flotation pH. At a pH above 5.5, no pyrochlore flotation is achieved. At this pH, microcline, limonite and aegirine were floated. It appears that the use of alkyl sulphate at slightly acidic to alkaline pH number of gangue minerals can be selectively floated from pyrochlore. At a pH between 1.5 and 3.0, alkyl sulphate floats pyrochlore and zircon, whereas floatability of limonite, microline and aegirine is greatly reduced (Figure 22.4). 100
Zircon
Niobium
80
Recovery (%)
Albite 60 Biotite
40
20
0 0
40
80
120
160
200
Sodium oleate additions (g/t)
Figure 22.3
Effect of sodium oleate on flotation of niobium from pegmatite ores.
118
22.
Flotation of Niobium
100 Zircon
Recovery (%)
80
60 Niobium
40
Biotite
20
Microcline
0 0
100
200
300
400
500
Na-alkyl sulphate additions (g/t)
Figure 22.4 of 1.4.
Effect of sodium alkyl sulphate on flotation of minerals from pegmatite ores at a pH
Using cationic flotation (C-14 amine and amine hydrochloride) method, no selectivity between pyrochlore and gangue minerals is achieved. Amine flotation, therefore, cannot be successfully applied for flotation of pyrochlore. In recent years, new technology has been developed for beneficiation of niobium from pegmatitic ores that contain nepheline, feldspar, fluorite and aluminosilicates [9]. A line of new collectors, known as the PLV and PM series, was developed that is highly selective for pyrochlore–zircon flotation from pegmatite ores. Collectors from the PLV series are mixtures of alkyl sulphosuccinamate and ester phosphates modified with sodium alkylsul phate. Collector PLV28 was successfully developed for the beneficiation of niobium zircon from feldspar-containing ores. The reagent scheme developed for the beneficiation of this ore is shown in Table 22.6 and the metallurgical results in Table 22.7. Good feldspar depression was achieved using oxalic acid and magnesium fluorosilicate. Small additions of a low-molecular-weight acrylic acid improved concentrate grade. Separation of pyrochlore and zircon from the bulk concentrate was possible. The separation method is discussed in Chapter 23. Collectors from the PM series were specifically developed for beneficiation of niobium ores that contain nepheline/cyanite as the major gangue minerals. The collector is com posed of a mixture of phosphate ester collector (SM15, Clariant) and phosphonic acid treated with octanol. From an ore that assays 0.5% Nb2O5, a concentrate grade of 49% Nb2O5 at a recovery of 73% was achieved.
22.4
Refractory Niobium Ores
119 Table 22.6
Reagent scheme developed for beneficiation of niobium from feldspar-containing pegmatite ores. Reagents
Additions (g/t)
Depressants and modifiers Oxalic acid Magnesium fluorosilicate Accumer 9400 Collectors PLV28 Diesel fuel
pH
Roughers
Cleaners
700–900 300–400 150
200–300 150–200 100–200
80 150–200
150 50
4.5
Table 22.7 Metallurgical results obtained on feldspar-containing niobium ores from the Kanyaka deposit in Africa Product
Weight (%)
Bulk Nb/Zr cleaner concentrate Bulk Nb/Zr tailing Magnetics Slimes Feed
22.4
2.85 86.25 2.50 8.40 100.00
Assays (%)
% Distribution
Nb2O5
ZrO2
Nb2O5
ZrO2
20.5 0.07 0.40 0.44 0.70
25.6 0.05 2.5 1.50 2.25
83.5 9.1 1.4 6.0 100.0
83.9 4.8 7.1 4.2 100.0
REFRACTORY NIOBIUM ORES
There are several fairly large niobium deposits around the world that belong to the refractory ore type. Some of these deposits can be found in Brazil, Africa and Greenland. Typically, these ores are heavily oxidized and mostly contain iron oxides and aluminium silicates. A typical example of such a deposit is the Mrima Hill deposit found in southeast Kenya, which was a case study in which new technology was examined. The Mrima Hill deposit is considered to be one of the richest deposits in the world. The niobium in this ore occurring as pyrochlore is concentrated in a feringinous residue formed by intense weathering of an underlying carbonatite. During the weathering process, pyro chlore has been altered to a microcrystalline form which breaks up into very fine particles. The principal gangue minerals are goethite with some hematite and magnetite, aluminosi licates and apatite. Minor amount of barite and ilmenite are also present in the ore. Previous studies conducted on this ore [10] indicated that a saleable-grade concentrate can be produced but at a very low niobium recovery (18–20%). Most of the niobium losses occur in the
120
22.
Flotation of Niobium
–8 μm fraction (between 34% and 43%). Research work on the Mrima ore was conducted during 1984–1985 [11]. The research work was designed to (a) develop a flowsheet that would minimize the niobium losses in the fine fraction, and (b) develop a reagent scheme that will float ultra-fine pyrochlore. The main task in flowsheet development is the desliming size to reduce niobium losses in the slime fraction. The generalized flowsheet used in this case study is shown in Figure 22.5.
Ore +4 mm
4 - mm screen
Deslime 1 –4 mm 65 - mesh screen
Grind Deslime 2 +65 m
–65 m
65 - mesh screen
Deslime 3
Slime Slime Slime
+65 m Low - intensity magnetic separation
Conditioner 1 Conditioner 2
Magnetics P2O5/Ba rougher
Slimes
P2O5/Ba scavenger Conditioner 1 Conditioner 2
P2O5/Ba 1st cleaner
P2O5/Ba 1st cleaner scavenger
Nb2O5 rougher
Nb2O5 scavenger
Nb2O5 1st cleaner
Nb2O5 1st cleaner scavenger
Bulk apatite barite concentrate Nb2O5 2nd cleaner
Nb2O5 3rd cleaner
Nb2O5 4th cleaner
Nb2O5 cleaner concentrate
Figure 22.5
Mrima case study flowsheet.
Nb2O5 tailings
22.4
Refractory Niobium Ores
121 Table 22.8
Effect of desliming size on niobium losses in the slime fraction Number of desliming stages
Dispersant
Desliming size (µm)
% Nb2O5 recovered in the slimes
2 2 3 3
Na2SiO3/Calgon (1:1) Na2SiO3/Calgon (1:2) AQ2 AQ2
6 5 4 3
44 32 22 14
The apatite barite bulk flotation was accomplished with a mixture of tall oil fatty acid and sulphonate (Aero 827) at an alkaline pH. Sodium silicate and caustic tapioca starch were used for pyrochlore depression during the bulk apatite barite flotation stage. In the desliming stage, various dispersants and a number of desliming stages were examined. The niobium losses in the slime fraction were closely related to the desliming size. Table 22.8 shows the relationship between desliming size and niobium loss in the size fraction. The niobium loss in the size fraction is significantly reduced at a 3 μm desliming size. The combined sands from the three desliming stages (Figure 22.4) was fed to the niobium flotation circuit. The successful niobium flotation was very dependent on: (a) type of dispersant used, (b) type of depressant system and (c) type of collector. The effect of collector type on niobium rougher recovery is illustrated in Table 22.9. The results obtained indicated that cationic flotation of pyrochlore was not successful. Dispersant AQ4 has a pronounced effect on niobium metallurgical results. Dispersant/ depressant AQ4 is composed of the following individual reagents: 60% orthodihydrox ybenzene (Catacol), 30% low-molecular-weight acrylic acid (Accumer 2400) and 10% hexametaphosphate. The AQ4 provides excellent pulp dispersion and slime depression during niobium flotation. The niobium grade–recovery relationship using different levels of AQ4 is shown in Figure 22.6. The final metallurgical results obtained in continuous locked cycle testing are shown in Table 22.10.
Table 22.9 Effect of collector type on niobium rougher recovery. Collector
Duomac T Duomac T/Ethofat C25 PLV28 PLV29
Dispersant
AQ4 AQ4 AQ4 AQ4
Niobium rougher concentrate % Grade
% Recovery
6.6 3.8 7.2 6.6
33.5 39.6 72.2 78.5
122
22.
Flotation of Niobium
100
0 g/ t
t g/
ne
⇐
AQ4 Concentration
60
40
0 20
no
Nb2O5 recovery (%)
80
40
20
0 0
10
20
30
40
50
Niobium concentrate grade (% Nb2O5)
Figure 22.6 Effect of level of dispersant/depressant AQ4 on niobium grade–recovery relationship using collector PLV29.
Table 22.10 Locked-cycle test results obtained from the Mrima Hill niobium refractory ore Product
Weight (%)
Nb2O5 cleaner concentrate Nb2O5 combined tails P2O5/BaSO4 concentrate Magnetics Slimes Head
22.5
3.37 51.5 12.0 18.0 15.5 100.0
% Niobium Assays
Distribution
46.3 0.71 0.20 0.6 2.2 2.4
65.1 15.9 0.1 4.5 14.2 100.0
PLANT PRACTICES IN BENEFICIATION OF PYROCHLORE ORES
There are several operating plants treating pyrochlore-containing ores from carbonatite and pegmatite ores. Operating plants that treat carbonatite ores described in this chapter include St. Honore Niobec, Canada, and OKa, Quebec, Canada. The operating plant that treats pegmatite ore is Araxa (Brazil).
22.5
Plant Practices in Beneficiation of Pyrochlore Ores
22.5.1
123
St. Honore Niobec operation
The research and development work began in the late 1960s and early 1970s. The Niobec plant was designed and put into operation in 1975. Early in the operation of the plant, tall oil fatty acid was used for calcite/dolomite flotation. In the niobium circuit, oxalic acid and HF depressant system was used. The pyrochlore collector used was Duomac T manufactured by Akzo Nobel. The initial flowsheet included calcite/dolomite circuit flotation and desliming the calcite tailing followed by niobium flotation and cleaning. From the niobium concentrate, pyrite was removed using a niobium depression system with starch and pyrite flotation using xanthate. In late 1970, the HF was replaced with H2SiF6. Although good concentrate grade was achieved (i.e. 5–62% Nb), the plant recovery was relatively low and ranged from 55% to 63% Nb2O5. The major niobium losses in the plant occurred in the (a) slimes, (b) calcite concentrate and (c) cleaner tailings. A portion of coarse columbite was usually lost in the cleaner tailings. In the early 2000s, detailed research work was carried out with the objective of improving the plant’s metallurgical results. As a result of these studies, a new flowsheet was developed and introduced into the plant. The current flowsheet includes a new desliming circuit, where two desliming stages were introduced. With the use of a double desliming stage, the desliming size was reduced from a P80 of 12 μm to 5 μm. This resulted in a reduction in niobium losses in the slime fraction from 15% to 6% Nb2O5. The niobium circuit flowsheet (Figure 22.7) was modified to include (a) thickening of the deslimed calcite tailing before flotation, and (b) retreatment of the niobium cleaner tailing for extra niobium recovery. With respect to the reagent scheme, the following modifications were made: • Emulsified fatty acid with soda ash and silicate was used in the calcite circuit. Xanthate was added to the emulsion, where pyrite was floated with the calcite/ dolomite concentrate. Using this calcite/dolomite system, the calcite/ dolomite recovery to the calcite concentrate increased from 55% to 80%, respectively. • A partially neutralized aliphatic mono-amine was used (collector CES 109) in the niobium circuit instead of diamine. The current Niobec reagent scheme is shown in Table 22.11. The metallurgical results over a 6-month period is shown in Table 22.12. 22.5.2
Oka operating plant
The Oka plant located in Quebec, Canada, has been in operation for past several years. The flowsheet and reagent scheme are similar to those used at the Niobec operation with the exception that the pyrochlore collector used involved a mixture of amines Duomac T: Ethofat C25 manufactured by Akzo Nobel. The concentrate grade obtained assayed 57.5% Nb2O5 at a recovery of 67.3%.
124
22.
Flotation of Niobium
Ore feed Primary grinding
65 - mesh –65 m screen
Deslime 1 Deslime 2
Slimes Slimes
+65 m
Slimes
sands Secondary grinding
Thickener
o/f
u/f Conditioner
Ca/Mg rougher
Conditioner Ca/Mg scavenger
Ca/Mg 1st cleaner
Nb2O5 rougher
Nb2O5 scavenger
Nb2O5 1st cleaner
Nb2O5 1st cleaner scavenger
Ca/Mg 2nd cleaner Nb2O5 2nd cleaner Deslime
Ca / Mg concentrate
Nb2O5 3rd cleaner
sands Nb2O5 scalper
Nb2O5 4th cleaner
Nb2O5 5th cleaner
Nb2O5 cleaner concentrate
Figure 22.7
Niobium plant flowsheet.
Final tails
References
125 Table 22.11 Current Niobec reagent scheme Additions (g/t)
Reagent
Depressants and modifiers Oxalic acid H2SiF6 Na2SiO3 CuSO4 Collectors CS-109 NC3
Calcite flot
Niobium Ro
Niobium Cl
– – 200–300 200
600–700 300–400 – –
300–500 100–200 – –
– 500
400–500 –
100–150 –
Table 22.12 Plant results obtained at the Niobec plant Product
Weight (%)
Nb2O5
SiO2
Fe
Recovery (% Nb2O5)
Nb2O5 final concentrate Nb2O5 combined tails Slimes Calcite concentrate + magnetics Feed
0.92 53.38 10.20 35.50 100.00
61.2 0.14 0.33 0.20 0.75
2.2 – – – –
1.8 – – – –
75.4 10.6 4.5 9.5 100.0
REFERENCES 1. Bulatovic, S., An Investigation into Recovery of Pyrochlore from St-Honore Niobium ore (Canada), Report of Investigation, 2003. 2. Bulatovic, S., Process Development for Beneficiation of Oka Niobium-containing (Quebec) Ore, Report of Investigation, 2006. 3. Desrochers, C., Traitement du Minerai de St-Honore, Centre de Rescherches Minerales (MRN), 1971. 3. Pavlor, D.A., Flotation of Niobium from Pegmatitic Ores, Tsvetnie Metally, No. 8, 1976. 4. Bushel, C.H.G., and Fackson, H.E., Flotation Process, US Patent 2,975,895, 1961. 5. Desrochers, C., and Dessureaux, S., Report on Pilot Plant Testing, St-Honore project, 1973, MRN. 6. Arthur, W.L., and Kent, F.M., Columbium Flotation Process, US Patent 2,875,896, 1975. 7. Polkin, S.I., Obogaschenie Rud Redkih and Blagorodnik, Metalov Moskow, Nedra, 1987. 8. Bulatovic, S., Research and Development of Niobium Flotation from Pegmatitc Ore, SGS Report of Investigation, 2007. 9. Harris, P.M., Investigation into Recovery of Niobium from Mrima Hill Deposit, Institution of Mining and Metallurgical, 10, 1966. 10. Bulatovic, S., Development of a Treatment Process for Beneficiation of Mrima Hill Ore, Report of Investigation, 1985.
– 23 –
Flotation of Tantalum/Niobium Ores
23.1
INTRODUCTION
There are approximately 130 different minerals that contain tantalum and niobium, from which about 80 are Ta/Nb only. The other minerals contain tantalum and niobium in the form of impurities. There is very little information available on beneficiation of Ta/Nb containing ores. In actual practice, there are three basic methods for production of Ta/Nb concentrate: (a) physical preconcentration, (b) combination of physical preconcentration and flotation and (c) direct flotation. In most cases, Ta/Nb ores contain significant quan tities of zircon and rare earth ores (REO).
23.2
CHARACTERISTICS OF Ta/Nb MINERALS OF ECONOMIC VALUE
The Ta/Nb minerals of economic value can be divided into three main groups: (a) tantoloniobites, (b) titanotantaloniobites and (c) tintanotantaloniolites containing uranium. Table 23.1 lists the major Ta/Nb minerals from these three groups. Ta/Nb minerals often occur as impurities in ilmenite, rutile, cassiterite, wolframite and perovskite, most of which contain REE. Because tantalite and columbite have similar chemical properties, they often replace each other, and are usually found as isomorph mixtures. Tantalum and niobium can also be found as separate minerals. Tantalite and microlite are primary sources of tantalum.
23.3
GEOLOGICAL AND MINERALOGICAL FEATURES OF Ta/Nb ORES
There are about five different geological identities of ores that contain tantalum and niobium. The following is a brief description of each ore type. Ores of magnetic origin. Tantalum/columbite granites are of economic interest when the columbite content of the ore ranges from 0.001% to 0.01% and the tantalite up to 0.2%. These deposits are most common in Nigeria (Africa). Because they are a low-grade ore, they do not represent significant economic value.
127
128
Table 23.1 Ta/Nb minerals of economic value Mineral
Formula
Assays (%) Nb2O5
Ta2O5
Specific gravity (g/cm3)
Hardness
Magnetic susceptibility
23.7–77 2.0–4.0 0.3–6.1 –
1–40 44–84 60–72 72–83
5.3–6.6 6.7–8.3 5.9–7.3 7.6–7.9
6 6 7 6
Low Low Low Low
Titanotantaloniobites Ilmenorutile Struverite Loparite Pyrochlore Microlite
(Ti,Fe,Nb)O2 (Ti,Fe2+,Ta)O2 (Na,Ca,Sr,Ce)(Ni,Ti)O3 (Na,Ca,Th,Tr2)(Nb,Ta,Ti) (Na2,Ca,Th,Tr)2(Ta,Ti,Nb2)2(O,OH,F)7
0.9–4.2 7.0 11.6 37–65 7.7
0.4–1.4 36 0.7 0–5.9 68–77
4.6–5.0 5–6 5.0 4–5 5–6
6 6.5 5.0 5–5.3 –
Very low Very low Very low Very low Very low
31.3 23–45 47 3.8–4.7 23–32 27–46 7.5–20
5.9 0–2.8 17.3 0.0–4.7 0–6.9 2–27
4.5 3.7–5 5.5–6.8 4.5–6 5.2 5.8 0–23
4–6 4–4 5–6 5.5–6.5 5–6 5–6 4–7
Non-magnetic Non-magnetic Non-magnetic Non-magnetic Non-magnetic Non-magnetic Non-magnetic
Uranium-containing pyrochlore Gatecottolit Betafite (U,Ca,Th,Ce)(Nb,Ti,Ta)O9n(H2O) Fergusonite (Y,Er,Ce,U)(Nb,Ta,Tr)O4 Euxenite (Y,Ce,U,Ca,Th)(Ti,Nb,Ta)2O6 Eshinit (Ce,Ca,Th)(Ti,Nb)2O6 Samarskite (Y,Er,U,Ce,Th)4[(Ta,Nb)2O7]3 Polikraz (Y,Ce,Ca,U,Th)(Ti,Nb,Ta)2O6
Flotation of Tantalum/Niobium Ores
(Fe,Mn)(Nb,Ta)2O6 (Mn,Fe)(Ta,Nb)2O6 AlTaO4 Sn(Ta/Nb)2O7
23.
Tantaloniobite Columbite Tantalite Simpsonite Torolite
23.4
Flotation Characteristics of Tantalite–Columbite Minerals
129
Pegmatite deposits are the most abundant. They contain a variety of minerals including tantalum, niobium, lithium and beryllium, as well as REE and zircon. Metasomatic deposits are altered albite and granatoids. These are low-grade ores. Of economic interest is the carbonatites, which contain up to 1% combined Ta/Nb. Pneumatalitic-hydrothermal deposits contain Ta/Nb as isomorph impurities in cassiterite and wolframite. Ta/Nb from these ores is recovered in a tin and wolframite concentrate. Sedimentary deposits are the most important deposits of economic values. These depos its contain tantalocolumbite, columbite, samorskite and torolite. The Ta/Nb from these deposits is recovered using a gravity concentration method. These deposits also contain significant quantities of ZrO2 and REO.
23.4 FLOTATION CHARACTERISTICS OF TANTALITE–COLUMBITE MINERALS There is very little information or literature on flotation properties of tantalum and columbium minerals. Also, there are only a few operating plants that treat tantalum– columbium ore by flotation. Most of the commercial plants use a gravity beneficiation method. Studies conducted by Mechanabve Institute [1,2] indicate that tantalite and columbite can be floated using sodium oleate. Figure 23.1 shows the effect of level of sodium oleate
100 90 80
Recovery (%)
70 Garnet Tourmaline
Tantalum/Niobium Muscovite Albite
60
50 40 30 20 10 0 0
100
200
300
400
Sodium oleate addition (g/t)
Figure 23.1 Effect of level of sodium oleate on recovery of tantalite, columbite and associated gangue minerals.
130
23.
Flotation of Tantalum/Niobium Ores
100
Zircon
Recovery (%)
80
60
40
Niobium Tantalum
20
0 2
4
6
8
10
Flotation pH
Figure 23.2 collector.
Effect of pH on tantalite and columbite flotation using sodium alkyl sulphonate as
on recovery of tantalite, columbite and associated minerals. The data in Figure 23.1 show that tourmaline is recovered together with tantalite/columbite, while albite and muscovite remain depressed. Sodium alkyl sulphonate is also a collector for tantalite and columbite at a pH below 3.0 (Figure 23.2). At a pH above 3.0, flotation recovery of tantalite and columbite decreased rapidly. This collector was not selective towards gangue minerals, such as tourmaline and garnet. Cationic flotation of tantalite columbite has also been studied on several ore types that contain tourmaline, feldspar and muscovite as the major gangue minerals [3]. The effect of aliphatic mono-amine on flotation of Ta/Nb is presented in Figure 23.3. As can be seen from Figure 23.3, tourmaline and muscovite float readily with amine collectors. Preflotation of the tourmaline and muscovite before tantalite/columbite flotation was not successful. Mixtures of phosphate esters (SM15, Clariant, Germany) and succinamates (R845, Cytec, USA) modified with alkyl sulphate are proven to be effective tantalite–columbite collectors. These collectors (SM500 series) were examined on ore from a pegmatitc origin that contained tantalite and columbite with minor amounts of struverite. The results obtained using several collectors from the SM500 series are shown in Table 23.2. Excellent results were achieved with these collectors.
23.5
Practices in Beneficiation of Ta/Nb Ores
131
100 Muscovite Tourmaline
80
Niobium
Recovery (%)
Tantalum 60
40
20
0 0
50
100
150
200
250
Amine additions (g/t)
Figure 23.3
Effect of amine on Ta/Nb flotation from complex ore. Table 23.2
Effect of collectors from the SM500 series on tantalum flotation and upgrading Collector
Depressant
Rougher concentrate
Cleaner concentrate
Assays (%)
Assays (%)
% Distribution
% Distribution
Ta2O5 Nb2O5 Ta2O5 Nb2O5 Ta2O5 Nb2O5 Ta2O5 Nb2O5
500 g/t SM502 500 g/t SM504 500 g/t SM510 500 g/t SM515
Citric acid, MgSiF6 1.1
5.5
82.0
86.5
7.2
42.2
70.0
77.5
Citric acid, MgSiF6 0.9
4.6
84.0
88.3
6.1
37.3
75.5
78.2
Citric acid, MgSiF6 2.1
8.5
75.5
82.1
9.2
46.6
70.2
75.2
Citric acid, MgSiF6 1.2
6.0
87.0
90.0
6.8
37.0
77.5
82.0
23.5 23.5.1
PRACTICES IN BENEFICIATION OF Ta/Nb ORES
Introduction
A large portion of Ta/Nb concentrate production at industrial scale comes from gravity concentrating plants. Using gravity preconcentration, the fine Ta/Nb (i.e. −150 mesh) is not recovered and the recovery of Ta/Nb using gravity concentration is relatively low, ranging
132
23.
Flotation of Tantalum/Niobium Ores
from 55% to about 65%. With the development of new technology, it is possible to float Ta/ Nb from gravity tailings with a significant increase in overall metallurgy. Another major problem with beneficiation of Ta/Nb-containing ores is the presence of zircon in the ore, or in the gravity concentrate. There is now an effective process for Ta/Nb–Zr separation, which is developed after extensive research work. Beneficiation of Ta/Nb ores containing REEs belong to a group of complex ores. Beneficiation of these ores presents a challenge. The REEs are of primary value from this ore type, whereas Ta/Nb is of secondary value. 23.5.2
Gravity concentration
The principal method for beneficiation of Ta/Nb ores is gravity concentration. In principle, most of the Ta/Nb ores contain low-specific-gravity minerals of about 2.8–3.0 specific gravity (SG) (quartz, calcite, aluminosilicates, feldspar, etc.), whereas heavy minerals (Ta/Nb and other Ta/Nb mineral carriers) have SGs of 4–4.4, which is suitable for gravity preconcentration. A typical flowsheet used for gravity preconcentration [4] is shown in Figure 23.4. The ground ore is usually sized after grinding and gravity concentration is performed on the different size fractions. Examples of gravity concentration of Ta/Nb are the Green Bushes operation in Australia and the Bernic Lake deposit in Canada. These plants are still in operation, and are using a flowsheet similar to that shown in Figure 23.4. Results from the Green Bushes (Australia) operation are shown in Table 23.3. Most of the Ta/Nb losses in the tailing occurred in the fine −200 mesh fraction and cyclone overflow slimes. Bernic Lake (Canada) operates a gravity circuit using a flowsheet similar to that shown in Figure 23.4. The results obtained from this operation are presented in Table 23.4. The results obtained are slightly better than those obtained in the Western Australia concentrators.
23.6 23.6.1
FLOTATION
Background
Direct flotation of Ta/Nb from mainly tantalum-containing ore is not practiced in operating plants. Only a few currently operating plants have tested the possibility of using flotation to recover tantalum from gravity tailings. Although in recent years new technology has been developed, it has yet to be introduced into any operating plants. 23.6.2
Bernic Lake Ta/Nb flotation from gravity tails
In 1980, Bernic Lake introduced a flotation circuit to float Ta/Nb from the gravity tailings using succinamate collector at a pH of about 4.5 controlled by hydrochloric acid. The main depressants used included Na2SiF6 and oxalic acid.
23.6
Flotation
133
Ore Grinding
6-mesh screen
–6 m
48-mesh screen
+6 m Gravity table
–48 m
+48 m tail
Gravity table
concentrate
100-mesh screen
–100 m
+100 m Deslimer
tail
slime
concentrate
Gravity table
tail
concentrate Classifier 100 mesh
Regrinding
–100 m
+100 m
tail
Gravity cleaner
Gravity recleaner
tail
Gravity concentrate
Figure 23.4
Final tail
Generalized gravity concentration flowsheet.
Table 23.3 Greenbushes gravity circuit results Product
Gravity combined concentrate Gravity tailings Feed
Weight (%)
0.84 99.16 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
Ta2O5
Nb2O5
25.4 0.12 0.33
11.97 0.10 0.20
64.5 35.5 100.0
50.3 49.7 100.0
Using a flotation method, about 30% Ta2O5 was recovered at a concentrate grade of about 6%. This concentrate was returned to the gravity circuit.
134
23.
Flotation of Tantalum/Niobium Ores
Table 23.4 Bernic Lake gravity circuit results Product
Weight (%)
Gravity combined concentrate Gravity tailings Feed
0.78 99.22 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
Ta2O5
Nb2O5
34.8 0.13 0.40
9. 0.08 0.15
68.2 31.8 100.0
47.3 52.7 100.0
23.6.3 Flotation of Ta/Nb from Greenbushes gravity tailing Extensive laboratory testing was performed on the Greenbushes gravity tailing, followed by pilot plant testing at the mine site. The generalized final flowsheet, evaluated in the pilot plant tests, is shown in Figure 23.5. The reagent scheme used in the pilot plant included oxalic acid–acidified silicate Na2SiF6 gangue depressant system and collector composed of a mixture of phosphoric esters and alkyl sulphate modified with mineral oil. The metallurgical results obtained are presented in Table 23.5.
23.7
BENEFICIATION OF Ta/Nb ORES CONTAINING ZIRCON
There are a number of fairly large deposits that contain Ta/Nb associated with zircon in a complex gangue matrix. Over the past 10 years, extensive laboratory testwork has been performed on several deposits from the Middle East (Arabia), Africa (Malawi) and Brazil. A description of the new processes used for beneficiation of these ores is presented in the following section. 23.7.1 Development of a beneficiation process for Ta/Nb recovery from Ghurayyah ore – Saudi Arabia The ore The Ghurayyah ore has a complex mineral composition, and is a relatively fine-grained ore. The niobium present in the ore is represented by a variety of minerals, including pyro chlore, yttro-pyrochlore and columbite. The tantalum in the ore was associated with niobium and had a high Nb:Ta ratio. Based on the available mineralogical data, grinding between 100 and 50 µm would liberate the majority of tantalum and niobium minerals. This was confirmed during a laboratory testwork. The zircon present in the ore had a subhedral, non-crystalline structure, common to some volcanogenic deposits. An association of the zircon with silica and feldspar was observed. The gangue minerals in this ore were represented by silica and feldspars.
23.7
Beneficiation of Ta/Nb Ores Containing Zircon
135
Combined gravity tailings
100-mesh screen
–100 mesh
Deslimer
Slime
+100 mesh Regrind
Conditioner 2
Ta/Nb rougher
Ta/Nb scavenger
Ta/Nb 1st cleaner
Ta/Nb cleaner scavenger
Ta/Nb 2nd cleaner
Ta/Nb 3rd cleaner
NaOH conditioner
Gravity cleaner Gravity concentrate
Final tail
Figure 23.5 Final flotation flowsheet for Ta/Nb flotation from gravity tailings.
23.7.2
Beneficiation studies
Throughout the development testwork, emphasis was placed on finding an effective reagent scheme that would produce a bulk concentrate with satisfactory tantalum, niobium and zircon recoveries. The Ta/Nb–Zr separation study concentrated mainly on magnetic separa tion. The floatability of Ta/Nb and Zr depend on type of collector, modifier and depressant system used.
136
23.
Flotation of Tantalum/Niobium Ores
Table 23.5 Plant metallurgical results [5] Product
Weight (%)
Concentrate Tailings Feed (plant gravity tails)
0.45 99.55 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
Ta2O5
Nb2O5
14.8 0.053 0.12
8.00 0.064 0.10
55.3 44.7 100.0
36.0 64.0 100.0
The cationic flotation of Ta/Nb was not effective, as they tend to float feldspar-bearing minerals. The effect of some amines on Ta/Nb–Zr bulk flotation is illustrated in Table 23.6. The results showed that amines normally used for pyrochlore flotation did not work for flotation of Ta/Nb. Therefore, collector selection is very dependent on the type of niobium minerals present in the ore. Fatty acids and their variations were examined for selective flotation of zircon from Ta/Nb. No selectivity or zirconium flotation was achieved using fatty acids. The use of a blend of succinamate and phosphoric acid esters, modified with phosphonic acid, produced good results. These collectors are known as collectors from the PL500 series. Performance of the flotation circuit was related to the flotation pH. The effect of pH on Ta/Nb recovery is illustrated in Figure 23.6. The levels of collector were examined in several series of flotation tests. These results from the tests using different levels of collector is shown in Table 23.7. In order to achieve high bulk concentrate recoveries, relatively high additions of collector are required. A depressant system developed for beneficiation of Ta/Nb–Zr ores involves oxalic acid– hydrofluorosilicic acid and depressant SHQ. SHQ is a mixture of a low-molecular-weight acrylic acid and condensation product of disulphonic acid (Suspendol PKK, manufactured by Cognis, Germany). After the development of the final reagent scheme, a series of locked-cycle tests were performed using the flowsheet shown in Figure 23.7. Table 23.6 Effect of different amines on Ta/Nb recovery Product
Hydrogenated tallow amine I-H imidazole-amine N-tallow amine acetate Coco amine acetate Tallow diamine
Assays (%)
% Distribution
Ta2O5
Nb2O5
Ta2O5
Nb2O5
0.02 0.04 0.009 0.01 0.01
0.33 0.45 0.11 0.12 0.13
10.4 15.8 8.4 5.5 4.8
15.48 25.2 15.3 8.9 9.5
23.7
Beneficiation of Ta/Nb Ores Containing Zircon
137
100
Zircon
Recovery (%)
80
60
40
Niobium Tantalum
20
0 2
4
6
8
10
Flotation pH
Figure 23.6
Effect of pH on Ta/Nb rougher recoveries using collector PL519. Table 23.7
Effect of level of collector PL519 on Ta/Nb–Zr flotation Collector additions (g/t)
200 400 600 800
Assays (%)
% Distribution
Ta2O5
Nb2O5
ZrO2
Ta2O5
Nb2O5
ZrO2
0.18 0.11 0.08 0.10
2.62 1.69 1.10 1.22
9.7 5.75 2.60 3.80
48.5 70.9 78.4 82.5
49.3 77.8 83.2 86.6
55.5 84.2 92.3 96.0
The results obtained from the continuous locked-cycle tests are shown in Table 23.8. The beneficiation process developed for a Saudi Arabian Ta/Nb–Zr ore is considered as a new technology. 23.7.3
Separation of Ta/Nb and Zr
Due to the nature of Ta/Nb–Zr, separation using a flotation method is not possible. Research using magnetic separation was carried out under different operating conditions. The efficiency of the Ta/Nb–Zr separation was a function of a number of factors, including
138
23.
Flotation of Tantalum/Niobium Ores
Ore 200-mesh screen
–200 mesh
+200 mesh Grind Rod Mill
+200 mesh
200-mesh screen
–200 mesh
Slime
Conditioner Cyclones
Conditioner 1 Conditioner 2 Bulk rougher 1
Bulk rougher 2
Bulk 1st cleaner
Bulk cleaner scavenger
Bulk scavenger
tail
tail
Bulk 2nd cleaner
Bulk 3rd cleaner
Bulk 4th cleaner
Bulk cleaner concentrate
Figure 23.7
Bulk combined tails
Final flotation flowsheet used in the continuous locked-cycle tests.
23.7
Beneficiation of Ta/Nb Ores Containing Zircon
139
100 Tantalum/ Niobium
Recovery (%)
80
60
40
20
Zircon
0 0
2
4
6
8
10
HCl addition (kg/t)
Figure 23.8 Effect of level of hydrochloric acid on Zr removal using high-gradient magnetic separation (HGMS).
• • •
Acid pretreatment pH Stage separation at different magnetic field strengths Separate treatment of slime and sand fractions.
The acid pretreatment with HCl was the most critical parameter. The effect of level of HCl on zircon rejection is shown in Figure 23.8. Table 23.8 Results from the continuous locked-cycle tests Product
Bulk cleaner concentrate Bulk combined tail Slimes Head (calc) Bulk cleaner concentrate Bulk combined tail Slimes Head (calc)
Weight (%)
4.34 89.98 5.68 100.00 3.58 90.49 5.93 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
ZrO2
Ta2O5
Nb2O5
ZrO2
0.45 0.002 0.022 0.022 0.46 0.002 0.023 0.020
5.49 0.033 0.34 0.27 6.22 0.035 0.36 0.28
16.3 0.033 0.32 0.78 23.5 0.024 0.33 0.84
86.8 7.6 5.6 100.0 84.3 8.7 7.0 100.0
82.9 10.4 6.7 100.0 80.8 11.5 7.8 100.0
93.7 3.9 2.4 100.0 95.3 2.4 2.2 100.0
140
23.
Flotation of Tantalum/Niobium Ores
Table 23.9 Ta/Nb separation results using HGMS after acid treatment Product
Weight (%) Assays (%)
% Distribution
Ta2O5 Nb2O5 ZrO2 Ta2O5 Ind 30A magnetic 21.27 cleaner concentrate 30A magnetic 28.57 rougher concentrate 30A combined 71.43 non-magnetic 100.00 Feed (calc)
1.82
24.5
1.42
19.1
6.05
Nb2O5 O’all Ind
ZrO2 O’all Ind
O’all
84.7 71.4
80.3 64.9
51.1
12.2
88.7 74.8
83.9 67.8
13.8 13.1
11.3
16.1 13.0
86.2 82.2
0.07
1.47
30.6
0.46
6.49
25.4
9.8
4.8
100.0 86.4 100.0 80.8 100.0 95.3
The use of separate treatments for the sand and slime fractions was also beneficial for separation efficiency. The final Ta/Nb–Zr separation results are shown in Table 23.9.
23.8
BENEFICIATION OF Ta/Nb ORE FROM MALAWI, AFRICA
The ore used in this example contained a mixture of pyrochlore and columbite as the major niobium minerals. The tantalum is mainly associated with columbite. The major gangue minerals present in this ore were soda and potassium feldspars with small amounts of mica and quartz. Beneficiation of this ore using cationic flotation, normally employed for flotation of niobium, was not applicable for this particular ore, since most of the mica and feldspar floated with the niobium and tantalum. The effect of amine on Ta/Nb flotation is illustrated in Figure 23.9. The selectivity between Ta/Nb and gangue minerals using a cationic collector was very poor. 23.8.1
Experimental development testwork using alternative collectors
A number of different collectors, rather than cationic and anionic collectors, which did not perform well, were examined. The new collectors evaluated included (a) alkaline sulphates, (b) alkyl sulphates, (c) sulphosuccinamates and (d) phosphoric acid esters. Mixtures of these collectors, modified with branched alcohols or 1-octane sulphonic acid, gave good metallur gical results. The results obtained with the different collectors are illustrated in Table 23.10. Collector mixtures PLV26 and PLV28 achieved good rougher–scavenger bulk recoveries. 23.8.2
Effect of different depressant systems on Ta/Nb flotation
A number of different depressant combinations were examined during a laboratory develop ment test programme. Oxalic acid, citric acid and fluorosilicic acid were among the
23.8
Beneficiation of Ta/Nb Ore from Malawi, Africa
141
100 Feldspar 80
Recovery (%)
Albite 60
Tantalum/ Niobium
40
20
0 100
200
300
400
Amine addition (g/t)
Figure 23.9
Effect of amine acetate on the flotation of individual minerals.
Table 23.10 Effect of different collectors on Ta/Nb–Zr bulk flotation Collector
Alkaline sulphonate Sulphosuccinamate Phosphonic acid ester Collector PLV26 Collector PLV28
Assays (%)
% Distribution
Ta2O5
Nb2O5
ZrO2
Ta2O5
Nb2O5
ZrO2
0.2 0.3 0.35 0.44 0.50
2.5 3.3 3.80 6.20 6.90
6.6 8.2 8.8 12.8 14.3
33.4 40.5 45.5 75.5 85.5
41.5 50.6 55.7 78.3 88.6
48.8 55.5 60.2 90.2 95.3
depressants included in this evaluation, at different pH values. Combinations of oxalic acid and fluorosilicic acid were found to perform the best. Usually, fluorosilicic acid is added in the rougher flotation stage, at fixed additions only. An excess of H2SiF6 reduces Ta/Nb and Zr recoveries. Figure 23.10 shows the effect of levels of H2SiF6 on individual mineral recoveries. The optimum dosage of H2SiF6 was between 200 and 300 g/t. Floatability of niobium was the most sensitive mineral to the level of H2SiF6. The performance of oxalic acid was related to pH. The effect of pH on the grade– recovery relationship is shown in Figure 23.11. A pH above 5.0 was detrimental to Ta/Nb cleaning efficiency.
142
23.
Flotation of Tantalum/Niobium Ores
100
Zircon
Recovery (%)
80
Niobium
60
40
Tantalum 20
0 100
300
500
700
900
H2SiF6 addition (g/t)
Figure 23.10
Effect of levels of fluorosilicic acid on individual mineral recoveries.
100
Recovery (%)(average Ta/Nb)
80 pH = 3 60 pH = 4
40 pH = 6
20
0 0
4
8
12
16
20
24
28
Concentrate grade (%Ta+Nb)
Figure 23.11 Effect of pH on Ta/Nb on the grade–recovery relationship using 800 g/t oxalic acid.
23.8
Beneficiation of Ta/Nb Ore from Malawi, Africa
23.8.3
143
The treatment flowsheet, reagent additions and metallurgical results
The treatment flowsheet and reagent additions developed under laboratory conditions are presented in Figure 23.12. The results obtained in a continuous locked-cycle test are shown in Table 23.11. In spite of the number of attempts to selectively float Ta/Nb and Zr, it was not possible to either float Ta/Nb and depress the zircon or float the zircon and depress the Ta/Nb. In fact, it was found that the recovery of Ta/Nb in the bulk rougher–scavenger concentrate was strongly related to the recovery of zircon. Figure 23.13 shows this relationship.
Feed Ore
ground, deslimed
200 g/t H2SiF6 400 kg/t oxalic acid Conditioner 1 400 g/t PLV28 Conditioner 2 200 g/t oxalic acid 200 g/t PLV 28 Bulk rougher 1
Bulk rougher 2
100 g/t oxalic acid 200 g/t PLV 28 Bulk scavenger
100 g/t H2SiF6 200 g/t oxalic acid 50 g/t PLV 28 st
Bulk 1 cleaner
Bulk cleaner scavenger
100 g/t oxalic acid Bulk 2nd cleaner
100 g/t oxalic acid Bulk 3rd cleaner
100 g/t oxalic acid Bulk 4th cleaner
Bulk cleaner concentrate
Figure 23.12
Bulk final tail
Final treatment flowsheet and reagent scheme for beneficiation of Ta/Nb–Zr ores.
144
23.
Flotation of Tantalum/Niobium Ores
Table 23.11 Bulk metallurgical results obtained in a continuous locked-cycle test Product
Weight (%)
Bulk cleaner concentrate Bulk combined tails Slimes Feed
1.92 92.07 6.01 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
ZrO2
Ta2O5
Nb2O5
ZrO2
1.17 0.007 0.02 0.03
23.5 0.08 0.46 0.55
20.36 0.02 0.17 0.42
75.0 21.0 4.0 100.0
82.2 12.8 5.0 100.0
93.1 4.5 2.4 100.0
100
Zircon recovery (%)
95
90 Tantalum Niobium
85
80
75
70 50
Figure 23.13 concentrate.
23.9
60
70 80 Ta or Nb recovery (%)
90
100
Relationship between Ta/Nb and Zr recovery in the bulk rougher–scavenger
Ta/Nb–Zr SEPARATION FROM THE BULK CONCENTRATE
A method involving Ta/Nb depression and Zr flotation was developed, by which Ta/Nb depression and Zr flotation was performed. This method is illustrated in Figure 23.14. The metallurgical results obtained are presented in Table 23.12. The results indicated that good separation efficiency can be achieved using a heat starch separation method. The zircon collector used in this test programme (CES3) was a mixture of primary and secondary amines. Corn caustic cooked starch was used as a Ta/Nb depression during zircon flotation.
23.9
Ta/Nb–Zr Separation from the Bulk Concentrate
145
Bulk Concentrate 3 kg/t Na2SiO3 Conditioner (Heat 80°C) Dewatering / Washing
Effluent 500 g/t caustic starch
Conditioner 100 g/t CES3 collector
50 g/t CES2 collector ZrO2 rougher
ZrO2 scavenger
Ta/Nb concentrate
250 g/t caustic starch ZrO2 1st cleaner
ZrO2 1st cleaner tail
200 g/t caustic starch ZrO2 2nd cleaner
ZrO2 2nd cleaner tail
ZrO2 cleaner
concentrate
Figure 23.14
Flowsheet and reagent scheme used in Ta/Nb separation.
Table 23.12 Preliminary batch Ta/Nb–Zr separation test results Product
Ta/Nb concentrate–ZrO2 tails ZrO2 combined Cl tails ZrO2 concentrate froth product Feed (bulk cleaner concentrate)
Weight (%)
46.19 6.70 47.11 100.00
Assays (%)
% Distribution
Ta2O5
Nb2O5
ZrO2
Ta2O5
Nb2O5
ZrO2
2.10 0.1 0.50 1.21
42.2 3.82 0.16 22.8
0.47 7.7 32.8 21.3
80.1 0.6 19.3 100.0
85.5 11.2 3.3 100.0
10.2 22.6 67.2 100.0
146
23.
23.10
Flotation of Tantalum/Niobium Ores
Ta/Nb SEPARATION FROM REFRACTORY TIN GRAVITY INTERMEDIATE PRODUCTS
The Pitinga tin operation, located in Brazil [6], produces an appreciable amount of intermediate gravity product containing Ta/Nb and Zr. This intermediate product is highly refractory and contains a relatively large quantity of Fe-hydroxides, which are coated on the mineral surfaces of zircon, tantalum and niobium, making Ta/Nb–Zr separation using flotation difficult. Extensive laboratory testwork was conducted on this intermediate product resulting in the development of a new separation process, based on zircon depres sion and Ta/Nb flotation. The process consists of two distinct steps: 1. 2.
decoating the Ta/Nb–Zr minerals, and Ta/Nb flotation from the decoated Ta/Nb–Zr product.
23.10.1
Fe-hydroxide decoating
A number of different reagent combinations were examined, along with different flowsheet combinations. The Ta/Nb–Zr separation was strongly related to the amount of Fe-hydroxide decoated. Figure 23.15 shows the relationship between Fe-hydroxide removed, Ta/Nb concentrate grade and Zr content of the Ta/Nb concentrate.
50
25
45
Concentrate grade (%Ta+Nb)
20
35 30
15
25 20
10
15 5
10
ZrO2 Content of Ta/Nb concentrate
40
5 0 0
20
40
60
80
0 100
Fe(OH)2 removed (%)
Figure 23.15 Effect of amount of Fe-hydroxide removed on Ta/Nb grade and Zr content of the Ta/Nb concentrate.
23.10
Ta/Nb Separation from Refractory Tin Gravity Intermediate Products
147
Feed
Grinding
Screening (100 mesh)
+
–
1 kg/t NaOH 200 g/t AQ4
Scrubbing 1
Desliming 1 kg/t NaOH 200 g/t AQ4
Slime
sand Scrubbing 2
Desliming 1 kg/t H2SO4 200 g/t AQ4
Slime
sand Scrubbing 3
Desliming
Sand to flotation
Figure 23.16
Slime
Slimes
Grinding, scrubbing and desliming flowsheet.
Good separation efficiency was achieved after >80% of the Fe-hydroxide was removed. Good Fe-hydroxide removal was achieved with the use of alkaline acid scrubbing and desliming. The final decoating flowsheet is shown in Figure 23.16. 23.10.2
Ta/Nb–Zr separation
The reagent scheme developed for Ta/Nb–Zr separation (Table 23.13) involved oxalic acid–H2SiF6–AAC10 depressant system for zircon. Depressant AAC10 is a mixture of alginic acid, a low-molecular-weight acrylic acid and citric acid. This depressant is specifically designed to depress iron-containing gangue minerals.
148
23.
Flotation of Tantalum/Niobium Ores
Table 23.13 Reagent scheme developed for Ta/Nb–Zr separation Reagent
Additions (g/t)
Modifiers and depressants NaOH H2SO4 H2SiF6 Oxalic acid AQ4 AAC10 Collectors RS702
Pretreatment
Ta/Nb rougher
Ta/Nb cleaning
3000 1000 – – 500 –
– – 1500 1300 – 100
– – 800 800 – 350
–
600
250
Table 23.14 Results obtained using the reagent scheme shown in Table 23.13 Product
Ta/Nb cleaner concentrate Magnetics Ta/Nb combined tail Feed (washed sand)
Weight (%)
5.62 1.87 92.51 100.00
Assays (%)
% Distribution
Ta2O3
Nb2O5
ZrO2
Ta2O3
Nb2O5
ZrO2
4.93 0.13 0.09 0.36
38.6 1.55 0.64 2.80
1.6 6.65 36.8 34.3
76.9 0.9 22.2 100.0
77.5 1.0 21.5 100.0
0.3 0.4 99.3 100.0
The Ta/Nb flotation was accomplished using collector RS702. This collector is com posed of amine acetate, phosphoric acid esters and hydroxamate. Collector RS702 is a powerful collector, capable of floating a variety of niobium minerals that are contained in the flotation feed. Metallurgical results obtained from a continuous locked-cycle test are shown in Table 23.14. Using this new separation method, over 99% of the total zircon in the feed was rejected in the Ta/Nb final tailing. REFERENCES 1. Polkin, C.I., and Glatkin, U.F., Concentration of Tantalum Niobium Ores, M. Gosgortexizdat, 1963, 160–85. 2. Polkin, C.I., Concentration of Rare Earth Ores from Sedimentary Deposits, Obogaschenie Rud Retkih and Blagovodish Metalov (eds.), Izdatelstvo Nedra Moskva, pp. 268–275, 1987. 3. Bulatovic, S., Tantalum niobium flotation from complex ores, Report of Investigation, p. 185, 1989.
References
149
4. Fishman, M.A., and Sobolev, D.C., Practices in Concentration of Sulphide and Oxide Minerals, Gornoe Delo, Vol. 4, pp. 283–305. 5. Bulatovic, S., Pilot Plant tests on Greenbushes (Australia) Gravity Tailing, Report of Investiga tion, 2003. 6. Bulatovic, S., New process for Ta/Nb-Zr separation from Paranapanema (Pitinga) gravity inter mediate, Report of Investigation, 2006.
– 24 –
Flotation of REO Minerals
24.1
ORE AND MINERALS CONTAINING RARE EARTH OXIDE ELEMENTS (REOE)
There are about 250 minerals that contain REOEs, but only a few of these minerals are of any economic value. Most of them contain uranium, titanium, tantalum and niobium. Based on the composition of the REOE minerals, they are classified into two main groups [1]. These are: 1. 2.
The cerium group of REOEs, in which loparit, bastnaesite, parisit, monazite, eshipit and ortit are included. The yttrium group of REOEs, this group includes ytroparisite, fergusonite, samarskite, priorit, kenotime, gadolinite, amongst others.
Table 24.1 lists the major REO minerals of economic value. REO minerals are also divided into two sub-groups, complex and selective complex minerals, all containing lantanoids (from cerium to lutecium). The selective group contains elements from onto or the other group. Most of the products that come from REOEs are monacite, bastnaesite and euxenite. Monazite belongs to the phosphate group of REOEs, with low magnetic properties and bright yellow colour. Usually it is found in pegmatites and granites and also entrained in zircon, magneite and ilmenite. During decomposition of hard rock ores, monazite, due to its chemical stability, is contained in sand deposits together with ilmenite, zircon, magneite and other minerals. The minimum content of monazite found in a sand deposit is about 1%. Bastnaesite belongs to the carbonatite group of minerals that contain REOEs. Beside the cerium group of elements, bastnaesite also contains yttrium and europium. Typically, it contains 65–75% REOE. Bastnaesite is usually found in pegmatites, carbonatite and hydrothermal ore bodies in alkaline gangue minerals. Because it is poor chemically and stable, it is not found in mineral sand deposits. Euxenite is a titanotantalum/niobium-containing mineral and has a complex formula (Table 24.1) with variable chemical composition. It is usually found in sand deposits together with monazite, xenotime, zircon, beryl, columbite and other minerals. The major minerals that contain REOE include apatite, phosphates, sfen, perovskite, eudialite, pyrochlore and ortit, some of which contain significant quantities of REOE.
151
152
Table 24.1 REO minerals of economic value Formula
Relative REOE content
Monazite
(Ce,La…)PO4
50–68% (Ca,La…)2O3, 22–31% P2O5, 4–12% ThO2, �7% ZrO2, �6% SiO2 4.9
5.5
Bastnaesite
(Ce,La,Pr)[CO3]F
36–40% Ce2O3, 36% (La…Pr)2O3, 19–20% CO3, 6–8% F
4.5
4.5
Xenotime
YPO4
52–62% Y2O3, Ce, Er as impurities, Th, �5% U, 3% ZrO2 �, 9% SiO2
4.6
4.5
Parasite
Ca(Ce,La…)2[CO3]3F2
11% CaO, 26–31% Ce2O3, 27–30% (La,Nd)2O3, 24% CO2, 6% F
4.3
4.5
Yttrocerite
(Ca,Y,Ce,Er)F2-3H2O
19–32% Ca, 8–11% Ce, 14–37% Y, 37–42% F
3.8
4.5
Gadolinite
(Y,Ce2)Fe,BeSi2O10
10–13% FeO, 30–46% YO3, 25% SiO2, 5% (Ce,La…)2O2, 9–10% BeO
4–4.5
6.5–7
Ortit
(Ca,Ce)2(Al,Fe)3SiO2[O,OH]
6% Ce2O3, 7%(La…)O3, 4% BeO, 8% Y2O3
Loparit
(Na,Ca,Ce,Sr)2(Ti,Ta,Nb)2O6
39–40% TiO2, 34% (Ce,La…)2O3, 8–11% (Ta,Nb)2O5, 5% CaO, Cr,Th as impurities
4.8
6
Esxenit
(Y,Ce,Ca,U,Th)(Ti,Nb,Ta)2O6
18–28% (Y,Er)2O3, 0.2–3% (CeLa…)2O3, 16–30% TiO2, 4–47% Nb2O5, 1.3–33% Ta2O5, 0.4–12% U3O8
4.9
5.5
Fergusonite
(Y,Sr,Ce,U)(Nb,Ta,Ti)O4
46–57% (Nb,Ta)2O5, 31–42% Y2O3, 14% Er2O3, 1–4% ThO2, 1–6% UO2
5.6-6.2
5.5–6.5
Samarskit
(Y,Er,U,Ce,Th)4(Nb,Ta)6O21
6–14% Y2O3, 2–13% Er2O3, 3% Ce2O3, 0.7–4% (Pr,Nd)2O3, 27–46% Nb2O5, 1.8–27% Ta2O5, Sn, U, Fe as impurities
5.6–5.8
5–6
Priorit
(Y,Er,Ca,Th)(Ti,Nb)2O6
21–28% (Y,Er)2O3, 3–4% Ce2O3, 21–34% TiO2, 15–36% Nb2O5, 0.6–7% ThO2, 0–5% UO2
7.8–5
5.6
Eschynite
(Ce,Ca,Th)(Ti,Nb)2O6
15–19% Ce2O3, 0.9–4.5% (Y,Er)2O3, 21–24% TiO2, 23–32% Nb2O5, 0–7% Ta2O5, 11–17% ThO2
24.
Mineral
Flotation of REO Minerals
24.2
Flotation Properties of Cerium Group of Reoe Minerals
153
Loparite (Nb-mineral) contains, for example, three times more REOE than niobium. It represents titanotantalo-niobium REOE ore. Loparite is found in pegmatites and nephelinecontaining ores. Monazite, bastnaesite and loparite contain exclusively cerium group of REOEs. Other minerals containing REOE, such as fergusonite, priorite and samerskite are usually accessory minerals that contain tantalum, niobium, uranium and thorium. 24.2 24.2.1
FLOTATION PROPERTIES OF CERIUM GROUP OF REOE MINERALS Flotation properties of monazite and bastnaesite
From disseminated ores contained in mineral lenses, the recovery of bastnaesite and monazite is accomplished using flotation. The flotation properties of bastnaesite and monazite are similar to the gangue minerals contained in the bastnaesite and monazite, such as calcite, barite, apatite, tourmaline, pyrochlore and others, which represent difficul ties in selective flotation. However, in recent years, significant progress has been made in the flotation of both monazite and bastnaesite [2,3]. Monazite is readily floatable using cationic collectors such as oleic acid and sodium oleate in the pH region of 7–11. Monazite does not float readily using, for example, laurel amine or anionic collectors. Adsorption of the sodium oleate on the monazite increases with an increase in pH, indicating that monazite does not float in acid pH, while pyrochlore is readily floatable and is depressed at a pH greater than 10. Figure 24.1 shows the effect of pH on flotation of monazite, pyrochlore and zircon.
100 Monazite
Recovery (%)
80
60
40
Zircon 20
Pyrochlore 0 4
Figure 24.1
6
8 Flotation pH
10
12
Effect of pH on flotation of monazite, zircon and pyrochlore.
154
24.
Flotation of REO Minerals
100 Monazite
Recovery (%)
80
60
40 Pyrochlore
Zircon
20
0 0.0
Figure 24.2
0.5
1.0
1.5 2.0 2.5 Na2S addition (g/t)
3.0
3.5
4.0
Effect of Na2S on the recovery of monazite, zircon and pyrochlore.
It was found that Na2S�9H2O is a selective regulating agent during monazite flotation at additions of 2–3 kg/t Na2S, both zircon and pyrochlore are depressed while monazite floatability remains unchanged or, in the case of some ores, improves. Figure 24.2 [4] shows the effect of Na2S on the flotation of zircon, pyrochlore and monazite. Flotation properties of bastnaesite depend largely on the gangue composition of the ore and the impurities present in the mineral itself. Bastnaesite found in a carbonatite ore is recovered using fatty acid collector after heat pretreatment of the flotation feed. The effect of heat temperature on bastnaesite grade–recovery is illustrated in Figure 24.3. Floatability of bastnaesite found in barite–fluorite ores is extremely poor using either fatty acid flotation or sodium oleate. Research work conducted on an ore from Central Asia showed that the floatability of bastnaesite improved significantly after barite preflotation [5]. The flotation of bastnaesite from a carbonatite ore improved with the use of oleic acid modified with phosphate ester. The flotation of bastnaesite from deposits of pegmatitic origin can be successfully accomplished with several types of collectors, including tall oil modified with secondary amine, and tall oil modified with petroleum sulphonate-encompassing group. O R S R O
The effect of the tall oil modification on bastnaesite metallurgical results is presented in Table 24.2. Data shown in this table indicates that the use of a modified tall oil resulted in significant improvement in the metallurgical results of bastnaesite.
24.2
Flotation Properties of Cerium Group of Reoe Minerals
155
100 Heated, ambient 85°C
REO recovery (%)
80
60
40
20 Heated to 40°C
Unheated to 18°C 0 10
Figure 24.3
20
30 40 50 Concentrate grade (% REO)
60
70
Effect of heat temperature on bastnaesite grade–recovery relationship.
Table 24.2 Effect of tall oil modifications on bastnaesite flotation from pegmatitic ores Collector
Product
Tall oil fatty acid
Weight (%) Total % REO assays % REO recovery
REO concentrate 10.77 REO combined tail 89.23 Feed 100.00 Tall oil modified with REO concentrate 10.59 Secondary amine REO combined tail 89.41 (amine acetate) Feed 100.00 Tall oil modified with REO concentrate 10.45 Petroleum sulphonate REO combined tail 89.55 Feed 100.00
24.2.2
48.5 2.07 7.08 60.1 0.74 7.02 62.2 0.61 7.05
73.8 26.2 100.0 90.5 9.5 100.0 92.3 7.7 100.0
Flotation properties of REO-containing yttrium
There is very little literature relevant to flotation of REO-containing yttrium. Yttrocerite, gadolinite, fergusonite and priorit are often found in relatively complex ores containing quartz, chlorite and sericite. Two or all of the above minerals are found together in some deposits. Some of the complex deposits of hydrothermal origin contain zircon together with REO from yttrium groups. Usually the ores that contain yttrium group minerals belong to
156
24.
Flotation of REO Minerals
disseminated ores where liberation occurs at <74 µm size, so the only method available for beneficiation of these ores is flotation. Limited research studies [6] show that the minerals from the yttrium groups can be recovered using alkyl hydroxamate collectors which form complex reactions with REO. It has been found that yttrocerite and gadolinite readily float with hydrohamic acid at a pH of 9–10. The proposed treatment flowsheet for beneficiation of REO-containing yttrium is presented in Figure 24.4. Using the flowsheet shown above, a concentrate grade of 65% REO+Y2O3 at a 72–75% Y2O3 recovery can be achieved on some ores. Research work has shown that the efficiency
Feed Grinding
Desliming 1
Slimes
Desliming 2
Slimes
Sand Conditioning 1 Conditioning 2
Y2O3/REO rougher
Y2O3/REO scavenger
Conditioning
Y2O3/REO 1st cleaner
Y2O3/REO cleaner scavenger
Y2O3/REO 2nd cleaner
Conditioning
Y2O3/REO 3rd cleaner
Y2O3/REO scalper
Y2O3/REO 4th cleaner
Y2O3/REO concentrate
Figure 24.4
Total tailings
Generalized flowsheet for beneficiation of yttrium group of minerals using flotation.
24.2
Flotation Properties of Cerium Group of Reoe Minerals
157
of alkyl hydroxamate for flotation of yttrium group REO can be improved by changing the alkyl group to iso-alcohol of fraction C12–C16, for example, isododecil alcohol: CH (CH2)3 CH CH2 CH CH2OH C2H2
C2H5
This hydroxamate is selective towards calcite, fluorite and sericite. The yttrium group minerals that contain zircon also have highly complex mineral compositions. These ores contain fergusonite, euxenite and priorit besides other minerals that contain REO. Such deposits are found in Northern Canada (Thor Lake). Limited research work has been conducted on these ores, but have indicated that REO cannot be recovered using either fatty acid or sodium oleate. It was, however, found that a mixture of sulphosuccinamate and phosphate ester modified with alkylsulphate can recover REO and zircon efficiently. Figure 24.5 shows the effect of above collector mixture (KBX3) on REO recovery from complex REO–ZrO2 ores. Oxalic acid and fatty acid (FA3) were not so effective compared to collector KBX3. As can be seen from the data shown in Figure 24.5, poor results were achieved using either fatty acid or sodium oleate collector. In the case of REO-containing zircon, there is a strong relationship between zircon recovery and the recovery of REO from the yttrium group of REOs. This relationship is illustrated in Figure 24.6.
100 Collector KBX3 80
REO recovery (%)
FA3 fatty acid 60
Sodium oleate 40
20
0 0
Figure 24.5
100
200 300 400 Collector addition (g/t)
500
600
Effect of different collectors on REO recovery from complex REO–ZrO2 ores.
158
24.
Flotation of REO Minerals
100
REO recovery (%)
80
60
40
20
0 0
20
40 60 Zircon recovery (%)
80
100
Figure 24.6 Relationship between zircon and REO recovery in the bulk zircon REO concentrate.
This is due to the fact that zircon present in these ores contains a portion of REO as inclusions in the mineral itself. In a number of cases, the REO from the yttrium group contains significant amounts of pyrochlore and/or tantalum columbite. Both minerals usually float with the zircon and REO minerals.
24.3
24.3.1
FLOTATION PRACTICES AND RESEARCH WORK ON BENEFICIATION OF REO MINERALS
Introduction
A large portion of the REOs are produced from monazite- and bastnaesite-containing ores. In the majority of cases, bastnaesite and monazite ores are relatively complex and contain gangue minerals (calcite, barite, fluorite and apatite) with similar flotation properties as the monazite and bastnaesite. Monazite is also found in heavy mineral sands, which are usually recovered using physical concentration methods, such as gravity, magnetic and electrostatic separation. Some deposits in addition to REO contain zircon and titanium minerals. From these ores, REO and zircon can be recovered in bulk concentrate suitable for hydrometallurgical treatment.
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
24.3.2
159
Flotation practice in the beneficiation of bastnaesite-containing ores
The Mountain Pass (USA) operation treats a relatively complex ore. The major REO mineral is bastaenesite with minor amounts of synchisite, parasite and monazite. The major gangue minerals are calcite, barite, silicates, and dolomite. The amount of the individual gangue minerals in this ore are variable and change on a yearly basis. There are two major ore types treated at the Mountain Pass concentrator: (a) high calcite ore (35–45% CaO) and (b) a high barite–dolomite ore (so-called brown ore). Barite also contains significant quantities of strontium. Liberation of the Mountain Pass ore has been extensively studied on the mill feed ore and on the plant product. Grinding the ore to a K80 of about 56 µm is required to achieve liberation. Locking between the bastnaesite and calcite above 50 µm is common. Usually calcite/bastnaesite middlings reports to the final concentrate. Over the past 20 years, extensive studies were conducted in which different reagent schemes were evaluated. The following is a brief summary of the findings: • •
• •
Hydroxamic acid used as a collector has shown to give better selectivity than fatty acid. However, it has yet to be tested in an operating plant. Extensive work has been carried out to evaluate different fatty acids. There are contradictory conclusions among different researchers regarding the performance of different fatty acids. Studies performed by the US Bureau of Mines (Reno, NV, USA) confirmed that distilled acid gave results superior to those of linoleic acid or fatty acid containing rosin acid. Studies conducted at the University of New Mexico and at the Molycorp laboratory showed that distilled tall oil containing rosin acid gave results better than those of pure oleic acid. These differences are likely due to different flotation responses related to a variation in the mineralogy. With respect to different depressant studies, only a limited amount of work has been performed with Quebracho, tanic acid and different lignin sulphonates. Lignon sulphonates with a medium molecular weight were superior. Flotation temperature was the subject of numerous studies. It was concluded that heating the pulp with collector is the only way to selectively float bastnaesite. Heating the pulp with collector is believed to result in selective aggregation of bastnaesite in the form of repellent droplets, which may result in improved selectivity and in a reduction in slime interference.
The flowsheet used in the Mountain Pass, with reagent additions, is shown in Figure 24.7. The plant reagent scheme that is currently being used is presented in Table 24.3. Weslig is a lignon sulphonate with a molecular weight of about 20,000 and also contains ethylene oxide. Ethylene oxide serves the purpose of reducing the frothing properties of the Weslig and improves the Weslig depression efficiency, in particular, for barite. A typical example of metallurgical results obtained in the plant is shown below (Table 24.4). It should be noted that the plant results are variable and depend on the type of ore being treated. Typical distributions of REO in the Mountain Pass concentrate are shown in Table 24.5.
160
24.
Flotation of REO Minerals
Feed ore Na2CO3 Na2SiF6 Thermal conditioning Weslig Thermal conditioning Collector Thermal conditioning Collector Rougher
Scavenger
Weslig Na2SiF6 Conditioning Collector 1st cleaner
1st cleaner scavenger
Weslig
2nd cleaner
Weslig 3rd cleaner
Concentrate
Figure 24.7
Tailings
Mountain Pass (USA) plant flowsheet.
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
161
Table 24.3 Reagent scheme used at the Mountain Pass concentrator Reagent
Additions (g/t)
Soda ash (Na2CO3) Sodium fluorosilicate (Na2SiF6) Lignin sulphonate (Weslig) Tall oil fatty acid (P25A)
3000–4500 300–600 2400–3500 200–400
Table 24.4 Molycorp plant metallurgical results Product
Weight (%) Assays (%) REO
Final bastnaesite concentrate 9.38 Final bastnaesite tailing 90.62 Feed 100.00
% Distribution
Ce2O3 La2O3 BaSO4 CaO REO
64.1 31.4 2.28 1.06 8.09 3.9
22.2 0.74 2.76
2.7 26.3 26.3
Ce2O3
3.1 75.6 75.5 16.9 24.4 24.5 15.6 100.0 100.0
Table 24.5 Distribution of the REO in the Mountain Pass concentrate Element
% of Total REO content
Element
% of Total REO content
Lanthanum Cerium Praseodymium Neodymium Samarium Europium Gadolinium Terbium
33.2 49.1 4.34 12.0 0.790 0.118 0.166 0.0159
Dysprosium Holmium Erbium Thulium Ytterbium Lutetium Yttrium
0.0312 0.0051 0.0035 0.0009 0.0006 0.0001 0.0913
Beneficiation of barite, fluorite and bastnaesite from the Dong Pao deposit in Vietnam This ore is heavily weathered ore, with more than 30% of the bastnaesite contained in the –7 µm fraction. The major host minerals present in this ore are barite and fluorite. Table 24.6 shows the chemical analyses of the ore used in various research studies. The ore deposit is located in the Lai Chan Province of Vietnam, and was developed by Sumitomo Metal Mining Company (Japan).
162
24.
Flotation of REO Minerals
Table 24.6 Chemical analyses of the Dong Pao ore Element
Assays (%)
Total REO Cerium (Ce2O3) Lanthanum (La2O3) Barite (BaSO4) Fluorite (CaF2) Silica (SiO2) Alumina (Al2O3) Iron (Fe2O3) Calcium (CaO) Sodium (Na2O) Potassium (K2O) Titanium (TiO2) Phosphorus (P2O5) Manganese (MnO) Chromium (Cr2O3) Vanadium (V2O5) LOI
8.72 3.76 3.18 62.5 5.54 8.85 0.97 2.69 0.15 0.54 0.11 0.09 0.13 0.64 0.22 0.03 10.6
Because this ore was high in barite and fluorite, direct flotation of bastnaesite from the ore was not possible. It should be pointed out that fluorite has similar flotation properties as bastnaesite and depression of fluorite during bastnaesite flotation is difficult. Extensive research work [7] has been conducted on this ore, aimed at developing a commercial treatment process that would produce a high-grade REO concentrate. As a result, a unique flowsheet and reagent scheme were developed. The flowsheet that was developed for beneficiation of the Dong Pao ore involves sequential barite–fluorite–bastnaesite flotation. The flowsheet is presented in Figure 24.8. The ore was washed and deslimed before grinding. The fines from the washing con tained over 30% of the total bastnaesite present in the ore. The ground ore was first subjected to barite flotation followed by fluorite flotation. By floating the barite and fluorite ahead of the bastnaesite, about 70% of the total weight was removed from bastnaesite flotation feed. The bastnaesite flotation feed was upgraded from 8.5% REO to about 30% REO. The reagent scheme developed during extensive laboratory testing is presented in Table 24.7. This reagent scheme is unique in such a way that the collector and number of depressants involved are composed of a number of chemicals that provide improved selectivity during sequential flotation of barite and fluorite from bastnaesite. For flotation of barite, sodium silicate was used as a depressant and barium chlorite as a barite activator. Barite collector SR82 was composed of petroleum sulphonate, sodium alkyl sulphate and succinamate mixture. The collector was selective towards both fluorite and bastnaesite. Over 96% of the barite was recovered in a relatively high-grade concentrate.
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
163
Feed Grinding
Conditioning
Conditioning
CaF2 rougher
BaSO4 rougher
CaF2 scavenger
BaSO4 scavenger CaF2 1st cleaner
BaSO4 1st cleaner
BaSO4 1st cleaner scavenger CaF2 2nd cleaner
BaSO4 2nd cleaner
BaSO4 3rd cleaner
CaF2 3rd cleaner
CaF2 concentrate
Thermal conditioning BaSO4 4th cleaner
Thermal conditioning Thermal conditioning (75°C)
REO rougher
REO scavenger
BaSO4 concentrate
REO 1st cleaner REO tailings
REO 2nd cleaner
REO 3rd cleaner
REO concentrate
Figure 24.8
Flowsheet developed for beneficiation of the Dong Pao ore.
During fluorite flotation, Quebracho and lignin sulphonate mixture (MESB) was used with collector composed of a mixture of oleic acid and phosphoric ester. Collectors used for bastnaesite flotation included tall oil fatty acid modified with three ethylene tetra
164
24.
Flotation of REO Minerals
Table 24.7 Reagent scheme developed for beneficiation of the Dong Pao ore Reagent
Additions (g/t)
Depressants and modifiers Na2SiO3 BaCl2 NaF Al2(SO4)3 MESB Na2CO3 Citric acid MM4 Collectors SR82 AKF2 KV3 Fuel oil
BaSO4 circuit
CaF2 circuit
Ro
Cl
Ro
Cl
Ro
Cl
2500 500 – – – – – –
1200 400 – – – – – –
1500 – 300 600 20 – – –
1100 – 400 400 200 – – –
– – – – – 4000 1000 1000
– – – – – 1400 3500 1300
850 – – –
– – – –
– 300 – –
– – – –
– – 900 200
– – 200 –
REO circuit
amine. Depressant MM4 was a mixture of lignin sulphonate with a molecular weight ranging from 9000 to 20,000. The results obtained from the continuous locked-cycle tests are summarized in Table 24.8. The major contaminant of the bastnaesite concentrate was fluorite. Complete fluorite flotation was not possible without heavy losses of bastnaesite in the fluorite concentrate.
Table 24.8 Continuous locked-cycle test results Product
BaSO4 Cl concentrate CaF2 Cl concentrate REO Cl concentrate REO combined tail Head (calc) BaSO4 Cl concentrate CaF2 Cl concentrate REO Cl concentrate REO combined tail Head (calc)
Weight (%)
62.83 7.36 13.72 16.09 100.00 62.83 7.36 12.97 16.84 100.00
Assays (%)
% Distribution
BaSO4
CaF2
REO
BaSO4
CaF2
REO
95.8 3.11 8.55 3.47 62.2 95.8 3.11 6.55 4.00 62.0
0.67 44.4 14.8 0.50 5.80 0.67 44.4 16.6 0.61 5.94
0.61 7.57 45.9 6.79 8.33 0.61 7.57 48.4 6.13 8.25
96.8 0.4 1.9 0.9 100.0 97.2 1.4 1.4 1.1 100.0
7.3 56.3 35.0 1.4 100.0 7.1 54.9 36.2 1.7 100.0
4.6 6.7 75.6 13.1 100.0 4.7 6.7 76.1 12.5 100.0
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
24.3.3
165
Flotation practices in beneficiation of monazite
A large portion of monazite production comes from mineral sand deposits. In the beneficiation of monazite from mineral sand deposits that contain garnet, ilmenite, shell and silicates, the physical concentration and combination of physical preconcentra tion–flotation is used. Several reagent schemes using flotation were developed throughout various studies [8–10] and some have been confirmed in continuous pilot plants. Flotation of the Indian beach sand (monazite) India has very large deposits of monazite on the coastal shores of Kerala and Chennai. A typical mineral composition of this type of deposit is 60% ilmenite, 1.2% rutile, 5% zircon, 6.4% garnet, 4% silinanite, 16% quartz, 2.5–5% monazite and 1–7% shell. Research work involved different anionic collectors and pH during monazite flotation, along with the level of sodium silicate used as depressant. Experimental work conducted at different levels of sodium silicate (Table 24.9) indicates that sodium silicate is an excellent depressant for titanium, zircon and other gangue minerals while the monazite flotation is not affected. The collector used in this experiment was sodium oleate at additions of 300 g/t. In addition to sodium oleate, other fatty acid collectors were examined. The results are given in Table 24.10. From these data, the saturated fatty acid soap was a poor collector for monazite, as well as sodium laurate. The acintols (mixture of oleic and linoleic acids) were found to give better results compared to sodium oleate. This can be attributed to the presence of linoleic acid, which has two double bonds. Furthermore, the rate of monazite flotation increased with the acintol than with the sodium oleate. The monazite concentrate in these experiments contained some garnet and sillinmanite. In conclusion, it can be noted that the effect of pH on flotation of beach sand minerals is critical in selective flotation of monazite from other minerals. Table 24.9 Effect of sodium silicate on monazite flotation from Kerala and Chennai beach sand (India) Reagent additions (kg/t)
Flotation pH
Na2SiO3 NaOH 1 3 5 7 9 11
2.2 3.0 5.5 6.5 9.0 8.5
9.2 9.4 9.6 9.7 9.8 9.8
Monazite concentrate
Monazite tailings
Weight (%)
% Grade % Recovery Weight (%)
% Grade % Recovery
3.2 10.4 8.3 6.6 5.6 4.8
23.9 33.3 66.2 76.2 84.4 94.3
2.54 0.95 0.28 0.24 0.24 0.40
13.3 37.5 88.4 92.3 85.7 83.6
96.8 89.6 91.7 93.4 94.4 95.2
33.7 62.5 11.6 8.3 5.6 4.8
166
24.
Flotation of REO Minerals
Table 24.10 Effect of different collectors on monazite flotation from the Chennai beach sand Collector type Addition Monazite concentrate Monazite tailings (kg/t) Weight % Grade % Recovery Weight (%) % Grade % Recovery (%) Sodium laurate Sodium oleate Neofat 140 Acintol FA1 Acintol FA2 Acintol FAX
11.4 5.5 5.5 5.0 5.0 5.0
5.0 8.3 9.0 6.1 5.6 5.8
21.4 66.2 57.0 75.5 81.6 71.0
20.0 88.4 89.0 86.4 89.2 77.0
95.0 91.7 91.0 93.9 94.4 94.2
4.6 0.28 0.12 0.23 0.16 0.16
80.0 11.6 11.0 13.6 10.8 23.0
The monazite can be selectively floated from other minerals when using Na2O:SiO2 (1:1) at relatively high doses (i.e. 5 kg/t). Processing of the black sand monazite at Rosetta The mineralogy of the Rosetta Nile black sand monazite is relatively complex and contains a variety of different minerals. Table 24.11 shows the chemical analysis of the run-of-mine ore. The size distributions of the black sand ranged from 80 to 100 µm. Development testwork on the black sand included an examination of anionic and cationic collectors. Cationic collectors, such as Amine 22, Armac and Armac T, gave poor results. Selectivity was poor, even when using modified starches as gangue depressants. Testwork using monazite depression with lactic acid and flotation of the residual miner als with 3-lauril amine hydrochloride achieved a concentrate grading 75.5% monazite at a recovery of about 70%. Table 24.11 Analyses of the run-of-mine black sand Element
Assays (%)
Silica (SiO2) Titanium (TiO2) Calcium (CaO) Magnesium (MgO) Zircon (ZrO2) Manganese (MnO) Iron (Fe2O3) Alumina (Al2O3) Sodium (Na2O) Potassium (K2O) Phosphorus (P2O5) Monazite (REO)
13.35 25.8 2.71 1.75 3.72 2.82 39.84 9.24 0.21 0.02 0.10 2.20
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
167
Carboxylic collectors from the carboxylate group These collectors were examined at a pH of 10 (Cyanamid 700 series) and diluted pulp to about 15% solids. A monazite recovery of over 95% was obtained when using Cyanamid collector 710. Monazite activation using oxalate Experimental work was carried out on black sand in which the effect of sodium oxalate on monazite activation was examined. It should be noted that monazite is essentially a phosphate of cerium and lanthanum, where the possibility exists that sodium oxalate has an activating effect on monazite [11]. The use of sodium oleate as activator was studied with different sulphonate collectors (Table 24.12). It was shown that with the use of sulphonate collectors, sodium oxalate had a positive effect on monazite grade and recovery. Conditioning time with oxalate had a pronounced effect on monazite recovery. Figure 24.9 shows the effect of conditioning time with oxalic acid on monazite recovery. The data from the figure show that 2–4 min of conditioning time was sufficient to achieve maximum recovery of monazite using different monazite collectors. Flotation of Brazilian monazite ore The Brazilian monazite ore is found as beach sand along rivers in the Sao Goncalodo Sapucai region. As mentioned earlier, the flotation characteristics of monazite, zircon and rutile are similar, and separation of these minerals is difficult. The objective of this research work was to find a reagent scheme that would selectively float the monazite from the associated minerals (zircon) and rutile. Sao Goncalo ore assayed approxi mately 2.9% total ROE, 36.6% TiO2, 7.68% ZrO2, 15.6% SiO2 and 24.6% FeT. Experimental testing was performed with hydroxamate and sodium oleate as collectors. The only depressant used was sodium meta-silicate. Comparison of results with the different collectors is shown in Table 24.13. Hydroxamate was more selective compared to the results obtained using sodium oxalate. Sodium oxalate, however, gave better recoveries.
Table 24.12 Effect of different collectors on flotation of monazite using sodium oleate as the activator Collector
Additions (g/t)
% Monazite concentration
% Monazite recovery
Sulphonate 231 Aeropromoter 710 R260 R376 R276R R376
900 4000 600 650 700 900
91.0 92.1 85.1 90.5 85.5 90.2
90.9 98.5 96.5 85.0 90.5 93.3
168
24.
Flotation of REO Minerals
100
Monazite recovery (%)
80
Collector
60
R260H R276F
40
R231 20
0 0
Figure 24.9 collectors.
1
2 3 4 5 Conditioning time (min)
6
7
Effect of conditioning time with sodium oxalate on monazite recovery using different
Table 24.13 Effect of different collectors on monazite flotation using Brazilian beach sand Reagents
Hydroxamate = 140 g/t Na2SiO3 = 1200 g/t Sodium oleate = 525 g/t Na2SiO3 = 1398 g/t
Product
Feed Rougher Conc Rougher Tail Feed Rougher Conc Rougher Tail
Weight (%)
100.00 4.90 95.10 100.00 5.66 94.34
Assays (%)
% Distribution
RE2O3
RE2O3
3.15 57.69 0.34 2.92 49.07 0.16
100.0 89.77 10.23 100.0 94.98 5.02
Monazite flotation from complex ores There are several large deposits of complex monazite ores, some of which are located in South Africa and Western Australia. Major research and development testwork has been performed on the Mount Weld ore from Western Australia. The Mount Weld ore is highly complex with about 50% of the monazite being contained in the –25 µm fines. Haematite, Fe-hydroxides, phosphates and alumosilicates are the principal gangue minerals present in this ore.
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
169
Table 24.14 Head analyses of the Mount Weld ore Element
Assays (%)
Total REO Cerium (Ce2O3) Lanthanum (La2O3) Samarium (Sm2O3) Yttrium (Y2O3) Iron (Fe2O3) Alumina (Al2O3) Magnesia (MgO) Calcium (CaO) Phosphorus (P2O5)
15.50 9.54 4.21 0.39 0.30 60.5 15.5 4.60 10.8 2.66
The head analyses of the ore are shown in Table 24.14. Research studies – Ore preparation The major task involved during ore preparation is to remove the maximum amount of primary slimes with minimum loss of REO minerals to the slime fraction. The REO losses in the slime fraction are dependent on the desliming size. Minimum loss of REO to the slime fraction occurs when desliming is done at a K80 of about 4 µm. Figure 24.10 shows the effect of desliming size on REO loss in the size fraction.
Monazite recovery in slimes (%)
50
40
30
20
10
0 2
Figure 24.10
4
6 8 10 Desliming size K80 (µm)
12
14
Effect of desliming size on monazite losses in the slime fraction using dispersant DQ4.
170
24.
Flotation of REO Minerals
Table 24.15 Effect of different dispersants from the DQ series on monazite loss in the slime fraction Desliming size (µm)
Dispersant
4 4.2 4.1 4.0 4.3 4.0
Slime fraction
Type
Additions (g/t)
Weight (%)
% Monazite assay
% Monazite recovery
None DQ2 DQ3 DQ4 DQ6 DQ8
– 800 800 800 800 800
25.0 23.3 23.1 21.5 22.2 23.4
17.8 15.6 13.3 9.4 12.0 11.8
28.7 23.4 19.8 13.0 17.1 17.8
The use of DQ4 in the desliming stage has a significant impact on monazite loss to the slime fraction. Table 24.15 shows the effect of different dispersants on monazite loss in the slime fraction, using dispersants from the DQ series. These dispersants are a mixture of low-molecular-weight acrylic acids modified with surfactant. The lower monazite losses in the slime fraction were achieved using dispersant DQ4. Mineralogical examination of the slime fraction, in which dispersants were used, revealed that about 80% of the slime was composed of Fe-hydroxides and ultrafine 2–3 µm clay.
100 4 kg/t 80 Monazite recovery (%)
2 kg/t
60
0 kg/t
40
20
0 0
Figure 24.11
10 20 30 40 50 Monazite concentrate grade (% REO)
60
Effect of Na2S on the monazite rade–recovery relationship.
24.3
Flotation Practices and Research Work on Beneficiation of Reo Minerals
171
Flotation studies Flotation studies were carried out on ground, deslimed ore. The optimum grinding fineness was about K80 = 65 µm. A variety of collectors and depressant systems were examined. Modified fatty acid collectors performed the best on the Mount Weld ore. The use of Na2S�9H2O in the conditioning had a significant effect on monazite grade and recovery. Figure 24.11 shows the relationship between monazite grade and recovery at different levels of Na2S additions. The final flowsheet and reagent scheme developed for beneficiation of the Mount Weld ore is shown in Figure 24.12 for grinding and desliming, and in Figure 24.13 for flotation. The desliming was performed in three stages at 15% pulp density to the desliming feed cyclone. During flotation, the pulp was conditioned with reagents at about 60% solids. Collector CB110 is composed of a mixture of fatty acids modified with hydrocarbon oil and then oxidized. The final results obtained in continuous operation are presented below (Table 24.16).
Feed Scrubbing
Coarse
Washing
Grinding
Fines
Fines
DQ4 300 g/t
Conditioning
Final slimes to tailings
Cyclones
to flotation
Figure 24.12
Final grinding and desliming flowsheet.
172
24.
Flotation of REO Minerals
Ground deslimed ore 1000 g/t Na2SiO3 600 g/t Na2CO3 Conditioning 1 500 g/t dextrose/quebracho 600 g/t collector CB110 Conditioning 2 2000 g/t Na2S Conditioning 3 200 g/t collector CB110 REO rougher
REO scavenger
200 g/t dextrose/quebracho 100 g/t DA663 1000 g/t Na2S Conditioning
REO
1st
200 g/t collector CB110 cleaner REO 1st cleaner scavenger
200 g/t dextrose/quebracho 2000 g/t Na2S Conditioning
REO 2nd cleaner
200 g/t dextrose/quebracho 100 g/t Na2S Conditioning
REO 3rd cleaner
REO cleaner concentrate
Figure 24.13
REO combined tailings
Final flotation flowsheet with points and levels of reagent additions.
References
173
Table 24.16 Overall metallurgical results obtained on the Mount Weld ore Product
Weight (%)
% Monazite assay
% Monazite recovery
Cleaner concentrate Combined tail Slimes Feed
20.89 54.51 24.6 100.00
58.5 2.55 8.8 15.8
77.5 8.8 13.7 100.0
REFERENCES 1. Ginsburg, I.E., Zuravleva, L.N., and Ivanov, E.B., Rare Earth Elements and their Origin, USSR Research Institute of Mineral Raw Materials, Moscow, 1959. 2. Polkin, C.I., Beneficiation of Precious Metals and Rare Mineral Ores, Publisher Nedra, Moscow, pp. 336–370, 1987. 3. Bulatovic, S, US Patent 4,772,238, Froth flotation of bastnaesite, September 20, 1988. 4. Bulatovic, S., Process Development for Beneficiation of Mount Weld REO Ore, Report of Investigation, 1990. 5. Bulatovic, S., Process Development for Beneficiation of Barite, Fluorite, Bastnaesite Ore from the Dong Pao Deposit, Vietnam, Report of Investigation, 1995. 6. Fishman, M.A., Sobolev, D.C., Practices in Beneficiation of Sulphide and Rare Metals, vol. V, Gosudarstvenie Naucno-tehnicheskie Lzdatelstro Nanche Literature Moskra, pp. 330–380, 1963. 7. Bulatovic, S., Process Development for Beneficiation of the Dong Pao Ore (Vietnam), Report of Investigation, April 2002. 8. Viswanathan, K.V., Madhavan, T.R., and Majumdar, K.K., Selective Flotation of Beach Sand Monazite, Mining Magazine, Vol. 13, No. 1, 1965. 9. Farah, M.Y., and Fayed, L.A., Oxalate Activation in the Flotation of Monazite by Heavy Sulphonate Collector, Egypt Journal of Chemistry, Vol. 1, p. 2363, 1958. 10. Pavez, O., and Perez, A.E.C., Bench Scale Flotation of Brazilian Monazite, Mineral Engineering, Vol. 7, No. 12, pp. 1561–1564, 1994. 11. Plaksin, I.N., Study of Superficial Layers of Flotation Reagents on Minerals and the Influence of the Structure of Minerals on their Interactions with Minerals, International Mineral Processing Congress, paper #13, London, 1960.
– 25 –
Flotation of Titanium Minerals 25.1
INTRODUCTION
Titanium is the most abundant metal in the earth crust, and is present in excess of 0.62%. It can be found as dioxy titanium and the salts of titanium acids. Titanium is capable of forming complex anions representing simple titanites. It can also be found in association with niobium, silicates, zircon and other minerals. A total of �70 titanium minerals are known, as mixtures with other minerals and also impurities. Only a few of these minerals are of any economic importance. This chapter discusses flotation properties of major titanium minerals and beneficiation methods used in some operating plants. In recent years, a new technology has been developed for beneficiation of hard rock titanium minerals. This is also discussed in this chapter.
25.2
TITANIUM-BEARING ORES AND MINERALS
Out of the 70 known titanium minerals, only a few have any economic value. Table 25.1 shows the major titanium minerals of value. The most important titanium minerals are ilmenite, rutile and perovskie. Loparite is a major mineral for production of niobium and REO. 25.2.1
Ilmenite
Ilmenite has variable titanium content, depending on the iron content and other impurities, and ranges from 45% to 52% TiO2. It can be found in a variety of ore types, along with rutile, apatite, zircon, columbite, etc. 25.2.2
Ilmenorutile
This mineral contains up to 53% TiO2 and 32% Nb2O5 along with 14.5% Ta2O5. The composition of ilmenorutile is variable, and often is considered to be a niobium mineral.
175
176
25.
Flotation of Titanium Minerals
Table 25.1 Important titanium minerals of economic value Mineral
Formula
Theoretical grade % TiO2
Specific gravity (g/cm3)
Hardness
Ilmenite Rutile Ilmenorutilea Perovskite Sphene Loparite Lucoxene Titanomagnetiteb
FeTiO3 TiO2 (Ti,Nb,Fe)O2 CaTiO3 CaO�TiO2�SiO2 (Na,Ce,Sr,Ca)(Nb,Ti)O2 TiO2.TiO2.SiO2 Fe3O4�FeTiO3b
52.6 100.0 53 58.9 40.8 39.2 50–95 2–30
4.6–5.2 4.3 4.6–5.1 4.0 3.3–3.6 4.7–5.0 3.3–4.3 4.5–5
5–6 6 6 5.5–6 5.6 5.5–6 5–6 5.5–6
a b
If Nb content is high, this mineral belongs to a group of niobium minerals. Or Fe3O4�TiO2.
25.2.3
Rutile
Rutile is the most stable of all the titanium minerals. In a number of cases, rutile may contain impurities such as iron oxides, tin, chromium and vanadium. The rutile grade can range from 95% to 99% TiO2. 25.2.4
Perovskite
Perovskite is a calcium–titanium mineral and usually contains impurities of iron, chromium and aluminium. The theoretical grade can vary from 50% to 57% TiO2. Also, sometimes contains niobium (up to 11%) and tantalum. 25.2.5
Leucoxene
Leucoxene has a composition similar to that of rutile, and is a product of alterations of a number of titanium minerals, most often ilmenite and sphene. It contains higher amounts of titanium, compared to ilmenite, and can range from 61% to 75% TiO2.
25.3
CLASSIFICATION OF TITANIUM DEPOSITS
Titanium minerals have been recovered from both hard rock and sand deposits. Until 1945, most of the ilmenite and rutile produced commercially came from sand deposits, but nowadays, the production of ilmenite from rock deposits exceeds that of sand deposits. Rutile, however, is exclusively produced from sand deposits, although a new technology exists that recovers rutile from rock deposits.
25.4
Flotation Properties of Major Titanium Minerals
25.3.1
177
Rock deposits
Anorthositic deposits – nearly all of the known commercially important rock deposits of titanium minerals are associated with anorthositic or gabbroic rocks. There are three main types: (a) ilmenite–magnetite (titanoferous magnetite), (b) ilmenite–haematite, and (c) ilmenite–rutile. The ilmenite–magnetite deposits usually contain ilmenite and magnetite as granular intergrowths, which can be separated readily to yield ilmenite and magnetite concentrates. The ilmenite–haematite deposits usually contain these minerals in intimate intergrowths, and hemo-ilmenite concentrates can be produced from these ores. Miscellaneous deposits – there are a number of deposits around the world (USA, Canada, Brazil, Chile, etc.) with a variety of ore types, some of which have been extensively studied. Such deposits include (a) deposits of ilmenite disseminated in schist, (b) complex deposits of apatite–ilmenite (Canada) and (c) deposits of rutile, anastase and brookite in a pegmatic phase of alkaline rocks (USA, Chile). The major occurrence of anatase and ilmenite, found in weathered carbonatite bodies, are found in Brazil. Occurrences of rutile and ilmenite in carbonatite–feldspar rocks are found in Mexico and Chile, and in recent years have been subject to extensive investigations. 25.3.2
Sand deposits of titanium minerals
The most abundant titanium sand deposits are black sands in streams and on beaches of volcanic regions. The principal black minerals are magnetite, titanoferous magnetite and black silicates, chiefly angite and hornblend. It is quite difficult to produce an ilmenite suitable for pigment product from black sand, but other sand deposits that contain rutile, ilmenite and often monazite are found in Australia, USA, India and Africa. These deposits are either alluvial or marine in origin. From a beneficiation point of view, formation of hard rock and sand deposits, and their mineral composition, determines the beneficiation method.
25.4
FLOTATION PROPERTIES OF MAJOR TITANIUM MINERALS
Extensive research has been carried out mainly on ilmenite and, to a lesser degree, on flotation of rutile and perovskite. Flotation studies have been performed on titanium minerals from both hard rock and fine-grained sand deposits. 25.4.1
Flotation properties of ilmenite
Extensive research work has been carried out on ilmenite flotation from different ores [1–3], including hard rock and sand deposits. Because the chemical composition of ilmenite is unstable, flotation processing characteristics of ilmenite varies from one ore type to another. Figure 25.1 shows the flotation of ilmenite from different ore types at different pH levels using 200 g/t of oleic acid.
178
25.
100
Flotation of Titanium Minerals
Ore type 3
80
TiO2 recovery (%)
2 1
60
40
20
0 2
3
4
5
6
7
8
9
10
Flotation pH
Figure 25.1
Effect of pH on ilmenite flotation from different ore types using oleic acid as collector.
The data from Figure 25.1 indicate that ilmenite can be recovered at a wide pH range. There is, however, a difference in the floatability of ilmenite from different ore types. Ilmenite can be successfully floated using fatty acid tall oil collectors at alkaline pH or with sodium alkyl sulphate (C16H33OSO3Na) at acidic pH. Figure 25.2 shows the effect of pH on ilmenite flotation from a sand deposit using alkyl sulphate collector. Acid pretreatment of the ore before flotation had a positive effect on ilmenite flotation. Figure 25.3 shows the effect of different acids used in the pretreatment on ilmenite recovery in the rougher concentrate. The best metallurgical results were achieved using sulphuric acid in the pretreatment stage. Another collector examined for flotation of ilmenite was dodecylammonium chloride. Using this collector, ilmenite readily floated at a pH region between 3.5 and 6.5. The type of gangue depressant and modifier used during ilmenite flotation depends on the type of gangue present in the ore. Sodium silicate is commonly used as a gangue depressant. In a recent study [4], it was demonstrated that the effectiveness of silicates as depressants improved significantly with the use of acidified silicate. Figure 25.4 shows the effect of acidified silicate on the ilmenite grade–recovery relationship. Acidified silicate gave significantly better concentrate grades, compared to that obtained using silicate alone. It has been found that the use of Pb(NO3)2 as an ilmenite activator improved ilmenite floatability and selectivity towards gangue minerals. Experimental work was conducted on
25.4
Flotation Properties of Major Titanium Minerals
179
100
Fluorite
Recovery (%)
80
Ilmenite
60
40
20
Sphene 0 1
Figure 25.2
2
3
4
5 6 Flotation pH
7
8
9
10
Effect of pH on ilmenite flotation from mineral sands using alkyl sulphate as collector.
100
TiO2 recovery (%)
90
H2SO4 HCl
80
HNO3 70
60
50
40 0
500
1000
1500
2000
2500
3000
Acid addition (g/t)
Figure 25.3 Effect of type and level of acid in the acid pretreatment stage on titanium rougher flotation.
180
25.
Flotation of Titanium Minerals
100 Acidified Na2SiO3 addition
TiO2 recovery (%)
80
400 g/t
60
200 g/t 40 0 g/t 20
0 10
20
30
40
50
60
TiO2 concentrate grade (%)
Figure 25.4 Effect of acidified silicate additions on the ilmenite grade–recovery relationship.
Table 25.2 Results of ilmenite activation flotation using Pb(NO3)2 Product
Pb(NO3)2 (g/t)
Weight (%)
% TiO2 Assay
% TiO2 Distribution
Concentrate Middling 1 Middling 2 Middling 3 Tailings Feed Concentrate Middling 1 Middling 2 Middling 3 Tailings Feed
60
49.4 18.7 14.3 9.3 8.3 100.0 38.4 15.7 13.7 9.3 22.9 100.0
36.7 5.8 8.4 5.7 10.7 21.8 36.1 14.6 12.4 8.5 11.8 21.4
83.0 5.0 5.5 2.5 4.0 100.0 65.0 10.7 7.9 3.7 12.7 100.0
0
both beach sand and hard rock ilmenite. Table 25.2 compares batch test results obtained with and without the addition of Pb(NO3)2. Significant improvement in ilmenite recovery was realized when using small additions of Pb(NO3)2. The concentrate grade was similar in both experiments.
25.4
Flotation Properties of Major Titanium Minerals
25.4.2
181
Flotation properties of rutile
Flotation processing characteristics of rutile from hard rock ore and sand deposits are very much dependent on two major factors: (a) mineral composition of the ore and (b) impurities present in the rutile. Although it has been pointed out by some researchers that rutile can be floated using oleic acid, sodium oleate or other fatty acids in neutral pH, this is not the case when the ore contains calcite, feldspars and olivine. Most recently, a study of rutile ore from Chile containing feldspar indicated that rutile cannot be recovered using fatty acid as collector. Table 25.3 shows the effect of different collectors on rutile from an ore that contains calcite, feldspar and silicate as the major gangue minerals. The results indicated that modified sulphosuccinamate and a mixture of phosphate esters and sulphosuccinamate gave good results. However, using fatty acid did not effectively float the rutile. The sulphosuccinamate collector was extremely effective in flotation of rutile, as well as ilmenite and zircon from a fine sand deposit. Laboratory testing conducted on Wimmera heavy mineral sand from Australia indicated that the use of sulphosuccinamate achieved a high titanium recovery in the bulk cleaner concentrate. Table 25.4 shows the results obtained on the Wimmera heavy mineral sand. The sand was scrubbed and deslimed before flotation. Between 90% and 95% TiO2 was recovered using a 60 g/t addition of succina mate collector. Research has also been conducted in which steryl phosphonic acid (SPA) was examined in place of benzyl arsonic acid (BAA), which was used in an operating plant in China [5]. In this study, several collectors were examined, including sodium laurate, sodium dodecyl sulphate, amino acids, diphosphonic acid (SPA). It was discovered that SPA was the most effective and that aliphatic alcohol (i.e. octanol) was required to maintain the effectiveness of SPA. The use of emulsifier in the mixture was required to provide a suitable emulsion of the composite collector. A composite collector blended with a 1:1 ratio of SPA and octanol was found to be an effective collector for flotation of hard rock rutile ores.
Table 25.3 Effect of different collectors on rutile flotation from the Cerro Blanco rutile ore from Chile Collector
Oleic acid Tall oil fatty acid Sodium oleate Succinamate Phosphoric acid mixture
Addition (g/t)
Feed (% TiO2)
Rougher concentrate
Rougher tailing
% TiO2 Assay
% TiO2 Recovery
% TiO2 Assay
% TiO2 Recovery
3.31 3.22
25.5 20.8
48.5 44.3
1.81 1.90
51.5 55.7
3.10 3.28 3.10
18.6 40.5 46.6
39.5 88.6 96.5
2.01 0.40 0.11
60.5 11.4 3.5
182
25.
Flotation of Titanium Minerals
Table 25.4 Effect of succinamate collector on titanium rutile flotation using Wimmera heavy mineral sand from
Australia
Sand type
Product
East pit sand
South pit sand
25.4.3
Bulk cleaner concentrate Bulk rougher concentrate Bulk rougher tail Head (calc) Bulk cleaner concentrate Bulk rougher concentrate Bulk rougher tail Head (calc)
Weight (%)
7.81 10.29 89.71 100.00 8.36 9.00 91.00 100.00
Assays (%)
% Distribution
TiO2
ZrO2
TiO2
ZrO2
34.2 26.07 0.31 2.96 42.29 39.44 0.16 3.70
11.11 8.43 0.026 0.89 11.74 10.29 <0.01 0.99
90.3 90.6 9.4 100.0 95.7 96.1 3.9 100.0
97.0 97.4 2.6 100.0 99.0 99.1 0.9 100.0
Flotation properties of perovskite
A large deposit of perovskite was found recently in the USA (Powderhorn). Perovskite deposits are also known to be found in Russia (Cola Pennisula). There is little information available on research into flotation of perovskite conducted on ores from some Russian deposits [6]. These ores are relatively complex and contain a variety of gangue minerals including pyroxene, amphibole, olivine, nepheline, biotite and calcite. Flotation of perovskite was achieved with pretreatment of the flotation feed with H2SO4 followed by perovskite flotation with oleaic acid at a pH of 6.5–7.5. The use of sodium silicate as a depressant resulted in an increase in concentrate grade, up to �47% TiO2. Pilot plant tests on a perovskite ore showed that a perovskite concentrate assaying 48.5% TiO2 can be readily produced using distilled tall oil as collector. Most recently, development testwork was performed on a large perovskite deposit (Powderhorn) located in the USA. An effective beneficiation process was developed, where a concentrate assaying >50% TiO2 was achieved in the pilot plant confirmation tests [7]. During this development testwork, a number of different collectors were exam ined at different pH values. Figure 25.5 shows the effect of the different collectors on perovsikte flotation. The most effective collector was phosphoric acid ester modified with either fatty alcohol sulphate or petroleum sulphonate.
25.5
PRACTICES IN BENEFICIATION OF TITANIUM ORES
A large portion of titanium minerals (ilmenite, rutile) are produced from heavy mineral sands using physical preconcentration methods including gravity, magnetic and electro static separation. Over the past 30 years, advances have been made using flotation, where ilmenite, rutile and perovskite can be effectively recovered from both heavy mineral sands and hard rock ores using flotation methods.
25.5
Practices in Beneficiation of Titanium Ores
183
100
TiO2 recovery (%)
80
60
40
20 Phosporic acid ester modified with fatty alcohol sulphate
Phosphoric acid ester, unmodified Sodium oleate Tall oil fatty acid 0 1
2
3
4
5
6 7 8 Flotation pH
9
10
11
12
Figure 25.5 Effect of different collectors on perovskite flotation at different pHs (Powderhorn ore from the USA).
25.5.1
Practices in beneficiation of ilmenite ores using flotation
Titania A/S, Norway This is one of the oldest operations in the world. The mine and plant are located in the southern part of Norway. This ore can be classified as an ilmenorutile, with ilmenite and magnetite as the valuable minerals. The gangue consists mainly of feldspar, hypersthene and biotite. The secondary minerals present in this ore include pyrite, olivine and pyrrhotite. There are two major ore bodies: the Stogargen (old deposit) and Zellnes deposits. These two deposits are quite different in mineral composition. Numerous studies have been carried out on these two ore types to provide support for the operating plant. Over a period of years, the Titania A/S flowsheet has changed as the ore in the plant changed. The flowsheet that is currently being used is shown in Figure 25.6. This flowsheet utilizes a two-stage flotation method, where in stage 1, pyrite and apatite are recovered, followed by ilmenite flotation in stage 2. Before sulphide flotation, magnetite was removed using a low-intensity magnetic separation method. The reagent scheme used at the Titania A/S plant is shown in Table 25.5. The major problem associated with beneficiation of this ore was the fact that the apatite tended to float with the ilmenite concentrate. Two options were examined to control apatite flotation: (a) apatite flotation in the pyrite circuit using small amounts of tall oil, and (b) use of NaF to
184
25.
Flotation of Titanium Minerals
Feed Grinding Magnetic separation
Nonmagnetics
Conditioning
Conditioning
Pyrite flotation
P2O5 flotation
Pyrite cleaner
P2O5 cleaner
Pyrite concentrate
P2O5 concentrate
Magnetics to magnetics plant
Desliming
Slimes
Conditioning
TiO2 rougher
TiO2 scavenger
TiO2 1st cleaner
TiO2 1st cleaner scavenger
TiO2 2nd cleaner
TiO2 3rd cleaner
TiO2 cleaner concentrate
Figure 25.6
Titania A/S generalized plant flowsheet.
Tailings
25.5
Practices in Beneficiation of Titanium Ores
185
Table 25.5 Titania A/S reagent scheme Reagent additions (g/t)
pH
Pyrite-apatite
Ilmenite
Pyrite
Ilmenite
Depressants and modifiers Sodium carbonate NaF Nafaril emulsifier Fuel oil
0–200 – 5 75
– 200–300 60–100 500–600
9 – – –
5.5 – – –
Collectors Tall oil (refined) Ethyl xanthate
50–75 50
600–1300 –
– –
– –
Reagent
depress the apatite during the ilmenite cleaning operation. Both methods were capable of lowering the apatite content of the ilmenite concentrate, with an appreciable loss of ilmenite. In the early 1980s, Nobel (a reagent manufacturing company) developed selective apatite collectors (Lilaflot series) based on modified fatty acids, which were capable of removing apatite without any loss of ilmenite. The pH in the ilmenite circuit was controlled with the use of sulphuric acid. In 1980, the tall oil used in the pyrite circuit was replaced with Lilaflot 100 (modified fatty acid). The metallurgical results obtained in the plant are variable with respect to ilmenite recovery. The concentrate grade is usually maintained constant at about 44% TiO2, while ilmenite recovery ranges from 66% to 75% TiO2. Otanmaki, Finland The Otanmaki ore contains about 35% magnetite, 28% ilmenite, 1% pyrite and 35% silicate minerals. This ore contains an appreciable amount of fine ilmenite, most of which reports to the slime fraction. About 10–12% of the total ilmenite in this ore reports to the slime fraction. Initially, this plant was operated using a standard flowsheet involving three-stage desliming followed by pyrite flotation and ilmenite flotation from the pyrite tailing. Research work was carried out [8] to examine possible recovery of ilmenite using an agglomerated flotation method. The major objective of this study was to float the ilmenite without desliming using an agglomeration process. The major variables exam ined in this study included conditioning time and type of emulsifying agent. The conditioning time and conditioning power were critical in achieving high ilmenite recoveries. Figure 25.7 shows the effect of conditioning time and conditioning power on ilmenite recovery. Good results were achieved using attrition conditioning at reduced time. With the standard long conditioning time, up to 50 min was required to achieve a recovery of 90% TiO2.
186
25.
Flotation of Titanium Minerals
100 Conditioner
TiO2 recovery (%)
90
Attritioning
80
70 Standard 60
50 0
10
20 30 40 Conditioning time (min)
50
60
Figure 25.7 Effect of conditioning time and power on ilmenite recovery using agglomeration flotation.
From the various emulsifiers examined, an anionic emulsifier from the sulphonic acid group of polyglycol-ether of fatty alcohol and alkylphenol-polyglycol esters was used. The best results were achieved with the use of alkylphenol-polyglycol ester (Berol EMU27). The Otanamki ilmenite flowsheet without desliming is shown in Figure 25.8. This flowsheet has replaced the flowsheet that incorporated the desliming stage in late 1959. The reagent scheme used in this plant for agglomeration flotation included 800 g/t tall oil, 1500 g/t fuel oil, 800 g/t tall oil emulsion, 60 g/t Etoxol P19 and 50 g/t xanthate. The pH in the rougher flotation was maintained at 4.5 and the cleaners at 3.5 using H2SO4. A concentrate grade of 44% TiO2 at a recovery of 88% was produced using agglomera tion flotation, compared to a concentrate grade of 44% TiO2 with a recovery of 74% without agglomeration flotation. 25.5.2
Beneficiation of apatite–ilmenite ores (Sept Iles Mine, Canada)
In the late 1990s, extensive research work was carried out on a number of complex ilmenite ores resulting in the development of new technology capable of producing good-quality ilmenite concentrate with a respectable recovery. This section describes the treatment process that was developed for apatite–ilmenite ores using new technology [9].
25.5
Practices in Beneficiation of Titanium Ores
187
Feed Magnetic separation
Grinding
Magnetics
Non magnetics Pyrite flotation Conditioning 1 Conditioning 2 Conditioning 3
TiO2 rougher
TiO2 scavenger
TiO2 1st cleaner
TiO2 2nd cleaner
TiO2 3rd cleaner
TiO2 4th cleaner
TiO2 cleaner concentrate
Figure 25.8
Final tail
Generalized Otanamki plant flowsheet.
The Sept Iles ore contains economic quantities of apatite and ilmenite. About 6% of the titanium in this ore is represented by titanomagnetite. The major gangue minerals include feldspar, olivine, dolomite and aluminosilicate. This ore assayed 7.2% TiO2 and 4.25% P2 O 5 . During research development testing, a fairly large number of collectors were examined, mainly phosphoric acid esters that were modified with different secondary collectors. Figure 25.9 shows the effect of different collectors and pHs on ilmenite flotation.
188
25.
Flotation of Titanium Minerals
100
TiO2 recovery (%)
80
Collector
60
R260H R276F
SM14 phosphoric acid ester modified with alkyl sulphate SM15 phosphoric acid ester modified with petroleum sulphonate Mixture of phosphoric acid ester with succinamate
40
20
R231
0 2
3
4
5
6
7
8
9
Flotation pH
Figure 25.9
Effect of different collectors and pHs on ilmenite flotation.
Based on data shown in Figure 25.9, ilmenite recovery was a function of both pH and collector modifications. The optimum flotation pH was between 3 and 5. Phosphoric acid esters modified with petroleum sulphonate gave the highest recovery. Tall oils were also tested, but without success, as they were unselective towards olivine. A number of different depressants were also examined, mainly organic and inorganic mixtures, some of which had a pronounced effect on both apatite and ilmenite. The flowsheet used for apatite flotation is shown in Figure 25.10. The ore was ground to a K80 of 80 μm, followed by magnetic separation. The non-magnetic fraction was subjected to apatite flotation and upgrading. The ilmenite flowsheet is shown in Figure 25.11 and was specifically designed to reduce recirculation loads of gangue during cleaning operations. The reagent scheme that was developed for beneficiation of this apatite–ilmenite ore is shown in Table 25.6. The following is a description and function of the individual reagents: • Caustic tapioca starch was used for depression of ilmenite and iron oxides during flotation of apatite. • Soda ash was used for pH control. • Depressant A4 was used for depression of silicates, feldspar and olivine. This mixture consists of acidified silicate and ferrous sulphate (FeSO4) in a 90:10 ratio. This mixture is also highly effective in depressing silicates and olivine. • Fatty acid (FA2) was used as an apatite collector in saponified form.
25.5
Practices in Beneficiation of Titanium Ores
Ore
189
Combined non-magnetics
Grinding Conditioning 1 Magnetic separation
Conditioning 2
Magnetics P2O5 rougher
P2O5 scavenger
Regrinding
Mgnetic separation
Regrinding
Cleaner magnetics P2O5 1st cleaner
P2O5 1st cleaner scavenger
Desliming
Sand
P2O5 2nd Cleaner
Slime rd
P2O5 3 cleaner
P2O5 4th cleaner
P2O5 cleaner concentrate
Figure 25.10
P 2 O5 final tail
Grinding, magnetic separation and apatite flotation flowsheet.
• Oxalic acid was used for gangue depression during ilmenite flotation as a primary depressant. • Depressant SHQ was used as a tertiary depressant, mainly for magnesium-bearing minerals in the final TiO2-cleaning stages. It consisted of a mixture of Calgon glass, sodium silicate and Quebracho in a ratio of 40:40:20. • Acidified silica/AQ55D mixture was used as primary depressant during ilmenite flotation. This mixture consisted of 70% acidified silicate a 30% AQ55D reagent. • NaOH was used in the alkaline conditioning pulp pretreatment stage. • HCl was used in the acid pretreatment stage as pH modifier in the ilmenite flotation and cleaning stages. • Collector mixture D consisted of SM15/R845/R825 in a ratio of 45:45:10 and was used as the primary ilmenite collector.
190
25.
Flotation of Titanium Minerals
Feed NaOH conditioning
Desliming
Acid conditioning
Desliming
Slime 1
Slime 2
Conditioning 1 Conditioning 2
Ti rougher
Ti scavenger
Ti 1st cleaner
Ti 1st cleaner scavenger
Total tail
Ti 2nd cleaner
Ti 3rd cleaner
Ti 4th cleaner Acid conditioning Ti 5th cleaner Desliming
Ti 6th cleaner
Slime 3
Ti scalper
Ti cleaner concentrate
Figure 25.11 Ilmenite flotation flowsheet.
The metallurgical results obtained in a continuous pilot plant operation are presented in Table 25.7. Excellent apatite results were achieved. An ilmenite concentrate was produced suitable for pigment production.
25.5
Practices in Beneficiation of Titanium Ores
191
Table 25.6 Reagent scheme developed for beneficiation of apatite–ilmenite ore from the Sept Iles mine Reagent
Additions (g/t) P2O5 circuit
TiO2 circuit
Total
Depressants and modifiers Caustic starch Na2CO3 A4 HCl Oxalic acid Acidified silicate/AQ55D NaOH SHQ
1200 1000 300 – – – – –
– – – 2500 800 700 500 20
1200 1000 300 2500 800 700 500 20
Collectors and frothers FA2 (saponified) SM15/CA540/R825 MIBC
1200 – –
– 200 5
1200 200 5
Table 25.7 Pilot plant results Product
Weight Assays (%) (%)
% Distribution TiO2
TiO2 P2O5 Fe2O3 SiO2 MgO Overall Circ 9.78 P2O5 cleaner concentrate 7.66 TiO2 cleaner concentrate TiO2 combined 82.56 tail Head (calc) 100.00
0.52 40.7 47.6
0.92
0.06 53.3
P2O5 Fe2O3 SiO2
0.40 0.08
0.8
–
93.6
0.2
0.1
1.65 1.06
60.0
67.3
0.1
10.6
0.4
32.7
6.3
89.2
99.5
3.16
0.32 41.6
35.0
–
40.0
6.31
4.25 38.52 29.1
–
100.0
100.0 100.0 100.0 100.0
25.5.3 Ilmenite production from heavy mineral sands and chromium problems The ilmenite production from heavy mineral sands exclusively utilizes a physical separation method using magnetic separation, gravity concentration and electrostatic separation. Flota tion is practiced mainly for beneficiation of fine mineral sands containing rutile, ilmenite and zircon. The ilmenite that is produced in a number of operations in Western Australia, India and the USA is high in chromium, which makes the ilmenite unusable. This section discusses a new process that was developed for chromium removal from ilmenite concentrates.
192
25.
Flotation of Titanium Minerals
A sample used for testing was an ilmenite concentrate from Western Australia that assayed 0.4% Cr2O3 and about 58% TiO2, where the chromium in the concentrate was in the form of chromspinel with small quantities of chromite. Another sample used in the development testwork was ilmenite concentrate that only contained chromite. It was a known fact that flotation properties of both chromite and ilmenite are similar and they float equally well using either tall oil or amine collectors. Development testwork involved the examination of different ilmenite depressants and different chromium collec tors. Depressants examined in this study included corn starch, NaF and H2SiF6 at a low pH. Good ilmenite depression was achieved using H2SiF6, while the chromium was not affected. Similar results were achieved using NaF. A number of different chromium collectors were also examined, including R84, which is a sulphonate collector as the primary collector, and amine acetate as the secondary collector was found to be effective for chromium flotation. The most critical parameter for selective chromium flotation was the pH. Selective chromium flotation occurs at a very narrow pH region, 1–2.5. Figure 25.12 shows the effect of pH on chromium flotation. The final flowsheet that was developed for chromium removal is shown in Figure 25.13. The concentrate was scrubbed with alkaline followed by desliming. The deslimed concen trate was subjected to chromium flotation followed by a single cleaning stage. The reagent scheme that was developed for chromium flotation is shown in Table 25.8. The pH control was achieved using nitric acid. The presence of nitric acid appeared to improve selectivity. The results obtained with HCl and H2SO4 were not as good as those achieved using HNO3. Final metallurgical results obtained using selective chromium flotation, from an ilmenite concentrate, are shown in Table 25.9. An average of 80% Cr2O3 was removed from the ilmenite concentrate. The chromium assays of the ilmenite concentrate were reduced from 0.4% to 0.09% Cr2O3.
Cr2O3 recovery to its concentrate (%)
100
80
60
40
20
0 1
Figure 25.12
2
3 Flotation pH
4
5
Effect of pH on chromium flotation from an ilmenite concentrate.
25.5
Practices in Beneficiation of Titanium Ores
193
Ilmenite Concentrate
Scrubbing Slimes
Desliming
Conditioning 1 Conditioning 2
Cr2O3 rougher
Conditioning
Cr2O3 scavenger
Conditioning
Cr2O3 cleaner
Cr2O3 product
TiO2 product
Figure 25.13
Chromium flotation flowsheet.
Table 25.8 Chromium removal reagent scheme Reagent
Additions (g/t)
pH
Scrubbing
Cr2O3 Flotation
Ro
Cl
Depressants and modifiers NaOH HNO3 H2SiF6 Corn starch
200 – – –
– 500–800 800–1200 100–150
1.3 – – –
1.2 – – –
Collectors R840 Armac C
– –
250 40
– –
– –
194
25.
Flotation of Titanium Minerals
Table 25.9 Chromium flotation metallurgical results Test no.
A
B
Product
Weight (%)
Cr2O3 concentrate Cr2O3 tailing Head (calc) Cr2O3 concentrate Cr2O3 tailing Head (calc)
25.6
11.34 88.66 100.00 10.73 89.27 100.00
Assays (%)
% Distribution
Cr2O3
TiO2
Cr2O3
TiO2
2.50 0.086 0.36 3.37 0.093 0.44
51.6 59.4 58.5 51.7 60.8 59.8
78.7 21.3 100.0 81.3 18.7 100.0
10.0 90.0 100.0 9.3 90.7 100.0
PRACTICES IN RUTILE FLOTATION
In the past, most of the rutile was produced from heavy mineral sands using physical concentration, involving gravity, magnetic separation and electrostatic concentration. The physical preconcentration method cannot be applied to a fine heavy mineral sand or hard ore. In some cases, heavy mineral sand contains zircon, tantalum, niobium and other heavy minerals, where in most cases a flotation method is used. Over the past 20 years, a new technology was developed that can produce a high-grade rutile concentrate from hard rock ores. In addition, different methods have been developed by which rutile from bulk gravity concentrates containing zircon and other heavy minerals can be successfully separated. This section discusses methods of beneficiation of rutile from hard rock and fine heavy mineral sands. 25.6.1
Development and operation of zircon flotation at sierra rutile limited
Mining and mineral processing operations at the Sierra Leone (Africa) mine are based on a series of relatively large, highly complex ore bodies characterized by a wide variation in mineral composition and mineral size distributions. Over the years, Sierra Leone Limited has produced rutile concentrate and ilmenite concentrate using gravity, magnetic and electrostatic separation from the +250-mesh frac tion. There is a large portion of rutile and zircon contained in the –250-mesh fraction, which cannot be separated by physical concentration and the fine material is stockpiled over the years. In the early 1990s, development testwork was conducted by Hazen Research (USA) to develop a process for treatment of fine rutile/zircon sand using a flotation method [11]. After the development testwork was completed, the separation process was introduced into the Seirra Leone plant. Description of the zircon flotation process The fine –250-mesh product was preconcentrated using gravity (tabling) followed by zircon flotation and magnetic separation to produce rutile and ilmenite concentrate. The process flowsheet with points of reagent additions is presented in Figure 25.14. Using
25.6
Practices in Rutile Flotation
195
–100 mesh table concentrate 0.18 kg/t starch ~0.05 kg/t H2SO4 Conditioning 1 30 sec 0.11 kg/t ARMAC "C" Conditioning 2 30 sec, 3 stages pH 7.5 concentrate (quartz) 1 min/stage 30 min total 35% solids
SiO2 rougher
0.64 kg/t starch ~0.57 kg/t H2SO4 0.91 kg/t NaF
Tailing
Conditioning 1 30 sec 1 min retention
Zircon rougher
0.61 kg/t Armac "C" Tailing
Conditioning 30 sec/stage
1 min/stage 5 stages
0.05 kg/t Armac "C"
1 min/stage
3 min total
Zircon rougher (continued)
Tailing
Concentrate Dry and induced roll
Zircon 1st cleaner
Magnetics
Zircon concentrate
Tailing
Ilmenite
Non- magnetics
Rutile
Figure 25.14 Plant flowsheet with reagent additions for production of zircon, rutile and ilmenite from the Sierra Leone fines.
this flowsheet, the following concentrates were produced: (a) zircon concentrate that assayed 58% ZrO2, 0.8% TiO2 at a recovery of 85%; (b) rutile concentrate that assayed 0.8% ZrO2, 95.2% TiO2 at a recovery of 40% and (c) ilmenite concentrate assaying 0.65% ZrO2, 56% TiO2 at a recovery of 30%. Rutile/ilmenite-zircon bulk flotation and separation Several large deposits of fine mineral sands containing rutile, ilmenite and zircon exist in Australia (Wimmera mine) and in the Soviet Union. The rutile, ilmenite and zircon cannot be preconcentrated. In most cases, flotation was used which involved bulk flotation followed by titanium–zircon separation. Over the years, several effective processes have been developed for bulk flotation followed by titanium–zirconium separation. The type of
196
25.
Flotation of Titanium Minerals
method used is dependent on the type and mineralogy of the fine sand. The following section describes three major methods developed for bulk Ta/Zr flotation and separation. Method 1 – This method has been successfully used in the Soviet Union. The flowsheet with the type and levels of reagent additions is shown in Figure 25.15.
Sand
Scrubbing
Slimes Desliming 100 g/t oleic acid 2000 g/t oxidized fuel oil H2SO4 to pH 6.5 Silica Bulk flotation
Thickening
Effluent
5000 g/t Na2SiO3 Conditioning heating 400 g/t CuSO4 Conditioning heating, 60°C
ZrO2 flotation
Thickening
Tabling
Rutile concentrate
ZrO2 concentrate
Tails
Tailings
Figure 25.15 Flowsheet with reagent additions for beneficiation of fine mineral sands (Kola Peninsula, Soviet Union).
25.6
Practices in Rutile Flotation
197
The bulk flotation can be accomplished with the addition of small doses of oleic acid plus oxidized emulsion of fuel oil. The fuel oil is treated with 10% solution of NaOH at a temperature of 60–80°C for 1 h. The following method was used for rutile–zircon separation; the concentrate was thickened, followed by heat conditioning to 60°C. After the heat treatment, the zircon was floated without the addition of collector. The zirconium tailing is the rutile concentrate. The zircon concentrate was thickened, followed by gravity cleaning. In some cases, the heat-treated pulp is washed before zircon flotation. The following metallurgical results were obtained: Rutile product – 92.5% TiO2 at 90% recovery Zircon concentrate – 0.2% TiO2, 63% ZrO2 at 94% ZrO2 recovery. Method 2 – It involves bulk flotation of rutile, ilmenite and zircon followed by selective flotation of rutile and ilmenite and depression of zircon. Figure 25.16 shows the flowsheet with type of reagent additions used in selective flotation of titanium from zircon. The collector used was a mixture of oleic acid and kerosene in a ratio of 1:1. The mixture was aerated with oxygen during a period of 2 h before using. The advantage of using the oxidized mixture is that it desorbs easily from the mineral surfaces during separation. The metallurgical results obtained using Method 2 are shown in Table 25.10. Method 3 – It involves bulk titanium/zircon flotation using succinamate collector followed by bulk concentrate pretreatment and selective zircon flotation. This method was developed for beneficiation of the Wimmera heavy mineral sand from Australia [12]. The beneficiation flowsheet with type and level of reagents is shown in Figure 25.17. The sand preparation method had a significant impact on both collector consumption, as well as quality of the bulk concentrate. It was found that the mixture of Na2SiO3/tall oil addition to the scrubbing stage before desliming improved the slime decoating from the heavy mineral surface resulting in a significant improvement in concentrate grade. In addition, collector consumption was reduced by 50%. The mixture consisted of 70% Na2SiO3 and 30% tall oil fatty acid. The effect of the levels of silicate tall oil additions and conditioning times are presented in Table 25.11. In these tests, the mixture of Na2SiO3/tall oil was added to the scrubber before desliming. Collector used in the bulk circuit was sulphosuccinate. In the rutile circuit, phosphoric acid ester was used. Silica was rejected in a bulk talking. The overall metallurgical results obtained in the continuous operation are shown in Table 25.12. 25.6.2
Rutile flotation from hard rock ore
Over the past 10 years, new technology has been developed that allows flotation of rutile from complex hard rock ores. This new technology has been confirmed in continuous pilot plant operation. During the development testwork, ores from Mexico, Chile and Australia were studied. Guadalajara (Mexico) rutile ilmenite ore The Guadalajara titanium-bearing ore comes from a hard rock deposit consisting princi pally rutile and ilmenite. Over 85% of the rutile and ilmenite are liberated at relatively
198
25.
Flotation of Titanium Minerals
Sand Scrubbing
Desliming
Slimes
400 g/t Na2SiF6 550 g/t collector H2SO4 to pH 6.2 Conditioning
TiO2/ZrO2 flotation
1500 g/t Na2SiF6 H2SO4 to pH 4.5 Conditioning
TiO2 flotation
500 g/t oxalic acid Conditioning H2SO4 to pH 4.5 Rutile flotation
Rutile concentrate
Figure 25.16 depression.
Ilmenite concentrate
ZrO2 concentrate
combined tailings
Flowsheet and reagent additions used in selective titanium flotation and zircon
coarse grind, while the remaining 15% appears in the form of middlings, as complex intergrowths with non-opaque minerals. The major gangue minerals were plagioclase, feldspar, quartz, calcite and some apatite. The removal of apatite before titanium flotation along with calcite dolomite was required since the apatite tends to float with the titanium. The flowsheet (Figure 25.18) shows the final flowsheet developed for the beneficiation of the Guadalajara ore. This flowsheet consists of two flotation circuits: (a) gangue prefloat circuit, where the apatite and calcite are recovered, and (b) titanium flotation circuit, where
25.6
Practices in Rutile Flotation
199 Table 25.10
Results obtained using sequential rutile, ilmenite, and zircon flotation from bulk concentrate Product
Assays (%)
Rutile concentrate Ilmenite concentrate Zircon concentrate Tailings Head (calc)
% Distribution
TiO2
ZrO2
TiO2
ZrO2
90.1 48.76 0.57 0.48 8.8
0.4 0.2 64.0 0.1 2.8
40.0 55.0 0.5 4.5 100.0
1.0 1.0 95.0 3.0 100.0
a high-grade rutile and ilmenite concentrate were produced. The rutile concentrate produced was free of apatite and silicate. The reagent scheme developed for beneficiation of the Guadalajara ore is shown in Table 25.13. During gangue flotation, caustic corn starch was used to depress the titanium. Gangue flotation was accomplished using emulsified tall oil DO2. Over 87% of the apatite was recovered in a gangue concentrate. The gangue tailings were treated with acid followed by titanium flotation using oxalic acid + H2SiF6 as the gangue depressants. A new titanium collector composed of a mixture of fatty acid ester and sulphosuccinamate modified surfactant was used (PL519). This collector provides a high rate of titanium flotation and is selective towards the gangue minerals. Metallurgical results obtained in a continuous operation are shown in Table 25.14. A high-grade rutile and ilmenite concentrate were produced with respectable recoveries. 25.6.3
White Mountain titanium (Chile)
A large hard rock rutile deposit was discovered in central Chile. This ore is relatively complex with variable head grade of rutile ranging from 2% to 4% TiO2. The liberation of rutile occurs at about 100 mesh nominal size. The major gangue minerals present in this ore include feldspars, calcite and some silicates. Development work conducted over the past 3 years has identified a treatment process that will produce a high-grade rutile concentrate. The initial flowsheet is similar to that used for the Guadalajara ore. However, using this flowsheet, only a portion of the calcite was recovered and an appreciable amount of the rutile was lost in the gangue concentrate. An alternative, effective treatment process has been developed that produces excellent results. The flowsheet developed for beneficiation of the White Mountain titanium ore consist of two distinct circuits: (a) grinding, sizing and gravity preconcentration of the ore, and (b) rutile flotation from the gravity concentrate. This flowsheet includes gravity preconcentrate and flotation as shown in Figure 25.19. It should be noted that gravity preconcentration on the sized ground ore improved gravity performance. The flotation flowsheet included a triple open-circuit flotation and
200
25.
Flotation of Titanium Minerals
Sand 200 g/t Na2SiO3 / D40LR Scrubbing
Slimes
Desliming
100 g/t Na2SiO3 / D40LR Scrubbing Slimes Desliming 50 g/t F2875 H2SO4 to pH 3.5
20 g/t Bulk rougher flotation
Bulk scavenger flotation
H2SO4 to pH 3.5 Bulk cleaner flotation
500 g/t NaOH Conditioning
Effluent
Dewatering 800 g/t H2SiF6
600 g/t starch 300 g/t NaF
Slimes
Desliming
300 g/t oxalic acid Conditioning
Conditioning
30 g/t SM15
400 g/t amine H2SO4 to pH 3.0
20 g/t SM15 TiO2 rougher
ZrO2 rougher
TiO2 scavenger
100 g/t oxalic acid TiO2 cleaner
100 g/t starch H2SO4 to pH 3.0 ZrO2 cleaner
Magnetic separation
ZrO2 concentrate
Figure 25.17
Magnetics Ilmenite
Non-magnetics Rutile
Tailings
Flowsheet and reagent scheme for beneficiation of the Wimmera heavy mineral sand.
25.6
Effect of level of silicate tall oil mixture and conditioning time on bulk Ti/Zr bulk flotation Test no.
Na2SiO3/D40LR (g/t)
Conditioning time (min)
Collector (g/t)
Bulk rougher concentrate Weight (%)
185 186 187 188 190 191 195 193 184 198 196 197
200 200 200 400 400 400 400 400 400 400 400 400
5+5 5+5 5+5 5+5 5+5 5+5 5+5 5+5 5+5 10 + 10 10 + 10 10 + 10
50 70 80 50 70 80 40 50 60 30 40 50
7.7 8.8 9.3 8.5 9.0 10.7 7.9 8.6 8.7 7.6 8.4 8.8
Assays (%)
% Distribution
TiO2
ZrO2
TiO2
ZrO2
42.3 39.7 38.1 40.7 38.7 32.4 41.9 40.6 40.9 41.9 41.5 40.8
12.4 11.4 10.6 11.4 11.2 9.1 12.7 11.5 11.8 12.9 11.9 11.8
90.7 96.5 97.5 96.0 97.2 98.0 90.9 96.0 96.5 88.5 95.5 96.6
98.1 99.1 99.1 99.1 99.1 99.1 99.0 99.1 99.1 99.1 99.1 99.1
Practices in Rutile Flotation
Table 25.11
201
202
25.
Flotation of Titanium Minerals
Table 25.12 Results obtained on the Wimmera fine mineral sand (WIM150 ore) Product
Weight (%)
Zircon concentrate Rutile concentrate Ilmenite concentrate Combined tails Feed (calc)
1.55 2.86 2.57 93.02 100.00
Assays (%)
% Distribution
TiO2
ZrO2
TiO2
ZrO2
0.20 89.6 46.6 0.11 3.87
63.4 0.1 0.5 0.036 1.03
0.1 66.3 31.0 2.6 100.0
95.2 0.3 1.2 3.3 100.0
Ground deslimed ore Conditioning 1
Acid conditioning
Conditioning 2 Desliming Gangue rougher
Gangue scavenger
Gangue cleaner
Slimes
TiO2 rougher
TiO2 1st cleaner
TiO2 1st cleaner scavenger
Gangue 1st cleaner
Gangue 2nd cleaner
Gangue cleaner concentrate
TiO2 2nd cleaner
TiO2 3rd cleaner
TiO2 4th cleaner TiO2 cleaner concentrate High intensity magnetic separation Magnetics Ilmenite
Figure 25.18 ilmenite ore.
Non-magnetics Rutile
TiO2 tailing
Flowsheet developed for beneficiation of the Guadalajara (Mexico) hard rock rutile
25.6
Practices in Rutile Flotation
203 Table 25.13 Reagent scheme Additions (g/t)
Reagent
Gangue prefloat
Acid treatment
Titanium circuit
Depressants and modifiers Caustic corn starch Sulphuric acid (H2SO4) Hydrofluorosilicic acid (H2SiF6) Oxalic acid Sodium silicate (Na2SiO3) acid
900 – – – –
– 2000 – – –
– – 450 400 600
Collectors and frothers Fatty acid DO2 Collector PL519 MIBC
180 – –
– – –
– 120 20
Table 25.14 Overall results obtained in continuous operation Product
Weight (%) Assays (%) TiO2
TiO2 rutile concentrate 7.62 (12 AN M) TiO2 ilmenite concentrate 7.48 (12 AMAG) TiO2 combined concentrate 15.10 Gangue D concentrate 19.90 TiO2 combined tails 56.10 Primary slimes 7.80 Acid slimes 1.10 Feed 100.00
SiO2
% Distribution Fe2O3 P2O5 TiO2
SiO2
Fe2O3 P2O5
96.4
1.14
0.77 0.02
57.0
0.2
0.8
0.2
54.2
2.17 46.65 0.01
31.5
0.3
49.2
0.1
75.5 3.6 0.32 5.5 14.0 12.88
1.69 23.5 54.91 6.10 65.8 2.38 47.0 10.9 42.2 13.8 51.1 7.1
0.01 88.5 0.5 50.0 0.3 2.03 5.5 21.4 17.1 87.9 0.04 1.5 70.0 18.8 4.8 0.37 3.3 7.2 12.0 0.3 0.30 1.2 0.9 2.1 0.7 0.46 100.0 100.0 100.0 100.0
cleaning. This flowsheet was designed to provide a more effective rejection of gangue during rutile cleaning. The reagent scheme developed for the White Mountain titanium ore is shown in Table 25.15. Gangue depressants H2SiF6, oxalic acid and DAX1 were used. Depressant DAX2 is a mixture of low-molecular-weight acrylic acids designed specifically to depress calcite. A highly selective collector, KBX2, is a mixture of succinamate (Cytec’s R845) and phosphoric acid ester (Clariant’s SM15) modified with alkyl sulphate. The metallurgical results from the gravity preconcentration continuous pilot plant are shown in Table 25.16.
204
25.
Flotation of Titanium Minerals
Feed Slimes
RM
O/S −35m
−35m
To flotation
BM
−200m
+65m C
T
M
C
M
−65m
C
C
T
M
M
Final gravity tails −100m
RM
Figure 25.19
White mountain titanium flowsheet.
Over 56% of the feed was rejected in the gravity tailing with about 9% loss of the total titanium in the ore. The overall results, including gravity and flotation, are summarized in Table 25.17. A premium-grade rutile concentrate assaying 97.3% TiO2 was produced at an average recovery of 96% TiO2. This was a premium-grade rutile concentrate.
25.6
Practices in Rutile Flotation
205 Table 25.15
Reagent scheme for the White Mountain titanium rutile ore Additions (g/t)
Reagent
TiO2 rougher
TiO2 cleaner
Depressants and modifiers Hydrofluorosilicic acid (H2SiF6) Oxalic acid DAX1
300 200 –
400 275 250
Collectors KBX1 Fuel oil
600 300–500
50 –
Table 25.16 Results from the gravity preconcentration tests Test Product number
T1
T2
Combined –200 m and +200 m table concentrate +middlings Combined –200 m and +200 m table tails Slime Head (calc) Combined –200 m and +200 m table concentrate +middlings Combined –200 m and +200 m table tails Slime Head (calc)
Weight Assays (%) % Distribution (%) TiO2 SiO2 Fe2O3 CaO TiO2 SiO2
Fe2O3 CaO
42.28
8.05 63.0
0.85
0.28
91.2
41.7
31.5
41.7
56.23
0.53 64.8
1.32
0.29
8.0
57.0
65.4
56.7
1.49 100.0
2.13 56.5 2.38 3.73 63.94 1.14
1.6 1.3 3.1 0.9 0.31 0.29 100.0 100.0 100.0 100.0
41.98
7.92 64.0
0.90
0.31
89.8
42.7
30.2
42.6
56.24
0.59 62.3
1.48
0.30
9.0
55.7
66.6
55.2
1.78 100.00
2.67 57.8 3.70 62.9
2.23 1.25
2.2 0.38 1.3 1.6 3.2 0.31 100.0 100.0 100.0 100.0
206
Table 25.17 Overall results obtained in a continuous pilot plant operation Test number
F-3
3.37 96.63 100.00 3.22 96.78 100.00
Assays (%) TiO2
SiO2
97.2 0.47 3.17 97.4 0.61 3.72
0.74 65.4 64.7 0.79 62.84 60.8
% Distribution Fe2O3
CaO
TiO2
SiO2
Fe2O3
CaO
0.72 1.23 1.17 0.80 1.45 1.43
0.06 0.28 0.29 0.27 0.30 0.30
87.9 12.1 100.0 84.7 15.7 100.0
0.04 99.96 100.0 0.04 99.9 100.0
2.0 98.0 100.0 1.8 98.2 100.0
0.8 99.2 100.0 2.9 97.1 100.0
Flotation of Titanium Minerals
TiO2 concentrate non-magnetic Combined overall tails + slime Head (calc) TiO2 concentrate non-magnetic Combined overall tails + slime Head (calc)
Weight (%)
25.
F-4
Product
References
207
REFERENCES 1. Polkin, S.I., Concentration of Ores from Sand Deposits and Hard Rock, Izdatelstro Nedra 1987, pp. 1180–23. 2. Fan, X., and Rawson, N.A., The Effect of pb(NO3)2 on Ilmenite Flotation, Minerals Engineering, Vol. 13, No. 2, pp. 205–213, 1999. 3. Bulatovic, S., and Wyslouzil, D.M., Process Development for Treatment of Complex Perovskite, Ilmenite and Rutile Ore, Minerals Engineering, Vol. 12, No. 12, pp. 1407–1417, 1999. 4. Bulatovic, S., Process Development for Beneficiation of Apatite, Ilmenite Ore from Quebec, Canada, Report of Investigation, p. 320, July 2001. 5. Liu, Q.I., and Peng, Y., Development of Composite Collector for the Flotation of Rutile, Minerals Engineering, Vol. 12, No. 12, pp. 1419–1430, 1999. 6. Belash, F.N., and Gamilow, M.A., Perovskite Flotation Using Acid Pretreatment, Bulletin CIN Cvetnie Metaly, No. 21, 1959. 7. Bulatovic, S., Pilot Plant test on Perovskite Recovery from Powderhorn USA ore, Report of Investigation, 1987. 8. Runolima, U., How Otammaki Floats Ilmenite from Fnland Titaniferous Magnetite, Mining World San Francisco, pp. 49–55, 1957. 9. Bulatovic, S., Process Development for beneficiation of Complex Apatite–Ilmenite Ore from Quebec, Canada, Laboratory and Pilot Plant Studies, Report of Investigation, 1997. 10. Bulatovic, S., Chromium Removal from the Ilmenite Concentrate by Flotation from RZM Western Australia, Report of Investigation, 1993. 11. Davis, J.P., Wonday, S., and Keilj, A.K., Developoment and Operation fo Zircon Flotation at Sierra Rutile, 10th Industrial Mineral International Congress, San Francisco, pp. 65–71, 1992. 12. Bulatovic, S., Laboratory and pilot plant development testwork on recovery of titanium and zircon from Wimmera heavy mineral sand, Report of Investigation, p. 330, 1992.
Index A
B
AAC10, for tantalum/niobium and zircon separation, 147, 148t Acid pretreatment, for ilmenite flotation, 178, 179f Acidified silica/AQ55D, for apatite-ilmenite ore beneficiation, 189 Acintols, for Indian beach sand flotation, 165, 166t Acrylate, for oxide zinc ore flotation, 82, 82t Alaskan-type deposits, of PGM, 22 Alkyl hydroxamate, for yttrium group of REOE beneficiation, 156–157, 157f Alluvial deposits, of PGM, 22 Aluminum sulphate, as tin ore collector, 102 Amines for fluorite flotation, 163–164 for oxide zinc ore flotation, 72–73, 72t, 81, 81t for pyrochlore flotation, in carbonatite ores, 116, 116t for tantalite-columbite flotation, 130, 131f for tantalum/niobium recovery in Ghurayyah ore, 136, 136t in Malawi, Africa ores, 140, 141f Anglesite, flotation of, 70–72, 71f Anorthositic deposits, of titanium, 177 Antimony ore, flotation of gold-antimony ores, 10–11, 11t gold-containing, 5, 6f, 6t Apatite-ilmenite ores, beneficiation of, 186–190, 187f, 188f, 189f, 190f, 191t collectors for, 187–188, 188f AQ4, in niobium recovery, 121, 121t, 122f Armac C, for oxide zinc ore flotation, 72t, 73 Arsenical gold ores, flotation of, 11–13, 12f Arsenopyrite, pyrite separation from, 12–13, 12f Arsonic acid collector, for tin ores, 93, 93f, 94f Atacamate, flotation of, 51 Azurite, flotation of, 51
BAA. See Benzyl arsonic acid Barite, flotation of, 162 Barite-calcite gangue, for mixed sulphide oxide lead zinc ore beneficiation, 77, 77t, 78t Barite-fluorite ores, bastnaesite flotation in, 154, 161–164, 162t, 163f, 164t Barium chlorite, for barite flotation, 162 Barium sulfide (BaS), for oxide lead ore, 70 BaS. See Barium sulfide Base metal sulphide ores, gold flotation from, 13–15, 14f, 14t Bastnaesite, 151, 152t, 153 flotation of, 153–154, 153f, 154f, 155f, 155t, 159–164 from Dong Pao deposit, 161–164, 162t, 163f, 164t in Mountain Pass operation, 159, 160f, 161t Beneficiation of apatite-ilmenite ores, 186–190, 187f, 188f, 189f, 190f, 191t of cassiterite. See Cassiterite, beneficiation of of ilmenite. See Ilmenite, beneficiation of of oxide lead ores, 78, 80t of oxide zinc ores, 78–83, 79f, 80t, 81t, 82t, 83t of pyrochlore ores. See Pyrochlore ores, beneficiation plant practices for of tantalum/niobium ores. See Tantalum/ niobium ores, beneficiation of of tin ores. See Tin ores, beneficiation of of titanium minerals. See Titanium minerals, beneficiation of Benzyl arsonic acid (BAA), for rutile flotation, 181 Bernic Lake, tantalum/niobium flotation at, 132–133, 134t Bornite, 62 Brazilian monazite ore, flotation of, 167, 168t Busheld Complex, 21
209
210 C Calcite gangue, for mixed sulphide oxide lead zinc ore beneficiation, 75, 75t Carbon, preflotation of, 7 Carbonaceous clay ores, gold-containing, flotation of, 5, 7, 7t Carbonaceous gangue, preflotation of, 7 Carbonatite ores, 111–112, 112t bastnaesite flotation in, 154, 155f pyrochlore flotation from, 112–116, 113f, 113t, 114f, 115t, 116t, 117t Carboxylic collectors, for monazite flotation, 167 Carboxymethyl cellulose (CMC), for PGM recovery, 27, 30, 31f Cassiterite, 87 beneficiation of, 89–97 gravity method, 89–91, 90f gravity-flotation combination, 91, 92f practices in, 98–108 treatment process selection, 98 deposits of, 88–89 flotation of, 91–93 collectors and chemistry of, 93–96, 93f, 94f, 94t, 95f, 96f depressants for, 96–97, 97t floatability of, 98 introduction to, 87, 91–93 plant development and operation for, 98–108 at Renison, 99, 99t, 100f, 101t at Union, 100–101, 102t with gravity concentration, 90–91 CB110, in monazite flotation, 171, 173t Cerium group, of rare earth oxide elements, 151, 152t flotation properties of, 153–158. See also Bastnaesite, flotation of; Monazite, flotation of Cerussite, flotation of, 70–72, 71f Chalcopyrite, 26, 26f Chloritic tourmaline ore types, 89 Chromium, flotation of, 192, 192f, 193f, 193t, 194t Chromium deposits chemical analyses of, 35, 35t with PGM, 24, 25t flowsheet for, 40, 42f reagent practice in flotation of, 33, 35–38, 35t, 36f, 37t, 38t
Index
Chrysocolla, flotation of, 51 Clastic sedimentary deposits, gold recovery from, 2, 2t CMC. See Carboxymethyl cellulose Coarse-grained tin ores, 88 Cobalt ores. See Copper cobalt ores Collector mixture D, for apatite-ilmenite ore beneficiation, 189 Collectors for apatite-ilmenite ore beneficiation, 187–188, 188f for Brazilian monazite ore flotation, 167, 168t for chromium flotation, 192 for fluorite flotation, 163–164 for gold recovery, 15, 16t for ilmenite production from heavy mineral sands, 192 for Indian beach sand flotation, 165, 166t for Malawi, Africa tantalum/niobium ores, 140, 141t for mixed sulphide oxide lead zinc ore beneficiation, 75, 75t for monazite flotation, 153, 153f, 167 for oxide copper ores, 55–58, 57t, 58f, 58t for oxide lead ore flotation, 71–72 for oxide zinc ore flotation, 72–73, 72t, 73f, 74f, 81, 81t for perovskite flotation, 182, 183f PGM recovery and in chromium ores, 35–38, 36f, 37t, 38t in copper-nickel deposits, 32, 33f, 34t for pyrochlore flotation, in carbonatite ores, 116, 116t, 117t for rutile flotation, 181, 181t for tantalite-columbite flotation, 130, 131t for tantalum/niobium and zircon separation, 148, 148t for tantalum/niobium recovery, 136, 137t for tin ore flotation, 93–96, 93f, 94f, 94t, 95f, 96f at Huanuni concentrator, 103–105, 104t, 106t for White Mountain titanium ore, 203, 205f Columbite minerals, 111–112, 112t flotation characteristics of, 129–130, 129f, 130f, 131f, 131t Conditioning power, for Otanmaki ore ilmenite beneficiation, 185–186, 186f Conditioning time, for Otanmaki ore ilmenite beneficiation, 185–186, 186f
Index
Copper Cliff plant, platinum recovery in, 31–32, 33t Copper cobalt ores, 51–52 flotation of industrial practice in, 59–61, 59t, 60f, 61t, 62t introduction to, 47 Copper ores. See also Copper oxide gold ores; Mixed copper sulphide oxide ores; Oxide copper ores gold-containing, flotation of, 8–9, 8f, 9f, 9t Copper oxide gold ores, 48, 48t flotation of, 10, 10t Copper oxide mixed ore – type 1, 48, 48t Copper oxide mixed ore – type 2, 48, 48t Copper sulfate (CuSO4) for oxide zinc mineral activation, 80–81 as PGM activator, 27, 28f, 28t Copper sulphide oxide ores, mixed, 48, 48t Copper-lead-zinc ores, flotation of gold from, 15, 16t Copper-nickel deposits, PGM from, 23–24 flowsheet for, 39–40, 41f reagent practice in flotation of, 31–33, 32f, 33f, 33t Copper-zinc ores, flotation of gold from, 13–14, 14t Covellite, 62 Cuprite, flotation of, 50 Cyanidation, for gold recovery, 1–2 D DAX1, for White Mountain titanium ore, 203, 205f Depressant A4, for apatite-ilmenite ore beneficiation, 188 Depressant SHQ, for apatite-ilmenite ore beneficiation, 189 Depressants for apatite-ilmenite ore beneficiation, 188 for bastnaesite-containing ore flotation, 159 for chromium flotation, 192 for fluorite flotation, 164 gold recovery from copper and, 14, 14t at Huanuni concentrator, 103 for ilmenite production from heavy mineral sands, 192 for Indian beach sand flotation, 165, 165t
211
for Malawi, Africa tantalum/niobium ores, 140–141, 142f for mixed sulphide oxide lead zinc ore beneficiation, 75, 75t for oxide copper ores, 54–55, 55f, 56t for oxide zinc ore flotation, 73, 81–82, 82t PGM recovery and, in chromium ores, 35–38, 36f, 37t, 38t for pyrochlore flotation, in carbonatite ores, 114–116, 115t for Rosetta Nile black sand flotation, 166 for tantalum/niobium and zircon separation, 147, 148t for tantalum/niobium recovery, 136 for tin ore flotation, 96–97, 97t at Huanuni concentrator, 104, 106t for White Mountain titanium ore, 203, 205f Desliming in monazite ore preparation, 169–170, 169f, 170t in niobium recovery, 120–121, 120f, 121t Dicarboxilic acids, for tin ores, 96, 96f Disseminated deposits, of tin, 88 Dithiophosphate collectors for gold recovery, 4–5, 15, 16t for PGM recovery, 31 Dodecylammonium chloride, for ilmenite flotation, 178 Dolomitic gangue, for mixed sulphide oxide lead zinc ore beneficiation, 75, 75t Dolomitic oxide ores, recovery of, 60f, 61, 61t, 62t, 63f Dong Pao deposit, bastnaesite flotation in, 161–164, 162t, 163f, 164t DQ4, for monazite desliming, 170, 170t E EMF2, for pyrochlore flotation, 113, 113t Emulsifiers, for Otanmaki ore ilmenite beneficiation, 186 Eschynite, 152t Euxenite, 151, 152t F Fatty acid flotation method for apatite-ilmenite ore beneficiation, 188 for bastnaesite-containing ores, 159 for carbonatite ore flotation, 112–113, 113f, 113t
212
Fatty acid flotation method (Continued) for monazite ores, 171 for oxide zinc ores, 79–80 for tantalum/niobium recovery, 136 for yttrium group of REOE beneficiation, 157, 157f Fatty acid modification, of xanthate collectors, 56–58, 57t, 58f, 58t Fergusonite, 152t, 155 Flotation of bastnaesite. See also Bastnaesite, flotation of of cerium group of rare earth oxide elements, 153–158 of chromium, 192, 192f, 193f, 193t, 194t of gold ores. See Gold ores, flotation of of monazite. See also Monazite, flotation of of niobium. See niobium, flotation of nitrogen atmosphere method, for gold recovery, 7 of oxide copper ores. See Oxide copper ores, flotation of of oxide lead ores, properties of, 70–72, 71f, 71t of oxide lead silver ores, 83–86, 84f, 85f, 85t, 86t of oxide zinc ores, properties of, 72–74, 72t, 73f, 74f of PGM ores. See Platinum group metals, flotation of of pyrochlore. See Pyrochlore, flotation of of rare earth oxide elements. See Rare earth oxide elements, flotation of of tin. See Tin ores, flotation of two-stage method, for gold recovery, 7, 7t Flowsheet for apatite-ilmenite ore beneficiation, 188, 189f for chromium PGM-containing ores, 40, 42f for chromium removal, 192, 193f for Cu–Ni-containing PGM ores, 39–40, 41f for dolomitic copper oxide ores, 60f, 61, 61t, 62t, 63f for gold-containing copper ore, 9f for gold-containing mercury-antimony ore, 6f for Guadalajara rutile ilmenite ore flotation, 198–199, 202f for ilmenite ore beneficiation, 188, 190f for Inco metal PGM recovery, 32f for iron-hydroxide decoating, 147, 147f for mixed sulphide oxide lead zinc ore beneficiation, 75, 76f
Index
for Mrima case study, 120, 120f for niobium beneficiation, 123, 124f for Otanmaki ore ilmenite beneficiation, 185–186, 187f for oxide siliceous ore, 59, 60f for oxide zinc ore beneficiation practices, 78, 79f for REOE for Dong Pao deposit, 162, 163f at Mount Weld, 171, 171f at Mountain Pass operation, 159, 160f for sulphide-dominated PGM ores, 39, 40f for tantalum/niobium flotation in Ghurayyah ore, 136, 138f from gravity tailings, 134, 135f in Malawi, Africa ore, 143, 143f zircon separation from, 144, 145f for tin ores gravity beneficiation, 90f gravity-flotation beneficiation, 92f at Huanuni concentrator, 104t, 105t at Renison, 100f at San Rafael tin mine, 107t for Titania A/S plant ilmenite beneficiation, 183, 184f for White Mountain titanium ore, 199, 204f for yttrium group of REOE beneficiation, 156–157, 156f for zircon, rutile and ilmenite production, 194–195, 195f Fluorite, flotation of, 163–164 Fossil placer deposits, of PGM, 22–23 G Gadolinite, 152t, 155–156 Gangue constituents of mixed sulphide oxide lead zinc ore, 74–75 of oxide copper ores, 49 Gangue flotation, for Guadalajara rutile ilmenite ore flotation, 199 Geology, of gold ores, 2–3, 2t, 3t Ghurayyah ore, tantalum/niobium recovery from, 134–140, 136t, 137f, 137t, 138f, 139f, 139t, 140t Gold, recovery of, 1–2 Gold ores flotation of arsenical, 11–13, 12f carbonaceous clay-containing, 5, 7, 7t
Index
concluding remarks on, 15 in copper ores, 8–9, 8f, 9f, 9t copper-lead-zinc ores, 15, 16t copper-zinc ores, 13–14, 14t gold-antimony ores, 10–11, 11t introduction to, 1–2 lead-zinc ores, 13, 14f low-sulphide-containing, 4–5 in mercury/antimony ores, 5, 6f, 6t oxide copper-gold ores, 10, 10t properties for, 3–4, 4f geology and mineralogy of, 2–3, 2t, 3t Gold-antimony ores, flotation of, 10–11, 11t Gravity method for tantalum/niobium ore beneficiation, 132, 133f, 133t, 134t for tin ore beneficiation, 89–91, 90f at Huanuni concentrator, 103–105, 104f at San Rafael tin mine, 106 Gravity preconcentration method for gold recovery, 1–2 for PGM recovery, 22–23 for White Mountain titanium ore, 199, 203, 204f Gravity-flotation combination, for tin ore beneficiation, 91, 92f Greenbushes gravity tailing, tantalum/niobium flotation from, 134, 135f, 136t Grind-flotation techniques, metallurgical results of, 6t Guadalajara rutile ilmenite ore, rutile flotation of, 197–199, 202f, 203t H H2SiF6. See Hydrofluorosilicic acid Hard rock lodge deposits, of tin, 91 Hard rock ore, rutile flotation from, 197–199 Heavy mineral sands, ilmenite production from, 191–192, 192f, 193f, 193t, 194t Hemimorphite ore type, 68t, 69 Huanuni concentrator, tin ore flotation at, 103–105, 104f, 104t, 105f, 106t Hydrochloric acid for niobium flotation, 114, 114f for tantalum/niobium separation from zircon, 114, 114f Hydrofluorosilicic acid (H2SiF6) for Malawi, Africa tantalum/niobium ores, 140–141, 142f
213
for tantalum/niobium and zircon separation, 147, 148t for White Mountain titanium ore, 203, 205f Hydrometallurgical method, for gold recovery, 1–2 Hydrophobic gangue depressants, for PGM recovery, 30–31, 31f Hydrothermal deposits, of PGM, 21 Hydroxamic acid for bastnaesite-containing ores, 159 for Brazilian monazite ore flotation, 167, 168t for oxide copper flotation, 49, 49f, 50f I Ilmenite, 175, 176t. See also Apatite-ilmenite ores beneficiation of, 183–186, 184f, 185t, 186f, 187f in Otanmaki ore, 185–186, 185f, 186f Titania A/S plant, 183–185, 184f, 185t flotation of properties of, 177–180, 178f, 179f, 180f, 180t at Sierra Leone mine, 194–197, 195f, 196f, 198f, 199t, 200f, 201t, 202t production from heavy mineral sands, 191–192, 192f, 193f, 193t, 194t Ilmenite-haematite, 177 Ilmenite-magnetite, 177 Ilmenite-rutile, 177 Ilmenorutile, 111–112, 112t, 175, 176t Indian beach sand, monazite flotation of, 165–166, 165t, 166t Iridium. See Platinum group metals Iron-hydroxide decoating, 146–147, 146f, 147f K KBX1, for White Mountain titanium ore, 205t KBX2, for White Mountain titanium ore, 203 KM3 depressant, in PGM recovery, 35–37, 37t Kokoamine KK, for oxide zinc ore flotation, 72t, 73 Kolwezi concentrator, 57–58 dolomitic oxide ore at, 61, 62t oxide siliceous ore at, 59–60, 59t, 60f, 61t Komoto plant, 58 mixed copper sulphide oxide ores at, 62
214 L LAC2, 104, 104t Lead ores. See Copper-lead-zinc ores; Oxide lead ores Lead-zinc ores, flotation of gold from, 13, 14f Leucoxene, 176, 176t Loparite, 152t, 153 Low-sulphide-containing gold ores, flotation of, 4–5
Index
Morensky Reef operation, flotation rates from, 26–27, 26f Morensky-type deposits, of PGM, 21 Mount Weld ore, monazite flotation from, 168–169, 169t, 171, 171f Mountain Pass operation, bastnaesite flotation in, 159, 160f, 161t Mozley drum separators, 89, 106 Mrima Hill deposit, 119–121, 120f, 121t, 122f, 122t Muscovite, with amine collectors, 130, 131f N
M Magnetic ores, of tantalum/niobium, 127 Malachite, floatability of, 49–50, 50f, 51f Malawi, Africa tantalum/niobium ores, beneficiation of, 140–143, 141f, 141t, 142f, 143f, 144f, 144t Medium-coarse-grained tin ores, 88 Mercaptan, for oxide zinc ore flotation, 81, 81t Mercury ore, gold-containing, flotation of, 5, 6f, 6t MESB, for fluorite flotation, 163 Metasomatic deposits, of tantalum/niobium, 129 Mineralogy, of gold ores, 2–3, 2t, 3t Mixed copper sulphide oxide ores, 48, 48t industrial practice in beneficiation of, 62–63, 62t, 63f, 64t Mixed sulphide oxide lead zinc ores geological and mineralogical features of, 67–68 reagent scheme and plant practice for beneficiation of, 74–78, 75t, 76f, 77t, 78t MM4, for fluorite flotation, 164 Modifiers for oxide zinc ore flotation, 81–82 for pyrochlore flotation, in carbonatite ores, 114–116, 115t for tin ore flotation, at Huanuni concentrator, 104, 106t Monazite, 151, 152t, 153 flotation of, 153–154, 153f, 154f, 155f, 155t, 165–173 Brazilian ore, 167, 168t from complex ores, 168–169, 169t of Indian beach sand, 165–166, 165t, 166t ore preparation for, 169–170, 169f, 170t of Rosetta Nile black sand, 166, 166t studies for, 170f, 171, 171f, 172f, 173t
Na2S. See Sodium sulfide NaOH. See Sodium hydroxide Nchanga concentrator dolomitic oxide ore at, 61, 62t mixed copper sulphide oxide ores at, 62 Nickel sulphide deposits, PGM from, 23–24 reagent practice in flotation of, 31–33, 32f, 33f, 33t Niobium. See also Tantalum/niobium ores flotation of, 111–125. See also Pyrochlore ores, flotation of refractory ores of, 119–121, 120f, 121t, 122f, 122t Nitric acid, for niobium flotation, 114, 114f Nitrogen atmosphere flotation method, for gold recovery, 7 Norilsk Talnakh ore, 24 O Oka operating plant, pyrochlore ore beneficiation at, 123 Orthodihydroxybenzene, for pyrochlore flotation, in carbonatite ores, 116 Ortit, 152t Osmium. See Platinum group metals Otanmaki ore, ilmenite beneficiation of, 185–186, 185f, 186f Oxalic acid for apatite-ilmenite ore beneficiation, 189 for Malawi, Africa tantalum/niobium ores, 140–141, 142f for pyrochlore flotation, in carbonatite ores, 114, 115t for tantalum/niobium and zircon separation, 147, 148t for White Mountain titanium ore, 203, 205f
Index
Oxide cobalt ores, 51–52 Oxide copper ores chemical composition and physical structure of, 49 depressants for, 54–55, 55f, 56t flotation of industrial practice in, 59–61, 59t, 60f, 61t, 62t introduction to, 47 practice in beneficiation of, 52–58, 54f, 55f, 56t, 57t, 58f, 58t properties of, 49–51, 49f, 50f, 51f, 52f minerals and, 47–48, 48t sulphidizers for, 53–54, 54f surface layer mechanical strength of, 49 Oxide lead ores beneficiation practices for, 78, 80t of economic value, 69–70, 69t flotation properties of, 70–72, 71f, 71t geological and mineralogical features of, 68 Oxide lead silver ores flotation of, 83–86, 84f, 85f, 85t, 86t plant metallurgical results for, 85–86, 86t plant reagent scheme for, 84–85, 85t processing characteristics of, 83 properties of, 83 research and development on, 84, 84f, 85f Oxide ores, PGM flotation of, 38 Oxide siliceous ore, recovery of, 59–60, 59t, 60f, 61t Oxide zinc ores beneficiation practices for, 78–83, 79f, 80t, 81t, 82t, 83t of economic value, 69–70, 69t flotation of collectors for, 72–73, 72t, 73f, 74f, 81, 81t properties of, 72–74, 72t, 73f, 74f geological and mineralogical features of, 68–69, 68t reagent schemes for, 82–83, 82t, 83t Oxygen, xanthate absorption and, 4, 4f P Palladium. See Platinum group metals Parisite, 152t Pechenga Cala Peninsula, 24 Pegmatite ores, 111–112, 112t pyrochlore flotation from, 116–118, 117f, 118f, 119t of tantalum/niobium, 129
215
Pentlandite, 26, 26f Perovskite, 176, 176t flotation properties of, 182, 183f Petrosol 845, for copper recovery, 59, 60f PGE. See Platinum group elements PGM. See Platinum group metals pH chromium flotation and, 192, 192f ilmenite flotation and, 178, 178f, 179f malachite floatability and, 49, 50f monazite flotation and, 153, 153f Otanmaki ore ilmenite beneficiation and, 186 tantalum/niobium recovery and in Ghurayyah ore, 136, 137f, 137t in Malawi, Africa ore, 141, 142f Phosphonic acid, for tin ores, 93, 93f, 94f Phosphoric acid ester, for perovskite flotation, 182, 183f PL519, for tantalum/niobium recovery, 136, 137t PL520, 103, 104t Placer deposits of PGM, 21–22 of tin, 89–91 Platinum group elements (PGE), 19 deposits of, 23 Platinum group metals (PGM) chromium deposits with, 24, 25t flowsheet for, 40, 42f reagent practice in flotation of, 33, 35–38, 35t, 36f, 37t, 38t classification of, 19–20, 20t copper-nickel and nickel sulphide deposits with, 23–24 flowsheet for, 39–40, 41f reagent practice in flotation of, 31–33, 32f, 33f, 33t description of deposits of, 21–22 flotation of introduction to, 25 ores amenable to, 23 of oxide ores, 38 gold associated with, 3, 3t flotation of, 4 introduction to, 19 mineralogy and recovery of, 22–23 minerals of, 19–20, 20t plant practice in treatment of, 38–40, 39f, 40f, 41f, 42f reagent schemes for, 41, 42–44t sulphide-dominated deposits of, 23
216
Platinum group metals (PGM) (Continued) flotation properties of, 25–27, 26f flowsheet for, 39, 40f reagent practice in flotation of, 27–31, 28f, 28t, 29t, 30t, 31f PlV28, for pyrochlore flotation, in pegmatite ores, 118, 119t PM230, for gold recovery, 10, 10t PM303, for PGM recovery, 37–38, 38t PM306, for PGM recovery, 30, 30t Pneumatalitic-hydrothermal deposits, of tantalum/niobium, 129 Priorit, 152t, 155 Pyrite arsenopyrite separation from, 12–13, 12f gold recovery v., 8, 8f Pyrochlore ores beneficiation plant practices for, 122–125, 124f, 125t at Oka operating plant, 123 at St. Honore Niobec operation, 123, 124f, 125t flotation of, 112–119 from carbonatite ores, 112–116, 113f, 113t, 114f, 115t, 116t, 117t from pegmatite ores, 116–118, 117f, 118f, 119t pH and, 153, 153f general overview of, 111–112, 112t Pyrometallurgical method, for gold recovery, 1–2 Pyrophyllite, in cassiterite flotation, 97 Pyrrhotite, 26, 26f Q Quinolines, for pyrochlore flotation, in carbonatite ores, 116, 117t R R845, 107 Rare earth oxide elements (REOE) cerium group of, 151, 152t flotation properties of, 153–158 flotation of, 158–173 of bastnaesite-containing ores, 159–164, 160f, 161t, 162t, 163f, 164t introduction to, 158 of monazite, 165–173, 165t, 166t, 167t, 168f, 168t, 169f, 169t, 170f, 170t, 171f, 172f, 173t
Index
ore and minerals containing, 151–153, 152t yttrium group of, 151, 152t flotation properties of, 155–158, 156f, 157f, 158f Reagent schemes for apatite-ilmenite ore beneficiation, 188, 191t for cassiterite, 98 for chromium flotation, 192, 193t for Guadalajara rutile ilmenite ore flotation, 199, 203t for mixed sulphide oxide lead zinc ore beneficiation, 74–78, 75t, 76f, 77t, 78t for Mountain Pass operation, 159, 161t for Otanmaki ore ilmenite beneficiation, 186 for oxide copper ores, 52–58, 54f, 55f, 56t, 57t, 58f, 58t dolomitic ores, 61, 61t for oxide lead silver ores, 84–85, 85t for oxide zinc ores, 82–83, 82t, 83t for PGM-containing ores, 41, 42–44t from chromium deposits, 33, 35–38, 35t, 36f, 37t, 38t from Cu-Ni and Ni ores, 31–33, 32f, 33f, 33t in sulphide-dominated ores, 27–30, 28f, 28t, 29t, 30t for REOE for Dong Pao deposit, 162, 164t at Mount Weld, 171, 171f, 172f for siliceous oxide ores, 59, 59t at St. Honore Niobec operation, 123, 125t for tantalum/niobium flotation in Malawi, Africa ore, 143, 144t with zircon, 147–148, 148t for tin ore flotation at Huanuni concentrator, 104, 106t at San Rafael tin mine, 106, 108t for Titania A/S plant ilmenite beneficiation, 183–185, 185t for White Mountain titanium ore, 203, 205t Renison Bell tin mine, tin ore flotation at, 99, 99t, 100f, 101t REOE. See Rare earth oxide elements Reverse flotation method, for oxide zinc ores, 80 Rhodium. See Platinum group metals Rock deposits, of titanium, 177 Rooiberg mill, tin ore flotation at, 100–101, 102t
Index
Rosetta Nile black sand, monazite flotation of, 166, 166t RS702, for tantalum/niobium and zircon separation, 148, 148t Ruthenium. See Platinum group metals Rutile, 176, 176t flotation of Guadalajara rutile ilmenite ore, 197–199, 202f, 203t from hard rock ore, 197–199 practices in, 194–204 properties of, 181, 181t, 182t White Mountain titanium, 199, 203–204, 204f, 205t, 206t with zircon flotation at Sierra Leone mine, 194–197, 195f, 196f, 198f, 199t, 200f, 201t, 202t S Samarskit, 152t San Rafael tin mine, tin ore flotation at, 106–108, 107f, 108t Sand deposits, of titanium, 177 Sea water, as tin ore collector, 102–103, 103t Sedimentary deposits, of tantalum/niobium, 129 Sierra Leone mine, zircon flotation at, 194–197, 195f, 196f, 198f, 199t, 200f, 201t, 202t Silver ores. See Oxide lead silver ores SM500 collectors, for tantalum/niobium flotation, 130, 130t Smithsonite ore type, 68–69, 68t Soda ash, for apatite-ilmenite ore beneficiation, 188 Sodium alkyl sulphate for pyrochlore flotation, in pegmatite ores, 117, 118f for tantalite-columbite flotation, 130, 130f Sodium fluoride, for Titania A/S plant ilmenite beneficiation, 183, 185 Sodium hexametaphosphate, for pyrochlore flotation, in carbonatite ores, 114 Sodium hydroxide (NaOH), for apatite-ilmenite ore beneficiation, 189 Sodium oleate for Indian beach sand flotation, 165, 166t for monazite flotation, 153, 153f, 167, 167t for perovskite flotation, 182, 183f for pyrochlore flotation, in pegmatite ores, 117, 117f
217
for tantalite-columbite flotation, 129–130, 129f for yttrium group of REOE beneficiation, 157, 157f Sodium oxalate for Brazilian monazite ore flotation, 167, 168t for monazite flotation, 167, 167t, 168f Sodium pyrophosphate, for pyrochlore flotation, in carbonatite ores, 114 Sodium silicate (Na2SiO3) for barite flotation, 162 for ilmenite flotation, 178, 180f for Indian beach sand flotation, 165, 165t for oxide copper ore flotation, 54 for oxide zinc ore flotation, 81, 82t for pyrochlore flotation, in carbonatite ores, 116 as tin ore collector, 102 Sodium sulfide (Na2S) for monazite flotation, 154, 154f, 170f, 171 for oxide lead ore, 70, 71f, 71t for oxide zinc ore flotation, 81, 81t SPA. See Steryl phosphonic acid SR82, for barite flotation, 162 St. Honore Niobec operation, pyrochlore ore beneficiation at, 123, 124f, 125t Stanin, 88 Steryl phosphonic acid (SPA), for rutile flotation, 181 Stillwater Complex, 21 Stogargen deposit, 183 Sudbury area, 23 Sulfuric acid for ilmenite flotation, 178, 179f for niobium flotation, 114, 114f for perovskite flotation, 182 Sulphide ores gold recovery from, 2 platinum group metals in, 23 flotation properties of, 25–27, 26f flowsheet for, 39, 40f reagent practice in flotation of, 27–31, 28f, 28t, 29t, 30t, 31f Sulphidization - amine flotation, for oxide zinc ores, 81 Sulphidization process, for oxide copper ores, 49, 53–58, 54f, 55f, 56t, 57t, 58f, 58t Sulphidizers for oxide copper ores, 53–54, 54f for oxide lead ore flotation, 70, 71t
218
Sulphosuccinamate collectors for rutile flotation, 181, 182t for tin ores, 95–96, 96f Surface layer mechanical strength, of oxide copper ores, 49 T Tall oil modifications for apatite-ilmenite ore beneficiation, 188 bastnaesite flotation and, 154, 155t for perovskite flotation, 182, 183f Tall oil modifications (Continued) for Titania A/S plant ilmenite beneficiation, 183, 184f Tantalite minerals, flotation characteristics of, 129–130, 129f, 130f, 131f, 131t Tantalum/niobium ores. See also Columbite minerals; Tantalite minerals beneficiation of from Ghurayyah ore, 134–140, 136t, 137f, 137t, 138f, 139f, 139t, 140t from Malawi, Africa, 140–143, 141f, 141t, 142f, 143f, 144f, 144t practices for, 131–132, 133f, 133t, 134t zircon containing, 134–140, 136t, 137f, 137t, 138f, 139f, 139t, 140t flotation of, 132–134 at Bernic Lake, 132–133, 134t at Greenbushes gravity tailing, 134, 135f, 136t geological and mineralogical features of, 127, 129 introduction to, 127 minerals of economic value, 127, 128t tin gravity intermediate product separation of, 146–148, 146f, 147f, 148t zircon separation from, 137, 139–140, 139f, 139t, 140t from bulk concentrate, 144, 145f, 145t Tapioca starch, caustic, for apatite-ilmenite ore beneficiation, 188 Tellurides, of gold, 3, 3t flotation of, 3 Temperature, for bastnaesite-containing ore flotation, 159 Tenorite, flotation of, 50 3XD, for oxide copper ore flotation, 54 Tin ores beneficiation of, 89–97
Index
gravity method, 89–91, 90f gravity-flotation combination, 91, 92f practices in, 98–108 treatment process selection, 98 deposits of, 88–89 flotation of, 87–108 collectors and chemistry of, 93–96, 93f, 94f, 94t, 95f, 96f depressants for, 96–97, 97t at Huanuni concentrator, 103–105, 104f, 104t, 105f, 106t introduction to, 87, 91–93 plant development and operation for, 98–108 at Renison, 99, 99t, 100f, 101t at San Rafael tin mine, 106–108, 107f, 108t at Union, 100–101, 102t at Valkoomesky plant, 102–103, 103t at Wheal Jane, 92f, 101–102, 102t mineral composition of, 87–88, 88t tantalum/niobium separation from, 146–148, 146f, 147f, 148t Titania A/S plant, ilmenite beneficiation at, 183–185, 184f, 185t Titanium minerals beneficiation of, 182–192 apatite-ilmenite ores, 186–190, 187f, 188f, 189f, 190f, 191t from heavy mineral sands with chromium problems, 191–192, 192f, 193f, 193t, 194t ilmenite ores, 183–186, 184f, 185t, 186f, 187f rutile ores, 194–204 deposit classification of, 176–177 flotation properties of, 177–182 ilmenite, 177–180, 178f, 179f, 180f, 180t perovskite, 182, 183f rutile, 181, 181t, 182t introduction to, 175 ores of, 175–176, 176t Topaz, in cassiterite flotation, 96–97 Tourmaline with amine collectors, 130, 131f in cassiterite flotation, 96–97 2MD, for oxide copper ore flotation, 54 Two-stage flotation method, for gold recovery, 7, 7t TX26, 104, 104t TY3 collector, for oxide copper ore recovery, 57–58, 57t, 58t
Index U
219
for PGM recovery, 27–30, 29t, 30t Xenotime, 152t
Union, tin ore flotation at, 100–101, 102t Y V Valkoomesky plant, tin ore flotation at, 102–103, 103t Violarite, 26, 26f
Yttrium group, of rare earth oxide elements, 151, 152t flotation properties of, 155–158, 156f, 157f, 158f Yttrocerite, 152t, 155–156
W Z Water, cassiterite flotation and, 98 Wheal Jane Concentrator, tin ore flotation at, 91, 92f, 101–102, 102t White Mountain titanium ore, rutile flotation of, 199, 203–204, 204f, 205t, 206t Willemite ore type, 68t, 69 X Xanthate collectors for gold recovery, 3–5, 4f, 15, 16t PM230 v., 10, 10t for oxide copper ore recovery, 55–58, 57t for oxide lead ore flotation, 71–72 for oxide zinc ore flotation, 81, 81t
Zellness deposit, 183 Zinc ores. See Copper-lead-zinc ores; Copperzinc ores; Lead-zinc ores; Oxide zinc ores Zircon flotation of, at Sierra Leone mine, 194–197, 195f, 196f, 198f, 199t, 200f, 201t, 202t REOE-containing, recovery of, 157–158, 158f in tantalum/niobium ores beneficiation of, 134–140, 136t, 137f, 137t, 138f, 139f, 139t, 140t separation from bulk concentrate, 144, 145f, 145t separation of, 137, 139–140, 139f, 139t, 140t, 147–148, 148t
