Cytec Solutions 2013

DELIVERING TECHNOLOGY BEYOND OUR CUSTOMERS’ IMAGINATION ™

IN PROCESS SEPARATION

VOLUME 17

CYTEC CY TEC SOLUTIONS SOLUTIONS for Solvent Extraction, Extrac tion, Mineral Processing Processing and Alumina Processing

Letter from the Vice President To our Valued Customers, Advancements in our new product development effort as well as mergers and acquisitions over the past several years has transformed Cytec’s Cytec’s business portfolio. These changes have created a leading high growth specialty chemical business. What does this mean for our mining customers? We continue to collaborate with you to address address challenges and meet them with our technology and products. With economic challenges, changes in ore grades, and the demand for natural resources, Cytec is committed to finding sustainable, quality solutions to help you keep up with these challenges. To help meet the growing demand from our customers, Cytec has made significant investments in our manufacturing assets to improve reliability reliability and increase capacity. One of the most significant investments includes a several hundred million dollar investment to expand our site in Niagara Falls, Canada. This site produces both mining products and phosphine derivates. We are also investing in assets to increase R&D capabilities at other global sites. We at Cytec Cytec want to help bring about solutions to your current and future operation’s success and we have a well balanced portfolio of products and expertise that that are unmatched. The major benefits of our products include increasing revenue revenue through improved production, reducing operating costs, and reducing capital expenditure to build new plants. We have a significant focus on technology development and you can rely on Cytec to bring you the latest technology with our steady stream of new products that ensures you always have the best option by partnering with Cytec. In this edition, we highlight highlight some of these new product advances. These include a scale controlling solution for phosphoric acid product plants with our innovative PHOSFLOW® technology, an alternative to traditional hazardous modifiers with our AERO® 7260 HFP, and nitration residence with our ACORGA ® NR series reagents. We are also pleased to share that our MAX HT ® Bayer sodalite scale inhibitor was awarded the 2012 EPA Presidential Green Chemistry Challenge Award. Award. I have been with Cytec for over 18 years and in many roles that have helped prepare me for my new role to lead the In Process Separation business. Now, I am excited to lead a business that is focused on our valued customers and partners in the mining industry. I am dedicated along with my team to provide you with the service and solutions you are looking for now and in the future. Thank you for your interest and business,

Michael Radossich Vice President, In Process Separation

Letter from the Vice President To our Valued Customers, Advancements in our new product development effort as well as mergers and acquisitions over the past several years has transformed Cytec’s Cytec’s business portfolio. These changes have created a leading high growth specialty chemical business. What does this mean for our mining customers? We continue to collaborate with you to address address challenges and meet them with our technology and products. With economic challenges, changes in ore grades, and the demand for natural resources, Cytec is committed to finding sustainable, quality solutions to help you keep up with these challenges. To help meet the growing demand from our customers, Cytec has made significant investments in our manufacturing assets to improve reliability reliability and increase capacity. One of the most significant investments includes a several hundred million dollar investment to expand our site in Niagara Falls, Canada. This site produces both mining products and phosphine derivates. We are also investing in assets to increase R&D capabilities at other global sites. We at Cytec Cytec want to help bring about solutions to your current and future operation’s success and we have a well balanced portfolio of products and expertise that that are unmatched. The major benefits of our products include increasing revenue revenue through improved production, reducing operating costs, and reducing capital expenditure to build new plants. We have a significant focus on technology development and you can rely on Cytec to bring you the latest technology with our steady stream of new products that ensures you always have the best option by partnering with Cytec. In this edition, we highlight highlight some of these new product advances. These include a scale controlling solution for phosphoric acid product plants with our innovative PHOSFLOW® technology, an alternative to traditional hazardous modifiers with our AERO® 7260 HFP, and nitration residence with our ACORGA ® NR series reagents. We are also pleased to share that our MAX HT ® Bayer sodalite scale inhibitor was awarded the 2012 EPA Presidential Green Chemistry Challenge Award. Award. I have been with Cytec for over 18 years and in many roles that have helped prepare me for my new role to lead the In Process Separation business. Now, I am excited to lead a business that is focused on our valued customers and partners in the mining industry. I am dedicated along with my team to provide you with the service and solutions you are looking for now and in the future. Thank you for your interest and business,

Michael Radossich Vice President, In Process Separation

Solvent Extraction, Mineral Processing and Alumina Processing

Table of Contents: Solvent Extractions Crud Processing Improvements Using ACORGA ® CB 1000 Crud Busting Reagent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Use of ACORGA ® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Mineral Processing AERO®7260 HFP Depressant: Novel, Safe and Sustainable Alternative to Traditional Traditional Hazardous Modifiers – NaSH, Nokes, Na 2S, an and Cy Cyanide . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Rej ejec ecti tion on of Pyr Pyrit ite: e: Ch Chal alle leng nges es and and Su Sust stai aina nabl blee Che Chemi mica call Sol Solut utio ions ns . . . . . . . . . . . . . . . 23

Alumina Processing MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Performance Evaluation of CYFLOC ® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 35

Industrial Minerals Scale Sca le Contr Controll olling ing Chemi Chemical cal Addi Additiv tives es for Phosph Phosphori oricc Acid Acid Product Production ion Plants Plants. . . . . . . . 42

IN PROCESS SEPARATION

03

Crud Processing Improvements Using ACORGA® CB 1000 Crud Busting Reagent Tyler McCallum, Troy Troy Bednarski, and Matthew Soderstrom

Cytec has developed a unique crud treatment process utilizing both chemical and mechanical means to enhance the solid/liquid separation, improve recovered organic quality, and reduce operational costs. Crud (a complex solid stabilized emulsion of aqueous and organic) is a common concern in most solvent extraction processes[1,3]. If crud is allowed to to build up in the the circuit, aqueous and organic velocities within the settlers will increase, resulting in higher entrainments and operational costs. Crud movement between stages can cause continuities to flip and may require a significant reduction in plant flows or a plant shutdown to stabilize the operation[2]. To prevent these events, interfacial pumping is typically carried out to remove crud from the circuit and process it for organic recovery[4]. Crud processing can be very time consuming, and the recovered organic quality is often lower than desired with current processing methods. Cytec has developed a unique crud treatment process utilizing both chemical and mechanical means to enhance the solid/liquid separation, improve recovered organic quality, and reduce operational operational costs. The use of ACORGA ACORGA® CB 1000 crud busting reagent allows a rapid separation of solids from the organic phase. ACORGA CB 1000 is an SX qualified chemical additive, which aids in the recovery of organic from crud. The process involves breaking breaking the stabilized crud emulsion, freeing the associated organic, and settling the solids very very rapidly. This process allows operations to return clean organic back to the plant more efficiently and may enable operations to process more crud. In addition to the improvement in processing time, the crud buster process enables more efficient clay treatment and therefore can improve the quality of organic that is returned to the SX circuit.

04

The crud buster process involves mixing the crud with organic (under organic continuity) then breaking the c rud emulsion through the addition of hydrophilic solids (clay). Once the emulsion has been temporarily broken, the addition of ACORGA ACORGA CB 1000 will bind to the solids causing them to settle and preventing the emulsion from re-forming. Following the clay and ACORGA CB 1000 addition, the agitator may be stopped, allowing the phases to separate and quickly recover the majority maj ority of the organic freed from the crud emulsion. This organic can then be more efficiently clay treated and returned to the process quickly without the typical issues associated with filtration of an emulsion. The solids remaining after the primary separation (containing some residual organic and aqueous, which was freed from the crud emulsion) can then be processed using typical methods for a secondary solid/liquid separation and further organic recovery. The volume of the the secondary separation is substantially less; therefore limited time is required for processing. Any organic recovered from the secondary separation should also be subjected to clay treatment. The laboratory test shown in Figure 2 illustrates the effect of ACORGA CB 1000 in breaking the crud emulsion and freeing the associated organic. For this test, crud was dispersed in an organic continuous mix of diluent. The picture on the left is the organic continuous mix before clay addition; the middle picture is after addition of clay and ACORGA CB 1000; and the picture on the right is the immediate result after agitation was ended. As shown, a very clear organic phase is evident using the process and recovery of this organic can be quickly achieved.

Solvent Extraction, Mineral Processing and Alumina Processing

Crud Processing Improvements Using ACORGA ACORGA® CB 1000 Crud Busting Reagent

FIGURE 2: ACORGA CB 1000 MIXING AND SETTLING

Crud Buster™ Benefits Crud processing using ACORGA® CB 1000 can offer significant time savings due to the rapid chemical separation of organic from crud without requiring the initial step of using a press or centrifuge to break the crud emulsion. The organic that is quickly recovered is a very clean stream

largely free of suspended solids. This clean organic stream can then be clay treated more efficiently producing a high quality recovered organic. The small amount of ACORGA ACORGA CB 1000 remaining in the organic after the solid/liquid separation is removed by the clay during clay treatment.

Time Savings Eliminating the need of a press or centrifuge for the initial rupturing of the crud emulsion to free organic allows significant time time savings. The crud emulsion can blind filter cloths when using a plate and frame filter press (requiring

additional time to drop and recharge the press). Centrifuges are limited by the flow rate and crud volume to be processed. The crud buster process allows a rapid solid/liquid separation without the additional steps/equipment.

Total Suspended Solids (TSS) Current crud treatment methods (regardless if using a centrifuge or filter press) are often inefficient and frequently allow suspended solids to be left in the recovered organic. The return of organic with these now finely dispersed solids can be the cause of additional operational difficulties. The amount of solids remaining in the organic following mechanical processing can vary greatly and is dependent on

IN PROCESS SEPARATION

the equipment being utilized. High TSS in recovered organic is common. Figure 3 shows solids removed from the organic during each step of processing. The top row of pictures gives an indication of the TSS present in the organic phase after each step using a traditional filter press process without ACORGA CB 1000. The bottom row of pictures represents represents the crud buster process after each step.

05

Crud Processing Improvements Using ACORGA® CB 1000 Crud Busting Reagent

FIGURE 3: TSS OF STANDARD FILTER PRESS AND CRUD BUSTER PROCESS

Visually it is easy to see that the final organic product returned to the circuit post clay treatment was m uch cleaner

using the crud buster process than the process using only mechanical separation.

Interfacial Tension (IFT) Mechanical rupturing of crud often results in surface active species associated with crud being transferred to the organic, lowering the IFT and organic quality. This is in addition to the problem of solids often being redistributed.

the circuit organic. This reduction in organic IFT is true for operations using plate and frame filter presses or centrifuges. The figure also shows that both the plant organic and crud processed organic have the potential to be of higher quality with efficient clay treatment. Without clay treatment of the recovered organic, the associated surface active species from the crud are often returned to the circuit.

Figure 4 shows the interfacial tension of various organic samples from operating plants. Traditional mechanical rupturing processes return organic with a lower IFT than

40

    )    m    c 35     /    s    e    n    y     d     (    n    o     i    s 30    n    e     T     l    a     i    c    a     f    r    e    t 25    n     I

Plant Organic Recovered Organic from Crud

29.1

29.3 27

35.3

34.6

34.4

26.8

22.3

FIGURE 4: IFT OF CIRCUIT ORGANIC, RECOVERED ORGANIC, AND CLAY TREATED ORGANIC

20 Sample 1

06

Clay Treated Organic

30

29.2 27.6

34.9

Sample 2

Sample 3

Sample 4

Solvent Extraction, Mineral Processing and Alumina Processing

Crud Processing Improvements Using ACORGA® CB 1000 Crud Busting Reagent Operations that practice clay treatment of recovered organic typically only utilize 0.1 – 0.3 wt% clay due to plugging concerns. This is rarely sufficient to remove all surfactants from the organic, and the clay is often deactivated by

aqueous remaining with the organic. As shown in Figure 5, an excess of 2% clay is required to restore the organic IFT (of this specific plant organic) to its maximum value.

39 Recovery Organic Clay Treatment Curve     )    m    c     /    s    e    n    y     d     (    e    m     i     T    n    o     i    s    n    e     T     l    a     i    c    a     f    r    e    t    n     I

37 35 33 31 29

FIGURE 5: 27 0

1

2 3 4 Clay Concentration (wt%)

The use of ACORGA CB 1000 efficiently separates the organic from the solids/aqueous emulsion, enabling the

5

6

CLAY TREATMENT VS. INTERFACIAL TENSION

organic to be treated with the appropriate clay dosage without deactivation of the clay.

Benefits of Higher Organic Quality Pilot plant testing was completed to compare organic recovered by crud buster to organic recovered by

typical mechanical crud processing means. This work was completed using a 2E + 1S configuration at 6 lpm feed flow and results are shown in Table 1.

TABLE 1. Pilot Plant Comparison CB™ Processed Organic

Typical Processed Organic

33.5.9

29.2

Extract PDT – Org Cont. (seconds)

51

229

Extract PDT – Aq Cont. (seconds)

63

66

Strip PDT – Org Cont. (seconds)

50

191

Dispersion Band – Org Cont. (% of org depth)

0%

61.2%

Organic Entertainment

34% decrease

–

Aqueous Entertainment

18% decrease

–

1032

645

IFT (dynes/cm)

Cu:Fe Transfer Ration

IN PROCESS SEPARATION

07

Crud Processing Improvements Using ACORGA® CB 1000 Crud Busting Reagent The crud buster process (enabling efficient clay treatment) produced an organic with a higher IFT and better overall organic quality. This resulted in a significant improvement in phase disengagement times, dispersion band depth, organic and aqueous entrainments, and Cu:Fe selectivity.

Note: Lower Fe transfer (along with reduced aqueous entrainment) would be expected to result in a significant reduction in operating costs through electrolyte bleed reduction.

Conclusion Current crud treatment and organic recovery practices are often not efficient in producing a high quality organic product. Use of mechanical equipment to break the crud emulsion is effective, but often leaves suspended solids and surfactants in the organic. It is critical to clay treat recovered organic (although not always practiced). When clay treatment is performed, the clay concentration used is often lower than optimal because of concerns related to plugging of the filtration equipment. The resulting organic returned to the circuit leads to redistribution of solids, poor phase disengagement, and higher entrainments. Metallurgical performance can also be negatively impacted.

The crud buster process enables efficient clay treatment and results in a high quality recovered organic in a timely manner. Crud buster is expected to produce an organic with a lower TSS and a higher IFT than current processes. These improvements in organic quality have been shown to result in improved SX performance (break times, entrainments, kinetics, stage efficiency, Cu/Fe selectivity) and are expected to bring operational cost savings.

References 1. R.F. Dalton, C.J. Maes, and K.J. Severs, “Aspects of Crud Formation in Solvent Extraction Systems”, Arizona Conference of the AIME, Tucson, AZ., 1983. 2. Cytec Industries Inc., “Crud: How It Forms and Techniques for Controlling It”, Marketing Publication, 2006.

3. T. Burniston, J.N. Greenshields, and P.E. Tetlow, “Crud control in Copper SX Plants”, E&MJ, 1992, (Jan) pp. 32-35. 4. M. Cox, “Liquid-Liquid Extraction and Liquid Membranes in the Perspective of the Twenty-First Century”, Solvent Extraction and Liquid Membranes, 2008, pp. 1-19.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries. 08

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art Rodrigo Zambra*, Alejandro Quilodran, Gonzalo Rivera, and Osvaldo Castro

Given the relevance of the nitration threat in Chile due to high nitrate containing ores in some plants and the lack of an available practical solution for the industry, Cytec developed a superior line of modified aldoxime extractants. This work presents the results of studies of different solvent extraction operations in the north of Chile where nitration concerns are the greatest. While all copper solvent extraction operations have some nitrates present, this paper is focused on the four copper SX plants that have the potential for appreciable levels of nitrate ions in their leach solutions. Nitration is a phenomenon that initially attracted the interest of the copper mining industry in the late 90’s due to the experience at Lomas Bayas where they experienced significant nitration of the organic inventory. Since then the industry developed the position that ketoxime-based extractants were the best solution for operations with nitration risk. Nitrated oximes (ketoximes and aldoximes) form stable Cu complexes that prevent the stripping of copper. Once the oxime is nitrated, the oxime no longer works as an extractant because that portion of the organic no longer transfers copper.

Nitration is certainly a function of the nitrate concentration in aqueous solutions, but it is also a function of the acidity, temperature, redox potential, interfacial tension and the reactivity of the aqueous and organic phases. Nitration of oxime compounds leads not only to reduced copper transference capacity, but also increased phase disengagement times, reduced interfacial tension, increased entrainment and hydrolytic degradation. Given the relevance of the nitration threat in Chile due to high nitrate containing ores in some plants and the lack of an available practical solution for the industry, Cytec developed a superior line of modified aldoxime extractants. These products, known commercially as the ACORGA® NR series, provide nitration protection without reducing copper production capacity. Examples of the relative performance of ACORGA NR series extractants and ketoxime-based extractants are discussed next.

The nitration mechanism is shown below: NO3- + H2SO4 ⇔ HNO3 + HSO4-

(1)

HNO3 + H 2SO4 ⇔ NO2+ + H 2O + HSO4-

(2)

OH

N

OH R1 + NO2+

R2

OH O2N H

+

N

OH R1

R2

OH O2N

N

OH R1

R2

R1 H OR CH3 �ALDOXIME OR KETOXIME� R2 C9H19 OR C12H25 �NONYLALDOXIME OR DODECYLALDOXIME�

09

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art Case 1, Plant A The conditions at Plant A prior to substitution of the ketoxime extractant LIX ® 84I with the modified aldoxime ACORGA® NR10 are listed below: TABLE 1: Characterization of the Solutions at Plant A

Element

Units

PLS

g/L

4.60

36

pH/H2SO4

-/g/L

1.60

180

NO3-

ppm

1,890

63

ORP

mV

470

500

Cu

Spent

Simulations In order to compare the extraction efficiency of the reagents LIX 84I and ACORGA NR10, the extraction and stripping isotherms were created in the laboratory using real plant

solutions. McCabe Thiele analysis was then used to calculate the expected recovery for the configuration. The results are presented in Table 2.

TABLE 2: Results of Simulations with Plant Solutions (23% extractant).

Extractant

Efficiency [%] Train A

Efficiency [%] Train B

Lix 84I

89.23

77.51

NR10

95.21

87.33

The better extraction kinetics under high copper tenor and low pH conditions of ACORGA NR10 results in a 6% higher

copper recovery than LIX 84I extractant, which was used in the plant.

Accelerated Nitration Tests Several tests were then carried out in order to evaluate the behavior of the extractant in a possible nitration scenario. The properties of the evaluated PLS feed (which had adjusted values of pH and nitrate to make the solution more aggressive) are shown in Table 3. This PLS was mixed continuously in a 1:1 ratio at 40°C with three separate reagents prepared at 25 vol %: LIX 84I (ketoxime), LIX 860 (pure aldoxime), ACORGA NR10 (modified aldoxime) and Plant Organic (a blend of the regents appointed previously).

10

TABLE 3: PLS Conditions for the Accelerated Nitration Tests. Value

Units

Cu

2.50

g/L

NO3-

62.0

g/L

FeT

4.70

g/L

Cl-

10.30

g/L

P. Redox

752

mV

T°

40

C

pH

1.10

–

Solvent Extraction, Mineral Processing and Alumina Processing

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art The results presented in Figure 1 show that there was a strongest resistance to nitration when using the ACORGA

NR10 reagent (approx. 50%) compared to LIX® 84I, LIX® 860 and Plant Organic.

2.0

Ketoxime

1.8

    )     %     (    n    o     i    t    a    r    t     i     N

1.6

Unmodified Aldoxime Plant Organic

1.4

NR10

1.2 1.0 0.8 0.6 0.4 0.2 0

    2     1     0     2       9     0       2     1

    2     1     0     2       9     0       9     1

    2     1     0     2       9     0       6     2

    2     1     0     2       0     1       3     0

    2     1     0     2       0     1       0     1

    2     1     0     2       0     1       7     1

    2     1     0     2       0     1       4     2

    2     1     0     2       0     1       1     3

    2     1     0     2       1     1       7     0

    2     1     0     2       1     1       4     1

    2     1     0     2       1     1       1     2

    2     1     0     2       1     1       8     2

0.50 0.45    u     C     l    p    g  ,    r    e    p    p    o     C     l    a    u     d     i    s    e     R

0.40 0.35 0.30 0.25 0.20 0.15

FIGURE 1:

0.10

RESULTS OF THE ACCELERATED NITRATION TESTS BASED ON RESIDUAL COPPER AND NITROXIME

0.05 0

    2     1     0     2       9     0       2     1

    2     1     0     2       9     0       9     1

    2     1     0     2       9     0       6     2

IN PROCESS SEPARATION

    2     1     0     2       0     1       3     0

    2     1     0     2       0     1       0     1

    2     1     0     2       0     1       7     1

    2     1     0     2       0     1       4     2

    2     1     0     2       0     1       1     3

    2     1     0     2       1     1       7     0

    2     1     0     2       1     1       4     1

    2     1     0     2       1     1       1     2

    2     1     0     2       1     1       8     2

11

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art Case 2, Plant B The second case shows the laboratory and piloting test to compare the behavior of the ACORGA® NR 20 extractant and the reagent currently in use at the plant LIX 84I. This plant has a complex SX configuration, with two different PLS feeds: the heap leaching solution at 1.8 gpl

Cu and pH 2.0 and the ROM leaching solution at 1.6 gpl Cu and pH 1.6. The stage efficiency was measured to compare the performance of LIX 84I and ACORGA NR 20, with both feeds.

100 90

90 80

85.9 80.1

79.5

70 60 50

FIGURE 2:

40 Acorga NR 20 – HEAP

LIX 84 IC – HEAP

Acorga NR 20 – ROM

LIX 84 IC – ROM

STAGE EFFICIENCIES FOR PLS HEAP AND ROM

As shown in the graph above higher stage efficiencies were achieved with the ACORGA NR reagent.

Accelerated Nitration Tests The following products were tested, LIX ® 84I, ACORGA NR20 and a traditional aldoxime that is not nitration resistant, “unprotected reagent” under aggressive nitrating conditions. The evaluation took place over a period of 150 days. The PLS used in this study was modified to be highly nitrating. Impurities were added to a real PLS (chloride, iron, and nitrate) with a pH of 1.0, as shown in Table 5. The extractants were mixed in a 1:1 ratio, and the solution was submerged in a thermostatic bath at a temperature of 40°C with constant agitation.

12

TABLE 4: Characterization of the PLS COMPOSITION

MODIFIED PLS

Acidity

g/L

5.7

NO3-

g/L

58.8

FeT

g/L

3.08

Cl-

g/L

10.38

It can be clearly seen in Figure 2 that both the ACORGA NR20 and LIX 84I extractants had an appropriate resistance to nitration but the unprotected extractant had significant nitration before 80 days of mixing.

Solvent Extraction, Mineral Processing and Alumina Processing

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art

60

Acorga NR 20 Unprotected Reagent LIX 84IC

50     ) 40     %     (    n    o 30     i    t    a    r    t     i     N

20

FIGURE 3: RESULTS OF ACCELERATED NITRATION TESTS BASED ON RESIDUAL COPPER AND NITROXIME

10 0 0

1

2

3

4 Sample

5

6

7

8

Pilot Plant Evaluation The ACORGA NR20 extractant was then evaluated in a 100 cm3/min pilot plant utilizing two PLS solutions (Heap and ROM). The initial conditions for the pilot study are presented in Table 5. The configuration of the pilot plant corresponded to that of an industrial plant, and the extractant was added at 26.30 % for LIX ® 84I and 24.94% for ACORGA NR 20. The results of the tests are shown in Table 6.

that has favourable kinetics for mass transfer as compared to those for extractants based on ketoxime chemistry. As a result, a better mixing efficiency near the equilibrium point is achieved. In addition, the ACORGA® extractant tolerates a wider pH range, maintaining good chemical and metallurgical performance from pH 1.0 to 2.5. Table 6: Extraction Efficiency, Pilot Plant Results

Table 5: Pilot Plant Test Initial Conditions

HEAP

ROM

Spent

g/L

1.97

1.63

42

- / g/L

2.08

1.81

175

O/A E

-

0.95

0.95

O/A S

-

1.24

1.24

Cu PLS pH / H2SO4

The extraction efficiency results clearly show a better metallurgical performance for the ACORGA NR20 extractant, resulting in a 5.8% increase in copper recovery for the Heap and 8.8% increase for the ROM. Both extraction efficiencies are enhanced using the ACORGA NR20 extractant, which is based on a modified aldoxime

IN PROCESS SEPARATION

Extractant

HEAP Extraction Efficiency (%)

ROM Extraction Efficiency (%)

Ketoxime

89.44

67.24

ACORGA NR 20

95.28

76.05

In addition, the results for the selectivity of the ACORGA NR20 extractant conclusively confirm that the new reagent improves the plant selectivity by approximately 50%. The organic Fe loading for both the Heap and the ROM PLS streams are shown in Figure 4.

13

Use of ACORGA® NR Reagents in the Presence of Nitrate Ions in SX: The State of the Art 30

Heap-Ketoxime

Heap-NR20

ROM-Ketoxime

ROM-NR20

25 20    m    p    p  ,     3 15    +    e     F

10 5 0 0

10

20

30

40 50 60 Loaded Organic, %

70

80

90

100

30 25 20    m    p    p  ,     3 15    +    e     F

10

FIGURE 4: ORGANIC FE CO�EXTRACTION AS A FUNCTION OF COPPER LOADING FOR HEAP AND ROM SOLUTIONS

5 0 0

10

20

30

40 50 60 Loaded Organic, %

70

80

90

100

CONCLUSION Based on the results of the studies in the laboratory, and in the pilot plant, the following conclusions can be made: • There is a great increase in the extraction efficiency and transfer of copper when using the ACORGA NR extractant, mainly because it provides better performance at low pH and enhanced extraction kinetics, which help improve the stage efficiency.

• In all of the cases studied, the ACORGA NR reagent performed better in terms of copper recovery by at least two percentage points with a maximum difference of 8 percentage points. • Cu/Fe selectivity is also increased significantly (50%) by use of ACORGA NR extractants rather than ketoxime. • The ACORGA NR extractant offers protection for the plant organic inventory under nitration conditions, ensuring a similar or better response than the LIX 84I® extractant.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries. 14

AERO®7260 HFP Depressant : Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide Mukund Vasudevan and D.R. Nagaraj

Cytec has developed AERO®7260 HFP Depressant, a highly efficient and versatile sulfide mineral depressant with wide applicability.

Introduction NaSH/Nokes are commonly used modifiers in Cu-Mo separation systems. However, these materials present a significant safety and health hazard to humans and a potential environment risk. After listening to the industry’s need for safer alternatives, Cytec’s innovation laboratory in Stamford, CT USA, focused its resources on finding a solution which is described in this article.

Cu-Mo ore Cu ~ 0.5% Mo ~ 0.05%

Cu-Mo operations typically process ores rich in Cu sulfides (head grade 0.1-2%) and molybdenite (MoS 2, head grade 0.01- 0.05%) via an operation consisting of a) the bulk flotation circuit, followed by b) Mo circuit as seen in Figure 1.

Bulk Circuit Flotation Cu-Mo Bulk Concentrate (28% Cu, 1% Mo) Mo Circuit

Roasting Steam Cl2, O3, H2O2,etc

Pre-Treatment (Optional)

Cu Depressants

Conditioning with Cu Depressant Mo Rougher

Tails

Cu Conc

FIGURE 1: Conc

Mo Ro Conc

Mo Cleaner Circuit

The bulk flotation circuit is intended to produce a high grade Cu concentrate containing molybdenite values along with minor amounts of pyrite and some non-sulfide gangue. This concentrate is then processed in the Mo circuit to selectively

A GENERIC FLOW SHEET FOR A CU�MO CIRCUIT

float MoS2 while depressing Cu sulfides and pyrite. This selective Cu-Mo separation is accomplished with the use of depressants such NaSH, Nokes, and Na2S (and cyanide, in some instances) with NaSH as the most widely used. 15

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide NaSH, Nokes and Na2S depressants generate significant amounts of a toxic, lethal, and flammable gas, H2S. Cyanide, which is also used as a depressant is both poisonous and has the potential to generate HCN, a toxic and flammable gas. In order to insure the safety of workers, the surrounding communities and the environment, Cu-Mo plants require several safety measures including H2S alarms and exhaust hoods over flotation cells and other exposed areas. In addition, H2S monitors are required on all personnel entering these plants and workers must adhere to strict safety protocols which involve rigorous training and evacuation procedures. In spite of these measures, hazards still persist

and the industry is waiting for a safer, economically viable depressant which will provide the same metallurgical benefits. In response, Cytec has developed AERO®7260 HFP Depressant, a highly efficient and versatile sulfide mineral depressant with wide applicability as a selective depressant for Cu sulfides and pyrite and a safer alternative to NaSH, Na2S, and Nokes reagent. The following sections discuss in greater detail the issues with conventional depressants and benefits and application guidelines for AERO 7260 HFP in Cu-Mo separation.

Problems with Current/Conventional Depressant Technology NaSH has been the main Cu sulfide depressant used in Cu-Mo separations for many decades. However, due to the danger of generation of high concentrations of toxic, flammable, hazardous, and lethal H2S gas, NaSH poses significant issues in plant operations and poses a threat to the local environment. Transportation of 20 to 40 tons per day of 40% solution of NaSH present shipping and logistics issues both in urban and remote areas. Metallurgical performance with NaSH is also not robust, for instance, plants can observe large performance swings with changes

in ore mineralogy, and often pyrite depression with NaSH is inadequate even at very large dosages, creating a significant challenge in flotation operations. In the absence of a robust and economically viable alternative, NaSH (Na2S and Nokes in some plants) continues to be used extensively in Cu-Mo operations globally despite the hazards and all the safety concerns associated with it. AERO 7260 HFP is Cytec’s innovative solution to this challenging issue.

Advantages of AERO 7260 HFP • Depression Efficiency AERO 7260 HFP is a highly efficient depressant for Cu sulfides and pyrite which effectively replaces 50 to 90% of NaSH depending on the process conditions. • Dosage-Performance AERO 7260 HFP requires only 10% to 20% of the dosage of NaSH, providing similar metallurgical performance. • Stability and Ease of Handling – Stable and chemically inert reagent in storage, transportation, and under process conditions – Does not release H2S or other toxic gases, and is non-hazardous – Classified as non-hazardous to the environment 16

– No downstream or upstream effects to mineral processing – Easy-to-handle aqueous solution – Completely miscible in water • pH AERO 7260 HFP is effective in a wide pH range (6 to 12). • Staged Addition AERO 7260 HFP is long lasting reagent eliminating the necessity of staged addition down the bank in scavengers and cleaner cells as with NaSH. • Bulk Concentrate Pretreatment Eliminates pretreatment of bulk Cu-Mo concentrate with steam, acid and CO2 conditioning, attrition conditioning, etc.

Solvent Extraction, Mineral Processing and Alumina Processing

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide • Applicability and Other Advantages – Eliminates the need for N2 or covered cells. – Does not require extended conditioning time.

– Does not contain any phosphorous or arsenic, so is suitable in many MoS2 operations. Clearly, with such advantages, AERO®7260 HFP offers a significant technological step forward in minimizing human and environmental hazards in Cu-Mo separations.

Proven Performance of AERO 7260 HFP – Lab and Plant Data The cumulative Cu and Mo recoveries from the concentrate from a North American mine are shown in Figure 2. For this concentrate sample, 7.5 kg/T of NaSH was required to provide efficient Cu depression (Cu recovery ~ 10%) and

Mo recovery of greater than 95%. AERO 7260 HFP at 0.5 kg/T replaced approximately 65% of the NaSH dosage and provided comparable Cu depression.

100

Cu

Mo

80     ) 60     %     (    y    r    e    v    o    c    e     R 40

FIGURE 2:

20

CUMULATIVE CU AND MO RECOVERY FROM A CU�MO NORTH AMERICAN CONCENTRATE

0 NaSH 7.5 kg/T

IN PROCESS SEPARATION

NaSH 7.5 kg/T + 7260 0.52 kg/T

17

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide In Figure 3, the Cu and Mo recoveries for a Cu-Mo concentrate from an Asian mine are shown. Efficient Cu depression was achieved only when 44 kg/T of Na2S was used. Under these conditions, Cu recovery was about 20%

and Mo recovery was about 80%. The effect of 1.2 kg/T AERO®7260 HFP helped achieve even better Cu depression and Mo selectivity with only half the dosage of Na2S.

90 80

80.9

Mo

81.3

Cu

70 60

    )     %     ( 50    y    r    e    v    o    c 40    e     R

30 20

19.8 11.8

10

CUMULATIVE CU AND MO RECOVERY FROM A CU�MO ASIAN CONCENTRATE

0 Na2S 44 kg /T

18

FIGURE 3:

Na2S 22 kg /T, AERO 7260 HFP 1.2 kg/T

Solvent Extraction, Mineral Processing and Alumina Processing

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide Figures 4A and B show Cu, Mo, and Fe recoveries and grades for lab data using AERO®7260 HFP on another North American mine Cu-Mo cleaner concentrate. In terms of Cu depression, this concentrate required about 11 kg/T of NaSH; however the Fe depression was not efficient at this dosage. For efficient Cu and Fe depression, a higher dosage

99.2

A� 100

99

of 55 kg/T NaSH was required. The addition of 0.25 kg/T of AERO 7260 HFP plus 11 kg/T of NaSH significantly enhanced both Cu and Fe depression and Mo selectivity. This suggests that AERO 7260 HFP is highly effective in the depression of both Cu and Fe and enables mine operations to significantly reduce NaSH consumption, in this case by over 80%.

98.8 92

NaSH 55 kg/T

90

NaSH 11 kg/T NaSH 11 kg/T + 0.25 kg/T 7260

80 68.4

70     ) 60     %     (    y    r 50    e    v    o    c    e     R 40

47.2 36.9 36.9

33

30 20 10 0 Cu

Mo

Fe

Mo Concentrate 50.7

49.2

B� 50

45.4

40

    ) 30     %     (    e     d    a    r     G 20

FIGURE 4: 10 3.2 0

0.2

0.2

4.2

0.2

Cu

Mo

2.4

2ND CLEANER CIRCUIT LAB DATA OF CU, MO AND FE A� RECOVERY AND B� GRADE

Fe

Mo Concentrate

Clearly, the benefits of adding AERO 7260 HFP are observed by the improved metallurgical performance and substantially reduced dosage of NaSH.

IN PROCESS SEPARATION

19

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide Figure 5 provides the average Mo assay in the scavenger tails from another Cu-Mo plant. The overall objective in this plant was to significantly reduce or eliminate Nokes (1400 g/T) usage in its Mo circuit, while maintaining Mo recovery

(Mo < 0.2% in scavenger tails). With only about 100 to 200 g/T of AERO®7260 HFP, a significant volume of Nokes was replaced, while the key specifications were maintained.

.40 .35     ) .30     %     (     l     i    a     T .25    v    a    c     S    n .20     i    o     M    e .15    g    a    r    e    v     A .10

FIGURE 5: PLANT DATA FOR MO IN SCAVENGER TAILS AS A FUNCTION OF NOKES DOSAGES USED

.05 0 Nokes

Standard

50% Reduction

75% Reduction

Figure 6 shows the plant data when using AERO 7260 HFP in an on/off cycle on 3 consecutive days. The plot shows the percentage difference in Cu, Mo and Fe grades in the cleaner circuit with and without AERO 7260 HFP on any given day. In the off-cycle, only NaSH was being used to control the respective grades in order to meet production specifications. With NaSH only, both Mo and Cu specifications were achieved while Fe was above the specifications, i.e. sufficient

20

100% Reduction

pyrite depression was not achieved. With the addition of AERO 7260 HFP (on-cycle), all the specifications were achieved in addition to reducing the NaSH consumption by over 60%. Further, it was observed that Mo grades were significantly better in the on-cycle. This clearly suggests the benefits of AERO 7260 HFP in such operations. Moreover, through optimization, the NaSH dosage could be reduced by 80%, by adding only about 2 kg/T of AERO 7260 HFP.

Solvent Extraction, Mineral Processing and Alumina Processing

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide     ) 20     %     (     P     F     H     0     6 0     2     7     O     R     E     A -20     h    t     i    w     d    n    a     l -40    o    r    t    n    o    c    n    e    e -60    w    t    e     b    e    c    n -80    e    r    e     f     f     i     D     %-100

Fe

Cu

Mo

FIGURE 6: THREE CONSECUTIVE DAYS OF PLANT DATA USING AERO®7260 HFP IN AN ON/OFF CYCLE IN THE CLEANER CIRCUIT. THE % DIFFERENCE IN FE, CU, AND MO GRADES BETWEEN CONTROL �OFF CYCLE� AND WITH AERO®7260 HFP �ON CYCLE� IS SHOWN Day 1

Day 2

Day 3

General Guidelines for Application • The typical dosages to test AERO®7260 HFP is around 250-1500 g/T, and needs to be adjusted depending on the ore mineralogy and other process conditions. Higher dosages may be evaluated as needed. Optimization should be based upon Cu and pyrite depression, Mo selectivity, and economics. • The performance of AERO7260 HFP is best when air is used. Note: N2 can be used, however the performance advantages and benefits of AERO 7260 HFP may not be fully realized. • Pretreatments are not required with AERO 7260 HFP.

• AERO 7260 HFP should be added along with NaSH (or Nokes/Na2S). • Recommended conditioning times are 5 to 15 minutes. Longer conditioning times, e.g. 30 min or longer are not required. • AERO 7260 HFP can be added in the roughers, scavenger or cleaner stage, as needed. Usually, if the dosages are optimized, stage addition is not required. • AERO 7260 HFP can be added as-is, or may be diluted as needed.

Other Applications for AERO 7260 HFP AERO 7260 HFP is an excellent depressant for a variety of sulfide minerals, selectivity being dictated by dosage of AERO 7260 HFP and process conditions. Products based on AERO 7260 HFP have a wide range of applications including:

b) Depression of penalty/toxic elements in Cu and complex sulfide ores.

a) Rejection of gangue from sulfide concentrates: Depression of all sulfide minerals while floating Non Sulfide Gangue E.g. Ni-talc separation.

d) Depression of iron sulfides in Cu-pyrite and Zn-pyrite separations.

IN PROCESS SEPARATION

c) Enhancement of selectivity in Cu-Pb, Pb-Zn, Cu-Zn separations.

e) Depression of Cu sulfides and pyrite in Cu-Mo, Cu-graphite, Cu-F, Cu-Talc separations. 21

AERO®7260 HFP: Novel, Safe and Sustainable Alternative to Traditional Hazardous Modifiers – NaSH, Nokes, Na2S, and Cyanide

Conclusion AERO®7260 HFP is a novel, safer, versatile and highly effective Cu sulfide and pyrite depressant with broad applicability. This paper focuses on the application and benefits of using AERO 7260 HFP in Cu-Mo separations. The examples discussed in the paper include both lab and plant data which highlight the effectiveness of AERO 7260

HFP in depressing Cu sulfides and pyrite and improving the selectivity with respect to Mo. In addition to enhanced selectively, dosages of hazardous reagents such as NaSH, Nokes, and Na2S could be reduced by 60%-80% with relatively small dosage of AERO 7260 HFP (0.5 to 2 kg/T).

References D. R. Nagaraj, S. S. Wang, P. V. Avotins and E. Dowling, Structureactivity relationships for copper depressants, Trans. IMM, Sect C: Vol 95, 1986, pp. 17-26. D.R. Nagaraj, C.I. Basilio, R.-H. Yoon and C. Torres, The Mechanism of Sulfide Depression with Functionalized Synthetic Polymers, Proc. Symp. “Electrochemistry in Mineral and Metals Processing”, The Electrochemical Society, Princeton, Proceedings Vol 92-17, 1992, pp 108-128. Chander, S. 1988. Inorganic depressants for sulfide minerals. Chapter 14 in Reagents in Mineral Technology. Edited by P. Somasundaran and B.M. Moudgil. New York: Marcel Dekker.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries. 22

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions Mario Palominos* and Carmina Quintanar

In recognition of the growing interest in meeting sustainability challenges, Cytec has been focused on the creation of greener products (collectors, modifiers and frothers) and processes using the FLOTATİON MATRİX 100 ™ approach.

Abstract The mining industry is currently facing significant sustainability challenges in terms of dealing with difficultto-process low-grade resources. These ores are typically characterized by complex mineralogy and the presence of significant amounts of penalty gangue sulfide minerals and toxic elements. Among them, pyrite is a common challenge in many operations. Three chemical strategies for dealing with gangue sulfides and penalty elements include: a) selective flotation of value minerals while rejecting penalty minerals throughout the entire circuit; b) rejecting penalty minerals in an appropriate part of the circuit using selective depressants; and c) using a combination of selective collectors and depressants in appropriate parts of the circuit. New products and application technologies have been developed in recent years for implementing these strategies as dictated by the particular needs of a given plant. However, in recognition of the growing interest in meeting sustainability challenges, Cytec has been focused on the

creation of greener products (collectors, modifiers and frothers) and processes using the FLOTATİON MATRİX 100 ™ approach. Chemicals today play a critical role, not just in flotation, but in almost all areas of mineral processing. They will play an even greater role in tackling the challenges and achieving the goals of sustainable mineral processing, particularly in the areas of water efficiency and water resource management; waste reduction and remediation; minimizing environmental impact, safety and health risks (meeting and exceeding the requirements of regulations); energy efficiency; and dealing with difficult-to-process, low-grade mineral resources and reserves. Together, these challenges are often termed greener processing. There is also a growing desire to develop greener chemicals, a major challenge in itself. Different strategies for dealing with difficult-to-process low-grade resources in a sustainable manner are evaluated in order to determine the most efficient alternatives. The discussion includes an overview of recent developments at Cytec using case studies in which the application of selective collectors and polymeric modifiers, including the newer, greener chemistries, are demonstrated.

23

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions

Introduction In earlier years, pyrite content and other sulphide gangues were less of a problem in the mineral processing of copper, lead, zinc and other elements, mainly due to the lower content of this mineralogical species, the high content of the valuable minerals and the lower ecological sensitivity to gas emissions (principally SO2) coming from the smelter. The first goal was to achieve higher selectivity, which was achieved through the development of dithiophosphate alternatives to the well-known xanthates (introduced to the market in 1923). Subsequently, it was found that thionocarbamates (and most commonly the isopropyl ethyl derivative, IPETC), generally have a higher selectivity than the above-mentioned chemistries. A third stage in the development of selective collectors focused on xanthate esters and dithiocarbamates1. In parallel, the use of high pH 2 to depress pyrite was implemented (particularly as a cleaning step). Lime (CaO) was the depressant agent, and was used as a slurry (Ca(OH) 2 in preference to caustic soda (NaOH) or soda ash (Na 2CO3). Hence, the solution used was based on flotation at high pH (10-11) using a selective collector in the rougher stage and a very high pH (> 11) in the cleaning step. The solution was acceptable for the processing conditions at that time. However, the use of lime negatively affected the recovery of valuable secondary elements (e.g., molybdenum and gold). Currently, use of seawater is an additional limiting factor for the application of lime. A second alternative, employed now for several years, is based on the use of depressants for iron sulphides (mainly pyrite and pyrrhotite). Sodium cyanide yields some good results; however, secure handling and environmental issues make its use unattractive. Thus, sulphoxy depressants have been increasingly applied in recent years. A factor not always considered is the degree of activation of the pyrite, mainly by copper ions from altered or oxidized minerals. When pyrite is unactivated, it is possible to obtain good results using lime, sodium cyanide or sulphoxy species (such as sodium or ammonium sulphite or metabisulphite3).

24

When pyrite is activated, however, lime is much less effective, cyanide has its safety, health and environmental (SHE) issues and the sulphoxy species have to be used at high dosages. Furthermore, the degree of association of pyrite, particularly in conjunction with valuable species (copper, molybdenum, gold, lead, zinc etc.) must be considered. Selectivity should be for liberated pyrite in order to prevent the loss of any valuable species associated with the pyrite. Alternatives to inorganic depressants have also been utilized, including organic products from natural sources 4,5,6 (including quebracho, tannins and their derivatives) and ethylene diamine tetraacetic acid. In recent years, polymeric depressants have been developed that work effectively for both active and non-activated pyrites. These products are actually hydrophilic copolymers containing chemical functionality that is able to adhere selectively to iron sulphide species and lead to their depression. Importantly, polymeric depressants do not have the toxicity problems associated with the inorganic depressants, and they may be used at significantly lower doses. The need to process ores with higher iron sulphide content, the generally lower grades of valuable elements and the growing importance of secondary elements (molybdenum, gold, etc.), are driving greater interest in the use of selective collectors. In recent years, more selective reagents have been developed for the rougher stage in order to achieve selective flotation with high efficiency at this point, and thus minimise the use of depressants in the cleaning step. The compounds of interest have included structurally modified dithiocarbamates and thionocarbamates. These collectors have the advantage of being selective against liberated pyrite, but effective for the valuable elements associated with pyrite, such as copper, molybdenum, and others, thereby avoiding the loss of these valuable species related to the non-flotation of associated particles (middlings).

Solvent Extraction, Mineral Processing and Alumina Processing

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions

Methodology Mineral ore samples from South America were used to evaluate the application of selective collectors and polymeric depressants. The feed grades of the ores are listed in Table 1. TABLE 1: Feed Grades of the Mineral Ores Used in the Evaluation of Selective Collectors.

Ore

Copper content, %

Iron content, %

Molybdenum content, ppm

Ore-1

0.74

2.1

98

Ore-2

0.33

4.73

107

Ore-3

1.05

4.40

300

Experimental Procedure Laboratory flotation tests were conducted to simulate 1) just the Rougher stage and 2) the different stages of the plant (open cycle test). The flotation products were collected and analysed for copper, iron and molybdenum using atomic absorption analysis. These mass balance results allowed the

calculation of the metallurgical balance, and therefore the metallurgical recoveries, for each test. The conditions for the laboratory tests with the different mineral ores are described in Table 2.

TABLE 2: Laboratory Test Conditions for Each of the Ores

Conditions

Ore-1

Ore-2

Ore-3

Machine

Agitair L500

Denver

Wemco

pH

10.5

9.5

9.5

% Solids

34

34

30

Flotation time (min)

10

15

12

Grinding

30% + 100#Ty

20% + 100 #Ty

20% +65#Ty

Note that with Ore-3, when the standard collector was used, typical conditions for the cleaning stage were used (lime was added) and the pH was 11.5. However, lime was not added in

the cleaning stage for the other collectors tested with Ore-3 (final pH=8.7).

Results and Discussion The study with Ore-1 demonstrated the difference in the selectivity for iron for the different types of selective collectors: dithiophosphate (DTP),

IN PROCESS SEPARATION

isopropyl ethylthionocarbamate (IPETC) and a structurally modified thionocarbamate (SMTC).

25

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions

94

DT

IPET

SMTC

92 90    y    r    e    v 88    o    c    e     R    u 86     C

84

FIGURE 1: SELECTIVITY COMPARISON FOR COPPER MINERALS VS. PYRITE FOR 3 COLLECTOR TYPES

82 80 15

20

25

30

35

40

45

50

55

60

Fe Recovery

The difference in the performance of DTP and IPETC, as described above in the Introduction, can be readily seen in Figure 1. IPETC, one of the first selective collectors to replace the xanthates, provides good recoveries and better selectivity. Importantly, though, it can also be seen in Figure 1 that the structurally modified thionocarbamate AERO® XD-5002 promoter, which represents a new family of collectors developed by Cytec, is clearly advantageous in terms of its selectivity for copper minerals against pyrite (as represented by the Fe assay).

A complementary study was then conducted with a second ore (Ore-2) with different mineralogical characteristics. Again, a series of selective collectors was evaluated, including the structurally modified thionocarbamate AERO ® 9950 promoter, which provided the highest selectivity against iron and also the best copper recovery among the tested chemicals. The results of this study are presented in Figure 2, while the different collectors and their dosage levels used in the test are listed in Table 3. Collector-1 refers to the main collector that was added to the grind. Collector-2, when used, refers to a secondary collector added in the conditioning stage prior to initiation of flotation.

88.5 88 87.5

    )     %     ( 87    c    e     R    u     C 86.5

86

26

14.0

17.5

22.0

22.1

19.5

29.0

Fe Rec (%)

6.8

5.8

6.8

6.5

6.8

6.4

Mass Pull (weight %)

1

2

3

4

5

std

Test

FIGURE 2: EVALUATION OF SELECTIVE VERSUS NON�SELECTIVE COLLECTORS USING ORE 2

Solvent Extraction, Mineral Processing and Alumina Processing

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions at pH 11.5, the regular condition for depression when lime is used. For the evaluated alternatives, however, lime was not added in the cleaning stage, so that comparisons could be made with results obtained for the subsequent study using depressants (see below).

TABLE 3: Reagent Scheme for the Study Using Ore-2.

N°

Collector-1[M]

Collector-2 [C]

1

AP-9950; 20 g/t

–

2

XD-5002; 10 g/t

–

3

AP-9950; 15 g/t

MX-945; 5 g/t

4

MX-8522; 15 g/t

MX-945; 5 g/t

5

MX-7017; 15 g/t

MX-945; 5 g/t

STD

PAX; 20 g/t

–

[M]: Grind mill; [C]: Conditioning

The selective collectors evaluated with Ore-3 included AERO 9950 promoter (structurally modified thionocarbamate) and AERO® 9955 promoter (a mix of thionocarbamate and dithiocarbamate). Their performance was compared to that of the non-selective collector SIPX (sodium isopropyl xanthate), for which the standard conditions were used.

The third study included a cleaning stage (evaluated in an open cycle test). As indicated in the Experimental section, for the standard collector, the cleaning stage was conducted

The following figure (Figure 3) shows both the rougher and global recoveries (considering an open cycle test with two cleaning stages and a scavenger stage) conducted on Ore-3.

100 90

Cu

Cu-FC Grade =

Fe

Mo

80 70     ) 60     %     (    y    r 50    e    v    o    c    e     R 40

30

FIGURE 3:

20

COMPARATIVE STUDY BETWEEN A NON�SELECTIVE COLLECTOR �SIPX� AND TWO SELECTIVE COLLECTORS

10 0 Rougher

Final ST

Rougher

Final

Rougher

AP-

It can be seen in the figure that similar copper rougher recoveries were obtained for all three of the collectors, while in the rougher stage, the xanthate and AERO® 9950 promoter had similar molybdenum recoveries and the AERO® 9955 promoter provided a greater recovery. The iron recoveries in the rougher stage were significantly different, however. The xanthate had a high recovery (approximately 80%), followed by the AERO 9955 promoter (with a value near 65%), but the AERO 9950 promoter was the most selective (rougher Fe recovery of approximately 40%).

IN PROCESS SEPARATION

Final AP-

The overall recovery of iron for the xanthate was calculated to be 30% based on analysis of the final concentrate after the two cleaning steps and considering the classical cleaning at high pH. AERO 9955 promoter, meanwhile, had an overall iron recovery of close to 20%, while that of AERO 9950 promoter (the most efficient in the rougher stage) was approximately 15%. With these values, the grades obtained for the final concentrate in terms of the copper content were determined and are indicated in Figure 3. With both AERO 9950 promoter and AERO 9955 promoter, 27

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions the final copper concentrates reached or exceeded the requirements for commercial grade material. In addition, due to the lower pH, there was limited loss during recovery

of the by-product molybdenum (Mo) in the cleaning stage as compared to the reduction in the Mo recovery using the standard collector (SIPX).

Comparative Study with Different Pyrite Depressants Figure 4 shows the results using different depressants, such as lime (for the standard condition), sodium metabisulphite (MBS), which is currently used, particularly when seawater is used for the processing, and new polymeric organic depressants. The ore used in this work was the same as that

used to evaluate the selective collectors (Ore-3). In this study, the standard collector (xanthate) was used in all of the tests so that the effect of the different depressants could be evaluated in the cleaning stage.

100 90

Cu

Cu-FC Grade =

Fe

Mo

80 70     )     % 60     (    y    r 50    e    v    o    c    e     R 40

30

FIGURE 4:

20

INORGANIC AND ORGANIC DEPRESSANTS FOR PYRITE USING ORE�3

10 0 Rougher ST

Final

Rougher

Final

Coll-STD / A-7260; 50

Rougher

Final

Coll-STD / A-7260; 100

MBS and the polymeric depressant developed by Cytec, AERO® 7260 HFP depressant, were evaluated under similar conditions (pH = 8.5). The standard test used only lime as the depressant and was conducted at pH = 11.5. All three depressants were added at the regrind mill stage. Importantly, as can be seen in Figure 4, neither the standard or the alternative depressants reached the values necessary for commercial concentrate grades (Cu > 25%).

28

Rougher

Final

Coll-STD / MBS-Na; 300

The addition of the depressant in the rougher stage to simulate the effect of the selective collector was also evaluated. However, low depression of iron was observed. The most significant effect was that copper and molybdenum species were depressed at high levels.

Solvent Extraction, Mineral Processing and Alumina Processing

Rejection of Pyrite: Challenges and Sustainable Chemical Solutions

Conclusion The results presented above demonstrate that there are new alternatives available on the market that are even more selective than the classic collectors commonly used for pyrite and other sulphide mineral gangue and can address the increasing levels of these contaminants that are present in today’s mineral deposits.

In addition, it was also shown that it is more efficient to use highly selective collectors in the roughing stage, rather than to use collectors with low or medium selectivity in conjunction with depressants. In the latter case, high doses are typically required, particularly when using organic depressants, which were found to be inefficient and have the potential to negatively affect the recovery of both the main sulphide product and secondary products, such as molybdenum and gold.

References 1. Klimpel R. Richard, A discussion of traditional and new reagent Chemistries for a Flotation of Sulphide Minerals. Chapter 7, Reagents for Better Metallurgy, Society for Mining Metallurgy and Exploration Inc., Littleton, Colorado USA, 1964.

4. Pedro E. Sarquis, Adriana Moyano, Mercedes Gonzalez, Vanesa Bazán, Organic Depressant Reagent Effect on pyrite in Copp er Minerals Flotation, 8th International Mineral Processing Seminar (Procemin 2011), 109-116.

2. Yuqiong Li, Jianhua Chen, Duan Kang, Jin Guo, Depression of Pyrite in alkaline medium and subsequent activation by copper, Minerals Engineering 26 (2012) 64-69.

5. Maximiliano Zanin, Saeed Farrokhpay, Depression of Pyrite in Porphyry Copper Flotation, 8th International Mineral Processing Seminar (Procemin 2011), 135-143.

3. G.I. Dávila-Pulido, A. Uribe-salas, R. Espinosaa-Gomez, International Journal of Mineral Processing, 101 (2011) 71-74.

6. Jianhua Chen, Yuqiong Li,Ye Chen , Cu-S Flotation Separation via the combination of Sodium Humate and Lime in a low pH Medium, Minerals Engineering, 24 (2011), 58-63.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries.

IN PROCESS SEPARATION

29

MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption Morris Lewellyn, Alan Rothenberg, Calvin Franz, Frank Ballentine, Frank Kula, Luis Soliz, Qi Dai, and Scott Moffatt

As the premier advanced chemicals partner for the Alumina industry, Cytec specializes in producing products with the breadth and depth to advance all stages in the Bayer Process. Our product innovations have transformed the industry’s expectations regarding their technology suppliers and our strategy is to continue to develop solutions that will provide step changes in the industry. Our MAX HT® scale inhibitor, a revolutionary product that eliminates sodalite scale from heat exchangers, recently received the 2012 Environmental Protection Agency’s Presidential Green Chemistry award. The award recognizes companies that have pioneered sustainable technologies that incorporate the principles of green chemistry. MAX HT was developed to reduce or eliminate scaling from the evaporator and digester heaters in the Bayer process. This product has been successfully applied in 20 Bayer process plants worldwide, resulting in the significant benefits of increased heat transfer, reduced energy consumption and reduced acid waste from reduced heater cleanings. Based

on trial data from a number of plants, the estimated annual savings per ton of alumina produced are 0.26-1.3 Gj energy, resulting in 13-92 kg reduction in CO2 emissions, and 0.9-2.7 kg reduction in acid waste.* When these savings are applied to the total alumina production from the 20 plants, this leads to an estimated realized annual savings of 11-56 million Gj energy, 0.54-3.9 billion kg CO 2 emissions, and 38-116 million kg of acid waste reduction. * The range reflects the wide variety in the operation of Bayer plants around the world.

Introduction Cytec has developed a line of polymers for use as scale inhibitors in evaporator and digester heaters used in the Bayer process [1-8]. These products provide benefits by reducing or eliminating the scale formation in the heaters resulting in significantly higher heat transfer, reducing energy consumption and waste. These products have been successfully applied in a number of plants utilizing the Bayer process throughout the world [9-11]. This technology is also being assessed for sodalite scale elimination in the evaporation process for the treatment of other types of substrate [12]. The scale deposited in these heaters is sodium aluminosilicate – sodalite or DSP (desilication product). 30

This is a result of the silica that is present in bauxite ores as silicates, primarily clay minerals, that dissolves quickly under typical Bayer alumina digestion conditions. The Bayer liquor remains supersaturated in silica and this supersaturation is greatest after the alumina precipitation step, i.e. in the spent liquor. As the alumina-depleted liquor is reheated, the rate of silica precipitation in the form of sodalite increases markedly with increasing temperature due to faster kinetics[13]. This precipitation occurs as scaling on the inside of the heat exchange tubes and a significant loss of heat transfer occurs, leading to increased energy consumption, increased caustic losses, reduced liquor flows, reduced throughput, reduced evaporation, and reduced production.

Solvent Extraction, Mineral Processing and Alumina Processing

MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption Without MAX HT, the method used to manage the sodalite scale problem is to clean out the system whenever the heat exchanger performance drops below a certain level, typically about half the original heat transfer rate. This cleaning is generally accomplished with the use of a 5-10% sulfuric acid solution to dissolve the scale. The used acid constitutes a waste stream requiring disposal. In addition to the acid cleaning, much of the inter-stage piping is cleaned using mechanical means, such as jackhammers, to remove the scale.

Cytec performed plant trials to research energy savings. See Figures 4 & 5. For detail on the results of the trials please go to www.cytec.com. The results of the trials show potential energy and waste savings which was the basis for the awarding Cytec the 2012 U.S. EPA (Environmental Protection Agency) Presidential Green Chemistry Challenge Award for MAX HT. Without Antiscalant

With Antiscalant

The use of MAX HT is one way to make the Bayer process greener in terms of energy use and waste generation. MAX HT is commercial at twenty different plants across the globe, and under evaluation at a number of other plants. Many of these are double stream plants where the scale inhibitor can be used on both evaporator and digester heaters, but there are a number of single stream plants that find benefits from  just treating the evaporator heaters. DIRTY AND CLEAN HEAT EXCHANGERS FROM OPERATING WITHOUT AND WITH MAX HT ANTISCALANT AFTER 160 HRS. OF CONTINUOUS OPERATION, CORRESPONDING TO THE HEAT TRANSFER CURVES IN FIGS. 4 AND 5, RESPECTIVELY.

2700 2500  y = -8.325x + 1981.8

2300 2100 1900 1700 1500 1300 1100 900

FIGURE 4:

700

TYPICAL HEAT TRANSFER DECAY DURING 160 HRS. WITHOUT MAX HT

500 0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150 160

Hours

IN PROCESS SEPARATION

31

MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption

2300 2100

 y = -0.4267x + 1744.1

1900 1700 1500 1300 1100

FIGURE 5:

900

CONSTANT HEAT TRANSFER COEFFICIENT RESULTING FROM THE USE OF MAX HT

700 500 1

11

21

31

41

51

61

71

81

91

101 111 121 131 141 151 161

Hours

Benefits of MAX HT MAX HT sodalite scale inhibitor has been used successfully in a number of Bayer process plants. Typically, the on-stream time for a heater is increased from some 8-10 days to 45-60 days for digestion and 20-30 days to >150 days for evaporators. This ability to maintain a high heat transfer over a much longer life cycle between cleanings has resulted in a number of benefits. These benefits are summarized below: 1. Increased evaporation when used in the evaporator heaters. This leads to reduced caustic consumption and improved mud washing in the washer circuit because more water is available for efficient washing of the red mud and gibbsite crystals. The annual realized reduction of caustic is estimated to be 79,000-198,000 tons of 50% caustic. 2. Increased production. This is a result of an increased average flow due to being able to maintain the outlet temperature without having to reduce flow to accommodate a lower heat transfer rate. 3. Reduced energy consumption realized per annum. Savings in the range of 4.4-22.0 million tons of steam have been realized, which translates to 11-56 million Gj energy, or 0.54-3.9 billion kg CO 2. 32

4. Less direct steam to the digester when used in the digester heaters. By being able to maintain the maximum live steam heater outlet temperature, the need to add steam by direct injection in the digesters is reduced or eliminated, resulting in less extraneous dilution which impacts soda recovery and therefore caustic consumption. This also allows more mass in the digester in terms of liquor and bauxite leading to higher production. 5. Reduced digester and evaporator heaters cleaning and maintenance. This leads to a reduction in cost for the acid, labor, tube changes, etc. There is also less exposure of the workers to the associated hazards. The realized annual reduction in hazardous acid waste is 38-116 million kg. The number of cleaning cycles can be reduced from a range of 20-50 per year per heater train to less than 10 per heater train. 6. Steadier plant operation.

Solvent Extraction, Mineral Processing and Alumina Processing

MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption The value of the MAX HT technology is to reduce energy usage by 0.26-1.3 Gj, reduce CO 2 emissions by 13-92 kg, and reduce waste generation by 0.9-2.7 kg per ton of alumina produced. There are about 73 operating Bayer plants throughout the world, ranging in production capacity of

0.2 to 6 million tons of alumina annually, with the majority being in the 1.5 to 3 million tons capacity. The estimated annual environmental benefit for the 20 commercialized plants is shown in Table 6 along with the estimated global annual potential benefit based on 2011 figures.

TABLE 6. POTENTIAL AND REALIZED BENEFITS OF MAX HT TECHNOLOGY Energy (Gj)

CO2 Reduction (Kg)

Waste Reduction (Kg)

0.26 – 1.3

13 – 92

0.9 – 2.7

Realized savings (20 commercial plants)

11 – 56 million

0.54 – 3.9 billion

38 – 116 million

Potential savings (All 73 plants)

25 – 128 million

1.3 – 8.9 billion

86 – 263 million

Savings per ton of alumina produced

Conclusion 1. MAX HT provides estimated annual savings per ton of alumina produced of 0.26-1.3 Gj energy, resulting in 13-92 kg reduction in CO 2 emissions and 0.9-2.7 kg reduction in acid waste. 2. This more efficient use of energy results in increased evaporation in the evaporator heaters, leading to reduced caustic consumption and more efficient use of water.

3. The use of MAX HT leads to increased production due to an increase in average flow and reduced direct steam injection. 4. The use of MAX HT also results in reduced cleaning of digester and evaporator heaters resulting in reduced exposure of workers to the hazards and reduced acid waste.

References 1. D. Spitzer, A. Rothenberg, H. Heitner, and F. Kula, “Method of preventing or reducing aluminosilicate scale in a Bayer process,” U.S. patent 6,814,873B2 (2004). 2. D. Spitzer, A. Rothenberg, H. Heitner, and F. Kula, “Method and compositions for preventing or reducing aluminosilicate scale in alkaline industrial processes,” U.S. patent 7,390,415B2 (2008). 3. H. Heitner, “Silane substituted polyethylene oxide reagents and method of using for preventing or reducing aluminosilicate scale in industrial processes,” U.S. patent 7,674,385B2 (2010). 4. H. Heitner and D. Spitzer, “Hydrophobically modified polyamine scale inhibitors,” U.S. patent 7,999,065B2 (2011).

6. D. Spitzer, A. Rothenberg, H. Heitner, F. Kula, M. Lewellyn, O. Chamberlain, Q. Dai, and C. Franz, “A real solution to sodalite scaling problems,” Proceedings of the 7th International Alumina Quality Workshop, 2005, 153-157. 7. D. Spitzer, O. Chamberlain, C. Franz, M. Lewellyn, and Q. Dai, “MAX HT Sodalite Scale Inhibitor: Plant experience and impact on the process,” Light Metals, 2008, 57-62. 8. M. Lewellyn, A. Patel, D. Spitzer, C. Franz, F. Ballentine, Q. Dai, O. Chamberlain, F. Kula, and H. Chen, “MAX HT Sodalite Scale Inhibitor: Plant experience with first and second generation products,” Proceedings of the 8th International Alumina quality Workshop, 2008, 121-124.

5. D. Spitzer, A. Rothenberg, H. Heitner, F. Kula, M. Lewellyn, O. Chamberlain, Q. Dai, and C. Franz, “Reagents for the elimination of sodalite scaling,” Light Metals, 2005, 183-188.

IN PROCESS SEPARATION

33

MAX HT® Bayer Sodalite Scale Inhibitor: A Green Solution to Energy Consumption 9. A. Oliveira, J. Dutra, J. Batista, J. Lima, R. Diniz, and E. Repetto, “Performance appraisal of evaporation system with scale inhibitor application in Alunorte plant,” Light Metals, 2008, 133-136. 10. L. Riffaud, P. James, E. Allen, and J. Murray, “Evaluation of sodalite scaling inhibitor – A user’s perspective” (Paper presented at the International Symposium on Aluminum: From Raw Materials to Applications – Combining Light Metals 2006 and the 17th International ICSOBA Symposium, Montreal, Quebec, Canada, 1-4 October 2006), Bauxite & Alumina session, paper 29.7. 11. M. Kiriazis, J. Gill, and D. Stegink, “Evaluation of MAX HT® at Queensland Alumina LTD,” Proceedings of the 9th International Alumina Quality Workshop, 2012, 88-92.

12. L.N. Oji, T.L. Fellinger, D.T. Hobbs, H.P. Badheka, W.R. Wilmarth, M. Taylor, E.A. Kamenetsky, and M. Islam, “Studies of potential inhibitors of sodium aluminosilicate scales in high-level waste evaporation,” Separation Science and Technology, 43 (2008), 2917-2928. 13. J. Addai-Mensah, “Fundamentals of sodium aluminosilicate crystallization during process heat exchange,” (Paper presented at the World congress of Chemical Engineering, 7th, Glasgow, United Kingdom, 10-14 July 2005), 8 6781/1-86781/10. 14. L. Perez, “Mechanism of calcium phosphate scale formation and inhibition in cooling systems,” Calcium Phosphates in Biological and Industrial Systems, ed. Z. Amjad (Boston, Kluwer, 1998), 395-415.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries. 34

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) Luis Soliz, Renata Vinhas, Emiliano Repetto, Paulo Prado, Haunn-Lin T.Chen, Roberto Seno Junior, Andre Arantes, Rodrigo Santos, and Rodrigo Moreno

In the late 1980’s, Cytec Industries Inc. developed a series of hydroxamated polyacrylamide products (HXPAM) to be used as high performance flocculants for treating red mud in the Alumina Bayer Process 1,2,3. This product family, known as CYFLOC ® HX Series, has since become the most widely used flocculant on red mud settlers in the industry.

Introduction The applications of several flocculants from CYFLOC® HX ‘00 and ‘000 series have been evaluated in a variety of alumina plants worldwide. Since then, the alumina industry has realized a number of significant benefits from the use of CYFLOC HX flocculants.

The replacement of CYFLOC ULTRA HX-5300 by CYFLOC ULTRA HX-6000 series – HX-6100, HX-6200, HX-6300, and HX-6400 was due to further improvements in the manufacturing process. The new series will be available in the market in the near future.

After years of producing HXPAM emulsions, CYTEC Industries Inc. has developed a family of new hydroxamated polyacrylamide products5 – CYFLOC ULTRA HXPAMS. CYFLOC ULTRA HX-5300 was the first of a new series.

This paper describes the relative performance of CYFLOC ULTRA HX-5300 versus CYFLOC HX-3000 in the CBA alumina plant.

Background – CBA Bauxites Characterization There are two types of bauxites processed in CBA: Zona da Mata and Pocos de Caldas. Both bauxites are commonly blended and added into the process in different ratios according to the production plan. The Pocos de Caldas ore has shown to be more difficult to process than Zona da Mata ore due to the higher content of reactive silica as well as the lower amounts of iron and

available alumina content. Operational adjustments are required in order to control instabilities in the clarification stage when the blend contains more than 40% Pocos de Caldas bauxite. These instabilities include: difficulties in red mud settling, and a loss of compaction in the vessel, which leads to reduced throughput and consequently reduced production.

Benchmarking of CYFLOC ULTRA HX-5300 Performance In December of 2011, settling tests were conducted at CBA in order to determine the performance of CYFLOC ULTRA HX-5300 relative to the CYFLOC HX-3000 sample applied in the process used on the settlers and first washer at the time.

Samples of settler feed slurry were collected and tested in CBA’s laboratory. According to the process data, the blend of Zona da Mata and Pocos de Caldas bauxite at the time of testing was 71% and 29% respectively. 35

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) Figure 1 indicates that the manufacturing and laboratory samples of CYFLOC® ULTRA HX-5300 provided a significant improvement in the red mud settling rate and overflow clarity. Approximately 238 g/T of CYFLOC HX-3000 was required in order to deliver a settling rate of 10 m/h. Conversely, the same settling rate was achieved with approximately 194 g/T of CYFLOC ULTRA HX-5300. Findings from these tests showed that the dosage of flocculant was reduced by approximately 18% when CYFLOC ULTRA HX-5300 was applied versus CYFLOC HX-3000.

Similar to settling rate performance, samples of CYFLOC ULTRA HX-5300 also outperformed CYFLOC HX-3000 in clarity, as indicated in Figure 2. At a dosage of 225 g/T, CYFLOC ULTRA HX-5300 delivered supernatant clarity of approximately 80mg/L versus 140 mg/L when CYFLOC HX-3000 was applied. After positive results were achieved in the laboratory a plant trial was recommended in order to verify the product performance in the plant.

30 CYFLOC® HX-3000 CYFLOC® Ultra HX-5300 (Lab. Lot)

25

CYFLOC® Ultra HX-5300 (Manuf. Lot)     ) 20     h     /    m     (    e    t    a     R 15    g    n     i     l    t    t    e     S 10

5

FIGURE 1:

0

SETTLING RATE VERSUS TOTAL DOSE OF FLOCCULANT 100

150

200

250 Total Dose (g/T)

300

350

400

600 CYFLOC® HX-3000 CYFLOC® Ultra HX-5300 (Lab. Lot) CYFLOC® Ultra HX-5300 (Manuf. Lot)

500     )     L     / 400    g    m     (    s     d     i     l    o     S 300    w    o     l     f    r    e    v 200     O

FIGURE 2:

100

SUPERNATANT SOLIDS VERSUS TOTAL DOSE OF FLOCCULANT

0 100

36

150

200

250 Total Dose (g/T)

300

350

400

Solvent Extraction, Mineral Processing and Alumina Processing

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio)

Plant Application of CYFLOC® ULTRA HX-5300 In February 2012, a plant trial was conducted at CBA for a period of ten days. The purpose of the trial was to demonstrate that it is possible to control the settler stage with the application of CYFLOC ULTRA HX-5300 while achieving a dose reduction of approximately 18% versus CYFLOC HX-3000. Based on historical data of two previous months, the success criteria of the trial was to maintain process conditions similar to the prior conditions when CYFLOC HX-3000 was applied. These conditions included: − Maintain the overflow (O/F) solids equal to or less than, 80 mg/L

− Maintain underflow (U/F) density equal to or higher than, 1450 g/L − Maintain the interphase level at approximately 4.5 meters − Maintain U/F density equal to or higher than 1450 g/L − Do not negatively impact filtration operation (carry over measurements) The plant trial was carried out on the TD1 settler according to the diagram below:

Flocculant Overflow Settler Feed Slurry

Filtrate (KM)

Flocculant Mud Settler TD 1

Mud Washer

Mud Washer

Mud Washer

Mud Washer Red Mud Residue

FIGURE 3: CBA’S CLARIFICATION FLOW DIAGRAM

Results and Discussion During the trial the target of Pocos de Caldas bauxite content was 38% in the blend. However due to logistic issues, the real content of Pocos bauxite in the blend varied from 27% to 45%. That wide range of Pocos de Caldas bauxite c harge was an interesting aspect to be evaluated. Even when the amount to Pocos bauxite increased from 27% to 45%, it was possible to control the settler stage with the CYFLOC ULTRA HX-5300 application. This observation suggested that CYFLOC ULTRA HX-5300 has significant potential to allow for flexibility in bauxite savings. However, a longer term evaluation is required to confirm this hypothesis.

IN PROCESS SEPARATION

Figure 4 contains sixty days of overflow clarity data from a period prior to the trial as well as overflow clarity data that was collected during the ten day trial period. The data indicates that the overflow clarity was maintained at approximately 75mg/L in the TD 1 settler when CYFLOC Ultra HX-5300 was applied. The mean and the standard deviation in both periods were very similar. Over the last few days of the trial the data suggested that the overflow clarity increased to approximately 100 mg/L. This deterioration is attributed to inefficient dispersion of the flocculant in the vessel caused by a reduction of the feed rate.

37

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) HX-3000

ULTRA HX-5300

120    e 100    u     l    a     V     l    a 80    u     d     i    v     i     d 60    n     I

UCL = 98.72 X = 75.41 LCL = 52.09

40 1/12/11 8/12/11 15/12/11 25/12/11 1/1/12

8/1/12

15/1/12 22/1/12 29/1/12 5/2/12

Date HX-3000

ULTRA HX-5300

40    e 30    g    n    a     R    g 20    n     i    v    o     M 10

UCL = 28.65

0 1/12/11 8/12/11 15/12/11 25/12/11 1/1/12

MR = 8.77

FIGURE 4:

LCL = 0

OVERFLOW CLARITY CHART

8/1/12 15/1/12 22/1/12 29/1/12 5/2/12

Date

Specifically, the CBA measurement of the interphase level is taken from the top of the settler. The data from the period prior to the trial indicated that the interphase level HX-3000

was approximately 4.2 meters. Figure 5 illustrates that the interphase level remained stable and in some cases slightly improved to 4.6 meters when the new flocculant was applied. ULTRA HX-5300

UCL = 5.483    e    u     l    a     V     l    a    u     d     i    v     i     d    n     I

5 X = 4.567 4 LCL = 3.651 3 1/12/11 8/12/11 15/12/11 22/12/1129/12/11 5/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date HX-3000

ULTRA HX-5300

1.5    e    g    n    a 1.0     R    g    n     i    v    o 0.5     M

UCL = 1.125

0 1/12/11 8/12/11 15/12/11 22/12/1129/12/11 5/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date

38

MR = 0.344

FIGURE 5:

LCL = 0

INTERPHASE LEVEL CHART

Solvent Extraction, Mineral Processing and Alumina Processing

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) Similar to the overflow clarity and interphase level, the underflow density also remained stable during the trial. Figure 6 depicts the mud density data prior to the trial and

HX-3000

during the trial. The mean density prior to the trial and during the trial was approximately 1515 g/L and 1521 g/L, respectively.

ULTRA HX-5300

UCL = 1607.0

1600    e    u     l    a 1550     V     l    a    u     d     i    v 1500     i     d    n     I

X = 1521.1

1450

LCL = 1435.2 1/12/11 8/12/11 15/12/11 22/12/1129/12/11 5/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date HX-3000

ULTRA HX-5300

UCL = 105.5

100    e    g 75    n    a     R    g 50    n     i    v    o     M 25

MR = 32.3

FIGURE 6:

0

LCL = 0

UNDERFLOW DENSITY CHART

1/12/11 8/12/11 15/12/11 22/12/11 29/12/115/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date

The Filtration Rate parameter was also monitored. Figure 7 showed that the filtration rate remained stable during the trial. CYFLOC® Ultra HX-5300 delivered a filtration rate of approximately 0.97 m3/m2/h versus 0.94 m3/m2/h when CYFLOC HX-3000 was applied.

IN PROCESS SEPARATION

The carry over was also monitored to ensure that no flocculant was carried over to the filters. No problems or issues were reported from the security filtration stage.

39

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) HX-3000

ULTRA HX-5300

1.1

UCL = 1.0704

   e    u 1.0     l    a     V     l    a    u     d 0.9     i    v     i     d    n     I

X = 0.969 LCL = 0.8676

0.8

1/12/11 8/12/11 15/12/11 22/12/1129/12/11 5/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date HX-3000

ULTRA HX-5300

.20    e    g .15    n    a     R    g .10    n     i    v    o     M .05

UCL = 0.1246

0

MR = 0.0381

FIGURE 7:

LCL = 0

FILTRATION RATE CHART

1/12/11 8/12/11 15/12/11 22/12/11 29/12/115/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date

Figure 8 illustrates the flocculant dosage usage in the plant both prior to and during the trial. Statistical evaluation indicated that CYFLOC ® HX-3000 dosage was approximately 252 g/T of mud, while CYFLOC Ultra HX-5300 was 208 g/T.

40

This represented a dosage reduction of approximately 18% when CYFLOC Ultra HX-5300 was applied as opposed to CYFLOC HX-3000, even during the time period in which 45% Pocos de Caldas bauxite was added the process.

Solvent Extraction, Mineral Processing and Alumina Processing

Performance Evaluation of CYFLOC® ULTRA HX-5300 – A New HXPAM Red Mud Flocculant Applied in CBA (Companhia Brasileira de Aluminio) 350

HX-3000

ULTRA HX-5300

   e 300    u     l    a     V     l    a 250    u     d     i    v     i     d    n 200     I

UCL = 246.2 X = 207.9 LCL = 169.5

150 1/12/11 8/12/11 15/12/11 22/12/1129/12/11 5/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date HX-3000

ULTRA HX-5300

80

   e    g    n 60    a     R    g    n 40     i    v    o     M 20

UCL = 47.1

0

MR = 14.4

FIGURE 8:

LCL = 0

FLOCCULANT DOSAGE CHART

1/12/11 8/12/11 15/12/11 22/12/11 29/12/115/1/12 12/1/12 19/1/12 26/1/12 2/2/12 9/2/12

Date

Conclusion 1. The test indicated that CYFLOC ® ULTRA HX-5300 had equivalent settler performance as compared to CYFLOC HX-3000. 2. It was possible to control the settler stage while applying CYFLOC ULTRA HX-5300, even when the process had the addition of 45% Pocos de Caldas bauxite.

3. There were no problems or issues reported by the security filtration stage during the trial. 4. It was possible to reduce flocculant dosage by approximately 18% with the application of CYFLOC ULTRA HX-5300 relative to CYFLOC HX-3000.

References 1. Spitzer, D. P. and Yen, W. S., US 4,767,540 (1988) 2. Rothenberg, A. S., Spitzer, D. P., Lewellyn, M.E., and Heitner, H. I. New reagents for alumina processing, Light Metals, (1989), pp 91-96. 3. Spitzer, D. P., Rothenberg, A. S., Heitner, H. I., Lewellyn, M.E., Laviolette, L. H., Foster, T., and Avotins, P. V., Development of new Bayer process flocculants, Light Metals, (1991), pp 167-171.

4. Ryles, R. G. Avotins, P. V., SUPERFLOC® HX, a new technology for the alumina industry, Fourth International Alumina Quality Workshop, (1996), pp 75-86. 5. Rothenberg, A. S., Lewellyn, M.E., Chen, H-L.T., Magliocco, L., Sassi, T., Microdispersions of hydroxamated polymers and method of making them, Patent Application Filed.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries.

IN PROCESS SEPARATION

41

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants  John Carr, Lei Zhang, Matthew Davis, S.A. Ravishankar, and Greg Flieg

Cytec has developed a novel family of antiscalants, PHOSFLOW®, that reduces the fluorosilicate, calcium sulfate and aluminate scale formation in phosphoric acid operations allowing for l onger cycle times, increased production, and reduced maintenance time.

1. Introduction Accumulation of scale is a perennial problem in phosphoric acid production processes. The large quantities of dissolved and un-dissolved species present during the digestion and concentration process lead to the formation of scale deposits on the filters, pipelines carrying the acid, evaporators, heat exchangers and coolers [1-2]. Scale buildup can cause a number of operational issues such as plugging of equipment, inefficient feed rate to the evaporators, increased utility

costs, lost production due to downtime, and downgraded products from lower concentration due to poor heat transfer[3-4]. A rough calculation of potential production gain from extending the operating cycle of a heat exchanger is illustrated in Table 1 assuming 50 weeks/year operation with a capacity of 300 T P 2O5/day (i.e. 105 000 T/year or 12.5 T/h) with a shutdown frequency of every 2 weeks and a shutdown duration of 20 hours.

TABLE 1. Production Gain by Extending Cycle NEW SHUT DOWN FREQUENCY �FROM 2 WEEKS TO�

PRODUCTION TIME SAVED �HRS/YEAR�

MERCHANT GRADE ACID PRODUCTION INCREASE �T/YEAR�

�%�

4 weeks

250

3125

3.0

5 weeks

300

3750

3.6

6 weeks

333

4167

4.0

Cytec has developed a novel family of antiscalants, PHOSFLOW®, that reduces the fluorosilicate, calcium sulfate and aluminate scale formation in phosphoric acid operations allowing for longer cycle times, increased production, and reduced maintenance time. The technology has been demonstrated via three commercial scale plant trials at

42

two distinct points in the phosphoric acid production process. No negative downstream effects were observed. The antiscalant is easily applied to the process through a piston pump requiring minimal capital investment. This article will summarize the application of this technology to a commercial phosphoric acid heat exchanger and feed line.

Solvent Extraction, Mineral Processing and Alumina Processing

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants

2. Trial of PHOSFLOW® Technology in Phosphoric Acid Plant Heat Exchanger and Feed Acid Line The customer identified the feed acid line to the heat exchanger and the heat exchanger as areas that suffer from scale buildup and requires periodic cleaning to maintain acid flow. The typical cycle length of the heat exchanger was 3 weeks and a clean out of the feed acid line was necessary approximately 10 days into the cycle. The cleanout time for the feed acid line was 4 hours and 3 days for the heat exchanger. Cleanout of the heat exchanger entails mechanical cleaning with a high pressure water lance that takes approximately 15 hours with an average of 150 tubes visually plugged out of a bundle of 868 tubes. During a cycle, the flow rate of acid to the evaporator decays from an initial rate of approximately 50 m 3/hr to less than 39 m 3/hr. A two-trial evaluation was proposed where the first trial of the antiscalant would replicate their normal operating cycle time of 3 weeks. The data generated from the trial would be compared with historical data as well as data from a control cycle run immediately after to ensure similar environmental conditions and ore chemistry were encountered. If the results looked promising, a second trial would be conducted

looking to extend the cycle beyond the typical 3 weeks. The process parameters chosen as indicators of performance for the feed line and heat exchanger are listed below: • Heat transfer coefficient – Gives an indication as to the efficiency of heat transfer. Values greater than 450 W/m2 K are considered sufficient. • Percent valve open reading – the valve opening is modulated to attain a set point flow rate. When the line is scaled or a blockage develops, the valve will need to be fully open (100%) to achieve the set point flow. • Current load on recirculation pump in the heat exchanger. If the tubes in the heat exchanger become plugged, the current load on the recirculation pump will increase. • The number of tubes plugged in the heat exchanger and the time required for cleaning the heat exchanger. • The flow rate of acid through the system.

2.1. Results from Three Week Trial of PHOSFLOW® Antiscalant The antiscalant was dosed into the outlet side of the feed line pump carrying 28% P2O5 acid from a storage tank to a heat exchanger designed to increase the acid concentration to 54% P2O5. The dose was set at 50 ppm (vol/vol) for the

IN PROCESS SEPARATION

duration of the trial and monitored regularly. Figures 2-5 below show a comparison of the process parameters for the antiscalant treated cycle as well as the cycle immediately after designated as a control cycle.

43

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants     )     K     2    m     /     W     (    t    n    e     i    c     i     f     f    e    o     C    r    e     f    s    n    a    r     T    t    a    e     H

650 600 550 500 450 400 350 300 250 200 0

    )     K     2    m     /     W     (    t    n    e     i    c     i     f     f    e    o     C    r    e     f    s    n    a    r     T    t    a    e     H

100

200 300 Cycle Time (hours)

400

650 600 550 500 450 400 350

FIGURE 2: DATA FOR HEAT TRANSFER COEFFICIENT FOR �A� PHOSFLOW® TREATED CYCLE AND �B� CONTROL CYCLE

300 250 200 0

100

200 300 Cycle Time (hours)

400

0

100

200 300 Cycle Time (hours)

400

100     )    n    e    p     O     %     (    n    o     i    t     i    s    o     P    e    v     l    a     V

80 60 40 20

100     )    n    e    p     O     %     (    n    o     i    t     i    s    o     P    e    v     l    a     V

80 60

FIGURE 3: 40

DATA FOR VALVE POSITION FOR �A� PHOSFLOW® TREATED CYCLE AND �B� CONTROL CYCLE

20 0

44

100

200 300 Cycle Time (hours)

400

Solvent Extraction, Mineral Processing and Alumina Processing

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants 54.0 53.5     )    s    p    m     A     (     d    a    o     L    t    n    e    r    r    u     C

53.0 52.5 52.0 51.5 51.0 50.5 50.0 0

100

200 300 Cycle Time (hours)

400

54.0 53.5     )    s    p    m     A     (     d    a    o     L    t    n    e    r    r    u     C

53.0 52.5 52.0

FIGURE 4:

51.5

DATA FOR CURRENT LOAD ON RECIRCULATING PUMP FOR �A� PHOSFLOW® TREATED CYCLE AND �B� CONTROL CYCLE

51.0 50.5 50.0 0

100

200 300 Cycle Time (hours)

400

0

100

200 300 Cycle Time (hours)

400

55 50     )    r     h     /     3    m     (    e    t    a     R    w    o     l     F

45 40 35 30 25 20

55 50     )    r     h     /

45

    3

   m     (    e    t    a     R    w    o     l     F

40 35

FIGURE 5:

30

DATA FOR FEED ACID FLOW RATE FOR �A� PHOSFLOW® TREATED CYCLE AND �B� CONTROL CYCLE

25 20 0

100

IN PROCESS SEPARATION

200 300 Cycle Time (hours)

400

45

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants With the exception of the valve position parameter, all other data points indicated that the antiscalant effe ctively controlled the scale buildup in the feed acid line and heat exchanger, especially compared to the control cycle. It can be observed that the flow rate, heat transfer coefficient and current load were all maintained to a greater degree than in the control cycle. With respect to the feed acid line, it did require a washout as the data indicated a blockage occurred however it should be noted that the slope of the curve before jumping to 100% open and after the washout (where it returned to baseline) was very low in contrast to the control cycle. Additionally, the increase in valve position was not accompanied by a large drop off in flow rate as in

the control cycle indicating a reduced scaling rate. Thus a higher rate of production was realized in the antiscalant treated cycle. When the heat exchanger was opened and inspected, it was observed that only 18 tubes were plugged compared to a historical average of 150. Additionally, the cleanout using a high pressure water lance took approximately one hour compared with a historical average of 15 hours. The data and visual inspection results showed great promise for extending the cycle length of the heat exchanger, thus it was decided to continue the evaluation with a trial aimed at doubling the cycle time.

2.2. Results from Extended 6 Week Trial of PHOSFLOW® Antiscalant As with the 3 week trial, the antiscalant was dosed into the outlet side of the feed line pump carrying 28% P 2O5 acid from a storage tank to a heat exchanger designed to increase the acid concentration to 54% P 2O5. The dose was set at     )     K     2    m     /     W     (    t    n    e     i    c     i     f     f    e    o     C    r    e     f    s    n    a    r     T    t    a    e     H

50 ppm (vol/vol) for the duration of the trial and monitored regularly. Figures 6 and 7 show the data from the process parameters of interest for the antiscalant treated cycle.

650 600 550 500 450 400 350 300 250 200 0

200

400 600 Cycle Time (hours)

800

1000

100     )    n    e    p     O     %     (    n    o     i    t     i    s    o     P    e    v     l    a     V

80 60

FIGURE 6: 40

DATA FOR HEAT TRANSFER COEFFICIENT AND VALVE POSITION FOR EXTENDED EVALUATION OF PHOSFLOW®

20 0

46

200

400 600 Cycle Time (hours)

800

1000

Solvent Extraction, Mineral Processing and Alumina Processing

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants 54.0 53.5     )    s    p    m     A     (     d    a    o     L    t    n    e    r    r    u     C

53.0 52.5 52.0 51.5 51.0 50.5 50.0 0

200

400 600 Cycle Time (hours)

800

1000

55 50     )    r     h     /     3    m     (    e    t    a     R    w    o     l     F

45 40

FIGURE 7:

35

DATA FOR CURRENT LOAD ON RECIRCULATING PUMP AND FEED ACID FLOW RATE FOR EXTENDED EVALUATION OF PHOSFLOW®

30 25 20 0

200

400 600 Cycle Time (hours)

800

The data for the process parameters show a successful extension of the cycle time without as significant degradation in heat transfer coefficient, feed acid flow rate, or increase in pump current load. In the case of valve position, it can be observed that the line was not cleaned out sufficiently from the start and that it had already developed a blockage. Nevertheless, the antiscalant allowed for production level flow rates for a significant period of time before the line was washed. No additional line cleanouts were necessary during the remainder of the trial which lasted an additional 700 hours past the first cleanout. This is significantly more than

1000

their average cycle time from historical data of 486 hours so a mid-cycle wash was effectively eliminated. When the heat exchanger was opened and inspected, it was observed that only 66 tubes were plugged compared to a historical average of 150 tubes for a 3 week cycle. Additionally, the cleanout time using a high pressure water lance took approximately 2.5 hours compared with a historical average of 15. The operators observed that the scale was very soft and easy to remove from the heat exchanger tube walls.

3. Conclusion The plant trial results demonstrate that Cytec Industries PHOSFLOW® antiscalant technology provided a significant benefit in terms of controlling scale buildup within process pipelines and heat exchangers of a phosphoric acid plant.

IN PROCESS SEPARATION

Figures 8 and 9 summarize the results of the three trials and new process option available by applying the antiscalant. The trials were considered a success and are leading to further evaluations of the technology at this customer’s plant and others to help establish the robustness of the technology.

47

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants

Total Cycle Time (weeks) Regular Process

Hours Required to Clean Heat Exchanger

3.5

Regular Process

with PHOSFLOW

6.8

with PHOSFLOW

15.0 2.0

+94%

-87%

# of Heat Exchange Tubes Plugged Regular Process 3+ weeks with PHOSFLOW 3+ weeks

150 18

-88% 66

with PHOSFLOW 6+ weeks

-56%

FIGURE 8: COMPILATION OF PLANT TRIAL RESULTS FOR APPLICATION OF PHOSFLOW® ANTISCALANT TECHNOLOGY

Feed line   wash (4 hours) Baseline Process Process with PHOSFLOW

3 weeks

3 day boil-out

Feed line wash (4 hours) 3 weeks

Only one feed line wash / no 3 day boil-out 6+ weeks

Feed line wash (4 hours) FIGURE 9: COMPARISON OF CUSTOMER’S HISTORICAL PROCESS WITH ONE USING PHOSFLOW® ANTISCALANT

48

Solvent Extraction, Mineral Processing and Alumina Processing

Scale Controlling Chemical Additives for Phosphoric Acid Production Plants References [1]

Theys, T, 2003. Influence of the Rock Impurities on the Phosphoric Acid Process, Products and Some Downstream Uses, presented at the IFA Meeting of the Technical Committee , Abu Dhabi.

[3]

Behbahani R.M., Müller-Steinhagen, H., and Jamialahmadi M., 2006. Investigation of Scale Formation in Heat Exchangers of Phosphoric Acid Evaporator Plants, The Canadian Journal of Chemical Engineering, Vol 84, p. 189.

[2]

Behbahani R.M., Müller-Steinhagen, H., and Jamialahmadi M., 2003. Heat Exchanger Fouling in Phosphoric Acid Evaporators – Evaluation of Field Data, Heat Exchanger Fouling and Cleaning: Fundamentals and Application, Vol. 9, p. 60.

[4]

Jamialahmadi, M., Müller-Steinhagen, H., 2007. Heat Exchanger Fouling and Cleaning in the Dihydrate Process for the Production of Phosphoric Acid, Chemical Engineering Research and Design, Vol 85, p. 245.

For more information on this subject and other Cytec technologies, please visit our website at www.cytec.com. TRADEMARK NOTICE: The ® indicates a Registered Trademark in the United States and the ™ indicates a trademark in the United States. The mark may also be registered, subject of an application for registration, or a trademark in other countries.

IN PROCESS SEPARATION

49

Contact Us

Bradford, UK

Welland, Canada: Collectors, Extractants Longview, WA: Emulsions

Botlek, Netherlands: Dewatering Aids Werndorf, Austria: CGM

Mt Pleasant, TN: Extractants

Stamford, CT Willow Island, WV: Collectors, Dewatering Aids Atlanta, GA: CGM, Collectors Mobile, AL: Emulsions

Phoenix, AZ Houston, TX: Collectors, Frothers Atequiza, Mexico: Collectors, Anti-Scale

Rayong, Thailand: Emulsions, Flocculants, Blending

Antofogasta, Chile: Collectors, Blending & Warehousing

Australia: Collectors, Blending

São Paulo, Brazil Australia: Emulsions

Santiago, Chile

Manufacturing Sites Research Centers

Corporate Headquarters Cytec Industries Inc. Five Garret Mountain Plaza Woodland Park, NJ 07424 USA Tel: +1 973 357-3193 US Toll Free: +1 800 652-6013 Fax: +1 973 357-3117

North American Regional Office Cytec Industries Inc. 2085 E. Technology Circle Suite 102 Tempe, AZ 85284 USA Tel: +1 480 730-2000 Fax: +1 480 730-2245

50

South American Regional Offices

Europe, Africa and the Middle East Regional Office

Los Orfebres 305 La Reina, Santiago, Chile Tel: +56 2-560-7901 Tel: +56 2-560-7900 Fax: +56 2-273-6423

Cytec Industries BV38, Franko Str., Of. 4 Kiev, 01030 Ukraine Tel: +380-44-2302093 Fax: +380-44-2386076

Cytec Comércio de Materiais Compostos e Produtos Químicos do Brasil Ltda. Av. Dr. Cardoso de Melo, 1460-6 andar CJ 61-62 Ed Pillar CEP 04548-005 Vila Olimpia São Paulo SP, Brazil Tel: 55-11-5033-2828

Asia Pacific Regional Office Australia Regional Office Cytec Australia Holdings Pty Ltd. Suite 1, Level 1, Norwest Quay 21 Solent Circuit Baulkham Hills, NSW 2153 AUSTRALIA Tel: +61 2-9846-6200 Fax: + 61 2-9659-9776

China Regional Office 30A Junyao International Plaza 789 Zhao Jia Bang Road, Xuhui District Shanghai 200032 Tel: +86-21-6422-8920 Fax: +86-21-6422-8980

Solvent Extraction, Mineral Processing and Alumina Processing

Delivering Innovative Technologies for the Mining Industry Since 1907 1951 1907 Calcium Cyanamide American Cyanamid

1905

1910

1925

1937

AERO® brand Cyanide in leadzinc flotation

Mercaptobenzothiozole chemistry (AERO 400 series) as flotation collector

1915

1920

1925

1930

Sulfonates for non sulfide flotation

Synthetic alcohol and polyglycol frothers

1935

1940

1915

1930

AERO brand Cyanide in gold and silver cyanidization

Dithiophosphates AEROFLOAT ® for sulfide flotation

1945

1975

1983

1965

Hydroxamated PAMs for Bayer Process

1981

1989 SIMS pioneered for plant troubleshooting

1975

1980

1970

1978

New collector formulations for non-sulfide flotation

Synthetic polymers for Bayer liquor washers

1985

2013 CYFLOC TF-9000 Alumina hydrate flocculant

Polymeric depressants for Mg silicate minerals

AEROFLOAT® MX 500

1990

1995

2000

2005

2010

1999

2001

Alkyl hydroxamate collectors for oxides and clay

2010

ACORGA NR extractant (nitration resistant)

New pyrite depressants

1987

IN PROCESS SEPARATION

Chemical enhancement of magnetic separation (MagSep) technology

1994

Ester modified aldoxime

First generation neutral circuit promoters

2007 Alumina hydrate flocculant CYFLOC® TF-8000

ACORGA® OPT products

CYTEC is spun off from American Cyanamid

Selective gold collectors introduced (MTP)

Polymeric depressants for flotation

Synthetic polymers for flocculation

New chemistries AERO ® XD-5002 & MAXGOLD series for sulfide flotation

1988

1980

1960

2008

MAX HT® for scale control in alumina refineries

1992

CYANEX 301 extractant

CYANEX 272 extractant

AEROPHINE® flotation products

1970

2004

1984

CYANEX® 923 extractant

1965

FLOTATION MATRIX™ 100 for improving metals recovery

1955

1952

1948

2005

Nonyl salicylaldoxine (Aldoxime) Cu extractant

1950

Heavy media process for coal and other minerals

1976 Synthetic red mud flocculants for Bayer Process

1955

Cadmium removal agents for phosphoric acid production

CYANEX 600 (molybdenum extractant) AERO 7260 HFP mineral depressant

2015

2020

2012 PHOSFLOW® for scale control in phosphoric acid plants Chemical enhancement of electrostatic separation (AERO EZ reagents) ACORGA Crud Buster 1000 reagent CYANEX 572 solvent extraction reagent for Rare Earths

2011 Alumina CYFLOC HX Ultra flocculants XR Series - Xanthuate replacement chemistries ACORGA OR extractant (oxidative resistant)

51