evaluation of net smelter returns in the south african pge ... · there are currently several...

The Southern African Institute of Mining and Metallurgy

Platinum 2012

577

D.A. Cowen, V. Agnello, P.J. Petit

Evaluation of net smelter returns in the South African PGE industry by

application of base metal concentrate commercial treatment terms

D.A. Cowen Snowden Mining Industry Consultants

V. Agnello Snowden Mining Industry Consultants

P.J.Petit Snowden Mining Industry Consultants

Abstract

Toll and custom treatment of base metal concentrates have been performed for decades as

mines and concentrators are developed distant from capital and, energy-intensive smelting and

refining facilities. Commercial treatment terms have been developed and refined for over a

century, and are negotiated taking into account a complex array of quality and economic

factors.

Treatment of third-party concentrates in the South African platinum industry is a comparatively

recent development, brought about by the opening up of mineral rights and operations being

developed that are smaller than the critical size required to sustain capitalization of smelting or

refining plant. Published commercial terms have not been demonstrated to reflect the

complexities and subtleties of base metal concentrate markets.

This paper outlines some key elements and parameter ranges of base metal concentrate

treatment contracts, and applies similar principles to typical Merensky and UG2 platinum group

element (PGE) concentrates to determine potential value distribution between the mining and

treatment parties in tolling arrangements.

Introduction

Toll and custom treatment of base metal concentrates has been performed for decades as

mines and concentrators (miners) are developed distant from capital, energy-, and labour-

intensive smelting and refining facilities (processors). Commercial terms for the sale of copper,

nickel, zinc, and lead have been developed and refined for over a century and are negotiated

taking into account a complex array of quality and economic factors.

The ultimate objective of treatment and refining contracts is to endeavour to allocate the value

of the final saleable product and the costs of its production equitably. In order to provide

incentive for both parties (miners and processors), contractual arrangements must address the

risk-return balance of each business.


Platinum 2012

578

Miners are intrinsically risk takers, facing unique uncertainties in resource definition and

commodity prices, whilst processors are more akin to industrial operations with controllable

raw materials and cost structures. High margins in commodity boom periods provide reward for

mining risk, while the reward for processing operations, is stable returns over the life of the

capital asset.

Factors within the miners’ control comprise delivery of concentrate of a quality acceptable to

the processing facility. The processor’s control area includes the cost and efficiency of recovery.

Key aspects of the concentrate sales contract should therefore embody appropriate allocation

of operational efficiency benefits, processing costs, and metal price risk.

The South African platinum industry

Treatment of third-party concentrates in the South African platinum industry is a comparatively

recent development, brought about by the opening of up mineral rights and operations being

developed that are smaller than the critical size required to sustain capitalization of smelting or

refining plant. Published commercial terms have not been demonstrated to reflect the

complexities and subtleties of base metal concentrate markets1.

There are currently several independent concentrate-producing platinum group element (PGE)

miners in South Africa, and four major integrated miner/processors1. A survey of PGE projects

in South Africa indicates that the number of stand-alone miners could increase significantly

over the next decade.

Platinum-bearing reef is found in three primary geographic localities in South Africa, referred to

as the western limb (WL), eastern limb (EL), and northern limb (NL) of the Bushveld Complex.

The WL is centred on Rustenburg, the capital of PGE mining in South Africa, and consequently

the centre of processing operations. Development of mining operations on the EL and NL is

limited principally to concentrate production.

The characteristics of ores and their concentrates are markedly different between the three

primary economic PGE reefs, namely the Merensky Reef (MR) and the UG2 Reef on the WL and

EL and the Platreef (PR) on the NL. The PGE distributions differ not only between reef types, but

between areas for each reef type, as illustrated in Error! Reference source not found..


Platinum 2012

579

Figure 1-PGE percentage distribution based on reef type and locality (modified1,2)

Merensky concentrates have a higher Pt/Pd ratio and contain significant amounts of nickel and

copper, while UG2 concentrates contain a larger percentage of lesser PGMs, such as rhodium,

and are lower in base metal values. Platreef concentrates may be similar to Merensky

concentrates in terms of base metal sulphide hosting of PGEs, but have different PGE

distributions. The EL concentrates tend to have more palladium and base metals than WL

concentrates1. The range of assay values for the various types of concentrate is large, although

certain thresholds and trends are noted.

Generalized characteristics in resource statements of company annual reports are shown in

Table I.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Merensky WL Merensky EL UG2 WL UG2 EL Platreef NL

PGE

Distribution (%)

Pt% Pd% Rh% Au%


Platinum 2012

580

Table I-General resource characteristics1,2

Unit Western Limb Eastern Limb Northern Limb

Merensky Reef

Mineralogy Ni/Cu Sulphide Ni/Cu Sulphide N/A

Resource PGE Grade g/t 4E >5g/t >4g/t N/A

Pt content percent >60% >50% N/A

Resource Ni Grade percent >0.2% >0.2% N/A

UG2 Reef

Mineralogy Chromite Chromite N/A

Resource PGE Grade g/t 4E >4g/t >4.5g/t N/A

Resource Ni Grade percent 0.15%

Furthermore, the average recovery of metal to concentrate in flotation is variable across the

Limbs and across the reefs, with MR recoveries broadly being in the order of 5 per cent higher

than UG2 (for the same target grade) and Platreef concentrates being a further 5 per cent lower

(Table II).


Platinum 2012

581

Table II-Typical ex-mine PGM concentrates1

Assay Unit Merensky UG2 Platreef

PGM g/t 4E 200.0 200.0 120.0

Pt % of 4E 63.5 56.7 45.1

Pd % of 4E 28.1 29.4 45.7

Rh % of 4E 4.4 13.0 3.2

Au % of 4E 4.0 0.9 6.0

Ni % 6.0 1.4 4.9

Cu % 3.4 0.7 2.5

Cr2O3 % 0.6 3.0 0.3

Sulphur % 15–20 4–6 10–15

These myriad factors results in a complex set of variables that determine the value of a

concentrate.

PGE concentrate processing operations are capital intensive with a high level of fixed costs,

power and labour being the principal cost drivers. Whilst technology development may change

this structurally in the medium to long term, the current route for stand-alone miners is to

engage the major processors for sale or toll treatment of their concentrates.

Commercial terms6,5,6,8

Typical pricing clauses in a commercial concentrate sales contract include components of

distribution of metal value, through deductions and price participation, and allocation of

smelting and refining costs, through application of treatment and refining charges1. Penalty

charges are applied to account for elements in the concentrate that affect process recoveries or

costs.

For a copper concentrate this may take the form of:

Deductions7 – The amount deducted from concentrate grades to calculate the metal paid for

by a smelter, e.g. copper concentrates typically deduct 1.0-1.5 units (grade percentage points).

The unit deduction is related to the grade of concentrate, with higher-grade concentrates

attracting a higher unit deduction.

Thus a concentrate grading 28 per cent copper would be paid as grading 26.9 per cent copper

(effective 96 per cent payability). This is subject to a minimum deduction, expressed as a

percentage of contained metal. Thus a minimum deduction of 5 per cent term for a 28 per cent


Platinum 2012

582

copper concentrate would result in a maximum payability of 95 per cent of the contained metal

value to the miner.

Metal recovery in PGE smelting is typically reported to be between 92-95 per cent. and in

refining over 99 per cent. A deduction of 10 per cent (or 90 per cent payability) of contained

metal value would result in the processor receiving between 1-6 per cent of the contained

metal value, depending on smelting and refining operation recoveries.

Treatment charge (TC) – The charge paid by the miner to have its concentrate treated through

smelting to produce saleable metal. This is typically quoted in US$ per ton of concentrate, and

is theoretically a function of the cost of smelting. In practice, it fluctuates as a function of the

supply/demand balance in the concentrate market.

Refining Charge (RC) – The price paid by the miner to the processor for refining the contained

metals in their concentrate to produce a saleable metal. The refining charge is based on the

payable metal content, and is theoretically a function of the cost of refining. Similar to TCs, it

fluctuates as a function of the balance between supply of smelter matte and refinery capacity.

An analysis of 2011/2012 publicly-listed PGE companies’ annual financial statements and other

open-source documentation indicates that the cost of South African PGE smelting varies

between US$200-US$250 per ton concentrate, and refining US$250-300 per ton concentrate1.

Price participation – The price participation mechanism may increase TC/RC as the metal price

increases above an agreed base price, and occasionally reduces the TC/RC as the price

decreases below the agreed base price. This allows smelters to participate in rises/falls in the

metal prices, and is usually combined with ’floors’ and ’caps’ to limit the processor’s loss or

gains.

The authors have no evidence of price participation being applied in PGE smelting contracts in

South Africa. The number of economic metals in a PGE concentrate ’basket’ would require

extensive price participation negotiations, with potential benefits being cancelled by

uncorrelated or inversely-correlated metal price movements of the basket metals.

Penalty charges (penalty) – The penalty charge is specific to particular elements that result in

additional process activities being necessary to remove them from the final metal product, or

that need to be disposed of in an environmentally friendly manner. A penalty is usually specific

for a particular deleterious element, and is related to the cost of its removal and/or disposal, or

the extent to which it reduces the value of final metal product.

The key penalty component for platinum concentrates is chromite, which has negative effects

on both the smelting and refining processes9,10,12.


Platinum 2012

583

The authors have not found any publicly reported ranges for chromite penalties in the South

African PGE industry. As a proxy for chromite penalties, this paper assumes that the rate at

which chromite penalties are applied will be equal to the TC of treating a chromite-containing

concentrate blended with sufficient mass of zero-grade concentrate to reduce the grade of

overall chromite values to below penalty levels.

It has been reported that concentrates with chromite content of above 1.0 per cent result in

metallurgical difficulties in the smelting operation9,10,11. Thus this analysis assumes that a

chromite penalty equal to the base TC will be applied to a concentrate grading 2 per cent

chromite; and a penalty equal to three times the base TC will be applied to a concentrate

grading 3 per cent chromite, etc.

South African PGE concentrate evaluation3,13

A custom MS Excel model has been developed to evaluate the relative effects of changing

concentrate parameters on the payment a miner could expect to receive for typical PGE

concentrate, thereby simulating the range of expected returns to miners' net of smelting and

refining charges.

The ’base case’ models a typical Merensky reef concentrate of quality indicated in Error!

Reference source not found.. The parameters applied are indicated in Error! Reference source

not found..

Table III-Snowden base case for PGE content in reef types1

PGE Unit Merensky UG2 Platreef

4E grade g/t 200 200 120

Pt split per cent 60.0% 55.0% 45.0%

Pd split per cent 30.0% 30.0% 45.0%

Rh split per cent 5.0% 10.0% 3.0%

Au split per cent 5.0% 5.0% 7.0%

Ni grade per cent 6.0% 1.0% 5.0%

Cu grade per cent 3.5% 0.5% 2.5%

Cr grade per cent 1.0% 3.0% 0.0%


Platinum 2012

584

Table IV-Metal prices and process charges for PGEs,1

Metal Prices Cost unit Metal Price

Platinum USD per ounce 1,700

Palladium USD per ounce 600

Rhodium USD per ounce 2,500

Gold USD per ounce 1,800

Nickel US cents per pound 700

Copper US cents per pound 200

Payability

PGE payability per cent 90%

Ni payability per cent 85%

Cu payability per cent 85%

Treatment charge

TC (smelter) USD per ton 250

Refining charge6

Pt RC USD per ounce Pt 25

Pd RC USD perounce Pd 25

Rh RC USD per ounce Rh 75

Au RC USD per ounce Au 10

Ni RC US cents per pound Ni 80

Cu RC US cents per pound Cu 20

Penalty basis

Max Cr per cent 1.50%


Platinum 2012

585

The effects of different concentrate 4E grades, metal prices, PGE payability, TC, RC, and

chromite content are illustrated in Error! Reference source not found.–Error! Reference source

not found..

The charts present the gross unit metal value (GMV) of the concentrate per ton of concentrate,

based on the basket prices indicated, the net metal value (NMV) attributable to the miner on

the same basis (equal to GMV less deductions), TC/RC and penalties. In addition, the NMV:GMV

ratio is reflected as a percentage payment to the miner (PMT).

Figure 2-Effect of PGE concentrate grade on smelter return

The benefit to the miner producing a high-PGE concentrate grade is demonstrated, with the

unit value not only increasing threefold per ton of concentrate, but payability increasing over 7

percentage points over the range 50 g/t to 250 g/t 4E; the principal driver of the latter being

the treatment charge.

76%

81%

83%84%

85%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

50 100 150 200 250

GMV/NMV

(USD/t conc)

Mine payment

%

PGE Conc Grade (g/t 4E)

GMV NMV PMT


Platinum 2012

586

Figure 3-Effect of metal price sensitivity on smelter return

The effect of metal price fluctuations is typified by a percentage point variation in payment to

the miner, demonstrating its higher exposure to metal price risk/reward.

Figure 4-Effect of PGE payability on smelter return

83%84% 84%

85% 85%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

-20% -10% 0% 10% 20%

GMV/NMV

(USD/t conc)

Mine payment

%

Sensitivity Factor (%)

GMV NMV PMT

94%

89%

84%

79%

74%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

100% 95% 90% 85% 80%

GMV/NMV

(USD/t conc)

Mine payment

%

4E Payment (%)

GMV NMV PMT


Platinum 2012

587

For constant concentrate grades and metal prices, return to the miner will be directly

proportional to PGE payability. Smelter return, however, will be influenced by smelter

efficiency. The relationship between metal grade and payability has been excluded in this

analysis, but may be incorporated by linking payability directly to grade. This would encourage

miners to optimize concentrate grades, which will enable the smelter to operate more

efficiently.

Figure 5-Effect of treatment charge on smelter return

Payment variability over a wide range of treatment charges, between US$100-400 per ton of

concentrate, is typified only by a single percentage point either way. This is indicative of the

relative small quantum of smelter processing cost against the value of metal production, and is

a pointer to the value of efficient smelting.

86% 85% 85% 84% 84% 83% 83%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

100 150 200 250 300 350 400

GMV/NMV

(USD/t conc)Mine payment

%

Treatment Charge (USD/t conc)

GMV NMV PMT


Platinum 2012

588

Figure 6-Effect of refining charges on smelter return

Similar to treatment charges, the mine return is relatively insensitive to mining charge, when

compared to recovery/revenue factors.

Figure 7-Effect of chrome penalty on smelter return

85% 85% 84% 84% 83%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

-40% -20% 0% 20% 40%

GMV/NMV

(USD/t conc)

Mine payment

%

Sensitivity Factor (%)

GMV NMV PMT

84% 84%83%

82%80%

-

2,000

4,000

6,000

8,000

10,000

12,000

70%

75%

80%

85%

90%

95%

100%

0.0% 1.0% 2.0% 3.0% 4.0%

GMV/NMV

(USD/t conc)

Mine payment

%

Concentrate Chrome Content (%)

GMV NMV PMT


Platinum 2012

589

As modelled, the effect of chrome content is illustrated by an approximate 2 per cent reduction

in mine payment per percentage point above 1.5 per cent. Discussion with industry members

revealed that this is a typical structure in place in some contractsError! Reference source not

found.. It is notable that the return is asymmetric, with ’clean’ concentrate not attracting a

premium.

Error! Reference source not found. illustrates the effect of applying the standard parameters to

typical Merensky, UG2 and Platreef concentrates.

Figure 8-Effect of reef values on smelter return

The relative GMV value of concentrates differs as a result of metal content differences and

lower grade of PR concentrates. For the same grade of concentrate, absolute payment

percentages to miners for MR and UG2 are similar.

Error! Reference source not found. illustrates the different process charges for the three ore

types.

84% 84%

81%

70%

75%

80%

85%

90%

95%

100%

-

2,000

4,000

6,000

8,000

10,000

12,000

MR UG2 PR

Mine

Payment (%)GMV/NMV

GMV NMV PMT PCT


Platinum 2012

590

Figure 9-Commercial terms and charges for the three ore types

Chromite penalties for UG2 are largely offset by lower refining charges as a consequence of

lower base metal tenors in the UG2.

Conclusions

The strategy for maximizing return for a miner is to produce high-grade concentrate with low

penalties. Concentrate grade bears an inverse relationship to concentrate tonnage, and

concentrate quality (including chromite content) is controllable through concentrator

design/operating practice, operating, and capital cost. Consequently, there exists opportunities

for a miner to tailor its beneficiation operations to optimize returns under a given set of

concentrate sales terms. These possibilities should be considered at the pre-feasibility stage.

From a processor perspective, similar goals are beneficial, with high-grade clean concentrates

maximizing the smelter’s economic utilization. Further consideration could be given to the

economic value of other elements in a concentrate mix. For example, Platreef concentrate,

under the base terms, provides low payability to the miner, as a consequence of its low grade

and low payment when compared to Merensky and UG2 concentrates.

The benefit of Platreef as a source of sulphide concentrate is of important technical value, as a

diluent of chromite-bearing material has not been considered in the analysis. Smelting power

requirements are lower for Merensky and Platreef sulphide ores. There should be recognition

of this benefit in concentrate agreements.

PMT 1,072 PMT 962

PMT 609

TC 250 TC 250

TC 250

Cr R PEN 83

RC 293 RC 208

RC 197

-

200

400

600

800

1,000

1,200

1,400

1,600

1,800

MR UG2 PR

PMT

Components

(USD/t conc)

PMT TC Cr R PEN RC


Platinum 2012

591

From an Integrated producer perspective, separation and distinction of the above

considerations is important to ensure coherence in mining strategy and processing strategies,

particularly as the mix of ore availability, in asset portfolios change.

References

1. Anglo American Platinum, Impala Platinum, Lonmin Platinum, and Northam Platinum.

Annual Reports, 2008–2011. [Accessed May/June] 2012.

2. Lomberg, K. Mineral Resource and Reserve aspects of the Junior Miners on the Bushveld

Complex. http://www.samcode.co.za/downloads/MiningInAfrica/LOMBERG_2008.pdf

[Accessed 1 June 2012].

3. Snowden Mining Industry Consultants. Analysis of South African Smelter Costs and Metal

Prices. 2008–2012.

4. Cramer, L. What is your PGM concentrate worth? Journal of the Southern African Institute

of Mining and Metallurgy, Sep. 2008. pp. 387-394.

5. Minedesignwiki. Net smelter return.

https://www.minedesignwiki.org/index.php/Net_smelter_return [Accessed 8 March

2012].

6. Hay, M.P. and Roy, R. A case study of optimising UG2 flotation performance. Part 1:

Bench, pilot and plant scale factors which influence Cr2O3 entrainment in UG2 flotation.

Minerals Engineering, vol. 23, no. 11–13, 2010. pp. 855-867.

7. Söderström, U. Copper markets day. http://investors.boliden.com/afw/files/press/

boliden/Kokkola-2008-6_Smelters_Copper_US.pdf /November 2008. [Accessed 1 June

2012].

8. Goldie, R. Net smelter return models and their use in the exploration, evaluation and

exploitation of polymetallic deposits. Geoscience Canada, vol. 18, no. 4, Dec. 1991.

pp. 159-168.

9. Hundermark, R. et al. The smelting operations of Anglo American's platinum business: an

update. Southern African Pyrometallurgy 2011 International Conference, Cradle of

Humankind, South Africa, 6-9 March 2011. Southern African Institute of Mining and

Metallurgy, Johannesburg, 2011. pp 295-308.

10. Coetzee, V. Common-sense improvements to electric smelting at Impala Platinum.

Southern African Pyrometallurgy 2006 International Conference, Cradle of Humankind,

South Africa, 5-8 March 2006. Southern African Institute of Mining and Metallurgy,

Johannesburg, 2006. pp. 43-62.


Platinum 2012

592

11. Eksteen, J. Cracking a hard nut: An overview of Lonmin's operations directed at smelting

of UG2-rich concentrate blends. Journal of the Southern African Institute of Mining and

Metallurgy, vol. 111, Oct. 2011. pp. 681- 690.

12. Georgalli, G.A. and Anderson, D.K. Optimizing concentrate allocation to Anglo Platinum

smelters. The 4th International Platinum Conference, Platinum in transition ‘Boom or

Bust’. Southern African Institute of Mining and Metallurgy, Johannesburg, 2010.

pp. 91-96.

13. Barr, G. et al. On-site processing vs. sale of copper concentrates. ALTA 2005, Perth,

Australia, June 2005. pp. 1-16.

The Author

Dennis Cowen, Divisional Manager – Corporate, Snowden

Dennis has 28 years of experience in the mining industry. He has been consulting for the the

last 13 years in techno-economic valuations, cash flow and financial modeling, risk modeling

and alternative valuation methods for resource related projects and companies, in platinum,

coal, gold and base metals fields. Clients range from large corporations including Anglo

American, Anglo Platinum, BHP Billiton, Konkola Copper Mines, Ivanhoe Nickel and Platinum

together with numerous junior mining companies, financial institutions and consultancies. Prior

to entering consultancy Dennis was at Gold Fields for 4 years, heading Coal and Base Metals in

the Mineral Economics Division and managing Gold Fields Coal New Business and Acquisitions.

Dennis practiced as an engineer for 10 years, running his own Process Engineering consultancy,

heading Davy South Africa process engineering division, performing project management and

engineering for Scientific Plant and Design and performing metallurgical research and

development for Anglo American after being trained in plant operations at Vaal Reefs.

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown

/Description >>> setdistillerparams> setpagedevice

evaluation of net smelter returns in the south african pge ... · there are currently several...

Documents