Production of rough diamonds

Summary: The production of rough diamonds already reached a peak value more than 10 years ago. Predictions see a continuation of the market downturn for mined diamonds up to 2030. Nevertheless, the medium- and long-term market outlook is good, especially in view of the rising prices and improved technologies. This article provides an insight into current market developments and explains important trends.

1 Introduction

In 2016, the diamond jewelry sales market was at a level of US$ 80 billion while the market for industrial diamonds was US$ 25 billion. The volume relations, on the other hand, are significantly different. While only around 23 million carats (or 4.6 t) of cut and polished diamonds went into the jewelry market, some 16 billion carats or 3200 t of diamonds were produced for the industrial market. Only 12.5 t of the mined rough diamonds, i.e. less than 0.01 %, go into the production of industrial diamonds, 99.9 % of which are produced synthetically. In the meantime, the synthetic processes have achieved such quality that these diamonds can only be distinguished with difficulty from natural diamonds and are increasingly being used for the jewelry market.

Synthetic diamonds are now manufactured with the same crystal structure, chemical composition and optical and phy­sical properties as natural diamonds. However, the achievable size is still limited. Natural diamonds were formed in the earth’s mantle under very high pressures and temperatures at a depth of about 150 km and at temperatures of 1200 to 1400 °C. Volcanic activity transported the rough diamonds to the surface of the earth through very deep, mostly vertical vents (pipes). The diamond-bearing rock of such pipes is called kimberlite, named after the first such find in ­Kimberley, South Africa, in 1866. Kimberlite pipes are the primary deposits for rough diamonds. Erosion of the pipes produced secondary deposits (alluvial deposits) along rivers and coastal strips.

Rough diamonds have been found on all five continents. Approximately 500 such deposits have been identified. Fig. 1 shows the most important diamond mines currently in operation. The largest deposits are located in Australia, Africa (especially Angola, Botswana, DRC, Ghana, Namibia, Sierra Leone, South Africa and Zimbabwe), Canada and Russia. According to industry experts, the production of rough diamonds reached its peak value of 177 million carats (mcts) in 2005. Fig. 2 shows the development of rough diamond production in recent years. It can be seen that the production has decreased from a level of more than 170 mcts in 2006 to 125-130 mcts in the last 5 years. As a consequence of the ongoing closure of numerous mines and the small number of new finds on a worldwide scale, the future production of rough diamonds is bound to decrease in the long term.

The 2014 edition of “The Global Diamond Report”, published annually by Bain & Company, included an analysis of the rough diamond volumes at different production stages (Fig. 3). This shows that out of the mined production quantity of 130 mcts in 2013, 55-65 mcts went into the production of industrial diamonds (42-50 %) and 65-75 mcts into the production of jewelry diamonds (50-58 %). Due to the cutting and polishing of the jewelry diamonds, most of the mass volume became waste, and only 20-25 mcts (15-19 %) of the produced quantity were actually transformed into jewelry diamonds. Even more drastic are the relationships between prices, i.e. the value added chain (Fig. 4). In 2016, the global value of rough diamonds amounted to US$ 14.9 billion. Over the various value-adding stages up to the jewelry market the value rose to US$ 80 billion, representing an increase of 437 %.

2 TOP Producers of rough diamonds
and their projects

The production of rough diamonds is dominated by a small number of companies. The TOP 5 companies have a market share of about 70 %. The three market leaders in the sector are Alrosa, De Beers and Rio Tinto (Fig. 5). Alrosa, with its most recent production of 37.4 mcts has a market share of 29.4 %, followed by De Beers with 21.5 % and Rio Tinto with 14.2 %. The entry barriers to the market are relatively high. At least 5 to 7 years go by between the exploration of a diamond mine and its commissioning. However, in view of the very attractive average profit margins, at 17-25 %, new companies regularly enter the diamond mining business. The mining companies also enjoy a high bargaining power as regards the prices of their rough diamonds.

Alrosa accounts for more than 98 % of production in Russia and thus has practically a monopoly position in the country [1]. In addition, the company has a 32.8 % stake in the Catoca diamond mine in Angola. In Russia, Alrosa owns 1030 mcts of diamond reserves in the Yakutia Republic and the Arkhangelsk region and operates 8 open cast mines (Fig. 6), 4 underground mines and 15 alluvial mining areas. In the Aikhal Division, ­Alrosa produced a total of 12.2 mcts from three kimberlite mines in 2016, corresponding to 33 % of the company’s production quantity. The mines have been explored down to a depth of 1450 m. Alrosa plans to expand its production capacity to 40.4 mcts by 2021 by means of several so-called brownfield projects. The company‘s average production cost is 36 US$/carat, with no significant differences in cost per ct between the different mining methods, which is unusual.

De Beers, a company of the AngloAmerican Mining Group, produces rough diamonds in South Africa, Botswana, Canada and Namibia, in some cases in the form of joint ventures. In Botswana, the company has a 50 % stake in Debswana Diamond. In 2016, a total of 27.3 mcts with a value of US$ 5.6 billion were produced by the joint ventures, giving De Beers first place in the global sales ranking. Last year, the Gahcho Kué mine (Fig. 7) was commissioned in the cold north-west of Canada. This is the largest new mine to be put into operation in the last 13 years. Projections forecast that this mine will produce about 5 mcts annually over the next 12 years. Other new mine projects are the Cut-8 open cast mine, which will extend the operating lifetime of the Jwaneng mine in Botswana until 2035, the Venetia underground mine in South Africa, and offshore mining off the coast of Namibia, aided by the exploration and sampling ship MV SS Nujoma (Fig. 8).

Rio Tinto operates 2 diamond mines, Argyle in Western Australia and Diavik in Canada. In 2016, the company produced 17.95 mcts. The Argyle Mine alone, which has been in ­operation since 1983 and has meanwhile produced more than 800 mcts, accounted for 14 mcts. This mine’s peak production quantity was 42.8 mcts in 1994. In 2013, Argyle was converted to underground operation, with more than US$ 2.5 billion invested in the project. However, the mine’s future is uncertain and closure is expected by 2021 at the latest. Rio Tinto owns 60 % of the Diavik Mine (Fig. 9) in north-west Canada, with the other 40 % owned by the Dominion Diamond Corporation. Diavik has been in operation since 2003, with 3 kimberlite pipes producing diamonds so far. The 4th Pipe A21 is to follow in 2018. In 2016, the mine’s production quantity was 6.7 mcts, recovered from 2.2 million t of ore. The yield has meanwhile fallen to 3 cts/t of ore. By the end of the mine’s lifetime in 2023, more than 40 mcts will have been produced there.

The Dominion Diamond Corporation (DDC) produced 7.9 mcts in 2016. Approximately 5.2 mcts of this came from the Ekati Mine (Fig. 10) and the rest from the Diavik Mine. Ekati is Canada’s first combined open cast and underground mine. Up to January 2017, a total of 67.8 mcts were produced from its 6 kimberlite pipes. Three pipes are still being mined (Fig. 11). In the financial year 2023 (calendar year 2022), the Jay pipe is to be put into operation, extending the lifetime of the Ekati mine until 2033/34. Towards the end of the mine’s life cycle, the company is expecting to produce more than 9 mcts annually here. The total production quantity from 2018 onwards should be 110 mcts. DDC is aiming to reduce Ekati’s operating costs from about 170 C$/t of ore in 2017 to under 50 C$/t.

In 2016, Sociedada Mineira de Catoca (Catoca) produced a quantity of 7.2 mcts from the open cast mine in the Lunda Sul province of Angola. Catoca is a joint venture of the state diamond company ENDIAMA (32.8 % stake), Alrosa (32.8 %), China Sonangal (18 %) and the Brazilian Odebrecht Mining Services (16.4 %). The Catoca mine went into operation in 1997. In 2015, it produced 6.7 mcts, for which 9.8 million t of ore containing 0.69 cts/t were extracted. The kimberlite mine is regarded as one of the largest in the world and the mining depth is expected to reach 600 m. The expected mining life should assure operations until 2034. A further project, the Luaxe kimberlite mine, promises a potential of 350 mcts over a lifetime of 30 years, but requires investments of US$ 1.5 billion.

In addition to the TOP 5 mining companies, some 20 other companies around the world are involved in the production of rough diamonds. Larger companies include Petra Diamonds, Lucara Diamond and Gem Diamonds. In addition, there are a number of so-called junior miners such as Stornoway, Asa Resource Group, Firestone Diamonds, Kimberley Diamonds, KKG Group, Koidu Ltd., Rockwell Diamonds and Shore Gold, as well as various Chinese companies such as the Anhui Foreign Economic Construction Group. Some companies, like BHP Billiton and Glencore/Xtrata have meanwhile withdrawn from the rough diamond business sector.

In 2016, Petra Diamonds produced 3.7 mcts from its 5 mining operations (four in South Africa, one in Tanzania). By 2019, the company aims to increase its production quantity to 5.3 mcts. The Cullinan mine (Fig. 12) is expected to account for 2.2 mcts while the Finsch mine should produce 2.0 mcts. Lucara Diamonds key asset is the Karowe mine (Fig. 13) in Botswana. This opencast mine was commissioned in 2012 with investment costs of US$ 120 million. It produces 0.4 mcts, generating sales of about US$ 230 million per year, and has a lifetime expectancy up to 2026. Gem Diamonds owns the two diamond mines ­Letseng in Lesotho and Ghaghoo in Botswana, which produced a total of 0.11 mcts in 2016. The Stornoway Diamond Corporation is set to quickly enter the circle of larger companies thanks to its Renard mine in Canada, which was put into ­operation in 2016. Over a period of 14 years, this mine is expected to produce 1.6 mcts annually.

3 Technology trends

The processing of diamond-bearing ore is carried out in three stages: the ore crushing with liberation of the rough diamonds, their concentration and beneficiation, and finally recovery. The technological challenges in the production of rough diamonds have increased. For a long time, differences in the production processes were largely due to the investment budgets available and whether they involved kimberlite ores from the mine, ­alluvial deposits, or the tailings of such mines. The rising cost pressure has meanwhile led to the situation that the application of new technologies leads to more economical production and the achievement of higher yields.

Fig. 14 shows a modern processing line, as used in the Lucara Diamond Corp’s Karowe mine [2]. This mine went into operation in 2012. The ore processing line has a rated throughput of 2.5 million t. The flow chart shows the essential process stages from the ore delivery and primary crushing to the recovery of rough diamonds. The main comminution of the ore is carried out in an autogenous mill. A subsequent sizing screen provides 4 fractions, of which the 3 coarse fractions >8 mm are conveyed to a sensor-controlled XRT sorting system (X-ray transmission method) while the fine fraction passes through a DMS stage (Dense Media Separation) with downstream X-ray sorting process. At the end of the process the rough diamonds are recovered and large amounts of rejects and the tailings fraction are produced.

The treatment plant is modern in that much of the concentration, enrichment and recovery of the rough diamonds are carried out by means of sensor-assisted processes. The centerpiece is a group of 6 XRT sorting machines (Fig. 15) that each handle 150 t/h in the coarse flow range of >8-60 mm. Only the DMS stage for separating the fine fraction with subsequent magnetic separation of the FeSi fractions represents a conventional physical-chemical process. An overview of the conventional processes used in the process steps of the rough diamond production is described inter alia in [3]. This includes jigging machines, ­so-called De-Beers pans, dense media separators, foam flotation systems, grease tables, grease belts, etc. However, there is a clear trend in the industry towards sensor-assisted processes, so that the use of conventional DMS processes is being shifted more to the fine fractions, which results in considerable cost advantages and higher yields.

The design of crushing and grinding machines has an important role in the production process. In the flow chart, a secondary crusher is installed upstream of the autogenous mill in order to counter the reduced throughput due to harder feed material. The plant is also equipped with a tertiary crusher to liberate diamonds from the particle size fraction >20 - <30 mm. The autogenous mill can also be operated in closed circuit with a cone crusher. In many other modern processing plants high-pressure grinding rolls (Fig. 16), also known as roller presses, are used as secondary crushers. HPGRs together with autogenous mills distinctly reduce the specific energy requirement and ensure good liberation without crushing or fracturing the rough diamonds [4].

The subject of diamond breakage is playing an increasingly important role in the production of rough diamonds. The larger and more shapely the rough diamonds are, the higher are the prices that can be achieved. Corresponding analyses are carried out on a regular basis by some of the leading companies (Fig. 17), who are looking for ways of totally avoiding diamond breakage. One possibility is to select special grinding processes or machines that largely prevent diamond breakage, such as the already mentioned HPGR or modified cone crushers for post-comminution after autogenous grinding. A further possibility is to introduce an XRT sensor process directly downstream of the primary crusher or the HPGR in order to identify and discharge particularly large rough diamonds before the grinding stage.

Various X-ray methods are used for sensor-assisted sorting in the production of rough diamonds. These include X-ray transmission (XRT), X-ray luminescence (XRL) and X-Ray fluorescence (XRF) methods. One X-Ray sorting system is shown schematically in Fig. 18. The system consists of the feeding unit, a separation chamber with the sensor system and the discharge system. The sensors recognize the target material on the basis of its typical characteristics. These particles are discharged via a nozzle system by precisely directed compressed air pulses, or by jets of water in a wet process plant. The quality of the system depends on how fast and precisely the technology works. This type of sorting machine is many times faster than traditional technology or manual sorting. Even the smallest valuable mineral particles can be detected and discharged [4].

4 Market outlook

Although some new diamond mines or extensions of existing mines will be put into operation in the coming years, the global production of rough diamonds will continue to decline. The biggest downturn is due to the fact that the Rio Tinto mines are expected to reach the end of their operational lives in 2024 at the latest. None of the existing companies will be able to compensate for this decline. Alrosa will, at best, only be able to increase production volumes slightly, while the of De Beers outputs will continue to decrease. Correspondingly, the forecast maximum production in 2030 is 120 mcts. On the basis of prices remaining the same, this will reduce the market for rough diamonds to below US$ 15 billion. Demand, however, is expected to be between US$ 20 and 25 billion.

Nevertheless, there are numerous mine expansion projects. These include the Gahcho Kué mine of De Beers/Mountain Province, Lucara’s Karowe mine, Gem Diamonds’ Ghaghoo mine, Alrosa’s Karpinsky mine, Stornoway Diamond’s Renard mine, Shore Gold’s Star-Orion South mine, Namakwa ­Diamonds’ Kao mine, Diamond Corp’s Lace mine, Firestone Diamonds’ Liqhobong mine and Endiama/Alrosa’s Luaxe mine. The potential up to 2023 will be around 20 mcts and this will rise to a maximum of 25 mcts in 2026. However, due to the rather short life of some of these mines, such as the Gahcho Kué Mine of De Beers, which only went into operation in 2013 but is expected to close down in 2029, the overall production volume will decrease significantly. One very clear aspect is the importance of the Luaxe mine in the Catoca deposit of Angola (Endiama/Alrosa). In 2030, this mine will account for more than half the total production of all new mines going into operation up to then.

If rough diamond supply shortages actually occur in the foreseeable future, several scenarios are conceivable. One is that the use of rough diamonds for industrial purposes will decrease and that usage for jewelry requirements will increase. Another possibility is that prices for rough diamonds will rise. Both of these scenarios will provide an impetus for manufacturers of synthetic diamonds and, with appropriate marketing strategies, it can be expected that the use of artificial diamonds in the jewelry sector will increase significantly.

Literatur • Literature

[1] Harder, J.: Russian‘s resources – Overview of the mining industry in Russia. AT MINERAL PROCESSING, 09/2010, pp. 80-93

[2] Lynn, M.D. et all: Karowe Diamond Mine Botswana NI 43-101 Independent Technical Report (Amended), dated 4th February 2014 for Lucara Diamond Corp

[3] v.d. Westhuyzen, P. et all: Current trends in the development of new or optimization of existing diamond processing plants, with focus on beneficiation. The Journal of the Southern African Institute of Mining and Metallurgy, Vol.114, July 2014, pp. 537-546

[4] Harder, J: Opportunities – Trends in Dry Processing. AT MINERAL PROCESSING, 07-08/2013, pp. 55-66