含铁原料：喀麦隆缺的是铁，而不是轧制能力

3.1 Billets show where the shortfall lies

Cameroon’s imports of iron and steel (customs chapter 72) rose from 182,890 t in 2019 to 412,049 t in 2025, according to the national statistics institute; in 2025 alone they rose by 33% [S67, S68]. The composition of those imports is more revealing than the total. In 2023, the last year reported to the UN Comtrade database, semi-finished steel (HS 7207, mostly billets) amounted to 161,400 t worth US$151.7 million, three times the 2019 level [S05]. Indonesia supplied 125,500 t of it, or 78% [S05]. Imports of rebar, by contrast, were marginal because rebar imports have been prohibited since 2016, a ban that Customs confirmed in April 2023 [S69].

Source: INS Cameroun foreign-trade notes for chapter 72 tonnage [S67, S70, S71]; UN Comtrade for HS 7207 [S05]. The 2023 billet share of chapter 72 tonnage (about 53%) is derived.

The pattern is that of an industry that owns rolling mills but not enough steelmaking. Cameroon Tribune, citing the industry association, reported in May 2026 that domestic rebar capacity far exceeds the rebar market [S06]. Industry sources describe about 1.2 Mt a year of installed steel transformation capacity running at about half of its potential, and note that local scrap now covers only about 40% of mill needs [S32, S33]. In September 2026 a leading domestic producer secured a €40 million revolving facility from Afreximbank, reported by the financial press as intended to import billets [S33, S72].

Cameroon is not new to electric steelmaking: in 2023 Cameroon Tribune described a Douala producer’s electric arc furnace, of about 200,000 t of rebar a year, as the first of its kind in Central Africa [S73]. But domestic melting remains far smaller than domestic rolling.

The implication for technology is direct. A new producer adds the least value, and takes the most market risk, if it builds another re-rolling mill that buys imported billets. It adds the most value by making steel, the step the country lacks. A melt shop with a billet caster also has an option that a re-roller does not: it can supply billets to other mills.

3.2 Domestic scrap is a fraction of what a steelworks needs

Scrap supply grows with a country’s accumulated stock of steel in use: buildings, bridges, vehicles, machinery and packaging that reach the end of their lives. Cameroon’s stock is small, and so is its scrap. The only quantified national study, published in 2007 by researchers from the National Advanced School of Engineering in Yaoundé, estimated annual ferrous scrap arising at 53,000 to 61,833 t, of which 53% was collected by recyclers, 29% by informal dismantlers and 15% by exporters [S13]. Recorded scrap exports have since fallen to near zero: 3,675 t in 2019 and 15 t in 2023 [S05]. Scrap purchase prices rose by about 30% in the year to September 2026, according to the press [S32].

No more recent survey exists. A rough cross-check is possible: domestic mills produced about 260,000 t in 2022 [S74]; if they consume about 1.2 t of metallic charge per tonne of product (the RC-1 yields) and local scrap covers about 40% of it, domestic scrap collected is of the order of 0.1 Mt a year (derived; Annex A). That scrap is already consumed by existing mills. A new melt shop that competed for it would mainly raise its price.

Source: Cameroon Steel reference configuration RC-1 (planning estimate), indicative balances [S03]; scrap study [S13]; production and scrap-share inputs [S32, S74]; derivations in Annex A.

Figure 3.2 shows the scale of the gap. Cameroon Steel’s reference configuration (RC-1) plans 0.3 Mtpa of finished products in Phase 1, rising to 0.5 Mtpa and then to 1 Mtpa [S03]. With a metallic yield of 0.88 t of liquid steel per tonne charged and a casting and rolling yield of 0.94, those capacities need about 363,000 t, 604,000 t and 1.21 Mt of iron units a year (derived from RC-1 assumptions) [S03]. Even Phase 1 would need between three and four times the scrap that Cameroon collects today, and about six times the scrap arising estimated by the 2006 national study.

3.3 Imported scrap, DRI, HBI and pig iron fill the gap

An electric melt shop at a deep-water port can buy iron units on the international market. Each has a different profile.

Imported scrap is the cheapest iron unit per tonne of iron where it is available, but Africa as a whole was a net importer of ferrous scrap in 2025, importing 3.4 Mt and exporting 1.2 Mt [S41]. The large exporting regions are the European Union and the United States, which shipped 16.7 Mt and 11.8 Mt respectively in 2025, while Türkiye imported 18.8 Mt [S64]. Scrap prices follow those flows, and the quality of imported scrap (its residual copper and tin, its density and its contamination) varies widely.

Direct-reduced iron and hot-briquetted iron are the traded virgin iron units. International DRI trade was about 10.8 Mt in 2025 on trade data, or about 13.2 Mt including sanctioned origins; Russia exported more than 2.0 Mt of HBI and India 1.6 Mt, while the United States, Türkiye, India, Nepal and Italy were the largest importers [S04]. HBI accounted for 11.6 Mt, or 7.7% of world DRI output [S04]. For a buyer without its own reduction plant, HBI is the practical form: it can be shipped with surface ventilation, whereas cold DRI must travel under an inert atmosphere because it re-oxidises and self-heats when wet [S14, S15].

Merchant pig iron offers clean iron with high carbon, which adds chemical energy in the EAF [S43]. Its supply is concentrated in a few exporting countries and it carries the emissions of the blast furnace that made it.

Table 3.1 Iron-unit options for the Kribi melt shop, by role and risk

Source: Cameroon Steel; [S03, S04, S14, S15, S41, S43, S64].

3.4 How iron units change the melt shop’s performance

The choice of iron units is not only a supply question; it changes how the furnace operates.

Scrap is already metallic and melts with the least energy, but its residual elements (copper, tin, nickel, chromium) cannot be removed in the furnace and accumulate in each recycling cycle. For rebar these residuals are tolerable within limits; for wire rod destined for fine drawing, they matter more. DRI and HBI dilute residuals because they contain almost none.

DRI and HBI carry gangue (silica and alumina from the ore) and some unreduced iron oxide. The furnace must melt the gangue and flux it with lime, which increases slag volume, electrical energy and refractory wear, and reduces metallic yield. The ENERGIRON reference indicates that each additional 10% of DRI in the charge adds about 14.5 kWh per tonne of liquid steel (Chapter 5) [S20]. That penalty is the price of independence from scrap.

Pig iron and the carbon in DRI add chemical energy when oxidised with injected oxygen, which partly offsets the electrical penalty and helps form the foaming slag that protects the furnace walls.

The planning assumption in RC-1 is that imported DRI or HBI forms about 30% of the Phase 1 charge, within a range of 20 to 50%, with the balance in scrap [S03]. On that basis, Phase 1 would use about 109,000 t of DRI or HBI and 254,000 t of scrap a year (derived) [S03]. Because domestic scrap is scarce and already in use, RC-1 assumes that the plant can buy only about 30,000 t of it a year, within a range of 0 to 60,000 t bounded by the 2006 national estimate of scrap arising [S03, S13]; about 224,000 t of the scrap, and all of the DRI or HBI, would be imported. Any configuration that relied on more than about 60,000 t a year of domestic scrap would not be feasible on current evidence. RC-1 estimates inbound port tonnage of about 351,000 t a year in Phase 1 for imported metallics and consumables (derived) [S03].

3.5 Scrap quality and preparation decide yield

Scrap is not a single commodity. It arrives as heavy melting steel from demolition, light sheet and packaging, turnings and borings from machining, and shredded scrap from end-of-life vehicles and appliances. Each grade has a different density, cleanliness and residual content, and the melt shop’s yield, energy use and charging time depend on the mix. Dense scrap fills the furnace in fewer charges: a large modern furnace published in 2026 is designed to run on two bucket charges with scrap densities down to 550 kg/m³ [S75]. Light, dirty or wet scrap needs more charges, more energy and more fume treatment.

Preparation therefore belongs to the melt shop’s design. A scrap yard needs space to separate grades, shears or balers to cut and compact oversized and light material, magnets and cranes to build buckets to a recipe, and inspection at the gate for sealed containers, explosives and radioactive sources, which are standard hazards in scrap trading. Imported scrap should be bought to recognised grade specifications and checked on arrival; domestic scrap, much of it collected informally [S13], will need sorting and cleaning before it can be charged safely.

3.6 Receiving iron units through Kribi

Because most Phase 1 iron units will be imported, the plant’s metallics chain starts at the port. The Port Autonome de Kribi inaugurated its second container terminal, with a 715 m quay and 16 m of water depth, in May 2025 [S76]. A mineral terminal of 125 Mt a year was the subject of a memorandum of understanding signed on 4 November 2025 [S37], and in March 2026 the port’s Director General expected the mineral and hydrocarbon terminals to be completed within five to six years [S38]. Phase 1 therefore plans on the existing general-cargo and container terminals; White paper 02 examines the port’s bulk capacity. HBI and scrap move in bulk carriers or containers; both need covered or well-drained storage at the port and at the plant. RC-1 estimates that Phase 1 would bring in about 351,000 t a year of imported metallics and consumables, and about 74 truck loads a day if all movements go by road before a rail siding exists (derived) [S03]. The site layout keeps a reserve for a rail siding and logistics corridor [S03]. When a direct-reduction module is built, the same logistics chain will handle about 1.45 t of pellets for every tonne of DRI (Chapter 4).

3.7 A metallics strategy, not a single supplier

Iron-unit security will be one of the main operating risks of the Kribi melt shop, and it can be managed with a small number of disciplined practices.

Two sources for every iron unit. Phase 1 should be supplied under term contracts with at least two suppliers each for scrap and for HBI, from different regions, with spot purchases on top. That reduces exposure to a single supplier, freight lane or sanctions regime.

Storage sized for disruption. HBI and DRI must be stored dry and covered; scrap yards need space for grading and preparation. RC-1 reserves a scrap and metallics stockyard with a covered DRI/HBI store in Phase 1 [S03].

Quality at the gate. Every delivery should be sampled for metallisation, carbon and gangue (for DRI/HBI) or inspected and graded (for scrap), and the results linked to the heats in which the material is used (Chapter 8).

Formal domestic collection. Supporting a formal scrap collection and preparation chain, with weighbridges, shears and safe handling, improves quality and traceability and keeps domestic scrap in Cameroon.

A clear path to domestic iron. The decisive step is to make DRI at Kribi from Cameroonian ore. The mining convention of Cameroon Mining Company, Cameroon Steel’s sister company and holder of the Mbalam permit [S77], is reported to provide for 15% of production to be transformed locally; the same press report presents Cameroon Steel as the national steel hub of the programme [S36]. Chapter 4 examines what has to happen to the ore before that clause can feed a shaft furnace.

3.8 What the metallics balance means for the pathway

The metallics balance explains the shape of the pathway. Phase 1 can be built and run on purchased iron units because a 0.3 Mtpa melt shop needs about 0.36 Mt of them a year, a quantity the international market can supply. At 1 Mtpa the requirement rises to about 1.2 Mt a year. Buying that quantity indefinitely would expose the plant to import prices, freight and foreign-exchange risk for most of its costs; making a large part of it in Cameroon would retain the value added and reduce imports. Figure 3.2 shows that with domestic DRI at 70% of the charge, the full plant would need about 846,000 t of DRI and about 363,000 t of scrap a year (derived) [S03]. That is why RC-1 keeps a reserve for a direct-reduction module of about 1 Mt of DRI a year at full development, subject to ore quality and gas supply [S03].