Excitement over direct-reduced iron (DRI) is running high as the first new plant in half a decade has come into service and other projects are accelerating. At the same time, excitement is always a bit muted because no one makes DRI just to make DRI.
Mills make DRI to make steel. That may seem tautological, but it means that advances in process technology are beholden to other economic factors in the steelmaking process. Size is the key trend rather than technology, according to analysts. Some DRI projects are world-scale operations to supply the parent company and even commercial markets, while others are more like custom feed trains for individual electric-arc furnaces (EFs).
The business in North America and Europe is dominated by two main technology providers: Charlotte, N.C.-based Midrex Technologies Inc. and Tenova HYL, a subsidiary of Milan-based Tenova SpA. Rivalry is keen but not cutthroat, as the global investment in DRI means that competition is not a zero-sum game. The companies highlight completely different characteristics when differentiating their technology from one another. And in practice, some steelmakers license both processes.
Charlotte-based Nucor Corp. achieved a major milestone when its $750-million DRI plant in St. James Parish, La., started up late last year. The facility, which uses Tenova technology, is the largest DRI plant in the world and the first to operate in the United States since 2009. Nucor said within just a few days of starting operations that the DRI being produced was on specification. The company confirmed that the plant has been yielding 275 tons of DRI per hour, which it said is the highest output of any DRI plant in the world.
Along with its DRI plant in Point Lisas, Trinidad and Tobago, Nucor is two-thirds of the way toward its goal of producing 6 million to 7 million tons of high-quality scrap substitutes. By producing more of its own DRI, Nucor aspires to have tremendous flexibility in using the lowest-cost iron units in steelmaking by providing a significantly shorter and more secure supply chain for high-quality iron units, chairman, president and chief executive officer John J. Ferriola said. The DRI inputs also support the investments Nucor has made to produce higher-margin sheet, special bar and plate.
We have been very pleased with the performance of the DRI plant during its first few months of operation, Ferriola said. In the first three months of operation, the facility attained peak operating rates above 90 percent. More importantly, the quality of the initial output was outstanding, with metallization rates of 96 percent and carbon content exceeding 4 percent. Our mills that have received DRI from our Louisiana facility have been very pleased.
The company had said that a second line would be possible, but it has remained quiet about those tentative plans. There has been no official word from Nucor or any company affiliated with the project, but industry sources said that conditions in the global steel market militate against an immediate doubling of DRI despite the sparkling performance of the first train. There is broad expectation that a second line will be built in the longer term but not likely in the near term.
Ferriola is not the only one pleased with the progress at St. James.
Four is the most important number for Francesco Memoli, vice president of steelmaking at Coraopolis, Pa.-based Tenova Core Inc.Ñas in the 4-percent carbon content his companys process can reach. We are extremely pleased, almost surprised, at how well the plant is operating. We had no doubt that Nucor was the best client for such a challenging project, based on their culture. They know how to make things work. What surprised us just a little is the results from the output of their furnace goes beyond what we had hoped.
Memoli would not comment on further licenses or prospects, but acknowledged that there are so many factors in a steelmakers decision on inputs. It is not just raw materials, ore, gas and so forth. We as technology providers are in front of a market that is unpredictable. Lately, it has not been easy to understand, much less predict.
Tenova calls its DRI process Energiron and identifies the current version as ZR, or zero reformer. We are the only process to reach 4 percent or higher in carbon, Memoli said. Four percent is comparable to pig iron. That is why Nucor is enjoying their furnace performance so much. Our DRI competes with pig iron, not with shredded.
Carbon content is not the only factor in the DRI decisionÑsticking points are a sticking point. What matters is not just the percentage of carbon sticking in the pellet but whether the carbon is bonded. In our DRI, the molecule is iron carbide, Fe3C. The carbon is bonded, so when it is charged into the EF it is not lost to offgas. It gets into the bath. That bonding also makes it dramatically less reactive with water, so its safer to ship and store.
Tenovas technology was originally developed by Mexicos HYL Technologies SA de CV, and the companys DRI technology and engineering center is still in Mexicos hub of Monterrey, Nuevo Leon. The project management and procurement base, as well as Memolis office, is in Coraopolis. Forty-nine modules, or reactors, using successive generations of the Tenova process have been sold worldwide, with a total capacity of more than 36 million tons per year, according to Memoli. We like to say that the sun never sets on Monterrey.
In fairness, the sun does not set on Midrex, either. Nucors first DRI plantÑNu-Iron Unlimited in TrinidadÑand the plant being planned by Voestalpine AG near Corpus Christi, Texas, use Midrex technology.
The outlook for licenses is strong, Midrex marketing manager Christopher Ravenscroft said, but he declined to provide any specifics. We can erect a plant in just 18 months, but the sales process can take years. All I can say is that we are talking in earnest to several people in North America and worldwide.
Ravenscroft is particularly pleased about the Voestalpine project, a $759-million, 2-million-tonne-per-year greenfield plant due to begin service at the end of next year. The facility will make a form of DRI known as hot-briquetted iron (HBI). It will be the first HBI merchant plant in the U.S., the largest ever, and the first DRI plant of any kind built specifically for blast furnace feed, he said.
Linz, Austria-based Voestalpine has said it would ship half the output to its blast furnace operations in Europe and sell the other half on the open market. HBI, a compacted form of DRI, was developed to solve the problems of storage and transportation for conventional DRI, which can become a fire hazard in the presence of water in enclosed areas.
One factor that is often overlooked in the DRI equation is environmental, Ravenscroft said. There are emissions restrictions with financial penalties already in Europe and probably before long in North America. There is no technology of any kind on the market that can help blast furnace operators on environmental factors other than DRI. In looking at innovation and how the future of the DRI business is playing out, we are anticipating where economic and environmental factors are going.
HBIs greater density is better in the blast furnace because of its physical burden. But the most important factor is removing the oxygen, and charging HBI jump-starts that process, said Robert Hunter, Midrex manager of iron products applications. However, there is no one solution for the blast furnace or the EF. Mills have to be very flexible, and they will adopt the inputs that fit best.
Midrexs HBI reactors discharge hotter than others for EFs, Hunter said. That is a big advantage EF operators have realized. They are saving money by keeping the heat up.
The basic DRI concept is at least 60 years old but licensors say there are no quantum leaps in process technology cooking in the lab. Rather, providers continue to tweak the basic process while customizing plant sizes and other external variable to better fit DRI into the steelmakers supply chain.
Midrex has formed a partnership with Linz-based Siemens VAI Metal Technologies Gmbh & Co. to develop three different forms of hot transport for HBI, Hunter noted. A plant in Malaysia is using special containers to deliver HBI at 650 Celsius (about 1,200 degrees Fahrenheit). That is hundreds of degrees hotter than other processes.
The current technology drive for DRI is outside the main reactor, according to Ralph Smailer, owner and director of Pittsburgh-based MetServ Metallurgical Consulting. The main difference between the two top licensors is the process gas composition, which reflects the carbon content, he said, adding that the degree to which a user values carbon is a major factor in the process selection.
There also are rotary-hearth technologies that are appropriate for recovering iron units from waste; a rotary-kiln process predominates in India using lump ore and coal. Its not fancy, Smailer said, but India makes millions of tons of DRI that way. Another technology makes DRI from ore fines.
Smailer does not see any game-changers among those technologies, however. The trend in DRI is not technology per se but process. We are seeing huge world-scale plants like the Voestalpine project, or we see direct hot feed customized for a single EF. Remember, the demand for DRI in the first place comes from a shortage of high-quality scrap, so it is becoming a necessity for the EFs.
However, in blast furnaces DRI is a convenience. Ore is still preferred, but with HBI the operator can pick up an increase in productivity of about 12 percent, he said. But blast furnace operators are indifferent to carbon because they have coke. They like to have HBI that they can ship and handle in all circumstances, but they certainly dont want to pay for carbon at the cost of HBI.
The steel industry can be considered a competition to see who can procure ready-to-melt iron units at the lowest cost, Andrew Lane, an analyst at Chicago-based Morningstar Inc., said in his July industry report. Whether the iron units are from virgin iron ore, ferrous scrap or high-iron-content complementary feedstocks, iron-related unit costs are a major driver in determining a steelmakers position on the industry cost curve. Because steel products are largely undifferentiated and commodified, a steelmakers most likely route to earning an economic moat is by establishing itself as a definitive low-cost producer.
We estimate that a representative ton of steel produced solely by the use of DRI in an EF could be produced at a unit cost of $324. That is roughly 15 percent below the estimated $380-per-ton unit cost for steel produced from pig iron. For iron ore, we anticipate that iron ore spot prices will average $95 per short ton (62-percent fines, c.f.r. China) over our five-year explicit forecast period. We expect metallurgical coal prices to average $150 per short ton and natural gas prices to average $4.75 per million British thermal units (mmBTU), Lane said. Our mid-cycle 2018 price forecasts for these three raw materials are $93 per short ton, $166 per short ton and $5.40 per mmBTU, respectively (in nominal terms). Natural gas prices would have to climb to around $9.80 per mmBTU before Nucors DRI-based production method would no longer reflect favorable unit costs relative to steel production by the use of pig iron.