Skip to content ↓

Team aims for cleaner steelmaking

Steel makes up 95 percent of all the metal made in the world, but the negative environmental effects of its production are also large. The industry generates roughly five percent of all greenhouse gases produced by human activity in addition to major amounts of sulfur dioxide (the gas that contributes to acid rain) and a variety of other noxious byproducts.

Now imagine a steel industry with oxygen as its principal byproduct.

That is the dream of Professor Donald R. Sadoway of the Department of Materials Science and Engineering (MSE). To that end Professor Sadoway is developing a new technology for the production of steel that involves zapping a molten mixture of iron ore and other materials with electricity. Unlike the current technology for producing steel, the new process does not use the element at the root of steel's problems: carbon. The principal byproduct of the new system? Oxygen.

The work has been well received by the steel industry. Thomas J. Usher, president of US Steel, said at a talk at MIT that "Professor Sadoway's research. is laying the foundation required for carbon-dioxide-free iron production."

Professor Sadoway is quick to point out, however, that the technology is still in the early stages of research. Furthermore, it won't be truly viable until the world has an abundant supply of carbon-free electric power (e.g., nuclear, hydroelectric or solar). Otherwise, he said, "you're simply shifting the source of pollution."

But he stresses the importance of forging ahead. "We can't wait. We must start now to develop new technologies for the future."

And along the way, interesting things can happen. For example, last month Professor Sadoway and colleagues in his laboratory received a patent on a device that measures the electrical properties of the melts their technology revolves around. The device turns out to have a variety of applications with other liquids. (Professor Sadoway received a patent on the steelmaking technology in 1993.)

In addition, preliminary tests show that the technology itself could be used to remediate certain wastes and produce other metals like titanium.


Dubbed pyroelectrolysis, the technology uses electrical energy to do chemical work. Iron oxide (derived from iron ore, the precursor to steel) is fed into a reactor called an electrolysis cell where it is made to dissolve in a solution of other molten oxides. An electric current is passed through the cell from one end (the anode) to the other (the cathode). At the interface of the molten oxides and the cathode, pure liquid iron-the steel-is formed. The principal byproduct (oxygen) bubbles off the anode.

Carbon would still be added to the process as an ingredient (the basic definition of steel is iron plus a small amount of carbon), but it would not be used in the vast amounts it is today as a fuel and as a chemical reactant to extract the iron from its ore. Currently it takes about half a ton of carbon to make a ton of steel.

Electrolysis itself is nothing new-all of the world's aluminum is produced this way. And that is one advantage of the new process: it is based on a technology that metallurgists are already familiar with. The difference here is that no carbon is involved. (In aluminum production, the anode is made of carbon, which is consumed in the process thereby generating carbon dioxide.)

Pyroelectrolysis could have a number of other advantages for steel production. For example, it could produce higher-purity alloys. The carbon used in the current steelmaking process contains sulfur, a contaminant that leads to a lower-quality metal. The new technology would eliminate the steps currently in place to remove sulfur from steel, saving energy and money.

So why hasn't the technology been pursued before now for steel production? For one, there hasn't been an economic incentive, because of the low cost of carbon. In addition, relatively little is known about the chemical and electrical properties of the molten oxides used in pyroelectrolysis. Such information is critical to determining, among other things, what the electrical needs of the process will be.


Currently the researchers are exploring these and other problems. Susan Schiefelbein recently received the PhD for work on the electrical properties of the molten oxides that the iron oxide will be added to. Andrew Ducret, a postdoctoral associate in MSE, will take Dr. Schiefelbein's work one step further. He will measure the electrical properties of the molten oxides to which the iron oxide has been added. "This work really moves forward in tiny increments," Professor Sadoway said. "We're doing the underlying science for what could be."

The group is also searching for carbon-free materials with which to build the electrolysis cell. "We're using platinum as the anode in our test cells, but that wouldn't be feasible for industrial cells because it's too expensive," he said. To this end he has developed a methodology for materials selection and testing.

Other researchers working on the project are Kevin Rhoads, a lecturer in the School of Engineering, and MSE graduate student Naomi Fried. Dr. Rhoads' expertise played an important role in developing the device to measure electrical properties of the molten oxides. Ms. Fried is applying pyroelectrolysis to the production of titanium (see MIT Tech Talk, May 10, 1995).

At present the work is unfunded, although it has been supported in the past by the National Science Foundation and the Electric Power Research Institute. Why hasn't the steel industry helped?

"I think because this solution could be 20 years off, and industry doesn't like to fund things that are that far off," Professor Sadoway said. "They are working on nearer-term partial solutions."

The program on pyroelectrolysis will continue, though, perhaps in large part because of the researchers' enthusiasm.

Said Professor Sadoway, "I really love this work because it has science and engineering, and it could benefit society. If it works, we'll make a difference in peoples' lives."

A version of this article appeared in MIT Tech Talk on March 20, 1996.

Related Topics

More MIT News