Refractory erosion in iron melting is seldom attributable to mechanical wear. Erosion generally is the aftermath of slag’s chemical attack – chemical reactions between slag and refractory – producing reaction byproducts that enter the slag. The overall effect is refractory’s effective protection layer, the thickness of the refractory lining, is continually eaten away, eventually leading to failure.
Refractory industry ceramic engineers have long used a very precise, real-time method to observe refractory-molten metal interactions via X-ray, and to evaluate refractory resistance to slag attack
The Sessile Drop Test consists of placing a drop of molten iron on a refractory material to be tested, and placing the apparatus into a precisely controlled, high-temperature furnace containing an inert atmosphere where the interface between the droplet and refractory can be continuously viewed by real time x-ray. Initial reactions between the metal and refractory are minimal due to the inert atmosphere. Then, oxygen is added to the atmosphere, which instantly causes iron oxide (FeO) to form on the molten droplet’s surface. Once iron oxide is present, rapid interactions between the molten iron and the refractory begin.
Without iron oxide, chemical reactions between the refractory and the molten iron are very subdued, almost non-existent. Iron oxide is the catalyst that changes everything, which demonstrates iron oxide’s far-reaching role in initiating refractory chemical erosion.
Two-pronged attack — Refractory erosion results from two sources. First, iron oxide overwhelms the refractory’s individual constituents, quickly reacting with those ingredients and converting them into useless slag. This type of iron-oxide attack simply overwhelms refractories regardless of their quality or purity.
Next comes a chemical attack of silica or other impurities within the refractory that affects its overall performance. The precise nature of these attacks is beyond the scope of this analysis, but the chemical reaction’s overall effect on the refractory’s service and performance pales in comparison to iron oxide’s gruesome attack.
Refractory erosion can be effectively stopped, and refractory service life extended to unimaginable lengths, if iron oxide is controlled. Iron oxide-induced attack must be solved before the material's designed refractoriness can benefit the specific application.
Refractories are blended to incorporate the specific properties of each ingredient. When iron oxide doesn’t steam-roller the application, the specific refractory properties designed into new refractories enhance service life beyond current performance, raising it to new levels.
Cupolas operating with low iron-oxide slag and with specially formulated refractories are setting benchmarks for service life, and the melting personnel endorse this great improvement. With the new refractory and low iron-oxide slag, cupolas can be drained at the end of major campaigns and require no refractory repair. Slag has not built up anywhere, and the small metal-line slag clumps are chipped off quickly and easily. No overnight or weekend repair is needed — unheard of for normal cupola operation.
DeOX metal treatment is designed to chemically reduce iron oxide in slag, effectively stopping iron oxide’s reactions with refractory. It chemically converts iron oxide into inert byproducts and shields the molten bath with a type of surface barrier, thereby preventing contact with the atmosphere and stopping iron-oxide formation.
DeOX metal treatment accomplishes exactly that when added to electric furnaces or other molten iron baths. It does not melt but rather mixes into existing cover slag masses, eventually reacting with iron oxide as iron oxide enters the slag layer. DeOX instantly neutralizes iron oxide formed during the melting stage in electric furnaces, similar to what it accomplishes in cupolas.
In cupola melting, iron-oxide formation cannot be stopped. It always forms in the tuyere raceways. The secret to DeOX metal treatment is that it effectively neutralizes iron oxide as it forms. Tuyere raceways are the only location in the cupola where oxygen molecules exist, and iron oxide forms only in the tuyere raceway. Immediately above the raceways, oxygen molecules have been converted into carbon monoxide and carbon dioxide by the coke combustion process. Iron oxide will not form without the presence of oxygen.
DeOX metal treatment is the only process and/or material that neutralizes iron oxide in the raceway, and it effectively eliminates iron oxide throughout the cupola.
Harmful levels of iron oxide — Iron oxide levels in cupola and electric furnace slags vary from over 50% all the way down to 0.2% FeO. The 0.2% FeO nears the inert level of FeO’s refractory erosion tendencies. Cupolas and EFs operating with 0.2% FeO show almost no refractory erosion. As a result, there is unprecedented refractory service life.
The taphole refractory shows no erosion after 80,000 tons of melted iron. Cupola front-box refractory shows no erosion, no slag build-up, and requires no repair after many weeks of melting.
Two adjacent foundries — a gray iron foundry side-by-side with a ductile iron foundry — both using 35-ton vertical channel furnaces, showed dramatic differences in furnace refractory service life. One furnace was skimmed on a daily basis to maintain a “slag free” molten iron surface, and the other was allowed a full slag cover with low-FeO content in the cover slag.
The cover slag in the latter furnace was seldom removed during a three-year comparison study. The low-FeO slag furnace showed an eight-times minimum service-life improvement over the “slag-free” iron surface furnace.
It has been found that furnaces holding or melting low free-oxygen iron produce 75% less slag than what is typical of standard melting and holding furnaces. This goes hand in hand with maintaining a cover slag barrier to prevent re-oxidation and seldom removing any of the protective cover slag barrier.
Once iron oxide is controlled, little additional slag is produced because silicon and manganese are not oxidized. This simplifies maintenance of a protective cover-slag barrier. Low iron-oxide slag is physically different from normal slag, forming clumps that do not stick together or to refractory sidewalls, inlets, or outlets. It forms a “dry” cover-slag barrier.
The theory that SiC added to refractory improves service life due to its hardness and inertness to chemical attack has been disproven. In fact, SiC used in refractories is chemically reactive with iron oxide — the exact opposite of that old theory.
Furnace manufacturers and some refractory suppliers maintain that cover slag should be removed daily from furnaces and pour vessels, so that the molten metal surface will be clean of slag. This is misguided.
Slag continually forms when molten iron contacts the atmosphere. Because this is unavoidable, daily cleaning of the molten metal surface will produce a higher level of iron oxide in the freshly formed slag layer than if the slag simply accumulated. Freshly formed slag contains 70-90% iron oxide and the balance is SiO2. Over time, the silica component in the slag increases due to oxidation of silicon in the molten iron.
The iron oxide component in cover slag promotes the slag’s crustiness characteristic, making temperature monitoring and chemistry sampling difficult. Penetrating slag crusts for sampling can damage the refractory and lead to personnel injuries. It is troublesome that “green poling” is still used today to break up hard slags.
Low FeO cover success — Two ferrous foundries demonstrate what results with low iron-oxide slag cover.
Example 1. A Michigan foundry encountered severe refractory erosion in a 35-ton, horizontal drum holding furnace. At times, the furnace service life did not reach six months before refractory replacement became necessary. Another problem was that the cover slag layer frequently needed oxygen lancing to allow metal from the cupola to enter into the molten bath.
The cupola and furnace were converted to a low iron-oxide process — low free-oxygen iron. Furnace slag issues immediately disappeared, along with slag-line erosion in the furnace. Inductor loop plugging stopped, and outlet plugging stopped.
The furnace was slated for relining within four weeks. It was nearing its useful service life when the DeOX Metal Treatment technology was begun.
Adopting DeOX metal treatment forestalled the reline. Refractory erosion was halted, and the furnace went on to set an all-time record for service life. Simply eliminating iron oxide in the cover slag achieved all this.
Example 2. The 35-ton holding furnace mentioned above was held full of iron for 52 days during cupola shell replacement. Nothing special was done to the furnace during that time: No alloys were added, no special airtight sealing was applied. The furnace was full at the beginning of the hold period.
The metal chemistry at the start of the hold was 3.42% C; 2.03% Si. Hold temperature was 2,600°F.
After 52 days, the chemistry checked the same: 3.42% C; 2.03% Si. Because no chemistry decline over 52 days was unheard of, the chemistry buttons were checked at two other foundries via spectrograph and laboratory analysis. All confirmed the same chemistry.
The cover slag in the 35-ton furnace contained less than 1% FeO. Visible “sparkles” from embedded Mastermelt special SiC particles were observed in the cover slag.
When this furnace was being brought back on-line after the extended idling, iron in the spout was free of crusty slag and easily opened for metal sampling. Chill wedges taken from the spout, un-inoculated, checked at 5/32nds chill depth — very similar to inoculated iron but this iron had been dormant for 52 days. These are extraordinary results, never seen before.
A thorough analysis of every casting poured from the “held” iron revealed no scrap castings, a true DeOX metal treatment success story.
This is the fifth installment in a series of reports addressing the effects of oxygen in iron melting. Also see:
Harmful Effects of Molten Metal Oxidation, FM&T June 2020.
Controlling Iron Oxide to Stop Carbon and Silicon Losses, FM&T July 2020.
Controlling Molten Iron Chemistry and Metal Fluidity, FM&T August 2020
Treating Oxidation to Reduce Iron Casting Scrap, FM&T September 2020