*4.3. Industrial Perspective on Feasibility of Implementation*

The computational simulations conducted in this research indicate that pre-heating of natural gas prior to injection into the blast furnace could offer process benefits. However, modeling does not consider the effort required for industrial implementation and the associated difficulties that such an effort might present.

Several routes to pre-heating exist, recuperative (including waste heat), on-demand, and in-situ partial combustion; each route having a certain level of technical maturity, capability, controllability, as well as CAPEX and OPEX characteristics that are plant unique. Conceptually, most of the identified heating technologies would heat gas sourced singly from the main furnace distribution line, i.e., all the injected natural gas would be passed through a single recuperative heater, while the heating routes illustrated in Figure 10 would heat individually closer to the point of consumption.

Implementation for the single source is somewhat easier, although potentially more expensive, as much of the construction and commissioning could be completed while the furnace remains online. However, pre-heating via the blowpipe route would require an outage of suitable length to change all the blowpipes, or a planned approach to replace at regular intervals, which could take several years and leave the furnace with unbalanced pre-heating in the meantime.

Industrial consideration also needs to be given for material selection (pipes, valves, and seals need to be capable of operating at higher temperatures) and process control (which may have varying feasibility depending on the route chose, i.e., blowpipe heating). In addition, modified operating and maintenance practices would need to be developed, and safety protocols would also require careful review. One example would be RAFT control, which is widely adopted as a key variable in ironmaking. Typically, front-line operators use a simple empirical model such as the AISI RAFT model equation to determine real-time RAFT values. As currently defined, this model does not account for pre-heating of natural gas. Similarly, the well-known operating rules-of-thumb would need expansion to include the effect of pre-heating temperature on RAFT, top heat, fuel rate, and productivity.
