skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Simulation of blast-furnace raceway conditions in a wire-mesh reactor: interference by the reactions of molybdenum mesh and initial results

Abstract

A novel trapped air injection system has been built for a wire-mesh reactor to enable tests with short exposure times to air that are intended to simulate typical residence times in blast-furnace raceways. Initial tests have shown that the molybdenum wire-mesh sample-holder reacts with O{sub 2} under conditions intended for this work. By varying the proportions of solid MoO{sub 2} (weight gain), vapor phase oxides (weight loss) may form, depending on reaction conditions. Oxide formation pathways thus become relevant to coal weight loss determinations during experiments. If, in addition to solid MoO{sub 2} formation, significant formation of vapor phase oxides occurs, then the weight change is more complicated to understand and the impact on the O{sub 2} concentration cannot be unravelled. Furthermore, it turns out that O{sub 2}-scavenging by the mesh affects the amount of O{sub 2} that is available to react with the coal sample. It was concluded that it is only possible to conduct reliable tests under conditions which the favor the formation of solid MoO{sub 2} only, as this leads to a quantifiable weight gain. Its impact can then be accounted for in the evaluation of the experimental weight change. In the case of MoO{sub 2} formation,more » the impact of the mesh oxidation on the amount of O{sub 2} available to react with the sample can also be estimated. It has been found that the wire-mesh reactor, equipped with the trapped air injection system, can be used to obtain valid data at up to 1600{sup o} C and 0.5 MPa. This pressure is similar to that of the blast-furnace raceway, but the temperature is several hundred degrees lower. However, preliminary tests have shown that useful kinetic data on the extents of reaction can be obtained with the equipment, provided it is operated under conditions that minimize the formation of vapor phase Mo oxides. 18 refs., 13 figs., 3 tabs.« less

Authors:
; ; ;  [1]
  1. Imperial College London, London (United Kingdom). Department of Chemical Engineering
Publication Date:
OSTI Identifier:
20838349
Resource Type:
Journal Article
Resource Relation:
Journal Name: Energy and Fuels; Journal Volume: 20; Journal Issue: 6; Other Information: n.paterson@ic.ac.uk
Country of Publication:
United States
Language:
English
Subject:
01 COAL, LIGNITE, AND PEAT; BLAST FURNACES; COAL; INJECTION; CHEMICAL REACTORS; BENCH-SCALE EXPERIMENTS; MOLYBDENUM OXIDES; ERRORS; CHEMICAL REACTIONS

Citation Formats

Long Wu, N. Paterson, D.R. Dugwell, and R. Kandiyoti. Simulation of blast-furnace raceway conditions in a wire-mesh reactor: interference by the reactions of molybdenum mesh and initial results. United States: N. p., 2006. Web.
Long Wu, N. Paterson, D.R. Dugwell, & R. Kandiyoti. Simulation of blast-furnace raceway conditions in a wire-mesh reactor: interference by the reactions of molybdenum mesh and initial results. United States.
Long Wu, N. Paterson, D.R. Dugwell, and R. Kandiyoti. Fri . "Simulation of blast-furnace raceway conditions in a wire-mesh reactor: interference by the reactions of molybdenum mesh and initial results". United States. doi:.
@article{osti_20838349,
title = {Simulation of blast-furnace raceway conditions in a wire-mesh reactor: interference by the reactions of molybdenum mesh and initial results},
author = {Long Wu and N. Paterson and D.R. Dugwell and R. Kandiyoti},
abstractNote = {A novel trapped air injection system has been built for a wire-mesh reactor to enable tests with short exposure times to air that are intended to simulate typical residence times in blast-furnace raceways. Initial tests have shown that the molybdenum wire-mesh sample-holder reacts with O{sub 2} under conditions intended for this work. By varying the proportions of solid MoO{sub 2} (weight gain), vapor phase oxides (weight loss) may form, depending on reaction conditions. Oxide formation pathways thus become relevant to coal weight loss determinations during experiments. If, in addition to solid MoO{sub 2} formation, significant formation of vapor phase oxides occurs, then the weight change is more complicated to understand and the impact on the O{sub 2} concentration cannot be unravelled. Furthermore, it turns out that O{sub 2}-scavenging by the mesh affects the amount of O{sub 2} that is available to react with the coal sample. It was concluded that it is only possible to conduct reliable tests under conditions which the favor the formation of solid MoO{sub 2} only, as this leads to a quantifiable weight gain. Its impact can then be accounted for in the evaluation of the experimental weight change. In the case of MoO{sub 2} formation, the impact of the mesh oxidation on the amount of O{sub 2} available to react with the sample can also be estimated. It has been found that the wire-mesh reactor, equipped with the trapped air injection system, can be used to obtain valid data at up to 1600{sup o} C and 0.5 MPa. This pressure is similar to that of the blast-furnace raceway, but the temperature is several hundred degrees lower. However, preliminary tests have shown that useful kinetic data on the extents of reaction can be obtained with the equipment, provided it is operated under conditions that minimize the formation of vapor phase Mo oxides. 18 refs., 13 figs., 3 tabs.},
doi = {},
journal = {Energy and Fuels},
number = 6,
volume = 20,
place = {United States},
year = {Fri Dec 15 00:00:00 EST 2006},
month = {Fri Dec 15 00:00:00 EST 2006}
}
  • A wire mesh reactor has been modified to investigate reactions of coal particles in the tuyeres and raceways of blast furnaces. At temperatures above 1000{sup o}C, pyrolysis reactions are completed within 1 s. The release of organic volatiles is probably completed by 1500{sup o}C, but the volatile yield shows a small increase up to 2000{sup o}C. The additional weight loss at the higher temperature may be due to weight loss from inorganic material. The residence time in the raceway is typically 20 ms, so it is likely that pyrolysis of the coal will continue throughout the passage along the racewaymore » and into the base of the furnace shaft. Combustion reactions were investigated using a trapped air injection system, which admitted a short pulse of air into the wire mesh reactor sweep gas stream. In these experiments, the temperature and partial pressure of O{sub 2} were limited by the oxidation of the molybdenum mesh. However, the tests have provided valid insight into the extent of this reaction at conditions close to those experienced in the raceway. Extents of combustion of the char were low (mostly, less than 5%, daf basis). The work indicates that the extent of this reaction is limited in the raceway by the low residence time and by the effect of released volatiles, which scavenge the O{sub 2} and prevent access to the char. CO{sub 2} gasification has also been studied and high conversions achieved within a residence time of 5-10 s. The latter residence time is far longer than that in the raceway and more typical of small particles travelling upward in the furnace shaft. The results indicate that this reaction is capable of destroying most of the char. However, the extent of the gasification reaction appears limited by the decrease in temperature as the material moves up through the furnace. 44 refs., 12 figs., 6 tabs.« less
  • A numerical model has been developed and validated for the investigation of coal combustion phenomena under blast furnace operating conditions. The model is fully three-dimensional, with a broad capacity to analyze significant operational and equipment design changes. The model was used in a number of studies, including: Effect of cooling gas type in coaxial lance arrangements. It was found that oxygen cooling improves coal burnout by 7% compared with natural gas cooling under conditions that have the same amount of oxygen enrichment in the hot blast. Effect of coal particle size distribution. It was found that during two similar periodsmore » of operation at Port Kembla's BF6, a difference in PCI capability could be attributed to the difference in coal size distribution. Effect of longer tuyeres. Longer tuyeres were installed at Port Kembla's BF5, leading to its reline scheduled for March 2009. The model predicted an increase in blast velocity at the tuyere nose due to the combustion of volatiles within the tuyere, with implications for tuyere pressure drop and PCI capability. Effect of lance tip geometry. A number of alternate designs were studied, with the best-performing designs promoting the dispersion of the coal particles. It was also found that the base case design promoted size segregation of the coal particles, forcing smaller coal particles to one side of the plume, leaving larger coal particles on the other side. 11 refs., 15 figs., 4 tabs.« less
  • In this work, a numerical model is used to study the flow and coal combustion along the coal plume in a large-scale setting simulating the lance-blowpipe-tuyere-raceway region of a blast furnace. The model formulation is validated against the measurements in terms of burnout for both low and high volatile coals. The typical phenomena related to coal combustion along the coal plume are simulated and analyzed. The effects of some operational parameters on combustion behavior are also investigated. The results indicate that oxygen as a cooling gas gives a higher coal burnout than methane and air. The underlying mechanism of coalmore » combustion is explored. It is shown that under the conditions examined, coal burnout strongly depends on the availability of oxygen and residence time. Moreover, the influences of two related issues, i.e. the treatment of volatile matter (VM) and geometric setting in modeling, are investigated. The results show that the predictions of final burnouts using three different VM treatments are just slightly different, but all comparable to the measurements. However, the influence of the geometric setting is not negligible when numerically examining the combustion of pulverized coal under blast furnace conditions.« less
  • The various factors all influenced the quality of the coke. High temperatures and carbonizing rates reduced the yield of large coke lumps (>60 mm) and increased that of intermediate lump sizes (60 to 40 and 40 to 25 mm), thereby making the coke more uniform. Low temperatures had the opposite effect. Although the coke finishing temperature had no influence on the initial properties of the reducing gas (at the tuyeres) or the material to be reduced (the iron ore burden), differences between the properties of cokes made under differing conditions led to alterations in the reduction rate within the blastmore » furnace and the consumption of reducing agents per ton of iron made. The changes in blast-furnace performance indicate that the high-temperature coke improved the aerodynamic smelting conditions. This made it possible to increase the blast volume (driving rate) by 59 m/sup 3//min and extend the reaction periphery. The furnace conditions remained stable although the overall resistance factor was raised; the pressure drop and the dynamic gas pressure in the throat were higher. The opposite effects were obtained on using the low-temperature coke. The blast volume had to be lowered because of burden instability. Overall, the direct reduction factor fell by 3.2% with high-temperature coke and rose by 1.9% with the low-temperature grade. A cost effectiveness estimate on the blast-furnace use ofcoke made at higher flue temperatures and to higher finishing temperatures, without adjusting the carbonizing period, showed that iron smelting costs are reduced by 0.3%, because of the lower coke consumption and smelting costs.« less
  • Reactions of coke in the blast furnace are determined mainly by the nature of the coke itself which depends on the coal properties, its preparation and the coking conditions. In the blast furnace the coke reacts with alkali and alkaline earth metals in the burden. Preheated coal charges were found to absorb far less sodium and potassium carbonates than a conventional moist charge over an equivalent period, due to the smaller pore volume available.