Open Access Te Herenga Waka-Victoria University of Wellington
thesis_access.pdf (18.26 MB)

Reduction of New Zealand Titanomagnetite Ironsand by Hydrogen Gas in a Fluidised Bed System

Download (18.26 MB)
Version 2 2022-09-02, 01:31
Version 1 2021-12-09, 10:41
posted on 2022-09-02, 01:31 authored by Sigit Prabowo

Titanomagnetite (TTM) ironsand has been used to produce steel in New Zealand (NZ) for about 40 years. However, the current steelmaking process in NZ produces high emissions of CO2 because it uses coal as a primary reducing agent. The fluidised bed (FB) process allows the use of pure hydrogen gas to reduce ironsand, and as a result, does not produce CO2 gas. However, for conventional hematite ores, reduction in a FB system is usually limited by the onset of particle sticking at temperatures ≳ 800°C. This thesis investigates the reduction of NZ TTM ironsand by hydrogen gas in the FB system with a key focus on ore sticking behaviour.  Initially, this thesis reports preliminary fluidisation tests by nitrogen and helium gases at room temperature, carried out to determine key fluidisation parameters for ironsand powder. From these results, a laboratory-scale experimental FB reactor has been designed and built for the hydrogen reduction study at high temperatures. A key feature of the reactor is a novel in-situ sampling system, which enables extraction of multiple samples during a single experimental run without interrupting operation of the FB.  Quantitative X-ray diffraction (q-XRD) has been used to determine the metallisation degree of partially reduced samples. Phase evolution during the reaction has also been analysed using q-XRD alongside scanning electron microscopy/energy dispersed spectroscopy (SEM/EDS). Additionally, the water vapour compositions in the exhaust gas were calculated from the q-XRD data and also measured in real-time using a high-temperature humidity sensor.   The effect of various parameters has been investigated within the FB reduction experiments: hydrogen gas concentrations, hydrogen gas flow rate, bed mass, particle size, and temperature. The results indicate that across the entire range of controlled studied, the FB reduction rate of TTM ironsand is simply controlled by the rate of hydrogen gas supply. Interestingly, there were no occurrences of the sticking phenomenon at any point during the reduction by hydrogen gas at high temperatures of up to 1000°C. Sticking appears to be prevented by the formation of a protective titanium-rich oxide shell around each particle during the initial reduction stage. Importantly, this shell remains present throughout the reduction process, and as a result, the reduction reaction proceeds rapidly to completion with a metallisation degree of ~93%.  The influence of temperature on the reaction progress has also been investigated. The reduction pathway appears to vary within different temperature regimes. At low temperatures (750°C-800°C), TTM is directly reduced to metallic iron and ilmenite without any evidence of wüstite phase. At ‘intermediate temperatures’ (850°C-900°C) small amount of short-lived wüstite is observed. Some of the amount of TTM appears to be reduced to wüstite, and some is directly reduced to metallic iron. At high temperatures (≥ 950°C), approximately half of the initial TTM phase is quickly reduced to wüstite. After that point, wüstite is then reduced to metallic iron whilst the reduction of TTM stops. This is due to the enrichment of Ti species in TTM phase, which stabilises TTM crystal. Once wüstite has been fully reduced, the reduction of TTM then resumes.   Throughout the entire experimental program for this thesis, particle sticking was observed to occur only under two specific sets of experimental conditions. These were: reduction by 100% H2 gas at 1050°C (case A) and reduction by 7.5 mol.% H2O – 92.5 mol.% H2 at 950°C (case B). In both cases, sticking occurred as a sinter which nucleated at the reactor wall surface, while most particles remained fluidised as loose powder. The mechanism of these sticking cases has been analysed by XRD and SEM. The results suggest that silica from the quartz reactor wall reacted and bonded with Fe from particles to nucleate the initial sinter.   In summary, the findings in this thesis show that the hydrogen-FB process is highly effective in reducing NZ ironsand to a direct reduced iron (DRI) product. These findings open up the possibility of developing a new industrial FB technology for the direct reduction of NZ TTM ironsand, with extremely low CO2 emissions.


Copyright Date


Date of Award



Te Herenga Waka—Victoria University of Wellington

Rights License

Author Retains Copyright

Degree Discipline


Degree Grantor

Te Herenga Waka—Victoria University of Wellington

Degree Level


Degree Name

Doctor of Philosophy

Victoria University of Wellington Unit

Engineering at Victoria

ANZSRC Type Of Activity code


Victoria University of Wellington Item Type

Awarded Doctoral Thesis



Victoria University of Wellington School

School of Engineering and Computer Science


Bumby, Chris; Monaghan, Brian; Longbottom, Raymond