Faster Generation of Nanoporous Hematite Ore through Dehydration of Goethite under Vacuum Conditions

Abstract

Goethite (FeOOH)-based ore has become attractive to be utilized in ironmaking. As mildly dehydrated to remove its high combined water (CW), it changes to a nanoporous hematite ore. The nanopore contributes a significant increase in ore reduction reactivity due to the nanocontact between iron oxide and reducing agents such as C, CO, or H₂. However, the long dehydration time of goethite ore is still one problem. This study revealed the effect of vacuum condition on the mild-dehydration of goethite-based ore significantly reduces the ore dehydration time. The dehydration is finished within one hour at 300°C under high-vacuum condition (P=1.7 Pa), producing nanoporous hematite ore that is similar to under atmospheric one for 24 h. However, no significant decrease in the dehydration temperature under the high-vacuum condition. In contrast, in-situ heating TEM observation reveals that nanopore generation can occur at low temperatures under ultra-high vacuum conditions (P = 5.6×10⁻⁶ Pa). Slit pore generates at low temperatures that then eventually disappear by merging to bigger pores at the higher temperatures.

1. Introduction

Recently, goethite (FeOOH)-based ore has attractive potential to be utilized in ironmaking.¹⁾ As dehydrated to remove its high combined water (CW), goethite changes to nanoporous hematite.²⁾ The nanopore plays a crucial role in providing a nanocontact between iron oxide and reducing agents such as C, CO, or H₂, promoting a significant improvement in its reduction reactivity as well as its catalytic activities for fuel decomposition.^3,4,5,6,7) The previous study reported that when the goethite was dehydrated at 200–250°C, slit-like pores with a width less than 2 nm were formed along the [010] direction, as well as twin formation of hematite occurs.⁸⁾ Slit-like pores changed to spherical micropores (300–500°C) and eventually disappeared at higher temperatures. In the case of natural goethite, the optimum dehydration temperature for higher surface area and pore volume was 350°C, which was higher than that of 250°C for the synthetic goethite.⁹⁾ However, it was mentioned that the long time (24 h) is required for dehydration at 300°C.^6,7,9) No optimation have been done yet for the dehydration time, as it should be shortened to make it applicable for the industrial scale. One idea is increasing the driving force of the CW removal in the ore, decreasing the gas pressure (vacuum condition) will improve the dehydration performance. The purpose of the present study is to present our initial investigation of the effect vacuum condition on the dehydration of goethite.

2. Experimental Methods

2.1. Materials

FeOOH reagent and Australian Goethite A (GA) ore were used in this study. The properties of the materials are shown in Table 1.

Table 1. The properties of iron oxide and iron ore used as raw materials in this study.

Ore name	Code	CW [wt%]	T.Fe [wt%]	FeO [wt%]	SiO2 [wt%]	Al2O3 [wt%]
α-FeOOH reagent	FeOOH	10.1	62.9	–	–	–
Australian Ore A	GA	8.6	58.6	n/a	4.5	1.6

2.2. Vacuum Dehydration

Three vacuum conditions were introduced in this study: mild-, high-, and ultra-high-vacuum. The dehydration under a mild-vacuum condition was carried out in thermogravimetry (TG) equipped with a rotary vacuum pump. The FeOOH reagent sample was placed in the 150 μl alumina crucible TG sample pan. The illustration of the experimental apparatus of mild-vacuum dehydration (P = 28.8 kPa) is shown in Fig. 1(a).

Fig. 1.

The scheme of experimental apparatus for (a) mild-dehydration using thermogravimetric analysis (TGA) and (b) high-vacuum-dehydration using a degassing unit.

The high-vacuum dehydration experiment was conducted using a modified-Quantachrome Auto-degassing apparatus, which is schematically depicted in Fig. 1(b). The samples were put into a quartz bulb cell, then heated with an electrically controlled mantle heater, connected with a rotary vacuum pump (the vacuum pressure of 1.7 Pa). The final temperature was set at 300°C with a heating rate of 10°C min⁻¹. The cell was equipped with an in-situ pressure-temperature monitor. The samples after dehydration were soon analyzed using gas adsorption technique (Quantachrome Autosorb 6AG) to obtain the surface area, total pore volume, and pore size distribution.

The ultra-high vacuum dehydration (P = 5.6×10⁻⁶ Pa) was also conducted in this study; however, only the in-situ microstructural changes of the sample during heating from room temperature up to 400°C were observed using in-situ transmission electron microscopy (TEM) JEM-ARM1300 (JEOL). The heating rate was manually set at 10°C min⁻¹ and then hold for 10 min at the specific temperatures for observation. The summary of the different experimental conditions is summarized in Table 2.

Table 2. Summary of different experimental conditions on the goethite dehydration in this study.

Parameters	Experiments
Dehydration conditions	Atmospheric	Mild-vacuum	High-vacuum	Ultra-high-vacuum
Ultimate system pressure [Pa]	1.013×105	3.7×103	1.7	5.6×10−6
Experiment types	Atmospheric-thermogravimetry (TG); Atmospheric-furnace	Vacuum-thermogravimetry (TG)	Heating with in-situ pressure-temperature monitoring	In-situ TEM under heating
Sample applied	FeOOH reagent	FeOOH reagent	FeOOH reagent; GA ore	FeOOH reagent; GA ore
Sample mass [mg]	50 for TG; 500 for atmospheric furnace	50	250	<<1
Sample size	1 μm-FeOOH	1 μm-FeOOH; 1–2 mm-GA ore	1 μm-FeOOH; 1–2 mm-GA ore	200 nm-FeOOH; GA ore
Temp [°C]	RT to 300	RT to 300	RT to 300	RT to 400
Heating rate [K min−1]	10	10	10	10
Holding time	2 h TG; 24 h box furnace	2 h	1 h	10 min*
Obtained date	TG, DTG Pore analysis	TG, DTG	In-situ P-T, Pore analiysis	In-situ TEM images, electoron diffraction

*  The holding time in TEM observation is only to ensure the image stabilization before capturing TEM figure and electron diffractions.

3. Result and Discussions

3.1. Thermogravimetric Behavior of FeOOH Reagent under Mild-vacuum Dehydration

Figure 2 shows the comparison between the TG profiles of FeOOH reagent during dehydration under atmospheric and mild-vacuum conditions. Based on DTG peak profiles, under these different conditions, temperature shift was detected on FeOOH dehydration from 295.4°C (at atmospheric - air 100 ml/min) to 281.1°C (mild-vacuum 28.8 kPa) meaning that mild-vacuum can preliminary improve the kinetic of dehydration.

Fig. 2.

TG and DTG profiles of FeOOH reagent during dehydration under atmospheric and vacuum conditions (28.8 kPa).

3.2. Dehydration Behavior under High Vacuum Condition

Figure 3 shows the comparison of the dehydration performance between under high-vacuum (P = 1.7 Pa) and atmospheric dehydration process of FeOOH reagent. The dehydration performance, as CW Removal (CWR) in Fig. 3(a), means the removal of the combined H₂O from the FeOOH structure in the sample after dehydration, which is calculated using the following equation   

$$CWR=\frac{C{W}_{removed}}{C{W}_{initial}}\times 100\text{\%}$$(1)

  

$$CWR=\frac{(W_{s0}-W_{s1})}{0.101W_{s0}{X}_{FeOOH}}\times 100\text{\%}$$(2)

where W_s0 and W_s1 denote the mass of the FeOOH sample before and after dehydration, respectively [g]. X_FeOOH is the FeOOH content in the intial sample [-]. However, the data of FeOOH content in the case of iron ore was difficult to find. Therefore, this equation is merely applicable to FeOOH reagent only. The results of different dehydration conditions were also compared with the previous method (300°C, 101.3 kPa).⁹⁾ Interestingly, under high-vacuum condition (P = 1.7 Pa), the dehydration at 300°C for 1 h can remove 95 mass% of CW. It has a similar result with 24 h at the atmospheric one. Under vacuum conditions, porous hematite can be produced faster significantly. The Brunauer–Emmett–Teller (BET) surface area, total pore volume, and non-linear differential functional theory (NLDFT) pore size distribution of the FeOOH reagent samples before and after dehydration were observed as well in Fig. 3(b). The NLDFT pore size distribution shows that the pore size of 2–4 nm is generated within 1 h at 300°C under high-vacuum condition. The high-vacuum condition (P = 1.7 Pa) might not affect the main dehydration temperature because the significant CW removal still occurs at around 300°C. Vacuum dehydration only promotes a faster CW removal from oxide structure which is proceeded within 30 min after 300°C reached (in case of FeOOH reagent). In addition, XRD patterns of the FeOOH sample after dehydration under different conditions show that all FeOOH structure has already changed to Fe₂O₃ (Fig. 4).

Fig. 3.

Comparison of (a) dehydration performance and (b) pore properties between high-vacuum (P = 1.7 Pa) and atmospheric dehydration process of FeOOH reagent.

Fig. 4.

XRD patterns of FeOOH reagent and GA ore sample before and after dehydration under high-vacuum condition (P = 1.7 Pa).

Figure 5 shows the in-situ pressure-temperature profiles of FeOOH reagent and GA ore during dehydration at 300°C under high-vacuum condition. Both samples show the pressure increase along with the temperature rises from 100 to 300°C, meaning that the goethite dehydration, releasing water vapor, occurs rapidly in this temperature range. The total pressure increase is due to the water vapor released during heating. When the temperature is held at 300°C, the pressure drops back to its original state due to the dehydration already finished, with the FeOOH structures changes completely to Fe₂O₃. Interestingly, under the high-vacuum condition (P = 1.7 Pa), the dehydration seems to finish within 1 h at 300°C.

Fig. 5.

In-situ pressure-temperature profiles on high-vacuum (P = 1.7 Pa) dehydration of FeOOH reagent and GA ore.

Theoretically, two molecules of goethite decompose to become a molecule of hematite and a molecule of water (2FeOOH → Fe₂O₃ + H₂O_(g)). This decomposition reaction spontaneously starts at 83°C based on its thermodynamic properties. This decomposition process apparently consists of two steps: (1) solid-phase decomposition of FeOOH to Fe₂O₃ and adsorbed water (2FeOOH → Fe₂O₃ + H₂O_(ads)) which is an instant reaction; and (2) the adsorbed water removal as water vapor (H₂O_(ads) → H₂O_(g)) leaving the porous Fe₂O₃ structure. The mass transport phenomena of adsorbed water removal is typically similar to the drying process. A higher temperature is required to increase the overall rate in the actual goethite dehydration process. Furthermore, in general, the overall mass transfer of CW is driven by the concentration difference of water inside the pore and in the gas phase outside of pore. Thus, at the lower the gas pressure, the overall kinetic of the goethite dehydration increase significantly.

Figure 6 shows the comparison between the pore properties of iron oxide reagent and ores from high-vacuum dehydration at 300°C. The NLDFT pore size distribution shows the nanopores of 2 nm-pore sized are generated in FeOOH reagents within 30 min under vacuum condition. It then shifted to a larger pore size after longer holding time. This is also related to the BET surface area and total pore volume of FeOOH reagent, which is at 30 min higher than at 1 h. Meanwhile, in the case of natural ore, GA ore, after vacuum dehydration, the BET surface area and total pore volume are less than the FeOOH reagent. This is related to CW content in the ores, which are less than the FeOOH reagent. However, it looks like a long time is needed to generate nanopore in ores. XRD patterns of the FeOOH and GA ore dehydration (Fig. 7) also confirm that all FeOOH structures already change to Fe₂O₃ after dehydration, meaning that, as is heated, the goethite quickly changes to hematite structure.

Fig. 6.

Pore properties of iron oxide and GA ore from high-vacuum (P = 1.7 Pa) dehydration at 300°C.

Fig. 7.

XRD patterns of FeOOH reagent and GA ore sample before and after dehydration under high-vacuum condition (P = 1.7 Pa).

3.3. Dehydration Behavior under Ultra-high Vacuum Condition

The in-situ TEM results in Fig. 8 demonstrates the pore formation on FeOOH reagent and GA ore under ultra-high vacuum condition (P = 5.6×10⁻⁶ Pa). Interestingly, under this condition, goethite dehydration generating nanopore starts to occur at 100°C. The dehydration temperature seems closer to the ideal FeOOH dehydration temperature of 83°C. This is due to the driving force of CW removal under ultra-high vacuum conditions become much higher than other conditions. Small-mesopores are generated at 100–300°C then start to disappear significantly at 400°C by merging into the bigger pore size. In the case of natural iron ore (GA ore), the small-mesopore starts to be generated at 100°C. Most of the slit-shaped pores are generated at 200–300°C. At higher than 300°C, the slit-shaped pore starts to vanish, merging into a larger round-shaped (cylindrical) pore. Figure 9 shows the electron diffraction of FeOOH reagent and GA ore during dehydration under ultra-high vacuum condition (P = 5.6×10⁻⁶ Pa). In the case of FeOOH at RT, the electrons diffraction well matches to the FeOOH. At 100°C, although the electron diffraction still matches to FeOOH, a slight change in the diffraction pattern is observed. However, the FeOOH samples of 200–400°C were difficult for patterns analysis because the electron diffraction data were not clearly observed due to some drifting effect in TEM observation at higher temperatures. However, in the case of GA ore, it was observed that the diameter of the smallest ring increases, which reveals that at 200°C and above, the structure significantly changes from FeOOH to Fe₂O₃ structure. The electron diffraction data of 300°C agrees with the corresponding XRD data, although, the vacuum pressure was different. Finally, this study successfully demonstrated the effect of vacuum pressure on FeOOH dehydration, which offers a significant solution to reduce the time for low-grade ore preparation for future ironmaking purposes.

Fig. 8.

In-situ TEM images on ultra-high vacuum dehydration (P = 5.6 ×10⁻⁶ Pa) of FeOOH and GA ore. The white color represents the pore.

Fig. 9.

In-situ electron diffractions under TEM observation of FeOOH and GA ore at elevated temperatures. (Online version in color.)

4. Conclusion

Vacuum condition significantly shortens the time for dehydration of goethite-based ore generating nanoporous ore. The high-vacuum-dehydration (1 kPa) of goethite ore successfully generates the nanoporous hematite ore within one hour at 300°C. However, no significant changes in the dehydration temperature under the mild- and high-vacuum conditions. In contrast, using an in-situ TEM observation, under the ultra-high vacuum condition (P = 5.6×10⁻⁶ Pa), the dehydration process is significantly faster, and it has a possibility of lowering the dehydration temperature close to the theoretical ones. Small-mesopores are generated at 100–300°C then eventually vanished at higher temperatures by merging into the bigger pore size.

Acknowledgments

A part of this study was done under collaborative work with Nippon Steel Corporation. The pore size analysis was performed with Quantachrome Autosorb 6AG at Open Facility, Hokkaido University. The In-situ observation of microstructural change during heating processes was conducted using JEOL JEM-ARM-1300 at High-Voltage Electron Microscope Laboratory, Hokkaido University.

References

• 1)   Y.  Hata,  H.  Purwanto,  S.  Hosokai,  J.  Hayashi,  Y.  Kashiwaya and  T.  Akiyama: Energy Fuels, 23 (2009), 1128. https://doi.org/10.1021/ef800967h
• 2)   Y.  Kashiwaya and  T.  Akiyama: J. Nanomater., 2010 (2010), 235609. https://doi.org/10.1155/2010/235609
• 3)   K.  Miura,  K.  Miyabayashi,  M.  Kawanari and  R.  Ashida: ISIJ Int., 51 (2011), 1234. https://doi.org/10.2355/isijinternational.51.1234
• 4)   S.  Hosokai,  K.  Matsui,  N.  Okinaka,  K.  Ohno,  M.  Shimizu and  T.  Akiyama: Energy Fuels, 26 (2012), 7274. https://doi.org/10.1021/ef3013272
• 5)   R. B.  Cahyono,  G.  Saito,  N.  Yasuda,  T.  Nomura and  T.  Akiyama: Energy Fuels, 28 (2014), 2129. https://doi.org/10.1021/ef500201m
• 6)   A.  Kurniawan,  K.  Abe,  K.  Ohashi,  T.  Nomura and  T.  Akiyama: Fuel Process. Technol., 178 (2018), 156. https://doi.org/10.1016/j.fuproc.2018.05.034
• 7)   K.  Abe,  A.  Kurniawan,  K.  Ohashi,  T.  Nomura and  T.  Akiyama: ACS Omega, 3 (2018), 6151. https://doi.org/10.1021/acsomega.8b00958
• 8)   G.  Saito,  Y.  Kunisada,  T.  Nomura,  N.  Sakaguchi and  T.  Akiyama: Phys. Chem. Miner., 43 (2016), 749. https://doi.org/10.1007/s00269-016-0831-8
• 9)   G.  Saito,  T.  Nomura,  N.  Sakaguchi and  T.  Akiyama: ISIJ Int., 56 (2016), 1598. https://doi.org/10.2355/isijinternational.ISIJINT-2016-231