3. Results
3.1. Strength Tests
The following strength test results were obtained from the AMCOM GROUP LLC laboratory. The crushing strength was tested for seven briquette samples 67–81 mm in length. The average crushing strength in normal conditions was 24.64 kg/cm as seen in Table 4. The briquettes used in the mechanical drop strength test were sieved after the test, and 98.5% of the briquettes were more than 5 mm in size. The abrasion strengths for briquette fractions examined after 25, 50, 100, and 200 revolutions, defined as portions exceeding 5 mm in size, were 93%, 89%, 80%, and 64%, respectively. The abrasion strength test results are seen in Table 5.

Table 4. Mechanical crushing strength test results with calculated averages (AVG) and standard deviations (SD).

	Length (mm)	Force (kg)	Strength (kg/cm)	Strength (kg/cm2)
	77.0	183.0	23.8	196.3
	81.0	192.0	23.7	195.4
	75.0	184.0	24.5	202.1
	68.0	170.0	25.0	206.2
	73.0	189.0	25.9	213.6
	73.0	198.0	27.1	223.5
	67.0	151.0	22.5	185.6
AVG	73.4	181.0	24.6	203.2
SD	4.9	15.9	1.5	12.6

 

Table 5. Results from abrasion strength test carried out in a rotating drum.

Number of Revolutions	Content (%)
	>5 mm	>0.5 mm
25	93	95
50	89	90
100	80	82
200	64	69

3.2. Reduction Experiments for Auger Pressing Briquettes
The relative weight losses that occurred during the Experiments A–F are shown in Figure 7. Swelling of the auger pressing briquettes and the reference briquette during Experiments A–D and F are shown in Figure 8.

 

Figure 7. Weight losses (%) of the auger pressing briquettes (Experiments A–D), the pellets (Experiment E), and the reference briquette (Experiment F).

 

Figure 8. Swelling (%) of the auger pressing briquettes (Experiments A–D) and the reference briquette (Experiment F). Swelling of the pellets (Experiment E) was not studied.

Based on Figure 7 and Figure 8, weight loss appears to be most significant in the case of auger pressing briquettes during Experiments A and B, and the reference briquette swelled more sensitively than the auger pressing briquette. TGA data provide more detailed information on the time of the weight losses occurred. The weight change curves for each experiment as a function of time together with the temperature curve are demonstrated in Figure 9.

 

Figure 9. Weight change (%) curves as a function of time in the reduction experiments for the auger pressing briquettes (Experiments A–D), the pellets (Experiment E), and the reference briquette (Experiment F).

3.2.1. Uninterrupted Reduction Experiments
In uninterrupted experiments, referred to as Experiments A and B, reduction was performed up to the iron segment through all reduction stages presented in more detail below. As the last chemical reaction in the experiment, wüstite was reduced to metallic iron according to Equation (3).

As seen from Figure 7, the relative weight losses during Experiments A and B are almost equal: 37.0% and 35.9%. The weight loss curves for auger pressing briquettes seen in Figure 9 overlap almost completely. It can be concluded that the 40 min isothermal period did not have a major effect on weight change, although the weight decreased slightly during it. Thus, the iron contained in the briquette can be assumed to be completely reduced (RD = 100%) and the rest of the weight loss to be due to coal gasification.

At about 1000 °C, the precipitation of a white substance was observed on the glass of the lid of the reduction tube and on the stems of the sample basket. This was apparently zinc excretion. As mentioned above, the circulation behavior of zinc with a quite low boiling point of 908 °C is known to be a nuisance in the blast furnace process. Zinc is likely to be present as a compound, such as zinc oxide (ZnO). Of the samples used, the phenomenon occurred only in the case of auger pressing briquettes with a zinc content of 0.17 wt.-%. Other samples contained significantly less zinc.

The swelling of the briquettes was slight. The results for Experiment A and B were 5.8% and 10.7%, and the swelling behavior was difficult to detect visually. Its occurrence could not be detected during the reduction tests despite the camera running continuously. Instead of swelling, the briquette seemed to roll in the basket due to the rise in temperature. Minor changes were observed in the external dimensions of the briquette samples as they were measured before and after the experiments. However, the briquette samples seen in Figure 10 show similar cracking behavior during Experiments A and B, indicating that no significant cracking or other external changes occurred during the 40 min isothermal period at 1100 °C.

 

Figure 10. Cracks formed in the auger pressing briquette sample during (a) Experiment B and (b) Experiment A, i.e., the run with the isothermal period of 40 min.

3.2.2. Interrupted Reduction Experiments
In interrupted Experiments C and D, the relative weight losses were lower: 7.5% and 1.8%. Based on the weight losses, the reduction was very slow at these reduction stages, although it was faster with the auger pressing briquettes than with the reference samples. At first, reduction of hematite to magnetite occurred in both experiments according to Equation (4).

In Experiment C, the reduction of magnetite to wüstite was achieved according to Equation (5).

The swelling observed during Experiments C and D, about 1.2% and 0.5%, respectively, was so minor that the differences in measurements could be due to measurement inaccuracies. The measurement was hampered by the uneven shape of the briquette ends. No visually detectable changes were observed, except for a slight discoloration of the briquette in Experiment C. That is, as a result of magnetite being reduced to wüstite.

3.3. Reduction Experiments for Reference Samples
In Experiments E and F, the relative weight losses for the iron ore pellets and the reference briquette were 13.0% and 24.8%, respectively. It can be seen from Figure 9 that the reduction behavior differs from auger pressing briquettes. The pellets began to lose weight at just below 500 °C, which is later than for the other samples. At this point, it can be assumed that the reduction of hematite to magnetite occurred. The pellet started to react more effectively when the temperature approached 1100 °C. The weight losses that occurred in the beginning of the Experiments A–D and F are discussed in more detail in Section 4. The RD was calculated for the iron ore pellet based on the weight loss in Experiment E by utilizing Equation (1) and the chemical analysis of the pellet shown in Table 2. The RD of the pellet was 45.3%. No cracking was observed in the pellets.

Compared to the auger pressing briquette, the reference briquette lost relatively more weight during the isothermal period although the overall weight loss was smaller. The carbon content is lower than in the auger pressing briquette, so the sample apparently was not completely reduced. Weight loss was greatest above 950 °C. Figure 11 shows the swelling behavior of the reference briquette. The most significant swelling took place above 1000 °C. More swelling occurred than in the case of the auger pressing briquette: 23.7%. No cracking was observed. The results are in line with a similar study on punch-and-die industrial briquettes carried out by Kemppainen et al.

Figure 11. Swelling of the reference briquette with increasing temperature during Experiment F.