Effect of Carbonaceous Material Surface Texture on Iron Carburization Reaction under Loading Condition

Abstract

Scrap melting process is one of biggest energy required process in the ironmaking process with EAF. Carbon based scrap melting process has a potential problem about CO₂ emission. An aim of this study is finding an optimal surface texture of carbon source for scarp melting with reaction acceleration. Combinations between pure iron cylindrical block and several kinds of carbonaceous materials were evaluated with isothermal condition at 1673 K and keeping contact each other with loading condition. In order to avoid effect from object lower temperature range, rapid heating and quenching condition was applied to reach to 1673 K. As different surface texture carbonaceous material’s samples, coke, charcoal from eucalyptus and graphite were prepared. The coke needed obvious longer time to melt iron sample than other carbonaceous materials. It is indicated that coke ash has an obvious effect to prevent carbon transportation at reaction interface. Although charcoal has a better carbon structure for carburization reaction than graphite, the charcoal showed almost similar time for the melting as graphite. An effective contact area on interface of iron and carbon samples was estimated from surface observation by laser microscope and SEM-EDS. The effective area was decreased by existing of ash, porosity, and roughness. Decreasing of the effective contact area had obvious effect on carburization reaction as melting start temperature rising.

1. Introduction

Around 80% of Japanese crude steel production is coming from blast furnace and basic oxygen furnace route. However, this route of ironmaking causes a large amount of CO₂ emission as 13% of all over Japan.¹⁾ On the other hand, EAF steelmaking route has an advantage on mitigation of CO₂ emission because it does not include iron ore reduction process, though energy consumption for electric power generation should be counted. This advantage is provided from usage of scrap as iron resource. A scrap utilization technology should be immediately improved because of no natural resource in Japan and increasing of domestic accumulation of scrap.²⁾ Scrap-based ironmaking technology means not only EAF process but also other coal-based scrap melting root. For mitigation of CO₂ in the coal-based root, scrap melting process should be speed up in order that energy consumption would be decreased due to suppress excess energy. Scrap melting normally needs high temperature processing over 1600°C, because melting temperature of pure iron is 1538°C. This is one of the biggest reasons for large energy consumption in the scrap melting process. There is possibility to decrease the scrap melting temperature by carburization reaction. Increasing of carbon concertation in Fe–C alloy from 0 mass% to 4.3 mass% decrease the liquidas temperature from 1538°C to 1147°C.³⁾ Thus, effective carburization reaction is one of the ways for acceleration of scrap melting process. Carburization reaction is a carbon transport phenomenon from carbonaceous material to iron phase. This reaction has been researched by many researchers with many different aspects.^4,5,6,7,8) From the reports from them, it generally could be understood that higher fixed carbon and lower ash containing carbonaceous material has an advantage for the carburization reaction. And also, higher graphitization degree of fixed carbon makes obvious difficult to progress the reaction.⁵⁾ From these aspects, charcoal could be thought as one of the highly potential carbon sources.^6,7) However, it could not say these comparisons correctly evaluated reactivity difference of them because every carbonaceous material naturally has surface texture difference. For example, charcoal is “porous material”. Surface roughness and porosity directly effect on reaction interface area with iron phase. This point could have a strong effect at solid-solid reaction between carbonaceous material and iron phase in early stage of the carburization reaction. Contact condition of reaction interface with liquid formation, like iron melting due to carburization reaction, should change in case of no restraining force. In practical scrap melting process, scrap weight will be restraining force to keep contact between them. For keeping the constant contact condition during reaction, loading condition is applied in this study. In order to practically evaluate a carbon reactivity for carburization, an effective contact area on interface of iron and carbon was defined from surface texture observation by laser microscope and SEM-EDS. Iron melting start temperature due to carburization reaction under loading condition was discussed with the effective contact area for finding optimum condition of carbonaceous material.

2. Experimental Methodology

In order to investigate about the effect of carbon surface texture on iron carburization reaction, 3 kinds of carbonaceous materials, graphite, eucalyptus charcoal, and coke, were prepared. Amount of fixed carbon, volatile matter, and ash of them were given in Table 1. The volatile matters normally vaper at lower temperature than cohesive zone temperature around 1200°C. For excluding effect from volatile matters, the charcoal and coke were pre-treated to remove them at 1200°C for 3600 s under N₂ gas flow. After the pre-treatment, fixed carbon structure of each carbonaceous materials was evaluated by Raman spectroscopy. I_D/I_G ratio was applied for evaluation the carbon crystallinities of samples.⁹⁾ This ratio can be estimated from the spectrum intensity ratio between D band, 1350 cm⁻¹, and G band, 1575 cm⁻¹, of each carbon materials. The smaller ratio means graphitization degree is higher as shown in Table 1. Each carbonaceous material was sliced as 10 mm thickness substrate. Charcoal was cut by two kinds plane direction because plant has vascular bundles as illustrated in Fig. 1. Ash containing amount of coke is relatively higher than other materials. To control coke ash amount, ash removal treatment for sliced coke sample by soaking in acid solution carried out. The pre-heat-treated coke was soaked into HCl solution and HF solution during predetermined time for ash removal treatment.⁸⁾ Residues amount of ash were measured by gravimetric measurement after combustion at 1088 K under air atmosphere as JIS M 8812. The residues amount of ash after the acid treatment were dependent on duration of this treatment as shown in Fig. 2. After carbonizing wood, the vascular bundles made hollow pipes. “Charcoal A” was cut in vertical to the pipes. “Charcoal B” was cut in parallel to the pipes. This difference gave them different surface roughness. In order to unite surface roughness of carbon samples, they were polished until #3000 emery paper. In order to evaluate surface roughness of carbon samples, several evaluation methods were combined under following concept for effective contact area. Carbonaceous material surface is thought it consists of pore area, ash area, fixed carbon area as illustrated in Fig. 3. Occupation by the pore and ash area at the reaction interface decrease the effective contact area. The fixed carbon area naturally has surface roughness. These 3 kinds of areas were estimated from surface observation in 2 ways of magnifications. The pore area was identified with lower magnification observation, × 108, by confocal laser microscope (OLS4500, Olympus), and then remaining part assumed as matrix area consisting of ash and fixed carbon. The surface roughness of the matrix area, excluded pore area, was estimate with higher magnification observation, × 1080, by confocal laser microscope. This observation decides the effective contact area in the matrix area. The contact area included the ash area. In order to exclude the ash area from the matrix area, SEM-EDS observation (SU3500, Hitachi High Tech.) was carried out in middle magnitude, × 600, as shown in Fig. 4. From these information, the effective contact areas of each materials derived from Eq. (1) and the results are shown in Table 2.   

$$\begin{array}{l}\text{Effective}\mathrm{\hspace{0.17em} }\text{contact}\mathrm{\hspace{0.17em} }\text{area(\%)}=\text{100}\times \text{(1}-\text{Pore}\mathrm{\hspace{0.17em} }\text{ratio(}-\text{))}\\ \mathrm{\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{0.5em}}\mathrm{\hspace{0.17em} }\mathrm{\hspace{0.17em} }\times \text{(1}-\text{Ash}\mathrm{\hspace{0.17em} }\text{ratio(}-\text{))}\\ \mathrm{\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{0.5em}}\mathrm{\hspace{0.17em} }\mathrm{\hspace{0.17em} }\times \text{Actual}\mathrm{\hspace{0.17em} }\text{contact}\mathrm{\hspace{0.17em} }\text{area}\mathrm{\hspace{0.17em} }\text{ratio(}-\text{)}\end{array}$$(1)

Table 1. Main components and I_D/I_G values of carbon samples.

		Fixed Carbon (mass%)	Volatile matter (mass%)	Ash (mass%)	ID/IG (−)
Graphite	As raw	100	0	0
	Heat-treatment	100	–	0	0.29
Eucalyptus charcoal	As raw	79.3	17.7	3.0
	Heat-treatment	96.4	–	3.6	0.72
Coke	As raw	87.4	0.5	12.1
	Heat-treatment	87.9	–	12.1	0.58

Fig. 1.

Schematic illustration of 2 types of charcoal samples.

Fig. 2.

Remaining ash content in coke after ash removal treatment.

Fig. 3.

Schematic illustration of identification into 3 kinds areas, as pore, ash, and fixed carbon.

Fig. 4.

Schematic illustration of identification of ash region in matrix region with SEM-EDS.

Table 2. Surface texture evaluation results of carbon samples.

	Pore ratio (−)	Ash ratio (−)	Actual contact area ratio (−)
Graphite	0.036	0	0.684
Charcoal A	0.253	0.003	0.425
Charcoal B	0.324	0.007	0.574
Coke	0.693	0.074	0.445
Coke with ash removal 6 h	0.696	0.068	0.508
Coke with ash removal 24 h	0.711	0.042	0.491

Iron sample was obtained from a pure iron rod cutting into 5 mm thickness. The purity is 99.95+%, and the diameter is 9.5 mm. Combinations of 4 kinds carbon substrates and sliced iron rod were applied as specimens to compare difference of carburization reactivities among different surface texture carbonaceous materials. The specimen was put into graphite crucible with alumina and magnesia refractories as shown in Fig. 5. These blocks avoid an effect from graphite crucible and lid on the carburization reaction at the specimen’s interface. During reaction evaluation at high temperature, 0.1 MPa load was added from top of the lid in order to be keeping good contact between the iron and the carbon samples. The sample was set into a softening and melting simulator we developed.¹⁰⁾ The apparatus can rapidly heat and quench samples with high rate temperature control system, consisted of infrared furnace and quenching dry chamber, over 1000°C/min. In order to correctly evaluate reactivity of carbon samples toward iron, rapid heating and quenching condition was applied in this study. The sample was heated up to 1400°C with 1000°C/min and kept a certain time and quenched with 1000°C/min. The apparatus can measure a displacement amount of graphite lid on the assembly with time. This information gives the iron sample melting behaviour from thickness variation of it. Although iron melting due to carburization reaction surely consumes also carbon material and slightly decrease their thickness, the variation of carbon side was ignored in this study for simplify. From the displacement amount measurement, shrinkage degree of iron sample was derived with Eq. (2).   

$$\text{Shrinkage}\mathrm{\hspace{0.17em} }\text{degree(\%)}=\frac{Displacement\mathrm{\hspace{0.17em} }length\mathrm{\hspace{0.17em} }at\mathrm{\hspace{0.17em} }time t\left(mm\right)}{Maximum\mathrm{\hspace{0.17em} }Displacement\mathrm{\hspace{0.17em} }length\left(mm\right)}$$(2)

Fig. 5.

Schematic illustration of sample setting for softening and melting simulator.

Measured displacement lengths were considered about thermal expansion effect of samples and refractories from their individual measurement results without any reaction condition.

3. Results and Discussion

Figure 6(a) shows typical measurement result of shrinkage behaviour with time. 0 s in this study means heating start with infrared furnace from room temperature. Temperature tendency is shown with shrinkage degree in this figure. The shrinkage behaviour typically begins from before reach 1400°C and then after certain time reach to almost 100%. In order to precisely estimate the shrinking beginning and finish time, shrinkage rate was calculated from time tendency of shrinkage degree as shown in Fig. 6(b). When the shrinkage behaviour begins in certain time, the shrinkage rate will rapidly increase. This moment is defined as a shrinkage beginning time in this study, and a shrinkage beginning temperature can be estimated from the time with temperature tendency curve.

Fig. 6.

Shrinkage beginning temperature evaluation method from (a) shrinkage degree variation with time through (b) conversion to shrinkage rate with time.

Figure 7 shows comparison of shrinkage behaviours among 3 types of carbonaceous materials except coke. Result of graphite shows the lowest temperature of beginning of shrinkage behaviour and the largest shrinkage rate. Although a difference between Charcoal A and B is not obvious, Charcoal B shows slightly lower temperature of starting to shrink when the shrinkage beginning temperature is estimated. Results of coke are separately shown in Fig. 8 from other materials, because it needed very longer time than others to appear shrinkage behaviour. Results of ash removal coke results were shown in same figure. In case without ash removal treatment, the shrinkage degree could not reach to 100% even after 3000 s. The coke with longer time ash removal treatment needed shorter time to complete the shrinkage behaviour. Figure 9 shows relationship between ash content from Fig. 2 and shrinkage beginning temperatures of all samples. Although obviously negative effect of ash on the shrinking was showed in comparison among the cokes, graphite and charcoals could not show clear trend in this relationship. In order to evaluate effect of pores on the surface of each materials, effect of matrix area ratio, which were estimated from the lower magnification observation with Eq. (3), on the shrinkage beginning temperature is shown in Fig. 10.   

$$\text{Matrix}\mathrm{\hspace{0.17em} }\text{area(\%)}=\text{100}\times \text{(1}-\text{Pore}\mathrm{\hspace{0.17em} }\text{ratio(}-\text{))}$$(3)

Fig. 7.

Shrinkage degrees variations of graphite, charcoal A, and Charcoal B with time.

Fig. 8.

Effect of ash removal treatment on shrinkage degrees variations of cokes with time.

Fig. 9.

Effect of ash content in carbonaceous material on shrinkage beginning temperature.

Fig. 10.

Effect of matrix area ratio in carbonaceous material on shrinkage beginning temperature.

This simple comparison shows only difference between cokes and others, but it could not still clearly explain a reason of difference among graphite and 2 types of charcoals. In order to evaluate the difference among lower ash containing carbonaceous materials, the effective contact area was applied to compare the beginning temperatures as shown in Fig. 11. Difference between charcoals were clearly distinguished in this comparison. The bigger effective contact area gave the lower shrinkage beginning temperature. Unclear point on difference between charcoal B and graphite still remains through these evaluations. Although charcoal B has obviously the smaller effective contact area than graphite, they showed almost same temperature of the shrinkage beginning. It might be affected by fixed carbon structure difference. I_D/I_G values of experimental materials were shown in Table 1. If the value is large, the material would have large amount of random structure as not like as graphite. Our previous work indicated the random structure has an advantage for carburization reaction. From this point of view, it was thought fixed carbon structure of charcoal might have the advantage rather than graphite. However, it could not quantitatively explain difference from coke behaviour because it has middle value of I_D/I_G. Additionally, understanding of coke ash effect from the effective contact area will be complicate. Because their values are not clearly different among them. In case of comparison about coke ash amount as shown Fig. 9, coke’s results can show a certain trend. The quantitative relationship between fixed carbon structure and surface texture of carbonaceous material should be clarified in further next researches.

Fig. 11.

Relationship between effective contact area of each carbonaceous materials and shrinkage beginning temperature.

4. Conclusions

In order to find optimum surface texture of carbonaceous materials for practically advantageous on melting of bulky iron material, 4 kinds of carbon samples were evaluated as carbon sources by softening and melting simulator with rapid heating and quenching under loading condition. Following results were obtained.

Graphite has an obvious advantage for iron melting reaction rather than other materials in this experimental condition. It indicated smooth surface and low ash content have important roles for optimum material as carbon source.

Coke needed a quite longer time to melt iron sample than other materials. The disadvantage for iron melting came from coke ash, and ash removal treatment showed a positive effect on shrinkage beginning temperature due to carburization reaction.

The effective contact area could explain different reactivities among the experimental carbonaceous materials from the surface textures on the shrinkage behaviours except for difference between charcoals and graphite.

Acknowledgements

Authors express their gratitude to financial supports and scientific advices from the research group of SMART ironmaking system established in ISIJ.

References

• 1)   T.  Ariyama and  M.  Sato: ISIJ Int., 46 (2006), 1736.
• 2)   N.  Takamatsu,  K.  Yonezawa,  H.  Ueno,  W.  Takaki and  S.  Hayashi: Tetsu-to-Hagané, 100 (2014), 740 (in Japanese).
• 3)   H.  Baker,  H.  Okamoto,  S. D.  Henry,  G. M.  Davidson,  M. A.  Fleming,  L.  Kacprzak and  H. F.  Lampman: ASM Handbook, Volume 3: Alloy Phase Diagrams, ASM International, Materials Park, OH, (1992), 2_110.
• 4)   T.  Murakami and  K.  Nagata: Miner. Process. Extr. Metall. Rev., 24 (2003), 253.
• 5)   K.  Ohno,  T.  Maeda,  K.  Nishioka and  M.  Shimizu: ISIJ Int., 50 (2010), 53.
• 6)   M. W.  Chapman,  B. J.  Monaghan,  S. A.  Nightingale,  J. G.  Mathieson and  R. J.  Nightingale: ISIJ Int., 47 (2007), 973.
• 7)   K.  Ohno,  A.  Babich,  J.  Mitsue,  T.  Maeda,  D.  Senk,  H. W.  Gudenau and  M.  Shimizu: ISIJ Int., 52 (2012), 1482.
• 8)   K.  Ohno,  S.  Tsurumaru,  A.  Babich,  T.  Maeda,  D.  Senk,  H. W.  Gudenau and  K.  Kunitomo: ISIJ Int., 55 (2015), 1245.
• 9)   G.  Katagiri: Tanso, 183 (1998), 168 (in Japanese).
• 10)   A.  Kemppainen,  K.  Ohno,  M.  Iljana,  O.  Mattila,  T.  Paananen,  E.  Heikkinen,  T.  Maeda,  K.  Kunitomo and  T.  Fabritius: ISIJ Int., 55 (2015), 2039.