IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117

Please cite this article as: M. H. Hemmati, J. Vahdati Khaki, A. Zabett, An Investigation on Devolatilization of Non-Coking Coal and Non-Isothermal Reduction of Iron Oxide, TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117

International Journal of Engineering

J o u r n a l H o m e p a g e : w w w . i j e . i r

An Investigation on Devolatilization of Non-coking Coal and Non-isothermal Reduction of Iron Oxide

M. H. Hemmati a, J. Vahdati Khaki *a, A. Zabett b

a Department of Metallurgical Engineering, Iron and Steel Research Center, Ferdowsi University of Mashhad, Iran b Department of Metallurgical Engineering, Ferdowsi University of Mashhad, Iran

P A P E R I N F O

Paper history: Received 25 June 2014 Received in revised form 05 September 2014 Accepted 18 September 2014

Keywords: Volatile Matter Devolatilizaiton Non-isothermal Reduction Iron Oxide Non-coking Coal

A B S T R A C T

The devolatilization of a non-coking coal and the reduction of iron oxide fines by volatile matter (VM) were studied non-isothermally using thermogravimetry(TG) inargon atmosphere. The devolatilization of the coal showed five differentregions in terms of the rate of devolatilization. The maximum rate of devolatilization and the maximum weight loss occur between 640oC and 725oC furnace temperature.The effect of the heating rate and the coal particle size on devolatilizatoin were studied. Increasing the heating rate and theparticle size resulted in lower devolatilization.Non-isothermally reduction of iron oxide by VM in a multi-layered array was investigated. A reduction degree of 40 percent was reached while heating the pack from room temperature to 950oC.Three distinct regions of reduction were observed for reduction of Fe2O3. The XRD patterns were confirmed the stepwise reduction of iron oxide.

doi: 10.5829/idosi.ije.2015.28.01a.14

1. INTRODUCTION1

Attempts to innovate and develop direct reduction methods using solid-reductants have become an important prime in iron making. The main goal of these attempts is to replace the current pollutant methods with more environment-friendly ones. Several studies have been carried out to investigate the use of wood and coal as solid-reductant for iron oxide [1, 2]. In addition to the environmental consequences, however, their low performances challenge the applicability of these methods [3].Since volatile matter (VM) of coal is a source of pollution, its utilization decreases environmental problems while contributes to the process of reduction of iron oxide. Potential reducing ability of VM can be particularly effective at lower temperatures [4].

When coal is heated, a substantial weight loss occurs because of the evolution of volatile matter as shown below [5]: 11*Corresponding Author’s Email: vahdati@um.ac.ir (J. Vahdati Khaki)

Coal Coal char + Volatile Matter (hydrogen, water, hydrocarbons)

From kinetics and thermodynamics perspective, the reduction of iron oxide by coal has been the focus of several laboratory investigations [6-8]. In most of these works a mixture of coal and iron oxide as a composite pellet has been used for the study. They have rarely examined the role of VM on the reduction of iron ore systematically [4, 9-12].Konishi et al. [13] studied the effect of residual VM on reduction of iron oxide in composite pellet and concluded that the VM have an important role in the reduction at low temperature. Most of the studies have been carried out isothermally [14-16] and a few papers published on non-isothermal reduction [17-19].

In the present study, the devolatilization behavior of the coal was investigated. The role of VM on the reduction was examined in non-isothermal conditions. There are some advantages in non-isothermal experiment as they are more closely related to industrial condition. The period of time which iron oxide is exposed to VM at low temperature is very important in

113 M. H. Hemmati et al./ IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117

non-isothermal reduction and a main portion of the reduction is achieved in this stage. 2. MATERIALS METHOD A high-purity hematite powder with particle size of minus 53 micron was used in this study. The chemical composition of the hematite is presented in Table 1. Anon-coking coal was used with three different ranges in particle size-53micron, 149 to 177 and 840 to 1680 microns. The proximate analysis of the coal is given in Table 2.

A thermal gravimeter furnace (TGF) enable to use a 80 mm-height crucible was designed for the purpose of this study. The TGF was equipped with a weighting system enable to weight up to 500g with an accuracy of 0.01g.Figure 1 shows the schematic design of TGF. The thermocouple is placed close to the crucible as shown in the figure but it is important to note that the temperatures reported are the furnace temperature. TABLE 1. Chemical composition and particle size of iron oxide.

Mesh Fe2O3 CaO Al2O3 MgO SiO2 P2O5 Na2O

270 - 98.8 0.05 0.03 0.21 0.03 0.05 0.01

TABLE 2. Proximate analysis of coal Property moisture Ash Volatile

matter Fixed

carbon Weight percent

0.5 13 38 49

Figure 1. Schematic of thermogravimetery(TG)

Figure (a). Two-layered array for coal devolatilization study.(b) four-layered array for reduction study

The experiments were designed so that a multi-layered array could be used to study the reduction of iron oxide by VM. Figure 2a and b show the array of materials for the experiments; devolatilization behavior of coal (a) and the reduction of iron oxide by VM (b). As can be seen asplitting layer of alumina was used between the coal and the iron oxide to avoid a direct contact. Therefore, only the VM was acted as the reductant in the reduction process. In the sample used for devolatilization, enough alumina was used so that the total height of the array was equal to the one in the multi-layered array for the reduction of iron oxide. The crucible was made of stainless steel A310.

The crucible is placed in the furnace at room temperature. The furnace is heated with a constant power. The temperature-time diagram can be seen in Figure 3. The sampling rate for the temperature and weight was 1 Hz. The devolatilization percent and the reduction degree are calculated as follows: For devolatilization experiments: ∆WC = W2 – W1 (1)

( )( )

( )

% 100 CC

C

WW

W∆

∆ = ×

(2)

where W1 and W2respectively represent the weights of crucible before and after the experiment. WC is the total weight of the coal, ∆WC is the weight loss of coal and %∆W(c) is devolatilization percent. For the reduction experiments:

(3) ( )2 3 3 4 – Fe O CW W W+∆ = (4) ( ) ( ) ( )2 3 2 3 CFe O Fe O CW W W+∆ = ∆ −∆ (5) ( )2 2 3 2 3( ) (48 /160) O Fe O Fe OW Wη= ×

In which W3 and W4 are the weights of crucible before and after the reduction experiment, respectively. W(Fe2O3) is the weight of hematite, η(Fe2O3) is the purity of hematite, ∆W(Fe2O3+C) is the weight loss of multi-layered array crucible, ∆W(Fe2O3) is the weight loss of iron oxide layer, and WO2 is the total weight of the oxygen in the iron oxide. Finally, the degree of reduction is calculated by Equation (6):

%R.D. =[∆W(Fe2O3)/WO2]×100 (6)

M. H. Hemmati et al. / IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117 114

Figure 3. Diagram of temperature versus time of TGF.

Figure 4. Devolatilization of coal and the rate of devolatilization

TABLE 3. Comparison between devolatilization rate and weight loss at difference temperature ranges. region Lower

Temp. ( °C)

Upper Temp. ( °C)

∆T (°C)

Weight loss (%)

Dev .Rate

(%/°C)

a 350 500 150 2.5 1.92

b 500 640 140 10.5 4.83

c 640 725 85 16 5.36

d 725 825 100 7.5 1.67

e 825 950 125 3.5 0.86

3.RESULTS AND DISCUSSION 3. 1. Devolatilization of Coal Figure 4 shows the devolatilization behavior of the coal. The devolatilization begins at a temperature between 300°Cand400°C. Five different regions can be distinguished in Figure 4.From 350°C to 500°Cthe devolatilization rate is increasing,above500°C the rate increases much faster. The maximum rate of devolatilization is reached at 640°C.The rate is decreased with sharp slope at region (c) while it is almost constantbetween725°Cand 825°C. After this

constant rate of devolatilization the rate is decreased until all VM is evolved.

Most of the devolatilization occurs between 500°C and 725 °C (region b and c) which forms 69% of the total weight loss(Table 3). Soot is observed in the flue gas in region b which is indication of heavy hydro-carbon evolution at this range of temperature. Casel et. al [20] reported that light volatiles such as H2,CO,CO2 and H2O are released at lower temperatures. At a temperature above 640°C, the heavy hydrocarbons are cracked and light hydrocarbons are evolved. Nilson et al. [21] reported similar observation in their study.

3. 2. The Effect of Particle Size on Devolatilization As it can be seen in Figures 5(a) and (b), decreasing the particle size from 1680 micron to -53 micron increases the devolatilization as well as devolatilization rate. At higher temperatures the devolatilization rate of the coal for different particle sizes converges.

As devoltilization is a thermal activated process it can be promoted by any changes that enhance the heat transfer. Devoltilization generally involves three stages: (a) Heat transfer to coal particles. (b) Chemical reaction and devolatilization as a thermal decomposition process in which VM must be evolved from molecular structure. (c) Mass transfer of VM from coal particles [21]. Finer particles of the coal have faster heat transfer and higher surface energy. Hence, the devolatilization is faster and decomposition reactions occur more readily (Figure5a). 3. 3. The Effect of Heating Rate on Devolatilization Figure 6(a) shows the effect of heating rate on devolatilization. At lower rates, the deovlatilization begins at a lower temperature and the weight loss is more. Higher temperature leads to convergence of the devolatilization rates.

Duration of devolatilization at lower heating rate is longer than the one at higher heating rate for the same range of temperature. This causes a lower temperature gradient inside the coal for the lower heating rate. The lower temperature gradient and the longer time would result in more devolatilization as can be seen in figure 6(b).This fact is overexposed of course when the results is presented in terms of temperature changes. It is shown in Figure 6(b) that the maximum differences between the weight loss in two heating rates occurs close to the peak of devolatilization which confirms the coal devolatilization is a thermal activated process. At higher temperatures, the difference is lowered.

3. 4. Reduction by VM Figure 7 shows the reduction of iron oxide by VM in argon. The maximum reduction degree obtained by the VM-reductant is 40%. The curve of the reduction degree can be divided into three different regions. In the first region 11% reduction degree is reached. This stage is related to the reduction of Fe2O3 to Fe3O4. The third region corresponds with the

115 M. H. Hemmati et al./ IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117

reduction of Fe3O4 to FeO which begins at a higher temperature of the pack. The second region between the two distinct reduction stages is a period of heating time after completion of Fe2O3 reduction and before the beginning for Fe3O4 reduction.

Figure 5(a). Effect of particle size on weight loss of coal during devolatilization.(b)rate of devolatilization for different particle size

Figure 6(a). Effect of heating rate on devolatilization (b) the rate of devolatilization for two heating rate.

Figure 7. Reduction of iron oxide by the VM in argon

Figure 8. XRD patteren of (I)iron oxide (II) 11percent dergree of reduction corresponded to Figure 7(II) it contains:Fe2O3,Fe3O4. (III) 40percent degree of reduction corresponded to Figure 7(III) it contains FeO and Fe. According to the reactions:

3Fe2O3 + H2 = 2Fe3O4 + H2O

(1) ∆G0 = -RT ln(PH2O/PH2)

3Fe2O3 + CO = 2Fe3O4 + CO2

(2) ∆G0 = -RT ln(PCO2/PCO)

From 500°C to 800°C the Gibbs free energy is negative and the equilibrium ratios of H2O to H2 or CO2 to CO are large. So that, the reduction occurs even at low hydrogen or carbon monoxide partial pressures.

The rate of the reduction shows four different regions. Two peaks can be seen on the rate of the reduction curve. The first peak occurs at about 5% of the reduction degree and the rate decreases until the end of the reduction of Fe2O3 to Fe3O4.

Figure 8shows the XRD pattern of iron oxides for three points labeled on Figure 7. The point (I) is the hematite. Point (II) contains Fe3O4with some remaining Fe2O3.Point (III) contains FeO and Fe which is related to 40% reduction degree. The results show the reduction process is stepwise.

4. CONCLUSION

The devolatilization process showed five different regions in terms of the weight loss. The maximum rate of devolatilization and the maximum weight loss occured between 640oC and 725oC. Devolatilization was affected by the particle size and the heating rate. Finer particles had high surface energy and higher heat transfer which result in higher devolatilization. Lower rates of heating highly increased the devolatilization due to longer period of time and lower temperature gradient inside the coal pack.

M. H. Hemmati et al. / IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117 116

A reduction degree of 45 percent was obtained by utilizing VM in a non-isothermal heating condition up to 950oC. Reduction of iron oxide by VM at a multi-layered array was influenced by thermodynamics and kinetics of the iron oxide reduction. Devolatilization of the non-coking coal and the reduction of the iron oxide are both thermal activated processes which can be greatly affected by heat transfer. It can be concluded that the most probable rate-controlling step in both volatilization of the coal and the reduction of the iron oxide by VM is the heat transfer to the materials.

5. REFERENCES

1. Fortini, O., "Renewable energy steelmaking: On a new process for ironmaking", (2004).

2. Gupta, R., "Woodchar as a sustainable reductant for ironmaking in the 21st century", Mineral Processing Extractive Metall. Rev., Vol. 24, No. 3-4, (2003), 203-231.

3. Shivaramakrishna, N., Agrawal, B., Ray, A., Prasad, K., Bandopadhyay, P. and Gupta, S., "Production of high carbon sponge iron from ore--coal composite pellets", Transactions of the Indian Institute of Metals, Vol. 43, No. 2, (1990), 91-101.

4. Dey, S.K., BISWANATH, J. and Basumallick, A., "Kinetics and reduction characteristics of hematite-noncoking coal mixed pellets under nitrogen gas atmosphere", ISIJ international, Vol. 33, No. 7, (1993), 735-739.

5. Sampaio, R., Coal devolatilization in bath smelting slags, (1990), Carnegie Mellon University PhD thesis, Pittsburgh.

6. Sun, K. and Lu, W.-K., "Mathematical modeling of the kinetics of carbothermic reduction of iron oxides in ore-coal composite pellets", Metallurgical and Materials Transactions B, Vol. 40, No. 1, (2009), 91-103.

7. Sadrnezhaad, K., "Direct reduced iron: An advantageous charge material for induction furnace", Journal of Engineering, (1988), 1-2.

8. Towhidi, N., "Reoxidation rate of sponge iron pellets, briquettes and iron powder compressed to various compressions in air", International Journal of Engineering, Vol. 1, No. 2&3, 111-116.

9. Sharma, T., "Non-coking coal quality and composite pre-

reduced pellets", International Journal of Mineral Processing, Vol. 39, No. 3, (1993), 299-311.

10. Kamijo, C., Hoshi, M., Kawaguchi, T., Yamaoka, H. and Kamei, Y., "Production of direct reduced iron by a sheet material inserting metallization method", ISIJ International, Vol. 41, No., (2001), S13-S16.

11. Gudenau, H.W., Senk, D., Wang, S., DE MELO MARTINS, K. and Stephany, C., "Research in the reduction of iron ore agglomerates including coal and c-containing dust", ISIJ International, Vol. 45, No. 4, (2005), 603-608.

12. Bryk, C. and Lu, W., "Reduction phenomena in composites of iron-ore concentrates and coals", Ironmaking & Steelmaking, Vol. 13, No. 2, (1986), 70-75.

13. Konishi, H., Yamashita, A. and Usui, T., "Effect of residual volatile matter on reduction of iron oxide in carbon composite pellets", ISIJ International, Vol. 8, (2008), s142-s146.

14. Mookherjee, S., Ray, H. and Mukherjee, A., "Thermogravimetric studies on the reduction of hematite ore fines by a surrounding layer of coal or char fines: Part 1. Isothermal kinetic studies", Thermochimica Acta, Vol. 95, No. 1, (1985), 235-246.

15. Huang, B.-H. and Lu, W.-K., "Kinetics and mechanisms of reactions in iron ore/coal composites", ISIJ International, Vol. 33, No. 10, (1993), 1055-1061.

16. Seaton, C.E., Foster, J.S. and Velasco, J., "Structural changes occurring during reduction of hematite and magnetite pellets containing coal char", Transactions of the Iron and Steel Institute of Japan, Vol. 23, No. 6, (1983), 497-503.

17. Mookherjee, S., Ray, H. and Mukherjee, A., "Thermogravimetric studies on the reduction of hematite ore fines by a surrounding layer of coal or char fines: Part 2. Non-isothermal kinetic studies", Thermochimica Acta, Vol. 95, No. 1, (1985), 247-256.

18. Prakash, S. and Ray, H.S., "Reduction of iron ore under rising temperature and fluctuating temperature conditions", Thermochimica Acta, Vol. 111, (1987), 143-166.

19. Dutta, S. and Ghosh, A., "Study of nonisothermal reduction of iron ore-coal/char composite pellet", Metallurgical and Materials Transactions B, Vol. 25, No. 1, (1994), 15-26.

20. Casal, M., Díez, M., Alvarez, R. and Barriocanal, C., "Primary tar of different coking coal ranks", International Journal of Coal Geology, Vol. 76, No. 3, (2008), 237-242.

21. Nelson, P.F., Smith, I.W., Tyler, R.J. and Mackie, J.C., "Pyrolysis of coal at high temperatures", Energy & Fuels, Vol. 2, No. 4, (1988), 391-400.

M. H. Hemmati et al. / IJE TRANSACTIONS A: Basics Vol. 28, No. 1, (January 2015) 112-117 1940

An Investigation on Devolatilization of Non-coking Coal and Non-isothermal Reduction of Iron Oxide

M. H. Hemmati a, J. Vahdati Khaki a, A. Zabett b

a Department of Metallurgical Engineering, Iron and Steel Research Center, Ferdowsi University of Mashhad, Iran b Department of Metallurgical Engineering, Ferdowsi University of Mashhad, Iran

P A P E R I N F O

Paper history: Received 25 June 2014 Received in revised form 05 September 2014 Accepted 18 September 2014

Keywords: Volatile Matter Devolatilizaiton Non-isothermal Reduction Iron Oxide Non-coking Coal

چکیده

تحلیل ي کورهدما در یک اکسید آهن توسط مواد فرار در شرایط غیرهم نشو و احیاي فرایند خروج مواد فرار زغال سنگ ککمنحنی خروج مواد فرار از لحاظ میزان خروج مواد فرار و نرخ . ثقلی و در محیط آرگن مورد پژوهش قرار گرفت-حرارتیگراد ي سانتی درجه 725تا 640ي دمایی ي خروج مواد فرار در بازه بیشینه. مجزا تشکیل شده است ي ها از پنج منطقه خروج آن

افزایش . ي ذرات زغال سنگ و نرخ گرمایش زغال بر منحنی خروج مواد فرار مطالعه گردید اثر اندازه. در کوره حاصل شد. ن خروج مواد فرار شد ولی از تأثیر این کاهش با افزایش دما کاسته شدي ذرات موجب کاهش میزا سرعت گرمایش و اندازه

آرایش چند الیه، در .مورد استفاده قرار گرفت آرایش چند الیه براي بررسی احیاي غیرهمدماي اکسیدآهن توسط مواد فرار. بودهاي احیا ینامیک واکنشو ترمود نرخ احیا وابسته به دما. دما به دست آمد مدرصد به صورت غیره 40ي احیاي درجه

.اي بودن فرایند احیا را در دماهاي مختلف نشان داد مرحلهي ایکس الگوهاي پراش اشعه

doi: 10.5829/idosi.ije.2014.28.01a.14