Interfacial Kinetics of Nitrogen Dissolution in Molten Fe–Mn–C Alloys Using ¹⁵N–¹⁴N Isotope Exchange Reaction

Abstract

The kinetics of the dissolution of nitrogen into molten Fe–Mn–C alloys through the isotope exchange reaction at 1673 K was investigated. The rate of nitrogen dissolution into a molten Fe–Mn–C alloy at low oxygen concentrations ([mass% O] < 0.001) indicated a first-order reaction with respect to the partial pressure of nitrogen. In a molten Fe–Mn–C alloy, Mn had a positive effect on the dissolution rate of nitrogen owing to a strong thermodynamic affinity with nitrogen, while carbon produced the opposite effect. The present work on nitrogen dissolution in molten Fe–Mn–C alloys was compared to previous studies on the carbon-free Fe–Mn binary alloy, providing empirical evidence for the proportionality of the rate constant to the square of the Mn alloying element activity. It should also be noted that nitrogen dissolution into the molten Fe–Mn–C alloy could be described using a parallel reaction model for the ideal mixing between Fe and Mn, which is also discussed in detail.

1. Introduction

The industrial importance of nitrogen dissolution in molten steels with high alloy additions has recently led to substantial interest in the kinetics of interfacial reactions in molten ferro-alloy systems. Several previous studies on the interfacial reaction of nitrogen on liquid steel have indicated that the rate of interfacial reaction can be reasonably estimated using Langmuir’s ideal isotherm model for a first-order reaction. The isotope exchange technique showed that the rate of nitrogen dissolution was strongly dependent on the surface coverage by surface active elements such as O,^1,2,6) S,^1,6) Se,²⁾ and Te,²⁾ which retarded nitrogen dissolution on the surface. Recent studies suggest that Al,^3,5) Si,^3,5) B,^3,6) and Cu⁵⁾ can also retard nitrogen dissolution, whereas Ti,⁴⁾ Zr,⁴⁾ V,⁴⁾ and Cr⁴⁾ can accelerate this dissolution, owing to an affinity with the nitrogen in liquid iron alloys.

Considering the increase on the stacking fault energy of TWIP steel with nitrogen content,⁷⁾ nitrogen dissolution into a Fe–Mn–C alloy should be controlled during the steelmaking process. Ono et al.⁵⁾ and Han et al.⁶⁾ estimated that the rate constant of nitrogen dissolution increases with increasing manganese content in a Fe–Mn binary alloy up to 15 mass%, since Mn has a stronger thermodynamic affinity with nitrogen than iron. Han et al.⁶⁾ suggested an increasing rate constant of nitrogen dissolution with manganese content that changes at approximately 6 mass% of Mn content, owing to changes in the liquid structure of the molten Fe–Mn alloy. Although fundamental studies on the interfacial reaction rates in several binary ferro-alloys have been conducted,^3,8) nitrogen dissolution into a molten Fe–Mn–C alloy has yet to be fully understood due to its high vapor pressures.

In the present study, the effect of manganese and carbon on the dissolution rate of nitrogen into molten Fe–Mn–C ternary system within a wide range of Mn and C at 1673 K has been investigated to clarify the interfacial dissolution behavior of nitrogen using the isotope exchange reaction.

2. Experimental Method

2.1. Principles of Isotope Exchange Reaction

Since the nitrogen dissolution reaction on the surface can be expressed by Eq. (1), its reaction rate, R (in mol/cm²·s), can be defined by Eq. (2), as follows:   

$$N_{\text{2}}\text{(}g\text{)}=\text{2}{\underset{\_}{N}}_{\text{(}in metal\text{)}}$$(1)

  

$$R=k\cdot p_{N_{\text{2}}}^m$$(2)

where k_overall (mol/cm²·s·atm) is the rate constant of nitrogen dissolution reaction into the molten metal and m is the reaction order in terms of the nitrogen partial pressure, p_N2 (atm), respectively. Using the ³⁰N₂ nitrogen gas isotope, nitrogen dissolution into the molten metal results in varying concentrations of ²⁸N₂, ²⁹N₂, and ³⁰N₂.^{5,6,8,9,10,11)} The nitrogen dissolution reaction after the adsorption of ³⁰N₂ and dissociation will result in a higher evolution of ²⁹N₂. The reaction rate constant of nitrogen dissolution on the metal surface is expressed by increasing ²⁹N₂, as follows:^1,6)   

$$-ln\frac{{\hspace{0.17em}}^{29}{F}_{eq}-{\hspace{0.17em}}^{29}{F}_f}{{\hspace{0.17em}}^{29}{F}_{eq}-{\hspace{0.17em}}^{29}{F}_i}=p_{N_{\text{2}}}^{m-\text{1}}k\frac{ART}{v\text{/}t}$$(3)

where ²⁹F_eq is the equilibrium fraction of the ²⁹N₂ species, and ²⁹F_i and ²⁹F_f are the fractions of ²⁹N₂ in the ingoing and outgoing gases, respectively. A (cm²) is the area of the liquid steel-gas interface, V (cm³) is the gas volume, and t (s) is the reaction time. If a complete mixing of the gas is assumed, V/t can be replaced by the volumetric flow rate, V. In the present study, nitrogen dissolution in Fe-70%Mn-1.6%C was found to be first-order from the results of preliminary tests, which is comparable to the pure iron surface,⁹⁾ therefore it may be assumed that m is equal to unity. The rate constant can therefore be expressed by the relation given in Eq. (4).   

$$k=-\frac{\overline{V}}{ART}ln\frac{{\hspace{0.17em}}^{29}{F}_{eq}-{\hspace{0.17em}}^{29}{F}_f}{{\hspace{0.17em}}^{29}{F}_{eq}-{\hspace{0.17em}}^{29}{F}_i}$$(4)

2.2. Experimental Arrangement

The experimental apparatus and employed procedure was similar to those used in a previous investigation of nitrogen isotope exchange reactions described in detail elsewhere.⁶⁾ A nitrogen isotope (98% ³⁰N₂, Cambridge Isotope laboratories, Inc., MA, USA) was mixed with UHP N₂ (99.999% ²⁸N₂) at a ratio of 3:97, where the partial pressure of the combined nitrogen was maintained at 0.3 atm using an Ar (99.9999%) carrier gas.

Reagent-grade carbon (99.999%), manganese chips (99%), and electrolytic iron were mixed in an Al₂O₃ crucible at 1673 K. The total oxygen was controlled below 10 ppm. A quadrupole type mass spectrometer (model HPR 20, HIDEN Analytical LTD, Warrington, UK) was used in order to analyze the fraction of each outgoing gas component.

After equilibration, the crucible was withdrawn from the furnace and rapidly quenched under Ar. Nitrogen and oxygen were measured using a LECO^® TC-300 analyzer, and carbon content was determined using a LECO^® CS-200 analyzer. Manganese was analyzed using X-ray fluorescence spectroscopy (XRF, S4 Explorer; Bruker AXS, Wisconsin, USA).

3. Results and Discussion

3.1. Rate-Determining Step of Nitrogen Dissolution

The elementary reaction involved in the dissolution of nitrogen into molten metal can be described as follows:   

$$N_{\text{2}}\text{(}g\text{)}+\hspace{0.17em}\square \hspace{0.17em}=N_{\text{2}}^{ad}\mathrm{\hspace{0.17em} }(\mathrm{A}\mathrm{d}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})$$(5-a)

  

$$N_{\text{2}}^{ad}+\hspace{0.17em}\square \hspace{0.17em}=\text{2}{N}^{ad}\mathrm{\hspace{0.17em} }(\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{o}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})$$(5-b)

  

$$N^{ad}={\underset{\_}{N}}_{(inFe)}\mathrm{\hspace{0.17em} }(\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})$$(5-c)

  

$$N^{ad}+N^{ad}=N_{\text{2}}^{ad}+\hspace{0.17em}\square \hspace{0.17em}\mathrm{\hspace{0.17em} }(\mathrm{A}\mathrm{s}\mathrm{s}\mathrm{o}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})$$(5-d)

  

$$N_{\text{2}}^{ad}=N_{\text{2}}\text{(}g\text{)}+\hspace{0.17em}\square \hspace{0.17em}\mathrm{\hspace{0.17em} }(\mathrm{D}\mathrm{e}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})$$(5-e)

where □ refers to a vacant site on the molten metal surface, and the superscript ‘ad’ denotes an adsorbed nitrogen atom or molecule on the vacant site. It is generally accepted that the nitrogen dissociation step (5-b) is the rate-determining step at low surface active element concentration, i.e. < 0.015 mass% O.²⁾ Since the reaction represented by Eq. (5-b) is rate-determining, all other reactions are at virtual equilibrium and the rate equation expressed in Eq. (2) that also encompasses the relations given in Eq. (5) can be modified to Eq. (6).^2,4,5,6)   

$$R=k\cdot p_{N_{\text{2}}}=k_{\text{5}-b}\cdot a_{N_{\text{2}}^{ad}}\cdot a_{\square }=k_{\text{5}-b}\cdot K_{\text{5}-a}^{eq}\cdot a_{\square }^2\cdot p_{N_{\text{2}}}$$(6)

  

$$k=k_{\text{5}-b}\cdot K_{\text{5-}a}^{eq}\cdot a_{\square }^2=k^{°}\cdot a_{\square }^2$$(7)

where a_□ is the activity of vacant site in pure liquid iron and k°(= $k_{\text{5}-b}\cdot K_{\text{5}-a}^{eq}$ ) is the forward rate constant of nitrogen dissolution on pure liquid iron.^4,6)

If molten Fe–Mn alloy, where Mn is substitutional element, was ideal mixing states between iron and Mn on the surface,⁶⁾ it can be reasonably assumed by Eq. (8) using a parallel reaction model for nitrogen adsorption on a molten Fe–Mn alloy.   

$$k_{Fe-Mn}=k_{Fe}^{°}\cdot a_{\square \mathrm{\hspace{0.17em} }on\mathrm{\hspace{0.17em} }Fe}^2+k_{Mn}^{°}\cdot a_{\square \mathrm{\hspace{0.17em} }on\mathrm{\hspace{0.17em} }Mn}^2$$(8)

where k_Fe–Mn, $k_{Fe}^{°}$ and $k_{Mn}^{°}$ are the rate constant of nitrogen dissolution on the liquid Fe–Mn alloy, pure liquid iron and pure liquid Mn, respectively. a_{□ on Fe} and a_{□ on Mn} are the activities of the vacant sites on iron and Mn, respectively.

3.2. Effects of Mn and C in Molten Steel on Rate Constant of Nitrogen Dissolution

In the Fe–Mn–C ternary system, the nitrogen dissolution rate on the bare surface considering the low initial oxygen content ([mass% O] < 0.001) and the stabilization of sulfur in the form of MnS¹¹⁾ can ignore the effect of oxygen and sulfur.

The effect of manganese in molten Fe–Mn–C on the rate constant of nitrogen dissolution at 1673 K, compared with previous studies,^3,4,5,6,8) is shown in Fig. 1. The rate constant increases with increasing Mn, V, and Cr content, each of which has a stronger affinity with nitrogen than iron.^13,14) The carbon in molten Fe–Mn–C alloy retarded the rate of nitrogen dissolution at various fixed manganese contents at 1673 K, as shown in Fig. 2. The effect of carbon on the rate constant of nitrogen dissolution was previously explained by its sole effect on the activity of nitrogen in the metal, thus implying that carbon typically had no effect on dissolution kinetics.¹⁴⁾ This was correlated to the negligible effects on the activity of the vacant site described in Eq. (7).^4,15) However, the addition of carbon to the molten Fe–Mn–C significantly influenced the rate constant of nitrogen dissolution on the surface of manganese, owing to its strong interaction with Mn. Since the effect of carbon on nitrogen activity is negligible, the rate constant of nitrogen dissolution, which was affected by the addition of carbon, should be considered in terms of effects on the activity of other elements within the molten Fe–Mn–C ternary system.

Fig. 1.

Dependence of the rate constant of nitrogen dissolution in molten Fe (–C) on additional element content (i = Mn, Si, Cr, V, Ni) at 1673 K.

Fig. 2.

Dependence of the rate constant of nitrogen dissolution in molten Fe–Mn–C ternary system on carbon content at 1673 K.

3.3. Relationship between Dissolution Rate of Nitrogen and Activities of Component Elements

Figure 3 shows the k_Fe–i–C/ $k_{Fe}^{°}$ to have a linear correlation with the square of the activity of i (i=Mn, Cr and Ni)^16,17) at the surface of Fe–i–C alloy under the assumption that the activity of the vacant site at the surface directly corresponds to the activities of the component elements.⁵⁾ Thus, the activity of the vacant site can be directly correlated to the activities of component elements at the surface, reasonably. From the information contained in Fig. 3, the dissolution rate of nitrogen at Fe–Mn–C alloy can be expressed in terms of the activities of component elements, as Eq. (9).   

$$k_{Fe-Mn-C}=k_{Fe}^{°}\cdot a_{Fe}^{\text{2}}+k_{Fe}^{°}\cdot \text{453}\cdot a_{Mn}^{\text{2}}\mathrm{\hspace{0.17em} }\text{(}mol\text{/}c{m}^{\text{2}}\cdot s\cdot atm\text{)}$$(9)

where $k_{Fe}^{°}$ , which is equal to 2.66×10⁻⁶ mole/cm²·s·atm, was determined by extrapolation, the results of the rate constant of nitrogen dissolution on Fe–C at 1673 K. 453 in Eq. (9) was determined by the slope in Fig. 3, which suggests the dissolution rate of nitrogen at the surface of pure manganese is approximately 453 times greater than the surface of pure iron. Therefore, the rate constant of nitrogen dissolution in the molten Fe–Mn–C ternary system using the parallel reaction model can be rearranged according to Eq. (10)   

$$k_{Fe-Mn-C}=\text{2}\text{.66}\times {\text{10}}^{-\text{6}}\cdot a_{Fe}^{\text{2}}+\text{1}\text{.2}\times {\text{10}}^{-\text{3}}\cdot a_{Mn}^{\text{2}}\mathrm{\hspace{0.17em} }\text{(}mol\text{/}c{m}^{\text{2}}\cdot s\cdot atm\text{)}$$(10)

where 1.2×10⁻³ mol/cm²·s·atm was found to be the rate constant of nitrogen dissolution in pure manganese at 1673 K. A comparison of the experimental results and the derived nitrogen dissolution model for the molten Fe–Mn–C ternary system seem to be in good agreement, as shown in Fig. 4.

Fig. 3.

The relationship k_Fe–i–C/k°_Fe (=a_□²) as a function of (a) a_i and (b) a_i² at 1673 K.

Fig. 4.

The estimated rate constant of nitrogen dissolution on molten Fe–Mn–C ternary system at 1673 K.

4. Conclusion

The effect of manganese and carbon on the dissolution rate of nitrogen into molten Fe–Mn–C ternary system has been investigated for a wide range of Mn and C contents at 1673 K. The resulting data was used to clarify the interfacial dissolution behavior of nitrogen using the isotope exchange reaction. The following conclusions were drawn:

(1) The rate of nitrogen dissolution into molten Fe–Mn–C ternary system subject to low oxygen concentration (mass% O] < 0.001) indicated a first-order reaction with respect to the partial pressure of nitrogen.

(2) Mn in the molten Fe–Mn–C alloy had a positive effect on the dissolution rate of nitrogen, owing to a strong thermodynamic affinity with dissolved nitrogen, while carbon showed the opposite effect.

(3) Since the activity of the vacant site, a_□, at the metal surface directly corresponds to the activity of the component elements on that surface, the rate constant of nitrogen dissolution on molten Fe–Mn–C ternary system at 1673 K can be expressed using the parallel reaction model.

Acknowledgments

This work was supported by the BK21 and the Industrial Strategy Technology Development (No. 10033389, Development of e-FERA Technology) through a grant provided by the Ministry of Knowledge Economy.

References

• 1)   M.  Byrne and  G. R.  Belton: Metall. Trans. B, 14B (1983), 441.
• 2)   H.  Ono,  H.  Fukazu,  K.  Morita and  N.  Sano: Metall. Trans. B, 27B (1996), 848.
• 3)   K.  Morita,  T.  Hirosumi and  N.  Sano: Metall. Trans. B, 31B (2000), 899.
• 4)   H.  Ono,  K.  Morita and  N.  Sano: Metall. Trans. B, 26B (1995), 991.
• 5)   H.  Ono,  T.  Koyama and  T.  Usui: ISIJ Int., 43 (2003), 298.
• 6)   S. M.  Han,  J. H.  Park,  S. M.  Jung,  J. H.  Park and  D. J.  Min: ISIJ Int., 49 (2009), 487.
• 7)   B. X.  Huang,  X. D.  Wang,  L.  Wang and  Y. H.  Rong: Metall. Trans. B, 39A (2008), 717.
• 8)   S. M.  Han: Ph D thesis, Yonsei University, (2009), 99.
• 9)   A.  Kobayashi,  F.  Tsukihashi and  N.  Sano: ISIJ Int., 33 (1993), 1131.
• 10)   S. M.  Han,  J. G.  Park and  I.  Sohn: J. Non-Cryst. Solids, 357 (2011), 2868.
• 11)   K. I.  Staffansson: Metall. Trans. B, 7B (1976), 131.
• 12)  The Japan Society for the Promotion of Science: The 19th Committee on Steelmaking, Steelmaking Data Source book, Gordon and Breach Science Publishers, New York, (1988), 285.
• 13)   W. Y.  Kim,  J. O.  Jo,  T. I.  Chung,  D. S.  Kim and  J. J.  Pak: ISIJ Int., 47 (2007), 1082.
• 14)   S.  Ban-ya,  F.  Ishii,  Y.  Iguchi and  T.  Nagasaka: Metall. Trans. B, 19B (1988), 233.
• 15)   P. C.  Glaws and  R. J.  Fruehan: Metall. Trans. B, 16B (1985), 551.
• 16)   W.  Dresler: Trans. Inst. Min. Metall. Sec. C, 98 (1989), 61.
• 17)   J. F.  Elliott,  M.  Gleiser and  V.  Ramakrishna: Thermochemistry for Steelmaking, Vol. II, Addison-Wesley Publishing Company, Inc., London, (1963), 503.