Effect of reaction atmosphere on combustion synthesis of Ti 2 AlN

Abstract

Combustion synthesis of Ti₂AlN was carried out in air and pressurised N₂ respectively. The reaction atmosphere had an evident effect on the phase composition and microstructure of the products. By combustion synthesis in 2 MPa N₂, Ti₂AlN was synthesised as the major product, but in the samples prepared in air, Ti₂AlN was obtained only as a minor phase, and the predominant phases were Ti–Al intermetallics. The reaction mechanism has been studied, suggesting that TiN is a key intermediate in the formation of Ti₂AlN. It is proposed that, in the combustion synthesis of Ti₂AlN, the flux of N₂ plays an important role in determining both reaction kinetics and phase composition of the products.

Keywords

Combustion synthesis Reaction atmosphere

Introduction

Ternary carbide or nitride compounds are a new class of inorganic materials with merits of both ceramics and metals, such as low density, high thermal and electrical conductivity, good resistance against chemical corrosion and oxidation and easy machinability.1^–3 These ternary compounds have a general chemical formula of M_n+1AX_n, where n = 1, 2 or 3, M is an early transition metal, A is an A group (mostly IIIA and IVA) element and X is C or N, and a layered hexagonal lattice structure with two formula per unit cell. In each unit cell, near close packed M layers are interleaved by A layers, and X atoms fill in the octahedral interstitial sites of M octahedra. The M₆X octahedra are edge sharing and identical to those in the rock salt structure of the corresponding binary carbides.

Among the M_n+1AX_n compounds, several carbide phases like Ti₃SiC₂, Ti₃AlC₂ and Ti₂AlC have been extensively studied,4^–7 and a variety of preparation methods have been reported, including reaction sintering, chemical vapour deposition, mechanical alloying and combustion synthesis (also known as self-propagating high temperature synthesis). Compared with carbides, fewer results have been reported on ternary nitrides, such as Ti₂AlN and Ti₄AlN₃. It was reported that the electric conductivity of Ti₂AlN is nearly 50% higher than that of Ti₂AlC.8 In this case, Ti₂AlN can be superior to Ti₂AlC for applications such as electrodes and electric brushes.

In previous studies, ternary nitride phases were mostly prepared by reaction sintering.8^–10 As an example, Ti₂AlN was prepared by hot pressing or spark plasma sintering of (Ti+Al+TiN) or by hot isostatic pressing of (Ti+AlN) powders, and the sintering temperature was in the range of 1200–1400°C. Recently, Yeh et al.11 reported the combustion synthesis of Ti₂AlN in a N₂ atmosphere of 0·45–0·79 MPa with TiN and AlN as diluents, where the pressure of N₂ had a strong effect on the phase composition of the products.

In order to further investigate the effect of reaction atmosphere, this paper reports the combustion synthesis of Ti₂AlN in air and N₂ with a pressure of 2 MPa respectively. The phase composition and microstructure of the products obtained in the two different atmospheres are examined, and the effect of reaction atmosphere is discussed in detail. According to the experimental results, the formation mechanism of Ti₂AlN in combustion synthesis is proposed.

Experimental

Commercial Ti (99%, 300 mesh), Al (99%, 300 mesh) and AlN (O<1·5%, d₅₀ = 1·8 μm) powders were used as raw materials. Two starting compositions, which are noted as TAN-1 and TAN-2 respectively in Table 1, were prepared according to the following chemical reaction equations (1) (2) The raw materials were mixed in absolute ethanol and then homogenised by ball milling for 2 h using agate balls as milling media. The homogeneous slurry was fully dried at 80°C for 6 h to prepare the starting reactant powder mixture.

Table 1

Phase compositions of samples with different starting compositions and synthesised in different reaction atmospheres

Sample	Starting composition	Atmosphere	Combustion temperature/°C	Phase composition
TAN-1-A	2Ti+Al	Air	1015	Surface: Ti₃Al+Ti₂AlN+TiO₂+TiN+Ti₃AlN
TAN-1-A	2Ti+Al	Air	1015	Centre: TiAl₃+Ti+Ti₃Al+TiAl
TAN-1-N	2Ti+Al	2 MPa N₂	1863	Ti₂AlN+Ti₃Al+TiN+Ti₃AlN
TAN-2-A	2Ti+0·5AlN+0·5Al	Air	1152	Surface: Ti₃Al+Ti₂AlN+TiO₂+TiN+AlN
TAN-2-A	2Ti+0·5AlN+0·5Al	Air	1152	Centre: Ti₃Al+Ti₂AlN+Ti+TiAl+AlN
TAN-2-N	2Ti+0·5AlN+0·5Al	2 MPa N₂	1986	Ti₂AlN+TiN+AlN
TAN-2-N- HP	Hot pressed at 1400°C for 1 h under 25 MPa and in 0·1 MPa N₂	Ti₂AlN+TiN+AlN

The reactant powder mixture was naturally loaded in a porous graphite crucible with an inner diameter of 26 mm and a height of 45 mm. The porosity in the green powder compacts was nearly 70%. For combustion synthesis in air, the crucible was directly exposed to air. For combustion synthesis in pressurised N₂, the crucible was placed into a closed steel reaction chamber. The chamber was evacuated to a vacuum of <20 Pa and then filled with N₂ up to a pressure of 2 MPa. By passing an electric current of ∼20 A through a tungsten coil closely above the reactant powder, combustion reaction was triggered. The reaction temperature was measured by a W–Re thermocouple directly inserted into the centre of the samples. After the reaction was over and the samples were cooled down, the products were collected.

The reaction product of the sample TAN-2-N was pulverised into fine powders by ball milling for 6 h. Then, the power was hot pressed at 1400°C for 1 h in 0·1 MPa N₂ protective atmosphere to prepare dense bulk ceramic samples. During the hot pressing, the heating rate was 20°C min⁻¹, and the applied mechanical pressure was 25 MPa. After hot pressing, the power was switched off, and the sample was naturally cooled down with the furnace.

The phase composition of the as synthesised products was identified by X-ray diffraction (D8 Focus; Bruker, Germany) using Cu K_α radiation and with a scanning rate of 4° min⁻¹. The microstructure was examined by scanning electron microscopy (SEM; S-4300; Hitachi, Japan) and transmission electron microscopy (TEM; 2010F; JEOL, Japan). Chemical analysis of selected areas was carried out by energy dispersive spectroscopy (EDS; INCA; Oxford Instrument, UK). Vickers hardness was tested with a load of 49 N, and for each sample, at least eight separate tests were made.

Results and discussion

After combustion synthesis in air, the synthesised samples were very loose like the starting green powder compacts and could be readily pulverised into powders. At the surface layer and at the centre part of the samples, different phases were obtained. For example, in the sample of TAN-1-A, at the surface, the target product Ti₂AlN was synthesised, but the major phase was Ti₃Al, as shown in Fig. 1. At the same time, minor TiN, TiO₂ and Ti₃AlN were also found. At the centre of the sample, neither the ternary phase Ti₂AlN nor the simple nitrides such as TiN and AlN were synthesised, and only a variety of Ti–Al alloys (TiAl₃, Ti₃Al and TiAl) and elemental Ti were produced. Similar results were observed for the sample TAN-2-A, as shown in Table 1, except that some Ti₂AlN was obtained at the centre.

Figure 1

X-ray diffraction patterns of products collected from surface and centre of TAN-1-A synthesised in air

Compared with those produced by combustion synthesis in air, the samples synthesised in 2 MPa N₂ were much stronger and difficult to be pulverised. Moreover, the samples prepared in N₂ showed a higher uniformity in phase composition, where same phases were obtained at the surface layer and at the centre. In contrast to those prepared in air, in samples prepared in N₂, the target product Ti₂AlN was the major product, as shown in Fig. 2. That is to say, for the combustion synthesis of Ti₂AlN, it is desirable to improve the phase purity by changing the reaction atmosphere from air to pressurised N₂. In addition, Ti₂AlN, Ti₃Al, TiN and Ti₃AlN were produced in the sample TAN-1-N, and TiN as well as minor AlN was obtained in TAN-2-N (Table 1).

Figure 2

X-ray diffraction patterns of samples of TAN-1-N and TAN-2-N synthesised in 2 MPa N₂

According to the experimental results, the formation mechanism of Ti₂AlN in combustion synthesis is proposed as follows. At first, the combustion reaction begins with the nitridation of Ti (3) During the nitridation of Ti, much heat energy is released, and the temperature of the reactant powder drastically increases, resulting in the melting of Al and sometimes even Ti particles. In this case, a variety of Ti–Al intermetallic phases can be produced as follows (4) (5) (6) Then, the as formed TiN reacts with the Ti–Al alloys to yield Ti₂AlN (7) (8) For the cases with the addition of AlN diluents, Ti₂AlN can be synthesised by another reaction route, i.e. (9) From the above mechanism for the formation of Ti₂AlN, the different phase compositions observed in samples TAN-1-N and TAN-2-N can be explained. In TAN-1-N with no AlN and TiAl₃, reactions (8) and (9) are inactive, and thus, Ti₃Al cannot be consumed and will be left at last. In TAN-2-N with the addition of AlN diluents, Ti₃Al was not found as a final product because it can react with TiN and AlN to form Ti₂AlN. On the other hand, the addition of AlN in TAN-2-N decreases the content of Al in the starting composition, which then reduces the formation of Ti-Al alloys during the combustion reaction. Consequently, the consumption of TiN by reactions (7) and (8) is limited, leading to a higher proportion of TiN in TAN-2-N than that in TAN-1-N, as shown in Fig. 2.

From reactions (7)–(9), TiN is an important intermediate for the formation of Ti₂AlN. TiN is produced by the nitridation of Ti, where the reaction rate strongly depends on the pressure of N₂. At a pressure of 2 MPa, the N₂ present in the starting porous green powder compacts is absolutely not adequate for full nitridation of Ti, and the infiltration of N₂ is necessary. The flux of N₂ depends on the difference between the concentration or pressure of N₂ in the sample and that in the reaction chamber. During the combustion reaction, N₂ is consumed quickly, and thus, the pressure of N₂ in the sample is very low. Therefore, the flux of N₂ is mostly determined by the pressure of N₂ in the chamber, which plays a key role in the combustion synthesis of Ti₂AlN. For samples TAN-1-N and TAN-2-N synthesised in 2 MPa N₂, the flux of N₂ is enough to feed the nitridation reaction. Therefore, the Ti particles were fully nitridised, and no free Ti was found in the product (Fig. 2). For samples TAN-1-A and TAN-2-A synthesised in air, however, the partial pressure of N₂ is only ∼0·08 MPa, and the flux of N₂ is not adequate for the nitridation reaction. Consequently, the nitridation reaction was incomplete, and much free Ti remains (Table 1). Another problem involved in the combustion synthesis in air is the oxidation of Ti because of the presence of O₂ in air. The oxidation reaction will cause the formation of TiO₂, which was verified by experimental observations (Table 1). It is noticed that the oxidation was limited at the surface layer of the samples, and no TiO₂ was detected at the centre (Fig. 1). This is probably attributed to the retarded infiltration of O₂ in comparison with the fast combustion reaction rate. Similar results were observed for the combustion of pure Al, Ti and (Al+C+Si₃N₄) powders.12^–14

In the combustion reaction, the nitridation of Ti is a major source of the generation of heat energy. Hence, the combustion temperature is determined to a large extent by the nitridation reaction. In this way, for the samples synthesised in 2 MPa N₂, the combustion temperatures were much higher than those for the samples synthesised in air, as shown in Table 1. A higher combustion temperature will accelerate the nitridation reactions. Especially, when the temperature exceeds the melting point of Ti (1670°C), the nitridation of Ti takes place by gas–liquid mode instead of gas–solid one, and thus, the reaction rate can be greatly increased. As discussed before, TiN is an important intermediate for the formation of Ti₂AlN. A sufficient supply of TiN will speed up the consumption of Ti–Al intermetallics (such as Ti₃Al and TiAl) and promote the formation of Ti₂AlN by reactions (7)–(9). For this reason, the samples synthesised in 2 MPa N₂ showed a higher content of Ti₂AlN compared with the samples prepared in air. For comparison, the combustion temperature for the reaction 2Ti+Al+0·5N₂ = Ti₂AlN under a N₂ pressure of 0·79 MPa was reported to be nearly 1680°C,11 which is between the temperatures observed here in air and in 2 MPa N₂ respectively (Table 1).

In addition to the pressure of N₂, the structure of the powder impacts also affects the infiltration of N₂, which can be facilitated with the presence of a high fraction of porous channels. In this aspect, the melting and coalescence of the metallic particles will hinder the infiltration of N₂. This negative effect is probably not so severe for combustion synthesis in pressurised N₂ but crucial for combustion synthesis in air because of the low partial pressure of N₂ (∼0·08 MPa). In fact, even for the samples synthesised in air, the reaction temperatures were much higher than the melting point of Al (660°C). In this case, the melting of Al will certainly take place during the combustion reaction, where the coalescence of liquid Al droplets can delay the infiltration of N₂. For example, in the sample TAN-1-A, Ti₂AlN and TiN were obtained only at the surface layer but not found at the centre (Fig. 1) without sufficient N₂ supply because of slow infiltration. The coalescence of metallic droplets can be reduced by the addition of refractory nitride powders as diluents. In contrast to TAN-1-A, in sample TAN-2-A with AlN diluents, Ti₂AlN and TiN were also produced at the centre (Table 1).

Figures 3 and 4 show the SEM images of the samples synthesised in air and those in 2 MPa N₂ respectively. The samples synthesised in air showed a loose structure like the starting green powder compacts, where coarse unreacted Ti particles were visible. In sample TAN-2-A with AlN diluents, the average particle size was smaller, and the fraction of coarse Ti particles was lower than TAN-1-A, which can be attributed to the higher reaction temperature (Table 1) and larger reaction extent in the former, as discussed above. The samples synthesised in 2 MPa N₂ showed a porous network structure. In this structure, although huge pores of >50 μm and cavities of up to 500 μm occurred, the solid part was relatively dense and continuous. Such continuous network was rigid and difficult to be destroyed, offering the samples a high mechanical strength. The average grain size in the samples was below 2 μm.

Figure 3

Images (SEM) of samples synthesised in air: a, b TAN-1-A; c, d TAN-2-A

Figure 4

Images (SEM) of samples synthesised in 2 MPa N₂: a–d TAN-1-N; e–h TAN-2-N

In order to prepare bulk ceramics, the sample TAN-2-N was pulverised into fine powders by ball milling for 6 h and then consolidated by hot pressing. From the SEM observation, as shown in Fig. 5, after milling, the particles were well dispersed, and most of them have a size of 1–3 μm. No abnormal grains or agglomerates larger than 5 μm were observed. Characterisation by TEM confirmed the presence of Ti₂AlN phase with a hexagonal lattice structure. The power was hot pressed at 1400°C for 1 h under a mechanical pressure of 25 MPa and in 0·1 MPa N₂ atmosphere. After hot pressing, there was no clear change in phase composition, and the hot pressed bulk sample consisted of Ti₂AlN, TiN and minor AlN. This means that the as synthesised Ti₂AlN is stable and will not decompose at least below 1400°C. In SEM images of fracture surface shown in Fig. 6, few pores were found, revealing that the sample had a relatively low porosity. The bulk sample showed a microstructure mainly composed of equiaxed TiN and lamellar Ti₂AlN grains. The average size of the TiN grains was below 3 μm, and the thickness of most lamellar Ti₂AlN grains was nearly 1 μm. The EDS spectra for TiN and Ti₂AlN are shown in Fig. 6d and e respectively. To determine the semiquantitative chemical composition of Ti₂AlN phase, EDS analysis was carried out for eight individual Ti₂AlN grains, giving an average of Ti/Al = 2·07±0·11 (in molar ratio), which was in good consistence with the nominal composition of Ti/Al = 2.

Figure 5

Micrographs of pulverised powder of TAN-1-N: a SEM; b TEM

Figure 6

Images (SEM) and EDS spectra for sample TAN-2-N-HP: a overview of fracture surface of sample; b, c SEM images of equiaxed TiN and lamellar Ti₂AlN grains; d, e EDS spectra of TiN and Ti₂AlN respectively

The hardness test showed that the Vickers hardness of the hot pressed sample (TAN-2-N-HP) was 5·8±0·6 GPa. This is higher than the results (nearly 4 GPa) previously reported for the hardness of Ti₂AlN ceramics,8, 9 which is probably caused by the co-existence of a much harder TiN phase in the sample.

Conclusion

In the combustion synthesis of Ti₂AlN, the reaction atmosphere had an evident effect on the phase composition and microstructure of the products. By combustion synthesis in 2 MPa N₂, Ti₂AlN was synthesised as the major product. By combustion synthesis in air, however, the predominant phases were Ti–Al intermetallics, and Ti₂AlN was only a minor product. At the same time, the phase composition of the samples prepared in air was not uniform, where different phases were obtained at the surface layer and at the centre. The reaction mechanism in the combustion synthesis of Ti₂AlN was studied, suggesting that TiN acted as a key intermediate in the formation of Ti₂AlN. It was proposed that, in the combustion synthesis of Ti₂AlN, the flux of N₂ played an important role, affecting both the reaction kinetics and the phase composition of the products, which can probably explain the experimentally observed effect of reaction atmosphere.

Footnotes

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant no. 50932006 and 51002163) and Beijing Natural Science Foundation (grant no. 2112043).

References

Barsoum

: ‘The M_N+1AX_N phases: a new class of solids’, Prog. Solid State Chem., 2000, 28, 201–281.

Wang

, Zhou

: ‘Recent progress in theoretical prediction, preparation, and characterization of layered ternary transition-metal carbides’, Annu. Rev. Mater. Res., 2009, 39, 415–443.

Sun

: ‘Progress in research and development on MAX phases: a family of layered ternary compounds’, Int. Mater. Rev., 2011, 56, 143–166.

Barsoum

, El-Raghy

: ‘Synthesis and characterization of a remarkable ceramic: Ti₃SiC₂’, J. Am. Ceram. Soc., 1996, 79, 1953–1956.

Tzenov

, Barsoum

: ‘Synthesis and characterization of Ti₃AlC₂’, J. Am. Ceram. Soc., 2000, 83, 825–832.

Zhang

, Bao

, Zhou

: ‘Current status in layered ternary carbide Ti₃SiC₂: a review’, J. Mater. Sci. Technol., 2009, 25, 1–38.

Yeh

, Shen

: ‘Combustion synthesis of Ti₃AlC₂ from Ti/Al/C/TiC powder compacts’, J. Alloys Compd, 2008, 466, 308–313.

Barsoum

, Ali

, El-Raghy

: ‘Processing and characterization of Ti₂AlC, Ti₂AlN and Ti₂AlC_0.₅N_0.₅’, Metall. Mater. Trans. A, 2000, 31A, 1857–1865.

Lin

, Zhou

, Li

, Wang

, Zhou

: ‘Synthesis and microstructure of layered-ternary Ti₂AlN ceramic’, Scr. Mater., 2007, 56, 1115–1118.

10.

Yan

, Mei

, Zhu

, Tian

, Wang

: ‘Synthesis of high-purity bulk Ti₂AlN by spark plasma sintering’, Ceram. Int., 2008, 34, 1439–1442.

11.

Yeh

, Kuo

, Wu

: ‘Formation of Ti₂AlN by solid–gas combustion synthesis with AlN and TiN-diluted samples in nitrogen’, Int. J. Appl. Ceram. Technol., 2010, 7, 730–737.

12.

Gromov

, Vereshchagin

: ‘Study of aluminum nitride formation by superfine aluminum powder combustion in air’, J. Eur. Ceram. Soc., 2004, 24, 2879–2884.

13.

Liu

, Chen

, Zhou

, Ren

, Jin

, Ferreira

: ‘Dynamically controlled formation of TiN by combustion of Ti in air’, J. Am. Ceram. Soc., 2007, 90, 2918–2925.

14.

Mei

, Li

: ‘Synthesis of AlN–SiC solid solution through nitriding combustion of Al–C–Si₃N₄ in air’, Acta Mater., 2008, 56, 3543–3549.