Kinetics evaluation of passive oxidation at high temperature of silicon nitride used in the fabrication of lightweight metal matrix composites

doi:10.21203/rs.3.rs-2124521/v1

Download PDF

Research Article

Kinetics evaluation of passive oxidation at high temperature of silicon nitride used in the fabrication of lightweight metal matrix composites

https://doi.org/10.21203/rs.3.rs-2124521/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

In the present work, the passive oxidation kinetics of silicon nitride α-Si₃N₄ and (α + β)-Si₃N₄ powders at different temperatures from 1000 to 1200°C were evaluated. Oxidation of these powders was carried out to improve wettability and pressureless infiltration of AZ91E Mg-alloy. The oxidation kinetic constant (k_p) values and oxide layer depth were calculated. The oxidation kinetic constants k_p, were 9.11e^− 16 and 6.35e^− 16 kg²/m⁴s and the oxidation depth were 4.25 nm and 3.69 nm, for α-Si₃N₄ and (α + β)-Si₃N₄, respectively. Furthermore, passive surface oxidation of silicon nitride was performed in order to achieve affinity between silicon nitride and AZ91E magnesium alloy. Therefore, the manufacture of AZ91E/Si₃N₄ magnesium matrix composite materials was successfully achieved by spontaneous infiltration process at a temperature of 800°C.

passive oxidation

oxidation kinetics

silicon nitride

composites

AZ91E alloy

Changes in passive oxidation kinetics of silicon nitride were identify by the β-Si₃N₄ presence in the Si₃N₄ powders.
The SiO₂ amorphous coating produced by passive oxidation on the silicon nitride surface reduce the contact angle between Si₃N₄ and AZ91E magnesium alloy.
Passive oxidation of silicon nitride improves the wettability between Si₃N₄ and AZ91E magnesium alloy.

Silicon nitride (Si₃N₄) is considered one of the hardest advanced ceramic materials and is used in components subject to high temperatures and mechanical stress due to its very characteristic properties such as low coefficient of thermal expansion, high resistance to corrosion, high elastic modulus and excellent mechanical properties [1, 2]. Silicon nitride is a potential candidate as a reinforcement for the manufacture of light composite materials by mixing it with magnesium alloys [3, 4]. However, it has been shown that there is not a good affinity between silicon nitride and magnesium showing low wettability and promoting the formation of fragile phases such as MgAl₂O₄ when magnesium interacts with the nitrogen in the system [5, 6]. Few related systems have been achieved based on the use of techniques such as stir casting leading to magnesium matrix composite materials with low reinforcement content, in addition to exhaustive mixing control and the possibility of poor reinforcement distribution [7].

On the other hand, it is possible to increase the affinity of silicon nitride with alloys rich in magnesium due to the formation of SiO₂ on the surface of the Si₃N₄ particles which reduces the contact angle between them. Silicon nitride, despite having a high chemical stability, it has been shown to decompose silicon nitride into silicon and nitrogen at temperatures above 1000°C [8, 9]. Passive oxidation of silicon nitride occurs if the surrounding atmosphere is rich in oxygen, with a well-established reaction:

$$S{i_3}{N_4}(s)+3{O_2}(g) \to 3Si{O_2}(s)+2{N_2}(g)$$

$\Delta {G^o}= - 1981+0.21T(kJmo{l^{ - 1}})$

Where T is taken as the absolute temperature in kelvin (K). This kind of oxidation it is detected by gain mass in the silicon nitride due to the formation of a thin layer of SiO₂ and the release of nitrogen. Active oxidation, on the other hand, is the decomposition of silicon nitride into nitrogen and silicon monoxide according to the following reaction:

$$S{i_3}{N_4}(s)+{\raise0.7ex\hbox{$3$} \!\mathord{\left/ {\vphantom {3 2}}\right.\kern-0pt}\!\lower0.7ex\hbox{$2$}}{O_2}(g) \to 3SiO(g)+2{N_2}(g)$$

This oxidation only occurs if the atmosphere contains a low partial pressure of oxygen, and it is detected by the mass lost after the oxidation process. The transition between the two types of oxidations is well studied and the possibility of a change from passive to active is almost impossible [8, 10]. This type of surface treatment of silicon nitride has been studied since 1958 with the passivation surface treatment of silica by Wagner [11]. It has different purposes and application proposals like increasing the electrical properties of silica and the theoretical analysis about passive oxidation kinetics [12].

In this work it is proposed to perform and analyzed the oxidation behavior of two different silicon nitride powders through the measurement of parabolic constant rate and oxide layer depth of the SiO₂ layer. All this with the purpose to have enough oxidation of Si₃N₄ powders to have good wettability between the Mg-alloy and the SiO₂ of the superficial layer of Si₃N₄ powders. This analysis will give news results about oxidized silicon nitride systems and will propose an innovative application in the field of lightweight metal matrix composite (MMC) materials.

2.1. Si₃N₄ powders

The materials used for this study were silicon nitride powder with particle diameter range between 400 to 600 nm. Two different samples of silicon nitride were used and compared: α-Si₃N₄ and (α + β)-Si₃N₄ (≤ 99.9% Toshiba Materials Co. JPN and Sigma Aldrich Co. USA, respectively).

The oxidation kinetics of silicon nitride was carried out by means of superficial passive oxidation through thermogravimetric analyses. A Setaram Setsys Evolution 16/18 thermobalance, with a precision up to 0.003 mg, was employed. About 20 mg of sample was used in an atmosphere of dry air with a partial pressure of 21.5 kPa. The oxidation experiments were carried out at temperatures between 1000 to 1200°C to evaluate the kinetic oxidation behavior of silicon nitride.

The structural characterization of the silicon nitride powders before and after the passive oxidation was carried out by X-ray diffraction on a Bruker D8 Advance diffractometer. The samples were scanned at 2 theta angle from 10° to 90° with 0.020 increment and a step time of 0.6 s. The morphological characterization and oxidation distribution were carried out in a FEGSEM 7600F field emission scanning electron microscope (FE-SEM).

2.2. AZ91E magnesium alloy

The magnesium alloy used to evaluate the affinity between the oxidated surface of silicon nitride was AZ91E whose chemical composition in wt.% is Mg 90.65%, Al 8.64%, Zn 0.44%, Mn 0.15%, Si 0.14%. The structural characterization of the magnesium alloy (AZ91E) was performeed by X-ray diffraction on a Bruker D8 Advance diffractometer. Phases distribution in the AZ91E alloy were analyzed through scanning electron microscope (SEM).

2.3. AZ91/Si₃N₄ composites

To evaluate the affinity and wettability between the previously oxidated silicon nitride and the magnesium alloy, porous compacts of silicon nitride powders were prepared and oxidized at 1200°C for 30 minutes and then compacted at 90 MPa. The infiltration was performed at 800°C for 15 minutes to obtain a full fill metal matrix composite AZ91E/Si₃N₄. The morphological characterization, homogeneity, and phase distribution through scanning electron microscope (SEM) were carried out. Moreover, reaction products of the composite were analyzed through energy dispersive spectroscopy (EDS).

3.1. Microstructural characterization of Si₃N₄ powders

Figure 1 shows the X-Ray diffraction patterns of two kinds of Si₃N₄ powders used to compare the kinetics behavior of the passive oxidation. Figure 1a shows X-Ray results fot the high purity α-Si₃N₄ (≤ 99.9% Toshiba Materials Co. JPN) revealing only α-phase with the indexed card 04-005-5074. Figure 1b shows the (α + β)-Si₃N₄ of high purity (≤ 99.9% Sigma Aldrich Co. USA). The indexed cards of both phases (α + β) are 04-005-5074 and 04-007-2386, respectively.

Figure 2a and b show SEM images of the start α-Si₃N₄ powders while. Figure 2c and d show SEM images of the start (α + β)-Si₃N₄ powders. The particle size distribution was homogeneous and the average particle size was ⁓600 nm for both types of silicon nitride.

The transformation of the α-Si₃N₄ to β-Si₃N₄ phase occurs around 1300 to 1450°C. However, the hexagonal structure continues to prevail, the only difference is that the interplanar distance in the c axis of the unit cell for α phase (5.618 Ȧ) is almost double compared with the β phase (2.90 Ȧ). At room temperature β-Si₃N₄ is more stable than α-Si₃N₄ and the transformation of the α-Si₃N₄ to β-Si₃N₄ is also an irreversible process, therefore β-Si₃N₄ is a more commercial reagent [13].

3.2. Microstructural characterization of AZ91 alloy

Microstructural characterization of the AZ91E magnesium alloy was carried out by X-ray diffraction on a Bruker D8 Advance diffractometer. Figure 3a shows the X-ray pattern obtained, being the Mg₁₇Al₁₂ the only intermetallic detected. Figure 3b confirms the phases present in the AZ91E alloy which were analyzed through scanning electron microscope (SEM) by energy dispersive spectroscopy (EDS).

3.3. Passive oxidation of Si₃N₄ powders

Passive oxidation experiments were performed at different temperatures from 1000 to 1200°C to evaluate the oxidation kinetics. The first oxidation experiment was performed at 1000°C with a heating control of 7°C per minute. Figure 4 shows the comparative of mass gain rate between α-Si₃N₄ and (α + β)-Si₃N₄ with a similar behavior under 1000°C. It can be observed that the passive oxidation reaction starts at about 690°C even though the literature indicates that the temperature at which passive oxidation begins is at about 1000°C in an atmosphere of dry air [8]. Moreover, some authors mention that the temperature can vary from 600 to 950°C when the oxygen concentration is increased in the oxidizing atmosphere [9, 14, 15]. The present results indicate that the passive oxidation below 1000°C is negligible. Thus, it was decided to increase the heating rate to 30°C/min up to 1200°C.

According to Hou et al. [16], the passive oxidation occurs from the surface of the particle towards the nucleus, first transforming the superficial layers into SiO₂ and with an intact nucleus of silicon nitride. Yang et al. [9], also mentioned that a time of 30 minutes at 1200°C is enough for the particle to be completely oxidized. The superficial passive oxidation of silicon nitride can be observed as translucent zones on the surface of the particles and can also be identified by the slight formation of necks between particles due to oxidation. The kinetic oxidation analyses of powders of α-Si₃N₄ and (α + β)-Si₃N₄ were carried out at 1200°C for 5 hours. Figure 5 shows the morphological comparison between α-Si₃N₄ and (α + β)-Si₃N₄ oxidized powders. The morphological analysis by SEM reveals that α-Si₃N₄ and (α + β)-Si₃N₄ do not present relevant changes in terms of morphology, but they do in terms of particle size distribution because of sintering of particles due to passive oxidation. On the other hand, the α-Si₃N₄ promotes nucleation and slow fine-grain growth in the crystallized proportion of SiO₂. β-Si₃N₄ phase tends to promote coarse grain growth and the formation of defects in the interface with SiO₂ [17]. In Fig. 5, the necks formation between particles is evident and can be considered as a sintering process. The formation of necks between particles can improve densification as occurs with Al₂O₃, Y₂O₃ and MgO which are used to promote the formation of a glassy phase used in some sintering processes [8, 18].

Figure 6 shows the crystallographic comparison between α-Si₃N₄ and (α + β)-Si₃N₄ before and after 5-hour oxidation. Regarding the structural analysis (Fig. 6a and b), no changes were identified despite knowing that the surface is composed of amorphous SiO₂. The formation of SiO₂ promotes sintering and densification of particles. There is evidence of the crystallization of SiO₂ which favors the formation of microcracks at the interface [17]; although another work mentions that crystallization occurs at 870°C and SiO₂ transforms from α-quartz to α-tridymite [9]. In this work, evidence of crystallization greater than 5% by weight was not found and, therefore, it was not identified by the X-ray diffraction technique.

On the other hand, the kinetic data reported by other authors are limited to obtaining the kinetic oxidation constant k_p and the activation energy E. Moreover, no evidence was found in the literature about kinetic oxidation studies of β-phase silicon nitride on its influence in passive oxidation. Passive oxidation on silicon nitride has been mainly focused on obtaining a surface layer that facilitates the fluidity of Si₃N₄ slurries or promoting the formation of other phases such as Si₂N₂O [8, 19]. Therefore, there are mostly comparative studies between (α) and (β) silicon nitride that described the grain growth of the SiO₂ phase [17]. Even though, it has been demonstrated [20] that the formation of Si₂N₂O does not occur unless a secondary treatment is given to Si₃N₄. In this work, evidence confirms that passive oxidation of silicon nitride improves the wettability with magnesium alloys.

Figure 7 shows the thermogravimetric results for α-Si₃N₄ and (α + β)-Si₃N₄. It can be observed that the influence of β-phase is null during heating process. The amount of β-phase has more impact in the oxidation process at a given temperature. Because the β-phase of Si₃N₄ is more stable than the α-phase; once it is formed, it can no longer be transformed into α-phase [13]. This implies that the amount of β-phase in the system reduces the oxidation kinetics. Due to the difficulty of quantifying the β-phase content, experiments cannot be carried out by changing the β-phase content in the system. Further experiments are required to better observe the β-phase dependence on the oxidation process as literature shows that silicon nitride is treated indifferently; regardless the amount of Si₃N₄ phase in the system [16].

As has been observed in Fig. 5, there are no important changes on the surface in terms of morphology. However, α-Si₃N₄ and (α + β)-Si₃N₄ achieved 4.86% and 2.87% mass gain, respectively. The findings has not been reported in a most recent review on the subject [8]. Being data that has been omitted over time, it is possible to obtain the parabolic oxidation rate constant and the surface reaction depth [21, 22].

To perform the kinetic calculations, the specific surface areas for α-Si₃N₄ and (α + β)-Si₃N₄ were obtained as 12.629 and 8.598 m²/g, respectively. Surface area measurements were obtained using a QuantaChrome surface area analyzer. For the case of the parabolic oxidation rate constant, k_p, the following equation was used [21]:

$${W^2}={k_p} \cdot t$$

Where W² is the mass gain per unit area squared, expressed in g²/m⁴, k_p is the kinetic oxidation constant, expressed in kg²/m⁴s and t is time. The k_p is obtained from the slope by plotting W² versus time. On the other hand, for the calculation of the oxidation depth, Eq. (5) may be applied [22].

$$d=\left( {\frac{{\% W}}{{100S}}} \right)\left( {\frac{{M(r)}}{{M(p) - M(r)}}} \right)\left( {\frac{1}{\rho }} \right)$$

Where d is the depth in (nm) of the oxidized layer, %W is the mass gain rate, S is the surface area of the material in (nm²/g), ρ is the density of the material, M(r) is the molecular mass of Si₃N₄ and M(p) is the molecular mass of SiO₂. For the thermogravimetric analyses, the initial sample mass for α-Si₃N₄ and (α + β)-Si₃N₄ powders were 22.64 and 23.31 mg, respectively.

Figure 8 shows the oxidation kinetic constants obtained by the slope of the plot W² versus time from the α-Si₃N₄ and (α + β)-Si₃N₄ thermogravimetric results. The oxidation kinetic constant achieved for α-Si₃N₄ was 9.11x10^− 16 kg²/m⁴s and a Pearson R of 0.9889. The same procedure was carried out for (α + β)-Si₃N₄ powders obtaining an oxidation kinetic constant of 6.35x10^− 16 kg²/m⁴s. These results indicate a strong dependence of the kinetic oxidation constant with temperature when compared with the other values in the literature. For example, at temperatures in the range of 1823 to 1923 K, Hirai et al. [23] found a higher kinetic constant, when working with α-Si₃N₄ manufactured by CVD, whose values are between 5.83x10^− 11 and 2.22x10^− 10 kg²/m⁴s. Similar values are reported by Ogbuji and Fox [12, 24] who calculated the kinetic constants being in the range from 1.75x10^− 12 to 2.44x10^− 11 kg²/m⁴s at the same temperature range. Moreover, Butt et al. [25] reported kinetic constants between 5.86x10^− 25 to 1.71x10^− 22 kg²/m⁴s at 973 and 1173 K, respectively. For temperature ranges close to this work like the results obtained by Franz et al. [26] at 1273 and 1533 K, the kinetic constants are between 3.4x10^− 18 and 1.6x10^− 17 kg²/m⁴s. This corroborates the impact of temperature on the passive oxidation of silicon nitride.

The results of the oxidation depth, as well as the kinetic constants are shown in Table 1.

Table 1

Oxidation kinetics constant and oxidation depth for both α-Si₃N₄ and (α + β)-Si₃N₄ powders.
Powder	Kinetics constant (kg²/m⁴s)	Oxidation depth (nm)
α -Si₃N₄	9.11x10^− 16	4.25
(α + β)-Si₃N₄	6.35x10^− 16	3.69

As observed in Fig. 8, the slope of the curves of (α + β)-Si₃N₄ is lower which is likely due to the content of β-phase in the Si₃N₄ powders, this observation has not been fully studied in the past. The present results, based on thermogravimetric analyses, are in contradiction to the work of Backhaus-Ricoult et al. [17] who mentioned that the β-phase content increases the oxidation kinetics. Since it is well known that the β-phase is more stable, monocrystalline, and defect-free structure, the present results suggest that presence of β-phase decreases the oxidation rate as compared to α-Si₃N₄ alone. Backhaus-Ricoult et al. [17], also mentioned that the content of β-phase promotes a high density of defects in the oxidized interface and the formation of Sionite-type silicon oxynitride Si₂N₂O, because the oxidation mechanism is due to the decomposition of Si₃N₄ and the formation of amorphous SiO₂, a devitrification and crystallization of SiO₂ begins in the form of cristobalite that increases with the nitrogen content trapped during the decomposition of Si₃N₄. The cristobalite is easily formed at temperatures around 1300°C. Authors such as Narushima et al. [10] mention that the presence of Si₂N₂O decreases the oxidation kinetics, since it is a phase formed between the Si₃N₄ and SiO₂ layers; Si₂N₂O phase also limits the diffusion of oxygen through the interface. There is no evidence of the formation of defects between the crystallized SiO₂ grains for α-Si₃N₄. Trapped nitrogen is the one that plays the most important role during cristobalite grain growth and in its particle size distribution. In general terms, the β-phase promotes grain growth to large grains, which promotes the formation of defects as porosities and the formation of silicon oxynitride. In case of the α-phase, it promotes the nucleation of new grains with slow growth and fine grain size, avoiding the formation of defects and the formation of additional phases [17].

The present kinetic results were compared with other systems that have already been studied. Table 2 shows a comparison of the oxidation kinetic constants for different Si₃N₄ systems studied. As aforementioned no oxidation kinetic studies have been made taking into account the type of phase present in silicon nitride.

Table 2

Comparison of the oxidation kinetics constant and activation energy in different Si₃N₄ systems studied.
Type of Si₃N₄	Oxidation Method	Atmosphere	Temperature (K)	Time (ks)	Kinetics	Activation Energy (kJ·mol^− 1)	Ref.
CVD (α)	Furnace	Wet O₂	1073–1473		Parabolic	119.3	[27]
CVD (α)	Ellipsometry	Dry O₂	1373–1673	3.6–21.6	Parabolic	464	[28]
CVD (α)	Ellipsometry	Dry O₂	1273–1573	32.4	Parabolic 9.07 x10^− 17 m²·s^− 1	330.2	[29]
CVD (amorphous)	Thermogravimetry	O₂, 0.1 Mpa	1823–1923	36	Parabolic (< 1923 K) 1823 K: 1.5 x 10^− 11 kg²·m^− 4·s^− 1 1903 K: 5.8 x 10^− 11 kg²·m^− 4·s^− 1 Linear (1923 K) 1.6 x 10^− 7 kg·m^− 2·s^− 1	460	[23]
CVD (α)	Thermogravimetry	O₂, Air, 0.1 Mpa	1823–1923	36	Parabolic O₂ 1823 K: 5.8 x 10^− 11 kg²·m^− 4·s^− 1 1923 K: 2.2 x 10^− 10 kg²·m^− 4·s^− 1 Air 1823 K: 3.6 x 10^− 12 kg²·m^− 4·s^− 1	390	[23]
Powder*	Thermogravimetric	N₂-20%O₂	973–1173		Parabolic (Sample M-11) 973 K: 5.86 x 10^− 25 m²·s^− 1 1073 K: 2.11 x 10^− 23 m²·s^− 1 1173 K: 1.71 x 10^− 22 m²·s^− 1 Parabolic (Sample SN-E10) 973 K: 8.29 x 10^− 25 m²·s^− 1 1073 K: 4.26 x 10^− 24 m²·s^− 1 1173 K: 3.67 x 10^− 23 m²·s^− 1	230–260	[25]
Powder*	Thermogravimetric	N₂-20%O₂	1273–1473		Parabolic (Sample M-11) 1273 K: 1.14 x 10^− 21 m²·s^− 1 1373 K: 1.50 x 10^− 20 m²·s^− 1 1473 K: 2.03 x 10^− 19 m²·s^− 1 Parabolic (Sample SN-E10) 1273 K: 4.16 x 10^− 22 m²·s^− 1 1373 K: 1.72 x 10^− 20 m²·s^− 1 1473 K: 4.29 x 10^− 19 m²·s^− 1	400–540	[25]
Powder*	Furnace	Dry O₂	1273–1533		Parabolic 1273 K: 3.4 x 10^− 18 m²·s^− 1 1373 K: 6.7 x 10^− 18 m²·s^− 1 1473 K: 1.15 x 10^− 17 m²·s^− 1 1503 K: 1.4 x 10–¹⁷ m²·s^− 1 1533 K: 1.6 x 10^− 17 m²·s^− 1		[26]
Powder*	Furnace	Dry O₂	1503–1533		Parabolic 1503 K: 1.8 x 10–¹⁹ m²·s^− 1 1533 K: 3.2 x 10^− 19 m²·s^− 1		[26]
Powder*	Furnace	Wet O₂	1273		Parabolic 1273 K: 3.8 x 10^− 19 m²·s^− 1		[26]
Hot-Pressed*	Thermogravimetric	Dry O₂	1623–1760		Parabolic 1623 K: 3.84 x 10^− 11 kg²·m^− 4·s^− 1 1633 K: 5.22 x 10^− 11 kg²·m^− 4·s^− 1 1666 K: 1.01 x 10^− 10 kg²·m^− 4·s^− 1 1700 K: 1.74 x 10^− 10 kg²·m^− 4·s^− 1 1731 K: 3.28 x 10^− 10 kg²·m^− 4·s^− 1 1734 K: 3.41 x 10^− 10 kg²·m^− 4·s^− 1 1752 K: 6.43 x 10^− 10 kg²·m^− 4·s^− 1 1760 K: 7.63 x 10^− 10 kg²·m^− 4·s^− 1	450–980	[30]
CVD (α)	Thermogravimetric	Dry O₂	1473–1823	36	Parabolic 1473 K: 1.75 x 10^− 12 kg²·m^− 4·s^− 1 1573 K: 3.05 x 10^− 12 kg²·m^− 4·s^− 1 1673 K: 8.33 x 10^− 12 kg²·m^− 4·s^− 1 1773 K: 1.69 x 10^− 11 kg²·m^− 4·s^− 1 1823 K: 2.44 x 10^− 11 kg²·m^− 4·s^− 1	186 ± 44	[24]
CVD (α)	Thermogravimetric	Dry O₂	1473–1773	360		363	[12]
Hot-Pressed*	Volumetric Method	Dry O₂	1521–1731	25.2–86.4	Parabolic 1521 K: 1.27 x 10^− 10 kg²·m^− 4·s^− 1 1621 K: 2.32 x 10^− 9 kg²·m^− 4·s^− 1 1644 K: 1.62 x 10^− 9 kg²·m^− 4·s^− 1 1668 K: 1.07 x 10^− 9 kg²·m^− 4·s^− 1 1713 K: 1.27 x 10^− 9 kg²·m^− 4·s^− 1	440	[31]
Powder (α)	Thermogravimetric	Dry O₂	1473	18	9.11 x 10^− 16 kg²·m^− 4·s^− 1		This work
Powder (α + β)	Thermogravimetric	Dry O₂	1473	18	6.35 x 10^− 16 kg²·m^− 4·s^− 1		This work
* Unknown type of phase in the silicon nitride

3.4. Thermodynamic analysis of the AZ91E/Si₃N₄ system

Attempts have been made to use silicon nitride for the manufacture of metal matrix composite materials with high reinforcement contents, without favorable results using the pressureless infiltration technique [6]. This is due to the low wettability and high contact angle formed between the liquid magnesium and the silicon nitride. Some authors [5, 6] even mention that the interaction between magnesium and nitrogen to form Mg₃N₂ leads to the formation of brittle phases such as the MgAl₂O₄ spinel. Also, it is possible that Si₃N₄ reacts with Al to form AlN. Figure 9 shows a thermodynamic analysis of the AZ91E/Si₃N₄ system for the possible reactions among the system components.

3.5. Mechanical properties of the AZ91E/Si₃N₄ system

It has been found that some magnesium alloys by the addition of silicon nitride, in low percentages, improve certain mechanical properties such as hardness, elastic modulus, toughness, tensile strength, and the failure strain; most of these composites are fabricated by stir casting [3, 4]. In this work, some mechanical properties of the Mg-AZ91/Si₃N₄ composite were predicted, such as elastic modulus and hardness, which directly depend on reinforcement content. Mechanical properties such as the Young's modulus can be roughly predicted with models such as the mixture rule [32], the Halpin-Tsai model [33] and the Hashin & Shtrickman model [34]. The rule of mixture is the simplest model. Halpin-Tsai model is a mathematical model for the prediction of elasticity of composite material based on the geometry and orientation of the reinforcement and the elastic properties of the reinforcement and matrix. Hashin and Shtrickman model is based in the aspect ratio of the particles mainly. Figure 10a shows a graphic of the calculated elastic modulus using these models, giving values between 100 and 200 GPa for 50 wt.% of reinforcement. This is more than 100% higher as compared to the 45 GPa for monolithic alloy [35]. Figure 10b shows a graphic of the calculated hardness using the above cited models, giving values between 1042 and 1120 HV.

3.6. Fabrication of AZ91/Si₃N₄ composites by infiltration technique

Fabrication of AZ91E/Si₃N₄ composites by several techniques has been carried out but with very low reinforcement content. The low reinforcement content is related to the low wettability between magnesium and silicon nitride [3, 4, 7]. Therefore, magnesium matrix composite materials are limited to manufacturing processes such as stir casting, squeeze casting, ultrasonic vibration, twin roll casting, shear compaction processing, powder metallurgy, in-situ reaction synthesis, mechanical alloying and even spray forming. In addition, processes such as stir casting have some disadvantages like the possibility of obtaining an heterogeneous distribution of the reinforcement particles, segregation, unwanted inclusions, and even porosity [36]. However, magnesium alloys in manufacturing processes for MMC such as pressureless infiltration is little considered in most recent reviews [36–38]. Pressureless infiltration is one of the most complete manufacturing processes for MMC due to the ease in changing composition of the alloy and reinforcement, also, it is one of the most economical processes, in addition to not requiring machining due to its near-net shaping property. Pressureless infiltration process is made possible by the passive oxidation of silicon nitride, due to the formation of SiO₂ on the surface of the particles. Is well known that the contact angle between SiO₂ and Mg is around 56° [39], promoting wettability between the oxidized Si₃N₄ particles and the magnesium alloy AZ91E.

Figure 11 shows the micrographs and chemical mappings of AZ91E/Si₃N₄ compound manufactured by pressureless infiltration. The infiltration was carried out at 800°C for 15 min, obtaining a complete infiltration of the alloy in the porous Si₃N₄ preform made with the powders previously oxidized for 30 min at 1200°C. Complete infiltration by gravity (Fig. 11a-f) and by capillarity (Fig. 11g-l) were obtained due to the good wettability between the alloy and the SiO₂ of the superficial layer of Si₃N₄ powders. The compounds were manufactured in a 50–50% volume ratio, but with the advantage that this process facilitates the variation in reinforcement content.

The passive oxidation treatment becomes novel for the system AZ91/Si₃N₄ when it comes to magnesium matrix composite materials due that it improves the wettability and enables the pressureless infiltration process. However, there is only evidence of being used in alloys rich in titanium such as Mg-Ti-C and Mg-Ti-B reinforced with B₄C [36], or in the alloy AZ91E reinforced with AlN [40]. In the latter, the predominant limitation is temperature since it is only feasible to infiltrate at 900°C, which is a high temperature for magnesium alloys due to its volatility and flammability.

The magnesium metal matrix composites can reduce the weight in automotive and aerospace industry [41] in addition to reduce CO₂ emissions; and in the case of electric vehicles, their autonomy is increased [42]. The reduction on costs and manufacturing step processes increase the possible implementation of this kind of materials in the automotive industry [41]. Moreover, magnesium alloys are also capable of reducing weight in military vehicles and increase the ballistic impact resistance, however, their mechanical properties are limited [43, 44]. With the help of passive oxidized silicon nitride as a reinforcement, the mechanical properties of magnesium alloys can be improved for the above applications by means AZ91E/Si₃N₄ composites by pressureless infiltration.

Oxidation kinetics of silicon nitride (α-Si₃N₄) and (α+β)-Si₃N₄ powders were evaluated at temperatures from 1000 to 1200 °C. Passive oxidation of silicon nitride was applied in order to achieve wettability between Si₃N₄and AZ91E Mg-alloy.
The oxidation kinetic constants k_p at 1200 °C for 5 hwere 9.11e^-16 and 6.35e^-16 kg²/m⁴s, for Si₃N₄ (α) and Si₃N₄ (α+β), respectively.
Oxidation depths of 4.25 nm and 3.69 nm were achieved at 1200 °C for 5 h for α-Si₃N₄ and (α+β)-Si₃N₄, respectively.
AZ91/Si₃N₄ composites were successfully produced by pressureless infiltration process at temperature of 800 °C for 15 minutes. Which it is indicative that oxidation of Si₃N₄ surface was enough to achieve wettability between Si₃N₄and Mg-alloy (AZ91E).
A complete infiltration by gravity and by capillarity were obtained due to the good wettability between the Mg-alloy and the SiO₂ of the superficial layer of Si₃N₄ powders.
The compounds were fabricated with 50% volume reinforcement with the premise that this process enables the variation in reinforcement content.

Conflict of interest: No potential or similar conflict of interest was declared by the authors.

Data availability statements: The datasets analyzed and generated during the present study are fully included in the results of this paper declared by the authors.

Acknowledgements

The authors would like to thank to CIC-UMNSH for providing the financial support and facilities to develop this study. I.I. López-López expresses his gratitude to National Council of Science and Technology (CONACYT) for the scholarship support given to this research work.

I. I. López-López: Conceptualization, Investigation, Methodology, Writing – Original Draft. A. Contreras: Validation, Writing – Review & Editing. R. Morales-Estrella: Writing – Review & Editing. J. Lemus-Ruiz: Investigation, Supervision, Funding acquisition, Project administration.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the Coordinación de la Investigación Científica-UMSNH

Richerson D W. The Magic of Ceramics, 2nd edi,, John Wiley & Sons, Inc., New Jersey, USA, 2012.
Richerson D W. Modern Ceramic Engineering: Properties, Processing and Use in Design, 2nd edi,, CRC Press, New York, USA, 1992.
Paramsothy M, Chan J, Kwok R, Gupta M. Enhanced mechanical response of magnesium alloy ZK60A containing Si3N4 nanoparticles. Composites Part A: Applied Science and Manufacturing. 42(12), 2093–2100, 2011.
Paramsothy M, Chan J, Kwok R, Gupta M. Enhanced mechanical response of hybrid alloy AZ31/AZ91 based on the addition of Si3N4 nanoparticles. Materials Science and Engineering: A. 528(21), 6545–6551, 2011.
Wang S R, Geng H R, Wang Y Z, Zhang J C. Microstructure and fracture characteristic of Mg–Al–Zn–Si3N4 composites. Theoretical and Applied Fracture Mechanics. 46(1), 57–69, 2006.
Wang S R, Geng H R, Wang Y Z. Fabrication and machinability of Si 3 N 4 –Mg–Al–Zn (AZ91) composites. Materials Science and Technology. 22(2), 223–226, 2006.
Anand N, Ramachandran K K, Bijulal D. Microstructural, mechanical and tribological characterization of vacuum stir cast Mg-4Zn/Si3N4 magnesium matrix nanocomposite. Materials Today: Proceedings. 46(53), 9387–9391, 2021.
Bocanegra-Bernal M, Aguilar-Elguézabal A, Echeberria J. High temperature oxidation of silicon nitride based ceramics: a review, in Silicon Nitride: Synthesis, Properties and Applications, E.J. Hierra and J.A. Salazar, eds., Nova Science Publishers, Inc, New York, USA, 2012, pp. 1–31.
Huang Y, Ma L, Tang Q, Yang J, Xie Z, Xu X. Surface oxidation to improve water-based gelcasting of silicon nitride. Journal of Materials Science. 35(14), 3519–3524, 2000.
Narushima T, Goto T, Hirai T, Iguchi Y. High-Temperature Oxidation of Silicon Carbide and Silicon Nitride. Materials Transactions, JIM. 38(10), 821–835, 1997.
Wagner C. Passivity during the Oxidation of Silicon at Elevated Temperatures. Journal of Applied Physics. 29(9), 1295–1297, 1958.
Ogbuji L U J T, Opila E J. A Comparison of the Oxidation Kinetics of SiC and Si3 N 4. Journal of The Electrochemical Society. 142(3), 925–930, 1995.
Forgeng W., Decker B. Nitrides of Silicon. Metallurgical Society of American Institute of Mining, Metallurgical and Petroleum Engineers - Transactions. 212, 343–348, 1958.
Jiménez C, Perrière J, Vickridge I, Enard J P, Albella J M. Transformation of silicon nitride in oxygen plasma. Surface and Coatings Technology. 45(1–3), 147–154, 1991.
Warburton J B, Antill J E, Hawes R W M. Oxidation of Thin Sheet Reaction-Sintered Silicon Nitride. Journal of the American Ceramic Society. 61(1–2), 67–72, 1978.
Hou X, Chou K-C, Hu X, Zhao H. A new measurement and treatment for kinetics of isothermal oxidation of Si3N4. Journal of Alloys and Compounds. 459(1–2), 123–129, 2008.
Backhaus-Ricoult M, Guerin V, Huntz A M, Urbanovich V S. High-temperature oxidation behavior of high-purity α-, β-, and mixed silicon nitride ceramics. Journal of the American Ceramic Society. 85(2), 385–392, 2002.
Hirosaki N, Okada A, Mitomo M. Effect of oxide addition on the sintering and high-temperature strength of Si3N4 containing Y2O3. Journal of Materials Science. 25(3), 1872–1876, 1990.
Yang X, Li B, Zhang C, Wang S, Liu K, Zou C. Design and fabrication of porous Si3N4-Si2N2O in situ composite ceramics with improved toughness. Materials & Design. 110, 375–381, 2016.
Balat M, Flamant M, Male G, Pichelin G. Active to passive transition in the oxidation of silicon carbide at high temperature and low pressure in molecular and atomic oxygen. Journal of Materials Science. 27, 697–703, 1992.
Singhal S C. Thermodynamics and kinetics of oxidation of hot-pressed silicon nitride. Journal of Materials Science. 11(3), 500–509, 1976.
Jr G E. Oxidation behavior of silicon carbide. Journal of the American Ceramic Society. 41(4888), 347–352, 1958.
Hirai T, Niihara K, Goto T. Oxidation of CVD Si3N4 at 1550° to 1650°C. Journal of the American Ceramic Society. 63(7–8), 419–424, 1980.
Fox D S. Oxidation Behavior of Chemically-Vapor-Deposited Silicon Carbide and Silicon Nitride from 1200° to 1600°C. Journal of the American Ceramic Society. 81(4), 945–950, 1998.
Butt D P, Albert D, Taylor T N. Kinetics of thermal oxidation of silicon nitride powders. Journal of the American Ceramic Society. 79(11), 2809–2814, 1996.
Fränz I, Langheinrich W. Conversion of silicon nitride into silicon dioxide through the influence of oxygen. Solid State Electronics. 14(6), 499–505, 1971.
Deal B E, Grove A S. General relationship for the thermal oxidation of silicon. Journal of Applied Physics. 36(12), 3770–3778, 1965.
Du H, Tressler R E, Spear K E, Pantano C G. Oxidation Studies of Crystalline CVD Silicon Nitride. Journal of The Electrochemical Society. 136(5), 1527–1536, 1989.
Choi D J, Fischbach D B, Scott W D. Oxidation of Chemically-Vapor‐Deposited Silicon Nitride and Slnglem‐Crystal Silicon. Journal of the American Ceramic Society. 72(7), 1118–1123, 1989.
Cubicciotti D, Lau K H. Kinetics of Oxidation of Yttria Hot-Pressed Silicon Nitride. Journal of The Electrochemical Society. 126(10), 1723–1728, 1979.
Cubicciotti D, Lau K H. Kinetics of Oxidation of Hot-Pressed Silicon Nitride Containing Magnesia. Journal of the American Ceramic Society. 61(11–12), 512–517, 1978.
Contreras Cuevas A, Bedolla Becerril E, Martínez M S, Lemus Ruiz J. Metal Matrix Composites: Wetting and Infiltration, Springer International Publishing, Cham, 2018.
Halpin J C. Primer on Composite Materials Analysis, 2nd edi,, CRC Press, Ohio, USA, 1992.
Hashin Z, Shtrikman S. A variational approach to the theory of the elastic behaviour of multiphase materials. Journal of the Mechanics and Physics of Solids. 11(2), 127–140, 1963.
Czerwinski F. Magnesium Alloys - Design, Processing and Properties, InTech, 2011.
Chen L, Yao Y. Processing, microstructures, and mechanical properties of magnesium matrix composites: A review. Acta Metallurgica Sinica (English Letters). 27(5), 762–774, 2014.
Dey A, Pandey K M. Magnesium metal matrix composites - A review. ACADEMIA. 42, 58–67, 2015.
Babaremu K, Udoh M, Joseph O O, Fayomi O S I, John M E. Production of Magnesium Matrix Composite and the Corresponding Mechanical Properties: A Review. IOP Conference Series: Materials Science and Engineering. 1107(1), 012199, 2021.
Shi L, Shen P, Zhang D, Dong E, Jiang Q. Reactive wetting in liquid magnesium/silica and magnesium/silicon systems. Applied Surface Science. 274, 124–130, 2013.
Bedolla E, Lemus-Ruiz J, Contreras A. Synthesis and characterization of Mg-AZ91/AlN composites. Materials & Design. 38, 91–98, 2012.
Burd J T J, Moore E A, Ezzat H, Kirchain R, Roth R. Improvements in electric vehicle battery technology influence vehicle lightweighting and material substitution decisions. Applied Energy. 283(November), 116269, 2021.
Wenlong S, Xiaokai C, Lu W. Analysis of energy saving and emission reduction of vehicles using light weight materials. Energy Procedia. 88, 889–893, 2016.
Malik A, Nazeer F, Wang Y. A Prospective Way to Achieve Ballistic Impact Resistance of Lightweight Magnesium Alloys. Metals. 12(2), 2022.
Malik A, Wang Y, Huanwu C, Nazeer F, Khan M A. Post deformation analysis of the ballistic impacted magnesium alloys, a short-review. Journal of Magnesium and Alloys. 9(5), 1505–1520, 2021.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Kinetics evaluation of passive oxidation at high temperature of silicon nitride used in the fabrication of lightweight metal matrix composites

Archived Versions:

Version 1

Abstract

Figures

Highlights

1. Introduction

2. Materials And Methods

2.1. Si₃N₄ powders

2.2. AZ91E magnesium alloy

2.3. AZ91/Si₃N₄ composites

3. Results

3.1. Microstructural characterization of Si₃N₄ powders

3.2. Microstructural characterization of AZ91 alloy

3.3. Passive oxidation of Si₃N₄ powders

3.4. Thermodynamic analysis of the AZ91E/Si₃N₄ system

3.5. Mechanical properties of the AZ91E/Si₃N₄ system

3.6. Fabrication of AZ91/Si₃N₄ composites by infiltration technique

4. Conclusions

Declarations

References

Additional Declarations

Archived Versions:

Version 1

Kinetics evaluation of passive oxidation at high temperature of silicon nitride used in the fabrication of lightweight metal matrix composites

Archived Versions:

Version 1

Abstract

Figures

Highlights

1. Introduction

2. Materials And Methods

2.1. Si3N4 powders

2.2. AZ91E magnesium alloy

2.3. AZ91/Si3N4 composites

3. Results

3.1. Microstructural characterization of Si3N4 powders

3.2. Microstructural characterization of AZ91 alloy

3.3. Passive oxidation of Si3N4 powders

3.4. Thermodynamic analysis of the AZ91E/Si3N4 system

3.5. Mechanical properties of the AZ91E/Si3N4 system

3.6. Fabrication of AZ91/Si3N4 composites by infiltration technique

4. Conclusions

Declarations

References

Additional Declarations

Archived Versions:

Version 1

2.1. Si₃N₄ powders

2.3. AZ91/Si₃N₄ composites

3.1. Microstructural characterization of Si₃N₄ powders

3.3. Passive oxidation of Si₃N₄ powders

3.4. Thermodynamic analysis of the AZ91E/Si₃N₄ system

3.5. Mechanical properties of the AZ91E/Si₃N₄ system

3.6. Fabrication of AZ91/Si₃N₄ composites by infiltration technique