Aging Behavior of In-Situ TiB₂ Particles Reinforced 2014Al Alloy Composites

Abstract

In-situ TiB₂ particles reinforced 2014Al alloy matrix composite was prepared using an exothermic reaction process with K₂TiF₆ and KBF₄ salts. The age-hardening behaviors of in-situ TiB₂/2014Al composites were studied using Vickers-hardness measurement, DSC and TEM. The results indicated that TiB₂ particles had a considerable effect on the aging response of 2014Al alloy matrix in the composite. Due to the introduction of TiB₂ particles, the time to achieve peak hardness in the composite is shorter at three aging temperatures compared with that of 2014Al alloy. While the age-hardness efficiency of the composites was lower than that of 2014Al alloy. Formation of G.P. zones was suppressed in composite, which was believed to be due to the introduction of vast particle-matrix interfaces that acted as a sink for vacancies during quenching. Moreover, TiB₂ particles caused increase in thermal-diffusion activation energy of θ′ phase during aging, which made the precipitation of θ′ phase more difficult. And the number and size of precipitates in the composite were relatively less and smaller than that of 2014Al alloy.

1. Introduction

Most composites are based on age hardenable alloys, so that they may be heat-treated to develop the optimum balance of mechanical properties. Extensive studies have been carried on age-hardening behavior of particles reinforced Al alloy matrix composites, and many mature theory systems of the microstructure and strengthen mechanism were built.

Previous some studies showed the addition of particles may accelerate aging kinetics of the matrix alloy in composite.^1–4) Some others concluded a decrease or very little alteration in aging kinetics of the matrix alloy with reinforcement addition.^5–9) The change in aging kinetics during aging of composites due to addition of reinforced particles depends on type of particles including their size, morphology and volume fraction.^1,10–12) However, the age-hardness efficiency of the composite depends mainly upon the nature of the interface between matrix and reinforcement.¹³⁾

Particles reinforced Al alloy matrix composites usually are classified into two categories according to the addition methods of reinforcing particles: ex-situ particles reinforced composite and in-situ particles reinforced composite. The ex-situ particles reinforced composite is prepared by direct adding reinforced particles into matrix alloy, while during the preparation of in-situ particles reinforced composite reinforced particles are generated by the reaction between salts and matrix alloy. In case of ex-situ micro-particles reinforced Al alloy matrix composite, it has been conclusively shown that the addition of particles such as SiC leads to acceleration in the aging kinetics in comparison with the unreinforced matrix alloy. The acceleration is associated with the higher dislocation densities observed in the composite due to the introduction of particles into 2014Al alloy. Many dislocations may act as nucleation sites for precipitates and make solute atoms diffuse faster, which accelerates precipitate nucleating and growing up.^1,2,7) However, in contrast, different results have been reported when the size of reinforced particles is changed.⁸⁾ L. T. Jiang et al.⁸⁾ indicated that when the Al₂O₃ reinforced particles size changes from micro to sub-micro, with the same volume fraction, the number of the reinforcement will be more increased. Due to the volume effect and surface effect, the composite exhibited some unusual characteristics of lower dislocation density and the precipitation of age-hardening phases was suppressed. At present, in-situ particles reinforced Al alloy matrix composite has been the highlight of research interests due to clean interfaces, strong interfacial bonding, fine particles, which differs from that of conventional ex-situ particles reinforced composite. More importantly, the composite exhibits the advantages on mechanical properties and abrasive resistance over the ex-situ particles reinforced composites, which indicates that strengthening mechanism is different.^14–17) Aging strengthening is one of the most important strengthening mechanisms for Al alloy matrix in composite. So, in order to fully utilizing the potentiality of the composite, it is more urgent to study the aging behavior of in-situ TiB₂ particles reinforced composite in details. Earlier, some studies^18,19) have indicated the increase in mechanical properties of the composite in peak-aged conditions compared to that in as-cast conditions, and the enhancement in aging kinetics of matrix alloy due to the presence of TiB₂ particles. However, no systematic studies on age-hardness behavior of such composite and effect mechanism of TiB₂ particles on the aging response of Al alloy matrix were carried out.

The present work is aimed at studying age-hardness behavior of in-situ TiB₂/2014Al composite and the effect of TiB₂ particles on the aging response of 2014Al alloy matrix. Hence systematic studies on the age-hardening behavior and precipitate microstructures of the in-situ TiB₂/2014Al composite and 2014Al alloy was done by means of Vickers Hardness measurement, differential scanning calorimeter (DSC) and transmission electron microscopy (TEM).

2. Materials and Methods

2.1 Experimental materials

In-situ 5 vol%TiB₂/2014Al composite was prepared in advance by an exothermic reaction among K₂TiF₆, KBF₄ and 2014Al alloy (composition listed in Table 1) melts at 1123 K. Then the composite and 2014Al alloy were remelted and squeezed under 120 MPa, respectively. The specific process and squeeze casts have been reported in the previous paper.²⁰⁾ The specimens were taken from the top of the composite and 2014Al alloy casts, respectively. The microstructures of the composite were shown in Fig. 1. As shown in Fig. 1(a), TiB₂ particles were near discoid shape and their sizes were about 50 nm～1 μm²¹⁾. It can be seen from Fig. 1(b) that particles distributed homogeneously and no obvious defects were found.

Table 1 Chemical compositions of 2014Al alloy (mass.%).

Element	Cu	Si	Mn	Mg	Fe	Al
2014Al	4.80	0.78	0.89	0.77	0.11	Bal.

Fig. 1

Microstructures of the in-situ 5 vol% TiB2/2014Al composite: (a) morphology of particles (b) distribution of particles.

2.2 Experiment methods

The specimens of composite were solution treated at 773 K for 4 h and water quenched at room temperature. After solution treatment, the specimens were aged at 423, 443 and 463 K for periods up to 100 h. For comparison, 2014Al alloy was also treated as the above-mentioned methods. After heat treatment, all samples were stored at 253 K to prevent possible natural aging.

The age-hardening responses of the composite and 2014Al alloy were characterized using Vickers hardness measurement at 5 kg load (hence the hardness is represented as Hv5). Each hardness value presented was an average of triplicate specimens and five measurements per condition to ensure the accuracy of results. DSC experiments were performed by using a C404/6/7 thermal analyzer made in Germany. Heating rates of DSC tests was from 5 K･min⁻¹ to 20 K･min⁻¹, and test temperature from room temperature to 773 K for all DSC tests is used to study the variation of activation energy for precipitates. The DSC thermograms were normalized to mass of the composite and 2014Al alloy.

The TEM specimens of the composite and 2014Al alloy were prepared by ion thinning at an accelerating voltage of 5 kV and 13 deg with a liquid nitrogen cold stage after using dimpler to ground to a 10 μm thickness. Transmission electron microscopy was performed on a JEM-2100F transmission electron microscope (TEM).

3. Results and Discussions

3.1 Hardness testing

Figure 2 shows the hardness of in-situ 5 vol%TiB₂/2014Al composite and 2014Al alloy as a function of aging time at aging temperatures of 423, 443 and 463 K, respectively. All curves qualitatively show a similar development. Hardness of the two materials increases monotonically as a function of aging time and rises to a peak value, and then gradually decreases. In addition, it also can be seen that for both the composite and 2014Al alloy the peak-aging time decreases with increasing aging temperature, and the age-hardening kinetics becomes faster with increasing of aging temperature.

Fig. 2

Aging curves of the in-situ 5 vol%TiB₂/2014Al composite and 2014Al alloy at different aging temperatures.

However, the attainments of the peak hardness value and peak-aging time in Fig. 2 reveal conspicuous differences in the two materials. It is seen that at the aging temperature of 443 K, the composite reaches a peak hardness value of 163Hv5 after 12 h of aging. While the peak hardness of 2014Al alloy is 138Hv5 observed after 17 h of aging. It is obvious that the time to attain the peak hardness is reduced 5 h for the composite. This indicates that the introduction of TiB₂ particles accelerates aging kinetics of matrix alloy in the composite. And at three aging temperatures all curves have a similar variation trend, that is, the peak hardness of the composite is shifted towards shorter aging times compared to the 2014Al alloy.

Figure 3 gives the relative hardness increment of the composite as well as 2014Al alloy at three aging temperatures. It can be seen that for the composite and 2014Al alloy, the temperature with maximum hardness value are both 443 K, that is, when the aging temperature is 443 K, the hardness increment is the biggest for the two materials. In addition, it also can be seen that the peak hardness increment is lower at all temperatures for the composite compared with that of 2014Al alloy. When the aging temperature is 443 K, the peak hardness increment is ΔHv5 = 64 for 2014Al alloy, it is only ΔHv5 = 53 for the composite. It indicates that the composite shows a relatively weak aging response, that is, the age-hardening efficiency of the composite is lower than that of 2014Al alloy. In the case of in-situ TiB₂ particles reinforced Al alloy matrix composite aging strengthening have a smaller contribution to hardness, and is not the major hardness mechanism. The increase of hardness for the composite is mainly attributed to introduction of TiB₂ particles. This may be attributed to the introduction of TiB₂ particles, which influences or alters the volume fraction or shape of precipitates in matrix alloy, therefore affecting aging behaviors of composite. And the further discussion is carried out as follows.

Fig. 3

The relative hardness increments (the difference between the peak hardness and the solution treatment hardness) for the in-situ 5 vol%TiB₂/2014Al composite and 2014Al alloy at three aging temperatures.

3.2 DSC analysis

For comparison purposes, results from DSC thermograms of the in-situ 5 vol%TiB₂/2014Al composite and 2014Al alloy quenched into water are shown in Fig. 4. The curve from 2014Al alloy obviously shows three exothermic peaks respectively associated with the formation of G.P. zones, θ^′′ and θ^′ phases, as shown in Fig. 4(a), which is similar to 2024Al reported by C. Badini²²⁾. While the curve of composite have not obvious peak associated with G.P. zone in Fig. 4(b). There are only two exothermic peaks between the regions of 323 K and 600 K in Fig. 4(b). It can be concluded that the formation of G.P. zones in the composite is suppressed due to the introduction of TiB₂ particles.

Fig. 4

DSC curves of the as-quenched specimens of 2014Al alloy (a) and in-situ 5 vol%TiB₂/2014Al composite (b) at different heating rates.

The thermal-diffusion activation energy during precipitation processes of θ^′′ and θ^′ phases can be calculated according to the following equation of Augis and Bennett,^8,23,24)   

\[ \ln \left[ \left( \frac{T_{\rm p} - T_{\rm 0}}{Q} \right) \right] = \frac{E}{RT_{\rm p}} - \ln K_{0} \](1)

Where T_P is the exothermic peak temperature of precipitates, T₀ is the initial temperature, Q is the heating rate, K₀ is the reactive constant, E is thermal-diffusion activation energy and R is gas constant. The results are put in formula (1), and a graph of relationship between $\ln [(T_{\rm p} - T_{0})/Q]$ and $1/T_{\rm p}$ is given. Tan θ is the slope coefficient of the line:   

\[ \tan \theta = \frac{E}{R} \](2)

According to formula (2), thermal-diffusion activation energies of precipitates of θ^′′ and θ^′ phases during aging process of the composite and 2014Al alloy can be calculated. It can be seen from Fig. 5 that the thermal-diffusion activation energies of θ^′′ and θ^′ phases in the composite are 74 kJ･mol⁻¹ and 100 kJ･mol⁻¹ respectively, while those in 2014Al alloy are 66 kJ･mol⁻¹ and 79 kJ･mol⁻¹ respectively. It is obvious that the thermal-diffusion activation energies in composite are higher than that in 2014Al alloy. The result shows that due to the introduction of TiB₂ particles the precipitation of θ^′′ and θ^′ phases become more difficult and need more energy, which is similar to the results on ex-situ sub-micron Al₂O₃ particles reinforced Al matrix composite⁸⁾, while is different from the results about ex-situ micro-particles reinforced Al matrix composite.^7,23)

Fig. 5

Thermal-diffusion activation energies measured by DSC for the precipitation of θ^′′ and θ^′ phases in in-situ 5 vol%TiB₂/2014Al composite and 2014Al alloy.

3.3 TEM observation

In order to follow the microstructural changes associated with the aging process, TEM micrographs of 2014Al alloy and in-situ TiB₂/2014Al composite aged at 443 K for different periods are given in Fig. 6. As shown in Fig. 6(a), (b), the needle precipitates both in two materials aged at 443 K for 1 h are observed. However, for 2014Al alloy, a large number of precipitates are formed and the size is about 30～60 nm, while a few precipitates are observed in the composite and the size is relatively smaller. Figure 6(c), (d) shows the microstructures of 2014Al alloy and the composite aged at 443 K for 6 h. To some degree, the precipitates both in two materials grow up. But the number and size of precipitates in the composite are still less and smaller than that of 2014Al alloy. With continued aging (443 K/64 h), the size of precipitates both in 2014Al alloy and the composite becomes larger, and the former is 3 times larger than the latter, as shown in Fig. 6(e), (f). Figure 7 gives the sizes of precipitates for 2014Al alloy and the composite aged at 443 K for different periods. It is obvious that the existence of TiB₂ particles suppresses the growth of precipitates.

Fig. 6

Microstructure of 2014Al alloy (a)(c)(e) and in-situ 5 vol%TiB₂/2014Al composite (b)(d)(f) aged at 443 K for different aging time: (a)(b) 1 h, (c)(d) 6 h and (d)(e) 64 h.

It can be seen from Fig. 6(f) that TiB₂ particles with discoid shape distribute relatively homogeneous in the composite, while in microscopic view, there are still some dense and rare particle region. Figure 8 shows TEM micrographs of the rare and dense particle regions in the composite aged at 443 K for 64 h. It is obvious that the density of precipitates in rare particle region is higher than that of dense particle region.

3.4 Effect of TiB₂ particles on the aging behavior

Earlier studies^1–3,7) found that for most of ex-situ micro-particle reinforced composites, the addition of particles accelerates aging kinetics of the matrix alloy in composite. Most of researchers considered that the major reason for aging acceleration is a higher dislocation density in the composite due to the introduction of particles into matrix alloy. Many dislocations may act as nucleation sites for precipitates as well as provide solute atom diffusion alleyway, thus accelerating nucleation and growth of the precipitates.

However, the effect of TiB₂ particles on the aging response of 2014Al alloy matrix in the composite is obviously different from that of micro-particle on matrix alloy in the ex-situ micro-particle reinforced composites. Although the introduction of TiB₂ particles accelerates aging kinetics of 2014Al alloy matrix in composite, the age-hardness efficiency of the composites is lower than that of 2014Al alloy at the same aging temperature. In addition, the number and size of precipitates in the composite are relatively less and smaller compared with that in 2014Al alloy. In the composite the size of in-situ formed TiB₂ particles is smaller and about 50 nm～1 μm, which results in the surface area of unit volume increases swiftly compared with micro-particle. Therefore the notable characteristic caused by the smaller size of reinforcement is the introduction of vast interfaces in the composite. At the same time a lot of interfaces play the predominant role as vacancy sinks and can soak up vacancies, thus reduce the whole concentration of vacancies in 2014Al alloy matrix during quenching⁶⁾. For the 2014Al alloy the equilibrium phase was θ phase, and precipitation process is: solid solution α-phase → G.P. zone → θ^′′ phase → θ^′ phase → θ equilibrium phase. The formation of G.P. zone relies on the presence of vacancies to aid the diffusion of the solute atoms⁶⁾. Thus, the decrease in vacancies concentration has a strong influence on the nucleation of G.P. zone and the formation of G.P. zone is inhibited, which strongly impacts on aging properties of the in-situ composites in low temperature. The result of DSC curves shows that there are obvious exothermic peaks of G.P. zone for 2014Al alloy, as shown in Fig. 4(a), while no obvious peak associated with G.P. zones is observed at four different heating rates for the composite in Fig. 4(b). The area of peak in the DSC curve gives the reaction enthalpy that is directly related to the molar heat of reaction and the volume fraction of the precipitating phase⁶⁾. The curves in Fig. 4 show that enthalpy of G.P. zone formation is significantly different for the 2014 Al alloy and composite. The enthalpy of G.P. zone formation in the composite is significantly less than that in 2014Al alloy at four different heating rates, i. e., the volume fraction of the G.P. zone is decreasing in the composites. This shows that the introduction of the in-situ formed TiB₂ particles introduces in a lot of interfaces in the 2014Al alloy matrix, which reduces the whole concentration of vacancies in 2014Al alloy matrix, therefore suppressing the formation of G.P. zones.

It is obvious that precipitates in the composite are less in number and finer in shape, while precipitates in 2014Al alloy is normally more in number and has visible tendency to grow, as shown in Fig. 6 and Fig. 7. In addition, it is observed that the density of precipitates in rare particle region is relatively higher than that of dense particle region in Fig. 8. It can be concluded that the precipitation of θ^′ phase in the composite is inhibited due to the introduction of TiB₂ particles. The introduction of TiB₂ particles introduces in a lot of interfaces in the 2014Al alloy matrix, which severely segments the 2014Al alloy matrix, and therefore the solute atoms diffusion is inhibited during aging process. So, aging precipitation phases are universally less in number and finer in shape in the composite compared with that of 2014Al alloy. In addition, the morphology of TiB₂ particle has also effect on the aging precipitation behavior of 2014Al alloy matrix. Earlier studies show²⁵⁾ that for micro-particles with sharp corner reinforced Al matrix composite, a high dislocation density is formed in the matrix alloy due to the bigger thermal mismatch stress between particles and matrix. While, when the particle size reduces to submicron or nanometer, the thermal mismatch stress is so weak that it is difficult to form dislocation structure. In the in-situ TiB₂/2014Al composite the morphology of TiB₂ particle is near discoid shape and its size is small (50 nm～1 μm). So, the thermal mismatch stress between particle and alloy matrix is lower, and it is difficult to form dislocation in the alloy matrix. The lower dislocation density in the composite can't provide solute atoms diffusion alleyway, which inhibits precipitate nucleating and growing up. In the meantime, the lack of G.P. zones has also a considerable effect on the later stages of aging process. The suppression of G.P. zone formation during the early aging stage means that no nuclei are produced for θ^′ formation until a considerably later stage of the aging process. Therefore aging precipitation phases in the composite are less in number and finer in shape compared with that of 2014Al alloy.

Fig. 7

Size of precipitates in 2014Al alloy and in-situ 5 vol%TiB₂/2014Al composite aged at 443 K under different periods.

Fig. 8

Microstructures of the rare (a) and dense (b) particle regions in in-situ 5 vol%TiB₂/2014Al composite aged at 443 K/64 h.

4. Conclusions

The age-hardening behavior in-situ TiB₂/2014Al composite and effect of in-situ formed TiB₂ particles on the aging characteristics of TiB₂/2014Al composite are investigated. The main conclusions are:

(1) The time to achieve peak hardness in the composite is shorter than that in 2014Al alloy at three aging temperatures.

(2) The age-hardening efficiency of composite is lower than that of 2014Al alloy. The reason is believed to be due to the suppression of G.P. zone and θ′ phase formation in the 2014Al alloy matrix of composite.

(3) Aging temperature has a great effect on the aging hardening process. With the increasing of temperature, the precipitation process is accelerated, and the time to reach peak hardness is shorted.

For the composite and 2014Al alloy, the aging temperatures with maximum hardness value are both 443 K.

(4) The existence of TiB₂ particles in 2014Al alloy matrix suppresses the formation of G.P. zones since the introduction of vast particle-matrix interfaces acts as a sink for vacancies and soaks up vacancies during quenching. Moreover, due to the presence of TiB₂ particles the thermal-diffusion activation energy of θ′ phase is increased during aging process, which makes the precipitation of θ′ phase more difficult. This may be attributed to the introduction of vast particle-matrix interfaces, lower dislocation density and suppression of G.P. zone formation in the composite compared with that in 2014 Al alloy. The number and size of precipitates in the composite are relatively less and smaller.

Acknowledgements

The authors wish to acknowledge the support of the National High Technology Research and Development Program of China (863 project) (No. 2008AA03A239) and the Science and Technology Plan projects of Education Department of Jiangxi province (GJJ150417). At the same time, the authors are grateful to Instrumental Analysis Center and State Key Laboratory of Metal Matrix Composite of Shanghai Jiao Tong University for help in microstructure analysis.

REFERENCES

• 1)  S. Suresh, T. Christman and Y. Sugimure: Scr. Metall. Mater. 23 (1989) 1599–1602. 10.1016/0036-9748(89)90136-1
• 2)  T.G. Nieh and R.F. Karlak: Scr. Metall. Mater. 18 (1984) 25–28. 10.1016/0036-9748(84)90083-8
• 3)  L. Lu, M.O. Lai and Y. Su: Scr. Mater. 45 (2001) 1017–1023. 10.1016/S1359-6462(01)01128-9
• 4)  A. Mandal, R. Maiti, M. Chakraborty and B.S. Murty: Mater. Sci. Eng. A 386 (2004) 296–300. 10.1016/S0921-5093(04)00938-4
• 5)  A.N. Abdel-Azim, Y. Shash, S.F. Mostafa and A. Younan: J. Mater. Process. Technol. 55 (1995) 140–145. 10.1016/0924-0136(95)01946-4
• 6)  K.C. Chen and C.G. Chao: Metall. and Trans. A 26 (1995) 1035–1043. 10.1007/BF02670599
• 7)  A. Daoud and W. Reif: J. Mater. Process. Technol. 123 (2002) 313–318. 10.1016/S0924-0136(02)00103-6
• 8)  L.T. Jiang, M. Zhao, G.H. Wu and Q. Zhang: Mater. Sci. Eng. A 392 (2005) 366–372. 10.1016/j.msea.2004.09.048
• 9)  W. Li, J.P. Long, S. Jing, B.L. Shen, S.J. Gao and M.J. Tu: J. Mater. Eng. Perform. 12 (2003) 19–22. 10.1361/105994903770343420
• 10)  B. Dutta and M.K. Surappa: Scr. Metall. Mater. 32 (1995) 731–736. 10.1016/0956-716X(95)91594-F
• 11)  T.R. Mcnelley and P.N. Kalu: Scr. Metall. Mater. 25 (1991) 1041–1046. 10.1016/0956-716X(91)90498-P
• 12)  D. Srinivasan and M.K. Surappa: Scr. Metall. Mater. 27 (1992) 1139–1144. 10.1016/0956-716X(92)90588-6
• 13)  N. Wang, Z.R. Wang and G.C. Weatherly: Metall. and Trans. A 23 (1992) 1423–1430. 10.1007/BF02647325
• 14)  Z. Sadeghian, B. Lotf, M.H. Enayati and P. Beiss: J. Alloy. Compd. 509 (2011) 7758–7763. 10.1016/j.jallcom.2011.04.145
• 15)  Z. Sadeghian, B. Lotfi, M.H. Enayati and P. Beiss: J. Alloy. Compd. 509 (2011) 7758–7763. 10.1016/j.jallcom.2011.04.145
• 16)  D.G. Zhao, X.F. Liu and Y.C. Pan: J. Mater. Process. Technol. 189 (2007) 237–241. 10.1016/j.jmatprotec.2007.01.028
• 17)  C.F. Feng and L. Froyen: J. Mater. Sci. 35 (2000) 837–850. 10.1023/A:1004729920354
• 18)  A. Mandal, M. Chakraborty and B.S. Murty: Mater. Sci. Eng. A 489 (2008) 220–226. 10.1016/j.msea.2008.01.042
• 19)  A. Mandal, R. Maiti and M. Chakraborty: Mater. Sci. Eng. A 386 (2004) 296–300. 10.1016/S0921-5093(04)00938-4
• 20)  J. Xue, J. Wang, Y.F. Han and B.D. Sun: Mater. Trans. 53 (2012) 2119–2128. 10.2320/matertrans.M2012179
• 21)  J. Xue, J. Wang, Y.F. Han, C. Chen and B.D. Sun: Trans. Nonferrous Met. Soc. China 22 (2012) 1012–1017. 10.1016/S1003-6326(11)61277-6
• 22)  C. Badini, F. Marino and E. Verne: Mater. Sci. Eng. 191 (1995) 185–191. 10.1016/0921-5093(94)09637-C
• 23)  X.F. Wang, G.H. Wu, L.T. Jiang and D.L. Sun: Acta Materiae Compositae Sinica 21 (2004) 61–67 (in Chinese).
• 24)  J.A. Augis and J.E. Bennett: J. Therm. Anal. 13 (1978) 283–292. 10.1007/BF01912301
• 25)  L.T. Jiang, G.H. Wu and D.L. Sun: J. Nonferrous Metals (China) 12 (2002) 516–520.