Trans. Nonferrous Met. Soc.

China 24(2014) 2282−2288

Effects of external stress aging on morphology and

precipitation behavior of θ'' phase in Al−Cu alloy

Shang FU1,2, Dan-qing YI1,2, Hui-qun LIU1,2, Yong JIANG1,2, Bin WANG1,2, Zhan HU1,2
1. School of Materials Science and Engineering, Central South University, Changsha 410083, China;
2. Key Lab of Nonferrous Materials (Ministry of Education), Central South University, Changsha 410083, China
Received 17 October 2013; accepted 3 April 2014

Abstract: The exposure of Al−5Cu alloy to an external stress with normal aging was carried out. The effects of external stress-aging
on the morphology and precipitation behavior of θ'' phase were investigated by transmission electron microscopy (TEM), differential
scanning calorimetry (DSC) and first principle calculation. The size of the θ'' phase precipitated plates in stress-aging (453 K, 6 h, 50
MPa) is 19.83 nm, which is smaller than that of those present (28.79 nm) in stress-free aging (453 K, 6 h). The precipitation process
of θ'' phase is accelerated by loading external stress aging according to the analysis of DSC results. The apparent activation energy
for the external stress-aging is 10% lower than the stress-free one. The first principle calculation results show that the external stress
makes a decrease of 6% in the interface energy. The effects of the stress on aging process of the alloy are discussed on the basis of
the classical theory. The external stress changes the morphology and precipitation behavior of θ'' phase because the critical nucleation
energy is decreased by 19% under stress aging.
Key words: Al−Cu alloy; stress aging; morphology; precipitation behavior; first principle calculation; interface energy

integral manufacture [11]. The precipitation phase

1 Introduction morphologies of the super-alloy [12] and the zircaloy [13]
were reported under external stress aging. HOSFORD
Precipitation strengthening via aging treatment is an and AGRAWAL [14] firstly introduced this technique
important technique of enhancing the mechanical into the processing of aluminum alloy and CHEN et al
properties of age-hardenable aluminum alloys [1,2]. And [15] found the microstructure of the precipitation phase
the age-hardening response is mainly dominated by the existing to be anisotropic or stress-orientation effect.
precipitation behavior and the morphology of the ZHU and STARKE [16] considered that this
meta-stable and stable precipitation phase [3]. As a stress-oriented phenomenon depended on the alloy
typical type of age-hardenable aluminum alloy, 2xxx composition, aging parameters and the stress level. The
series Al−Cu alloy presents one of the most famous possible explanations [15,16] of this phenomenon were
examples with the aging precipitation sequence the atomic diffusion and the precipitation/matrix lattice
(supersaturate solid solution−G.P. zone−θ″ phase−θ' misfits. However, these explanations were not fully
phase−θ phase) [4,5]. And the θ″ phase is considered to accepted since the explanations cannot satisfy all the
be the first independent precipitation phase [6]. The reported results.
previous works were focused on the optimization of the Generally, the energy of matrix and interface
age-hardening response by changing the aging influences significantly the precipitation nucleation−
processing parameters and the alloy composition [7−10]. growth behavior according to the classical
A few yeas ago, a novel external stress aging transformation theory. But the hypothesis explanations
combing the external stress-field and traditional artificial mentioned above have not still given a clear description
aging was conducted into the age-hardenable alloy to for the interface energy between precipitation phase and
improve the mechanical properties and the process of matrix under external stress because the evaluation of the

Foundation item: Project (2012CB619506) supported by the National Basic Research Program of China; Project (51071177) supported by the National
Natural Science Foundation of China
Corresponding author: Dan-qing YI; Tel: +86-731-88830263; E-mail: yioffice@csu.edu.cn
DOI: 10.1016/S1003-6326(14)63345-8
 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288 2283
interface energetic would be a challenge for the
experimentalists. The first principle calculation is a
useful method to investigate the energetic properties on
the bulk material and interface between the secondary
phase and matrix [17].
The aims in this work are to investigate the
precipitation behavior and morphology of the θ″ phase Fig. 1 Al matrix/precipitation θ″ phase interface model
under different external stress levels by DSC and TEM
characterization methods. The effect of potential change 3 Results and discussion
of external stress on the interface energy and the critical
nucleation energy is assessed by the first principle 3.1 Age-hardening response and morphology of
calculation based strategy according to the classical precipitation under external stress
nucleation theory. The hardness versus aging−time curves under
different external stress levels are exhibited in Fig. 2.
2 Experimental and computational method The peak-aging time is 6 h from the curves and it shows
that the peak-aging time is irrelevant with the external
The material used in this study was an Al−Cu alloy stress 50 MPa since it is in the elastic stress range which
melted by induction furnace. The chemical composition would only produce a strain at 0.006. During the
of the alloy is listed in Table 1. under-age period, the hardness values of all different
external stress levels samples are improved dramatically
Table 1 Chemical composition of nominal Al−5Cu alloy (mass
since the amount of the meta-stable phase precipitated
fraction, %)
from the supersaturated solid solution. Additionally, the
Cu Mg Mn Fe Si Zn Al
major precipitation phase in this period is θ″ phase (the
4.99 0.01 0.008 0.01 0.026 0.012 Bal. nominal composition of the θ″ phase is Al3Cu) according
to the 453 K aging temperature. Meanwhile, the higher
The ingots of the alloy were broken-down and external stress-level sample had the higher hardness
rolled to sheets with the thickness of 2 mm. The sheets compared with 0 stress level (or stress-free aging). It
were solution heat treated at 773 K for 1 h, followed by means that the external stress makes the precipitation
water-quenching, and the sheets were then subjected to response more significantly and rapidly. The traditional
the external stress aging. The temperature for the precipitation strengthening view holds that the strong
external stress aging was determined as 453 K and the precipitation response is caused by the decreasing of the
external stress levels were 0, 50 MPa (50 MPa was in the precipitation barrier energy [5]. Generally speaking, the
elastic load period according to the stress−strain curve external stress is considered to be a sort of energy source
yet). Hardness measurement was applied to determining to have an impact on the aging processing response.
the peak-aging time for each aging condition. The load
for the hardness measurement was 1500 g. The TEM
specimens of 3 mm in diameter were electro-polished
with 30% nitric acid and 70% methanol solution at 253
K. TEM observation was performed on FEI Tecnai G2 20
transmission electron microscope operated at 200 kV.
Differential scanning calorimetry (DSC, NETZSCH
DSC 200 F3) was performed on the specimens to
investigate the transformation temperature and the θ″
phase precipitation behavior under different heat rates
10, 15, and 20 K/min. The first principle calculation was
applied to determining the interface energy between Al
matrix and the θ″ phase. All these calculations (including
the bulk property optimization, the activity of Cu in Al
under stress/strain and the interface energy as mentioned Fig. 2 Age-hardening curves of Al−5Cu alloy at 453 K
above) were determined by the Vienna ab initio
simulation package (VASP) [18]. The parameter of the The TEM microstructure and the corresponding
K-point mesh was followed by the energy convergent selected area electron diffraction patterns of θ″ phase for
principle. The interface model (including a 12Å vacuum all samples in the peak-aging condition are shown in
layer) is schematically shown in Fig. 1. Fig. 3. Along the {100} zone axis of the electron beam,
2284 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288
the bright field image shows two “edge-on” {100} average θ″ phase length under 50 MPa external stress
variants of the θ″ phase plates with the stress-free level level is determined as 19.83 nm while the result for
(0). However, the external stress-aged (453 K, 50 MPa) stress-free aging is 28.79 nm. According to Fig. 4,
sample obtained a stress-orientation precipitation external stress makes the θ″ phase plate length smaller,
phenomenon that only one variant of the θ″ phase plate is and the mode zone move from 30 nm to 20 nm in
visible. It would be in an agreement with the previous statistics distribution level. This implies that the external
work [16]. stress would promote the nucleation of θ″ phase,
resulting in the formation of finer and more θ″ phase.

Fig. 3 TEM images of θ″ phase and its selected area electron

diffraction patterns of alloy aged at 453 K for 6 h under 0 (a) Fig. 4 θ″ phase plate length distribution (number fraction)
and 50 MPa (b) external stress levels counted by TEM images

It is clearly observed that the morphology of the θ″ 3.2 θ″ phase precipitation behavior under external
phase under higher stress level is quite different that stress
under the stress-free ones. The θ″ phase in higher stress- The DSC (heat flow) curves under all different
level (50 MPa) sample is finer than that in stress-free (0) external stress levels at heating rate of 15 K/min are
aged sample; the volume fraction of the θ″ phase is displayed in Fig. 5. The endothermic peak1 (peak 1a and
higher than that in stress-free aging condition. So, the peak 1b) and the exothermic peak 2 (peak 2a and peak
finer and more volume fraction θ″ phase under external 2b) correspond to the G.P. zone’s dissolution period and
stress contributed to the precipitation strengthening the θ″ phase precipitation transformation, respectively,
effect, resulting in the improvement of hardness. according to the work by HAYOUNE and HAMANA
Additionally, 4−6 bright-field TEM photographs [19].
(about 1000 θ″ phase plates counted) in peak-aging The exothermic peak 2b moves to the lower
condition were used to make a precipitation phase length temperature since the precipitation behavior is influenced
distribution statistics figure and evaluate the effect of by the external-stress. Similarly, the endothermic peak1
external stress (0, 50 MPa) aging on the morphology of shifts to the higher temperature because an amount of the
θ″ phase by quantitative method. The precipitation phase G.P. zone is too little to detect in lower temperature. The
length distribution statistics is exhibited in Fig. 4. The θ″ phase precipitation rate results are shown in Fig. 6.
 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288 2285
This supports the consideration that the sample under the stress-free aging. It would be a strong evidence to
external stress has more amounts and quick precipitation show that the precipitation behavior is accelerated by
of θ″ phase. external stress.

Table 2 Activation energies of θ″ phases obtained by Kissinger

method
External β/ Activation
Tp/K
stress/MPa (K·min−1) energy/(kJ·mol−1)
10 540.1
0 15 551.3 99.2
20 555.2
10 519.1
50 15 531.9 90.1
20 533.9

Fig. 5 DSC curves of samples at peak-aged condition under

external stress

Fig. 7 Plots of ln(β/Tp2) versus 1/Tp for θ″ phase precipitation

under external stress

3.3 Interface energy and critical nucleation energy

Fig. 6 Precipitation rate curves of samples under external stress
calculations under external stress aging
According to the classic nucleation theory, the
Furthermore, the apparent activation energy is
driving force is the change of the system Gibbs free
determined according to the Kissinger method [20,21], as energy, and the main obstruction in this case is interface
shown in Eq. (1), which can be used to evaluate the energy. The nucleation energy for age-hardening
precipitation ability. aluminum alloy could be determined according to
⎛β ⎞ E 1 Eq. (2):
ln ⎜ 2
⎜T ⎟⎟ = − k +C (1)
⎝ ⎠ R T 4π 3
Wr = r (ΔEchem + ΔEstrain ) + 4 πr 2 γ (2)
3
where β is the heating rate; T is the temperature at the
DSC heat flow curve peak; Ek is the activation energy; R where Wr is the nucleation energy; r is the nucleation
is the gas constant; C is the integration constants. radius; ΔEchem is the chemical potential; ΔEstrain is the
strain energy; γ is the interface energy between
With the different heating rates (β= 10, 15 and 20
precipitation phase and the matrix. And the critical
K/min) and the peak (the exothermic peak 2) temperature
nucleation energy could be expressed as Eq. (3) when the
Tp listed in Table 2, the scatter plot of ln(β/Tp2) vs (1/Tp)
critical nucleation radius r* is determined.
is shown in Fig. 7. The apparent activation energy under
50 MPa stress-aging is 90.1 kJ/mol and the activation 16 π γ3
energy is 99.2 kJ/mol in stress-free aging condition from Wr (r * ) = (3)
3 (ΔFchem + ΔFelastic ) 2
Fig. 7, respectively. The apparent activation energy for
the external stress aging is about 10% lower than that for The chemical potential change (ΔEchem) under
2286 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288
external stress (transforming into strain (ε)) can be decreases gradually by the effect of the external
obtained by Eq. (4) according to the work of RAVI et al stress/strain from 0.88 J/m2 to 0.82 J/m2. As a result, the
[22] where the μ is the chemical potential. And the influence of external stress is more significant on the
results are listed in Table 3 according to the first interface energy than on the chemical potential and the
principle calculation method. The change in chemical elastic strain energy. When the aging temperature is 453
potential is not apparent according to the calculation K, the critical nucleation energy under 50 MPa external
results. stress level is 19 % lower than that under level 0
ΔEchem (ε ) = μ (Al3Cu (ε )) + according to Eq. (3). And the precipitation behavior
under external stress could be stimulated as the
(31 − 3) μ (Al(ε )) − μ (Al31Cu (ε )) (4)
nucleation energy becomes low. The θ″ precipitation
phase would be finer and the precipitation processing
Table 3 Results of chemical potential and chemical potential would be easier, which is supported by the experiment
change results.
External strain/ μ(Al3Cu)/ μ(Al)/ μ(Al31Cu)/ ΔEchem(ε)/
stress level eV eV eV eV
0 −30.6519 −3.7446 −119.9346 −15.5671
0.002 −30.6518 −3.7445 −119.9315 −15.5662
0.004 −30.6517 −3.7441 −119.9194 −15.5665
0.006 −30.6512 −3.7434 −119.8985 −15.5678

The change of the elastic strain energy (ΔEstrain)

could be described by the Eshelby’s elastic inclusion
model [23] and the mathematic expression is shown as
Eq. (5).

ΔEstrain = − Eδ 2 (5)
Fig. 8 Calculated activity αCu in Al under different strain/stress
Under stress-free aging condition, the misfit δ could levels
be determined as −0.052 between θ″ phase [24] and
matrix, the calculated elastic modulus is 77 GPa.
However, the misfit δ was determined to be −0.058 under
50 MPa external stress. Therefore, the elastic strain
energy would be similar for the all external stress levels
(0, 50 MPa) to some extent. This means that lower
external stress level has little influence on the strain
energy.
The interface energy can be expressed as Eq. (6).
1
γ interface (ε ) = [ E0 (ε ) − N Al μAl (ε ) − N Cu μCu (ε )]
2A
1 1
= {E0 (ε ) − N Al μ (ε ) − N Al in Al3Cu μAl3Cu (ε ) −
2A Al
3
1 Fig. 9 Calculated interface energy and nucleation energy
( N Cu in Al3Cu − N Al in Al3Cu )[ μ Cu (ε ) + KT ln α Cu in Al ]}
3 change in all strain/stress levels
(6)
where A is the area defined in the interface model; E0 is In addition, it is implied that the interface energy is
the total energy of interface model; μi and Ni are the the major factor to influence the nucleation processing.
chemical potential and the atom number of species i, The change of the critical nucleation energy is larger
respectively; ε is the applied strain under external stress; than that of the apparent precipitation activation energy
αCu in Al is the activity of Cu in Al matrix; αCu in Al can be since the effect of the external stress would be different
determined as shown in Fig.8 according to Ref. [25]. during the coarsening processing of precipitation.
The interface energy is calculated by the first
principle calculation (VASP software packages), and the 4 Conclusions
interface energy results in different strains are shown in
Fig. 9. The results show that the interface energy 1) The applied external stress contributes to the
 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288 2287
improvement of hardness of Al−Cu alloy. The size of θ″ [9] CHEN Z G, ZHENG Z Q. Microstructural evolution and ageing
phase under 453 K, 6 h, 50 MPa stress aging is 19.83 nm, behaviour of the low Cu−Mg ratio Al−Cu−Mg alloys containing
scandium and lithium [J]. Scripta Materialia, 2004, 50: 1067−1071.
which is smaller than that of the θ″ phase (28.79 nm) in
[10] XU G F, ZHOU M S, YIN Z M. Effects of Sc addition on
stress-free condition (453 K, 6 h).
age-hardening and precipitation behavior of Al−3.0% Cu alloy [J].
2) The precipitation process of θ″ phase is Mining and Metallurgical Engineering, 2006, 26(2): 92−95, 99.
accelerated by external stress aging according to the shift [11] ZHU A W, STARKE E A. Materials aspects of age-forming of
of the DSC heat flow exothermic peak. And the Al−XCu alloys [J]. Journal of Materials Processing Technology, 2001,
precipitation under 50 MPa stress is dramatically 117(3): 354−358.
accelerated compared with the stress-free aged one. The [12] OBLAK J M, PAULONIS D F, DUVALL D S. Coherency
apparent activation energy for the external stress-aging is strengthening in Ni base alloys hardened by DO22 γ′ precipitates [J].
Metallurgical Transactions, 1974, 5(1): 143−153.
90.02 kJ/mol, 10% lower than that for stress-free aging.
[13] LOUTHAN M R, ANGERMAN C L. Thermo-mechanical effects on
3) The first principle calculation results show that
hydride precipitation in zircaloy-2 [J]. Metallurgical Transactions,
the external stress has an obvious influence on the 1973, 4(7): 1763−1765.
interface energy. The interface energy under external [14] HOSFORD W F, AGRAWAL S P. Effect of stress during aging on
stress condition is 0.82 J/m2, 6% lower than the interface the precipitation of θ’ in Al-wt pct Cu [J]. Metallurgical and
energy (0.88 J/m2) in stress-free aging condition. The Materials Transactions A, 1975, 6(3): 487−491.
critical nucleation energy is 19% lower than that in [15] CHEN Da-qin, ZHENG Zi-qiao, LI Shi-chen, CHEN Zhi-guo.
stress-free aging condition. The change of nucleation Mechanism of stress aging in Al−Cu(−Mg−Ag) alloys [J].
Transactions of Nonferrous Metals Society of China, 2004, 14(4):
energy determines the precipitation behavior and
779−784.
morphology of θ″ phase.
[16] ZHU A W, STARKE E A. Stress aging of Al−xCu alloys experiments
[J]. Acta Materialia, 2001, 49(12): 2285−2295.
Acknowledgements [17] LAN G Q, JIANG Y, YI D Q, LIU S J. Theoretical prediction of
This paper was supported by the High Performance impurity effects on the internally oxidized metal/oxide interface: The
Computing Center of Central South University, China. case study of S on Cu/Al2O3 [J]. Physical Chemistry Chemical
Physics, 2012,14: 11178−11184.
[18] KRESSE G, FURTHMÜLLER J. VASP the guide [EB/OL].
References http://cms.mpi.univie.ac.at/vasp/vasp/ vasp.html. 2013.
[19] HAYOUNE A, HAMANA D. Structural evolution during
[1] SMALLMAN R E, BISHOP R J. Modern physical metallurgy and non-isothermal ageing of a dilute Al−Cu alloy by dilatometric
materials engineering: Science, process applications [M]. 6th ed. analysis [J]. Journal of Alloys and Compounds, 2009, 474(1−2):
Butterworth-Heinemann Press, 1999. 118−123.
[2] SON S K, TAKEDA M, MITOME M, BANDO Y, ENDO T. [20] GHOSH K S, GAO N. Determination of kinetic parameters from
Precipitation behavior of an Al−Cu alloy during isothermal aging at calorimetric study of solid state reactions in 7150 Al−Zn−Mg alloy
low temperatures [J]. Material Letter, 2005, 59(6): 629−632. [J]. Transactions of Nonferrous Metals Society of China, 2011, 21(6):
[3] JANG J H, NAM D G, PARK Y H, PARK I M. Effect of solution 1199−1209.
treatment and artificial aging on microstructure and mechanical [21] LI Hai, WANG Xiu-li, SHI Zhi-xin, WANG Zhi-xiu, ZHENG
properties of Al−Cu alloy [J]. Transactions of Nonferrous Metals Zi-qiao. Precipitation behaviors of Al−Mg−Si−(Cu) aluminum alloys
Society of China, 2013, 23(3): 631−635. during continuous heating [J]. The Chinese Journal of Nonferrous
[4] XU Guo-fu, PENG Xiao-yan, LIANG Xiao-peng, LI Xu, YIN Metals, 2011, 21(9): 2028−2034. (in Chinese)
Zhi-min. Constitutive relationship for high temperature deformation [22] RAVI C, WOLVERTON C, OZOLIŅŠ V. Predicting metastable
of Al−3Cu−0.5Sc alloy [J]. Transactions of Nonferrous Metals phase boundaries in Al−Cu alloys from first-principles calculations
Society of China, 2013, 23(3): 1549−1555. of free energies: The role of atomic viberations [J]. Europhysics
[5] SILCOCK J M, HEAL T J, HARDY H K. Structural aging Letters, 2006, 73(5): 719−725.
characteristics of binary aluminum−copper alloys [J]. Journal of the [23] WANG Hong-wei, YI dan-qing, CAI Jin-ling, WANG Bin. Effect of
Institute of Metals, 1954, 82: 239−248. stress aging on mechanical properties and microstructures of 2E12
[6] LOECHTE L, GITT A, GOTTSTEIN G, HURTADO I. Simulation of aluminum alloy [J]. The Chinese Journal of Nonferrous Metals, 2011,
the evolution of GP zones in Al−Cu alloys: An extended 21(12): 3019−3025. (in Chinese)
Cahn–Hilliard approach [J]. Acta Materialia, 2000, 48(11): [24] SKROTZKI B, SHIFLET G J, STARKE E A. On the effect of stress
2969−2984. on nucleation and growth of precipitates in an Al−Cu−Mg−Ag alloy
[7] BOURGEOIS L, DWYER C, WEYLAND M, NIE J F. The magic [J]. Metallurgical and Materials Transactions A, 1996, 27:
thicknesses of theta’ precipitates in Sn-microalloyed Al−Cu [J]. Acta 3431−3444.
Materialia, 2011, 60(2): 633−644. [25] JIANG Y, SMITH J R, EVANS A G. Temperature dependence of the
[8] RINGER S P, SAKURAI T, POLMEAR I J. Origins of hardening in activity of Al in dilute Ni(Al) solid solutions [J]. Physical Review B,
aged Al−Cu−Mg−(Ag) alloys [J]. Acta Materialia, 1997, 45(9): 2006, 74: 224110.
3731−3744.
2288 Shang FU, et al/Trans. Nonferrous Met. Soc. China 24(2014) 2282−2288

外场应力对 Al−Cu 合金中 θ″相形貌及

析出行为的影响
傅 上 1,2，易丹青 1,2，刘会群 1,2，江 勇 1,2，王 斌 1,2，胡 湛 1,2

1. 中南大学 材料科学与工程学院，长沙 410083；

2. 中南大学有色金属材料科学与工程教育部重点实验室，长沙 410083

摘 要：对 Al−5Cu 铝合金在时效过程中施加弹性加载，并采取透射电镜、热分析研究外应力时效作用下 θ″析出

相的形态和析出行为。结果表明：在 50 MPa 的外力作用下，在 453 K 时效 6 h 后的析出相尺寸是 19.83 nm，小
于其常规时效( 453 K，6 h)的尺寸 28.79 nm。热分析结果揭示施加应力后时效析出过程得到促进，同时析出激活
能结果表明外应力作用使得激活能下降 10%。第一性原理计算结果发现，外应力作用使得析出相与基体的界面能
下降 6 %；结合经典形核相变理论分析讨论，临界形核功受到外应力作用下降 19 %，从而引起了析出相形态和析
出行为的改变。
关键词：Al−Cu 合金；应力时效；形态；析出行为；第一原理计算；界面能
(Edited by Hua YANG)