materials

Article
Development of Precipitation Hardening Parameters
for High Strength Alloy AA 7068
Julia Osten 1 , Benjamin Milkereit 1,2 , Michael Reich 1, *, Bin Yang 1 , Armin Springer 3 ,
Karina Nowak 4 and Olaf Kessler 1,2
1 Materials Science, Faculty of Mechanical Engineering and Marine Technology, University of Rostock,
Justus-von-Liebig-Weg 2, 18059 Rostock, Germany; werkstofftechnik@uni-rostock.de (J.O.);
benjamin.milkereit@uni-rostock.de (B.M.); bin.yang@uni-rostock.de (B.Y.); olaf.kessler@uni-rostock.de (O.K.)
2 Competence Centre ◦ CALOR, Department Life, Light & Matter, Faculty of Interdisciplinary Research,
University of Rostock, Albert-Einstein-Str. 25, 18059 Rostock, Germany
3 Electron Microscopic Centre, University Medical Centre Rostock, Strempelstraße 14,
18057 Rostock, Germany; armin.springer@med.uni-rostock.de
4 Fraunhofer Research Institution for Large Structures in Production Engineering IGP, Albert Einstein-Str. 30,
18059 Rostock, Germany; karina.nowak@igp.fraunhofer.de
* Correspondence: michael.reich@uni-rostock.de; Tel.: +49-381-498-9490




Received: 19 December 2019; Accepted: 5 February 2020; Published: 19 February 2020


Abstract: The mechanical properties after age hardening heat treatment and the kinetics of related
phase transformations of high strength AlZnMgCu alloy AA 7068 were investigated. The experimental
work includes differential scanning calorimetry (DSC), differential fast scanning calorimetry (DFSC),
sophisticated differential dilatometry (DIL), scanning electron microscopy (SEM), as well as hardness
and tensile tests. For the kinetic analysis of quench induced precipitation by dilatometry new
metrological methods and evaluation procedures were established. Using DSC, dissolution behaviour
during heating to solution annealing temperature was investigated. These experiments allowed for
identification of the appropriate temperature and duration for the solution heat treatment. Continuous
cooling experiments in DSC, DFSC, and DIL determined the kinetics of quench induced precipitation.
DSC and DIL revealed several overlapping precipitation reactions. The critical cooling rate for a
complete supersaturation of the solid solution has been identified to be 600 to 800 K/s. At slightly
subcritical cooling rates quench induced precipitation results in a direct hardening effect resulting in a
technological critical cooling rate of about 100 K/s, i.e., the hardness after ageing reaches a saturation
level for cooling rates faster than 100 K/s. Maximum yield strength of above 600 MPa and tensile
strength of up to 650 MPa were attained.

Keywords: AlZnMgCu; AA 7068; aluminium alloys; precipitation hardening; dilatometry;

differential fast scanning calorimetry (DFSC); differential scanning calorimetry (DSC)

1. Introduction
AlZnMgCu alloys attain highest strength levels among aluminium alloys and are particularly
used in aviation applications [1]. The strength is typically adjusted through precipitation hardening,
respectively, age hardening [2,3]. The age hardening process comprises three major steps: Solution
annealing, quenching, and ageing [1]. The solution heat treatment intends to dissolve relevant alloying
element atoms and is setting the initial structural condition for the following cooling or quenching.
Due to quenching the alloying element atoms shall be retained in solution resulting in a supersaturation.
From this unstable state, nanometre sized precipitates form during the final ageing treatment, which
hinder the dislocation glide and thereby increase the strength of the alloy. To obtain the intended

Materials 2020, 13, 918; doi:10.3390/ma13040918 www.mdpi.com/journal/materials

Materials 2020, 13, 918 2 of 14

properties, for instance high strength or sufficient ductility, a number of heat treatment parameters for
the age hardening process must be carefully selected. This includes reasonable heating and cooling
rates as well as temperatures and durations of the solution annealing and the ageing step. In terms of
technological application the cooling step is critical as undesired quench induced precipitation during
cooling might severely reduce the strength as well as the ductility after ageing [4–6]. Until the end of
the last century, the process parameters of age hardening were predominantly acquired by empirical
ex situ tests [1,7]. Nowadays the kinetics of quench induced precipitation can be investigated by in
situ differential scanning calorimetry (DSC, e.g., [8]), as well as using in situ dilatometry (DIL) [9].
Conventional DSCs are limited to cooling rates of about 5 K/s. However, high strength aluminium
alloys have critical cooling rates for a complete supersaturation of the solid solution of up to some
hundreds of K/s. Basically DIL should allow the in situ cooling analysis of precipitation with cooling
rates of up to about 100 K/s [9]. Thus, in this work, the DIL method is further developed in order to
enlarge the cooling rate range and to enhance the sensitivity. Another sophisticated technique for the
kinetic analysis of quench induced precipitation is the chip-sensor based differential fast scanning
calorimetry (DFSC [10]) using a differential reheating method (DRM) [11].
This work intends to enable a sound selection of appropriate heat treatment parameters for the
high alloyed aluminium alloy AA 7068 based on the kinetic analysis of the related solid-solid phase
transformations. The major objective was adjusting a high strength to allow the usage of AA 7068
as a material for solid self-piercing rivets [12,13] applied for medium to high strength AlMgSi and
AlCuMg sheet material [14]. Therefore, the kinetic behaviour of dissolution and precipitation reactions
during heating, cooling, and annealing was investigated by in situ DSC as well as in situ differential
DIL in a broad range of scanning rates and durations. In addition, the DFSC-DRM was used to reveal
the critical cooling rate. Quench induced precipitations of slower cooling rates were investigated by
SEM. The continuous cooling precipitation diagram of AA 7068 has been recorded. Moreover, the
hardness and tensile properties of selected states including ageing/stretching after quenching were
tested resulting in a complete set of suitable age hardening heat treatment parameters, which obtain
high strength for AA 7068.

2. Materials and Methods

The investigated aluminium wrought alloy AA 7068 (AlZn7.5Mg2.5Cu2) was supplied as a rolled
bar stock with a diameter of 8 mm. The mass fractions of alloying elements are given in Table 1
(analysed by optical emission spectrometry). This composition was chosen as it allows to adjust very
high strength. It should be noted, that the Cu content is at the lowermost boundary, whereas the Zn
content is slightly above the boundary. The initial heat treatment condition was T651, which is solution
annealed, quenched, stress relieved by stretching, and artificially aged.

Table 1. Mass fraction of alloying elements of the investigated material in %.

Mass Fraction in %
Aluminium Alloy
Si Fe Cu Mn Mg Cr Zn Ti Zr
AA 7068 0.052 0.076 1.60 0.008 2.96 0.013 8.43 0.022 0.084
AMS 4331 ≤0.12 ≤0.15 1.6–2.4 ≤0.10 2.2–3.0 ≤0.05 7.3–8.3 ≤0.10 0.05–0.15

DSC was performed to analyse the kinetic dissolution and precipitation behaviour during heating
of the initial conditions T651 and “as-quenched”. In particular, a temperature was determined which is
suitable for solution annealing. Additionally, the kinetic behaviour of quench-induced precipitation was
analysed via DSC. Two instruments were used, namely a Setaram 121 DSC (Setaram, Caluire-et-Cuire,
France) and a Pyris Diamond DSC (PerkinElmer, Waltham, MA, USA). The DSC experiments were
performed as described in references [15,16]. For each condition at least four samples and two baselines
were measured.
Materials 2020, 13, 918 3 of 14

In addition to DSC, the quench-induced precipitation was analysed by DIL, which basically allows
detection of precipitation reaction in Al alloys [17–19], although their volume effect is very small.
Materials 2020, 13, x FOR PEER REVIEW 3 of 14
Beyond that it enables an extension of the accessible cooling rates [9]. In this work the dilatometric
method has further
dilatometric methodbeenhas developed
further beentodeveloped
enlarge thetoassessable
enlarge thecooling rate cooling
assessable range and
ratetorange
increase
and theto
sensitivity for the detection of precipitation during cooling.
increase the sensitivity for the detection of precipitation during cooling.
Dilatometric
Dilatometric measurements
measurements were were performed using aa Bähr
performed using Bähr 805A/D
805A/D instrument
instrument with
with aa linear
linear
variable differential transducer sensor
variable differential transducer sensor (LVDT),(LVDT), utilising silica push rods to transmit the change inin
silica push rods to transmit the change
length/elongation,
length/elongation,see seeFigure
Figure1.1.

Figure 1. Schematic setup of the Bähr dilatometer in alpha-mode.

Figure 1. Schematic setup of the Bähr dilatometer in alpha-mode.
The specimens (hollow cylinders in 4 mm diameter, 10 mm length, and thickness 1 mm or 0.2 mm)
The specimens (hollow cylinders in 4 mm diameter, 10 mm length, and thickness 1 mm or 0.2 mm)
were inductively heated by an induction coil. An additional inner hollow coil is perforated and thus
were inductively heated by an induction coil. An additional inner hollow coil is perforated and thus
used for inert gas quenching. For cooling up to 10 K/s nitrogen and for more rapid cooling helium
used for inert gas quenching. For cooling up to 10 K/s nitrogen and for more rapid cooling helium
were used as inert and cooling gases. Hollow rods allow an internal cooling by slight gas volumes
were used as inert and cooling gases. Hollow rods allow an internal cooling by slight gas volumes
flow through hollow samples. The temperature of the sample was controlled by S-type thermocouples
flow through hollow samples. The temperature of the sample was controlled by S-type
spot-welded to the sample surface. Heating to solution annealing at 480 ◦ C for 10 min was carried out
thermocouples spot-welded to the sample surface. Heating to solution annealing at 480 °C for 10 min
at 2 K/s. Cooling was realised at five different cooling rates, namely 0.1, 1, 10, 30, and 100 K/s. For each
was carried out at 2 K/s. Cooling was realised at five different cooling rates, namely 0.1, 1, 10, 30, and
experiment, a fresh sample was used. The same heat treatments were done on 99.9995% pure Al
100 K/s. For each experiment, a fresh sample was used. The same heat treatments were done on
(Al5N5) alternating with AA 7068 samples. It is expected that the Al5N5 samples only show the thermal
99.9995% pure Al (Al5N5) alternating with AA 7068 samples. It is expected that the Al5N5 samples
expansion of the aluminium matrix together with the push rods and can therefore be used for baseline
only show the thermal expansion of the aluminium matrix together with the push rods and can
measurement. The baseline is new for dilatometry of Al alloys and allows to correct the signal for the
therefore be used for baseline measurement. The baseline is new for dilatometry of Al alloys and
basic
allowsexpansion
to correctofthe
thesignal
aluminium
for thematrix lattice. Beyond
basic expansion of thethat, it contains
aluminium all additional
matrix influences
lattice. Beyond that, itof
the device on the dilatometry result, e.g., additional length changes of the silica push
contains all additional influences of the device on the dilatometry result, e.g., additional length rods. The strain ε
ischanges
evaluated dividing
of the the length
silica push rods. change ∆Lεby
The strain is the initial sample
evaluated dividing length L0 , ε =
the length ∆L/L0ΔL
change . For
byathe
reduction
initial
and a smoothing of raw data a stepwise linear description with 500 points between 50 and 478 ◦ C has
sample length L0, ε = ΔL/L0. For a reduction and a smoothing of raw data a stepwise linear description
been
withused. The pure
500 points Al baseline
between 50 and 478strain εBL been
°C has was subtracted
used. The pure fromAlthe strainε εSwas
samplestrain
baseline (seesubtracted
Figure 2a)
BL
and the strain difference has been derived with respect to temperature (d∆ε/dT). The derivative is
from the sample strain ε (see Figure 2a) and the strain difference has been derived with respect to
further smoothed by the SSavitzky–Golay [20] filter method of the software OriginPro 2018 which
temperature
performs (dΔε/dT).
a local The derivative
polynomial regressionistofurther smoothed
determine by the Savitzky–Golay
the smoothed [20] filter method
value for each temperature point.
of the
The software
resulting curveOriginPro 2018
can finally which
be used forperforms a local
evaluation, polynomial
e.g., of regression
precipitation start andto enddetermine the
temperatures
smoothed value for each temperature point. The resulting curve can finally be used for evaluation,
(Figure 2b). As a consequence of the baseline subtraction, the zero-level of the derivative curve should
e.g., of precipitation start and end temperatures (Figure 2b). As a consequence of the baseline
now be the reference value (being zero), allowing easy determination of the characteristic temperatures.
subtraction, the zero-level of the derivative curve should now be the reference value (being zero),
For each condition at least three samples and two baselines were measured by DIL. The dilatometric
allowing easy determination of the characteristic temperatures. For each condition at least three
experiments in general showed a very good reproducibility. Hence, one representative curve of three
samples and two baselines were measured by DIL. The dilatometric experiments in general showed
cooling tests was used for further discussion.
a very good reproducibility. Hence, one representative curve of three cooling tests was used for
As the in situ detection of precipitation during cooling by dilatometry is limited to cooling rates
further discussion.
of 100 K/s, even faster cooling experiments were performed by Differential Fast Scanning Calorimetry
(DFSC) applying the Differential Reheating Method (DRM) introduced in [11,21]. In addition, hardness
Materials 2020, 13, 918 4 of 14

testing (HV1) in the as-quenched state as well as in the artificially aged state were performed. Tensile
testing was performed using round tensile specimen with a screw head, a diameter of 5 mm and a
gauge length of 25 mm, in accordance with DIN 50125:2016 form B following the test method B out of
DIN EN 2020,
Materials ISO 13,
6892-1:2017. (Testing machine Z50, ZwickRoell GmbH & Co.KG, Ulm, Germany).
x FOR PEER REVIEW 4 of 14

derivative of strain difference in 10-2/K

0.00 0.0020
0.0 Cooling with 0,1 K/s cooling 0.1 K/s
-0.02 0.0018 red: Savitzky-Golay smoothing
-0.2

strain difference in %
-0.04 0.0016
-0.4 -0.06 0.0014
strain in %

-0.6 -0.08 0.0012

-0.10 0.0010
-0.8
Al5N5 -0.12 0.0008
-1.0 7068 0.0006
-0.14
-1.2 0.0004
-0.16
-1.4 0.0002
-0.18
0.0000
-1.6 -0.20
0 100 200 300 400 500 -0.0002
0 100 200 300 400 500
(a) temperature in °C
temperature in °C (b)

Figure2.2.Evaluation
Figure Evaluation procedure
procedure dilatometric analysis:Alloy
dilatometric analysis: Alloy7068,
7068, solution
solution annealing
annealing °C ◦10min,
480480 C 10 min,
and
andcooling
coolingrate
rate 0.1
0.1 K/s. (a) raw
K/s. (a) raw strain
straindata
dataofofsample
sampleand
andbaseline
baseline measurement
measurement as well
as well as strain
as strain
differenceofofthese
difference thesetwotwoand
and (b)
(b) derivation
derivation of
ofstrain
straindifference.
difference.

The microstructural
As the in situ detection changes caused by
of precipitation quench-induced
during precipitation
cooling by dilatometry were
is limited to investigated
cooling rates by
of 100 K/s,
scanning even faster
electron cooling (SEM).
microscopy experiments
Sampleswereforperformed
SEM and by Differential
energy FastX-ray
dispersive Scanning Calorimetry
spectroscopy (EDS)
(DFSC) were
analysis applying the Differential
cold embedded Reheating
in epoxy Method
resin and (DRM)
prepared introduced
by standard in [11,21].
grinding andIn addition,
polishing with
hardness testing (HV1) in the as-quenched state as well as in the artificially aged
water-free, ethanol-based lubricants [22]. For final polishing, a 0.05 µm oxide polishing suspension was state were
performed.
used. Tensile testing
The embedded was performed
and polished samples wereusing round tensile
mounted on a SEMspecimen with adhesive
carrier with a screw head, a
conductive
diameter
carbon tapeof(Co.
5 mm and a gauge
PLANO, length
Wetzlar, of 25 mm,Additionally,
Germany). in accordancethe
with
SEMDIN 50125:2016
samples wereform
coatedB following
with carbon
the test method B out of DIN EN ISO 6892-1:2017. (Testing machine Z50, ZwickRoell
(Leica EM SCD 500, Co. Leica, Wetzlar, Germany). SEM samples were analysed by field emission GmbH & Co.KG, SEM
Ulm, Germany).
®
(MERLIN VP Compact, Co. Zeiss, Oberkochen, Germany) equipped with an EDS detector (XFlash
The microstructural changes caused by quench-induced precipitation were investigated by
6130, Co. Bruker, Berlin, Germany). For SEM imaging, an Everhart–Thornley-type high-efficiency
scanning electron microscopy (SEM). Samples for SEM and energy dispersive X-ray spectroscopy
secondary electron detector (Co. Zeiss, Oberkochen, Germany) was used, applying an accelerating
(EDS) analysis were cold embedded in epoxy resin and prepared by standard grinding and polishing
voltage of 5 kV. The particle size was determined with the Imagic ims Client software. Representative
with water-free, ethanol-based lubricants [22]. For final polishing, a 0.05 µm oxide polishing
areas of the samples were analysed and mapped for elemental distribution on the basis of the EDS
suspension was used. The embedded and polished samples were mounted on a SEM carrier with
spectra
adhesive obtained
conductiveat acceleration
carbon tapevoltages of upWetzlar,
(Co. PLANO, to 20 kVGermany).
and evaluated by the QUANTAX
Additionally, the SEM samples ESPRIT
Microanalysis software (version 2.0, Co. Bruker, Berlin, Germany).
were coated with carbon (Leica EM SCD 500, Co. Leica, Wetzlar, Germany). SEM samples were
The different
analysed analysis techniques
by field emission SEM (MERLIN used in Compact,
® VP correlation with
Co. theOberkochen,
Zeiss, particular applied
Germany) heat treatments
equipped
are illustrated
with an 2020,
Materials EDS13, in Figure
detector 3.
(XFlash
x FOR PEER REVIEW6130, Co. Bruker, Berlin, Germany). For SEM imaging, an Everhart– 5 of 14
Thornley-type high-efficiency secondary electron detector (Co. Zeiss, Oberkochen, Germany) was
used, applying an accelerating voltage of 5 kV. The particle size was determined with the Imagic ims
Client software. Representative areas of the samples were analysed and mapped for elemental
distribution on the basis of the EDS spectra obtained at acceleration voltages of up to 20 kV and
evaluated by the QUANTAX ESPRIT Microanalysis software (version 2.0, Co. Bruker, Berlin,
Germany).
The different analysis techniques used in correlation with the particular applied heat treatments
are illustrated in Figure 3.

Figure3.3.Schematic
Figure Schematic temperature-time
temperature-time profile.
profile.

3. Results and Discussion

3.1. DSC Heating Curves: Determination of an Appropriate Solution Annealing Temperature

The sequence of endothermic dissolution and exothermic precipitation reactions during heating
of alloy 7068 has been analysed for various heating rates. Previous DSC melting tests with 0.3 K/s
Materials 2020, 13, 918 5 of 14
Figure 3. Schematic temperature-time profile.

3. Results and Discussion

3.

3.1.
3.1. DSC Heating Curves: Determination of an Appropriate Solution Annealing Temperature
Temperature
The
The sequence
sequence ofof endothermic
endothermic dissolution
dissolution and
and exothermic precipitation
precipitation reactions
reactions during
during heating
heating
of
of alloy
alloy 7068
7068 has
has been
been analysed
analysed for
for various
various heating
heating rates.
rates. Previous DSC melting tests with 0.3 K/s K/s
revealed the solidus temperature to be 491 ◦ C [23]. Thus, the temperature range analysed extends
revealed the solidus temperature to be 491 °C [23]. Thus, the temperature range analysed extends
from ◦ C (sometimes from −70 ◦ C) ◦ C,
from 2020 °C (sometimes from −70 °C) up to 485 °C, keeping the sample in the solid state. improved
state. For improved
readability
readability the DSC curves shown in Figure 4 are shifted, showing the curve with the lowest heating
the DSC curves shown in Figure 4 are shifted, showing the curve with the lowest heating
rate
rate on
on top.
top. Endothermic
Endothermic changes are plotted
plotted upwards.
upwards. The dissolution behaviour of two different
different
initial
initial conditions,
conditions, namely
namely T651
T651 (Figure
(Figure 4a)
4a) and
andas-quenched
as-quenched(Figure
(Figure4b),
4b),was
wasinvestigated.
investigated.

(a) initial condition T651 (b) solution-annealed and water-quenched

0.1 Jg-1K-1

0.1 Jg-1K-1
0.01 K/s 0.01 K/s
endo
excess specific heat capacity

excess specific heat capacity

0.1 K/s 0.1 K/s

1 K/s 1 K/s
endo
heating curve

-100 0 100 200 300 400 500 -100 0 100 200 300 400 500
temperature in °C temperature in °C
Figure 4. Differential
Figure 4. Differential scanning
scanning calorimetry
calorimetry (DSC)
(DSC) heating
heating curves
curves of
of alloy
alloy 7068:
7068: (a)
(a) initial
initial condition
condition
T651
T651 and
and (b)
(b) solution-annealed
solution-annealed and
and water-quenched.
water-quenched.

The heating DSC curves reveal a complex dissolution and precipitation behaviour with alternating
endo and exothermic regions. The single reactions severely overlap, making a deconvolution of single
reactions difficult. However, between the two different initial conditions substantial differences are
seen particularly within the temperature region below 300 ◦ C. The T651 state mainly shows dissolution
reactions of the initial aged state (precursors of η-Zn2 Mg, S-Al2 CuMg, etc.). The quenched state
shows alternating precipitation and dissolution reactions according to the precipitation sequence of
AlZnMgCu alloys (also precursors of η-Zn2 Mg, S-Al2 CuMg, etc.). Above about 300 ◦ C dissolution
processes of more stable precipitates (η-Zn2 Mg, S-Al2 CuMg, etc.) dominate in both cases. A suitable
solution annealing temperature can therefore be determined from the slow warming DSC curves
of 0.01 K/s, since the DSC signal of the main solution process drops abruptly towards zero at high
temperatures of 472 ◦ C. This means, the dissolution process is finished, and the heating rate specific
solvus temperature is about 472 ◦ C. It can therefore be assumed that an equilibrium solvus temperature
precipitation sequence of AlZnMgCu alloys (also precursors of η-Zn2Mg, S-Al2CuMg, etc.). Above
about 300 °C dissolution processes of more stable precipitates (η-Zn2Mg, S-Al2CuMg, etc.) dominate
in both cases. A suitable solution annealing temperature can therefore be determined from the slow
warming DSC curves of 0.01 K/s, since the DSC signal of the main solution process drops abruptly
towards zero at high temperatures of 472 °C. This means, the dissolution process is finished, and the
Materials 2020, 13, 918 6 of 14
heating rate specific solvus temperature is about 472 °C. It can therefore be assumed that an
equilibrium solvus temperature is slightly below 472 °C. In the next step the influence of soaking
within thebelow
is slightly temperature
472 ◦ C. range between
In the next solvus
step the and solidus
influence temperature
of soaking was
within the assessed by
temperature subsequent
range between
DSC cooling curves.
solvus and solidus temperature was assessed by subsequent DSC cooling curves.

3.2.
3.2. DSC
DSC Cooling
Cooling Curves:
Curves: Assessing
Assessingthe
theInfluence
Influence of
of Solution
Solution Annealing
Annealing Temperature
Temperature and Duration
To
To allow a sound selection of the solution annealing temperature and duration, DSC cooling
curves
curves after
aftersolution
solutiontreatment
treatmentatat470,
470,475,
475,and
and 480 °C◦were
480 C were compared,
compared, seesee
Figure 5a. 5a.
Figure In general, the
In general,
DSC cooling curves are very similar. They show at least two overlapping precipitation
the DSC cooling curves are very similar. They show at least two overlapping precipitation reactions reactions
(details are given
given ininChapter
Chapter3.3.).
3.3.).However,
However,inindetail
detailit it
cancan
bebe seen
seen that
that after
after solution
solution treatment
treatment at
at 470
◦ C for
470 °C 20
formin,
20 min, an instantaneous onset of precipitation with onset of cooling occurs.
an instantaneous onset of precipitation with onset of cooling occurs. This indicates an This indicates
an incomplete
incomplete dissolution
dissolution of main
of the the main alloying
alloying elements,
elements, actingacting as nuclei
as nuclei for precipitation
for precipitation [24]. To[24]. To
ensure
ensure a complete
a complete solution,solution, the temperature
the temperature for the solution
for the solution treatment treatment
was set atwas ◦ Cat
475set 475 °C
(−480 ◦ C).
(−480 °C).
In the In
next
the next step the soaking duration was varied between 10 and 60 min, see Figure
step the soaking duration was varied between 10 and 60 min, see Figure 5b. The DSC cooling curves 5b. The DSC cooling
curves
from any from any soaking
soaking durationduration weresimilar.
were highly highly 30 similar.
min at30475 min◦ Cat
(10475
min°Cat(10 ◦ C) at
480min 480defined
were °C) were as
defined
solutionas solution duration.
treatment treatment duration.

(a) 0.5 (b) 0.5

470°C 20´ 475°C 10´
475°C 30´ 475°C 30´
0.4 480°C 20´ 0.4 475°C 60´
excess cp in Jg-1K-1

0.3
excess cp in Jg-1K-1

0.3

0.2 0.2

0.1 0.1
exo exo
0.0 0.0
cooling curve
-0.1 -0.1
100 200 300 400 500 100 200 300 400 500
temperature in °C temperature in °C

Figure 5. DSC cooling curves of alloy 7068 (cooling rate 0.01 K/s): (a) different solution annealing
Figure 5. DSC cooling curves of alloy 7068 (cooling rate 0.01 K/s): (a) different solution annealing
temperatures and (b) different solution annealing times.
temperatures and (b) different solution annealing times.
3.3. Kinetic Behaviour of Quench Induced Precipitation Analysed by DSC, DIL, DFSC, and Hardness Testing
3.3. Kinetic Behaviour of Quench Induced Precipitation Analysed by DSC, DIL, DFSC, and Hardness Testing
The kinetic behaviour of quench-induced precipitation during cooling with various rates from the
The kinetic
solution behaviour
annealing of quench-induced
temperature precipitation
has been analysed by DSCduring cooling
(Figure with
6a) and various rates
dilatometry from 6b).
(Figure the
solution annealing temperature has been analysed by DSC (Figure 6a) and dilatometry (Figure
The cooling curves in Figure 6 are shifted and the slowest rates are shown on top. Be aware of the 6b). The
cooling
differentcurves in Figure
rates used 6 are
in DSC shifted
and and the slowest rates are shown on top. Be aware of the different
dilatometry.
rates The
usedDSCin DSC and dilatometry.
cooling curves show exothermic precipitation when the signal exceeds the zero level,
since only exothermic precipitation reactions can occur during cooling. DSC cooling (Figure 6a) reveals
a relatively sharp onset after slight undercooling (below the solvus) followed by a very broad reaction
shoulder, similar to other highly concentrated AlZnMgCu alloys [8,25]. This undercooling is in the
range of few K at the slow rate of 0.01 K/s and increases with increasing cooling rate. At the rate of 3 K/s
the onset of precipitation is seen at about 440 ◦ C and thus the undercooling is about 30 K. The second
broad exothermic precipitation during slow cooling ranges from about 465 to 180 ◦ C. This indicates a
severe overlap of multiple exothermic precipitation reactions. With an increasing cooling rate, the peak
area decreases, i.e., in total the precipitation reactions are supressed. The critical cooling rate cannot be
quantified by conventional DSC measurement, since at the fastest rate of 3 K/s a precipitation enthalpy
of about 20 J/g is still measured.
Materials 2020, 13, 918 7 of 14
Materials 2020, 13, x FOR PEER REVIEW 7 of 14

(a) cooling after SHT 475 ° C 30 min (b) cooling after SHT 480 ° C 10 min

0.003 K/s

1x10-5
0.01 K/s
0.1 K/s
0.5 Jg-1K-1

0.03 K/s

difference deviation in K-1

1 K/s
excess specific heat capacity

0.1 K/s
decrease in length
DL-
10 K/s
0.3 K/s

increase in length DL+

1 K/s
100 K/s
exo
3 K/s

0 100 200 300 400 500 0 100 200 300 400 500
temperature in ° C temperature in ° C
Figure 6.6.Cooling
Coolingcurves
curves of alloy
of alloy 7068 7068 after solution
after solution annealing:
annealing: (a) (b)
(a) DSC and DSC and (b) dilatometry.
differential differential
dilatometry.
Differential dilatometry has been found to be useful for studying higher cooling rates of up to
100 K/s
TheandDSC thus can extend
cooling curves the show bandwidth
exothermic of precipitation
analysed cooling whenrates. Figureexceeds
the signal 6b compares
the zero cooling
level,
dilatometry curves in the range of 0.1 to 100 K/s. A relative shortening
since only exothermic precipitation reactions can occur during cooling. DSC cooling (Figure 6a) of the sample is plotted upwards.
It has been
reveals described
a relatively earlier
sharp [9] that
onset aftercertain
slight precipitation
undercoolingreactions
(below the cause an increase
solvus) followedin the
by atotal
veryvolume
broad
and thusshoulder,
reaction an elongationsimilarof the sample,
to other while
highly precipitation
concentrated of a different
AlZnMgCu phase
alloys might
[8,25]. Thiscause a shrinkage
undercooling is
of the volume
in range ofand fewtherefore a shortening
K at the slow rate of 0.01of the
K/ssample. That means,
and increases opposing volume
with increasing changes
cooling rate. caused
At the rate
by3precipitation
of K/s the onsetofofdifferent phases
precipitation is might
seen atsuperimpose
about 440 °Ceach and other andundercooling
thus the a signal value is of zero30
about does not
K. The
necessarily
second broadmean that no reaction
exothermic is occurring.
precipitation duringTherefore
slow coolingcooling dilatometry
ranges from about bears465
theto same180 issue like
°C. This
heating DSC
indicates in terms
a severe of itsofsignal
overlap multipleinterpretation
exothermic[16]. precipitation reactions. With an increasing cooling
rate, For cooling
the peak areaofdecreases,
aluminium alloy
i.e., 7068,
in total thedilatometry
precipitation reveals veryare
reactions similar resultsThe
supressed. compared to DSC.
critical cooling
rate cannot be quantified by conventional DSC measurement, since at the fastest rate of 3 K/s in
For instance, comparing the cooling curves of 0.1 K/s, reveals a sharp onset of precipitation resulting a
a peak of about
precipitation 435 ◦ C in
enthalpy ofboth
about cases.
20 J/gAlso, the inflection towards the shoulder 410 ◦ C are revealed as
for measured.
is still
characteristic
Differentialtemperature
dilatometry in both
has cases. For 1 K/s
been found theuseful
to be general forcurve shapes
studying are similar,
higher cooling too. In addition,
rates of up to
dilatometry
100 K/s and helps
thus canto discuss
extendtemperature
the bandwidth regions of overlapping
of analysed coolingreactions.
rates. Figure The latter is observed
6b compares at a
cooling
cooling rate of
dilatometry 10 K/sinwhere
curves the dilatometer
the range of 0.1 to 100 reveals
K/s.a Afirst precipitation
relative shortening region of between
the sample about is 440 and
plotted
320 ◦ C causing
upwards. It hasabeen
shortening
describedof the sample.
earlier Thiscertain
[9] that is followed by another
precipitation reaction
reactions region
cause an between
increase in about
the
320 tovolume ◦
150 C,and causing
total thusanother elongation.
an elongation of the sample, while precipitation of a different phase might
causeAs a very positive
a shrinkage of theaspect
volume cooling dilatometry
and therefore helps to substantially
a shortening of the sample.expand the cooling
That means, rate
opposing
spectrumchanges
volume up to 100 causedK/s. by
Forprecipitation
high alloyedofAlZnMgCu alloy AA
different phases 7068
might even this fast
superimpose cooling
each other rate
andof a
100 K/svalue
signal revealsofa zero
precipitation
does notsignal betweenmean
necessarily aboutthat
380 and 220 ◦ C. This
no reaction finding shows
is occurring. that thecooling
Therefore upper
critical cooling
dilatometry bearsratethe
is above
same issue100 K/s.
like Faster
heating cooling
DSC inofterms
hollow of samples
its signalmay be possible[16].
interpretation in quenching
dilatometry.
For cooling However,
of aluminiumthe decreasing
alloy 7068, stiffness with reveals
dilatometry deceasing verywall thickness
similar resultscomplicates
compared tolength DSC.
measurements.
For Thus, fastthe
instance, comparing scanning
cooling calorimetry
curves of 0.1(DFSC) and hardness
K/s, reveals testing
a sharp onset of after cooling as
precipitation well as
resulting
after cooling ◦ C for 16 h) are used to identify the upper critical cooling rate.
in a peak of and
about artificial
435 °Cageing
in both (130cases. Also, for the inflection towards the shoulder 410 °C are
revealed as characteristic temperature in both cases. For 1 K/s the general curve shapes are similar,
too. In addition, dilatometry helps to discuss temperature regions of overlapping reactions. The latter
 up to 100 K/s. For high alloyed AlZnMgCu alloy AA 7068 even this fast cooling rate of 100 K/s reveals a
precipitation signal between about 380 and 220 °C. This finding shows that the upper critical cooling rate
is above 100 K/s. Faster cooling of hollow samples may be possible in quenching dilatometry.
However, the decreasing stiffness with deceasing wall thickness complicates length measurements.
Thus, fast scanning calorimetry (DFSC) and hardness testing after cooling as well as after cooling and
Materials 2020, 13, 918 8 of 14
artificial ageing (130 °C for 16 h) are used to identify the upper critical cooling rate.
Figure 7 shows the results obtained by the DFSC-DRM method. Three individual samples were
usedFigure 7 showsthe
to determine theprecipitation
results obtained by theduring
enthalpy DFSC-DRMcooling method.
at variousThree individual
rates. One issue samples
of thewere
used
used
methodto determine
is, that thethe precipitation
samples are veryenthalpy
small with during cooling
only about 100atbyvarious
100 byrates.
40 µ mOne issue of no
3. Currently thedevice
used
method is, thatwhich
is available the samples
allowsare tovery small with
determine onlysamples’
these about 100mass.by 100Thus,
by 40 theµm3normalization
. Currently no deviceof the
iscalorimetric
available which allows
signal, to determine
which is normally thesecarried
samples’outmass.via Thus, the normalization
the sample mass, hasofto thebecalorimetric
based on
signal, which isInnormally
assumptions. this work,carried
we wereout via
ablethetosample
measure mass,
the has to berate
cooling based
of 3onK/s
assumptions.
in both, DSC Inand
this DFSC.
work,
we
Wewere able toused
therefore measure the cooling
the value of the rate of 3 precipitation
specific K/s in both, DSC and DFSC.
enthalpy We therefore
obtained used the value
by DSC (integral of the
of the specificDSC
normalised precipitation enthalpy
curve, average obtained
value = 21 by DSC
J/g) to (integral
calibrateofthe thespecific
normalised DSC curve,
precipitation average
enthalpies
value
measured= 21 byJ/g)DFSC.
to calibrate
This isthethespecific
raw DFSC precipitation enthalpies
signal of each samplemeasured by DFSC.
in µ J as shown This 7a
in Figure is the raw
divided
DFSC
by thesignal of each
specific sample inenthalpy
precipitation µJ as shown of 21 in Figure 7a divided
J/g obtained by DSCby the
to specific
estimateprecipitation
the individual enthalpy
DFSC
sample
of masses. With
21 J/g obtained by DSC these estimated
to estimate the masses
individual andDFSC
the cooling rates applied,
sample masses. the DFSC
With these signals
estimated masses are
normalised
and the cooling andrates
the values
applied, plotted in Figure
the DFSC 7b are normalised
signals obtained. Figureand the 7 shows
valuessubstantial
plotted in deviations
Figure 7b
between
are obtained. the Figure
three samples. Those deviations
7 shows substantial deviations arebetween
related theto the
threetiny sample
samples. size deviations
Those and probably are
originate from slight variation in the chemical composition (by local segregations) or from variations
related to the tiny sample size and probably originate from slight variation in the chemical composition
in the
(by localdensity of nucleation
segregations) or fromsites (dispersoid
variations in the distribution and particularly
density of nucleation the amount
sites (dispersoid of grain
distribution
boundary
and surface
particularly thewithin
amount theofsamples). Nevertheless,
grain boundary surfaceall samples
within the show a similar
samples). trend andall
Nevertheless, thesamples
specific
precipitation
show a similarenthalpy
trend andisthe dropping towards zeroenthalpy
specific precipitation with increasing
is dropping cooling
towardsrate. The
zero upper
with critical
increasing
coolingrate.
cooling rate The
was upper
foundcritical
to be about
cooling600rate
K/swasto 800 K/s,towhich
found be aboutallows a complete
600 K/s supersaturation
to 800 K/s, which allows of a
the alloy.supersaturation
complete This is one of the of highest
the alloy.critical
This iscooling
one of rates ever found
the highest forcooling
critical an aluminium
rates ever alloy,
found compare
for an
[11,21,25,26].
aluminium alloy, compare [11,21,25,26].

(a) 8 sample 1 (b) 25 sample 1

7
absolute enthalpy change in µ J

specific enthalpy change in J/g

sample 2 sample 2
6 sample 3 20
sample 3
5
15 critical cooling rate
4 critical cooling rate
» 600–800 K/s » 600–800 K/s
3 10
2
1 5
0
0
-1
1000 100 10 1 1000 100 10 1
cooling rate in K/s cooling rate in K/s
Figure 7.7. Differential
Figure Differential fast
fastscanning
scanningcalorimetry
calorimetry(DFSC)
(DFSC)results
resultsofofthree
threesamples
samples7068:
7068:(a)
(a)absolute
absolute
enthalpychange
enthalpy change and
and (b)
(b) specific enthalpy change.
change.

To
Tocompare
comparethethe
precipitation
precipitationenthalpies fromfrom
enthalpies DSCDSCand DFSC with hardness
and DFSC values of
with hardness the distinct
values of the
cooling
distinctrates Figure
cooling 8 isFigure
rates plotted.
8 isAverage,
plotted. minimum, and maximum
Average, minimum, values are shown
and maximum values forarethe enthalpies.
shown for the
The averaged
enthalpies. hardness
The averaged values are plotted
hardness valueswith
are their standard
plotted deviations
with their standard from six singlefrom
deviations indentations.
six single
The
indentations. The cooling rate axis is scaled logarithmic with high cooling rates on the left alignment
cooling rate axis is scaled logarithmic with high cooling rates on the left in order to allow in order to
with
allowthealignment
time axis with
of thethe
following
time axiscontinuous cooling precipitation
of the following diagramprecipitation
continuous cooling in Figure 9. diagram in
The
Figure 9. specific precipitation enthalpies determined for slow cooling rates are quite high (up to 40 J/g),
which is reasonable for this high concentrated alloy. With increasing cooling rate, the total precipitation
enthalpy is decreasing as precipitation is suppressed. This behaviour is common for age hardening
aluminium alloys [25]. As indicated above the upper critical cooling rate, which is the slowest cooling
rate at which any quench-induced precipitation is supressed completely, is about 600 to 800 K/s. This is
the highest value ever reported for an aluminium wrought alloy, which is reasonable, as AA 7068 is a
quite high concentrated alloy. However, the hardness after ageing indicates a technical critical cooling
rate to be just about 100 K/s. This means, that after artificial ageing (130 ◦ C for 16 h) there is no further
increase in hardness with an increasing cooling rate above 100 K/s. This can be attributed to a certain
direct hardening effect of quench induced precipitation for cooling rates above 100 K/s. This indeed is
confirmed by the as-quenched hardness, which first rises with increasing cooling rate, then shows
 Materials 2020, 13, 918 9 of 14

a maximum/plateau for cooling rates between about 30 to 300 K/s. For even faster cooling rates the
as-quenched hardness finally drops again when approaching the physical upper critical cooling rate.
A direct hardening contribution of quench-induced precipitation has already been shown for alloy
AA7150 [27]. In detail for AA7150 very similar hardness values in similar cooling rate regions were
achieved. For AA7150 the direct hardening effect of quench induced precipitation was confirmed by
tensile
Materials 2020,tests
13, xinFOR
thePEER
as-quenched
REVIEW condition and is primarily referred to the quench-induced precipitation9 of 14
Materials 2020,
of thin Zn13,and
x FOR
CuPEER REVIEW platelets [27,28].
rich Y-phase 9 of 14
250 SHT DSC 475° C 30 min 50
250 SHTSHT DFSC 475°CC30
DSC 475° 10min
min 50
SHT DFSC 475° C 10HV
min after 130°C 16h
200 HV after 130°C 16h 40

J/gJ/g
200 water
40
HV1

in
quenching
150
water Dh from DSC 30
HV1

in
DhDh
quenching
150 Dh from DSC 30
hardness

HV as quenched

change
hardness

HV as quenched

change
100 20
100 20

enthalpy
enthalpy
(ice) water
50 quenching
(ice) water
10
50 quenching 10
Dh from DFSC
0 Dh from DFSC 0
010 10 10 10 10 10 10 10 10 0
4 3 2 1 0 -1 -2 -3 -4

104 103 102 cooling

101 10 0
rate in10
K/s-1
10-2 10-3 10-4
cooling rate in K/s
Figure 8. Enthalpy change of alloy 7068 from quench induced precipitation as well as hardness in as-
Figure
Figure
quenched 8. Enthalpy
8. Enthalpy
and change
agedchange of alloy
of alloy
condition. 7068
7068 fromquench
from quenchinduced
induced precipitation
precipitation as
aswell
wellas
ashardness
hardnessinin as-
as-quenched and aged condition.
quenched and aged condition.
SHT DFSC & DIL: 480° C 10 min SHT DSC: 475° C 30 min
500
SHT DFSC & DIL: 480° C 10 min SHT DSC: 475° C 30 min
500
a
400
in °inC ° C

a
400 a a+X
a a+X
temperature

300
temperature

300
200
200
CCR
100
CCR0 4 12 21 37 38 39
100
0
201 4
198 12
194 21
132 37
99 38
73 39
53
0 201 198 194 132 99 73 53
00.1 1 10 100 1000 10000 100000
0.1 1 10 100 1000 10000 time in s
100000
HV1 after ageing 130 ° C 16 h time in s
DHV1
htotalafter
in J/gageing 130 ° C 16 h
Dhtotal in J/g
Figure 9. Continuous Cooling Precipitation Diagram of AA 7068 from DSC, dilatometry (DIL), and
Figure 9. Continuous Cooling Precipitation Diagram of AA 7068 from DSC, dilatometry (DIL), and
DFSC. Within the reaction α → α + X, the X comprises any quench induced phase.
FigureWithin
DFSC. 9. Continuous
the reaction α  Precipitation
Cooling Diagram of
α + X, the X comprises anyAA 7068 from
quench DSC,
induced dilatometry (DIL), and
phase.
AllWithin
DFSC. coolingthe
results
reaction α
(DSC, DIL,
α + DFSC,
X, the Xand hardness)
comprises anyare summarised
quench inducedinphase.
the continuous cooling
precipitation
The specific (CCP) diagramenthalpies
precipitation of aluminium alloy 7068for
determined inslow
Figure 9, being
cooling a powerful
rates are quitetool for (up
high practical
to 40 J/g),
heat
The treatment
specific as well as for
precipitation heat treatment
enthalpies simulation
determined [29].
for slow cooling rates are quite high
which is reasonable for this high concentrated alloy. With increasing cooling rate, the total precipitation (up to 40 J/g),
which is reasonable
enthalpy for this
is decreasing high concentrated
as precipitation alloy. With
is suppressed. increasing
This cooling
behaviour rate, thefor
is common total
ageprecipitation
hardening
enthalpy is alloys
aluminium decreasing as indicated
[25]. As precipitation
aboveis suppressed. This cooling
the upper critical behaviour is which
rate, common forslowest
is the age hardening
cooling
aluminium alloys [25]. As indicated above the upper critical cooling rate, which is
rate at which any quench-induced precipitation is supressed completely, is about 600 to 800 K/s. the slowest cooling
This
rate at which any quench-induced precipitation is supressed completely, is about
is the highest value ever reported for an aluminium wrought alloy, which is reasonable, as AA 7068600 to 800 K/s. This
was confirmed by tensile tests in the as-quenched condition and is primarily referred to the quench-
induced precipitation of thin Zn and Cu rich Y-phase platelets [27,28].
All cooling results (DSC, DIL, DFSC, and hardness) are summarised in the continuous cooling
precipitation (CCP) diagram of aluminium alloy 7068 in Figure 9, being a powerful tool for practical
heat treatment
Materials as well as for heat treatment simulation [29].
2020, 13, 918 10 of 14

3.4. Microstructure of Different Cooled Samples through Scanning Electron Microscopy (SEM)
3.4. Microstructure of Different Cooled Samples through Scanning Electron Microscopy (SEM)
The microstructures of alloy 7068 after various cooling rates have been analysed by SEM-
The microstructures
secondary electron images of and
alloyselected
7068 after various
images arecooling
shownrates have 10
in Figure beenforanalysed by SEM-secondary
cooling rates ranging from
electron
0.01 to 10 images and selected
K/s. Next images are matrix
to the aluminium shown which
in Figure 10 fordark,
appears cooling rates
there areranging
two main fromtypes
0.01 to
of
10 K/s. Nextappearing
precipitates to the aluminium
brighter:matrix which on
Precipitates appears
grain dark, there are
boundaries andtwo main types
precipitates of precipitates
inside the grains.
appearing
Particularlybrighter:
for the Precipitates
slower rates, onthe
grain boundaries
grain boundary and precipitates
particles become inside the large.
fairly grains.Within
Particularly
those
for the slower rates, the grain boundary particles become fairly large. Within
particles often even brighter particles are visible, which are probably Fe rich primary particles. those particles often even
brighter
Obviously, particles are visible,
the quench induced which are probably
particles Fe rich primary
become smaller particles.
with increasing Obviously,
cooling the quench
rate. This applies
induced
especiallyparticles
to the grainbecome smaller
boundary with increasing
precipitates. However, cooling rate. This
precipitates applies
inside especially
the grains from to the
0.01 tograin
1 K/s
boundary precipitates. However, precipitates inside the grains from 0.01 to 1 K/s first
first appear to increase in size and particle density, but with further increasing rate these particles appear to increase
in size and
become particle density,
substantially smaller.but withthis
Within further
work increasing
no detailed ratephasethese particles
analysis becomeout.
was carried substantially
According
smaller.
to previous Withinworkthis work no detailedofphase
precipitation the analysis
following wasphases
carried out.
can According
be expected: to previous work
S-Al2CuMg,
precipitation
η-Mg(Zn,Al,Cu) of 2the following
, ZnCu phases can
rich Y-phase be expected:
platelets S-Aland
[8,27,28,30], 2 CuMg, η-Mg(Zn,Al,Cu)
potentially 2 , ZnCu rich
also precursor/cluster
Y-phase
formation platelets
[31,32].[8,27,28,30],
The latter twoandparticle
potentially
typesalso
are precursor/cluster
probably too small formation [31,32]. by
to be detectable TheSEM.
latterThus,
two
particle
it seemstypes
likelyare
thatprobably too small
the majority to be
of the detectable
particles by SEM.
observed Thus,
in our SEMit seems
worklikely that the
are either S-Almajority
2CuMg or of
the particles observed
η-Mg(Zn,Al,Cu) 2 phase inparticles.
our SEMEDS workindicates
are either anS-Al 2 CuMg or
enrichment ofη-Mg(Zn,Al,Cu)
Zn, Mg, and Cu 2 phase
in anyparticles.
quench
EDS
inducedindicates
particleanobserved
enrichment of Zn,
in SEM Mg, and
(Figure Cu thus
11) and in any quench
does induced
not allow particle observed
to distinguish between in theSEM
two
(Figure 11) and thus does not allow to distinguish between the two
phases. Within the CCP diagram (Figure 9) all quench induced precipitation reactions arephases. Within the CCP diagram
(Figure
summarised9) all to as α 
quench α + X. precipitation reactions are summarised to as α → α + X.
induced

Figure 10. SEM-secondary electron images of alloy 7068: Cooling rates (a): 0.01 K/s; (b): 0.1 K/s; (c) 1
K/s; and (d) 10 K/s.
Materials 2020, 13, x FOR PEER REVIEW 11 of 14

Materials 13, SEM-secondary

2020,10.
Figure 918 electron images of alloy 7068: Cooling rates (a): 0.01 K/s; (b): 0.1 K/s; (c) 11
1 of 14
K/s; and (d) 10 K/s.

Figure
Figure 11.
11. SEM-EDS
SEM-EDS maps
maps for
for Zn,
Zn, Mg,
Mg, and
and Cu
Cu of
of alloy
alloy 7068
7068 after
after cooling
cooling at
at (a)
(a) 11 K/s
K/s and
and (b)
(b) 10
10 K/s.
K/s.

3.5. Mechanical
3.5. Mechanical Properties
Properties after
after Ageing:
Ageing: Tensile
Tensile Tests
Tests
The tensile
The tensileproperties
propertiesofofaluminium
aluminiumalloyalloy 7068
7068 were
were tested
tested forfor different
different ageing
ageing durations
durations at 130
at 130 °C.
◦ C.Plain quenched samples were compared with samples which have seen an additional stretching
Plain quenched samples were compared with samples which have seen an additional stretching of 3%
of 3%quenching
after after quenching and ageing,
and prior prior ageing, see Figure
see Figure 12. For 12.each
For condition
each condition six samples
six samples were were
tested.tested.
The
The results
results fromfrom tensile
tensile tests,
tests, yield
yield strength
strength Rp0,2
Rp0,2 , tensilestrength
, tensile strengthRR , andfracture
m,m and fracture elongation
elongation are are
summarised in Table 2 for T6 and T651 states. Next to the average values the standard
summarised in Table 2 for T6 and T651 states. Next to the average values the standard deviations out deviations
outsix
of of tests
six tests
are are shown.
shown. Maximum
Maximum yield
yield strength
strength andand tensilestrength
tensile strengthwere
wereobtained
obtainedfor
for an
an ageing
ageing
◦
treatment at 130 C for 10 h following on the stretching operation (T651). These strengths are slightly
Materials 2020, 13, 918 12 of 14

lower than for aluminium alloy 7068 in the T6511 state according to AMS4311. However, it must be
considered that that the Cu content of the investigated batch of alloy 7068 is at the lowermost boundary
(Table 1). Beyond that an additional stretching after aging (T6511) may result in a further increase
in yield strength compared to T651. The fracture elongation is relatively high in all cases, while it is
decreasing with increasing ageing duration.

Figure 12. Mechanical properties of alloy 7068: With or without stretching 3%, different ageing times 1,
3, 10, and 30 h at 130 ◦ C.

Table 2. Parameters from tensile testing: Heat treatment with or without stretching.

T6 475 ◦ C 30 min, Quenching T651 475 ◦ C 30 min, Quenching in

in Water, 130 ◦ C 30 h Water, Stretching 3%, 130 ◦ C 10 h
yield strength in N/mm2 605 ± 5 626 ± 2
tensile strength in N/mm2 649 ± 2 651 ± 3
elongation at fracture in % 9.8 ± 0.8 10.0 ± 1.2

4. Conclusions
Based on the experimental work comprising differential scanning calorimetry, differential
dilatometry, differential fast scanning calorimetry, scanning electron microscopy, as well as hardness
and tensile tests after age hardening heat treatment of a highly concentrated AlZnMgCu alloy AA 7068
the following findings can be stated:

• An appropriate temperature for solution treatment is determined to be 475 ◦ C (−480 ◦ C), as the
heating rate specific solvus at 0.01 K/s is 470 ◦ C;
• Cooling DSC after soaking durations of 10 to 60 min at 475 ◦ C showed no differences, and a
soaking duration of 30 min was chosen;
• The application of differential dilatometry allows the in-situ detection of quench induced
precipitation in a wide cooling rate range of 0.1 K/s to 100 K/s. At fast cooling of 100 K/s
dilatometry still reveals quenched induced precipitation;
• While DSC only shows one very broad precipitation peak for a wide range of cooling rates,
1
dilatometry indicates the overlap of several quench induced precipitation reactions;
• The physical upper critical cooling rate (CCR) at which quench induced precipitation is fully
suppressed was determined by DFSC analysis to be about 600 to 800 K/s;
Materials 2020, 13, 918 13 of 14

• A technological CCR is found to be ≈ 100 K/s (i.e., hardness after artificial ageing remains on
a saturation level for higher cooling rates including WQ). This indicates, that quench induced
precipitation at cooling rates faster than 100 K/s contributes to a direct hardening effect;
• For the investigated batch of AA 7068 the age hardening treatment comprising of solution heat
treatment 475 ◦ C 30 min, water quenching, stretching 3%, and ageing (130 ◦ C 10 h) results in a
yield strength Rp0.2 of 626 N/mm2 , a tensile strength Rm of 651 N/mm2 , and a fracture elongation
of 10%.

Author Contributions: The experiments were designed by J.O., M.R., O.K. and B.M.; DSC and hardness tests
were performed by J.O.; dilatometry by M.R.; DFSC by B.Y.; The tensile tests were done by K.N.; SEM analyses
was done by A.S. and Marcus Frank. All authors contributed to the discussion of the results. The manuscript was
written by B.M., J.O., and M.R. All authors have revised the text and given their approval for the final version of
the manuscript.
Funding: The project’ Joining of high-strength aluminium alloy structural components by means of aluminium
self-piercing rivets’ is funded by the German Federal Ministry for Economic Affairs and Energy based on a
resolution of the German Parliament (grant number 19432BR) via ‘Stifterverband Metalle e. V.’.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Polmear, I.J. Light Alloys. From Traditional Alloys to Nanocrystals, 4th ed.; Elsevier Butterworth-Heinemann:
Amsterdam, The Netherlands, 2006; ISBN 0750663715.
2. Hornbogen, E. Hundred years of precipitation hardening. J. Light Met. 2001, 1, 127–132. [CrossRef]
3. Hornbogen, E. Precipitation hardening—The oldest nanotechnology. Metall-Berlin 2001, 55, 522–526.
4. Baba, Y. Influence of Additional Elements on the Quench-Sensitivity and Nucleation of Precipitates in
Al-Zn-Mg Alloys. Nippon Kinzoku Gakkaishi 1967, 31, 910–915. [CrossRef]
5. Baba, Y. Quench-Sensitivity and Age-Hardening of Al-Zn-Mg Alloys Containing Trace Elements. Trans. Jpn.
Inst. Met. 1968, S9, 356.
6. Conserva, M.; Fiorini, P. Interpretation of quench-sensitivity in Al-Zn-Mg-Cu alloys. Metall. Trans. A 1973,
4, 857–862. [CrossRef]
7. Shuey, R.T.; Tiryakioğlu, M. Quenching of Aluminum Alloys. Quenching Theory and Technology, 2nd ed.; CRC
Press: Boca Raton, FL, USA, 2011.
8. Starink, M.J.; Milkereit, B.; Zhang, Y.; Rometsch, P.A. Predicting the quench sensitivity of Al-Zn-Mg-Cu
alloys: A model for linear cooling and strengthening. Mater. Des. 2015, 88, 958–971. [CrossRef]
9. Milkereit, B.; Reich, M.; Kessler, O. Detection of Quench Induced Precipitation in Al Alloys by Dilatometry.
Mater. Sci. Forum. 2016, 877, 147–152. [CrossRef]
10. Zhuravlev, E.; Schick, C. Fast scanning power compensated differential scanning nano-calorimeter: 1.
The device. Thermochim. Acta 2010, 50, 1–13. [CrossRef]
11. Yang, B.; Milkereit, B.; Zhang, Y.; Rometsch, P.A.; Kessler, O.; Schick, C. Continuous cooling precipitation
diagram of aluminium alloy AA7150 based on a new fast scanning calorimetry and interrupted quenching
method. Mater. Charact. 2016, 120, 30–37. [CrossRef]
12. Mucha, J. The effect of material properties and joining process parameters on behavior of self-pierce riveting
joints made with the solid rivet. Mater. Des. 2013, 52, 932–946. [CrossRef]
13. Jäckel, M.; Grimm, T.; Niegsch, R.; Drossel, W.-G. Overview of Current Challenges in Self-Pierce Riveting of
Lightweight Materials. Proceedings 2018, 2, 384. [CrossRef]
14. Gröber, D.; Georgi, W.; Sieber, M.; Scharf, I.; Hellmig, R.J.; Leidich, E.; Lampke, T.; Mayr, P. The effect of
anodising on the fatigue performance of self-tapping aluminium screws. Int. J. Fatigue 2015, 75, 108–114.
[CrossRef]
15. Milkereit, B.; Kessler, O.; Schick, C. Recording of continuous cooling precipitation diagrams of aluminium
alloys. Thermochim. Acta 2009, 492, 73–78. [CrossRef]
16. Osten, J.; Milkereit, B.; Schick, C.; Kessler, O. Dissolution and precipitation behaviour during continuous
heating of Al-Mg-Si alloys in a wide range of heating rates. Materials 2015, 8, 2830–2848. [CrossRef]
Materials 2020, 13, 918 14 of 14

17. Hengge, E.; Enzinger, R.; Luckabauer, M.; Sprengel, W.; Würschum, R. Quantitative volumetric identification
of precipitates in dilute alloys using high-precision isothermal dilatometry. Philos. Mag. Lett. 2018,
98, 301–309. [CrossRef]
18. Luckabauer, M.; Hengge, E.; Klinser, G.; Sprengel, W.; Würschum, R. In Situ Real-Time Monitoring of
Aging Processes in an Aluminum Alloy by High-Precision Dilatometry. In Magnesium Technology 2017; The
Minerals, Metals & Materials Series; Solanki, K., Orlov, D., Singh, A., Neelameggham, N., Eds.; Springer:
Cham, Switzerland, 2017; pp. 669–674. [CrossRef]
19. Resch, L.; Klinser, G.; Hengge, E.; Enzinger, R.; Luckabauer, M.; Sprengel, W.; Würschum, R. Precipitation
processes in Al–Mg–Si extending down to initial clustering revealed by the complementary techniques of
positron lifetime spectroscopy and dilatometry. J. Mater. Sci. 2018, 53, 14657–14665. [CrossRef]
20. Savitzky, A.; Golay, M.J.E. Smoothing and Differentiation of Data by Simplified Least Squares Procedures.
Anal. Chem. 1964, 36, 1627–1639. [CrossRef]
21. Zohrabyan, D.; Milkereit, B.; Kessler, O.; Schick, C. Precipitation enthalpy during cooling of aluminum alloys
obtained from calorimetric reheating experiments. Thermochim. Acta 2012, 529, 51–58. [CrossRef]
22. Milkereit, B.; Jonas, L.; Schick, C.; Kessler, O. The continuous cooling precipitation diagram of an aluminum
alloy en AW-6005A. HTM J. Heat Treat. Mat. 2010, 65, 159–171. [CrossRef]
23. Osten, J.; Reich, M.; Milkereit, B.; Kessler, O. Adapting age hardening parameters of high-alloyed 7xxx
aluminium alloys through thermal analysis. In Proceedings of the 16th International Conference on
Aluminium Alloys, Montreal, QC, Canada, 17–21 June 2018.
24. Fröck, H.; Milkereit, B.; Wiechmann, P.; Springer, A.; Sander, M.; Kessler, O.; Reich, M. Influence of
Solution-Annealing Parameters on the Continuous Cooling Precipitation of Aluminum Alloy 6082. Metals
2018, 8, 265. [CrossRef]
25. Milkereit, B.; Starink, M.J.; Rometsch, P.A.; Schick, C.; Kessler, O. Review of the Quench Sensitivity of
Aluminium Alloys: Analysis of the Kinetics and Nature of Quench-Induced Precipitation. Materials 2019,
12, 4083. [CrossRef]
26. Milkereit, B.; Kessler, O.; Schick, C. Quench Sensitivity and Continuous Cooling Precipitation Diagrams.
In Encyclopedia of Aluminum and Its Alloys; CRC Press: Boca Raton, FL, USA, 2019; ISBN 9781351045636.
27. Zhang, Y.; Weyland, M.; Milkereit, B.; Reich, M.; Rometsch, P.A. Precipitation of a new platelet phase during
the quenching of an Al-Zn-Mg-Cu alloy. Sci. Rep. 2016, 6, 23109. [CrossRef]
28. Liu, S.; Li, Q.; Lin, H.; Sun, L.; Long, T.; Ye, L.; Deng, Y. Effect of quench-induced precipitation on
microstructure and mechanical properties of 7085 aluminum alloy. Mater. Des. 2017, 132, 119–128. [CrossRef]
29. Reich, M.; Kessler, O. Quenching Simulation of Aluminum Alloys Including Mechanical Properties of the
Undercooled States. Mater. Perform. Charact. 2012, 1, 104632. [CrossRef]
30. Deschamps, A.; Texier, G.; Ringeval, S.; Delfaut-Durut, L. Influence of cooling rate on the precipitation
microstructure in a medium strength Al-Zn-Mg alloy. Mater. Sci. Eng. A 2009, 501, 133–139. [CrossRef]
31. Schloth, P. Precipitation in the High Strength AA7449 Aluminium Alloy: Implications on Internal Stresses on
Different Length Scales. Ph.D. Thesis, EPFL, Lausanne, Switzerland, 2015.
32. Schloth, P.; Wagner, J.N.; Fife, J.L.; Menzel, A.; Drezet, J.M.; van Swygenhoven, H. Early precipitation during
cooling of an Al-Zn-Mg-Cu alloy revealed by in situ small angle X-ray scattering. Appl. Phys. Lett. 2014,
105, 101908. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).