E ec s o wis channel angula p essing on s uc u e and p ope ies
o bime allic Al/Cu clad composi es
Radim Kocich ⁎
Depa men o Ma e ial Fo ming, Facul y o Me allu gy and Ma e ials Enginee ing, VŠB TU, Os a a 17. Lis opadu 15, 70833 Os a a-Po uba, Czech Republic
HIGHLIGHTS
•TCAP was used o composi e p epa a-
ion o he fi s ime e e .
•Single pass TCAP impa ed significan
g ain efinemen and ha dening.
•Double pass TCAP in oduced homoge-
nisa ion o imposed s ain.
•Well bonded Al/Cu clad composi es
we e success ully ab ica ed.
GRAPHICAL ABSTRACT
abs ac a icle in o
A icle his o y:
Recei ed 15 Sep embe 2020
Recei ed in e ised o m 21 Oc obe 2020
Accep ed 21 Oc obe 2020
A ailable online 24 Oc obe 2020
Keywo ds:
Clad composi e
Fini e elemen me hod
Twis channel angula p essing
TCAP
E ec i e s ain
Residual s ess
This s udy documen s he e ec s o he wis channel angula p essing (TCAP) on he s uc u e and selec ed
p ope ies o Al/Cu clad composi es. P ocessing o he designed clad composi e ia single, as well as double
pass TCAP was pe o med expe imen ally, and simula ed ia he fini e elemen me hod. The de o ma ion beha -
iou p edic ed ia he simula ion was e ified expe imen ally by 3D Mic o-CT scanning, and by s uc u e analy-
ses. The esul s e ealed ha he single pass in oduced significan de o ma ion s eng hening o he Al/Cu
composi e componen s, and a e he second pass, he e ec i e imposed s ain in he Al shea h app oached
he alue o 5. The composi e shea h also exhibi ed mo e o less homogeneous dis ibu ion o esidual s ess o
he lowes obse ed absolu e alues, and he smalles obse ed a e age g ain size - sligh ly abo e 1.5 μm. The
Cu g ains in he s uc u es o he wi es wi hin he ex uded composi es also efined significan ly, especially
a e he second pass, o almos 3 μm. Howe e , he analysed pa ame e s exhibi ed sligh a ia ions ac oss he
c oss-sec ions o he composi e bille s. This can be a ibu ed o phenomena ela ed o he (di e ences in he)
plas ic flow. The in ensi e imposed shea s ain also impa ed inc ease in mic oha dness and changes in he mal
conduc i i y.
© 2020 The Au ho (s). Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
1. In oduc ion
The end in esea ch and inno a ions in a ious comme cial and in-
dus ial b anches, including medicine, ci il enginee ing, anspo a ion,
oil indus y, ae ospace indus y, o ene ge ics is o de elop mode n
s uc u al ma e ials wi h op imized mechanical p ope ies, especially
as ega ds s eng h and duc ili y [1–3]. Among he me hods applied
du ing he esea ch a e echnologies imposing plas ic de o ma ion, he
g ain size o me allic ma e ials using which can e ec i ely be educed
down o he ul a-fine-g ained (UFG) scale, o nano-scale. The challenge
is, howe e , finding he p ocessing pa ame e s p o iding ma e ials ea-
u ing op imum a ios be ween he s eng h and duc ili y. Gene ally, in-
c easing he imposed s ain esul s in dec easing he g ain size, which
Ma e ials and Design 196 (2020) 109255
⁎Co esponding au ho .
E-mail add ess: adim.kocich@ sb.cz.
h ps://doi.o g/10.1016/j.ma des.2020.109255
0264-1275/© 2020 The Au ho (s). Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Con en s lis s a ailable a ScienceDi ec
Ma e ials and Design
jou nal homepage: www.else ie .com/loca e/ma des
impa s inc ease in s eng h, bu dec ease in plas ici y [4,5]. On he
o he hand, nume ous wo ks documen ed ha his gene al p esupposi-
ion can be o e come by modi ying he p ocessing condi ions [6,7].
The up- o-da e esea ch has p o en ha plas ic de o ma ion can be
success ully applied o con ol he de elopmen o s uc u e and p op-
e ies, no only o con en ional alloys [8], bu also o o he me allic
compounds [9]. Mo eo e , op imized SPD p ocessing has been shown
o be able o p oduce equi ed s uc u e pa e ns in me allic ma e ials
[10]. Al hough he in oduc ion o he momechanical p ocessing o a -
ious p oduc ion echnologies enables o efine he s uc u e e y e ec-
i ely, his app oach has ce ain limi a ions. By his eason, he
uncon en ional se e e plas ic de o ma ion (SPD) p ocesses a e
esea ched wi h he aim o p epa e a ious UFG ma e ials. The le el o
(shea ) s ain which can be imposed by SPD me hods is a beyond
ha o con en ional me hods o plas ic de o ma ion, such as olling,
o ging, d awing, and ex usion. Despi e he ac ha he g oup o SPD
me hods has been esea ched o mo e han 4 decades, hei popula i y
is s ill high gi en by hei posi i e e ec s on ma e ials' s uc u e and
p ope ies. Mo eo e , his „ op-down“app oach o s uc u e efine-
men [11] can be applied o p epa e oid and po osi y- ee bulk UFG
ma e ials in olumes su ficien o s uc u al applica ions [12,13].
The (sub)s uc u e de elopmen induced by SPD is complex and can
in ol e a ious de o ma ion mechanisms (e.g. disloca ion slip and
climb, de o ma ion winning, edis ibu ion o elemen s and phases
and hei ans o ma ion, g ains o a ions, g ain g ow h, e c.) ha can
con ibu e o he final g ain efinemen . The ac i a ion o he pa icula
de o ma ion mechanisms and hei ex en , as well as he esul ing (sub)
s uc u e de elopmen , is p ima ily de e mined by he selec ed p ocess
and p ocessing his o y, bu also by he in insic p ope ies o he pa ic-
ula ma e ial(s acking aul ene gy, e c.). Wi h he in en ion o (dynam-
ically) con ol he subs uc u e de elopmen o he p ocessed ma e ials,
qui e a as numbe o me hodsbased on he applica iono simple shea
as he p ima y de o ma ion mechanism has been designed (mo e de-
ailed cha ac e iza ion o he in oduced me hods can be ound in nu-
me ous publica ions, e.g. [14,15]). The gene al goal, ega dless he
applied p ocessing me hod, is o ab ica e he final ma e ial wi h he
lowes possible numbe o p ocessing s eps while simul aneously im-
posing he highes possible homogeneously dis ibu ed s ain.
The de elopmen o new inno a i e ma e ials in oduces new possi-
bili ies o applica ion o he SPD me hods. Whe eas he ea ly expe i-
men al wo ks deal wi h pu e me als and comme cially a ailable
alloys, he con empo a y esea ch in he field o p ocessing me hods o-
cuses on he p epa a ion o g adien s uc u es and lamina es [16], and
composi e ma e ials [17], also deno ed as hyb id ma e ials; he la e is
appealing p ima ily due o he possibili y o combine ma e ials wi h
a he con adic o y p ope ies wi hin a single composi e. Va ious me-
allic sys ems, such as Cu-Ag [18], ca bon nano ubes (CNT-Al) [19], Al-
s eel [20,21], Ti/Cu [22], s eel/Al/Cu [23], Mg/Al/s eel [24], me allic
glasses [25], o Al/Cu [26] can be selec ed o p epa e UFG composi e
ma e ials.
The las men ioned sys em o composi e ma e ials is wi hin he
ocus o esea che s especially o i s p ospec i e usage as mode n elec-
ic conduc o s. The p ima y ad an age o Al/Cu composi es/hyb ids lies
in he possibili y o combine he mos con enien p ope ies o bo h he
componen s, i.e. he high s eng h and excellen elec ical conduc i i y
o Cu wi h he ligh weigh and low cos o Al, in a single mode n ma e-
ial. Cu-based clad conduc o s a e gene ally e y e ficien o applica-
ions using he al e na ing cu en (AC), due o he occu ence o “skin
e ec ,”whe eas Cu-based mul ifilamen conduc o s can be ad an a-
geously used o applica ions using he di ec cu en (DC) and
low- equency AC, due o he addi ional pa allel pa hs o elec ons.
Howe e , he cha ac e is ics o he composi e a e non-negligibly a -
ec ed also by he cha ac e o he componen me als mu ual in e ace.
P e iously published s udies ha e documen ed ha inc easing he em-
pe a u e a ec ing he composi e (e.g. ia pos -p ocess hea ea men )
abo e 300 °C induces o ma ion o b i le in e me allics a mu ual
in e aces, which can consequen ly de e io a e he final p ope ies o
he composi e [26,27]. Ne e heless, he selec ed p epa a ion/p ocess-
ing echnology can influence i s o ma ion and p o ide he possibili y
o op imise he wid h and composi ion o he in e mixing zone, which
e en ually defines he p ope ies o he p epa ed hyb id ma e ial.
The esea ch wo ks epo ing he e ec s o high shea s ain and
high hyd os a ic p essu e on he (sub)s uc u e de elopmen , in e -
di usion, and mechanisms o o ma ion o he mu ual in e aces wi hin
clad composi es and hyb ids a e sca ce. The s udies documen ing he
applica ion o high shea s ain o p epa a ion o composi e ma e ials
epo ed he usage o me hods based on olling, such as asymme ical
oll bonding [28] and accumula i e oll bonding [29], cyclic ex usion
comp ession [30], o high p essu e o sion (HPT) [31]. Se e al wo ks
ha e documen ed he applica ion o me hods based on equal channel
angula p essing (ECAP) [32,33].
This wo k p esen s a de ailed s udy cha ac e izing p ocessing o an
Al/Cu clad composi e using he wis channel angula p essing (TCAP)
me hod. As documen ed by p e iously published s udies [34–40], due
o he h ee independen s ain pa hs, TCAP enables o impose ela i ely
high alues o homogeneous shea s ain in o he p ocessed bille
wi hin a single pass. Howe e , he men ioned s udies only deal wi h
single-phase ma e ials (aluminium, coppe ). Since his me hod has
no been applied o p epa e composi e ma e ials be o e, he ocus o
he he ein p esen ed s udy is no only on he expe imen al p epa a ion
o he composi e, bu also on he expe imen al and nume ical in es iga-
ion o he de o ma ion beha iou o bo h he composi e componen s
du ing mul iple TCAP p ocess ( wo subsequen passes). The (sub)s uc-
u e de elopmen and mechanical and selec ed physical p ope ies o
he p epa ed composi e, as well as he p ocessing pa ame e s such as
in ensi y and dis ibu ion o he imposed s ain, and plas ic flow ela ed
o he occu ence and de elopmen o esidual s ess, a e also
cha ac e ized.
2. Ma e ial and me hods
The aim o he s udy was o p o ide a mo e de ailed desc ip ion o
he TCAP echnology and i s e ec s on p epa a ion o Al/Cu clad com-
posi es. The schema ic depic ion o he TCAP die can be seen in Fig. 1a.
The die geome y was defined by he ollowing angles: φ= 90°,ω=
90°, β= 40°, ψ= 20°. In o de o in es iga e he e ec s o TCAP, wo
consequen passes we e pe o med; mul iple TCAP was pe o med
wi h he applica ion o de o ma ion ou e A(de ails can be ound e.g.
in [40]). The composi e sample consis ed o a ec angula Al shea h
and fi e ci cula ein o cing Cu wi es (p ocessed sample is depic ed in
Fig. 1b).
2.1. Nume ical modelling
The fi s pa o he s udy deals wi h he nume ical analysis o bo h
he consequen TCAP passes. Fo ge NxT comme cial so wa e was
used o analyse he de o ma ion beha iou o he ex uded composi e
and, a he same ime, i s componen me als. The pu pose o his pa
o he esea ch wo k was o s udy he e ec s o he in e ac ion o he di -
e en componen me als and hei plas ic flows on he magni ude o
e ec i e imposed s ain, (in)homogenei y o s ain ac oss he c oss-
sec ion o he ex uded sample, as well as possible occu ence o esid-
ual s ess, i s alues and localiza ion.
Since he p ospec i e applica ion o he ein o ced Al/Cu composi es
is in he elec o echnics, he he mal p ope ies o he componen s in
which a e o he u mos impo ance, FEM, oge he wi h he La ice
Mon e Ca lo (LMC) me hod, we e also applied o in es iga e he he -
mal conduc i i y o he composi e ex uded ia he fi s TCAP pass.
The pa ial assembly used o he he mal simula ion is depic ed in
Fig. 2a. The p edic ed he mal p ope ies we e subsequen ly e alua ed
ia ul asound expe imen al measu emen s, and used as he bounda y
condi ions o he ollowing simula ion o de o ma ion beha iou .
R. Kocich Ma e ials and Design 196 (2020) 109255
2
To enable compa ison o he simula ion wi h he subsequen expe -
imen , he nume ical simula ions we e pe o med using an assembly
he geome ical dimensions o he componen s and mechanical p ope -
ies o he p ocessed me als in which we e iden ical o he expe imen .
Bo h he ex ude and die we e defined as igid pa s. The bille compo-
nen s we e meshed wi h e ahed al elemen s and he clad composi e
sample was cha ac e ized by a mesh wi h 168,742 nodes in o al. Pa ic-
ula ly, he Cu wi es we e meshed wi h a e y fine mesh i.e. each o he
wi es was defined by 20,475 nodes. The mesh a he mu ual Cu/Al in e -
aces was fine han in he emaining olumes o wi es. Since se e e
shea de o ma ion was expec ed o p oceed du ing he ex usion p o-
cess, au oma ic e-meshing was implemen ed. The ex usion speed o
he used hyd aulic p ess was = 3 mm/s, and he ic ion was de e -
mined by he Coulomb ic ion o μ= 0.02. Bo h he pa ame e s we e
selec ed based on ou p e ious s udy [39], mu ual compa ison o nu-
me ical p edic ions and expe imen s in which was pe o med in o de
o acqui e he alues o he bounda y pa ame e s co esponding o
he eal expe imen al p ocess.
The elas ic-plas ic model wi h he New on–Raphson con e gen al-
go i hm was applied o de e mine he pa ame e s in he simula ion. The
s ess-s ain cu es o he expe imen ally used ma e ials (Fig. 2b),
which we e impo ed o he ma e ial flow s ess da abase o he simu-
la ion so wa e, we e acqui ed using a o sion es pe o med a oom
empe a u e wi h he s ain a e o 0.1 s
−1
on SETARAM de ice –a
se o-hyd aulic o sion plas ome e . The de o ma ion beha iou o he
componen s o he p ocessed composi e was desc ibed ia he
Haensel-Spi el equa ion (Eq. (1)),
σ ¼Aem1TTm8εm2em4=ε1þεðÞ
m5Tem6εε
:m3ε
:m7Tð1Þ
whe e εis equi alen s ain, ε
:is equi alen s ain a e, Tis empe a u e,
and Aand m
1
o m
8
a e eg ession coe ficien s. The alues o he indi id-
ual coe ficien s o Cu a e A= 411.19 MPa, m
1
o m
4
a e −0.00121,
0.21554, 0.01472, −0.00935, espec i ely, and m
5
÷m
8
a e 0. The
alues o he indi idual coe ficien s o Al a e A= 151.323 MPa, m
1
o
m
4
a e −0.00253, 0.21142, 0.03177, −0.00654, m
5
÷m
8
a e 0.
The bounda y condi ions in he nume ical simula ion we e defined
by he ini ial empe a u e o 25 °C, and he alues cha ac e izing he
empe a u e beha iou o coppe , aluminium, and he die, i.e. Young's
modulus, Poisson's a io, he mal expansion coe ficien , he mal con-
duc i i y, specific hea , emissi i y and densi y, which we e o Cu de-
fined as he ollowing cons an s ( espec i ely): 111 (GPa), 0.3,
1.7 × 10
−5
(K
−1
), 394 (W/(m K)), 398 (J/kg K), 0.7 and 8960 (kg/m
3
),
and o Al as ( espec i ely): 72 (GPa), 0.3, 2.4 × 10
−5
(K
−1
), 250
(W/(mK)), 1230 (J/kg K), 0.03 and 2800 (kg/m
3
). Mu ual ic ion be-
ween he wo composi e componen s a hei in e aces was defined
by T esca's nolube ic ion (coe ficien mba e = 8.000000e-01).
The de o ma ion beha iou o he Al/Cu clad composi e was mon-
i o ed in wo pe pendicula planes wi hin he ex uded bille o be
able o eliably obse e he plas ic flow o he ma e ial being
Fig. 1. Schema ic depic ion o TCAP die (a), p ocessed Al/Cu clad composi e (b).
Fig. 2. Model used o he he mal conduc i i y simula ion (a); s ess-s ain cu es used o FEM analysis (b); loca ions o analysed planes (c).
R. Kocich Ma e ials and Design 196 (2020) 109255
3
ex uded. The fi s moni o ing g id was supe imposed on a longi udi-
nal plane passing h ough he axis o he sample, whe eas he second
moni o ing plane was loca ed pe pendicula ly o he sample longi u-
dinal axis (see Fig. 2c o bo h). Bo h he plana g ids we e c ea ed
in fixed loca ions wi hin he sample and depic ed he same loca ions
wi hin he bulk o he ma e ial du ing he en i e ex usion p ocess.
In o he wo ds, he unde o med g id wi hin he o iginal bille was de-
o med acco ding o he pe o med de o ma ion o he pa icula ma-
e ial olume. To allow de ailed e alua ion o he in es iga ed
pa ame e s, he supe imposed g ids we e defined by e y fine squa e
cells (0.5 × 0.5 mm
2
).
2.2. P ac ical expe imen
The second pa o he esea ch wo k ocuses on he expe imen al
ealiza ion o he TCAP p ocess. The ma e ials selec ed o manu ac u e
he composi e we e comme cially pu e Al (99.97%) wi h he chemical
composi ion (in w %) o 0.125 Fe, 0.10 Si, 0.020 Zn, 0.020 Cu, 0.015
Mn, 0.015 Mg, 0.015 Ti, and CP Cu (99.97%) wi h he chemical compo-
si ion (in w %) o 0.0074 Ni, 0.0058 Sn, 0.0031 Fe, 0.0030 Zn, 0.0023 Si.
P io o comple ing he composi e, bo h he Al shea h and Cu wi es
we e annealed in a u nace a 500 °C o 30 min in o de o elimina e
possible e ec s o p e ious de o ma ion his o y. The dimensions o he
assembled composi e samples we e iden ical o he dimensions in he
simula ion assembly, i.e. 12 mm × 12 mm squa e c oss-sec ion, and
130 mm leng h. The ex usion was ca ied ou using a hyd aulic p ess
a oom empe a u e (25 °C) and ex usion a e o 5 mm/ s; MoS
2
was used as he lub ican .
To be able o expe imen ally in es iga e he ma e ial flow o he ex-
uded composi es du ing bo h he consequen TCAP passes and com-
pa e he esul s wi h he p edic ed ones, X- ay mic o compu ed
omog aphy (Mic o-CT) was applied. The Mic o-CT measu emen s
we e pe o med using he GE phoenix | ome|x m 300 sys em equipped
wi h a 300 kV/500 W mic o ocus X- ay ube, and DXR250 fla panel de-
ec o wi h high con as o 2048 × 2048 pixels (200 × 200 μm
2
pixel
size). The omog aphic measu emen s we e pe o med a oom em-
pe a u e o 22 °C wi h he exposu e ime o he de ec o o 333 ms;
2700 p ojec ions we e acqui ed o he single 360° o a ion. The applied
pa ame e s we e accele a ion ol age o 200 kV, and X- ay ube cu en
o 140 μA. The oxel esolu ion o he acqui ed CT da a se was 48 μm.
The omog aphic econs uc ion [41] was pe o med using own CT sys-
em e alua ion so wa e (GE phoenix da os|x 2.0), and he analyses and
isualiza ion o hecomposi es' inne s uc u es we e pe o med ia he
VG S udio MAX 3 so wa e.
Mechanical p ope ies o bo h he TCAP-ed composi es a e single
and double pass we e p ima ily in es iga ed ia mic oha dness p ofile
measu emen on ans e sal c oss-sec ional cu s h ough he bille s
using a Zwick/Roell mic oha dness es e . The load was 200 g, and he
loading ime o a single inden was 10 s.
De ailed s uc u e analyses o he TCAP-ed clad composi es we e
pe o med ia scanning elec on mic oscopy (SEM), elec on backsca -
e di ac ion (EBSD) in pa icula . P epa a ions o he samples o he
analyses we e ca ied ou by g inding on SiC pape s ollowed by me-
chanical, and finally elec oly ical polishing. The s uc u e in es iga ions
we e pe o med using a Tescan Ly a 3 FIB/SEM mic oscope equipped
wi h a No dlys Nano EBSD came a. The samples we e scanned unde
he il o 70° wi h he scan s ep o 0.5 μm o bo h he Cu and Al compo-
nen s. The acqui ed scans we e u he e alua ed using he AZ ec C ys-
al and ATEX so wa e [42]. The p esence o esidual s ess wi hin he
p ocessed s uc u es was e alua ed ia EBSD image analysis de e min-
ing he p esence o esidual s ess based on Kikuchi lines pa e nsfinally
p o iding he in e nal g ains miso ien a ions in he scale om 0° (neg-
ligible miso ien a ions, i.e. negligible p esence o esidual s ess), o 15°
(high miso ien a ions, i.e. occu ence o esidual s ess), in ainbow col-
ou scheme.
3. Resul s and discussion
3.1. De o ma ion beha iou
The p ima y ocus o he nume ical analysis was o in es iga e he
de o ma ion beha iou o bo h he componen me als, as well as im-
posed s ain and i s homogenei y, ma e ial plas ic flow, and dis ibu ion
o esidual s ess ac oss he c oss-sec ion o he ex uded bille s. To en-
able compa ison o he nume ical p edic ion and eal expe imen , he
de o ma ion beha iou s o he componen me als du ing bo h he
TCAP passes we e also moni o ed ia Mic o-CT scans (see Fig. 3a and
b o single and double TCAP pass, espec i ely).
As documen ed by he CT scans acqui ed a e bo h, single and dou-
ble TCAP passes, he ex uded composi es exhibi ed di e ences in hei
de o ma ion beha iou s. Bo h he TCAP passes esul ed in significan
plas ic de o ma ion o he Al shea h and Cu wi es. Ne e heless, di e -
ences can be obse ed, especially a he ends o he Cu wi es (depending
on he numbe o passes). A e he fi s pass, he ends o he pai o
wi es si ua ed in he uppe pa o he composi e we e ben upwa ds,
whe eas he ends o he pai o wi es loca ed in i s bo om pa we e
ben downwa ds (Fig. 3a). Fo bo h he uppe and bo om pai s o
wi es, he obse ed bending o igina ed du ing passing h ough he
main de o ma ion zone (MDZ), i.e. h ough he inle o he ho izon al
channel pa . As he esul o he shea s ain a ec ing he bille in he
MDZ, he plas ic flow in he uppe hal o he composi e (c oss-sec ion)
was as e han in he bo om hal . By his eason, he bo om pai o
wi es was ben in he opposi e di ec ion han he uppe pai . Al hough
u he passing o he composi e h ough he emaining pa s o he
Fig. 3. Compa ison o Al/Cu clad composi e shapes ia Mic o-CT a e : fi s pass (a); second pass (b).
R. Kocich Ma e ials and Design 196 (2020) 109255
4
channel esul ed in sligh co ec ion o he bending o he wi es, hese
we e no elimina ed. The second TCAP pass hen in oduced bending
o he majo i y o he Cu wi es in a single di ec ion (Fig. 3b).
The di e en de o ma ion beha iou s o he inpu and ou pu ends
o he Al/Cu clad composi e can be a ibu ed o he influencing ac o s
a ec ing he indi idual bille ends. When passing h ough he MDZ,
he ou pu end o he composi e is in a pe manen con ac wi h he ma-
e ial o he ex usion punch, which gene a es ce ain p essu e on he
bille and in oduces ic ion limi ing he ee mo emen o bo h, he
Cu wi es and Al shea h. On he o he hand, he ee o wa d mo emen
o he inpu end o he clad composi e is no es ic ed du ing and a e
passing h ough he MDZ. Howe e , he ee flow o he Cu wi es was
mo e o less limi ed by he su ounding Al shea h, he limi ing e ec
on he bending o he inse ed wi es o which inc eased du ing he sec-
ond pass due o de o ma ion s eng hening o he Al ma ix. De o ma-
ion s eng hening du ing he second pass also occu ed wi hin he Cu
wi es (accumula ed especially a he ben ends) and con ibu ed o
inc eased flow s ess.
The discussed ma e ial beha iou is in acco dance wi h he cha ac-
e is ic plas ic flows o bo h he componen me als du ing he indi idual
passes. As can be seen in Fig. 4a and b, he p edic ed ma e ial plas ic
flows o he single and double TCAP pass in which a e depic ed, espec-
i ely, ce ain di e ences in he plas ic flow eloci ies o igina ed in he
p ima y de o ma ion zone ( wis ). The wis pa o he die imposed se-
e e shea s ain p ima ily o he Al shea h and caused i s ou e pe iph-
e y egion o flow quicke han he Cu wi es. The de ails depic ed in
Fig. 4a also show he decele a ing e ec s o ic ion occu ing be ween
he ex uded bille and die, and be ween he Al shea h and indi idual
Cu wi es. The MDZ hen in oduced he shea s ain in a g ea e ex en
also o he Cu wi es. The di e ences in he plas ic flows in he indi idual
egions o he ex uded bille (mani es ed as he di e en shapes and
bending o he ends o he indi idual Cu wi es) o igina ed in he MDZ,
he uppe pa o he Al shea h in he icini y o he Cu wi es in which
exhibi ed he endency o flow quicke han he bo om pa o he com-
posi e. In o he wo ds, he shea on he indi idual Al/Cu in e aces man-
i es ed di e en ly in di e en loca ions o he clad composi e. These
di e ences we e mos p obably in oduced by he wis pa o he
die, since ma e ials p ocessed ia he con en ional ECAP p ocess do
no ypically exhibi such localiza ion o he imposed shea s ain. This
supposi ion is suppo ed by he u he p esen ed di e ences in he e -
ec i e s ain and esidual s ess dis ibu ions, as well as by he cha ac-
e o he mu ual in e aces poin ing o he endency o o m mechanical
locking (see e.g. [43,44]). La ypo e al. [45] also obse ed unexpec ed
(helical) shape o he ein o cing Al fib e du ing hei nume ical and ex-
pe imen al s udy o wis ex usion (TE) o a composi e consis ing o Cu
shea h and Al fib e. Acco ding o hei supposi ion, such beha iou can
be a ibu ed o ou -o -plane dis o ions o he sample c oss-sec ion, es-
pecially when pe o ming mul i-pass TE. Despi e he ac ha he he ein
p esen ed composi e consis s o a e e se s acking sequence (Al shea h
and Cu wi es compa ed o Cu shea h and Al wi es), simila endencies
we e obse ed – he endency o he Cu wi es o bend, and also axially
o a e (see Fig. 3 non pe ec ci cula shapes o some Cu wi es c oss sec-
ions in which can be seen). Howe e , subsequen passing h ough he
MDZ caused subs an ial supp ession o he e ec s o p e ious passing
h ough he wis sec ion on he helical-like shapes o he ein o cing
wi es. The FEA confi med ha o ex like flow was pa ially obse ed
on he c oss-sec ions, oo (seen in Fig. 4a and b). Possible easons o
hese findings a e he di e en shapes o p ocessed samples ( ounded
ec angle s. squa e), as well as he posi ion and numbe o he Cu
wi es, and geome ical design o he die.
Fig. 4b also shows ha he fi s TCAP pass p o oked se e e deg ada-
ion o he mesh supe imposed h ough he Al shea h pe pendicula ly
o he composi e axis. In o he wo ds, he in ensi y o plas ic flow in
he in e -a ea be ween he pe iphe al and axial egions o he Al shea h
inc eased subs an ially a e passing h ough he MDZ. The obse ed
plas ic flow g adien , i.e. inhomogenei y, was also he mos p obable
cause o he di e ences in he imposed s ain obse ed be ween he
Al shea h and Cu wi es (see Sec ion 3.2). The esul s demons a e ha
he wis zone a ec ed non-negligibly he Al shea h no only in i s pe-
iphe al egions, bu also in he in e -a ea, i.e. in he icini y o he indi-
idual wi es. Mo eo e , he pa icula a ea also exhibi s no able plas ic
flow in he axial di ec ion al eady in he wis zone. The subsequen
passing h ough he MDZ esul ed in he inc ease in he plas ic flow e-
loci y o he Al shea h. Howe e , he flow was also a ec ed by he ic-
ion a mu ual in e aces o he Al shea h and die channel. On he o he
hand, he e ec o his ac o was educed du ing he second TCAP pass
by he op imized selec ion o hede o ma ion ou e, which also esul ed
in educ ion in he o plas ic flow inhomogenei y o he Cu wi es.
3.2. Imposed s ain
A e he fi s TCAP pass, he ex uded bille c oss-sec ion e iden ly
ea u ed ce ain inhomogenei y o heimposed s ain (Fig. 5a). Whe eas
he Al shea h exhibi ed mo e o less homogeneous s ain dis ibu ion
(Se ~ 2.5), he Cu wi es exhibi ed a ia ions in he imposed s ain alues
ac oss hei c oss-sec ions, which also mani es ed as s ain g adien in
Fig. 4. Analyses o ma e ial's plas ic flow wi hin Al/Cu clad composi e du ing fi s pass: longi udinal g id (a); pe pendicula g id (b).
R. Kocich Ma e ials and Design 196 (2020) 109255
5

he diagonal di ec ion. The p edic ed imposed s ain inhomogenei y
ac oss he Cu wi es c oss-sec ions was ela ed o hei o ien a ions in
ela ion o he indi idual de o ma ion zones du ing passing h ough
he TCAP die (i.e. he selec ed de o ma ion ou e). The acqui ed esul s
a e in acco dance wi h he abo e discussed di e ences in he plas ic
flows o bo h he componen me als, by he e ec o which mu ual slip-
ping o he adjoining componen me als occu ed (especially when
passing h ough he MDZ). The gene ally mo e in ense plas ic flow o
he Al shea h esul ed in he localized endency o he plas ic flow o
he Cu wi es o delay; his di e ence in plas ic flow eloci ies o he
me als p ima ily o igina ed in he highe flow s ess o Cu. E en ually,
he Cu wi es ea u ed significan ly lowe alues o he imposed s ain
han he su ounding Al shea h.
As can be seen in Fig. 5b, he c oss-sec ion o he Al shea h a e he
second TCAP pass also exhibi ed homogeneously dis ibu ed imposed
s ain, al hough he absolu e alues we e significan ly highe (Se ~5).
Ne e heless, he dis ibu ion o he imposed s ain ac oss he Cu
wi es c oss-sec ions exhibi ed subs an ial di e ences when compa ed
o he fi s pass; he applied de o ma ion ou e A esul ed in he inc ease
in s ain homogenei y wi hin he Cu wi es, as well as in he inc ease in
he o e all imposed s ain alues (Se ~ 2.3). In o he wo ds, o a ion o
he bille be o e en e ing he MDZ dec eased he s ain g adien ac oss
he c oss-sec ions o he Cu wi es, which also signifies ha inc ease in
he quali y o he mu ual con ac o bo h he me als a hei in e aces
occu ed (i he quali y o he con ac o bo h he me als was simila
as du ing he fi s TCAP pass, he s ain g adien would ha e been com-
pa able o he g adien occu ing a e he fi s pass).
Mu ual compa ison o he nume ically p edic ed (Figs. 5)andexpe -
imen ally acqui ed (Fig. 3) ex uded clad composi e geome ies showed
sa is ac o y co ela ion, which documen s ha he bounda y condi ions
o he simula ion we e eliably defined.
3.3. S ess s a e
The p edic ed esul s showed ha he e ec s o bo h he passes on
he p esence and dis ibu ion o esidual s ess wi hin he ex uded
composi es we e di e en . As seen in Fig. 6a, al eady he fi s TCAP
pass a ec ed he s ess s a e wi hin he Al/Cu composi e. Du ing passing
h ough he fi s de o ma ion zone ( wis ), he composi e exhibi ed
o ex-like flow (especially in he pe iphe al egions o he ou e Cu
wi es), which consequen ly in oduced inhomogeneous s ess dis ibu-
ion and ensile s ess maxima loca ed p ima ily in he ou e Cu wi es.
These phenomena mo e o less agg a a ed he plas ic flow in he axial
composi e egion and caused he axial wi e o exhibi no significan
s ess maxima. In o he wo ds, he slowe plas ic flow o he Al shea h
in he axial egion o he composi e esul ed in he axial Cu wi e o
Fig. 5. P edic ed imposed s ain a e : fi s pass (a); second pass (b).
Fig. 6. P edic ed esidual s ess dis ibu ion a e : fi s pass (a), second pass (b).
R. Kocich Ma e ials and Design 196 (2020) 109255
6
ea u e lowe alues o ensile s ess and mo e homogeneous s ess dis-
ibu ion when compa ed o he ou e Cu wi es.
The s ess dis ibu ion ac oss he composi e c oss-sec ion changed
a e passing h ough he MDZ, i.e. edis ibu ion o esidual s ess oc-
cu ed as he esul o changes in he plas ic flow in oduced by he
MDZ. A e passing h ough he MDZ, all he Cu wi es ea u ed ensile
s ess wi h inhomogeneous dis ibu ion, especially he uppe pai o
wi es exhibi ed local maxima o ensile s ess (see he de ail in
Fig. 6a). The ac ha he uppe pa s o bille s ex uded ia ECAP-
based me hods exhibi quicke plas ic flow compa ed o hei lowe
pa s is well known [46]. As he esul o he agg a a ed plas ic flow in
he bo om egion o he composi e when passing h ough he MDZ,
he bo om pai o wi es exhibi ed highe alues o ensile s ess and
mo e subs an ial s ess inhomogenei y.
The occu ence o comp essi e s ess wi hin he Al ma ix in he i-
cini y o he Cu wi es documen s he abo e discussed supposi ion o di -
e en eloci ies o plas ic flows o bo h he componen me als. In o he
wo ds, he Al shea h, being mo e suscep ible o he imposed s ain han
he Cu wi es, exhibi s comp essi e s ess in he loca ions adjoining o
he wi es. This phenomenon is he p ima y consequence o ic ion oc-
cu ing a mu ual Al/Cu in e aces. Cu, ea u ing highe flow s ess, has
he endency o agg a a e he plas ic flow o he adjoining Al, by he e -
ec o which cha ac e is ic c oss-like shape o esidual s ess occu -
ence de elops (see he de ail in Fig. 6a). On he o he hand, he
egions in which he Al shea h is in a di ec con ac wi h he walls o
he ex usion channel we e cha ac e ized wi h ela i ely homogeneous
dis ibu ion o s ess o low alues. The eason o his phenomenon was
mos p obably he abo e men ioned occu ence o o ex like flow [40].
The second TCAP pass in oduced significan changes no only in
he esidual s ess dis ibu ion, bu also in i s alues; he s ess s a e
wi hin he Al/Cu clad composi e du ing he second pass di e ed
significan ly al eady a e passing h ough he wis zone. As can be
seen in Fig. 6b, comp essi e s ess s a ed o domina e h oughou
he Al shea h c oss-sec ion, despi e he ac ha he ou e Cu wi es
s ill ea u ed localized maxima o ensile s ess. Con inuing ex usion
o he bille h ough he MDZ esul ed in ce ain esidual s ess ho-
mogenisa ion ac oss he c oss-sec ions o bo h he composi e compo-
nen s, especially he Cu wi es exhibi ed mo e uni o m dis ibu ion o
ensile s ess. Fig. 6b documen s ha , compa ed o he fi s pass, he
c oss-sec ional a eas o he majo i y o Cu wi es exhibi ed p e ailing
ensile s ess, he maximum alues o which we e de ec ed wi hin
he uppe pai o wi es, bu also wi hin he bo om le and axial
wi e. The changes we e p ima ily impa ed by he selec ed de o ma-
ion ou e (A), by he e ec o which he posi ions o all he ou pe-
iphe al wi es changed du ing ex usion. In o he wo ds, he en i e
c oss-sec ion was o a ed by 90° when compa ed o he fi s pass,
i.e. he wi e loca ed in he uppe le posi ion a e he fi s pass was
in he bo om le posi ion a e he second pass, e c., by he e ec
o which homogenisa ion o esidual s ess ac oss he c oss-sec ion
o he en i e composi e bille , i.e. bo h he shea h and wi es, occu ed.
The ac ha bo h he de o ma ion zones, wis and MDZ, con ibu ed
o he educ ion o inhomogenei y o esidual s ess ac oss he c oss-
sec ion o he clad composi e needs o be s essed, oo.
3.4. S uc u e analyses
To e i y he p edic ed esul s, de ailed s uc u e analyses we e sub-
sequen ly pe o med on he eal ex uded bille s (e.g. Fig. 1b). Table 1
depic s he expe imen ally measu ed a e age g ain sizes (a g. max. e -
e diame e in μm) wi hin he indi idual Cu wi es a e bo h, single and
double TCAP passes. The able shows no able di e ences in g ain sizes
be ween he indi idual passes, bu also sligh di e ences de ec ed be-
ween he wi es wi hin he indi idual ex uded composi es.
Bo h hepasses e iden ly esul ed in subs an ial g ain efinemen , as
he o iginal g ain size in he p e-annealed Cu was almos 40 μmina e -
age (see he s uc u e scan shown in Fig. 7a depic ing la ge ec ys al-
lized g ains ea u ing he majo i y o high angle g ain bounda ies –
HAGBs –highligh ed in black). Focusing on he wi es a e single TCAP
pass, hei a e age g ain size a ied be ween 6 and almos 8 μm and
he la ges g ains we e de ec ed in he axial wi e. This phenomenon
can be a ibu ed o he agg a a ed plas ic flow in he axial composi e
egion, as discussed in Sec ion 3.1. The supposi ion is also suppo ed
Table 1
G ain sizes wi hin indi idual wi es a e single and double TCAP.
TCAP 1 pass a g. wi e g ain size [μm] TCAP 2 passes a g. wi e g ain size [μm]
Uppe le Uppe igh Uppe le Uppe igh
6.04 6.58 3.79 4.16
Axial Axial
7.94 4.92
Bo om le Bo om igh Bo om le Bo om igh
6.42 6.12 4.28 3.87
Fig. 7. O iginal p e-annealed Cu s uc u e wi h highligh ed high angle g ain bounda ies (HAGBs) (a); OIM o s uc u e o Al shea h a e second pass wi h eminiscence o o ex-like
flow (b).
R. Kocich Ma e ials and Design 196 (2020) 109255
7
by he alue o he a e age g ain size de ec ed in he axial wi e a e he
second TCAP pass, which was also he la ges o all he wi es wi hin he
bille , howe e smalle han he a e age axial wi e g ain size obse ed
a e he fi s pass. This beha iou demons a es he a ou able e ec
o he bending, i.e. MDZ, on homogenisa ion o he imposed s ain, i.e.
s uc u e. As men ioned p e iously and discussed in de ail e.g. in [38],
he wis de o ma ion zone p ima ily in oduces shea s ain, i.e.
impa s de o ma ion esul ing in g ain efinemen and de o ma ion
s eng hening, o he pe iphe al egions o he ex uded bille and i s e -
ec diminishes owa ds he axis o he bille . On he o he hand, he
bend de o ma ion zone, i.e. MDZ, p ima ily in oduces he shea s ain
in o he axial egion o he ex uded bille as he plas ic flow in i s pe-
iphe al egions is agg a a ed by he e ec s o ic ion and he geome y
o he channel bend. Gi en by hese ac o s, he s ain imposed ac oss
he c oss-sec ion o he axial/pe iphe al ex uded composi e egions
ends o equalise wi h inc easing numbe o TCAP passes, which esul s
in mo e homogeneous dis ibu ion o he imposed s ain (see
Sec ion 3.2), and also mo e homogeneous g ain size dis ibu ion;
Table 1 depic s ha he g ains size ange o he Cu wi es a e he sec-
ond pass was be ween 3.8 and less han 5 μm, he la ges g ains o
4.92 μm in diame e we e again de ec ed in he axial Cu wi e.
As ega ds he indi idual wi es, he smalles g ains we e obse ed in
he uppe le wi e o bo h he ex uded bille s. Howe e , he alues
we e compa able o he g ain size alues de ec ed in he bo om igh
wi es. Simila ly, he a g. g ain size alues de ec ed in he bo om le
and uppe igh wi es o bo h he ex uded composi es we e compa a-
ble. This mino di e ences occu ing in diagonals can p ima ily be a -
ibu ed o he wis de o ma ion zone in oducing he o ex-like
flow, he eminiscence o which was obse ed in he s uc u e o he
Al shea h (see Fig. 7b he o ien a ion image map –OIM –depic ing
he pe iphe al egion o he Al shea h o he bille ex uded ia wo
TCAP passes in which is shown). Fig. 7b also depic s he se e e de-
o med g ains wi hin he Al shea h, he a g. size o which a e he sec-
ond pass dec eased o 1.66 μm (i.e. nea o UFG s uc u e). The ac ha
he obse ed g ain size was he fines wi hin he Al shea h (compa ed o
he Cu wi es) can be a ibu ed o i s lowe flow s ess and highe sus-
cep ibili y o he imposed s ain, bo h esul ing in he highes e ec i e
Fig. 8. G ains miso ien a ions, i.e. esidual s ess, o : uppe le wi e a e fi s pass (a); second pass (b); axial wi e a e fi s pass (c), second pass (d); Al shea h a e second pass (e).
Fig. 9. P edic ed and measu ed he mal conduc i i y alues.
R. Kocich Ma e ials and Design 196 (2020) 109255
8
imposed s ain (see Fig. 5b) and consequen ly o he fines g ain size.
The dec easing g ain size also a ec ed o he mechanical and u ili y
p ope ies, as u he discussed in Sec ions 3.6 and 3.7.
3.5. Residual s ess
Fig. 8a shows he g ains miso ien a ions in he ange om 0° o 15°
poin ing o he p esence o esidual s ess o he s uc u e o he uppe
le wi e a e he fi s pass, while Fig. 8b shows he g ains miso ien a-
ions wi hin he s uc u e o he uppe le wi e a e he second pass.
Bo h he figu es ea u e mo e o less homogeneously dis ibu ed a eas
wi h high miso ien a ions (i.e. ed colou ), depic ing he egions
exhibi ing esidual s ess. Howe e , he p esence o esidual s ess
was highe in he s uc u e o he wi e subjec ed o wo TCAP passes,
which co esponds o he p edic ed esul s (see Fig. 6a and b). Fig. 8c
and d hen show he esidual s ess dis ibu ion in he axial wi e a e
he fi s and second pass, espec i ely. The s ess dis ibu ion wi hin
he axial wi e was compa able a e bo h he passes, howe e , i s mag-
ni ude was sligh ly highe a e he second pass. Mu ual compa ison o
he s uc u es o he pe iphe al wi es and he axial wi es e eals ha
he p esence o esidual s ess was mo e e iden in he pe iphe al
wi es o bo h he passes, which co esponds o he p edic ed esidual
s ess dis ibu ion, as well as he discussed plas ic flow beha iou . Ac-
co ding o he p edic ed esul s, he Al shea h exhibi ed he lowes ab-
solu e alues o esidual s ess (± ela i e o 0) when compa ed o
he Cu wi es, i s dis ibu ion was also mo e homogeneous han wi hin
he wi es. The acqui ed scans e ified he p edic ion, as he Al shea h
o he composi e bille p ocessed ia wo TCAP passes exhibi ed mo e
o less homogeneous esidual s ess dis ibu ion ea u ing a sca ce oc-
cu ence o he loca ions wi h inc eased miso ien a ions (Fig. 8e).
3.6. The mal p ope ies
The p esen ed s udy also in ol ed b ie nume ical and expe imen al
e alua ion o he c oss-sec ional he mal conduc i i y o he composi e
ex uded ia single TCAP pass (used model can be seen in Fig. 2a). Mo e-
o e , he conduc i i ies o he used CP Al and CP Cu we e e alua ed o
e i y co ec se ings o he bounda y condi ions o he nume ical
analysis o de o ma ion beha iou . The measu ed he mal conduc i i-
ies o he CP Al and CP Cu we e 246.8 and 399.2 W/(m·K), espec i ely,
which co esponds well wi h he bounda y condi ions o he pe o med
simula ion (see Sec ion 2.1).
Fig. 9 depic s he compa ison o he c oss-sec ional he mal conduc-
i i y alues p edic ed ia he LMC and FEM me hods, as well as he
alues measu ed expe imen ally in pa allel and se ies se ing. The p e-
dic ed conduc i i ies o he composi e ex uded ia single pass we e
263.3 and 260.8 in a e age, depending on he selec ed me hod (see
Fig. 9), and he measu ed alues we e 264.2 and 253.8 in he pa allel
and se ies se ing, espec i ely. The measu ed alues e iden ly in-
c eased when compa ed o he CP Al, and dec eased compa ed o he
CP Cu. The ela i ely low o e all c oss-sec ional he mal conduc i i y
alue o he ex uded composi e was no only gi en by he mino con-
en o Cu wi hin he bille , bu also by he pe o med p ocessing,
which esul ed in accumula ion o de o ma ion s eng hening and
g ain efinemen , bo h in oducing obs acles such as disloca ions, possi-
ble o ma ion o he mally s able p ecipi a es, as well as la ice dis o -
ions, consequen ly esul ing in dec ease in he mal conduc i i y [47].
3.7. Mic o-ha dness
Las bu no leas , he Vicke s mic oha dness o bo h he ex uded
composi es was e alua ed o expe imen ally in es iga e he e ec s o
he imposed s ain on he mechanical p ope ies. The o iginal HV alues
o he p e-annealed me als we e 58.6 o he Cu, and 37.4 o he Al. The
mic oha dness maps o he composi es a e depic ed in Fig. 10aandb
o he bille s ex uded ia single and double TCAP pass, espec i ely.
The mic oha dness e iden ly inc eased a e bo h he TCAP passes;
he mos no able inc ease was obse ed a e he fi s pass. The HV in-
c ease a e he second pass was less significan , howe e , con a y o
he bille a e single pass, all he Cu wi es wi hin he bille a e double
pass exhibi ed he maximum HV alues. This phenomenon co esponds
o he p edic ed alues o he imposed s ain wi hin he Cu wi es, which
we e non-homogeneous a e he fi s pass ( a ia ions o e ec i e
s ain be ween 0.4 and 1.8, see Fig. 5a), and hen equalized a e he sec-
ond pass (mo e o less homogeneous e ec i e s ain o 2.5, see Fig. 5b).
The inc ease in mic oha dness is also in acco dance wi h he obse ed
g ain efinemen impa ed by he imposed se e e shea s ain.
4. Conclusions
The s udy p esen ed nume ical and expe imen al analysis o Al/Cu
clad composi es p epa ed ia single and double pass wis channel an-
gula p essing (TCAP). Bo h he passes impa ed significan shea s ain
in o bo h he composi e componen s; he maximum e ec i e s ain
eached o 5 wi hin he Al shea h a e he second pass. The Al compo-
nen wi hin his bille also ea u ed he fines a e age g ain size o
sligh ly mo e han 1.5 μm, and he lowes occu ence o esidual s ess.
Nume ical p edic ion e ealed ha he mul iple-pass p ocessing
Fig. 10. C oss-sec ional mic oha dness map a e : fi s pass (a); second pass (b).
R. Kocich Ma e ials and Design 196 (2020) 109255
9