scieee Science in your language
[en] (orig)

Effects of crystallographic texture on subsurface fatigue crack generation in Ti-Fe-O alloy at low temperature

Abstract

Subsurface microcracks developed in a groove-rolled and cold-swaged Ti-Fe-O alloy were characterized to clarify the generation of subsurface fatigue crack. In addition, the effects of crystallographic texture on subsurface crack initiation and growth were discussed. A considerable number of microcracks were detected in the beta grains, alpha grains, and at the alpha-beta interface. The microcracks in the beta grains grew negligibly into the neighboring alpha grains along the basal plane. This was because these grains were oriented with their c-axis almost perpendicular to the loading axis. The {10 (1) over bar0}(alpha) fiber texture prevents the formation of basal facet and its growth on the basal plane. The stress concentration around the microcrack in the beta grains could assist the growth of the microcrack into neighboring alpha grains along the prismatic plane (which is inclined to the loading axis at a suitable angle) or occasionally at a {10 (1) over bar0}(alpha) twist boundary. The {10 (1) over bar0}(alpha) fiber texture assisted microcrack growth, and thereby, formed aligned faCets and yield longer microcrack length. The combination of the shear stress and opening stress on {10 (1) over bar0}(alpha) results in a Mode II or III microcrack and causes microcrack growth on the prismatic plane in the neighboring grain.

Read accessible full text

Effects of crystallographic texture on subsurface fatigue crack generation in Ti-Fe-O alloy at low temperature

Author: Umezawa, Osamu
Publisher: Iron and Steel Institute of Japan
Year: 2022
DOI: 10.2355/isijinternational.ISIJINT-2021-381
Source: https://dspace.vsb.cz/bitstreams/787beb62-c043-40b5-bf78-3a86e41b9a17/download
593
ISIJ In e na ional, Vol. 62 (2022), No. 3, pp. 593–601
h ps://doi.o g/10.2355/isijin e na ional.ISIJINT-2021-381
* Co esponding au ho : E-mail: [email p o ec ed]
© 2022 The I on and S eel Ins i u e o Japan. This is an open access a icle unde he e ms o he C ea i e Commons
A ibu ion-NonComme cial-NoDe i s license (h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
E ec s o C ys allog aphic Tex u e on Subsu ace Fa igue C ack
Gene a ion in Ti–Fe–O Alloy a Low Tempe a u e
Osamu UMEZAWA1,2)* and Weibo LI3)
1) Facul y o Enginee ing, Yokohama Na ional Uni e si y, 79-5 Tokiwadai, Hodogaya, Yokohama, 240-8501 Japan.
2) Cen e o Ad anced Inno a ion Technologies, Vysoká Škola Báňská - Technical Uni e si y o Os a a, 17. lis opadu 15, 708
33 Os a a-Po uba, Czech Republic.
3) G adua e School o Enginee ing, Yokohama Na ional Uni e si y. Now a Cen al Resea ch Labo a o ies, Nihon Pa ke izing
Co., L d., 2784 Okami, Hi a suka, Kanagawa, 254-0012 Japan.
(Recei ed on Augus 18, 2021; accep ed on No embe 8, 2021; J-STAGE Ad ance published da e:
Decembe 29, 2021)
Subsu ace mic oc acks de eloped in a g oo e- olled and cold-swaged Ti–Fe–O alloy we e cha ac e ized
o cla i y he gene a ion o subsu ace a igue c ack. In addi ion, he e ec s o c ys allog aphic ex u e on
subsu ace c ack ini ia ion and g ow h we e discussed. A conside able numbe o mic oc acks we e
de ec ed in he
β
g ains,
α
g ains, and a he
α
-
β
in e ace. The mic oc acks in he
β
g ains g ew negligibly
in o he neighbo ing
α
g ains along he basal plane. This was because hese g ains we e o ien ed wi h
hei c-axis almos pe pendicula o he loading axis. The 10 10
{}
α
ibe ex u e p e en s he o ma ion o
basal ace and i s g ow h on he basal plane. The s ess concen a ion a ound he mic oc ack in he
β
g ains could assis he g ow h o he mic oc ack in o neighbo ing
α
g ains along he p isma ic plane
(which is inclined o he loading axis a a sui able angle) o occasionally a a 10 10
{}
α
wis bounda y. The
10 10
{}
α
ibe ex u e assis ed mic oc ack g ow h, and he eby, o med aligned ace s and yield longe
mic oc ack leng h. The combina ion o he shea s ess and opening s ess on 10 10
{}
α
esul s in a Mode
II o III mic oc ack and causes mic oc ack g ow h on he p isma ic plane in he neighbo ing g ain.
KEY WORDS: i anium alloys; high cycle a igue; ex u e; subsu ace c ack; elec on backsca e di ac-
ion.
1. In oduc ion
Subsu ace a igue c ack ini ia ion si es in nea -
α
ype
and
α
-
β
ype i anium alloys commonly consis o ans-
g anula c ack ( ace ) o c acks ( ace s) unde bo h no mal
cyclic a igue and dwell a igue. Each ace is i ed o he
α
g ain mo phology o mic os uc u e and he quasi-clea age
ace s ha a e mos ly o med on o nea he basal plane.1–7)
Thei inclina ion o he p incipal s ess axis e ealed a wide
ange o a ia ions owing o he ex u e o he alloys,2)
al hough he mic oc ack g ow h o each ma e ial showed
a dependence on he o ien a ion o g ain s uc u e and
p incipal s ess axis.3–5,8) The disloca ion mo emen in he
α
phase is es ic ed o he p ima y slip plane and is ai ly
plana so ha disloca ion a ays on 10 10 1120
{}
a e
piled-up in he icini y o he g ain bounda ies. The local
s ess concen a ion nea he
α
g ain bounda y owing o
he e ogeneous slip may cause subsu ace c ack ini ia ion.9,10)
Fu he mo e, ex u e and mic o ex u e (mac ozone) s ongly
a ec subsu ace c ack ini ia ion and high-cycle a igue
s eng h.11–14) The s ess edis ibu ion be ween “so ”
(plas ically de o med and wi h a high Schmid ac o ) and
“ha d” (elas ically de o med and wi h a low Schmid ac o )
α
g ains owing o s ain incompa ibili y a hei bounda y
is also a ac o ha causes he de elopmen o a ace in he
ha d g ain unde cyclic loading.10,15–19) Mul iple mic oc acks
we e de ec ed in he mac ozone o ien ed wi h hei main
c-axis ex u e componen . Mo eo e , he dominan c ack
g ew in he mac ozones owing o mic oc ack coalescence.
A simila
α
-g ain o ien a ion p omo es cyclic s ain accu-
mula ion, because localized de o ma ion in one g ain can be
s aigh o wa dly accommoda ed in adjacen g ains.14) Thus,
he c ys allog aphic o ien a ions in neighbo ing g ains and
he mic o ex u e play impo an oles in he gene a ion o
mic oc acking owing o he s ain incompa ibili y and s ess
edis ibu ion du ing a igue p ocesses.20)
The mic oc acking on (0001) in a ha d
α
-g ain may
be gene a ed by he combina ion o he opening s ess
and shea s ess componen s unde cyclic loading. This is
because he shea s ess componen p omo es he de elop-
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ 594
men o he slip band on (0001). The no mal di ec ion o
he (0001)
α
quasi-clea age ace s is inclined a 15° o 60°
o he p incipal s ess axis,1,2,5,11) because he no mal s ess
on (0001) is excep ionally high when he ace is no mal o
he applied s ess.10,15) Howe e , he ace o ma ion modes
a lowe s ess le els di e and depend signi ican ly on
he mic os uc u es o he alloys. The mac ozones whose
main c-axis ex u e componen is almos pe pendicula o
he p incipal s ess axis obs uc mic oc ack g ow h.12) The
10 10
{}
α
ace was de ec ed, a he han he (0001)
α
ace .
This was pa icula ly so in he dwell a igue es o highly
ex u ed alloys wi h <0001> pe pendicula o he ensile
axis.19) The p isma ic c ack was consis en wi h a high
Schmid ac o o p isma ic slip. Howe e , i is no e iden
as o which ace ini ia es he o ma ion o c ack.19,21)
The e inemen o he
α
-g ain s uc u e and/o g ain
dis ibu ion wi h andom o ien a ion (elimina ion o mic o-
ex u e) can e ec i ely lowe he maximum s ess concen-
a ion a a bounda y. This is because a sho e slip leng h
esul s in subs an ial imp o emen s in a igue s eng h.3,22) A
Ti-6Al-4V olled ma e ial whose
α
-g ain s uc u e displayed
a andom o ien a ion dis ibu ion owing o a highe wo king
a io exhibi ed a igue s eng h ha was signi ican ly highe
han ha o a o ged ma e ial wi h a colony s uc u e a low
empe a u e.22,23)
The nea -
α
ype Ti–Fe–O alloy which p o ided a cha ac-
e is ic s uc u e o
α
phase wi h dispe sed small amoun o
β
phase clea ly exhibi ed subsu ace a igue c ack ini ia ion a
77 K, whe e he ace was de ec ed as a (0001)
α
c ack.10,24,25)
C yogenic empe a u e a igue could be mos a o able o
cha ac e ize he subsu ace c ack gene a ion, because he
subsu ace c ack ini ia ion and he s ain incompa ibili y
p e ail mo e a lowe empe a u es wi h highe he c i ical
esol ed shea s esses (CRSSs) o indi idual slip sys ems
and hei di e ences.26) Such he e ogeneous slip de o ma-
ion manne e ealed ha he in e nal s ess in he
α
g ain
had accumula ed no mal o (0001)
α
by he ull cons ain s
model analysis, in which esul ed in (0001)
α
c acking.27)
Ti–Fe–O alloy olled ma e ials p o ided a mic o ex u e
(mac ozone) wi h a he clea ly di ided in o eco e ed
(so )
α
g ain and ec ys allized (ha d)
α
g ain egions.
Fu he mo e, he mic os uc u e o he Ti–Fe–O alloy was
modi ied by g oo e- olling and cold-swaging o e ine he
α
-g ain s uc u e and a oid he mic o ex u e (mac ozone).
A ine globula
α
-g ain s uc u e and a weak 10 10
{}
α
ibe
ex u e we e achie ed. Howe e , he e was no imp o emen
in he a igue s eng h a 107 cycles.24) Al hough he (0001)
α
ace p o ided an o igin o he subsu ace c ack ini ia ion,
he 10 10
{}
α
ace s mainly co e ed he c ack ini ia ion si e.
The e o e, in he p esen s udy, he cyclic de o ma ion
s uc u e and mic oc acks de eloped in he modi ied mic o-
s uc u e o he Ti–Fe–O alloy we e cha ac e ized o discuss
he subsu ace c ack gene a ion.
2. Expe imen al
2.1. Ma e ial and Fa igue Tes ed Samples
The a igue es ed samples o a cold-swaged and annealed
nea -
α
ype Ti-0.994%Fe-0.386%O (mass%) ma e ial we e
examined in his s udy. The de ails o he ma e ial we e
desc ibed in he e e ence 24). The ec angula ba s cu om
he hick pla e a e ho - o ged (1 273 K hea ing) and ho -
olled (1 123 K hea ing) we e g oo e- olled (1 023 K hea -
ing, equi alen s ain
η
=1.3) and cold-swaged (
η
=0.48)
in o a ound ba wi h a diame e o 20 mm. The ound ba
was annealed a 1 023 K o 3.6 ks ollowed by ai cooled,
and he cylind ical es pieces we e cu pa allel o he olling
di ec ion (RD) di ec ion.
Fa igue es s we e pe o med using a c yogenic se o-
hyd aulic a igue machine unde load con ol a a s ess
a io, R o 0.01, in liquid ni ogen (77 K) and in ambien
ai (293 K).24) The es condi ions a e gi en in he e e ence
23). Table 1 lis s he ensile p ope ies o he es ma e ial.
Figu e 1 shows maximum cyclic s ess s. he numbe o
cycles o ailu e (N ) da a o he es ma e ial.24) The sam-
ples ha ailed a 77 K and 293 K in highe cycles exhibi ed
subsu ace c ack ini ia ion (solid plo s in Fig. 1). He ein,
hei maximum cyclic s ess was signi ican ly lowe han
0.2% p oo s ess,
σ
0.2. The samples ha ailed by subsu -
ace c ack ini ia ion we e selec ed o he p esen analyses.
2.2. Analyses
The ac u ed samples we e cu o nea he subsu ace
c ack ini ia ion si es pa allel o he di ec ion o ini ial c ack
g ow h as shown in Fig. 2.21) The mic os uc u es a ound
mic oc acks benea h he c ack p opaga ion planes we e
analyzed in he longi udinal c oss-sec ion. The mechanically
and elec ochemically polished su aces we e examined by
scanning elec on mic oscopy (SEM). Elec on backsca e
di ac ion (EBSD) pa e n analysis in SEM was employed
o de e mine he mic os uc u e a ound he mic oc ack.
The hin disk o ansmission elec on mic oscopy
(TEM) analysis was sec ioned om benea h he ac u e
su ace pe pendicula o he p incipal s ess axis and was
mechanically g ound. Each disk diame e was measu ed and
i s maximum cyclic s ess,
σ
max, was calib a ed, because he
Fig. 1. S–N da a o he es ed ma e ial a 77 K and 293 K.
Table 1. Tensile p ope ies o he es ed ma e ial a 77 K and
293 K.
Tempe a u e
(K)
0.2% p oo
s ess,
σ
0.2 (MPa)
Ul ima e ensile
s eng h,
σ
B (MPa)
To al
elonga ion (%)
77 1 257 1 326 11.3
293 659 799 29.5
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ595
Fig. 2. Schema ic illus a ion o analysis samples cu om he
ailed specimens.
Fig. 3. Mic os uc u e o es ma e ial: EBSD image quali y maps
o he RD plane (a) and o a longi udinal sec ion (b).
Fig. 4. Disloca ion s uc u e in
α
and
β
g ains o he es ed ma e-
ial on he RD plane.
Fig. 5. Disloca ion s uc u es in
α
g ains a e cyclic de o ma ion a 293 K: (a)
σ
max = 549 MPa, N = 388 010 cycles,
and (b)
σ
max = 597 MPa, N = 72 424 cycles.
a igue es specimens had an hou glass shape. TEM oils
we e p epa ed by elec ochemical winje polishing a 243 K
in a s i ed solu ion o 6% pe chlo ic acid, 35% bu anol,
and 59% me hanol. A JEM-2100F elec on mic oscope
equipped wi h a double- il goniome e s age was ope a ed
a 200 keV.
2.3. Mic os uc u e
Figu e 3 shows he mic os uc u e o he ma e ial as
EBSD image quali y maps. The ma e ial consis ed o
equiaxed
α
g ains and ine od-like
β
g ains, which we e
andomly dis ibu ed along he RD. The a e age
α
-g ain
size was app oxima ely 13
μ
m. Indi idual disloca ions we e
dis ibu ed in bo h
α
and
β
g ains as shown in Fig. 4, so
ha he equiaxed
α
g ains could be eco e ed ully. The
α
mic os uc u e showed a weak 10 10
{}
ibe ex u e in which
he maximum pole densi y o 10 10
{}
was less han h ee.24)
Fu he mo e, he c-axis is andomly dis ibu ed pe pendicu-
la o he p incipal s ess axis (RD).24)
3. Resul s
3.1. Cyclic De o ma ion S uc u e
A e cyclic de o ma ion a 293 K and 77 K, disloca ion
a ays we e de eloped in he
α
g ains a each s ess le el
as shown in Figs. 5(a) and 6(a). A ays we e blocked a
he g ain bounda ies o he
α
-
β
in e aces as indica ed by
an a ow in Fig. 6(a). Du ing he cyclic de o ma ion, p e-
exis ing mo eable disloca ions ea anged hemsel es and
mul iplied in he
α
g ain. Ce ain disloca ion loops we e
obse ed on he p isma ic plane a low empe a u es (Fig.
6(b)). When a disloca ion sou ce emi s a se ies o disloca-
ions all lying on he same slip plane, disloca ions pile up
behind he leading disloca ion and in e ac elas ically. The
dominan de o ma ion mode is ia he p isma ic slip sys em,
and he glide p ocess is ma kedly plana . Nea ly all he dis-
loca ions in coplana a ays obse ed in he
α
g ains (Figs.
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ 596
5(a) and 6(a)) we e o he sc ew ype as discussed in he
e e ences 9) and 10), and he mo emen was es ic ed o
hei slip planes. Disloca ions can pass h ough he bounda -
ies and in o he neighbo ing g ains a highe s ess le els.
Tangled disloca ions we e also de ec ed in
α
g ains a a
highe s ess le el a bo h es empe a u es (see Figs. 5(b)
and 6(c)). In addi ion, many disloca ions wi h mul iple slip
sys ems we e obse ed in he
β
g ains a bo h es empe a-
u es a any s ess le el (see Fig. 7).
3.2. Mic oc acks in
ββ
G ains
Rega dless o he s ess le el, a conside able numbe
o mic oc acks and oids we e de ec ed nea he ac u e
su ace on he c oss-sec ion o he ailed specimens a bo h
es empe a u es (see Fig. 8). Mos o he mic oc acks we e
de ec ed in he
β
g ains wi hin a dep h o 100
μ
m om
he ac u e su ace. All he de ec ed
β
mic oc acks we e
oughened a he
α
-
β
in e ace. A ew o oids we e also
de ec ed in he
α
g ains and a he
α
-
β
in e aces. Figu e
9 shows a mic oc ack gene a ed and g own a he
β
g ain
bounda y whe e he miso ien a ion be ween he
β
1 and
β
2
g ains was app oxima ely 42° (Figs. 9(b) and 9(d)). The ips
o he mic oc ack a e a he iple poin s o
α
1-
β
1-
β
2 and
α
2-
β
1-
β
2, and he localized plas ic de o ma ion is gene a ed
in neighbo ing
α
1 and
α
2 g ains (Fig. 9(c)). The c-axis o
each
α
g ain is nea ly pe pendicula o he p incipal s ess
axis (Fig. 9(b)). The Schmid ac o s in he indi idual
α
g ains a e high as M=0.46 (in
α
1) and M=0.48 (in
α
2) o
he p isma ic slip sys em (Table 2) and M=0.07 and 0.05,
espec i ely, o he basal slip sys em.
3.3. G ow h o Mic oc acks in o
αα
G ains
Al hough mos o he mic oc acks in he
β
g ains we e
a es ed in he g ains, a ew o he mic oc acks we e
obse ed o ha e g own in o he neighbo ing
α
g ains.
Th ee mic oc acks (labeled as A, B and C in Fig. 10(a))
we e o med in indi idual
β
g ains. Mic oc ack A g ew in o
he neighbo ing g ains
α
3 and
α
4 (Fig. 10(b)). Acco ding
o he c ys allog aphic analysis, he mic oc ack g ew in o
he
α
3 g ain along i s p isma ic plane, which was inclined
Fig. 6. Disloca ion s uc u es in
α
g ains a e cyclic de o ma ion a 77 K wi h
σ
max = 978 MPa, N = 13 480 cycles.
Fig. 7. Disloca ions in
β
g ain o CS ma e ial a e cyclic de o ma ion: (a) 293 K,
σ
max = 513 MPa, N = 2 ×107 cycles,
no b oken, and (b) 77 K,
σ
max = 1 038 MPa, N = 6 590 cycles.
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ597
Fig. 8. Mic oc acks (indica ed by a ows) in he longi udinal sec ion benea h he ac u e su ace: (a) SEI (77 K,
σ
max =
995 MPa, N = 8 455 cycles), (b), (c) magni ied images indica ed in (a), and (d) SEI (77 K,
σ
max = 1 161 MPa,
N = 4 060 cycles).
Fig. 9. Mic oc ack (indica ed by an a ow) o med a he
β
-g ain bounda y in he longi udinal sec ion benea h he ac-
u e su ace (77 K,
σ
max = 995 MPa, N = 8 455 cycles): (a) SEI, (b) IPF o ien a ion map o (a), (c) KAM map o
he a ea indica ed in (b), and (d) miso ien a ion p o ile o line AB. (Online e sion in colo .)
o he p incipal s ess axis a app oxima ely 29° (see Table
3). Bo h he shea and ensile s esses on he c ack plane
we e ela i ely high. Mic oc ack A also g ew along he
10 10
{}
wis g ain bounda y be ween he g ains
α
4 and
α
9, which we e almos pe pendicula o he p incipal s ess
axis. The g ain
α
5 showed a c ys al o ien a ion ha was
almos iden ical o ha o g ain
α
4. Howe e , mic oc ack
B did no p opaga e in o he neighbo ing g ains
α
5 and
α
6. De o ma ion winning is supp essed as oxygen con en
inc eases. Howe e , 1122 1123
{}
comp essi e wins and

ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ 598
10 12 10 11
{}
ensile wins we e de ec ed in he g ains
α
6,
α
8, and
α
9 nea he mic oc ack ips, as shown in Fig. 10(b).
3.4. Localized Slips and Voids a
αα
-g ain Bounda ies
Figu e 11 shows a oid a he
α
-g ain bounda y. The
g ain
α
10 shown in Fig. 11(b) was o ien ed wi h a highe
p isma ic Schmid ac o (0.49). Fu he mo e, se e al p is-
ma ic slip bands de eloped he e. The g ain
α
11 was
o ien ed wi h p isma ic and basal Schmid ac o s o 0.42
and 0.25, espec i ely. The s ess concen a ion a he g ain
bounda y owing o he plana slip bands in
α
10 could
ha e gene a ed he oid. Al hough no mic oc acks we e
gene a ed a he oid, a localized p isma ic slip band was
emi ed om he oid in
α
11. Because he iaxial s ess
concen a ion ields we e de eloped a ound he oids, he
localized slip bands could ha e been de eloped by shea
s ess componen s. Howe e , no slip-o o he bands was
de ec ed unde any s ess le el.
Fig. 10. Mic oc ack (indica ed by an a ow) ha was o med a he
β
g ain bounda y and i s g ow h in o neighbo ing
α
g ains in he longi udinal sec ion benea h he ac u e su ace (77 K,
σ
max = 995 MPa, N = 8 455 cycles): (a)
SEI, (b) IPF o ien a ion map o (a). (Online e sion in colo .)
Fig. 11. Void o med a he
α
-g ain bounda y in he longi udinal sec ion benea h he ac u e su ace (293 K,
σ
max =
539 MPa, N = 1 166 619 cycles): (a) SEI and (b) IQ map. (Online e sion in colo .)
Table 2. Schmid ac o s o p isma ic slips o he
α
1 and
α
2
g ains neighbo ing he mic oc ack be ween
β
g ains
shown in Fig. 9 (
θ
: inclina ion o plane no mal o p inci-
pal s ess axis).
α
g ain Plane Slip sys em Schmid ac o , M
θ
(°)
α
1
P1
a
3
01 10 2110
()


0.08 6.8
P2
a
3
10 10 12 10
()


0.46 55.4
P3
a
3
1100 1 120
()


0.38 74.8
α
2
P4
a
3
01 10 2110
()


0.12 7.7
P5
a
3
10 10 12 10
()


0.48 52.8
P6  a
3
1100 1 120
()


0.36 67.3
Table 3. Schmid ac o s o p isma ic slip o he g ain
α
3 shown
in Fig. 10(b).
Plane Slip sys em Schmid ac o , M
θ
(°)
P7
a
3
01 10 2110
()


0.38 29.2
P8
a
3
10 10 12 10
()


0.44 35.9
P9
a
3
1100 1 120
()


0.06 86.4
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ599
4. Discussion
4.1. De o ma ion Modes
The dominan de o ma ion mode in
α
i anium is he
10 10 1120
{}
slip ope a ion (pa icula ly a lowe em-
pe a u es), and he glide p ocess is ma kedly plana . The
disloca ion mo emen s a e es ic ed o hei slip planes,
and c oss-slip om a p ism plane on o ano he is di icul .
He e ogeneous mic oplas ici y owing o plana slip and
es ic ed sys ems causes s ain incompa ibili y a a bound-
a y. A p ocess o s ess dis ibu ion be ween weak and
s ong g ains owing o s ain incompa ibili y may be he
cause o he gene a ion o (0001) mic oc acks in i anium
alloys unde a igue loading. Co-plana a ays in
α
g ains
may pass h ough he bounda y and in o he neighbo ing
g ain o be blocked a he g ain bounda y whe e s ain
incompa ibili y is de eloped. Because he ma e ial exhibi s
10 10
{}
ibe ex u e, a shea mode is likely o be de eloped
on bo h (0001) and 10 10
{}
planes by s ain incompa ibili y.
The elaxa ion unde he simple shea mode is mo e s aigh -
o wa d han ha unde he ension mode. Fu he mo e, he
localized slip on hose planes unde he simple shea mode
may assis he g ow h o mic oc acks. The e o e, he e ine-
men o he
α
-g ain s uc u e should be accompanied by a
andom dis ibu ion o g ain o ien a ion o an imp o emen
in he a igue s eng h o
α
and
α
-
β
ype i anium alloys in
which subsu ace c ack ini ia ion is dominan .
Al hough ewe de o ma ion wins we e de ec ed in he
α
g ains, 1122 1123
{}
in comp ession and 10 12 10 11
{}
in
ension we e ope a ed nea he mic oc ack ips (Fig. 10(b)).
In gene al, winning occu s in polyc ys alline
α
i anium
and i s alloys because o an inadequa e numbe o slip
sys ems o accommoda e an a bi a y plas ic s ain. Fou
winning modes ha e been epo ed in
α
i anium. He e,
10 12 10 11
{}
and 1121 1 126
{}
a e ex ension wins,
and he 1122 1123
{}
and 10 11 10 12
{}
a e con ac ion
wins.28,29) The p edominan winning mode a oom em-
pe a u e is he 10 12
{}
ensile win, which co esponds o
a o a ion o 85° a ound he axis. A 77 K, he amoun o
10 12
{}
winning inc eased conside ably, and he 1121
{}
ensile and 1122
{}
comp essi e wins we e also obse ed.30)
Because he 1121
{}
win can o m in egions comple ely
cons ained by su ounding g ains owing o a shea s ess
highe han a c i ical alue,31) he 1122 1123
{}
and
10 12 10 11
{}
ope a ions we e de ec ed in his expe imen .
These ope a ions elaxed he in e nal s ess no mal o he
(0001).
4.2. Mic oc acks in
ββ
G ains
Less a en ion has been paid o he ole o he
β
phase in
subsu ace a igue c ack ini ia ion. Ruppen e al.4) p oposed
ha he piled-up disloca ions in
α
-g ain-induced slip in he
neighbo ing
β
phase and he o ma ion o a Co ell clea age
kni e {001} in he
β
phase p o ided a si e o subsu ace
c ack ini ia ion. Howe e , no expe imen al e idence o
clea age c acking has been epo ed. A quali a i e mic o-
mechanical model was also p oposed in which mic oc acks
we e gene a ed in a specimen’s in e io by he coalescence
o shea -induced ca i ies o med a he
α
-
β
in e aces along
localized slip bands.32) Howe e , ca i y nuclea ion was
de ec ed in a lamella s uc u e wi h hin
β
la hs unde high
applied s ess.
Void (o mic oc ack) nuclea ion in
β
g ains was de ec ed
a he
α
-
β
in e aces. I de eloped s ain incompa ibili y
be ween a eco e ed
α
(so ) g ain and a ec ys allized
α
(ha d) g ain in he c oss- olled Ti–Fe–O alloy a e cyclic
de o ma ion.25) The s ain incompa ibili y be ween ec ys al-
lized
α
g ains and
β
g ains could ha e induced mic oc ack
(o oid) nuclea ion, and he ea angemen o disloca-
ions in he neighbo ing eco e ed
α
-g ains enhanced he
s ain incompa ibili y. In he p esen s udy, a conside able
numbe o mic oc acks in
β
g ains we e de ec ed in highly
s ained egions such as nea he ac u e su ace. This is
no wi hs anding ha ew combina ions o so -
α
-g ain-
β
-g ain-ha d-
α
-g ain we e o med in he ma e ial because
o he ibe ex u e. The mic oc ack gene a ed a he iple
poin o
α
-
β
1-
β
2 could ha e p opaga ed along he
β
g ain
bounda y, as shown in Fig. 8(d). The densi y o disloca ions
in he
β
g ains as shown in Fig. 7 was highe han ha in
he
α
g ains.33) Because he hickness o
β
g ains was la ge
han ha o he c oss- olled ma e ial,25) he longe mean
ee slip leng h in he
β
phase could ha e induced a highe
local s ess concen a ion a he
α
-
β
in e ace, pa icula ly
a he iple poin o
α
-
β
1-
β
2 (see Fig. 9). This could ha e
caused he mic o oid gene a ion a he iple poin . The eby,
he opening s ess on he
β
g ain bounda y ha is nea ly
pe pendicula o he loading di ec ion could ha e assis ed
he mic oc ack g ow h om he mico oid, as illus a ed in
Fig. 12. Figu e 13 shows he o ien a ions o he indi idual
β
g ains in which oids we e de ec ed. The dis ibu ed
β
g ains de ia ed om <111> because o he high Schmid
Fig. 12. Schema ic illus a ion o mic oc ack g ow h a
β
-g ain
bounda y om he mic o oid a iple poin .
Fig. 13. O ien a ion dis ibu ion o indi idual oids de ec ed in
β
g ains. (Online e sion in colo .)
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ 600
ac o s o he {011}<111> slip. Mos o he mic oc acks
o med a he
β
g ain bounda ies e ealed a miso ien a ion
o 40°–60°, as shown in Fig. 13. Fu he mo e, Young’s
modulus o
β
-Ti alloys is lowe han ha o
α
-Ti (e.g.
app oxima ely 106 GPa o comme cially pu e i anium),
al hough Young’s modulus depends on he alloy composi-
ion.34) A he elas ic o plas ic de o ma ion s age, he p is-
ma ic slip sys ems in
α
phase we e ac i a ed p e e en ially
a he han slip sys ems in
β
phase owing o lowe CRSS.35)
The s ess in he
α
phase hen eleases and ans e s o he
β
phase. I also suppo s he local yielding in he
β
phase in
he low plas ic-s ain egime.
4.3. In luenceo Tex u eonMic oc ackGene a ion
The s ess concen a ion by he mic oc acks de eloped a
he
β
-g ain bounda y may igge (0001) mic oc acking in
he neighbo ing
α
g ain. Al hough (0001)
α
mic oc acks con-
s i u ed he main c ack in he c oss- olled ma e ial,25) ew
(0001)
α
mic oc acks we e de ec ed in he p esen ma e ial.
The subsu ace c ack ini ia ion si es (pa icula ly a 77 K)
consis ed mos ly o 10 10
{}
α
ace s and we e a he de oid
o (0001)
α
ace s.24) The
α
mic os uc u e in he p esen
ma e ial showed a weak 10 10
{}
ibe ex u e, and he c-axis
o he
α
g ains we e andomly dis ibu ed and we e nea ly
pe pendicula o he p incipal s ess axis. The e o e, bo h he
shea s ess and no mal s ess on he basal plane we e low,
and negligible (0001)
α
mic oc acking could be gene a ed in
he neighbo ing g ain. Al hough a (0001)
α
mic oc ack p o-
ides a s ess ield wi h mul iaxial modes in he neighbo ing
g ain, he combina ion o shea s ess and opening s ess on
10 10
{}
α
may also esul in Mode II o III mic oc acking.
The localized shea s ess as well as he opening (Mode I)
s ess owing o he p incipal s ess is likely o ha e caused
he mic oc ack g ow h on he p isma ic plane in he neigh-
bo ing g ain. In addi ion, Ban ounas e al.12) obse ed ha
α
g ains whose main c-axis ex u e componen was almos
pe pendicula o he loading di ec ion unc ioned as ba -
ie s o ace ed c ack g ow h. Then, 10 10
{}
α
ace s may
appea ,24) and he modi ied mic os uc u e o small
α
-g ains
wi h c-axis andomly dis ibu ed no mal o he loading axis
p e en s he g ow h o (0001)
α
ace .
Bo h su ace and subsu ace c acks o med along a p is-
ma ic plane in
α
g ains o ien ed wi h a e y high Schmid
ac o ha e been epo ed o unc ion as ini ia ion si es.5,21,36)
The mic oc ack o med along a p isma ic plane was in an
α
g ain and p opaga ed negligibly o he neighbo ing g ain.21)
Tha is, he p isma ic c ack o ma ion is consis en wi h he
su ace oughening mechanism a he han he ace o ma-
ion in he specimen in e io . Sacke e al.19) obse ed ha
p isma ic c acks in he in e io o a specimen we e almos
pe pendicula o he ensile loading di ec ion. This can be
explained by he s ess edis ibu ion models such as he
E ans-Bache model.19) In he p e ious s udy,24) he models
o ming 10 10
{}
α
ace o ace s we e p oposed, whe e he
(0001)
α
ace may gi e a igge o gene a e he subsu ace
c ack ini ia ion si e wi h 10 10
{}
α
ace s. In he p esen
wo k, he mic oc acks in he
β
g ains p opaga ed negligibly
in o hei neighbo ing
α
g ains, al hough hey displayed a
high Schmid ac o (see he g ains
α
1 and
α
2 in Fig. 9 and
Table 2). In addi ion, no mic oc acks we e de ec ed on he
p isma ic plane ha was nea ly pe pendicula o he loading
axis. Howe e , mic oc acks we e de ec ed a he 10 10
{}
α
wis bounda y and on he p isma ic planes inclined o he
loading axis, as shown in Fig. 10. Conside ing he edis ib-
u ed shea s ess caused by he s ess concen a ion a ound
he
β
-g ain mic oc ack and he s ain incompa ibili y a he
Fig. 14. Schema ic illus a ion o subsu ace a igue c ack ini ia ion: (a) a model based on he subc ack obse a ion and
(b) a modi ied model a 77 K in e . 24). (Online e sion in colo .)
ISIJ In e na ional, Vol. 62 (2022), No. 3
© 2022 ISIJ601
α
-g ain bounda y, he local p isma ic slip in he neighbo ing
g ain should be selec ed on he p isma ic plane wi h a high
Schmid ac o . Howe e , he in ense p isma ic slip o slip-
o in he neighbo ing g ain may ha e caused he mic oc ack
g ow h. The e o e, a combina ion o high ensile s ess and
shea s ess on he p isma ic plane is necessa y o
β
-g ain
mic oc ack p opaga ion. Thus, a simple explana ion o he
10 10
{}
α
ace o ma ion model is p oposed, as shown in
Fig. 14(a). When he inclined 10 10
{}
α
mic oc ack g ows
in he
α
1 g ain o i s bounda y, he mic oc ack may p opa-
ga e in o he neighbo ing
α
2 g ain along he p isma ic plane
because o he c oss slip. The combina ion o shea s ess
(Mode II o III) and opening s ess (Mode I) on 10 10
{}
α
may esul in mic oc acking on he p isma ic plane, which
is nea ly pe pendicula o he loading axis.
Because de o ma ion cons ain s in he
α
g ains on he
specimen su ace a e mo e lexible han he in e nal ones,
s ain incompa ibili y may be in oduced nea he speci-
men su ace. In ou p e ious wo k, we discussed a model
o subsu ace c ack ini ia ion based on he ac og aphical
analysis, whe e he ole o he
β
phase was omi ed.23) In he
model, he sc ew disloca ions o 10 10 1120
{}
pile-ups in
an
α
-g ain nea he specimen su ace in oduce he mic o-
c acking by (0001)
α
slip-o , as shown in Fig. 14(b). He ein,
he
β
mic oc ack may assis in he o ma ion o he shea
s ess on (0001)
α
. Because he (0001)
α
mic oc ack also p o-
ides a s ess ield wi h mul iaxial modes in he neighbo ing
g ain, he combina ion o shea s ess and opening s ess on
10 10
{}
α
esul in Mode II o III mic oc acking.
5. Conclusions
The cha ac e is ics o subc acks in Ti–Fe–O alloy o med
by high-cycle a igue a low empe a u e we e in es iga ed.
The phenomenological de ails o mic oc ack gene a ion,
pa icula ly he igge and beginning o he mic oc ack, a e
cla i ied as ollows:
(1) Mos o he mic oc acks we e in
β
g ains and a ew
p opaga ed in o
α
g ains along he p isma ic plane. A ew
oids we e p esen in he
α
g ains and a he
α
-
β
in e aces.
(2) The s ess concen a ion a ound he mic oc acks in
he
β
g ains assis s mic oc acking on he 10 10
{}
α
ace .
(3) The combina ion o shea s ess and opening s ess
on 10 10
{}
α
esul s in Mode II o III mic oc acking on he
p isma ic plane, which is pe pendicula o he loading axis.
Fu he mo e, he 10 10
{}
α
ex u e signi ican ly assis s mic o-
c ack g ow h on he 10 10
{}
α
plane.
Acknowledgmen
This pape was suppo ed by p ojec No. CZ.02.1.01/0.0/
0.0/17_049/0008441, “Inno a i e The apeu ic Me hods o
Musculoskele al Sys em in Acciden Su ge y” wi hin he
Ope a ional P og amme Resea ch, De elopmen and Educa-
ion inanced by he Eu opean Union and by he s a e budge
o he Czech Republic.
REFERENCES
1) O. Umezawa and K. Nagai: ISIJ In ., 37 (1997), 1170. h ps://doi.
o g/10.2355/isijin e na ional.37.1170
2) W. Li and O. Umezawa: Key Eng. Ma e ., 741 (2017), 76. h ps://doi.
o g/10.4028/www.scien i ic.ne /KEM.741.76
3) A. L. Pilchak, R. E. A. Williams and J. C. Williams: Me all. Ma e .
T ans. A, 41 (2010), 106. h ps://doi.o g/10.1007/s11661-009-0064-2
4) J. Ruppen, P. Bhowal, D. Eylon and A. J. McE ily: Fa igue Mecha-
nisms, ASTM STP 675, ed. by J. Fong, Ame ican Socie y o Tes -
ing and Ma e ials, Philadelphia, (1979), 47. h ps://doi.o g/10.1520/
STP35884S
5) I. Ban ounas, D. Dye and T. C. Lindley: Ac a Ma e ., 57 (2009),
3584. h ps://doi.o g/10.1016/j.ac ama .2009.04.018
6) M. R. Bache, M. Cope, H. M. Da ies, W. J. E ans and G. Ha ison:
In . J. Fa igue, 19 (1997), Suppl. 1, S83. h ps://doi.o g/10.1016/
S0142-1123(97)00020-0
7) V. Sinha, M. J. Mills and J. C. Williams: Me all. Ma e . T ans. A, 37
(2006), 2015. h ps://doi.o g/10.1007/s11661-006-0144-5
8) Y. Mine, S. Ka ashima, R. Ding, P. Bowen and K. Takashima: Sc .
Ma e ., 165 (2019), 107. h ps://doi.o g/10.1016/j.sc ip ama .2019.02.029
9) O. Umezawa, K. Nagai and K. Ishikawa: Ma e . Sci. Eng. A, 129
(1990), 223. h ps://doi.o g/10.1016/0921-5093(90)90269-9
10) H. Yokoyama, O. Umezawa, K. Nagai, T. Suzuki and K. Kokubo:
Me all. Ma e . T ans. A, 31 (2000), 2793. h ps://doi.o g/10.1007/
BF02830339
11) I. Ban ounas, D. Dye and T. C. Lindley: Ac a Ma e ., 58 (2010),
3908. h ps://doi.o g/10.1016/j.ac ama .2010.03.036
12) I. Ban ounas, T. C. Lindley, D. Rugg and D. Dye: Ac a Ma e ., 55
(2007), 5655. h ps://doi.o g/10.1016/j.ac ama .2007.06.034
13) K. Le Bia an , S. Pommie and C. P ioul: Fa igue F ac . Eng.
Ma e . S uc ., 25 (2002), 527. h ps://doi.o g/10.1046/j.1460-
2695.2002.00480.x
14) C. J. Szczepanski, S. K. Jha, J. M. La sen and J. W. Jones: Me all.
Ma e . T ans. A, 39 (2008), 2841. h ps://doi.o g/10.1007/s11661-
008-9633-z
15) M. R. Bache, W. J. E ans and H. M. Da ies: J. Ma e . Sci., 32
(1997), 3435. h ps://doi.o g/10.1023/A:1018624801310
16) M. R. Bache: In . J. Fa igue, 25 (2003), 1079. h ps://doi.o g/10.1016/
S0142-1123(03)00145-2
17) A. N. S oh: P oc. R. Soc. Lond. A, 223 (1954), 404. h ps://doi.
o g/10.1098/ spa.1954.0124
18) M. R. Bache, W. J. E ans, B. Suddell and F. R. M. He ouin: In .
J. Fa igue, 23 (2001), Suppl. 1, 153. h ps://doi.o g/10.1016/S0142-
1123(01)00124-4
19) E. E. Sacke , L. Ge main and M. R. Bache: In . J. Fa igue, 29 (2007),
2015. h ps://doi.o g/10.1016/j.ij a igue.2006.12.011
20) W. J. E ans and M. R. Bache: In . J. Fa igue, 16 (1994), 443. h ps://
doi.o g/10.1016/0142-1123(94)90194-5
21) F. B idie , P. Villechaise and J. Mendez: Ac a Ma e ., 56 (2008),
3951. h ps://doi.o g/10.1016/j.ac ama .2008.04.036
22) O. Umezawa, K. Nagai and K. Ishikawa: Te su- o-Hagané, 76 (1990),
924 (in Japanese). h ps://doi.o g/10.2355/ e su ohagane1955.76.6_924
23) O. Umezawa and K. Ishikawa: Ma e . Sci. Eng. A, 176 (1994), 397.
h ps://doi.o g/10.1016/0011-2275(92)90353-C
24) O. Umezawa, T. Yuasa and W. Li: ISIJ In ., 58 (2018), 1332. h ps://
doi.o g/10.2355/isijin e na ional.ISIJINT-2017-673
25) W. Li, O. Umezawa and N. Koga: ISIJ In ., 58 (2018), 359. h ps://
doi.o g/10.2355/isijin e na ional.ISIJINT-2017-514
26) K. Kishida, J. G. Kim, T. Nagae and H. Inui: Ac a Ma e ., 196 (2020),
168. h ps://doi.o g/10.1016/j.ac ama .2020.06.043
27) M. Mo i a and O. Umezawa: Ma e . T ans., 52 (2011), 1595. h ps://
doi.o g/10.2320/ma e ans.L-M2011815
28) M. H. Yoo: Me all. T ans. A, 12 (1981), 409. h ps://doi.o g/10.1007/
BF02648537
29) L. Wang, Y. Yang, P. Eisenloh , T. R. Biele , M. A. C imp and
D. E. Mason: Me all. Ma e . T ans. A, 41 (2010), 421. h ps://doi.
o g/10.1007/s11661-009-0097-6
30) S. Mahajan: Encyclopedia o Ma e ials: Science and Technology (2nd
ed.), ed. by K. H. J. Buschow e al., Else ie , Ams e dam, (2002), 1.
h ps://doi.o g/10.1016/b0-08-043152-6/01809-x
31) C. J. Bee e s and M. D. Halliday: Me . Sci. J., 3 (1969), 74. h ps://
doi.o g/10.1179/msc.1969.3.1.74
32) P. Le anc, V. Doque , M. Ge land and C. Sa azin-Baudoux: Ac a
Ma e ., 56 (2008), 4450. h ps://doi.o g/10.1016/j.ac ama .2008.04.060
33) W. Li and O. Umezawa: P oc. 13 h Wo ld Con . on Ti anium, ed. by
V. Venka esh e al., TMS, Wa endale, PA, (2016), 843. h ps://doi.
o g/10.1002/9781119296126.ch144
34) T. Ozaki, H. Ma sumo o, S. Wa anabe and S. Hanada: Ma e . T ans.,
45 (2004), 2776. h ps://doi.o g/10.2320/ma e ans.45.2776
35) R. Shi, Z. Nie, Q. Fan and G. Li: J. Alloy. Compd., 688 (2016), 787.
h ps://doi.o g/10.1016/j.jallcom.2016.07.105
36) J. E e ae s, D. Gon cha o , B. Ve linden and M. We e s: In . J.
Fa igue, 98 (2017), 203. h ps://doi.o g/10.1016/j.ij a igue.2017.01.043