MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
49
Re ealing dynamic‑mechanical
p ope ies o p ecipi a es
in a nanos uc u ed hin ilm using
mic omechanical spec oscopy
Ma kusAl eide ,* MichaelMeindlhume , TobiasZiegelwange ,
Ros isla Daniel, Joze Keckes, and DanielKiene
Nanos uc u ed ma e ials wi h hei ema kable p ope ies a e key enable s in many
mode n applica ions. Fo example, indus ial d y-milling p ocesses would no be
as widely sp ead wi hou he use o ha d, wea - esis an me al ni ide coa ings o
p o ec he cu ing ools. Howe e , imp o ing hese nanos uc u ed hin ilms wi h
ega d o dynamical p ope ies is demanding as p obing espec i e pa ame e s o
(sub-)mic on laye s wi hou any subs a e in luence is s ill challenging. To ex end
he scien i ic oolbox o such spa ially con ined sys ems, a no el me hodological
app oach based on esonance peak measu emen s o a can ile e - ansduce
sys em e med mic omechanical spec oscopy (µMS) is de eloped and applied
o a Al0.8C 0.2N model sys em. The mainly wu zi e ype supe sa u a ed Al0.8C 0.2N
sys em showed p ecipi a ion o cubic C N a g ain bounda ies and local C
a ia ions upon annealing a 1050°C. This was accompanied by an inc ease in he
p e iously unknown damping capabili y o 63pe cen and an inc ease in Young’s
modulus by 36pe cen .
Impac s a emen
The e is a wide a ie y o applica ions o nano- o
mic ome e -sized hin ilms in oday’s enginee ing
echnology, om he mal ba ie - and wea - esis an
coa ings in u bines and bea ings, o e di usion ba -
ie s and hea sinks in mic oelec onic de ices, o op i-
cally ac i e laye s in lase s o mi o s. The mechanical
p ope ies o such hin ilms a e o en imes go e ned
by hei he mal his o y, leading o ei he in en ional
o undesi ed changes in he mic os uc u e (e.g., he
o ma ion o p ecipi a es). While he in es iga ion o
such ea u es is usually cons ic ed o s a ic analysis
using high- esolu ion echniques, such as ansmission
elec on mic oscopy, unde s anding hei impac on
dynamic p ope ies o he ilm emains a challenge.
Howe e , hese a e highly ele an in many enginee -
ing applica ions whe e cyclic beha io is common,
such as high-speed d y milling. In he p esen wo k,
we in es iga e he change in mechanical damping
capabili y upon annealing o a 6-µm hin AlC N ilm,
commonly used in demanding d y-milling applica-
ions, using mic omechanical spec oscopy (µMS)
o can ile e -shaped specimens. A e a ca e ully
adjus ed hea ea men , he ilm exhibi s he o ma-
ion o cubic C N p ecipi a es in an o he wise wu zi e
AlC N ma ix, which leads o a p e iously unknown
bene icial inc ease in damping capabili y o he ilm.
In oduc ion
Ha d nanoce amic coa ings a e commonly
used in mode n enginee ing applica ions
whe e du abili y and he mal s abili y ma -
e , such as cu ing inse s o high-speed
d y milling1,2 o he mal ba ie coa ings o
u bines.3–5
Commonly, such sys ems a e based on
ansi ion-me al ni ide hin ilms such
as TiN and C N deposi ed by plasma-
assis ed physical apo deposi ion ech-
niques (e.g., ca hodic a c e apo a ion).6
Adding Al in o TiN and C N leads o he
o ma ion o Ti1−xAlxN and C 1−xAlxN,
espec i ely, which ha e signi ican ly
highe ha dness and oxida ion esis -
ance compa ed o hei ep esen a i e
bina y coun e pa s.7 In u n, cubic (B1)
Ti1−xAlxN and C 1−xAlxN o m me as a-
ble supe sa u a ed solid solu ions, whe e
Al eplaces Ti o C up o x = 0.678,9
o 0.7,7 espec i ely. Upon annealing,
hese hin ilms unde go a complex em-
pe a u e and ( esidual) s ain-dependen
decomposi ion p ocesses,10,11 which al e
hei unc ional and mechanical p ope -
ies.7 Typically, o Ti1−xAlxN spinodal
decomposi ion in o Ti- ich and Al- ich
cubic zones is obse ed a empe a u es
be ween 800° and 900°C leading o a
sligh inc ease o ha dness12 and ough-
ness,13 while a e exceeding 950°C, he
o ma ion o wu zi e (B4) Al(Ti)N is
p omo ed, esul ing in a signi ican d op
in mechanical p ope ies.12–14 Compa a-
bly, cubic C 1−xAlxN (x < 0.7) di ec ly
© The Au ho (s) 2023
doi:10.1557/s43577-023-00549-w
Ma kusAl eide , Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia; [email p o ec ed]
MichaelMeindlhume , Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia
TobiasZiegelwange , Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia
Ros isla Daniel, Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia
Joze Keckes, Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia
DanielKiene , Depa men o Ma e ials Science, Mon anuni e si ä Leoben, Leoben, Aus ia
*Co esponding au ho
Impac A icle
50 MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
decomposes in o cubic C (Al)N and wu zi e Al(C )N a
empe a u es abo e ~700–900°C, whe e he o me ans-
o ms la e on a empe a u es su passing 1000°C and
1350°C in o C 2N and me allic C unde he elease o
gaseous ni ogen.15
In he case o C 1−xAlxN hin ilms wi h x > 0.7, he ilms
o m p edominan ly a wu zi e B4 c ys al s uc u e du ing
deposi ion,9 whe e C eplaces Al in he wu zi e c ys al
s uc u e. The e, decomposi ion s a s a signi ican ly highe
empe a u es (>950°C), o ming again cubic C (Al)N and
wu zi e Al(C )N composi e.16,17 As shown o Al0.9C 0.1N,
acuum annealing a 1100°C esul ed in he o ma ion o
globula cubic C (Al)N and elonga ed cubic C N p ecipi a es
a in a- and in e g anula posi ions, which, in u n, simul ane-
ously enhanced Young’s modulus, ha dness, ac u e s ess,
and ac u e oughness.17
The hickness o hese ilms is ypically in he ange
o a ew mic ome e s, which makes he cha ac e iza ion
o mechanical p ope ies wi hou subs a e in luence gen-
e ally challenging. In ecen yea s, a ious me hods ha e
been de eloped o ob ain eliable expe imen al da a, om
ha dness and modulus using nanoinden a ion,18 o e ac-
u e p ope ies using pilla spli ing19 o mic ocan ile e
bending20 o esidual s esses using laye emo al21–23 o
c oss-sec ional x- ay nanodi ac ion echniques.24–26 How-
e e , all o hese app oaches de e mine quasis a ic ma e ial
pa ame e s, whe eas in mos applica ions (e.g., high-speed
d y milling), hese sys ems a e loaded wi h a a he s ong
dynamic componen due o he high o a ion speeds. In
ac , dynamic p ope ies such as damping can ha e a ben-
e icial o de imen al impac on he pe o mance. Highe
damping capabili ies enable he dissipa ion o ib a ional
ene gy, which could occu h ough sligh eccen ici ies and
misalignmen s.
Fu he mo e, damping con ols dissipa ion o ene gy gi en
e e sible elas ic de o ma ion and is he e o e nondes uc-
i e independen o measu emen echnique. This means ha
damping measu emen s can be used o de e mine inhe en
changes due o he mal o en i onmen al his o y in a sys em,
and because a change in damping capabili y is commonly he
di ec esul o mic os uc u al changes i can be used o p obe
mic os uc u al modi ica ions, such as p ecipi a ion o g ain
g ow h.
Al hough such in es iga ions can be expe imen ally eal-
ized on whole de ices (e.g., cu ing inse s), he esul s a e
challenging o esol e in o he indi idual cons i uen s o
he sys em (e.g., subs a e o ha d nanoce amic coa ing).
The e o e, i would be desi able o ha e a echnique capa-
ble o esol ing he damping beha io o hese indi idual
cons i uen s.
In his a icle, we showcase such an expe imen al app oach,
e med mic omechanical spec oscopy (µMS, ecen ly in o-
duced in Re e ence 27), on a Al0.8C 0.2N coa ing on cemen ed
ca bide as commonly ound in he high-speed d y-milling indus-
y. The measu emen s a e based on de e mining esonance
peaks o an in si u nanoinden a ion ansduce sys em in con ac
wi h a can ile e -shaped mic ospecimen o he coa ing ma e ial
inside a scanning elec on mic oscope (SEM). The absence o
ai -damping inside he SEM chambe enables a high- esolu ion
analysis o he esonance peak and he e o e de e mina ion o he
damping capabili y o such spa ially con ined specimens. The
measu ed change in damping upon hea ea men o he coa ing
is co ela ed o he o ma ion o p ecipi a es using µMS in es-
iga ions in conce wi h complemen a y ansmission elec on
mic oscopy (TEM) and x- ay di ac ion (XRD), espec i ely.
Resul s
Mic os uc u al andchemical analysis
Figu e 1a shows a c oss-sec ional SEM image o he
Al0.8C 0.2N ilm in as-deposi ed condi ion. The e, indi idual
b igh ea u es co espond o esidual C xAly d ople s om
he deposi ion p ocess. Upon annealing o only 5 min a
1050°C, he mic os uc u e exhibi s ob ious p ecipi a ion,
as e iden by he da k and b igh (depending on channeling
con as ) ea u es in Figu e 1b. De ailed scanning ansmis-
sion elec on mic oscope (STEM) images in conjunc ion wi h
ene gy-dispe si e spec oscopy (EDS) analysis, as depic ed
in Figu e 1c e eal a laye ed mic os uc u e wi h a sligh
undula ing magni ude o Al and C con en s, espec i ely.
This laye o ma ion is known o be esul an o he dis-
ibu ion o chemical species inside he deposi ion plasma
in conjunc ion wi h he o a ion o he subs a e in and ou
o he line o sigh o he ca hode.28–30 Fu he mo e, i is
e iden ha d ople s (Figu e 1c, b igh ea u e) appea o
be mainly in e me allic C xAly, as he N con en is dimin-
ished in ha egion, while i is app oxima ely cons an e e-
ywhe e else. Thus, wi h he excep ion o indi idual d ople s,
he ilm appea s o be ully homogenous, sugges ing a e y
high con en o supe sa u a ed wu zi e ype Al0.8C 0.2N, as
de ailed in p e ious wo ks.17 A e hea ea men he ilm
exhibi s a clea decomposi ion, displayed in Figu e 1d, whe e
dis inc C - ich p ecipi a es a e e iden . These appea o
nuclea e p e e en ially a he posi ions o p e iously highe
ini ial C con en (i.e., along he C -en iched laye s).
Phase analysis
To analyze he phase con en s in as-deposi ed and hea - ea ed
condi ions, XRD in es iga ions we e conduc ed on he espec-
i e ilms and a e summa ized in Figu e 2. The e he peak
posi ions o wu zi e AlN and cubic C N a e deno ed by
g een del oids and blue squa es, espec i ely. The posi ions
o he WC subs a e (g ay iangles) and he TiN bonding
laye (o ange squa es) a e addi ionally added o e e ence.
I is e iden ha he dis inc AlN peaks (100: 33.211°; 101:
37.924°; 110: 59.338°) a e shi ed owa d lowe alues in he
as-deposi ed s a e and mo e owa d hei equilib ium posi ions
upon annealing, in acco dance wi h p e ious wo ks on compa-
able composi ion.17 Howe e , a con ibu ion o he emo al
o esidual s ess by de ec annihila ion as a consequence o
he hea ea men canno be excluded om he peak shi o
MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
51
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
he AlN la ice.17 In ac , he peak shi oge he wi h he con-
comi an inc ease in magni ude o he 200 C N peak a 43.693°
(Figu e 2, op igh inse ) sugges s ha C di uses ou o he
supe sa u a ed Al0.8C 0.2N s uc u e, lea ing a ela i ely highe
con en o Al in he wu zi e AlN la ice, while he p ecipi a ed
C o ms cubic C N c ys als.
Mic omechanical spec oscopy
Mic omechanical spec oscopic (µMS) in es iga ions we e con-
duc ed on ou as-deposi ed and h ee hea - ea ed can ile e -
shaped specimens. This echnique was de eloped ecen ly in a
p e ious wo k27 and is capable o e ealing changes in damping
capabili y o e y con ined olumes by in es iga ing he shape
o he i s esonance peak o he used inden a ion se up. Fig-
u e 3a depic s an SEM mic og aph o he specimen and inden a-
ion se up wi h ele an geome ic dimensions: leng h L, heigh
W, and hickness B (in imaging di ec ion). Quan i a i e alues
a
b
cd
Figu e1. Scanning elec on mic oscope images o (a) he as-deposi ed and (b) hea - ea ed Al0.8C 0.2N ilms, espec i ely. The
lowe pa o he igu e con ains scanning ansmission elec on mic oscope mic og aphs, wi h co esponding ene gy-dispe si e
spec oscopy da a o N (blue, op igh ), Al (g een, lowe le ), and C ( ed, lowe igh ) o (c) as-deposi ed and (d) hea - ea ed
s a es, espec i ely.
Figu e2. X- ay di ac ion da a o he hin- ilm samples in as-depos-
i ed (blue) and hea - ea ed ( ed) s a es, espec i ely. The inse de ails
he 200 peak o cubic C N, showing e iden inc ease o his phase
upon annealing.
52 MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
a e ga he ed by i ing he esonance peak based on a physical
single deg ee o eedom (SDOF) oscilla o model, as shown
schema ically in Figu e 3b using he lm i package ( e sion
1.0.0) wi hin he Py hon 3.8 p og amming en i onmen .27
Ini ial da a a e ga he ed as dynamic compliance c o
he whole sys em as a unc ion o equency, whe eby
c(
ω
)=�
x
(
ω
)/�
F , wi h displacemen ampli ude Δx and load
ampli ude ΔF. To compa e he specimens among hemsel es, i
is use ul o no malize he adial equency by he esonance e-
quency
ω0=√
k
/m
and he compliance by he s a ic compliance
c0(
ω
=
0
)=
1
/
k , wi h mass m and s i ness k. The esul ing da a
a e summa ized in Figu e 4, whe ein he as-deposi ed specimens
a e depic ed by open symbols and he hea - ea ed ones a e indi-
ca ed by illed symbols. The damping capabili y o a ma e ial is
commonly quan i ied as in e se quali y ac o Q−1, which is he
a io be ween dissipa ed wo k ΔW and o al conduc ed wo k W
du ing one oscilla ion cycle and can be w i en as:
wi h he iscous damping cons an b. In Figu e 4 he a e age
esonance cu e o all as-deposi ed specimens (Qad
−1 = 1.0
6 × 10–2 ± 1 × 10–4) is depic ed wi h a blue solid line and he
a e age esonance cu e o all hea - ea ed specimens (Qh
−1 =
1.124 × 10–2 ± 1 × 10–4) is gi en by a dashed ed line. As he
damping o he sys em is la ge in compa ison o ha o he
indi idual can ile e s, obse ing he di e ence is a he chal-
lenging in he whole da a g aph. The e o e, he yellow inse
in Figu e 4 ( op igh co ne ) de ails he peak ip egion and
shows he e iden di e ence in peak heigh o he wo speci-
men s a es. Fu he mo e, all o he as-deposi ed s a es show
a e y dis inc damping peak on he igh -hand side o he
main esonance peak a an absolu e a e age adial equency
posi ion o ω = 1849 ± 4 Hz, which is no e iden in he hea -
ea ed specimens. These peaks a e shown as de ailed inse
1
Q
−1
=
�
W
2
π
W
=
b
√km
in Figu e 4 (lowe igh co ne ). No e ha hose a e sligh ly
shi ed among hemsel es as he esonance equency o he
indi idual can ile e s is no iden ical due o mino geome ical
a ia ions.
To ob ain he damping capabili y o he ma e ial sys em inde-
penden o he p obing inden e one needs o ake in o accoun
he wo k ha is dissipa ed by he inden e sys em ou o con ac
(i.e.,
Q−
1
s=
Q
−
1
−
Q−1
i
)
,
whe eby he subsc ip s s and i deno e
he specimen and inden e , espec i ely. The indi idual specimen
in e se quali y ac o Qs
−1 is summa ized in Figu e 5a, whe e he
da a o each as-deposi ed can ile e a e depic ed by open blue
hexagons (le -hand side) and ha o each hea - ea ed can ile e
by open ed squa es ( igh -hand side). All unce ain y es ima es
a e conside ed based on unco ela ed inpu quan i ies31 wi h geo-
me ic measu emen e o s o ± 3 px (±50 nm) and all pa ame e
e o s aken as one s anda d de ia ion, gi en by he i ing p oce-
du e. The illed da a poin s in Figu e 5 depic he in e se a iance
weigh ed a e age o all indi idual µMS expe imen s o he wo
espec i e ma e ial s a es. The e he e o ba s deno e he s and-
a d e o o he weigh ed a e age. F om Figu e 5a i is e iden
ha he damping capabili y inc eases qui e signi ican ly by abou
66% om Qs,ad
−1 = 2.3 × 10–3 ± 2 × 10–4 o he as-deposi ed s a e o
Qs,h
−1 = 3.8 × 10–3 ± 3 × 10–4 o he hea - ea ed s a e, espec i ely.
Fu he mo e, µMS expe imen s allow o he de e mina ion
o Young’s modulus based on he equency shi o he peak, as
his co esponds o he can ile e specimen s i ness ks. How-
e e , o de e mine he indi idual specimen’s s i ness om he
o al s i ness k wi hou he con ibu ions o he inden e sys em
Figu e3. (a) Scanning elec on mic oscope mic og aph o a
can ile e -shaped specimen posi ioned inside he Al0.8C 0.2N ilm wi h
geome ic pa ame e s leng h L and heigh W. (b) The single deg ee
o eedom oscilla o model, depic ing he s i ness k and damping
elemen s b o inden e ( ed), con ac (g een), and specimen (blue),
espec i ely, as well as (c) he educed sp ing model o Young’s
modulus de e mina ion.
Figu e4. Resonance peaks o all es ed specimens, no malized by
esonance equency ω0 and s a ic compliance o he o al sys em
c o al (ω = 0). The as-deposi ed specimens a e deno ed by open
symbols and he hea ea ed by illed symbols, espec i ely. The
solid blue line ep esen s he a e age esonance shape o he whole
sys em o all as-deposi ed da a wi h Q−1 = 0.0106, whe eas he
dashed ed line ep esen s he same o all hea - ea ed da a wi h
Q−1 = 0.0124. Two inse s show de ails o he esonance peak o bo h
s a es and an occu ing damping peak in as-deposi ed s a e a highe
esolu ion, espec i ely.
MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
53
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
and he con ac be ween ip and specimen su ace, one needs
o conside a educed model as shown in Figu e 3c, whe e
he specimen s i ness ks and con ac s i ness kc a e se ies-
connec ed and he inden e s i ness ki is in pa allel o bo h o
hem. This is due o he ac ha bo h he specimen and he
inden e ip a e connec ed wi h he igid base inside he SEM,
gi ing hem a common g ound in analogy o elec ic ci cui s.
The specimen s i ness ks is calcula ed as ollows:
While he inden e s i ness is ob ained du ing he cali-
b a ion ou ine o he ins umen , he con ac s i ness is a
unc ion o he con ac o ce and needs o be measu ed on
he base o he can ile e .32 Fo he gi en expe imen s a an
a e age cons an con ac o ce o 500 µN a con ac s i ness
o kc = 44,030 N/m was de e mined. Using he ac ha du -
ing he expe imen only small o al load line displacemen s o
2
k
s
=
1
k
−
k
i
−
1
k
c
−1
.
≈250 nm, leading o small o a ion angles o ≈1.5°, a e used,
one can u ilize he simple Eule –Be noulli beam heo y33 o
calcula e he Young’s modulus E as:
The da a a e summa ized in Figu e 5b, whe e he same colo
scheme and unce ain y p opaga ion calcula ions as o he Q−1
da a a e employed. I is e iden ha he modulus inc eases by
abou 36% om Ead = 205 ± 6 o Eh = 280 ± 9 GPa, om he
as-deposi ed o he hea - ea ed s a es, espec i ely.
Discussion
The discussion will be s uc u ed in o sec ions ega ding phase
o ma ion, change in damping capabili y and inc ease in mod-
ulus, espec i ely.
Cubic C N phase o ma ion
The e iden phase o ma ion o he cubic C N phase a e
annealing, as de ailed in a ecen p e ious wo k on a sim-
ila composi ion (Al0.9C 0.1N),17 is d i en by he supe -
sa u a ion o C a oms inside he equilib ium AlN c ys al
s uc u e as a mainly wu zi e ype Al0.8C 0.2N solid solu-
ion (Figu e 2).
To es ima e he amoun o cubic C N phase ha o ms
upon annealing, Rie eld e inemen (Powde Cell 2.434) was
used on he da a p esen ed in Figu e 2, u ilizing he ac ual
c ys allog aphic spacing o he simila supe sa u a ed wu z-
i e Al0.9C 0.1N s uc u e (a = 313.3 pm, c = 500.3 pm17) as
ini ia ion pa ame e s. A e sub ac ion o he dis inc peaks
o he TiN bonding laye and he WC subs a e he con en s
o he wu zi e- ype ma ix s uc u e compu e o 94 and 85%,
while he cubic C N con en s enume a e o 6 and 15% o
he as-deposi ed and hea - ea ed ma e ial s a es, espec-
i ely. This sugges s ha al eady in he as-deposi ed s a e a
mino amoun o cubic C N is p esen , bu upon annealing he
amoun inc eases by abou 9 pe cen . As he mic os uc u e
is a he homogeneous in he as-deposi ed s a e (Figu e 1c),
hese mic os uc u al egions seem o be below he esolu ion
limi o he ga he ed STEM images, so in he ange o a ew
nanome e s. Fu he mo e, he ac ha only 15% C N is e i-
den a e annealing sugges s ha s ill abou 5% o C emain
inco po a ed inside he wu zi e AlN ma ix, as he d i ing
o ce o he C o di use ou is educed by he elaxa ion o
eigens esses, due o de ec annihila ion and seg ega ion as
e iden om he AlN peak shi (Figu e 2). As complemen a y
analysis, mul iple STEM mic og aphs o he hea - ea ed s a e
(simila o Figu e 1d) ha e been analyzed by image h eshold-
ing (ImageJ 1.54s35) using manual h eshold inpu s o ob ain
es ima es o he a ea ac ion o he C N phase. The esul s
anged om app oxima ely 5 o 16%, which includes he in lu-
ence o human inpu pa ame e s as well as locally di e en
C N con en ha could occu due o he s ochas ic dis ibu ion
o C . Howe e , hese alues show easonable ag eemen wi h
3
E=
4ksL3
BW
3
.
Figu e5. Summa y o (a) damping capabili y Qs−1 and (b) Young’s
modulusE o all specimens o as-deposi ed (blue hexagons) and
hea - ea ed ( ed squa es) s a es; espec i ely. The open symbols
deno e he indi idual da a ga he ed om µMS expe imen s, whe eas
he illed symbols deno e he in e se a iance weigh ed a e age o
he speci ic ma e ial s a es.

54 MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
he Rie eld analysis and co obo a e ha he majo amoun
o cubic C N o ms hese e iden app oxima ely 10–20-nm-
sized p ecipi a es.
Elas ic modulus change
The elas ic modulus o he as-deposi ed mainly wu zi e
Al0.8C 0.2N is conside ably lowe a 205 ± 6 GPa han he elas ic
modulus o pu e wu zi e AlN would be wi h ~ 301 GPa36 (cal-
cula ed by he Voig –Reuss–Hill (VRH) app oxima ion37,38)
om simula ions o ~340 GPa39 om expe imen al measu e-
men s. One could a gue ha his dec ease o elas ic modulus is
a esul o he la ge uni cell gi en he C inco po a ion. Using
he p opo ionali y ela ions be ween elas ic modulus E, bulk
modulus K, and he second de i a i e o he binding ene gy
o e uni cell olume
6
2U
6
V
2 , as commonly desc ibed by he
Bi ch–Mu naghan equa ion o s a e,40,41 in conjunc ion wi h
he di e ence be ween supe sa u a ed and elaxed uni cell size
om Meindlhume e al.17 ( oughly 0.5%), i is possible o
es ima e he change o elas ic modulus o be app oxima ely
2 pe cen . This would mean he modulus change based on he
inc eased uni cell size in he as-deposi ed Al0.8C 0.2N in com-
pa ison o he equilib ium AlN, would amoun o only abou
6 GPa ( om 301 o 295 GPa), which seems no enough o a
di e ence o explain he disc epancy obse ed he ein. How-
e e , conside ing on he o he hand he ac ha C is mos
likely subs i u ional on Al posi ions wi hin he wu zi e s uc-
u e, one can a gue ha he change in bonding ene gy be ween
he species could al e he elas ic esponse. Al hough he p e-
cise bonding ene gies o he indi idual species combina ions
inside he c ys al s uc u e a e nea impossible o ob ain, bind-
ing ene gies o wo-a omic s uc u es a e abula ed and can ac
as a i s -o de es ima e.42 The s eng hs o ei he Al–N bonds
(3.81 eV) and C –N bonds (3.92 eV) di e only sligh ly by
abou 3%, whe eas he Al–Al bonds (2.74 eV) a e signi ican ly
s onge han he Al–C bonds (2.31 eV) by abou 16 pe cen .
As he elas ic p ope ies o c ys als a e a di ec esul o he
bonding s eng hs, his could ac as a sou ce o he educed
modulus gi en he o ced inco po a ion o C wi hin he wu z-
i e AlN s uc u e. Conside ing now he hea - ea ed specimens
sugges s wo possible sou ces o an inc eased modulus. Fi s
he p ecipi a ion o la ge cubic C N egions wi h an inhe en ly
highe elas ic modulus o 358 GPa43 (using again he VRH
app oxima ion37,38), o second he s a is ical inc ease in
s onge Al–Al bonds, wi hin he wu zi e AlxC 1−xN s uc u e,
due o dissolu ion o he C ou o he ma ix.
The ini ial elas ic modulus in he as-deposi ed Al0.8C 0.2N
is also in good ag eemen wi h he single-c ys al ab ini io
calcula ions by May ho e e al.,44 who ound elas ic moduli
anging om 188 o 195 GPa o he wu zi e AlxC 1−xN sys-
em, espec i ely.
To alida e he p esen esul s, nanoinden a ion expe imen s
we e pe o med on he ilms wi hou any p io su ace p epa a-
ion. As nanoinden a ion expe imen s only de e mine a educed
modulus, hey a e no di ec ly compa able o he µMS esul s
as hese de e mine he ac ual ma e ial’s elas ic modulus. The e-
o e, he Poisson’s a io o AlN ν = 0.24 was assumed o cal-
cula e he ma e ial’s inden a ion moduli as Ei,ad = 266 ± 34 GPa
and Ei,h = 326 ± 82 GPa o he as-deposi ed and hea - ea ed
s a es, espec i ely. While a simila end o modulus inc ease
is e iden in he nanoinden a ion da a, he ough su ace o he
unpolished ilms as well as he unknown mic os uc u al ea-
u es unde nea h he inden s (e.g., d ople s [see Figu e 1c]) lead
o a a he la ge sca e in compa ison wi h he µMS esul s.
One could imp o e he nanoinden a ion esul s by p io pol-
ishing s eps. Howe e , gi en ha he sys em is only 6 µm in
hickness and a ce ain hickness mus be e ained o nanoin-
den a ion o de e mine a subs a e-in luence ee modulus,
his could be a he challenging. Simila limi a ions apply o a
c oss-sec ional app oach.45 The e o e, i appea s ha al hough
he p epa a ion ia FIB milling is mo e cumbe some, he mod-
ulus de e mina ion using µMS could be bene icial in sys ems
wi h a high su ace oughness o small geome ic spacing, such
as he Al0.8C 0.2N laye as s udied he ein. Fu he mo e, he
ac ha no ini ial Poisson’s a io es ima e is necessa y means
ha unknown changes in his a io upon p ecipi a ion o o he
mic os uc u al changes ha e no in luence on he inal esul .
Change indamping capabili ies
Commonly he damping capabili y is educed a e anneal-
ing as dissipa i e elemen s such as esidual de ec s a e
ei he emo ed (e.g., disloca ions,46 acancies)47 o elaxed
(e.g., g ain bounda ies).27,48,49 The e o e, he p esen inc ease
o damping by abou 66% upon annealing seems a he coun-
e in ui i e. Especially, conside ing he small damping peak a
ω = 1849 ± 4 Hz in he as-deposi ed s a e, which is no appa -
en a e hea ea men sugges s ha some easy o ac i a e
anelas ic elaxa ion p ocesses we e emo ed (e.g., annealing
o de ec s om a me as able, o a mo e s able (equilib ium)
s a e ook place). Independen o he ac ual unde lying p o-
cess, hese peaks a e known o ollow an A henius ela ion-
ship,50,51 as:
whe e , 0 a e he elaxa ion ime and limi elaxa ion ime, H
is an ac i a ion en halpy, k is he Bol zmann cons an , and T is
he empe a u e. While classical in e nal ic ion expe imen s
change empe a u e as well as equency o measu e 0 and H,
one can use he in e se o he Debye equency o wu zi e AlN
as 0 ≈ 10–13 –10–15 52,53 o ob ain a i s -o de app oxima ion o
H. In conjunc ion wi h he cons an oom empe a u e T = 22°C
(ai condi ioned), he ac i a ion en halpy o a peak wi h a elax-
a ion ime o = 0 ω−1 s, equa es o H ≈ 0.57 – 0.69 eV. Com-
monly, c ys alline dissipa i e mechanisms can be sepa a ed in o
h ee main ca ego ies, namely sho - ange di usion, o exam-
ple, Snoek elaxa ion,54 disloca ion-based mechanisms,55–57 o
g ain-bounda y (phase bounda y)-based mechanisms.58–60 Con-
side ing ha any ype o oscilla o y disloca ion mo ion in his
ha d nanoce amic coa ing is unlikely o occu unde he small
load ampli ude o 5 µN, which co esponds o a maximum s ess
4
=
0
e
H
kT
,
MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
55
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
ampli ude o app oxima ely 10 MPa a he base o he can ile e ,
his mechanism can be neglec ed.
I bulk di usion p ocesses would ha e a majo con ibu-
ion o he damping peak, he es ima ed ac i a ion en halpy
should ag ee wi h ac i a ion ene gies o he co esponding
di usional p ocess.
Densi y unc ional heo y calcula ions o acancy di usion
wi hin wu zi e AlN sugges an ac i a ion ene gy o app oxi-
ma ely 2.33 eV61 and Okumu a e al.62 measu ed he ac i a ion
ene gy o impu i y di usion (Si and O) wi hin wu zi e AlN as
app oxima ely 1.3 eV using posi on annihila ion. None o hese
ene gies a e low enough o be compa able wi h he es ima e o
0.57–0.69 eV om he obse ed damping peak. The e o e, he
p ocesses co ela ed wi h his damping peak mos likely esul s
o he nonequilib ium c ys al s uc u e in he as-deposi ed s a e
(i.e., me as able a om posi ions due o he o ced inco po a ion o
C o e y small local wu zi e-cubic s uc u al in e aces ha a e
al eady p esen ), as sugges ed by he XRD da a (Figu e 2). This
is u he suppo ed by he ac ha acancy o ma ion ene gies
seem o be gene ally educed in un elaxed (i.e., de o med), in
compa ison o he elaxed (i.e., annealed s a e in (Ti,Al)N/AlN
in e aces), as epo ed om ab ini io simula ions.63
The disappea ance o his small peak upon annealing
would imply a lowe amoun o dissipa ing p ocesses, and i
he p ocesses co esponding o his peak would con ibu e o
he o e all damping in a majo ex en , also he main esonance
damping peak should dec ease, leading o a lowe Q−1 in he
hea - ea ed s a e.
Howe e , he obse ed p ecipi a ion o he C N phase
esul s in an inc ease in phase bounda y a ea, which can also
con ibu e o he damping capabili y o he sys em. Schoeck64
de i ed an analy ical exp ession o he change in Q−1 upon
p ecipi a ion based on he shape change, elas ic modulus
misma ch, and occu ence o incohe en phase bounda ies,
espec i ely. He a gued ha he i s wo cases seem un ea-
sonable due o he ac ha a change in shape would need o
inco po a e some kind o di usional componen o anelas ic
elaxa ion o occu , and a misma ch in elas ic modulus leads
o ins an aneous changes (no ime dependence— he e o e no
anelas ic elaxa ion) upon load e e sal, lea ing only elaxa-
ion p ocesses wi hin incohe en bounda ies as easonable
sou ces o damping. His ea men was based on a quasi-
degene a e ellipsoidal egime wi h hal -axis a, which con ains
he incohe en bounda y and leads o:
whe e τ, τi a e a global homogeneous and a local ac ing shea
s ess, espec i ely, and V is he o al specimen olume. This
a he abs ac geome ical desc ip ion is challenging o co -
ela e wi h expe imen al da a. Idealizing he sys em by assum-
ing ha he homogeneous global shea s ess is a he a shea
s ess dis ibu ion being equal o he shea s ess dis ibu ion
ac ing on he in e aces emo es τ and τi om he equa ion.
5
Q
−1
=
1
τ2
8(1−ν)
3π
(
2
−
ν
)

ia3
iτ2
i
V
≈
0.0194Vp ecipi a es
V
,
Fu he mo e, assuming ha all p ecipi a es a e dodecahed al
in shape and a e o he same size allows o es ablish a ela ion-
ship be ween he ellip ical hal -axis a and he (a e age) p e-
cipi a e olume Vp ecipi a es, by se ing he c oss-sec ional a ea
o he ellipsoid and one o he 12 sides’ a eas o he dodeca-
hed on o be equal, as schema ically depic ed in Figu e 6a.
Based on his simpli ied sys em one can es ima e he inc ease
in olume ac ion o p ecipi a es wi hin he specimen based
on he change in damping capabili y (ΔQ−1 ≈ 1.5 × 10–3) o
be app oxima ely 7.7%, which ag ees con incingly well wi h
he inc ease o he cubic C N phase deduced om Rie eld
analysis (9%), and sugges s ha he dissipa ing p ocesses
wi hin incohe en in e aces ha e a majo con ibu ion o he
obse ed damping inc ease upon annealing. Simila beha -
io has been p e iously obse ed only in me allic sys ems
(Ni–Al,65 Cu–O,66 Al–Cu,67 Al–Mg68), whe e semi-cohe en
o incohe en p ecipi a ion o ma ion lead o he occu ence
o a empe a u e independen elaxa ion peak in in e nal ic-
ion expe imen s.
Al hough i is gene ally no known whe he cubic C N
p ecipi a es o m comple ely incohe en ly wi hin he wu z-
i e AlN ma ix, he p ecipi a es ha o m on g ain bounda-
ies will mos likely ha e a leas one incohe en in e ace,
as i will be nea ly impossible o exhibi wo (o mo e—
conside ing iple junc ions) cohe en in e aces du ing
o ma ion be ween wo andomly o ien ed g ains. This is
shown ep esen a i ely in he high- esolu ion TEM mic o-
g aph in Figu e 6b, whe e a C N p ecipi a e wi h wo dis-
inc ly di e en c ys al o ien a ions on ei he side (uppe ,
lowe ) is depic ed. The lowe bounda y is de ailed in Fig-
u e 6c, wi h he zoomed egion on he le -hand side and he
co esponding Fou ie - il e ed image (using only he i s -
o de peaks) on he igh -hand side. The e, he con inua ion
o he la ice planes as e idenced in he Fou ie - il e ed
image demons a es cohe ency be ween he p ecipi a e and
he ma ix. In Figu e 6d, a same-sized egion o he uppe
bounda y as well as he co esponding Fou ie - il e ed
image a e depic ed. While he comple e h ee-dimensional
s uc u e o he bounda y is no esol able, he e iden loss
o cohe ency in he image sugges s an incohe en bound-
a y. This, in conjunc ion wi h he a guable absence o o he
majo dissipa i e mechanisms makes anelas ic elaxa ion
wi hin incohe en cubic C N–wu zi e AlN bounda ies he
mos likely mechanism o he obse ed inc ease in o e all
damping capaci y.
Al hough his no el me hodology p obes a a he wide
ensemble o inhe en de ec s, he analysis and sepa a ion o
hese indi idual de ec s’ s uc u es is no s aigh o wa d. Fo
chemically simple ma e ials (i.e., single species), an a emp
o ma ch he ob ained Q−1 alues o di e en de ec ypes
could be he co ela i e molecula dynamics app oach69 whe e
indi idual de ec s (disloca ions, acancy clus e s, GBs) unde
load esul in a ying phase shi s be ween shea s ess and
s ain, which can be di ec ly co ela ed o damping magni ude.
56 MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
Summa y andconclusion
In he p esen wo k an Al0.8C 0.2N hin ilm o abou 6-µm
hickness was deposi ed on cemen ed ca bide by ca hodic a c
e apo a ion as commonly ound in comme cially a ailable
high-speed d y-milling ools. A hea ea men o 1050°C
o only 5 min led o he o ma ion o cubic C N p ecipi a es
wi hin he o he wise wu zi e Al(C )N ma ix, as e idenced
by TEM and XRD in es iga ions. The dynamic-mechanical
p ope ies o he ilm in bo h ma e ial s a es, as-deposi ed
and hea - ea ed, we e in es iga ed using he no el µMS
me hodology and e ealed an inc ease o Young’s modulus
by abou 36% om Ead = 205 ± 6 GPa o Eh = 280 ± 9 GPa, as
well as an inc ease in damping capabili y o abou 66%, om
Qs,ad
−1 = 2.3 × 10–3 ± 2 × 10–4 o Qs,h
−1 = 3.8 × 10–3 ± 3 × 10–4,
upon annealing. The modulus om µMS expe imen s is in
ag eemen o heo e ical and expe imen al li e a u e alues, as
well as nanoinden a ion esul s, which alida es he new me h-
odology. The lowe sca e o he µMS esul s (app oxima ely
3%) in compa ison o he nanoinden a ion da a (up o 25%),
sugges s ha his echnique could be bene icial in sys ems wi h
high su ace oughness and/o a limi ed spa ial ex en , such as
he hin- ilm sys em s udied he ein. Fu he mo e, he e alua-
ion is independen o he o en imes unknown Poisson’s a io.
The inc ease in damping upon annealing is mos likely a esul
o he o ma ion o incohe en phase bounda ies be ween he
cubic C N p ecipi a es and he wu zi e Al(C )N ma ix. This
p e iously un esol able inc ease in damping can be bene i-
cial conside ing he applica ion o high-speed d y milling,
as i de ains mechanical ib a ion and he e o e s ess spikes,
which can lead o ac u e o he ha d coa ing. Fu he mo e,
i unde lines ha he no el
µMS me hodology is able
o esol e mic os uc u al
changes, e en in he g ain/
phase bounda y egime o
e y con ined olumes o
only a ew mic ome e s.
This makes i a use ul
complemen a y echnique
o b idge he gap be ween
high- esolu ion single-
bounda y in es iga ions
(i.e., by TEM and mac o-
scopic in es iga ions on
he whole de ice).
Ma e ials
andme hods
Thin‑ ilm deposi ion
andhea ea men
The ini ial Al0.8C 0.2N hin
ilms we e deposi ed on a
cemen ed ca bide (WC,
10 w % Co) subs a es
by ca hodic a c e apo a ion (alpha 400p, oes alpine ei ele
Vaco ec GmbH, Düsseldo , Ge many) a a bias ol age o
UB = − 100 V, subs a e empe a u e Ts = 475°C, and ni o-
gen p essu e pN2 = 4 Pa. The 10 × 10 × 5 mm3 mi o -polished
cemen ed ca bide subs a es we e moun ed a a dis ance
o ~100 mm om he ca hodes in a one old o a ion holde
ope a ed a a speed o 2 pm and we e plasma cleaned be o e
deposi ion. Fi s , a ~200-nm TiN bonding laye was depos-
i ed u ilizing a single Ti ca hode o inhibi di usion be ween
subs a e and ilm du ing annealing ollowed by Al0.8C 0.2N
deposi ed om i e sin e ed Al0.8C 0.2 ca hodes. The same
deposi ion condi ions we e used h oughou he whole depo-
si ion p ocess. The inal ilm hickness was measu ed using
SEM o be ~6 µm.
To ob ain mic os uc u al changes indi idual samples
we e annealed o 5 min a 1050°C in a acuum a mosphe e
(panneal = 5 × 10–4 Pa) using a con en ional high acuum u -
nace (HTM Ree z GmbH, Be lin, Ge many) ope a ed a hea -
ing and cooling a es o 0.5 K/s.
Elec on mic oscopy andx‑ ay di ac ion
Mic os uc u al in es iga ions we e conduc ed using an SEM
(Leo 1525, Ca l Zeiss AG, Obe kochen, Ge many) ope a ed
a 3 kV using seconda y elec on as well as backsca e elec-
on imaging. Fu he mo e, elec on anspa en oils o he
ilm c oss sec ion we e p epa ed ia ocused ion beam mill-
ing (FIB, Helios NanoLab 660, The mo Fishe Scien i ic Inc.,
Wal ham, Mass., USA) and analyzed using a TEM (Ti an
Themis, The mo Fische Scien i ic Inc., Wal ham, Mass.,
ab
d
c
Figu e6. (a) Idealized dodecahed al p ecipi a e shape wi h side leng hs as well as Schoeck’s64 ellipsoidal
egime o hal -axis leng ha. (b) High- esolu ion ansmission elec on mic oscopemic og aph o a C N p e-
cipi a e wi h de ails o (c) a cohe en and (d) an incohe en phase bounda y, whe e he igh -hand side images
o he de ails a e Fou ie - il e ed, espec i ely.
MRS BULLETIN • VOLUME 49 • JANUARY 2024 • m s.o g/bulle in
57
REVEALiNg dYNAMic‑MEchANicAL pROpER iEs O pREcipi A Es iN A NANOs RUc UREd hiN iLM UsiNg MicROMEchANicAL spEc ROscOpY
USA) ope a ed a 200 kV accele a ing ol age and a p obe
cu en o 1 nA equipped wi h a de ec o o ene gy-dispe si e
spec oscopy (ChemiSTEM Supe -X spec ome e , The mo
Fishe Scien i ic Inc., Wal ham, Mass., USA) o ob ain local
chemical in o ma ion. XRD phase analysis o he samples was
pe o med using a i e-ci cle x- ay di ac ome e (Sma Lab,
Rigaku Co., Tokyo, Japan) equipped wi h Cu–Kα adia ion, a
pa abolic mul ilaye mi o in he p ima y beam, and a second-
a y g aphi e monoch oma o . The XRD cha ac e iza ion was
ca ied ou in g azing incidence geome y wi h 5° incidence
angle.
Nanoinden a ion andmic omechanical spec oscopy
Nanoinden a ion expe imen s we e conduc ed on bo h ma e ial
s a es wi hou p io su ace p epa a ion using a G200 Nanoin-
den e (KLA Co po a ion, Milpi as, Cali ., USA) wi h a dia-
mond Be ko ich ip. Dep h-sensi i e ha dness and educed
modulus we e ga he ed using con inuous s i ness measu e-
men a a equency o 45 Hz, a loading a e o 0.05 s−1, and
was a e aged be ween 170- and 190-nm pene a ion dep h.
Mic omechanical spec oscopy was conduc ed ollowing
he p ocedu e de eloped in p e ious wo ks27 on can ile e -
shaped specimens wi h a geome y o ~2 × 3 × 12 µm3 p o-
cessed by FIB (1540XB, Ca l Zeiss AG, Obe kochen, Ge -
many) ope a ed a 30 kV and wi h subsequen ly educing
cu en s om 10 nA o 50 pA. The expe imen s we e con-
duc ed using a Hysi on PI85 (B uke Co po a ion, Bille ica,
Mass., USA) ansduce wi h a nanoDMA III upg ade, which
exhibi s an inhe en esonance equency a 114 Hz, and was
equipped wi h a 5-µm-wide conduc i e diamond wedge ip
(Syn on-MDP AG, Nidau, Swi ze land). The specimens we e
loaded o a con ac o ce o 500 µN and he oscilla ion ampli-
ude was kep cons an a 5 µN, while he equency sweep
was conduc ed om 200 o 300 Hz wi h a andomized e-
quency pa e n and h ee epe i ions wi h di e ing equency
spacing o ep oducibili y checks and o ob ain a highe eso-
lu ion owa d he con ac esonance peak. The combined ip
and sha mass as well as he inden e s i ness we e measu ed
du ing he usual calib a ion ou ine as m = 805.63 mg and
ki = 419.62 N/m, espec i ely, and he inhe en damping o
he inden e is Q−1
i = 8.34 × 10–3 ± 3.7 × 10–4.
Acknowledgmen s
This p ojec has ecei ed unding om he Eu opean Resea ch
Council (ERC) unde he Eu opean Union’s Ho izon 2020
esea ch and inno a ion p og amme (G an No. 771146
TOUGHIT). The CzechNanoLab p ojec LM2018110 unded
by MEYS CR is g a e ully acknowledged o he inancial
suppo o he measu emen s/sample ab ica ion a CEITEC
Nano Resea ch In as uc u e. Expe imen al suppo wi h he
x- ay equipmen by J. Tod , he nanoinden a ion equipmen
by V. Maie -Kiene , and he ansmission elec on mic oscopy
in es iga ions by J. Zalesak a e g a e ully acknowledged.
Funding
Open access unding p o ided by Mon anuni e si ä Leoben.
Da a a ailabili y
The aw/p ocessed da a equi ed o ep oduce hese indings
a e a ailable om he co esponding au ho upon easonable
eques .
Con lic o in e es
On behal o all au ho s, he co esponding au ho s a es ha
he e is no con lic o in e es .
Open access
This a icle is licensed unde a C ea i e Commons A ibu-
ion 4.0 In e na ional License, which pe mi s use, sha ing,
adap a ion, dis ibu ion and ep oduc ion in any medium o
o ma , as long as you gi e app op ia e c edi o he o iginal
au ho (s) and he sou ce, p o ide a link o he C ea i e Com-
mons license, and indica e i changes we e made. The images
o o he hi d pa y ma e ial in his a icle a e included in he
a icle’s C ea i e Commons license, unless indica ed o he -
wise in a c edi line o he ma e ial. I ma e ial is no included
in he a icle’s C ea i e Commons license and you in ended
use is no pe mi ed by s a u o y egula ion o exceeds he
pe mi ed use, you will need o ob ain pe mission di ec ly
om he copy igh holde . To iew a copy o his license, isi
h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/.
Re e ences
1. P.H. May ho e , C. Mi e e , L. Hul man, H. Clemens, P og. Ma e . Sci. 51, 1032
(2006). h ps:// doi. o g/ 10. 1016/j. pma s ci. 2006. 02. 002
2. C. Mi e e , Comp ehensi e Ha d Ma e ials (Else ie , Ams e dam, 2014), ol. 2, p.
449. h ps:// doi. o g/ 10. 1016/ B978-0- 08- 096527- 7. 00035-0.
3. J.G. Thaka e, C. Pandey, M.M. Mahapa a, R.S. Mulik, Me . Ma e . In .27, 1947
(2020). h ps:// doi. o g/ 10. 1007/ s12540- 020- 00705-w
4. A.G. E ans, D.R. Mumm, J.W. Hu chinson, G.H. Meie , F.S. Pe i , P og. Ma e . Sci.
46, 505 (2001). h ps:// doi. o g/ 10. 1016/ S0079- 6425(00) 00020-7
5. M.R. Begley, H.N.G. Wadley, Ac a Ma e . 60, 2497 (2012). h ps:// doi. o g/ 10. 1016/j.
ac am a . 2011. 12. 039
6. J. Ve e , Su . Coa . Technol. 76–77, 719 (1995). h ps:// doi. o g/ 10. 1016/ 0257-
8972(95) 02499-9
7. M. Kawa e, A.K. Hashimo o, T. Suzuki, Su . Coa .Technol.165, 163 (2003). h ps://
doi. o g/ 10. 1016/ S0257- 8972(02) 00473-5
8. M. Kawa e, A. Kimu a, T. Suzuki, J. Vac. Sci. Technol. A20(2), 569 (2002). h ps://
doi. o g/ 10. 1116/1. 14485 10
9. A.E. Rei e , V.H. De linge , B. Hanselmann, T. Bachmann, B. Sa o y, Su . Coa .
Technol. 200, 2114 (2005). h ps:// doi. o g/ 10. 1016/j. su c oa . 2005. 01. 043
10. M. Meindlhume , S. Klima, N. Jäge , A. S a k, H. H uby, C. Mi e e , J. Keckes, R.
Daniel, Sci. Rep. 9, 18027 (2019). h ps:// doi. o g/ 10. 1038/ s41598- 019- 54307-7
11. L. Rogs öm, J. Ullb and, J. Alme , L. Hul man, B. Jansson, M. Odén, Thin Solid
Films 520, 5542 (2012). h ps:// doi. o g/ 10. 1016/j. s . 2012. 04. 059
12. P.H. May ho e , A. Hö ling, L. Ka lsson, J. Sjölén, T. La sson, C. Mi e e , L. Hul man,
Appl. Phys. Le . 83, 2049 (2003). h ps:// doi. o g/ 10. 1063/1. 16084 64
13. M. Ba osik, C. Rumeau, R. Hahn, Z.L. Zhang, P.H. May ho e , Sci. Rep. 7, 16476
(2017). h ps:// doi. o g/ 10. 1038/ s41598- 017- 16751-1
14. L. Chen, J. Pauli sch, Y. Du, P.H. May ho e , Su . Coa . Technol. 206, 2954 (2012).
h ps:// doi. o g/ 10. 1016/j. su c oa . 2011. 12. 028
15. H. Willmann, P.H. May ho e , P.O.Å. Pe sson, A.E. Rei e , L. Hul man, C. Mi e e , Sc .
Ma e . 54, 1847 (2006). h ps:// doi. o g/ 10. 1016/j. sc ip ama . 2006. 02. 023
16. P.H. May ho e , H. Willmann, A.E. Rei e , Su . Coa . Technol. 202, 4935 (2008).
h ps:// doi. o g/ 10. 1016/j. su c oa . 2008. 04. 075
17. M. Meindlhume , T. Ziegelwange , J. Zalesak, M. Hans, L. Lö le , S. Spo , N. Jäge , A.
S a k, H. H uby, R. Daniel, D. Holec, J.M. Schneide , C. Mi e e , J. Keckes, Ac a Ma e .
237, 118156 (2022). h ps:// doi. o g/ 10. 1016/j. ac am a . 2022. 118156
18. S.J. Bull, J. Phys. D Appl. Phys. 38(24), R393 (2005). h ps:// doi. o g/ 10. 1088/
0022- 3727/ 38/ 24/ R01