scieee Science in your language
[en] (orig)

Morphological features in aluminum nitride epilayers prepared by magnetron sputtering

Abstract

The aim of this study is to characterize the surface topography of aluminum nitride (AlN) epilayers prepared by magnetron sputtering using the surface statistical parameters, according to ISO 25178-2:2012. To understand the effect of temperature on the epilayer structure, the surface topography was investigated through atomic force microscopy (AFM). AFM data and analysis of surface statistical parameters indicated the dependence of morphology of the epilayers on their growth conditions. The surface statistical parameters provide important information about surface texture and are useful for manufacturers in developing AlN thin films with improved surface characteristics. These results are also important for understanding the nanoscale phenomena at the contacts between rough surfaces, such as the area of contact, the interfacial separation, and the adhesive and frictional properties.

Read accessible full text

Morphological features in aluminum nitride epilayers prepared by magnetron sputtering

Author: Stach, Sebastian; Sobola, Dinara; Talu, Stefan; Kaspar, Pavel; Tománek, Pavel; Giovanzana, Stefano; Grmela, Lubomír
Publisher: De Gruyter Open
Year: 2015
DOI: 10.1515/msp-2015-0036
Source: https://dspace.vut.cz/bitstreams/b95444f4-058d-42f2-a2e0-1eaf159e2861/download
Ma e ials Science-Poland, 33(1), 2015, pp. 175-184
h p://www.ma e ialsscience.pw .w oc.pl/
DOI: 10.1515/msp-2015-0036
Mo phological ea u es in aluminum ni ide epilaye s
p epa ed by magne on spu e ing
SEBASTIAN STACH1, DINARA DALLAEVA2, ¸STEFAN ¸T˘
ALU3∗, PAVEL KASPAR2,
PAVEL TOMÁNEK2, STEFANO GIOVANZANA4, LUBOMÍR GRMELA2
1Uni e si y o Silesia, Facul y o Compu e Science and Ma e ials Science, Ins i u e o In o ma ics,
Depa men o Biomedical Compu e Sys ems, ul. B˛edzi´
nska 39, 41-205 Sosnowiec, Poland
2B no Uni e si y o Technology, Facul y o Elec ical Enginee ing and Communica ion, Physics Depa men , Technická 8,
616 00 B no, Czech Republic
3Technical Uni e si y o Cluj-Napoca, Facul y o Mechanical Enginee ing, Depa men o AET, Discipline o Desc ip i e
Geome y and Enginee ing G aphics, 103-105 B-dul Muncii S ., Cluj-Napoca 400641, Cluj, Romania
4Uni e si y o Milan-Bicocca, 20125 Milano, I aly
The aim o his s udy is o cha ac e ize he su ace opog aphy o aluminum ni ide (AlN) epilaye s p epa ed by magne on
spu e ing using he su ace s a is ical pa ame e s, acco ding o ISO 25178-2:2012. To unde s and he e ec o empe a u e
on he epilaye s uc u e, he su ace opog aphy was in es iga ed h ough a omic o ce mic oscopy (AFM). AFM da a and
analysis o su ace s a is ical pa ame e s indica ed he dependence o mo phology o he epilaye s on hei g ow h condi ions.
The su ace s a is ical pa ame e s p o ide impo an in o ma ion abou su ace ex u e and a e use ul o manu ac u e s in de-
eloping AlN hin ilms wi h imp o ed su ace cha ac e is ics. These esul s a e also impo an o unde s anding he nanoscale
phenomena a he con ac s be ween ough su aces, such as he a ea o con ac , he in e acial sepa a ion, and he adhesi e and
ic ional p ope ies.
Keywo ds: aluminum ni ide epilaye ; a omic o ce mic oscopy; magne on spu e ing; subs a e; su ace oughness
© W oclaw Uni e si y o Technology.
1. In oduc ion
As opposed o bulk ma e ial hin laye s p ope -
ies s ongly depend on he su ace condi ion. Mo -
phological ea u es and associa ed wi h hem s a is-
ical pa ame e s ( olume and spa ial) de ine su ace
p o ile o ilms opog aphy and cha ac e ize hei
quali y. Mic o- and nano-aspe i ies e alua ion is
impo an in he e os uc u es p epa a ion. Topog-
aphy o he ilms also in luences he su ace o he
deposi ed hin con ac laye s.
Aluminum ni ide is an A3B5g oup wide band
gap (6.3 eV) semiconduc o [1] wi h a wu zi e
c ys al s uc u e. I s expe imen al c ys al la ice pa-
ame e s a e: a =3.11 Å, c =4.98 Å, yielding a
elaxed c/a a io o 1.6 [2]. The e is a conside -
able ionic componen in AlN bonding because o
∗E-mail: s e an_ [email p o ec ed]
elec onega i i ies (1.6 and 3.0) o Al and N a oms.
The a omic adii a e 1.25 Å o Al and 0.70 Å
o N [3]. One aluminum a om is su ounded by
ou ni ogen a oms, making up a dis o ed e a-
hed on wi h h ee bonds and one bond in he di-
ec ion o he c-axis [4]. The elec onega i i ies o
Al and N a e e y di e en : 1.6 and 3.0, espec-
i ely, and so he e is a signi ican ionic compo-
nen o he bonding in AlN. O he p ope ies o
AlN a e: di ec band gap (6.2 eV), he mal conduc-
i i y (285 W·m−1·K−1), coe icien o he mal ex-
pansion (4 o 5 × 10−6K−1), e ac i e index (1.8
o 2.2) and dielec ic cons an (8.5) [4].
P ominen ea u es o aluminum ni ide make
his ma e ial one o he mos desi able in mod-
e n elec onics. I is cha ac e ized by in e es ing
ibological [5] p ope ies, high alue o ha dness
and high he mal conduc i i y, mode a e piezoelec-
ici y, low dielec ic and acous ic losses [6], high
176 SEBASTIAN STACH e al.
esis ance o empe a u e and s abili y in co osi e
medium [7], good hea dissipa ion [8], high di-
elec ic cons an , mode a ely high elec omechan-
ical coupling coe icien [9], low coe icien o
he mal expansion [4], high elas ic s i ness [2],
non- oxici y [10], elec ical eliabili y [11], ligh
weigh [12], high usion empe a u e [13], high e-
ac i e index, anspa ence in isible ligh [14].
All hese cha ac e is ics in combina ion wi h la ge
op ical band gap make AlN sui able o applica-
ions in high powe and high equency de ices,
su ace acous ic wa e il e s, insula ing [7], pas-
si a ing, cladding laye s [15] and op ical de ices
(blue ligh emi ing diodes, sho wa eleng h lase s,
ul a iole ligh de ec o s [16], compac disks, lase
diodes, phase shi li hog aphy masks, AlN/GaN
mul ilaye de ices [17], o g ow h o GaN laye s
on Al2O3and on 6H–SiC, which a e also o in-
e es as pe spec i e ma e ials), elec oluminescen
applica ions o e a wide wa eleng h ange [18],
acous ic-op ic de ices.
Oxidized AlN can subs i u e adi ional passi-
a ing ilms, such as silicon ni ide and silicon
oxide, a p- ype sola cells [13], selec i e de ec-
o s [19], ield emi e s o la panel displays, high-
speed ansis o s [20]. As no ed Auge e al. [7], i
can be applied as an in e media e bu e laye o
op ical and elec onic de ices and o u he ab i-
ca ion o he e os uc u es.
AlN hin ilms can be applied o me al-
oxide-semiconduc o ield-e ec ansis o [21],
CMOS echnologies, mic oelec omechnical and
mic oop omechanical sys ems ab ica ion [22,23],
elec onic packaging, wa eguides, senso s [24],
elec o-acous ic applica ions (mic oac ua o s, il-
e s, esona o s, acous ic modula o s, su ace
acous ic wa e de ices) [6,11,25]. The e a e a lo
o me hods o hin ilms ab ica ion, such as pulsed
lase deposi ion, eac i e molecula beam epi axy,
acuum a c/ca hodic a c deposi ion, DC/RF eac-
i e spu e ing, ion beam spu e ing, me al-o ganic
chemical apo deposi ion e c. [26].
The way o AlN o ma ion is de e mined
by subsequen applica ions. Magne on spu e -
ing is one o physical hin ilm deposi ion ech-
niques and is cha ac e ized by low- empe a u e
deposi ion, ease o syn hesis, low cos , non-
oxici y, good quali y ilms wi h a ai ly smoo h
su ace [4]. Magne on spu e ing is employed in
hin ilms p oduc ion o di e en applica ions in
mic oelec onics, pa ially in eg a ed ci cui s and
in op oelec onics. By changing o p ocess pa am-
e e s (bias- ol age, subs a e empe a u e, gas mix-
u e, p essu e, spu e cu en , applied ol age [27])
i is possible o con ol he ilm mic os uc u e
and i s mechanical, op ical, and elec ical p ope -
ies [28]. This me hod allows p ocessing he sub-
s a e and ilm su aces, such as e ching, deposi-
ion, annealing, and ion implan a ion [29].
By magne on spu e ing i is possible o ab i-
ca e ha d, wea - esis an , low ic ion, co osion e-
sis an , deco a i e coa ings and coa ings wi h spe-
ci ic op ical, o elec ical p ope ies [30], hin ilm
senso s, pho o ol aic hin ilms (sola cells), me al
can ile e s and in e connec s [31]. The c i ical ac-
o o piezoelec ic AlN hin ilm is i s c ys al o i-
en a ion and mo phology [4].
Low-cos AlN-on-silicon pho onic ci cui s a e
excellen subs i u es o complemen a y me al-
oxide-semiconduc o -compa ible pho onic ci cui s
o building new unc ional op omechanical de-
ices ha a e ee om ca ie e ec s [9].
The s udies o AlN hin ilm could be di ided
in o he in es iga ions which emphasize chemi-
cal and physical p ope ies (mechanical, elec ical,
magne ic) [6,9,17,21,22], analysis o he ilms
s uc u e [4,7,12,15,16,19,23–25,32] and com-
bined, showing he dependences o he p ope ies
on he ilms mo phology [2,11,13,14,26,33–
35]. The s uc u e o he hin ilms can u e ly di e
om he s uc u e o bulk ma e ial and ha e di e -
en s uc u al pe ec ion.
The su ace ea u es, such as de ec concen a-
ion, a e e y impo an in he case o hin ilms.
Some p ope ies o hin ilms s ongly depend on
he s uc u e. The s udy was ca ied ou on he ap-
plicabili y o s a is ical pa ame e s o aluminum
ni ide hin ilms cha ac e iza ion. This analysis
ep esen s ins uc i e in o ma ion o hin ilms
manu ac u ing and p ocessing.
Mo phological ea u es in aluminum ni ide epilaye s p epa ed by magne on spu e ing 177
2. Ma e ials and me hods
2.1. Ma e ials
The p ocesses used o aluminum ni ide o -
ma ion by magne on spu e ing include d y e ch-
ing, ion implan a ion o he subs a e and spu e -
ing o he a ge . We used magne on spu e ing
o aluminum a ge o hin ilms ab ica ion since
his me hod is compa ible o semiconduc o p o-
cesses. The ounda ion o his me hod is spu e ing
o a ge ma e ial (high-pu i y aluminum) by ions
o wo king gas in plasma o glow discha ge. A -
gon (A 2)was used o a ge spu e ing. Plasma
occu s in acuum a e applying high ol age be-
ween ca hode- a ge and anode and ac s as a con-
duc ing medium. The essen ial cons i uen s o he
p ocess a e ca hode, anode and magne ic sys em
o plasma con inemen nea he ca hode su ace.
High spu e ing a e and accu acy o he composi e
ecu ence a e he ad an ages o magne on spu -
e ing. The p essu e in he wo king chambe was 3
o 7 × 10−2Pa, cu en densi y o discha ge 5 o
7 × 10 A/m2. These pa ame e s de ine he a e
o condensa ion which consequen ly in luences he
s uc u e o he ilms. The ilms can be ob ained by
spu e ing o polyc ys alline aluminum ni ide a -
ge , bu i is possible o con ol hin ilms s uc u e
by gas composi ion in he chambe . The ni ogen
gas was used o c ea ion o wo k ambience and
o accomplish ni ogen inco po a ion in AlN ab-
ica ion. Magne on spu e ing de ice allows ab i-
ca ion o ma e ials wi h good s oichiome y. This
me hod was desc ibed by Dallae a e al. [36]. Alu-
minum ni ide ilms we e deposi ed on sapphi e
subs a es (Al2O3)a di e en subs a e empe a-
u es. Monoc ys alline sapphi e is one o he ha d-
es oxides, wi h high ha dness a high empe a u es,
good he mo-physical p ope ies and i is op ically
anspa en in he isible ligh o nea -in a ed e-
gion [36].
The subs a es we e cleaned by d y e ching in
o de o emo e he aces o polishing and hen
he ni idiza ion o nea su ace a ea was ca ied
ou . This helped o c ea e he bu e laye be ween
Al2O3and AlN. The subs a e empe a u e has a
s ong impac on he g ow h beha io [27] and
mo phology o he ilms. The empe a u e o he
subs a e is a e y impo an pa ame e because i s
changing causes a change in ene ge ic ba ie nea
he subs a e su ace. In addi ion, i in luences he
cha ac e o ilm g ow h: sizes o nucleolus and a e
o nuclea ion.
We ob ained he ilms a h ee empe a u es
(1000 K, 1300 K and 1500 K) o he subs a e. The
ilm becomes smoo h a low wid h and i consis s
o la ge numbe o small nuclei. These islands a e
h ee-dimensional (3-D) and hey ha e no iceable
sizes in he no mal di ec ion o g ow h plane [37].
2.2. Me hods
The ilms ha e been cha ac e ized by a omic
o ce mic oscopy (AFM) and scanning elec on
mic oscopy (SEM) in o de o desc ibe he 3-D
g ow h cha ac e o hin ilms and o in es iga e
hei su ace mo phology. 3-D su ace oughness
o he epilaye s was in es iga ed by a omic o ce
mic oscope N eg a P ima (NT-MDT, Russian Fed-
e a ion) in semi-con ac mode.
Can ile e s, model NSG01 DLC (AFM
“Golden” Silicon P obes) [38] wi h he ollow-
ing nominal speci ica ions: esonan equency
150 kHz, o ce cons an 5.1 N/m, leng h 125 µm,
wid h 30 µm and hickness 2 µm, we e used.
The ip speci ica ions we e as ollows: e ahed al
shape, heigh 14 µm, cu a u e adius 6 nm, and
cone angle a he apex 7° o 10° [38].
All measu emen s we e pe o med in he same
oom, a oom empe a u e (298 ± 1 K) and
(50 ± 1 %) ela i e humidi y. The measu emen s
we e epea ed ou imes o each sample on di e -
en e e ence a eas, o alida e he ep oducibili y
o hese ea u es.
2.3. Su ace cha ac e iza ion using quan-
i a i e pa ame e s
The 3-D su ace cha ac e is ics a e highly com-
plex and hei unde s anding plays a de e mi-
nan ole in ecognizing physico-chemical in e -
ac ions and in e acial p ope ies a he nanome-
e ange, in ol ed in manu ac u ing p ocess [39–
41]. Su ace ex u e is he epe i i e o andom de-
ia ion om he nominal su ace ha o ms he
178 SEBASTIAN STACH e al.
3-D opog aphy o he su ace and includes:
oughness (nano-and mic o oughness); wa iness
(mac o oughness); lay and laws [39].
The 3-D su ace oughness o hin ilms can
be desc ibed using he classical desc ip i e ap-
p oach [42–47]. De ailed su ace cha ac e iza ion
o he hin ilms was ob ained using six quan i a-
i e pa ame e s (heigh , unc ional, spa ial, hyb id,
unc ional ( olume) and ea u e pa ame e s), ac-
co ding o ISO 25178-2:2012 [48,49].
2.4. S a is ical analysis
S a is ical analyses we e pe o med using he
G aphPad InS a e sion 3.20 compu e so wa e
package (G aphPad, San Diego, CA, USA) [50].
The Kolmogo o -Smi no es was used o as-
sess he no mal dis ibu ion o quan i a i e a i-
ables. Compa isons among di e en a eas wi hin
he same sample we e pe o med using indepen-
den samples T- es . When s a is ical signi icance
was ound, he di e ence be ween wo g oups was
u he compa ed using he Mann-Whi ney U es .
Di e ences wi h a P alue o 0.05 o less we e con-
side ed s a is ically signi ican .
3. Resul s
The ep esen a i e wo-dimensional (2-D) mi-
c o opog aphic images o AlN laye s on he sap-
phi e subs a e ob ained a 1000 K, 1300 K and
1500 K, o a scanning squa e a ea o 5 µm × 5 µm,
a e shown in Fig. 1.
As known, he e a e se e al possible s uc u es
o AlN c ys als: g anula , wo m-like, and a colum-
na su ace o g ains [23]. In ou case o cleaned
and ni idized sapphi e subs a e, he columna
s uc u e has been expec ed. I has been expec ed
ha he g ains a e o ien ed along he g ow h di ec-
ion [34].
The he e os uc u e o (0001)AlN /(0001)Al2O3
was s udied by scanning elec on mic oscopy
(SEM) wi h he SEM sys em (FEI Quan a 200).
This measu emen showed he occu ence o AlN
ilm on he Al2O3subs a e. In he c oss-sec ional
image (Fig. 2) he e a e c ys alline columna g ains
o AlN. Thei ops a e la .
(a)
(b)
(c)
Fig. 1. Rep esen a i e 2-D mic o opog aphic AFM
images o AlN epilaye s on sapphi e subs a es
ob ained a : a) 1000 K, b) 1300 K and c) 1500 K;
scanning squa e a ea o 5 µm × 5 µm; e ical
ange o he displayed da a (nm) and he colo
ba s a e shown on he igh side o he AFM
images.
Mo phological ea u es in aluminum ni ide epilaye s p epa ed by magne on spu e ing 179
Fig. 2. A SEM image o he AlN epilaye on he sap-
phi e subs a e in c oss-sec ion.
The g aphical s udy o olume pa ame e s (su -
ace): Vmp, V c, Vmc and V based upon he Ab-
bo cu e calcula ed on he su ace associa ed wi h
Fig. 1, is shown in Fig. 3.
The cu e o ma e ial sha e p o ides impo an
in o ma ion abou he condi ion o he su ace in
e ms o i s ope a ional sui abili y. I is known ha
o he op imal unc ionali y o he su ace, i is e-
qui ed o be p og essi e o p og essi e- eg essi e.
The g aphical s udy o Sk pa ame e s associ-
a ed wi h Fig. 1is shown in Fig. 4.
The peak coun his og ams associa ed wi h
Fig. 1a e shown in Fig. 5.
Quali a i e con ol o he image includes bo h
measu ing o he su ace ea u es and i s dis ibu-
ion along he su ace. The da a in Fig. 5indica e
he uni o m g ow h o su ace ea u e sizes wi h he
inc easing o subs a e empe a u e.
Heigh -heigh co ela ion unc ion p o ides
conside able in o ma ion abou opog aphy condi-
ion (such as es ima ion o co ela ion a eas, sizes
o g ains and holes, and he cha ac e o hei dis-
ibu ion). The heigh -heigh co ela ion unc ions
associa ed wi h Fig. 1a e shown in Fig. 6.
La e al o ce mode o AFM p o ides edge-
p ecision scanning o he su ace o ea u e em-
phasis. Fig. 7allows isual es ima ion o he su -
ace appea ance (small egula lines a e he scan-
ning a e ac s as no il e was applied in o de o
obse e he eal objec s shape).
A summa y o he s a is ical pa ame e s esul s
is p esen ed in Table 1.
(a)
(b)
(c)
Fig. 3. Face o AFM 3-D images (le side) and g aph-
ical s udy o olume pa ame e s: Vmp, V c,
Vmc and V ( igh side) based upon he Ab-
bo cu e calcula ed on he su ace. Two bea -
ing a io h esholds a e de ined (using he e -
ical ba s ha a e d awn wi h do ed lines). By
de aul , hese h esholds a e se a bea ing a ios
o 10 % and 80 %. The i s h eshold, p1 (de-
aul : 10 %), is used o de ine he cu le el c1
(and p2 de ines c2, espec i ely). AlN epilaye s
on sapphi e subs a es ob ained a : (a) 1000 K,
(b) 1300 K and (c) 1500 K.
4. Discussion
A omic o ce mic oscopy, scanning elec on
mic oscopy and desc ip i e analysis ha e been em-
ployed o cha ac e ize he ilms mo phology o alu-
minum ni ide (AlN) epilaye s p epa ed by mag-
ne on spu e ing on he sapphi e subs a es.
While he AFM measu emen s (Fig. 1) p o ide
a su ace appea ance and a ounda ion o u he
ma hema ical p ocessing, he SEM image (Fig. 2)
shows he inne s uc u e co esponding o AlN
g ow h in c oss-sec ion.

180 SEBASTIAN STACH e al.
Table 1. S a is ical pa ame e s o AlN epilaye s on he sapphi e subs a es ob ained a : (a) 1000 K, (b) 1300 K and
(c) 1500 K, based on ISO 25178-2:2012.
S a is ical pa ame e s Symbol Samples a 1000 K Samples a 1300 K Samples a 1500 K
Values Values Values
Heigh Pa ame e s
Roo mean squa e heigh Sq [nm] 3.18 4.39 3.17
Skewness Ssk [–] 4.27 −0.089 0.254
Ku osis Sku [–] 26.5 4.27 2.70
Maximum peak heigh Sp [nm] 28.2 19.1 9.87
Maximum pi heigh S [nm] 6.31 19.4 9.64
Maximum heigh Sz [nm] 34.5 38.4 19.5
A i hme ic mean heigh Sa [nm] 1.60 3.24 2.58
Func ional Pa ame e s
A eal ma e ial a io Sm [%] 100 100 100
In e se a eal ma e ial a io Smc [nm] 1.27 5.18 4.27
Ex eme peak heigh Sxp [nm] 2.73 9.81 5.27
Spa ial Pa ame e s
Au o-co ela ion leng h Sal [µm] 0.290 0.302 0.135
Tex u e-aspec a io S [–] 0.537 0.696 0.0545
Tex u e di ec ion S d [°] 111° 0.356° 0.363°
Hyb id Pa ame e s
Roo mean squa e g adien Sdq [–] 0.0371 0.0818 0.0866
De eloped in e acial a ea a io Sd [%] 0.0682 0.333 0.370
Func ional Pa ame e s (Volume)
Ma e ial olume Vm [µm³/µm²] 0.000544 0.000286 0.00016
Void olume V [µm³/µm²] 0.00181 0.00545 0.00441
Peak ma e ial olume Vmp [µm³/µm²] 0.000544 0.000286 0.00016
Co e ma e ial olume Vmc [µm³/µm²] 0.00122 0.00333 0.00298
Co e oid olume V c [µm³/µm²] 0.00165 0.00482 0.00414
Pi oid olume V [µm³/µm²] 0.000153 0.000634 0.000271
Fea u e Pa ame e s
Densi y o peaks Spd [1/µm²] 0.805 6.44 4.92
A i hme ic mean peak cu a u e Spc [1/µm] 2.98 6.96 5.58
Ten poin heigh S10z [nm] 11.5 19.4 13.2
Fi e poin peak heigh S5p [nm] 8.10 12.7 6.86
Fi e poin pi heigh S5 [nm] 3.40 6.66 6.33
Mean dale a ea Sda [µm²] 1.33 0.121 0.119
Mean hill a ea Sha [µm²] 0.692 0.120 0.122
Mean dale olume Sd [µm³] 0.000317 0.000027 0.000027
Mean hill olume Sh [µm³] 0.000082 0.000033 0.000057
Mo phological ea u es in aluminum ni ide epilaye s p epa ed by magne on spu e ing 181
(a)
(b)
(c)
Fig. 4. G aphical s udy o Sk pa ame e s o AlN epi-
laye s on he sapphi e subs a es ob ained a : (a)
1000 K, (b) 1300 K and (c) 1500 K.
(a)
(b)
(c)
Fig. 5. The peak coun his og ams o AlN epilaye s on
he sapphi e subs a es ob ained a : (a) 1000 K,
(b) 1300 K and (c) 1500 K.
182 SEBASTIAN STACH e al.
(a)
(b)
(c)
Fig. 6. The heigh -heigh co ela ion unc ions o
AlN epilaye s on he sapphi e subs a es ob-
ained a : (a) 1000 K, (co ela ion leng hs: Lx
=0.319 µm, Ly =0.356 µm); (b) 1300 K,
(co ela ion leng hs: Lx =0.317 µm, Ly =
0.212 µm); (c) 1500 K, (co ela ion leng hs: Lx
=0.224 µm, Ly =0.0815 µm).
Di e en nano-aspe i ies wi h a speci ic heigh
p obabili y dis ibu ion and aspe i y cu a u e o
di e en aspe i y sizes a e dis ibu ed o e he
3-D su ace o all he samples. The esul s show
ha he e is an in e dependence be ween he em-
pe a u e o a subs a e du ing deposi ion p ocess
and ilm mo phology. Also, he e is a connec ion
be ween 3-D su ace oughness pa ame e s and he
ilm ex u e (Table 1) since he ex u e could be
conside ed as a measu e o su ace oughness [51].
The esul s shown in Fig. 3and 4include in-
o ma ion abou olume and Sk pa ame e s. The e
is a dec ease o Vmp and consequen ly Spk al-
ues. I indica es he educ ion o he sudden peaks,
(a)
(b)
(c)
Fig. 7. AFM la e al o ce mode images o AlN epilay-
e s on he sapphi e subs a es ob ained a : (a)
1000 K, (b) 1300 K and (c) 1500 K.
and om physical poin o iew, i means he
homogeniza ion o g ain g ow h. Reduced peak
heigh o co e a io (Spk/Sk), acco ding o he da a
o Fig. 4, shows inc easing o ex u e ampli ude
symme y.
Mo phological ea u es in aluminum ni ide epilaye s p epa ed by magne on spu e ing 183
The peak coun his og ams (Fig. 5) show
alongside wi h dec easing o opog aphy heigh s
(x-axis), he wide peaks dis ibu ion, which means
ha he su ace ea u es become mo e compa able
(y-axis) in size a he highes subs a e empe a u e.
The esul s show he dec easing o co ela ion
leng h, caused by educ ion o cha ac e is ic dis-
ance, whe e he co ela ed bonds a e los be ween
he opog aphy ea u es (Fig. 6). Fo he highe
subs a e empe a u e du ing he deposi ion p o-
cess, he ilm is ep esen ed by well ex u ed a eas.
The same esul s ha e been ob ained by Gue e o
and Ga cía [25] who ob ained he AlN laye s by
chemical apo deposi ion and no ed ha he su -
ace mo phology became smoo h wi h inc easing
o deposi ion empe a u e.
On he o he hand, adhesion s eng h o pa -
icles and subs a e depends on empe a u e. In-
c easing o empe a u e by hea ing he subs a e in-
c eases he in e ac ion a he in e ace a he nea -
su ace a ea.
I can be no ed a s ong dependence be ween
he ene ge ic pa ame e s o he ilm g ow h (nucle-
a ion) and he mo phology o he p epa ed ilms.
The s a is ical pa ame e s o 3-D su ace ough-
ness analysis also show ha he s a is ical desc ip-
ion o he ilms is in ag eemen wi h he da a ob-
ained by Dallae a e al. [36].
5. Conclusions
The s a is ical pa ame e s o 3-D su ace ough-
ness ha e been used o he unc ional pe o mance
p edic ion and quali y con ol o aluminum ni ide
(AlN) epilaye s p epa ed by magne on spu e ing
on he sapphi e subs a es. These esul s a e also
impo an o unde s anding he nanoscale phe-
nomena a he con ac be ween ough su aces,
such as he con ac a ea, he in e acial sepa a ion,
and he adhesi e and ic ional p ope ies.
Acknowledgemen s
Resea ch desc ibed in he pape was inancially sup-
po ed by he Eu opean Cen e o Excellence CEITEC
CZ.1.05/1.1.00/02.0068, by p ojec Senso , In o ma ion and
Communica ion Sys ems SIX CZ.1.05/2.1.00/03.0072 as well
as by g an FEKT-S-14-2240".
Re e ences
[1] VASHISHTA P., KALIA R.K., NAKANO A., RINO J.P.,
J. Appl. Phys., 109 (2011), 033514:1-8.
[2] LITIMEIN F., BOUHAFS B., DRIDI Z., RUTERANA P.,
New J. Phys., 4 (2002), 64.1.
[3] NORTHRUP J.E., DIFELICE R., NEUGEBAUER J.,
Phys. Re . B, 55 (20) (1997), 13878.
[4] KAR J.P., BOSE G., Aluminum Ni ide (AlN) ilm based
acous ic de ices: ma e ial syn hesis and de ice ab ica-
ion, in M.G. BEGHI (Ed.), Acous ic Wa es – F om Mi-
c ode ices o Helioseismology, InTech, 2011, p. 563.
[5] GUDA A.A., MAZALOVA V.L., YALOVEGA G.E.,
SOLDATOV A.V., J. Su . In es .-X-Ray+, 3 (3) (2009),
460.
[6] AHMAD M.A., PLANA R., IEEE Mic ow. Wi el. Co., 19
(3) (2009), 140.
[7] AUGER M.A., VAZQUEZ L., JERGEL M., SANCHEZ
O., ALBELLA J.M.,Su . Coa . Tech., 180 – 181 (2004),
140.
[8]
h p://www.dupon .com/p oduc s-and-
se ices/elec onic-elec ical-ma e ials/
uses-and-applica ions/mic oci cui -
ma e ials.h ml
.
[9] XIONG C., PERNICE W.H.P., SUN X., SCHUCK C.,
FONG K.Y., TANG H.X., New J. Phys., 14 (095014)
(2012), 1.
[10]
h p://www.coo s ek.com/ esou ce-
lib a y/lib a y/8510-1843-Aluminum_
Ni ide_Subs a es.pd
.
[11] LASPINA L., NANVER L.K., SCHELLEVIS H.,
IBORRA E., CLEMENT M., OLIVARES J., ESSDERC,
(2007), 354.
[12] YOSHIDA M., OKUMIYA M., ICHIKI R., TEKMEN C.,
KHALIFA W., TSUNEKAWA Y., HARA T., J. Plasma
Fusion. Res., 8 (2009), 1447.
[13] GARCIA-MENDEZ M., MORALES-RODRIGUES S.,
MACHORRO R., DELACRUZ W., Re . Mex. Fis, 54
(4) (2008), 271.
[14] STAFINIAK A., MUSZY ´
NSKA D., SZYSZKA A.,
PASZKIEWICZ B., PTASI ´
NSKI K., PATELA S.,
PASZKIEWICZ R., TŁACZAŁA M., Op . Appl., 39 (4)
(2009), 717.
[15] KARMANN S., SCHENK H.P.D., KAISER U., FISSEL
A., RICHTER W.O., Ma e . Sci. Eng. B-Ad ., 50 (1997),
228.
[16] JAGANNADHAM K., SHARMA A.K., WEI Q.,
KALYANRAMAN R., NARAYAN J., J. Vac. Sci. Technol.
A, 16 (5) (1998), 2804.
[17] JONES D.J., FRENCH R.H., MULLEJANS H.,
LOUGHIN S., DORNEICH A.D., CARCIA P.F., J.
Ma e . Res., 14 (11) (1999), 4337.
[18] ISHIKAWA R., LUPINI A.R., OBA F., FINDLAY
S.D., SHIBATA N., TANIGUCHI T., WATANABE K.,
HAYASHI H., SAKAI T., TANAKA I., IKUHARA Y.,
PENNYCOOK S.J., Sci. Rep.-UK, 4 (2014), 3778.
[19] AUGER M.A., VÁZQUEZ L., SÁNCHEZ O., JERGEL
M., CUERNO R., CASTRO M., J. Appl. Phys., 97 (12)
(2005), 123528.