applied
sciences
A icle
Expe imen al S udy on he Su ace P ope ies o
Nanoalumina-Filled Epoxy Resin Nanocomposi es
Deesy Pin o 1, Ana M. Ama o 2,* and Luís Be na do 3
1CQM—Cen o de Química da Madei a, Uni e si y o Madei a, Campus da Pen eada, 9020-105 Funchal,
Po ugal; deesy.pin o@s a .uma.p
2CEMMPRE, Depa men o Mechanical Enginee ing, Uni e si y o Coimb a, 3030-788 Coimb a, Po ugal
3Depa men o Ci il Enginee ing and A chi ec u e, Cen e o Ma e ials and Building
Technologies (C-MADE), Uni e si y o Bei a In e io , 6201-001 Co ilhã, Po ugal; [email p o ec ed]
*Co espondence: ana.ama [email p o ec ed]
Recei ed: 3 Janua y 2020; Accep ed: 18 Janua y 2020; Published: 21 Janua y 2020


Abs ac :
This a icle p esen s an expe imen al s udy on he su ace p ope ies o epoxy esin
nanocomposi es (EPNCs) manu ac u ed wi h a he mose ing epoxy esin (EP)–bisphenol A diglycidyl
e he (BADGE)–2-[[4-[2-[4-(Oxi an-2-ylme hoxy)phenyl]p opan-2-yl]phenoxy]me hyl]oxi ane) and
illed wi h alumina nanopa icles (NPs). The NPs consis o p e ea ed (wi h a silane agen ) alpha
alumina wi h i egula shapes and a 100 nm maximum size. Th ee weigh ac ions o NPs we e s udied:
1, 3, and 5 w . (%). Two di e en epoxy (EP) esins we e manu ac u ed, one cu ed and pos cu ed wi h
bis (4-aminophenyl) me hane (DDM); and ano he one cu ed wi h 3-dodec-2-enyloxolane-2,5-dione
(DDSA) +8-me hyl-3a,4,7,7a- e ahyd o-4,7-me hano-2-benzo u an-1,3-dione (MNA). The we abili y
and he su ace oughness o he ob ained EPNCs we e s udied h ough he measu emen o con ac
angles and opog aphic images ob ained wi h a omic o ce mic oscopy (AFM), espec i ely. Signi ican
in luence o bo h he loading o NPs and used cu ing agen s was obse ed. EPNCs cu ed wi h DDM
we e shown o be hyd ophobic o 0, 1, and 3 w . (%) and hyd ophilic o 5 w . (%). Maximum
su ace oughness was obse ed o 5 w . (%). EPNCs cu ed wi h DDSA+MNA we e shown o
be hyd ophilic o 0 and 1 w . (%) and hyd ophobic o 3 and 5 w . (%). The su ace oughness
dec eased as he weigh ac ion o NPs inc eased un il 3 w . (%), and hen inc eased o 5 w . (%).
Keywo ds:
epoxy esin nanocomposi es; nano-alumina; cu ing agen s; con ac angle; su ace
oughness
1. In oduc ion
The mose ing epoxy esins (EP) sys ems ha e been widely used as ma ices o composi e
ma e ials in se e al indus ies (e.g., ai c a , au omo i e, ae ospace, shipbuilding, and ci il cons uc ion
indus ies). In pa icula , hey ha e been used o me al subs i u ion in se e al enginee ing applica ions
(e.g., elec onic de ices, ex iles, and machine y) and also as base ma e ials o wea - esis an coa ing,
adhesi es, and ad anced ib ous polyme ic composi es (PCs), among o he s [1–4].
In spi e o hei excellen pe o mances, EPs ha e some d awbacks, such as b i le ailu e
(because plas ic de o ma ion is cons ained [
5
]), low oughness (weak esis ance o c ack ini ia ion
and p opaga ion), low low, high coe icien o linea he mal expansion, sh inkage, low he mal
conduc i i y, and high sensi i i y o c acks, which highly limi hei applica ion in some demanding
ields [
1
–
5
]. In o de o sol e hese d awbacks, a conside able amoun o esea ch has been ca ied ou
o modi y EPs, o ins ance by changing hei molecula s uc u e o inc ease he c osslink densi y,
leading o highe s i ness and s eng h [
6
]. In addi ion, he oughness o EPs can be imp o ed by
inco po a ing oughening agen s such as liquids, igid o hyb id igid ubbe y pa icles, o un eac i e
Appl. Sci. 2020,10, 733; doi:10.3390/app10030733 www.mdpi.com/jou nal/applsci
Appl. Sci. 2020,10, 733 2 o 12
nano ha dene s (i.e., nanopa icles (NPs)) a e y low con en s (usually desc ibed in e ms o he
weigh ac ion (w . (%)) o olume ac ion ( ol. (%)) [
2
,
3
,
7
–
11
]. In ecen yea s, special a en ion
has been paid o EPs illed wi h NPs. These nano ille s (e.g., silicon dioxide (SiO
2
) [
12
–
14
], i anium
dioxide (TiO
2
) [
15
,
16
], aluminum oxide (Al
2
O
3
) [
17
–
19
]) a e cha ac e ized by hei speci ic shape and
size (nanoscale), high su ace a eas, su ace p e ea men , and he deg ee o dispe sion in o he EP
ma ix [1,2,4,20,21].
In ecen yea s, many s udies ha e demons a ed ha he inclusion o ino ganic NPs in o
EPs has he capabili y o imp o e he s i ness, oughness, lexu al modulus, ac u e oughness,
s eng h, ha dness, and many o he p ope ies o he inal PCs [
1
–
3
,
7
,
8
], which a e necessa y in
many enginee ing applica ions. These imp o emen s a e achie ed wi hou sac i icing he basic
p ope ies o EPs and wi h low pe cen ages o NP loading [
3
,
8
,
17
,
22
–
24
]. Among he a ailable
NPs ha ha e been s udied and ha can be used o ein o ce EPs, alumina NPs cons i u e a good
solu ion. The good esul s published in ecen yea s, combined wi h he low cos o alumina NPs
when compa ed o o he NPs ( o ins ance, alumina NPs a e cheape han i anium NPs), i s high
modulus (abou 300 GPa), and high he mal esis ance in compa ison wi h o he s me al oxides
NPs, show ha unc ionalized alumina NPs a e a iable and e y p omising solu ion o be used as
nano ein o cemen s o PCs [
1
,
2
,
4
,
8
,
22
,
24
]. Among he cu ing agen s used as ha dene s o EPs, some
s udies show ha bis (4-aminophenyl) me hane (DDM) [
1
,
2
,
25
–
28
] o 3-dodec-2-enyloxolane-2,5-dione
(DDSA) +8-me hyl-3a,4,7,7a- e ahyd o-4,7-me hano-2-benzo u an-1,3-dione (MNA) [
28
] a e sui able
o EP sys ems.
Bo h he physicochemical and opog aphical p ope ies o su ace ma e ials a e impo an o
adhesion and coa ings applica ions [
29
,
30
]. EP adhesi es illed wi h NPs ha e also been ecen ly
in es iga ed [
20
,
21
,
31
,
32
]. I was ound ha he inclusion o NPs can imp o e he we ing beha io ,
su ace oughness, wea esis ance, and shea s eng h, which a e impo an p ope ies o he esul ing
epoxy esin nanocomposi es (EPNCs) o be used as adhesi es o coa ings.
The we abili y o solid su aces, including PCs, can be s udied h ough he con ac angle (CA)
o he sp eading a ea o a liquid o e he su ace. The con ac angle,
θ
, is he angle con en ionally
measu ed h ough he liquid, whe e a liquid– apo in e ace mee s a solid su ace (see Figu e 1).
Smalle CAs o la ge sp eading a eas imply highe we abili y [
20
]. The CA on a ough su ace o
he same ma e ial, also called he appa en (o measu ed) con ac angle
θ∗
(see Figu e 1), is smalle .
The e o e, he ma e ial is mo e we able. This is because he su ace a ea is inc eased due o oughness
( he eal su ace a ea is highe han he geome ical su ace a ea). The e o e, we abili y and oughness
a e s ongly ela ed. The we ing cha ac e is ics o su ace ma e ials using unc ionalized NPs depend
s ongly on he NP dis ibu ion wi hin he su ace laye s [21].
Appl. Sci. 2020, 10, 733 2 o 12
inco po a ing oughening agen s such as liquids, igid o hyb id igid ubbe y pa icles, o un eac i e
nano ha dene s (i.e., nanopa icles (NPs)) a e y low con en s (usually desc ibed in e ms o he
weigh ac ion (w . (%)) o olume ac ion ( ol. (%)) [2,3,7,8–11]. In ecen yea s, special a en ion
has been paid o EPs illed wi h NPs. These nano ille s (e.g., silicon dioxide (SiO2) [12–14], i anium
dioxide (TiO2) [15,16], aluminum oxide (Al2O3) [17–19]) a e cha ac e ized by hei speci ic shape and
size (nanoscale), high su ace a eas, su ace p e ea men , and he deg ee o dispe sion in o he EP
ma ix [1,2,4,20,21].
In ecen yea s, many s udies ha e demons a ed ha he inclusion o ino ganic NPs in o EPs
has he capabili y o imp o e he s i ness, oughness, lexu al modulus, ac u e oughness, s eng h,
ha dness, and many o he p ope ies o he inal PCs [1–3,7,8], which a e necessa y in many
enginee ing applica ions. These imp o emen s a e achie ed wi hou sac i icing he basic p ope ies
o EPs and wi h low pe cen ages o NP loading [3,8,17,22–24]. Among he a ailable NPs ha ha e
been s udied and ha can be used o ein o ce EPs, alumina NPs cons i u e a good solu ion. The good
esul s published in ecen yea s, combined wi h he low cos o alumina NPs when compa ed o
o he NPs ( o ins ance, alumina NPs a e cheape han i anium NPs), i s high modulus (abou 300
GPa), and high he mal esis ance in compa ison wi h o he s me al oxides NPs, show ha
unc ionalized alumina NPs a e a iable and e y p omising solu ion o be used as
nano ein o cemen s o PCs [1,2,4,8,22,24]. Among he cu ing agen s used as ha dene s o EPs, some
s udies show ha bis (4-aminophenyl) me hane (DDM) [1,2,25–28] o 3-dodec-2-enyloxolane-2,5-
dione (DDSA) + 8-me hyl-3a,4,7,7a- e ahyd o-4,7-me hano-2-benzo u an-1,3-dione (MNA) [28] a e
sui able o EP sys ems.
Bo h he physicochemical and opog aphical p ope ies o su ace ma e ials a e impo an o
adhesion and coa ings applica ions [29,30]. EP adhesi es illed wi h NPs ha e also been ecen ly
in es iga ed [20,21,31–32]. I was ound ha he inclusion o NPs can imp o e he we ing beha io ,
su ace oughness, wea esis ance, and shea s eng h, which a e impo an p ope ies o he
esul ing epoxy esin nanocomposi es (EPNCs) o be used as adhesi es o coa ings.
The we abili y o solid su aces, including PCs, can be s udied h ough he con ac angle (CA)
o he sp eading a ea o a liquid o e he su ace. The con ac angle, θ, is he angle con en ionally
measu ed h ough he liquid, whe e a liquid– apo in e ace mee s a solid su ace (see Figu e 1).
Smalle CAs o la ge sp eading a eas imply highe we abili y [20]. The CA on a ough su ace o
he same ma e ial, also called he appa en (o measu ed) con ac angle *
θ (see Figu e 1), is smalle .
The e o e, he ma e ial is mo e we able. This is because he su ace a ea is inc eased due o
oughness ( he eal su ace a ea is highe han he geome ical su ace a ea). The e o e, we abili y
and oughness a e s ongly ela ed. The we ing cha ac e is ics o su ace ma e ials using
unc ionalized NPs depend s ongly on he NP dis ibu ion wi hin he su ace laye s [21].
Figu e 1. De ini ion o con ac angle.
The su ace oughness o PCs can be analyzed by a omic o ce mic oscopy (AFM), which is a
ela i ely new echnique used o he su ace cha ac e iza ion o polyme s. By using AFM, i is
possible o ob ain images o a non-conduc ing polyme su ace and s udy i s mechanical p ope ies.
Compa ed wi h o he mic oscopy echniques, AFM does no in ol e any chemical e ching, s aining,
Figu e 1. De ini ion o con ac angle.
The su ace oughness o PCs can be analyzed by a omic o ce mic oscopy (AFM), which is
a ela i ely new echnique used o he su ace cha ac e iza ion o polyme s. By using AFM, i is
possible o ob ain images o a non-conduc ing polyme su ace and s udy i s mechanical p ope ies.
Compa ed wi h o he mic oscopy echniques, AFM does no in ol e any chemical e ching, s aining,
Appl. Sci. 2020,10, 733 3 o 12
o elec on beam adia ion, which can damage he polyme su ace [
6
]. F om he AFM opog aphic
images, oughness coe icien s can be compu ed, which ep esen he a io o he eal su ace a ea o
he geome ical su ace a ea o iden ical ex e nal dimensions o he sample.
We abili y and oughness a e impo an p ope ies o be ensu ed o many applica ions, such as
o pain ing and gluing componen s, o which he bonding o adhesi es and coa ing subs ances a e
inc eased o we able and oughened su aces.
This a icle p esen s an expe imen al s udy on he we abili y and he su ace oughness o EPNCs
manu ac u ed wi h a he mose ing EP, which in u n is manu ac u ed wi h wo di e en cu ing agen s
(DDM and DDSA+MNA) and illed wi h unc ionalized alumina (Al
2
O
3
) NPs wi h di e en weigh
a ios, namely 1, 3, and 5 w . (%). To s udy he we abili y, CAs we e measu ed. To s udy he su ace
oughness, AFM opog aphic images we e i s ob ained. Nex , wo ampli ude pa ame e s we e
e alua ed, namely he a i hme ic a e age oughness (
Ra
) and he oo mean squa e oughness (
Rq
).
These pa ame e s we e conside ed as measu emen s o he su ace oughness, because AFM su ace
p o iles can also be used o assess he nano ille dispe sion [
32
]. The ob ained esul s, namely he mean
alues o CAs, he images o su ace oughness om AFM, and he ela ed ampli ude pa ame e s, a e
p esen ed and discussed wi h espec o he in luence o he NP loading. The esul s a e di e en ia ed
o he wo di e en he mose ing EP sys ems, which we e ob ained by using he wo di e en cu ing
agen s (DDM and DDSA+MNA).
I should be no ed ha no p e ious s udies ocused on he we abili y and su ace oughness o
EPNCs sys ems simila o hose s udied he ein we e ound in he open li e a u e. Fo his eason,
he p esen ed s udy can be conside ed o iginal, and he ob ained esul s a e o g ea impo ance o
u u e s udies.
2. Ma e ials and Expe imen al P ocedu es
2.1. Raw Ma e ials
The used EP polyme ic ma ix was a 2-[[4-[2-[4-(Oxi an-2-ylme hoxy)phenyl]p opan-2-
yl]phenoxy]me hyl]oxi ane (D.E.R.
™
332) based on bisphenol A diglycidyl e he (DGEBA, also
known as BADGE). This he mose ing EP p esen s uni o m pe o mance, low iscosi y, low chlo ide
con en , and ligh colo . Two cu ing agen s we e used o ob ain wo di e en he mose ing
EP sys ems: bis(4-aminophenyl)me hane (DDM), and 3-dodec-2-enyloxolane-2,5-dione (DDSA) +
8-me hyl-3a,4,7,7a- e ahyd o-4,7-me hano-2-benzo u an-1,3-dione (MNA). Fo he EP cu ed wi h
DDSA+MNA, he ca alys N,N-dime hyl-1-phenylme hanamine (BDMA) was applied du ing he
cu ing eac ion. All he aw chemicals we e pu chased om Sigma-Ald ich Co.
The alumina NPs, wi h 99.99% pu i y, we e pu chased om NanoshellTM LLC. The NPs a e o
i egula shape and p esen an a e age size less han 100 nm. The speci ic su ace a ea o he NPs is
o ~ 20 m
2
/g. Figu e 2a shows a mic og aph o he as- ecei ed NPs ob ained using elec on scanning
mic oscopy (SEM), using a FEI Quan a 400 FEG E SEM mic oscope. Figu e 2b shows a g aph o he
NP size dis ibu ion, as gi en by he supplie . The as- ecei ed alumina NPs inco po a ed a su ace
p e ea men ( unc ionalized NPs) wi h a silane agen (3-Aminop opyl) ie hoxysilane (APTES) o
ensu e high dispe sion in o he EP polyme ic ma ix.
Appl. Sci. 2020,10, 733 4 o 12
Appl. Sci. 2020, 10, 733 4 o 12
(a) (b)
Figu e 2. Alumina nanopa icles (NPs): (a) scanning elec on mic oscopy (SEM) mic og aph
(x60957); (b) NP size dis ibu ion.
2.2. P epa a ion o he Epoxy Resin Nanocomposi es
Two di e en manu ac u ing p ocesses we e pe o med because wo di e en cu ing agen s
we e used (DDM and DDSA+MNA) o p oduce he EP sys ems.
In he i s manu ac u ing p ocess (MP1), he s oichiome ic amoun o he he mose sys em
cons i uen s was 3.51:1 o DER332/DDM, which was based on Ba della [33] and Belle i [34]. In he
second manu ac u ing p ocess (MP2), he chosen s oichiome ic amoun o he cons i uen s was he
same as ecommended by Ma ine e al. [35], namely 1:85:0.15:0.04 o DER332/DDSA/MNA/BDMA.
In bo h manu ac u ing p ocesses, he as- ecei ed unc ionalized alumina NPs we e added in o he
mix u e a e he p ehea ing o he EP ma ix and be o e he addi ion o he cu ing agen s. Th ee
weigh ac ions o alumina NPs we e conside ed o his s udy: 1, 3, and 5 w . (%). In addi ion, a nea
EP sys em (0 w . (%)) was also manu ac u ed o compa ison. Fo bo h MP1 and MP2, he cu ing
s age was ca ied ou in an o en (P ecision Scien i ic Napco acuum o en model 5831) unde acuum
condi ions (20 mm Hg) a 60 °C o 24 h [33]. Fo MP2, an addi ional pos cu ing cycle wi h wo
in e al ime s eps was pe o med on he cu ed samples: i s ly o 2 h a 100 °C, and secondly o
addi ional 3 h a 180 °C [35].
The ob ained EPNCs we e allowed o cool a oom empe a u e. A e 24 h, he samples we e
demolded and cu in o de o ob ain small geome ical samples o he measu emen s.
Addi ional de ails abou he successi e s eps used in MP1 and MP2 o ab ica e he EPNCs can
be ound in p e ious wo ks om he au ho s [1,2,28].
2.3. Expe imen al P ocedu es
CAs on EPNCs samples (wi hou sa u a ion) we e measu ed wi h a con ac angle es e using
he sessile d op echnique (DSA100 d op shape analyze , K üss GmbH [36]). Figu e 1 illus a es he
CA,
θ
, o an ideal (smoo h) su ace [37]. In his s udy, he su ace o he EPNCs samples whe e no
polished and he appa en (measu ed) CA,
*θ
, on a ough su ace was ob ained. The measu emen s
we e pe o med a oom empe a u e and on d y samples. The p obe liquid consis ed o ul apu e
wa e in a dosing sy inge o 500 µL wi h a needle 1.81 mm diame e . The d op olume anged om
1 o 50 µL on he solid su aces o EPNC samples. The esul s we e based on he a e age alues o a
leas i e eplica es.
Topog aphic images o he EPNCs su aces we e ob ained om AFM by using a Nanosu
FlexAFM Sys em (Flex-Axiom). EPNC samples we e cleaned be o ehand by wiping he su ace wi h
99% e hanol. Nex , hey we e d ied wi h a pape owel o emo e he excess e hanol and subjec ed o
0
5
10
15
20
25
30
[0,20]
[20,40]
[40,60]
[60,80]
[80,100]
[100,120]
[120,140]
[140,160]
[160,180]
[180,200]
[200,220]
[220,240]
[240,260]
[260,280]
Con en s (%)
Size (nm)
Figu e 2.
Alumina nanopa icles (NPs): (
a
) scanning elec on mic oscopy (SEM) mic og aph (x60957);
(b) NP size dis ibu ion.
2.2. P epa a ion o he Epoxy Resin Nanocomposi es
Two di e en manu ac u ing p ocesses we e pe o med because wo di e en cu ing agen s we e
used (DDM and DDSA+MNA) o p oduce he EP sys ems.
In he i s manu ac u ing p ocess (MP1), he s oichiome ic amoun o he he mose sys em
cons i uen s was 3.51:1 o DER332/DDM, which was based on Ba della [
33
] and Belle i [
34
]. In he
second manu ac u ing p ocess (MP2), he chosen s oichiome ic amoun o he cons i uen s was he
same as ecommended by Ma ine e al. [
35
], namely 1:85:0.15:0.04 o DER332/DDSA/MNA/BDMA.
In bo h manu ac u ing p ocesses, he as- ecei ed unc ionalized alumina NPs we e added in o he
mix u e a e he p ehea ing o he EP ma ix and be o e he addi ion o he cu ing agen s. Th ee
weigh ac ions o alumina NPs we e conside ed o his s udy: 1, 3, and 5 w . (%). In addi ion, a
nea EP sys em (0 w . (%)) was also manu ac u ed o compa ison. Fo bo h MP1 and MP2, he cu ing
s age was ca ied ou in an o en (P ecision Scien i ic Napco acuum o en model 5831) unde acuum
condi ions (20 mm Hg) a 60
◦
C o 24 h [
33
]. Fo MP2, an addi ional pos cu ing cycle wi h wo in e al
ime s eps was pe o med on he cu ed samples: i s ly o 2 h a 100
◦
C, and secondly o addi ional 3
h a 180 ◦C [35].
The ob ained EPNCs we e allowed o cool a oom empe a u e. A e 24 h, he samples we e
demolded and cu in o de o ob ain small geome ical samples o he measu emen s.
Addi ional de ails abou he successi e s eps used in MP1 and MP2 o ab ica e he EPNCs can be
ound in p e ious wo ks om he au ho s [1,2,28].
2.3. Expe imen al P ocedu es
CAs on EPNCs samples (wi hou sa u a ion) we e measu ed wi h a con ac angle es e using
he sessile d op echnique (DSA100 d op shape analyze , K üss GmbH [
36
]). Figu e 1illus a es he
CA,
θ
, o an ideal (smoo h) su ace [
37
]. In his s udy, he su ace o he EPNCs samples whe e no
polished and he appa en (measu ed) CA,
θ∗
, on a ough su ace was ob ained. The measu emen s
we e pe o med a oom empe a u e and on d y samples. The p obe liquid consis ed o ul apu e
wa e in a dosing sy inge o 500
µ
L wi h a needle 1.81 mm diame e . The d op olume anged om 1
o 50
µ
L on he solid su aces o EPNC samples. The esul s we e based on he a e age alues o a
leas i e eplica es.
Topog aphic images o he EPNCs su aces we e ob ained om AFM by using a Nanosu
FlexAFM Sys em (Flex-Axiom). EPNC samples we e cleaned be o ehand by wiping he su ace wi h
99% e hanol. Nex , hey we e d ied wi h a pape owel o emo e he excess e hanol and subjec ed o a
Appl. Sci. 2020,10, 733 5 o 12
je o comp essed ai un il comple e d yness. The measu emen s we e pe o med wi h apping mode
(phase con as ) wi h a scanning a e be ween 1 and 2.5 Hz using a non-con ac long e lex (NCLR)
p obe (
k=
48
N/m
,
=
190
kHz
). Fo each EPNC sample, mul iple images om a eas o 10
×
10
µm2
we e acqui ed a di e en loca ions.
Based he ob ained opog aphic images o he EPNCs su aces and using Gwyddion so wa e
o analyze AFM da a, ampli ude pa ame e s we e e alua ed. These pa ame e s cha ac e ize he
su ace based on he e ical de ia ions o he oughness map om he mean su ace. Basically, hey
educe all o he in o ma ion om he opog aphic images o a single numbe . In his s udy, wo
ampli ude pa ame e s we e e alua ed [
37
]: he a i hme ic a e age oughness
Ra
(also called AA o
CLA; Equa ion (1)) and he oo mean squa e oughness
Rq
(also called
R ms
; Equa ion (2)). These
ampli ude pa ame e s a e widely used in he li e a u e.
Ra=1
nxny
nx
X
i=1
ny
X
j=1
[Z(i,j)−Za e](1)
Rq=
u
u
u
u
u
u
nx
P
i=1
ny
P
j=1
[Z(i,j)−Za e]2
nxny(2)
In he p e ious equa ions, he meaning o he pa ame e s a e he ollowing ones:
i
and
j
co esponds o pixels in
x
and
y
di ec ion, espec i ely,
nx
and
ny
a e he maximum numbe o pixels in
he wo di ec ions,
Z(i
,
j)
ep esen s he opog aphy da a o he su ace (a e specimen il -co ec ion),
and Za e ep esen s he heigh o he a e age su ace.
3. Resul s and Discussion
3.1. Con ac Angle
The appa en CA,
θ∗
, was measu ed o e alua e he we abili y o he EPNCs manu ac u ed
wi h a he mose ing epoxy esin ha dened wi h wo di e en cu ing agen s and illed wi h di e en
pe cen ages o alumina NPs (1, 3, and 5 w . (%)). The we abili y is ecognized sa is ac o y when he
CA is less han 90◦[38] ( he solid su ace is conside ed o be hyd ophobic i he CA is abo e 90◦, and
conside ed supe -hyd ophobic i he CA is abo e 120
◦
), which means ha he sample su ace abso bs
wa e . Figu e 3p esen s he ob ained esul s o he appa en con ac angle θ∗.
Appl. Sci. 2020, 10, 733 5 o 12
a je o comp essed ai un il comple e d yness. The measu emen s we e pe o med wi h apping
mode (phase con as ) wi h a scanning a e be ween 1 and 2.5 Hz using a non-con ac long e lex
(NCLR) p obe ( 48 N/mk=, 190 kHz =). Fo each EPNC sample, mul iple images om a eas o
2
10 10 m×
μ
we e acqui ed a di e en loca ions.
Based he ob ained opog aphic images o he EPNCs su aces and using Gwyddion so wa e o
analyze AFM da a, ampli ude pa ame e s we e e alua ed. These pa ame e s cha ac e ize he su ace
based on he e ical de ia ions o he oughness map om he mean su ace. Basically, hey educe
all o he in o ma ion om he opog aphic images o a single numbe . In his s udy, wo ampli ude
pa ame e s we e e alua ed [37]: he a i hme ic a e age oughness a
R (also called AA o CLA;
Equa ion (1)) and he oo mean squa e oughness q
R (also called ms
R; Equa ion (2)). These
ampli ude pa ame e s a e widely used in he li e a u e.
11
1(, )
y
xn
n
aa e
ij
xy
RZijZ
nn ==
=−
 (1)
[]
2
11
(, )
y
xn
n
a e
ij
q
xy
Zi j Z
Rnn
==
−
=

(2)
In he p e ious equa ions, he meanings o he pa ame e s a e as ollows: i and
j
co espond
o pixels in
x
and y di ec ions, espec i ely, while
x
n and
y
n a e he maximum numbe o
pixels in he wo di ec ions; (, )
Z
ij ep esen s he opog aphy da a o he su ace (a e specimen
il co ec ion) and a e
Z
ep esen s he heigh o he a e age su ace.
3. Resul s and Discussion
3.1. Con ac Angle
The appa en CA, *
θ, was measu ed o e alua e he we abili y o he EPNCs manu ac u ed
wi h a he mose ing epoxy esin ha dened wi h wo di e en cu ing agen s and illed wi h di e en
pe cen ages o alumina NPs (1, 3, and 5 w . (%)). The we abili y is ecognized sa is ac o y when he
CA is less han 90° [38] ( he solid su ace is conside ed o be hyd ophobic i he CA is abo e 90°, and
conside ed supe -hyd ophobic i he CA is abo e 120°), which means ha he sample su ace abso bs
wa e . Figu e 3 p esen s he ob ained esul s o he appa en con ac angle *
θ.
123
112
98
85
74
68
113
104
0
20
40
60
80
100
120
140
0
20
40
60
80
100
120
140
CS 1 w .(%) 3 w .(%) 5 w .(%)
DDSA+MNA (θ
∗
[deg])
DDM (θ
*
[deg])
DDM DDSA+MNA
Figu e 3.
Appa en con ac angles o epoxy esin nanocomposi es (EPNCs) using
bis(4-aminophenyl)me hane (DDM) and 3-dodec-2-enyloxolane-2,5-dione (DDSA) +
8-me hyl-3a,4,7,7a- e ahyd o-4,7-me hano-2-benzo u an-1,3-dione (MNA) as ha dene s.

Appl. Sci. 2020,10, 733 6 o 12
F om Figu e 3, i is possible o obse e ha he con ol sample (CS) (nea EP) using DDM as
ha dene is supe -hyd ophobic and displays CAs o abou 123
±
3
◦
. Wi h he inclusion o alumina
NPs, he hyd ophobici y dec eases by a ound 9%, 20%, and 31% o 1, 3, and 5 w . (%), espec i ely,
when compa ed wi h CS. Addi ionally, o 1 and 3 w . (%) samples, he EPNCs become hyd ophobic,
while o 5 w . (%) i u ns hyd ophilic. The opposi e end is de ec ed o samples wi h DDSA+MNA
ha dene . In ac , CS using DDSA+MNA is obse ed o be hyd ophilic wi h a CA o a ound 74
±
3
◦
,
and a sligh di e ence is obse ed wi h he inclusion o 1 w . (%) alumina NPs. Howe e , wi h u he
inc ease o he w . (%) NPs, he we abili y dec eases and he EPNCs become hyd ophobic, which
is in acco dance wi h he obse a ion o Hill e al. in 2019 [
21
]. F om Figu e 3, he majo inc ease
o abou 53% compa ed wi h CS was obse ed o 3 w . (%) alumina NPs. Fo 5 w . (%) alumina
NPs, he inc ease was abou 40% when compa ed wi h CS. The dec ease in he CA sugges s ha
he we ing abili y o he EPNCs inc eased [
20
]. Acco ding o Ama o e al. [
1
], be e dispe sion o
NPs was achie ed in he case o 1 w . (%) NPs, while a 3 and 5 w . (%), some NPs agglome a ions
we e obse ed, which explains he inc ease o CA o such pe cen age loadings o he case wi h
DDSA+MNA ha dene . In ac , acco ding o Radiom e al. [
39
], a highe NP concen a ion leads o a
la ge a e age pa icle size and induces highe CA. In he case o EPNCs wi h DDM as he ha dene ,
be e we ing abili y was obse ed o 5 w . (%), while in he case o EPNCs wi h DDSA+MNA, a
smalle CA was obse ed o 1 w . (%). When compa ed wi h CS, he EPNCs wi h DDSA+MNA
con aining 1 w . (%) alumina NPs showed he be e CA. The obse ed dec ease in he con ac angle
o he case o he DDM ha dene could due o he ype o ma ix he mose ing sys em used. This is in
acco dance wi h Syaku e al. [40], who s a ed ha he hyd ophobici y o he su ace is in luenced by
he ype o ma ix and he composi ion o he ille cons i uen .
3.2. Su ace Roughness
The su ace oughness o he EPNCs was s udied h ough AFM. The a i hme ic a e age oughness
Ra
and he oo mean squa e oughness
Rq
we e compu ed om mul iple AFM opog aphical images
o a eas o 10
×
10
µm2
a di e en loca ions o he EPNC su ace. The ob ained esul s we e a e aged
and he esul s a e p esen ed in Table 1. F om his able, he ends obse ed o bo h ampli ude
pa ame e s (
Ra
and
Rq
) a e e y simila . Fo his eason, in Figu e 4only he esul s ela ed wi h
Ra
a e
p esen ed g aphically.
Table 1. Roughness o EPNC su aces.
Sample w/w%Ranm S .De . nm Coe . a .% Rqnm S .De . nm Coe . a . %
DDM_0 0 197.6 26.8 13.6 267.7 59.0 22.0
DDM_1 1 202.9 5.4 2.7 243.7 5.2 2.1
DDM_3 3 68.7 12.5 18.2 95.1 16.8 17.7
DDM_5 5 233.6 16.3 7.0 289.8 40.4 13.9
DDSA+MNA_0 0 331.6 56.7 17.1 487.1 120.8 24.8
DDSA+MNA_1 1 256.2 55.7 21.7 324.3 68.8 21.2
DDSA+MNA_3 3 17.4 2.7 15.5 21.6 3.6 16.7
DDSA+MNA_5 5 136.0 16.9 12.4 186.5 22.4 12.0
S .De —S anda d de ia ion; Coe . Va .—S .De ./Value.
Appl. Sci. 2020,10, 733 7 o 12
Appl. Sci. 2020, 10, 733 7 o 12
Figu e 4. A i hme ic a e age oughness o EPNCs using DDM and DDSA+MNA as ha dene s.
F om Table 1 and Figu e 4, i is possible o obse e ha o bo h ha dene s (DDM and
DDSA+MNA), a gene al dec ease o he su ace oughness ( a
R
) is obse ed un il 3 w . (%) is eached,
ollowing by an inc ease o 5 w . (%). In he case o DDM ha dene , he highe alue o he su ace
oughness is ob ained o 5 w . %, while o he case o DDSA+MNA ha dene he highe alue is
ob ained o he CS. F om Table 1, i is obse ed ha EPNCs wi h 3 w . (%) show dec eases in a
R
(
q
R
) o a ound 65% and 64% (95% and 96%) when compa ed wi h he CS wi h DDM and
DDSA+MNA, espec i ely.
Acco ding o Figu e 4, he EPNC su ace ha shows he highes a e age su ace oughness is
he CS; o bo h DDM and DDSA+MNA, he is he highes a e age su ace oughness alues a e o
supe -hyd ophobic and hyd ophilic EPNCs, espec i ely. Fo DDM, he highes a e age su ace
oughness was o he su ace co esponding o he EPNC wi h 5 w . (%) NPs, which was hyd ophilic;
and o EPNC his co esponded o he 1 w . (%) NP o he case o DDSA+MNA, which was also
hyd ophilic. Fo he DDM ha dene , he esul s a e in ag eemen wi h Uelzen and Mulle [41], who
s a ed ha o smoo h hyd ophilic su aces, he we abili y is imp o ed by oughening hem, while
o smoo h hyd ophobic su aces he opposi e is obse ed. In addi ion, CA will inc ease by
oughening he su aces. Howe e , in his s udy and o EPNCs wi h DDSA+MNA, he opposi e
e ec is obse ed. In ac , conside ing he hyd ophobic su aces o EPNCs wi h 3 and 5 w . (%) NPs,
he con ac angle dec eases wi h he oughness. Howe e , Uelzen and Mulle [41] s udied ilms,
which could explain why he beha io is di e en . Acco ding o he au ho ’s knowledge, he e a e
no s udies conce ning he s udy o CA and oughness o nanoalumina- illed epoxy esin
nanocomposi es.
Figu es 5 and 6 p esen ep esen a i e 2D and 3D AFM images o EPNCs cu ed wi h DDM and
DDSA+MNA, espec i ely. F om Figu es 5 and 6, i is possible o obse e some a iabili y o he
su ace oughness o he s udied EPNCs on he selec ed egions. Howe e , he in luence in he
oughness esul s is no e y high, since he coe icien s o a ia ion p esen ed in Table 1 a e gene ally
below 20%.
Figu e 4. A i hme ic a e age oughness o EPNCs using DDM and DDSA+MNA as ha dene s.
F om Table 1and Figu e 4, i is possible o obse e ha o bo h ha dene s (DDM and DDSA+MNA),
a gene al dec ease o he su ace oughness (
Ra
) is obse ed un il 3 w . (%) is eached, ollowing by
an inc ease o 5 w . (%). In he case o DDM ha dene , he highe alue o he su ace oughness is
ob ained o 5 w . %, while o he case o DDSA+MNA ha dene he highe alue is ob ained o he
CS. F om Table 1, i is obse ed ha EPNCs wi h 3 w . (%) show dec eases in
Ra
(
Rq
) o a ound 65%
and 64% (95% and 96%) when compa ed wi h he CS wi h DDM and DDSA+MNA, espec i ely.
Acco ding o Figu e 4, he EPNC su ace ha shows he highes a e age su ace oughness is
he CS; o bo h DDM and DDSA+MNA, he is he highes a e age su ace oughness alues a e
o supe -hyd ophobic and hyd ophilic EPNCs, espec i ely. Fo DDM, he highes a e age su ace
oughness was o he su ace co esponding o he EPNC wi h 5 w . (%) NPs, which was hyd ophilic;
and o EPNC his co esponded o he 1 w . (%) NP o he case o DDSA+MNA, which was also
hyd ophilic. Fo he DDM ha dene , he esul s a e in ag eemen wi h Uelzen and Mulle [
41
], who
s a ed ha o smoo h hyd ophilic su aces, he we abili y is imp o ed by oughening hem, while o
smoo h hyd ophobic su aces he opposi e is obse ed. In addi ion, CA will inc ease by oughening
he su aces. Howe e , in his s udy and o EPNCs wi h DDSA+MNA, he opposi e e ec is obse ed.
In ac , conside ing he hyd ophobic su aces o EPNCs wi h 3 and 5 w . (%) NPs, he con ac angle
dec eases wi h he oughness. Howe e , Uelzen and Mulle [
41
] s udied ilms, which could explain
why he beha io is di e en . Acco ding o he au ho ’s knowledge, he e a e no s udies conce ning
he s udy o CA and oughness o nanoalumina- illed epoxy esin nanocomposi es.
Figu es 5and 6p esen ep esen a i e 2D and 3D AFM images o EPNCs cu ed wi h DDM
and DDSA+MNA, espec i ely. F om Figu es 5and 6, i is possible o obse e some a iabili y o
he su ace oughness o he s udied EPNCs on he selec ed egions. Howe e , he in luence in he
oughness esul s is no e y high, since he coe icien s o a ia ion p esen ed in Table 1a e gene ally
below 20%.
Appl. Sci. 2020,10, 733 8 o 12
Appl. Sci. 2020, 10, 733 8 o 12
(a)
(b)
Figu e 5. Rep esen a i e 2D and 3D a omic o ce mic oscopy (AFM) images o EPNCs wi h DDM:
(a) con ol sample (CS); (b) 5 w . (%) sample.
(a)
(b)
Figu e 6. Rep esen a i e 2D and 3D AFM images o EPNCs wi h DDSA+MNA: (a) CS; (b) 3 w . (%)
sample.
Figu e 7 illus a es he compa ison be ween he oughness ( a
R
) and he CA (
*θ
). F om he
igu e, i is possible o conclude ha in he case o EPNCs wi h DDM ha dene , he we abili y is
imp o ed o he mo e hyd ophilic su aces wi h he inc ease in oughness. Fo EPNCs wi h
DDSA+MNA ha dene , he lowe alue o he oughness co esponds o he highe alue o CA.
These esul s a e in ag eemen wi h Syaku e al. [40], e i ying ha he su ace hyd ophobici y o
Figu e 5.
Rep esen a i e 2D and 3D a omic o ce mic oscopy (AFM) images o EPNCs wi h DDM: (
a
)
con ol sample (CS); (b) 5 w . (%) sample.
Appl. Sci. 2020, 10, 733 8 o 12
(a)
(b)
Figu e 5. Rep esen a i e 2D and 3D a omic o ce mic oscopy (AFM) images o EPNCs wi h DDM:
(a) con ol sample (CS); (b) 5 w . (%) sample.
(a)
(b)
Figu e 6. Rep esen a i e 2D and 3D AFM images o EPNCs wi h DDSA+MNA: (a) CS; (b) 3 w . (%)
sample.
Figu e 7 illus a es he compa ison be ween he oughness ( a
R
) and he CA (
*θ
). F om he
igu e, i is possible o conclude ha in he case o EPNCs wi h DDM ha dene , he we abili y is
imp o ed o he mo e hyd ophilic su aces wi h he inc ease in oughness. Fo EPNCs wi h
DDSA+MNA ha dene , he lowe alue o he oughness co esponds o he highe alue o CA.
These esul s a e in ag eemen wi h Syaku e al. [40], e i ying ha he su ace hyd ophobici y o
Figu e 6.
Rep esen a i e 2D and 3D AFM images o EPNCs wi h DDSA+MNA: (
a
) CS; (
b
) 3 w .
(%) sample.
Figu e 7illus a es he compa ison be ween he oughness (
Ra
) and he CA (
θ∗
). F om he igu e,
i is possible o conclude ha in he case o EPNCs wi h DDM ha dene , he we abili y is imp o ed o
he mo e hyd ophilic su aces wi h he inc ease in oughness. Fo EPNCs wi h DDSA+MNA ha dene ,
he lowe alue o he oughness co esponds o he highe alue o CA. These esul s a e in ag eemen
wi h Syaku e al. [
40
], e i ying ha he su ace hyd ophobici y o epoxy esin is in luenced by he
Appl. Sci. 2020,10, 733 9 o 12
chemical su ace o componen s. Also, Kubiak e al. [
42
] obse ed ha he oughness had a s ong
in luence on he we abili y o enginee ing su aces.
Appl. Sci. 2020, 10, 733 9 o 12
epoxy esin is in luenced by he chemical su ace o componen s. Also, Kubiak e al. [42] obse ed
ha he oughness had a s ong in luence on he we abili y o enginee ing su aces.
In ac , i was expec ed ha as he oughness inc eases, he CA should also inc ease. This can be
explained by he “lo us e ec ”, whe eby lo us lea es epel wa e because o he p esence o small
bumps a he mic o- and nanome e scale on he su ace [43]. In his case, a d op o wa e is suppo ed
be ween he peaks o he p o ube ances, which is he sho es con ac , he e o e acqui ing a mo e
sphe ical shape ( ha is, wi h a highe CA). The obse ed di e en beha io o he es ed samples
can be explained by he cu ing agen s and manu ac u ing p ocedu es used. In ac , in he case o
DDSA+MNA, a pos cu ing cycle was pe o med o manu ac u e he EPNCS, while o he case o
DDM he EPNCs, hese we e no subjec ed o pos cu ing. Acco ding o Ama o e al. [2], he deg ee
o enhancemen o a speci ic p ope y is highly dependen on he ype o cu ing agen used, and also
on he s a e o he dispe sion o he NPs h oughou he EP esin ma ix.
Figu e 7. Roughness e sus con ac angle.
In p e ious s udies, a co ela ion be ween he mesoscopic s uc u e and we abili y has been
epo ed o bionanocomposi e ilms [44,45]. In pa icula , i was obse ed ha he inclusion o
nano ubes in o he s udied composi e ilms led o a signi ican dec ease o he CA. The au ho s
explained his obse a ion as being due o he p esence o hyd ophilic nano ubes on he ilm su ace.
In his s udy, unc ionalized alumina NPs wi h APTES we e used (Sec ion 2.1). Hence, amine g oups
exis a he su ace o he NPs, which make hem hyd ophilic. F om his poin o iew, he esul s in
his s udy should also show a gene al dec ease o he CA due o he inclusion o he alumina NPs in
he EP ma ix. This is obse ed o EPNCs wi h DDM, bu no in gene al o EPNCs wi h
DDSA+MNA (Figu e 7). As p e iously men ioned, hese di e ences can be explained by he cu ing
agen s and manu ac u ing p ocedu es used.
4. Conclusions
This wo k s udied he su ace p ope ies (namely he we abili y and oughness) o EPNCs
manu ac u ed wi h di e en cu ing agen s (DDM and DDSA+MNA) and ein o ced wi h alumina
NPs. The weigh ac ions o alumina NPs we e 1, 3, and 5 w . (%).
F om he ob ained esul s, i was concluded ha he CA dec eases as he weigh ac ion o
alumina NPs inc eases in he case o EPNCs wi h DDM as he cu ing agen , p esen ing hyd ophobic
beha io ( 90 * 120<θ <

) un il 3 w . (%) and hyd ophilic beha io ( *90θ< ) o he 5 w . (%)
sample. Fo EPNCs wi h DDSA+MNA as he ha dene , complex beha io is achie ed and he
addi ion o alumina NPs p omo es a dec ease in CA in he CS un il 1 w . (%), wi h bo h being
hyd ophilic, ollowed by an inc ease o CA o 3 w . (%) NPs and a small dec ease o 5 w . (%) NPs.
Fo he wo las NP loadings, he EPNCs we e shown o be hyd ophobic. The highe loading o
alumina NPs (5 w . (%)) we e shown o ha e a nega i e e ec on he hyd ophobici y o EPNCs wi h
DDM as he ha dene , which became hyd ophilic.
Figu e 7. Roughness e sus con ac angle.
In ac , i was expec ed ha as he oughness inc eases, he CA should also inc ease. This can
be explained by he “lo us e ec ”, whe eby lo us lea es epel wa e because o he p esence o small
bumps a he mic o- and nanome e scale on he su ace [
43
]. In his case, a d op o wa e is suppo ed
be ween he peaks o he p o ube ances, which is he sho es con ac , he e o e acqui ing a mo e
sphe ical shape ( ha is, wi h a highe CA). The obse ed di e en beha io o he es ed samples
can be explained by he cu ing agen s and manu ac u ing p ocedu es used. In ac , in he case o
DDSA+MNA, a pos cu ing cycle was pe o med o manu ac u e he EPNCS, while o he case o
DDM he EPNCs, hese we e no subjec ed o pos cu ing. Acco ding o Ama o e al. [
2
], he deg ee o
enhancemen o a speci ic p ope y is highly dependen on he ype o cu ing agen used, and also on
he s a e o he dispe sion o he NPs h oughou he EP esin ma ix.
In p e ious s udies, a co ela ion be ween he mesoscopic s uc u e and we abili y has been
epo ed o bionanocomposi e ilms [
44
,
45
]. In pa icula , i was obse ed ha he inclusion o
nano ubes in o he s udied composi e ilms led o a signi ican dec ease o he CA. The au ho s
explained his obse a ion as being due o he p esence o hyd ophilic nano ubes on he ilm su ace.
In his s udy, unc ionalized alumina NPs wi h APTES we e used (Sec ion 2.1). Hence, amine g oups
exis a he su ace o he NPs, which make hem hyd ophilic. F om his poin o iew, he esul s in
his s udy should also show a gene al dec ease o he CA due o he inclusion o he alumina NPs in he
EP ma ix. This is obse ed o EPNCs wi h DDM, bu no in gene al o EPNCs wi h DDSA+MNA
(Figu e 7). As p e iously men ioned, hese di e ences can be explained by he cu ing agen s and
manu ac u ing p ocedu es used.
4. Conclusions
This wo k s udied he su ace p ope ies (namely he we abili y and oughness) o EPNCs
manu ac u ed wi h di e en cu ing agen s (DDM and DDSA+MNA) and ein o ced wi h alumina
NPs. The weigh ac ions o alumina NPs we e 1, 3, and 5 w . (%).
F om he ob ained esul s, i was concluded ha he CA dec eases as he weigh ac ion o
alumina NPs inc eases in he case o EPNCs wi h DDM as he cu ing agen , p esen ing hyd ophobic
beha io (90
◦<θ∗<
120
◦
) un il 3 w . (%) and hyd ophilic beha io (
θ∗<
90
◦
) o he 5 w . (%)
sample. Fo EPNCs wi h DDSA+MNA as he ha dene , complex beha io is achie ed and he addi ion
o alumina NPs p omo es a dec ease in CA in he CS un il 1 w . (%), wi h bo h being hyd ophilic,
ollowed by an inc ease o CA o 3 w . (%) NPs and a small dec ease o 5 w . (%) NPs. Fo he wo
las NP loadings, he EPNCs we e shown o be hyd ophobic. The highe loading o alumina NPs (5
w . (%)) we e shown o ha e a nega i e e ec on he hyd ophobici y o EPNCs wi h DDM as he
ha dene , which became hyd ophilic.