applied
sciences
A icle
The High-Veloci y Impac Beha iou o Ke la
Composi e Lamina es Filled wi h Co k Powde
Ana Ma ins Ama o 1,* , Paulo Nob e Balbis Reis 2, Ines I añez 3, Sonia Sánchez-Saez 3,
Shi ley Kalamis Ga cia-Cas illo 3and En ique Ba be o 3
1CEMMPRE, Depa men o Mechanical Enginee ing, Uni e si y o Coimb a, 3030-788 Coimb a, Po ugal
2C-MAST, Depa men o Elec omechanical Enginee ing, Uni e si y o Bei a In e io ,
6201-100 Co ilhã, Po ugal; [email p o ec ed]
3Depa men o Con inuum Mechanics and S uc u al Analysis, Uni e si y Ca los III o Mad id,
28911 Leganés, Spain; [email p o ec ed] (I.I.); [email p o ec ed] (S.S.-S.);
[email p o ec ed] (S.K.G.-C.); [email p o ec ed] (E.B.)
*Co espondence: [email p o ec ed]; Tel.: +351-239-790-700
Recei ed: 25 June 2020; Accep ed: 31 Augus 2020; Published: 3 Sep embe 2020


Abs ac :
The li e a u e epo s bene i s when he co k powde ob ained om indus ial by-p oduc s
is used as he ille o composi e lamina es. Fo example, while he a igue li e is insensi i e o he
p esence o co k in he esin, signi ican imp o emen s a e achie ed in e ms o o low- eloci y impac
s eng h. Howe e , in e ms o ballis ic domain, he li e a u e does no ye epo any s udy abou
he e ec o inco po a ing powde ed co k in o esins. The e o e, his s udy in ended o analyse he
ballis ic beha iou and damage ole ance o Ke la /epoxy ein o ced composi es wi h ma ix illed by
co k powde . Fo his pu pose, high- eloci y impac s we e s udied on pla es o Ke la bi-di ec ional
wo en lamina es wi h su aces o 100
×
100 mm
2
. I was possible o conclude ha he minimum
eloci y o pe o a ion is 1.6% highe when he co k powde is added o he esin, bu conside ing
he dispe sion, his small di e ence can be neglec ed. In e ms o damage a eas, hey a e sligh ly
lowe when co k dus is added, especially o eloci ies below he minimum pe o a ion eloci y.
Finally, he esidual bending s eng h shows ha hese composi es a e less sensi i e o impac eloci y
han he samples wi h nea esin. In addi ion o hese bene i s, co k powde educes he amoun o
esin in he composi e, making i mo e en i onmen ally iendly.
Keywo ds: Ke la composi e lamina es; ballis ic impac ; co k powde ; mechanical es ing
1. In oduc ion
The indus ial wo ld is unde eno mous p essu e o use mo e sus ainable ma e ials in hei
p oduc s, in o de o alle ia e p oblems ela ed o ai pollu ion, global wa ming and sho ages o ossil
uels. Acco ding o s udies published in 2015, o example, he anspo sec o alone is esponsible o
25% o g eenhouse emissions in Eu ope, 16% in Aus alia and 23% in he USA [
1
], and his endency is
o inc ease i no hing is done o coun e i .
Fib e- ein o ced polyme s a e a ac i e candida es o eplace he adi ional ma e ials,
because hey ha e high speci ic s eng h and s i ness, and excellen a igue esis ance and s abili y.
Howe e , hese ma e ials a e also subjec ed o he same sus ainable p essu es, and oday, i is possible
o obse e a signi ican inc ease in he use o na u al ib es. They a e enewable, biodeg adable and
a ailable h oughou he wo ld. The a ious e iew a icles a ailable in he open li e a u e deno e his
eno mous in e es [2–7], e en in e ms o impac s eng h [8–10].
Na u al ib es a e s ong enough, ligh in weigh , abundan , non-ab asi e and cheap [
11
–
13
].
When compa ed o o he ib es, hey ha e a speci ic weigh ha is abou hal o a glass ib e’s weigh
Appl. Sci. 2020,10, 6108; doi:10.3390/app10176108 www.mdpi.com/jou nal/applsci
Appl. Sci. 2020,10, 6108 2 o 12
and a ensile modulus qui e simila o ha o a amid ib es [
14
]. Mo eo e , ib es wi h highe cellulose
con en , highe deg ees o polyme isa ion and lowe mic o ib illa angles exhibi highe ensile
s eng h and moduli [
15
]. On he o he hand, he main disad an ages a e hei lowe he mal s abili y,
which limi s hei use wi h some he moplas ics, and low esis ance o mois u e and dimensional
s abili y, which can lead o debonding and mic oc acking in he composi e [16].
Howe e , he e is eal e idence ha o es s in de eloped coun ies a e dwindling and o he na u al
p oduc s a e eme ging, especially hose ob ained om ag icul u al by-p oduc s o ag o-was e ma e ials.
In his con ex , he di e en co k was es esul ing om he manu ac u e o a ious co k-based p oduc s
a e e y a ac i e because hey main ain he same in insic cha ac e is ics o co k, such as low speci ic
weigh , g ea elas ici y, lexibili y and du abili y, speci ic s i ness and s eng h, impe meabili y o
liquid and gases, esis ance o wea and i e, dimensional s abili y and esis ance o eac i e agen s
and mic oo ganisms [
17
–
20
]. These co k was es p esen di e en densi ies, amoun s o mois u e,
g anulome ies, sizes, le els o ash con en and annin concen a ions [
21
,
22
], and all o hem we e
con enien ly cha ac e ized by Gil [
23
]. Fo example, co k powde is he mos impo an was e om co k
p ocessing, whe e only he co k-s oppe p oduc ion is esponsible o 25% o 30% o he aw ma e ial.
Acco ding o he Po uguese S anda ds NP-114 and NP-273, co k powde has dimensions lowe han
0.25 mm and he e a e di e en ypes acco ding o hei o igin: he g inding powde , om g anula ion
o p e-g inding; he cleaning powde , wi hou impu i ies; he inishing powde , om cu and sanding
ope a ions; he agglome a ed co k panels’ inishing powde ; he agglome a ed co k s oppe s and
disks’ inishing powde ; he insula ion co k boa d powde [
23
,
24
]. These powde s ha e been used
mainly as combus ion uels (due o i s high hea ing alue) [
22
,
23
]. A small ac ion a e also used as
illing agen s ( o imp o e he quali y o co k-s oppe s) [
23
,
24
]; in he p oduc ion o linoleum [
23
,
24
];
in applica ions in agglome a es [
23
,
24
]; as b ique es [
23
,
24
]; as ag icul u al subs a es [
23
,
24
]; and mo e
ecen ly, co k powde s ha e been used as pollu an adso ben s o as abso ben s in oil spli s due o
hei good adso p ion p ope ies [
24
–
26
], and in he p oduc ion o composi es due o hei physical
and mechanical p ope ies [27,28].
Sanchez-Saez e al. [
29
] de eloped an expe imen al s udy, whe e he mul i-impac beha iou o
agglome a ed-co k specimens was analysed. They obse ed a g ea capabili y o agglome a ed co k o
abso b ene gy e en a e se e al consecu i e impac s. The damage esis ance o sandwich s uc u es
wi h agglome a ed co k co e and lax/epoxy lamina es ace shee s subjec ed o low- and high- eloci y
impac s was e alua ed by Sa asini e al. [
30
]. They obse ed ha co k cell walls expe ienced buckling,
and in his con ex , he ene gy abso p ion a high s ains can be qui e conside able wi hou signs
o cell walls b eakage. This sugges an almos comple e eco e y o hei o iginal shape and size.
Mo eo e , he composi e s uc u es du ing i s se ice li e can be damaged due o impac loading,
and his damage in composi e ma e ial is o p ima y conce n because i leads o la ge educ ion in
s eng h and comp omises hei s uc u al in eg i y [30,31].
Reis e al. [
32
], o example, de eloped s udies in composi es inco po a ing co k powde , ob ained
om indus ial by-p oduc s, and obse ed bene i s in e ms o impac s eng h and glass ansi ion
empe a u e (T
g
). When 2.5 w .% was added o he polyes e esin, he impac s eng h inc eased
o a ound 18.5% and he T
g
o abou 7.3% compa ed o he alues ob ained wi h he p is ine esin.
In ano he expe imen al s udy [
33
], hese au ho s ob ained bene i s in e ms o elas ic ecupe a ion and
damaged a ea when co k powde was added o an epoxy esin, and highe pene a ion h eshold and
esidual ensile s eng h o he lamina es ha inco po a e he co k powde . Fo example, he elas ic
ecupe a ion inc eased by abou 26%, o a 21 J impac ene gy, and he damaged a ea dec eased by
a ound 15% o he same impac ene gy and compa ed o he alues ob ained o lamina es wi h nea
esin. Rega ding he pene a ion h eshold, hese alues we e abou 15% while he esidual s eng h
dec eased a ound 47% o lamina es subjec ed o an impac ene gy o 21 J [33].
In e ms o a igue p ope ies, independen ly o he lowe s a ic bending s eng h (a ound 10%
lowe when 3 w .% o co k powde was added o he esin), Reis e al. [
34
] obse ed, based on a
s a is ical analysis o a igue da a, ha a igue li e is insensi i e o he p esence o co k in o he esin.
Appl. Sci. 2020,10, 6108 3 o 12
Finally, due o he iscoelas ic p ope ies o co k and polyme ic ma ices, Reis e al. [
35
] pe o med a
s udy whe e he s ess elaxa ion and c eep beha iou o mul iphase composi es inco po a ing co k
powde we e analysed in de ail. Fo his pu pose, composi es wi h he same lay-up bu wi h di e en
ib es we e p oduced. They concluded ha highe alues o elaxa ion and c eep displacemen occu
when Ke la ib es a e used, bu , independen ly o he ib es, highe alues we e also obse ed when
co k powde was in oduced in o he esin.
The e o e, based on he bene i s achie ed wi h his indus ial by-p oduc , he main goal o his
wo k was o s udy he high- eloci y impac beha iou o Ke la composi e lamina es illed by co k
powde , in o de o ob ain ull knowledge o all loading modes. These ib es a e o pa icula in e es
o mili a y and ci ilian pu poses, due o hei high deg ee o oughness and good impac /ballis ic
pe o mance. In ac , a amid ib es ail by a se ies o small ib il ailu es ha abso b a signi ican
amoun o ene gy, esul ing in e y high oughness. Fo ha pu pose, his wo k ocused on he
ballis ic beha iou and damage ole ance o Ke la /epoxy ein o ced composi es, in which he epoxy
ma ix is illed wi h co k powde . The pe o a ion eloci y and ene gy abso p ion we e es ima ed om
expe imen al es s and damaged a ea was ob ained by bo h C-scan ul asound echnique and isual
inspec ion om impac ed specimens. In o de o e alua e he damage ole ance, bending a e impac
(BAI) es s we e ca ied ou .
2. Expe imen al P ocedu e
Composi e lamina e pla es in ol ing Ke la bi-di ec ional wo en ab ics ( a e a wi h 281 g/cm
2
)
and an Amp eg 22 epoxy esin wi h an Amp eg 22 ha dene s anda d (bo h supplied by Gu i ) we e
manu ac u ed wi h a use ul size o 300
×
300
×
3.3 mm
3
by hand lay-up echnique. Fo his pu pose,
nine ply lamina es, all in he same di ec ion, we e used. A e ha , he sys em was placed inside
a acuum bag and a load o 2.5 kN was applied o 48 h o ensu e a uni o m lamina e hickness.
Du ing he i s 10 h he bag emained a ached o a acuum pump o elimina e possible ai bubbles in
he composi e lamina es. Finally, he pos -cu e was ca ied ou in an o en a 45
◦
C o 48 h o gua an ee
a con enien cu ing s age.
Simila composi e lamina es wi h epoxy ma ix illed by co k powde we e also p oduced by he
same me hodology. The co k powde used consis s o a esidue ob ained du ing he inal sanding
phase o co k-s oppe s and i was collec ed a Amo im Co k S.A. (Valada, Po ugal) indus ial acili ies.
The co k powde p esen s a pa icle size, in e ms o pe cen ile, o d (0.1) =18.6
µ
m, d (0.5) =78.9
µ
m
and d (0.9) =208.3
µ
m; a bulk densi y o 0.1095 g.cm
−3
; and a humidi y o
≈
5.1%. The low mois u e
con en obse ed is explained by he ac ha he aw ma e ial om which he co k powde was
ob ained unde wen se e al hea ing and d ying ope a ions [
23
]. Rega dless o he low mois u e
con en obse ed, o imp o e he adhesion wi h he esin, he co k powde was s ill d ied in an o en
(He aus, model UT 6060) a 120
◦
C o 2 h and s o ed un il used in a desicca o . Finally, he co k
powde (3 w .% o he epoxy esin-ha dene mix u e) and epoxy esin we e mixed a 900 pm o
2 h, wi h esou ce o an elec ical mixe , in an ul asonic ba h o an uni o m dis ibu ion o he co k
pa icles in o he esin. In o de o a oid he o ma ion o ai bubbles, he mix u e was degassed in a
acuum o en, ollowed by addi ion o ha dene agen . Howe e , mo e de ails abou he co k ac ion
used and espec i e manu ac u ing p ocess can be ound in [32–34].
Acco ding o Pin o e al. [
24
], co k’s s uc u e is o med by hollow polyhed al p isma ic cells,
which ha e a honeycomb shape when obse ed om he adial di ec ion and ec angula when
obse ed om he ans e sal di ec ion [
9
]. The a ea o he p ism base is 4 o 6
×
10
−6
cm
2
, wi h a mean
p ism edge o 13–15
µ
m and a p im heigh o 30–40
µ
m. The mean cell olume is a ound 2
×
10
−8
cm
3
and he numbe o cells pe uni is 4 o 7
×
10
7
cm
−3
. The cell walls a e hin, wi h a hickness o 1 o
1.5
µ
m, in which is concen a ed he solid mass olume ac ion o he co k [
36
]. The co k powde
keeps he same cellula s uc u e in ac ; howe e , when he size o co k pa icles deceases, he numbe
o closed cells dec eases and he ex e nal su ace a eas o he pa icles inc ease, and consequen ly,
he numbe o open h ough-cu cells inc eases [
36
]. Based in s udies epo ed by Pe ei a [
36
], a co k
Appl. Sci. 2020,10, 6108 4 o 12
pa icle o olume 0.015 mm
3
(a ound 0.25
×
0.25
×
0.25 mm
3
) con ains abou 500 cells (7–9 cell pe
one ow) in which only a ac ion o 6 o 8 cells in one ow a e closed. This e idence and espec i e
pa icle shape and mo phology a e well epo ed by scanning elec on mic og aphs, in a e iew pape
compiled by Pe ei a [36], o co k pa icles simila o hose used in his s udy.
The high- eloci y impac s we e ca ied ou on pla es o Ke la bi-di ec ional wo en lamina es
wi h su ace o 100
×
100 mm
2
. The specimen’s size was big enough ha he damaged a ea did no
each he bo de o he lamina e. The p ojec ile used o impac es s was o s eel wi h a sphe ical
geome y o 7.5 mm diame e and a densi y o 7800 kg/m
3
. In o de o ca y ou he impac es s, a gas
gun (Figu e 1) was used o impulse he p ojec ile a he equi ed eloci y. This gas gun consis s o
se e al clea ly di e en ia ed zones: he load and accommoda ion a ea o he p ojec ile, he p essu e
chambe ha impulses he p ojec ile agains he a ge , a launch ube and a ame whe e he lamina e
is embedded.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 4 o 13
pa icle shape and mo phology a e well epo ed by scanning elec on mic og aphs, in a e iew
pape compiled by Pe ei a [36], o co k pa icles simila o hose used in his s udy.
The high- eloci y impac s we e ca ied ou on pla es o Ke la bi-di ec ional wo en lamina es
wi h su ace o 100 × 100 mm2. The specimen’s size was big enough ha he damaged a ea did no
each he bo de o he lamina e. The p ojec ile used o impac es s was o s eel wi h a sphe ical
geome y o 7.5 mm diame e and a densi y o 7800 kg/m3. In o de o ca y ou he impac es s, a
gas gun (Figu e 1) was used o impulse he p ojec ile a he equi ed eloci y. This gas gun consis s
o se e al clea ly di e en ia ed zones: he load and accommoda ion a ea o he p ojec ile, he p essu e
chambe ha impulses he p ojec ile agains he a ge , a launch ube and a ame whe e he lamina e
is embedded.
Figu e 1. Pho o o he equipmen used in he high- eloci y impac es s.
To impulse he p ojec ile, he p essu e o he gas used was egula ed and wo di e en ypes o
gas we e used: helium o achie e he highes eloci ies, and S a gon® a mix u e o a gon, ca bon
dioxide and oxygen manu ac u ed by NIPPON GASES ESPAÑA, S.L.U. (Mad id, Spain) o he
lowes eloci ies.
The impac es s we e eco ded wi h a high- eloci y came a (Figu e 2), which was used o pick
up essen ial in o ma ion such as he impac and esidual eloci ies. The came a used was a Pho on
Fas cam ATX, which is able o each 250,000 ames pe second. In hese es s he acquisi ion da a
eloci y was 90,000 ps. The came a was placed pa allel o he ame o essays. An ARRI ARRISON
12 PLUS lamp was used o ligh ing. This lamp has a powe o 1200 W.
A e he impac es s, non-des uc i e es ing (NDT) was ca ied ou on all impac ed specimens,
in o de o e alua e he damage ex ension by ul asonic inspec ion, speci ically C-scan. Fo his
pu pose, he ai -coupled scan echnique was used, because i is a good al e na i e o he adi ional
ul asonic imme sion C-scan echnique when he de ec ion o small de ec s (ma ix c acking,
ib e/ma ix in e ace debonding, e c.) is no equi ed, as in he p esen s udy [37,38]. The e o e, a
pai o ansduce s wi h 400 kHz cen e equency, 38 mm ocus and 25 mm ac i e diame e we e
used in a h ough- ansmission con igu a ion.
Figu e 1. Pho o o he equipmen used in he high- eloci y impac es s.
To impulse he p ojec ile, he p essu e o he gas used was egula ed and wo di e en ypes o gas
we e used: helium oachie e he highes eloci ies, andS a gon
®
amix u eo a gon, ca bon dioxideand
oxygen manu ac u ed by NIPPON GASES ESPAÑA, S.L.U. (Mad id, Spain) o he lowes eloci ies.
The impac es s we e eco ded wi h a high- eloci y came a (Figu e 2), which was used o pick
up essen ial in o ma ion such as he impac and esidual eloci ies. The came a used was a Pho on
Fas cam ATX, which is able o each 250,000 ames pe second. In hese es s he acquisi ion da a
eloci y was 90,000 ps. The came a was placed pa allel o he ame o essays. An ARRI ARRISON
12 PLUS lamp was used o ligh ing. This lamp has a powe o 1200 W.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 5 o 13
Figu e 2. Pho o o he high- eloci y impac es se up.
Finally, esidual s eng h was ob ained by h ee-poin bending (3PB) s a ic es s pe o med wi h
a spam o 80 mm in a Shimadzu AG-10 (Ri e wood D i e Columbia, SC, USA) uni e sal es ing
machine equipped wi h a 5 kN load cell, s ain a e o 5 mm/s and a oom empe a u e.
3. Resul s and Discussion
Figu e 3 shows he abso bed ene gy ega ding impac ene gy o all he specimens. The a ia ion
o he abso bed ene gy wi h impac ene gy is e y simila o bo h ype o pla es. Below pe o a ion,
he pla es abso b all he kine ic ene gy o he p ojec ile, mainly by damage. When he impac ene gy
inc eases abo e pe o a ion, he abso bed ene gy dec eases asymp o ically.
I was no possible o calcula e he minimum pe o a ion eloci y in a de e minis manne , as an
impac - eloci y in e al exis s in which he s uc u e may o may no be en i ely pe o a ed. In
addi ion, he esidual eloci y o he p ojec ile om he gas gun canno be ully con olled. Fo his
eason, a high le el o sca e was ound in he da a ob ained om he expe imen al es s a ound he
minimum pe o a ion eloci y. To ob ain his eloci y, a leas -squa es me hod was used, by i ing
Equa ion (1) which is simila o ha p oposed in he model o Rech and Ipson [39]. This me hod has
been used success ully o p edic he minimum pe o a ion eloci y in lamina e and sandwich
s uc u es subjec ed o high eloci y impac s [30,40–42].
𝑣𝑣𝑟𝑟=�0 𝑖𝑖𝑖𝑖 0 ≤𝑣𝑣𝑖𝑖<𝑣𝑣𝑏𝑏𝑏𝑏
𝑣𝑣𝑟𝑟=𝐴𝐴∙�𝑣𝑣𝑖𝑖𝑝𝑝−𝑣𝑣𝑏𝑏𝑏𝑏
𝑝𝑝�
1
𝑝𝑝
� 𝑖𝑖𝑖𝑖 𝑣𝑣𝑖𝑖>𝑣𝑣𝑏𝑏𝑏𝑏
(1)
whe e: is he esidual eloci y, i he impac eloci y, bl he minimum eloci y o pe o a ion and
p and A a e i ing pa ame e s.
Figu e 2. Pho o o he high- eloci y impac es se up.
Appl. Sci. 2020,10, 6108 5 o 12
A e he impac es s, non-des uc i e es ing (NDT) was ca ied ou on all impac ed specimens,
in o de o e alua e he damage ex ension by ul asonic inspec ion, speci ically C-scan. Fo his pu pose,
he ai -coupled scan echnique was used, because i is a good al e na i e o he adi ional ul asonic
imme sion C-scan echnique when he de ec ion o small de ec s (ma ix c acking, ib e/ma ix
in e ace debonding, e c.) is no equi ed, as in he p esen s udy [
37
,
38
]. The e o e, a pai o
ansduce s wi h 400 kHz cen e equency, 38 mm ocus and 25 mm ac i e diame e we e used
in a h ough- ansmission con igu a ion.
Finally, esidual s eng h was ob ained by h ee-poin bending (3PB) s a ic es s pe o med wi h a
spam o 80 mm in a Shimadzu AG-10 (Ri e wood D i e Columbia, SC, USA) uni e sal es ing machine
equipped wi h a 5 kN load cell, s ain a e o 5 mm/s and a oom empe a u e.
3. Resul s and Discussion
Figu e 3shows he abso bed ene gy ega ding impac ene gy o all he specimens. The a ia ion
o he abso bed ene gy wi h impac ene gy is e y simila o bo h ype o pla es. Below pe o a ion,
he pla es abso b all he kine ic ene gy o he p ojec ile, mainly by damage. When he impac ene gy
inc eases abo e pe o a ion, he abso bed ene gy dec eases asymp o ically.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 6 o 13
Figu e 3. Impac ene gy s. abso bed eloci y.
The e o e, he minimum eloci y o pe o a ion is abou 206.35 ± 4.20 m/s o Ke la lamina es
wi h nea esin, while o ha wi h esin illed wi h co k powde is a ound 209.71 ± 4.63 m/s.
Rega dless o he small di e ence in e ms o a e age alues (1.6%), bo h alues a e simila because
hey a e wi hin he expe imen al dispe sion. In ac , co k is a hyd ophobic ma e ial wi h a su ace
ene gy o 18 mN.m−1, while he ypical cu ed epoxies ha e su ace ene gy o abou 45 mN.m−1 and a
poo adhesion occu s o subs a es wi h a su ace ene gy o 30 mN.m−1 o lowe . Howe e , his
weakness is coun e ed by he la ge ex e nal su aces o he pa icles [36]. Rega dless o he
imp o emen s ob ained in e ms o su ace ene gy using su ace ea men s [43,44], Ba bosa e al. [44]
epo ha some o hem des oy pa o he honeycomb s uc u e o he co k cells (e osion o co k
cell walls). In his case, he damage caused o he pa icles can acili a e he pene a ion o he esin
wi h he consequen dec ease on he abso p ion o ene gy, because he damping e ec o co k is los .
The e o e, while he bonds ha occu when he adhesi e di uses in o cell walls o build chemical
bonds wi h he cell’s chemical componen s a e weake , he pa icles’ mo phology p omo es he
physical bonds wi h he pene a ion o he epoxy esin in o he open cells and consequen mechanical
in e locking. In his con ex , he co k pa icles a e su ounded by esin, and he closed cells ha
con ain ai inside can de o m when hey su e an impac o ce and abso b ha impac [44]. Fo his
eason, small pa icles show less imp o emen han la ge pa icles, because he numbe o open
h ough-cu cells inc eases, and he esin does no allow he cell de o ma ion. Consequen ly, hey
canno abso b as much ene gy. The e o e, he esponse o he co k/ esin composi e is in luenced by
he numbe o closed cells in he pa icle [44].
In addi ion o he bene i ob ained in e ms o impac , i is unques ionable ha he co k powde
educes he amoun o esin in he composi e, making i mo e en i onmen ally iendly. These esul s
a e in good ag eemen wi h he open li e a u e, whe e Reis e al. [34] obse ed, o example, ha
a igue li e is insensi i e o he p esence o co k powde in o he esin. Simul aneously, hese au ho s
ob ained bene i s in e ms o impac s eng h and glass ansi ion empe a u e [32,33].
To unde s and hese esul s, he ailu e mechanisms a e shown in de ail and Figu es 4 and 5.
They show ypical images ob ained om isual inspec ion o he back side o he pla es. Figu e 4
shows he specimens o Ke la and Ke la wi h co k powde impac ed a simila eloci ies, close o
he pe o a ion eloci y, while he specimens o Figu e 5 we e impac ed a highe eloci ies.
Figu e 3. Impac ene gy s. abso bed eloci y.
I was no possible o calcula e he minimum pe o a ion eloci y in a de e minis manne , as an
impac - eloci y in e al exis s in which he s uc u e may o may no be en i ely pe o a ed. In addi ion,
he esidual eloci y o he p ojec ile om he gas gun canno be ully con olled. Fo his eason,
a high le el o sca e was ound in he da a ob ained om he expe imen al es s a ound he minimum
pe o a ion eloci y. To ob ain his eloci y, a leas -squa es me hod was used, by i ing Equa ion (1)
which is simila o ha p oposed in he model o Rech and Ipson [
39
]. This me hod has been used
success ully o p edic he minimum pe o a ion eloci y in lamina e and sandwich s uc u es subjec ed
o high eloci y impac s [30,40–42].
=






0i 0≤ i< bl
=A· p
i− p
bl1
pi i> bl
(1)
whe e:
is he esidual eloci y,
i
he impac eloci y,
bl
he minimum eloci y o pe o a ion and p
and Aa e i ing pa ame e s.
The e o e, he minimum eloci y o pe o a ion is abou 206.35
±
4.20 m/s o Ke la lamina es wi h
nea esin, while o ha wi h esin illed wi h co k powde is a ound 209.71
±
4.63 m/s. Rega dless o

Appl. Sci. 2020,10, 6108 6 o 12
he small di e ence in e ms o a e age alues (1.6%), bo h alues a e simila because hey a e wi hin
he expe imen al dispe sion. In ac , co k is a hyd ophobic ma e ial wi h a su ace ene gy o 18 mN.m
−1
,
while he ypical cu ed epoxies ha e su ace ene gy o abou 45 mN.m
−1
and a poo adhesion occu s
o subs a es wi h a su ace ene gy o 30 mN.m
−1
o lowe . Howe e , his weakness is coun e ed by
he la ge ex e nal su aces o he pa icles [
36
]. Rega dless o he imp o emen s ob ained in e ms o
su ace ene gy using su ace ea men s [
43
,
44
], Ba bosa e al. [
44
] epo ha some o hem des oy
pa o he honeycomb s uc u e o he co k cells (e osion o co k cell walls). In his case, he damage
caused o he pa icles can acili a e he pene a ion o he esin wi h he consequen dec ease on he
abso p ion o ene gy, because he damping e ec o co k is los . The e o e, while he bonds ha occu
when he adhesi e di uses in o cell walls o build chemical bonds wi h he cell’s chemical componen s
a e weake , he pa icles’ mo phology p omo es he physical bonds wi h he pene a ion o he epoxy
esin in o he open cells and consequen mechanical in e locking. In his con ex , he co k pa icles a e
su ounded by esin, and he closed cells ha con ain ai inside can de o m when hey su e an impac
o ce and abso b ha impac [
44
]. Fo his eason, small pa icles show less imp o emen han la ge
pa icles, because he numbe o open h ough-cu cells inc eases, and he esin does no allow he
cell de o ma ion. Consequen ly, hey canno abso b as much ene gy. The e o e, he esponse o he
co k/ esin composi e is in luenced by he numbe o closed cells in he pa icle [44]
In addi ion o he bene i ob ained in e ms o impac , i is unques ionable ha he co k powde
educes he amoun o esin in he composi e, making i mo e en i onmen ally iendly. These esul s
a e in good ag eemen wi h he open li e a u e, whe e Reis e al. [
34
] obse ed, o example, ha a igue
li e is insensi i e o he p esence o co k powde in o he esin. Simul aneously, hese au ho s ob ained
bene i s in e ms o impac s eng h and glass ansi ion empe a u e [32,33].
To unde s and hese esul s, he ailu e mechanisms a e shown in de ail and Figu es 4and 5.
They show ypical images ob ained om isual inspec ion o he back side o he pla es. Figu e 4
shows he specimens o Ke la and Ke la wi h co k powde impac ed a simila eloci ies, close o
he pe o a ion eloci y, while he specimens o Figu e 5we e impac ed a highe eloci ies.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 o 13
(a)
(b)
Figu e 4. Damage a ea o he back side. (a) Specimen o Ke la impac ed a 231 m/s and (b) specimen
o Ke la + co k powde impac ed a 230 m/s.
(a)
(b)
Figu e 5. Image o damage a ea o he back side. (a) Specimen o Ke la impac o 438 m/s and (b)
specimen o Ke la + co k powde impac ed o 439 m/s.
F om he isual inspec ions o he back sides o he pla es (Figu es 4 and 5), i is isible ha he
ailu e mechanisms a e: ib e ailu e and delamina ion. Al hough ma ix c acking is no possible o
iden i y by isual inspec ion, his is one o he i s damage mechanisms in composi e lamina es, and
he e o e, i should appea since i induces o he ailu es modes, as delamina ion [45].
The images o he damaged a ea ha we e ob ained by C-scan in he samples subjec ed o
di e en impac eloci ies a e shown in Figu es 6 and 7. The impac eloci ies coincide wi h he
impac eloci ies o he samples showed in Figu es 4 and 5. In all specimens inspec ed by C-scan, i
was possible o obse e ha he damaged a eas had ci cula shapes, a beha iou ha also appea s in
wo en glass lamina es [46].
Figu e 4.
Damage a ea o he back side. (
a
) Specimen o Ke la impac ed a 231 m/s and (
b
) specimen
o Ke la +co k powde impac ed a 230 m/s.
F om he isual inspec ions o he back sides o he pla es (Figu es 4and 5), i is isible ha he
ailu e mechanisms a e: ib e ailu e and delamina ion. Al hough ma ix c acking is no possible o
iden i y by isual inspec ion, his is one o he i s damage mechanisms in composi e lamina es, and
he e o e, i should appea since i induces o he ailu es modes, as delamina ion [45].
The images o he damaged a ea ha we e ob ained by C-scan in he samples subjec ed o
di e en impac eloci ies a e shown in Figu es 6and 7. The impac eloci ies coincide wi h he impac
eloci ies o he samples showed in Figu es 4and 5. In all specimens inspec ed by C-scan, i was
possible o obse e ha he damaged a eas had ci cula shapes, a beha iou ha also appea s in wo en
glass lamina es [46].
Appl. Sci. 2020,10, 6108 7 o 12
Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 o 13
(a)
(b)
Figu e 4. Damage a ea o he back side. (a) Specimen o Ke la impac ed a 231 m/s and (b) specimen
o Ke la + co k powde impac ed a 230 m/s.
(a)
(b)
Figu e 5. Image o damage a ea o he back side. (a) Specimen o Ke la impac o 438 m/s and (b)
specimen o Ke la + co k powde impac ed o 439 m/s.
F om he isual inspec ions o he back sides o he pla es (Figu es 4 and 5), i is isible ha he
ailu e mechanisms a e: ib e ailu e and delamina ion. Al hough ma ix c acking is no possible o
iden i y by isual inspec ion, his is one o he i s damage mechanisms in composi e lamina es, and
he e o e, i should appea since i induces o he ailu es modes, as delamina ion [45].
The images o he damaged a ea ha we e ob ained by C-scan in he samples subjec ed o
di e en impac eloci ies a e shown in Figu es 6 and 7. The impac eloci ies coincide wi h he
impac eloci ies o he samples showed in Figu es 4 and 5. In all specimens inspec ed by C-scan, i
was possible o obse e ha he damaged a eas had ci cula shapes, a beha iou ha also appea s in
wo en glass lamina es [46].
Figu e 5.
Image o damage a ea o he back side. (
a
) Specimen o Ke la impac o 438 m/s and (
b
)
specimen o Ke la +co k powde impac ed o 439 m/s.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 8 o 13
(a)
(b)
Figu e 6. Images o damage a eas by C-scan. (a) Specimen o Ke la impac o 231 m/s and (b)
specimen o Ke la + co k powde impac ed o 230 m/s.
(a)
(b)
Figu e 7. Images o damage a eas by C-scan. (a) Specimen o Ke la impac o 438 m/s and (b)
specimen o Ke la + co k powde impac ed o 439 m/s.
The specimens impac ed a highe eloci ies, a ound 300 m/s, showed ib e b eakage close o he
impac poin , as epo ed in Figu es 5 and 7. Simul aneously, he C-scan images clea ly show ha he
damage caused by he bulle s no only causes local damage (mac o damage), bu also a ec s he
icini y o he damaged a eas clea ly seen in di e en colou s (mic o damage), which ag ees wi h he
s udies de eloped by Dhakal e al. [47]. Mo eo e , he ex ension o damage is mo e loca ed a ound
he impac poin a highe eloci ies, as can be seen in Figu e 7. In his con ex , he ex ension o
damage a ea de e mined by he C-scan inspec ion is la ge han he damage es ablished by he isual
inspec ion on bo h sides o he lamina e ( on and back ace).The e o e, in addi ion o mac o-damage,
2918.31 mm²2571.03 mm²
1235.87 mm²629.81 mm²
41.52 mm²
Figu e 6.
Images o damage a eas by C-scan. (
a
) Specimen o Ke la impac o 231 m/s and (
b
) specimen
o Ke la +co k powde impac ed o 230 m/s.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 8 o 13
(a)
(b)
Figu e 6. Images o damage a eas by C-scan. (a) Specimen o Ke la impac o 231 m/s and (b)
specimen o Ke la + co k powde impac ed o 230 m/s.
(a)
(b)
Figu e 7. Images o damage a eas by C-scan. (a) Specimen o Ke la impac o 438 m/s and (b)
specimen o Ke la + co k powde impac ed o 439 m/s.
The specimens impac ed a highe eloci ies, a ound 300 m/s, showed ib e b eakage close o he
impac poin , as epo ed in Figu es 5 and 7. Simul aneously, he C-scan images clea ly show ha he
damage caused by he bulle s no only causes local damage (mac o damage), bu also a ec s he
icini y o he damaged a eas clea ly seen in di e en colou s (mic o damage), which ag ees wi h he
s udies de eloped by Dhakal e al. [47]. Mo eo e , he ex ension o damage is mo e loca ed a ound
he impac poin a highe eloci ies, as can be seen in Figu e 7. In his con ex , he ex ension o
damage a ea de e mined by he C-scan inspec ion is la ge han he damage es ablished by he isual
inspec ion on bo h sides o he lamina e ( on and back ace).The e o e, in addi ion o mac o-damage,
2918.31 mm²2571.03 mm²
1235.87 mm²629.81 mm²
41.52 mm²
Figu e 7.
Images o damage a eas by C-scan. (
a
) Specimen o Ke la impac o 438 m/s and (
b
) specimen
o Ke la +co k powde impac ed o 439 m/s.
Appl. Sci. 2020,10, 6108 8 o 12
The specimens impac ed a highe eloci ies, a ound 300 m/s, showed ib e b eakage close o he
impac poin , as epo ed in Figu es 5and 7. Simul aneously, he C-scan images clea ly show ha
he damage caused by he bulle s no only causes local damage (mac o damage), bu also a ec s he
icini y o he damaged a eas clea ly seen in di e en colou s (mic o damage), which ag ees wi h he
s udies de eloped by Dhakal e al. [
47
]. Mo eo e , he ex ension o damage is mo e loca ed a ound he
impac poin a highe eloci ies, as can be seen in Figu e 7. In his con ex , he ex ension o damage
a ea de e mined by he C-scan inspec ion is la ge han he damage es ablished by he isual inspec ion
on bo h sides o he lamina e ( on and back ace).The e o e, in addi ion o mac o-damage, i is e iden
ha he impac also induces in e nal mic o-damage ha is no isible o he naked eye, bu should be
conside ed when quan i ying he damage size [47].
Figu e 8shows he ex ension o damage a ea o all specimens measu ed by C-scan. In all
s uc u es, he la ge damaged a ea was ound o eloci ies a ound he minimum pe o a ion eloci y.
Fo eloci ies below he minimum pe o a ion eloci y, he damaged a ea inc eases wi h eloci y, while
o eloci ies g ea e han he pe o a ion eloci y, he damaged a ea dec eases wi h inc easing eloci y.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 9 o 13
i is e iden ha he impac also induces in e nal mic o-damage ha is no isible o he naked eye,
bu should be conside ed when quan i ying he damage size [47].
Figu e 8 shows he ex ension o damage a ea o all specimens measu ed by C-scan. In all
s uc u es, he la ge damaged a ea was ound o eloci ies a ound he minimum pe o a ion
eloci y. Fo eloci ies below he minimum pe o a ion eloci y, he damaged a ea inc eases wi h
eloci y, while o eloci ies g ea e han he pe o a ion eloci y, he damaged a ea dec eases wi h
inc easing eloci y.
Figu e 8. Ex ension o damage a ea measu ed by C-scan.
This beha iou was obse ed by Bui ago e al. [48] o glass wo en lamina es, and i is p oduced
by he bending o he pla e due o he ans e se wa e a el, which leads o delamina ion. Since he
con ac ime is maximal a he pe o a ion eloci y, he dis ance a elled by he ans e se wa e is
also maximal, and hus he delamina ed a ea is oo [46,49]. On he o he hand, when he damaged
a eas a e compa ed be ween lamina es wi h nea esin and esin illed by co k powde , in all
condi ions he e is a clea endency o he las ones o ha e smalle damaged a eas. Fo example,
compa ing he damaged a eas shown in Figu es 6 and 7, i is possible o obse e ha o impac
eloci ies a ound 230 m/s, hey a e a ound 11.9% lowe o lamina es wi h co k powde . Howe e ,
his alue was a ound 45.7% lowe o impac eloci ies o 438 m/s. Ne e heless, o he highes
impac eloci ies, he damaged a eas end o be simila . This beha iou allows one o a i m ha he e
a e changes in he ene gy abso p ion mechanisms, al hough he e is no much change in he ballis ic
beha iou (Figu e 3). In ac , co k has a high capaci y o abso b ene gy, and as epo ed abo e, he
esponse o he co k/ esin composi e is in luenced by he numbe o closed cells in he co k pa icles.
Sanchez-Saez e al. [29] obse ed a g ea capaci y o abso b ene gy e en a e se e al consecu i e
impac s. In addi ion, Sa asini e al. [30] obse ed in hei s udies ha he abso p ion o ene gy in high
de o ma ions can be qui e conside able wi hou signs o b eaking o he cell walls, because an almos
comple e eco e y o hei o iginal shape and size occu s.
Finally, Figu e 9 shows he esidual bending s eng h, whe e σ is he a e age bending s ess o
each eloci y and σ0 is he a e age bending s ess ob ained om he 3PB s a ic es s o each lamina e.
Figu e 8. Ex ension o damage a ea measu ed by C-scan.
This beha iou was obse ed by Bui ago e al. [
48
] o glass wo en lamina es, and i is p oduced
by he bending o he pla e due o he ans e se wa e a el, which leads o delamina ion. Since he
con ac ime is maximal a he pe o a ion eloci y, he dis ance a elled by he ans e se wa e is also
maximal, and hus he delamina ed a ea is oo [
46
,
49
]. On he o he hand, when he damaged a eas a e
compa ed be ween lamina es wi h nea esin and esin illed by co k powde , in all condi ions he e is a
clea endency o he las ones o ha e smalle damaged a eas. Fo example, compa ing he damaged
a eas shown in Figu es 6and 7, i is possible o obse e ha o impac eloci ies a ound 230 m/s, hey
a e a ound 11.9% lowe o lamina es wi h co k powde . Howe e , his alue was a ound 45.7% lowe
o impac eloci ies o 438 m/s. Ne e heless, o he highes impac eloci ies, he damaged a eas
end o be simila . This beha iou allows one o a i m ha he e a e changes in he ene gy abso p ion
mechanisms, al hough he e is no much change in he ballis ic beha iou (Figu e 3). In ac , co k has
a high capaci y o abso b ene gy, and as epo ed abo e, he esponse o he co k/ esin composi e
is in luenced by he numbe o closed cells in he co k pa icles. Sanchez-Saez e al. [
29
] obse ed a
g ea capaci y o abso b ene gy e en a e se e al consecu i e impac s. In addi ion, Sa asini e al. [
30
]
obse ed in hei s udies ha he abso p ion o ene gy in high de o ma ions can be qui e conside able
wi hou signs o b eaking o he cell walls, because an almos comple e eco e y o hei o iginal shape
and size occu s.
Appl. Sci. 2020,10, 6108 9 o 12
Finally, Figu e 9shows he esidual bending s eng h, whe e
σ
is he a e age bending s ess o
each eloci y and
σ0
is he a e age bending s ess ob ained om he 3PB s a ic es s o each lamina e.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 10 o 13
Figu e 9. Impac eloci y s. a io o bending s ess.
Rega dless o he lowe s a ic bending s eng h (a ound 13.7% less when 3 w .% o co k powde
was added o he esin), which ag ees wi h he s udies de eloped by Reis e al. [34], i is possible o
obse e ha lamina es inco po a ing co k powde a e less sensi i e o impac eloci y han he
samples wi h nea esin. This is explained by he lowe damaged a eas ha we e p e iously epo ed
o lamina es wi h co k powde , due o he high capaci y o abso b ene gy om he ille s. In addi ion
o ha , Ba bosa e al. [44] obse ed ha co k pa icles ac like obs acles o c ack p opaga ion. This
e idence also explains he be e bending pe o mance ob ained o lamina es wi h co k powde . On
he o he hand, and independen ly o he lamina e, i is also e iden ha he maximum bending
s eng h dec eases o eloci ies below he minimum pe o a ion eloci y, and ha he e is a endency
o i o emain cons an o eloci ies g ea e han he pe o a ion eloci y. Howe e , he obse ed
dec ease was mo e exp essi e o lamina es wi h nea esin due o he g ea e damage a eas obse ed
(as epo ed abo e).
4. Conclusions
The high- eloci y impac and pos impac beha iou s o Ke la /epoxy composi es, in which he
epoxy ma ix is illed wi h co k powde , ha e been s udied.
In e ms o a e age alues, he minimum eloci y o pe o a ion was 1.6% highe o lamina es
wi h co k powde , bu his di e ence can be neglec ed in iew o he dispe sion ob ained in he
expe imen al es s. The damage a eas inc ease o eloci ies below he minimum pe o a ion eloci y,
bu he opposi e end occu s o eloci ies g ea e han he pe o a ion eloci y. Howe e , ega dless
he impac eloci y, he C-scan images e eal ha lesse damage a eas a e ob ained o lamina es
wi h co k powde , especially o impac eloci ies close o pe o a ion eloci y. Consequen ly,
conside ing he esidual bending s eng h, hese lamina es a e less sensi i e o impac eloci y han
he samples wi h nea esin. In addi ion o hese bene i s, co k powde educes he amoun o esin
in he composi e, making i mo e en i onmen ally iendly. Finally, in addi ion o hese bene i s,
lamina es wi h co k powde a e mo e en i onmen ally iendly due o hei lowe pe cen ages by
weigh o esin.
Au ho Con ibu ions: A.M.A. pe o med he NDT es s, analysed he esul s and helped o w i e he
manusc ip ; P.N.B.R. p oduced he lamina es and helped o w i e he manusc ip . I.I., S.S.-S., S.K.G.-C. and E.B.
Figu e 9. Impac eloci y s. a io o bending s ess.
Rega dless o he lowe s a ic bending s eng h (a ound 13.7% less when 3 w .% o co k powde was
added o he esin), which ag ees wi h he s udies de eloped by Reis e al. [
34
], i is possible o obse e
ha lamina es inco po a ing co k powde a e less sensi i e o impac eloci y han he samples wi h
nea
esin. This
is explained by he lowe damaged a eas ha we e p e iously epo ed o lamina es
wi h co k powde , due o he high capaci y o abso b ene gy om he ille s. In addi ion o ha ,
Ba bosa e al. [
44
] obse ed ha co k pa icles ac like obs acles o c ack p opaga ion. This e idence also
explains he be e bending pe o mance ob ained o lamina es wi h co k powde . On he o he hand,
and independen ly o he lamina e, i is also e iden ha he maximum bending s eng h dec eases
o eloci ies below he minimum pe o a ion eloci y, and ha he e is a endency o i o emain
cons an o eloci ies g ea e han he pe o a ion eloci y. Howe e , he obse ed dec ease was mo e
exp essi e o lamina es wi h nea esin due o he g ea e damage a eas obse ed (as epo ed abo e).
4. Conclusions
The high- eloci y impac and pos impac beha iou s o Ke la /epoxy composi es, in which he
epoxy ma ix is illed wi h co k powde , ha e been s udied.
In e ms o a e age alues, he minimum eloci y o pe o a ion was 1.6% highe o lamina es
wi h co k powde , bu his di e ence can be neglec ed in iew o he dispe sion ob ained in he
expe imen al es s. The damage a eas inc ease o eloci ies below he minimum pe o a ion eloci y,
bu he opposi e end occu s o eloci ies g ea e han he pe o a ion eloci y. Howe e , ega dless he
impac eloci y, he C-scan images e eal ha lesse damage a eas a e ob ained o lamina es wi h co k
powde , especially o impac eloci ies close o pe o a ion eloci y. Consequen ly, conside ing he
esidual bending s eng h, hese lamina es a e less sensi i e o impac eloci y han he samples wi h
nea esin. In addi ion o hese bene i s, co k powde educes he amoun o esin in he composi e,
making i mo e en i onmen ally iendly. Finally, in addi ion o hese bene i s, lamina es wi h co k
powde a e mo e en i onmen ally iendly due o hei lowe pe cen ages by weigh o esin.