Modern Applied Science; Vol. 15, No. 1; 2021 ISSN 1913-1844 E-ISSN 1913-1852

Published by Canadian Center of Science and Education

32

Identification of 'Defects' in Epoxy Matrix Composites and Carbon Fiber by Ultrasound. Interaction of the Ultrasonic Beam with the

Material Gaston Sanglier1, Eduardo J. Lopez1, Sonia Cesteros1 & Roberto A. Gonzalez1

1 Architecture and Design Department, Engineering Area, Escuela Politécnica Superior, Universidad CEU San Pablo, Montepríncipe Campus, 28668, Boadilla del Monte, Madrid, Spain Correspondence: Gaston Sanglier, Engineering Area, Escuela Politécnica Superior, Universidad CEU San Pablo, Montepríncipe Campus, 28668, Boadilla del Monte, Madrid, Spain. Received: September 2, 2020 Accepted: October 28, 2020 Online Published: December 22, 2020 doi:10.5539/mas.v15n1p32 URL: https://doi.org/10.5539/mas.v15n1p32 Abstract Non-Destructive Testing (NDT) methods, and especially, ultrasounds have gone from being a mere laboratory curiosity to an indispensable tool in the industry as a primary means of determining the level of quality achieved in its products (ASM, 1989; Barbero, 1999). This study will identify and apply the main physical phenomena of interaction of an ultrasonic wave in a composite material, to see if through this type of waves, you can detect defects of the type of porosity or delaminations in these materials. The percentages of reflected and transmitted waves in real cases of defects in the composite material will also be studied. It will be shown if the frequencies and intensities of the waves are adequate to find this type of defects or imperfections in the material. The theoretical study of the ultrasonic wave seeks to help researchers in the development of equipment that uses the methodology of immersion ultrasound for the inspection of materials in the search for 'defects' and to understand the physics of the test. Keywords: ultrasound, carbon fiber, epoxy matrix, porosity, delamination, NDT techniques 1. Introduction In the last decades the operating requirements have increased, while at the same time trying to reduce the weight of the structures and mechanical components used for industrial purposes. This has led to the need to use advanced materials that have high mechanical properties along with a decrease in weight (Inasmet, 1998). The use of fiber reinforced polymeric matrix composites has replaced in many cases conventional materials (steel, plastic, aluminum, concrete, etc.). These materials have been used preferably in the aeronautical and space industry for the manufacture of floor panels, spoilers, rudders, depth rudders, interior and exterior ailerons, space shuttles, satellites, etc. where they have had a greater development because they are structures that require high values of resistance and specific rigidity, and in which the weight factor entails a great decrease in the cost (Sanglier et al., 2003; Ramírez & Col, 1982; Wróbel & Wierzbicki, 2005; Ramírez López et al., 1996). Fiber-reinforced composite materials are also used in other sectors such as construction, wind turbines, marine, elite sports, automotive and other sectors such as medical and military (ImieliÉska et al., 2004; Rojek, Stabik, & Wróbel, 2005; Ochelski, 2004). These composite materials are characterized by high rigidity and mechanical resistance, high fatigue resistance, corrosion resistance, low weight and the possibility of selecting the appropriate orientation of the sheets for each specific application. Their low thermal conductivity and high dimensional stability give these materials a very interesting alternative in applications subject to low temperature conditions. The structural components of this type are made up of plate and/or beam type elements, which are subjected to loads perpendicular to their plane that originate a state of work in bending in which tensile and compressive stresses appear (Padmanabhan & Kishore, 1995; Potel et al., 1998; Scarponi & Briotti, 2000; Vaccaro & Akers, 1996). For example, these tensile states can appear in the aerodynamic loads exerted on the wings of an aircraft.

mas.ccsen

Howeverdefects unthat can chas been resistanceFor all thtechniqueinteractioreflected 2. MethoTo carry and a thirresistancemanufactcured laywide. Sixteen la(test piecin the foufrom the Below arcorresponspecimenAttached three typeinformatidifferenceobservatitype is intamplitude

net.org

r, another aspenderstood as

cause the failuobserved expe to compressihese reasons, e, in this case on of a wave and transmitt

ods out the study

rd with delame carbon fibertured in an aut

yer of 0.13 mm

ayers were usee) of 3mm aft

urth layer (at 0surface, in thee some imagends to the refern with medium

are some imaes of panels anion to capturees in attenuaton conditionstrinsically insee to oscillate.

ect of great imporosity, dela

ure of the mateerimentally thion (Cantwell

it is necessathe ultrasoundwith the comed wave in a d

, several paneminations) of cor AS4 preimprtoclave, curingm with a fiber w

ed, overlappedter the curing p0.75 mm frome central part oes of the specimrence specime

m porosity, andages taken by nalyzed. The ie it in a singltion or travel s (Huang et alensitive and thHowever, des

Figure 1.

Mod

mportance andamination or aerial in case ofhat the delaminet al., 1986; G

ary to know hds, those poss

mposite materidetermined im

els have been omposite materegnated in theg pressure 6 baweight per are

d in a unidirecphase. Each la

m the surface) of the test piecmens used in en (well made d finally, the sultrasound us

image is a verle glance. It mtime, and dig

l., 1998; Mourhe signal detecspite the low s

Different typ

dern Applied Sc

33

d where this staccumulation f receiving impnations produ

Gray et al., 199how to ident

sible defects atial and to be

mperfection or

manufacturederial based one form of unidars, volumetricea of 145 gr/m

ctional way (dayer has a thicand the delamce). the test. The swith no defec

specimen P2 osing total immry powerful inmakes it possgital support turitz, Townsedcted is usuallysensitivity the

pes of specime

cience

tudy is going of material (f

pact loads on tuced by this ty95; Baker, Jonify, using somttending to theable to calcu

r defect.

d (one of refern a modified edirectional tapc fiber content

m2. The size of

direction of theckness of 0.18

minations were

specimen withcts), the specimon the left cor

mersion transmnstrument that sible to easilyechniques allo

d, & Shah Khy quite affectedre is a high de

ens in the ultra

to be centeredfiber cuts, resithe structure o

ype of impact cnes, & Calliname NDT (Noe study of the ulate and inter

rence, anotherepoxy matrix 8e with a curint of 62% and nf the panels wa

e fibre 0º) with75 mm, then t

e made in the

h the letter R imen P1 on the responds to th

mission with 1allows to inte

y compare adjow the technian, 2000; Kind by noise or iefect capacity.

asonic test

Vol. 15, N

d is the identifin, fiber break

of composite mconsiderably r

an, 1985; Bishon-Destructivephysical phen

erpret the perc

r with medium8552 togetherng temperaturenominal thicknas 34 cm high

h a final panelthe porosity weighth layer (

in the lower cright correspo

he delaminatio1 MHz frequeegrate a great jacent areas wician to easilyno, 1979). A tinstability that.

No. 1; 2021

fication of kage, etc.) material. It reduce the

hop, 1985). e Testing) nomena of centage of

m porority r with high e of 180ºC ness of the

h by 29 cm

l thickness was caused (at 1.5 mm

entral part onds to the ons. ncy of the amount of

with small y optimize test of this t causes its

mas.ccsen

Attached specimen

Figure 2.

Figure 2 defects) ofigure 2 (rwith medin red ma

In figure defects hdelaminasizes (3x3test piece

net.org

are some imans.

Reference tes

(left) shows aof the materialright) it is obs

dium porosity harks this circum

3, an ultrasonhave been geations caused b3mm, 5x5mme.

ages taken by u

st tube on the l

an ultrasonic il under study. served an ultrahave been genmstance, in ad

Figure 3. nic inspectionenerated. A kby Teflon bag

m, 7.5x7.5 mm,

Mod

ultrasound usi

left (A) and P1reinfor

inspection usiIt is observed

asonic inspectinerated in the ddition to the u

P2 test tube wn using immerkind of dark gs, application, 9x5 mm 10x

dern Applied Sc

34

ing the 1 MHz

1 test tube withrced with carbing immersionthe unidirectiion using immfourth layer oupper central

with delaminatrsion transmissquares clear

n of release agx10 mm) place

cience

z total immers

h medium poron fiber n transmissionionality of the

mersion transmof the test tube

zone.

tion zones in tssion of the P2rly aligned in

gents, resin stied between lay

sion transmiss

rosity on the ri

n of the referefiber at 0º alon

mission of the tor sample. Es

the sample 2 specimen isn four rows cking, fiber cu

yers have been

Vol. 15, N

sion of the thre

ight (B) of epo

ence specimenong the whole test tube P1 wspecially the l

s shown whereappear. Thesuttings, etc. o

n placed in lay

No. 1; 2021

ee types of

oxy matrix

n (without sample. In

where areas ower zone

e different se defects, f different

yer 8 of the

mas.ccsen

2.1 TheorUltrasounNesvijskilayer) in tThe stabiabsence obetween sbe done a

FigIt will be will occuwave we between 2microscopmoleculeabsorptioal., 1996;Next, som1987; KaWave freall particlAcoustic zone. Thifrequencyliquids anthis influea very use

where E iand axial Acoustic The resisimpedancvibration small chathe wavepressure calculated

net.org

retical Foundand measuremei, 2000; Dunegthe coupling flility of the ouof turbulencestylus nozzle aautomatically

gure 4. Left: wtaken into ac

ur that will be refer to frequ

20 and 100 KHpy. These ults capable of

on and dispersi; Hervfeld & Lme important apranos & Priequency (f): is les and is idenvelocity (C): is velocity is y and any wand gases, on prence can be neeful condition

is the moduludeformationsimpedance (Z

stance that thce, impedanceof the wave.

ange in sound p. If, on the cois high, sinced through the

ations and Coents have beengan & Tetelm

fluid, which is utput signal is. The design and sample is (Maskov et al

water tank immcount that whnecessary to

uencies higherH), band fromtrasonic wavevibrating. Hoion phenomenLitovitis, 1959parameters w

estner, 1987; Hthe number o

ntical to the freis the propagaa characteris

avelength. Horessure, tempeeglected. For t

n for ultrasonic

us of elasticity s and ρ is the dZ): not all mate materials oe of the acousIf a medium hpressure; that

ontrary, the ime the mediumformula:

Mod

oncepts n made by imman, 1971), in talmost alway

s directly relatof the probe 20 mm. Usingl., 2000; Mars

mersion systemhen an ultrason

know for a br than 20 KH,

m 0.2 to 25 MHes can propagowever, thesena, so not all m9).

will be definedHarvey, 1963;f oscillations oequency of thation velocity stic of the mawever, this is

erature and oththis reason, thic testing. The a𝐶

(N/m2), µ is tdensity of the erials will behppose to the stic wave or shas a low impeis, the medium

mpedance is him offers great𝑍 𝜌.

dern Applied Sc

35

mersion testinthis case the sps water with thted to the quanozzle is crit

g this techniqush, 2002; Vad

m. Right: Meanic beam passbetter interpret

within this baHz and frequengate through ae waves are amedia are suita

d for the study; Mundry, 196of a particle p

he generator, wof the wave fo

aterial and, ins not strictly her parametersis speed can bacoustic veloc. µ. µ . µ

the Poisson comaterial (Kg/

have the samepassage of th

specific acousedance, its mam will offer litigh, the mass t resistance to. 𝐶 (Kg

cience

ng of the specipecimen and the necessary aality of the watical as it canue, panel or precantos, Migu

asured on the pses through a mtation of the rand we can spncies above 10any material mattenuated by able for transm

y to be carried68). per second. Wiwhich can be cor a given con

n general, is caccurate, sincs. However, ine taken approx

city for a longi

(m/s)

oefficient (dim/m3).

e way when thhe ultrasonic stic impedancass elements wttle resistance elements wil

o elastic defo

g/m2.s)

imen (Silk, 19he probe are im

additives to preater jet, and m

n cause disturrobe inspectionuel, & Gallego

part by immermaterial, certaresults. When peak of nearby00 MHz whichmedium, wheall materials

mitting ultraso

d out (Alonso

ithin the samechosen arbitrarndition, for exaconstant for a ce such speedn many solid mximately as a citudinal wave

mensionless) q

ey are crossedwave is wha

ce. It is a resiswill vibrate at to the elastic l vibrate slow

ormations. Exp

Vol. 15, N

982; Huang etimmersed (76 revent corrosiomore specificarbances. The n is much fasto, 1995).

rsion transmisain physical ph

n we speak of y ultrasound (h is the field o

ere there are as according toound (Ramirez

& Finn, 1987

e wave, it is tharily. ample, in a coma given materid depends, espmaterials, suchconstant of theis given by th

quotient betwe

d by an ultrasoat is known astance that ophigh speed, wdeformations

wly, even if thxperimentally,

No. 1; 2021

t al., 2000; mm water

on damage. ally to the separation ter and can

sion henomena ultrasonic

(frequency of acoustic atoms and o different z Lopez et

7; Pollock,

e same for

mpression ial, at any pecially in h as metals, e material, e formula:

(1)

een lateral

onic wave. as acoustic pposes the with only a

caused by he acoustic

it can be

(2)

mas.ccsen

which indcalled "acKg/m2.s).On the othmedia, a propagatethe reflecgraphical

It can be medium IWe call R

Both are in relationThe energtheory of

It can be don whichspecific a

and the so

where w iwe can pu

in such a

net.org

dicates that thcoustically har. Impedance inher hand, if a part of the enes in the seconcted one, It thelly exposed as

Fseen from fig

II and C is theR the reflectio

dimensionlessn to the incidegy balance expf propagation o

deduced that fh the wave affeacoustic energ

ound intensity

is the angular ut the followin

way that it ca

he acoustic imrd" (steel withn solid bodiesflat acoustic w

nergy of the wnd medium me transmitted os follows:

Figure 5. Perpgure 5 above, e energy reflecn coefficient a

s coefficients ent intensity. pressed in inteof acoustic wa

from the pointects since the

gy is given by:

y by the follow

frequency, A ng expression

an be put:

Mod

mpedance is a h Z = 45.106 Ks is generally hwave falls perpave is reflecte

maintaining its one and Z1 an

pendicular incithat A is the

cted on the boand T the tran𝑅𝑇that express th

ensities remaiaves it is obtai𝑅𝑇t of view of thvalues of R an

: 𝐸wing formula:𝐼is the amplitu

ns of the coeffi𝑅 𝑇𝑅′𝑇′

dern Applied Sc

36

a constant of tKg/m2.s) in conhigher than inpendicularly oed and returnsdirection and

nd Z2 the respe

idence of the incident ener

oundary surfacnsmission coef𝑅 ; 𝑇

the percentage

ins: 𝐼 𝐼 ined that: 𝑅 ;

he acoustic inteand T do not ch

𝜌𝑉

𝜌𝐶𝑉ude and P is theficients of refle;

;

cience

the material. Mntrast to "acou

n liquids and, ion a flat and sos in the same dd sense. If Ii isective acousti

sound wave orgy, B is the ece I, so that A fficient that ar

e of reflected a 𝐼 and there

;

ensities, it is inhange when e

𝜌𝑤 𝐴𝑍𝑉 𝑍𝑤

e sound pressuection and tran

= √𝑅 = 𝑇

Materials withustically soft" in these, higheoft surface thatdirection of th the intensity c impedances

on a flat surfacenergy transm= B + C.

re given by:

and transmitte

efore can be p

ndifferent the exchanged betw

𝑊/𝑤 𝐴 𝑊ure. Taking innsmission of s

Vol. 15, N

h a high impe(water with Z

er than in gaset separates two

he incident, anof the inciden

s of the materi

ce mitted from me

ed intensity re

put: R + T = 1

side of the limtween Z1 and Z

/𝑚

𝑊/𝑚

nto account thisound pressur

No. 1; 2021

edance are Z = 1,5.106 es. o different nother part nt wave, Ir ials, this is

edium I to

(3)

(4)

spectively

. From the

(5)

(6)

mit surface Z2. As the

(7)

(8)

is equation, re:

(9)

(10)

(11)

(12)

mas.ccsenet.org Modern Applied Science Vol. 15, No. 1; 2021

37

From the expression 11 it is deduced firstly that the reflected acoustic pressure will be of the same amplitude, whatever the side of the limit surface on which the wave is incident, that is to say, independently of the sequence of both materials; although in the case of being Z2>Z1, R' will be positive, which indicates that the incident wave and the reflected wave are in phase and, in the opposite case (Z2<Z1), R' will be negative, which indicates an inversion of the phase of the reflected wave in relation to the incident wave. From the expression 12 it is deduced that although the acoustic pressure transmitted in phase with the incident, it will not be independent of the sequence of the two materials, so that if Z2>Z1, then T'>1, which indicates that its amplitude will be greater than that of the incident wave and, in the opposite case (Z2<Z1, T'<1) less. Finally, the balance of sound pressure, in contrast to energy or sound intensity, can be put as

Pi + Pr = Pt, either 1 +R’ = T’ which implies that for balance to be maintained, the sum of the pressures must be the same on both sides of the interface. 2.2 Methodology Applied to the Real Case The determination of possible defects in composite materials widely used in the industrial sector, is the problem to be addressed once established the theoretical foundations of the physical interaction of an ultrasonic wave with matter. It is important that the use of these materials at a technological and industrial level guarantees that they do not have defects that reduce their mechanical properties considerably, especially in critical structures. The main defects to be evaluated are:

• The punctual or generalized porosity that will be basically air inclusions within the polymeric matrix that could have been produced by a low pressure application in the resin curing process in its manufacturing process or by a movement between layers after the resin has started its molecular crosslinking process (Heru et al., 1997).

• In epoxy matrix carbon fiber laminates, low energy impacts cause damage that can result in dents, matrix cracking, fiber to matrix delamination and fiber breakage. Of all these, delamination is probably the most harmful due to the difficulty in detecting it and the reduction it causes in the properties (Baker; Jones & Callinan, 1985; Wróbel, Wierzbicki & Pawlak, 2005). The possible delaminations of a carbon fiber layer will be evaluated, these will be identified by the presence of air due to the detachment of successive layers of fiber due to lack of adhesion between them (Cantwell, Curtis, & Morton, 1986; Miyano et al., 1994).

In Table 1, some physical and mechanical properties of some materials that will be used in the ultrasonic inspection process to be developed on the chosen composite material have been collected. Table 1. Physical and mechanical data on some materials

Material Densidad Ρ (Kg/m3)

Módulo elástico E (Gpa)

Coeficiente de Poisson (adimensional)

Resina epoxy Modificada 8552

1100 3,4 0,4

Fibra de carbono (AS4 alta resistencia)

1790 330 0,1

Agua 1000 --- --- Aire 1,22 --- ---

In the technical inspection performed by ultrasound we will find, first the water layer, since it is an inspection by immersion in water, then the polymer matrix formed by an epoxy resin type, the carbon fiber used as reinforcement, and finally, the presence of air in the form of porosity or delamination. The speed of sound propagation in the different media (CL) and the acoustic impedance (Z) will be determined, since the amount of reflected and transmitted sound will be obtained from these parameters. It will be taken into account that the speed of sound in air is 340 m/s and in water 1435 m/s, and therefore, it is not necessary to calculate them.

mas.ccsen

2.2.1 InspWe are gomaterial. have beenthe panel the combWe will spropagatea frequenThe studyand then,The follothe ultras

Figure 6.

Phase 1: WUsing theTable 1 fofollowingTable 2. A

Using forresults suTable 3. P

In this caindicates incident w

net.org

pection for Pooing to try to We have alre

n able to deterdue to lack of

bination of watstart the ultrases). It will be t

ncy of 1 MHz y will be carrie in a second pwing figure sh

sonic inspectio

. Distribution

Water + epoxe formulas of or the two matg results are reAcoustic velo

M

AR

rmulas 11 andummarized in Percentage of

ase, Z2>Z1 is that T'>1, w

wave. The tran

orosities determine if i

eady seen thatrmine them byf pressure durter, epoxy resonic inspectiotaken into accothat will proded out in two pphase, the epoxhows in schemon to locate th

of layers in th

xy resin the speed of

terials that areeached: city and impe

Material

Agua esina epoxy 8

d 12 of the refTable 3. In th

f reflected andMaterial

Agua Resina epox

positive, indhich implies nsmitted wave

Mod

t is possible tot one of the pay means of imming the forminin and air.

on study takingount that the in

duce an intensiphases, that is,xy resin will b

matic form thehe porosity in t

he study of poflat s

sound propage analyzed in th

edance of wateVeCL

148552 25flected and trais case the wa

d transmitted w

xy 8552 icating that ththat the ample has an acous

dern Applied Sc

38

o detect if theranels (P1) hasmersion ultrasng process, it w

g into accountnspection willity of 200 dB., first the matebe studied wite components othe material.

orosity with pesurface of the

gation (1) and his phase, mat

er and epoxy relocidad acúst

L (m/s) 435 573,59 ansmitted sounater will be mawave from wa

ReflectanciaR’ (%) 32,55

he incident anlitude of the stic pressure o

cience

re are air poros been preparesounds as showwill appear in

t the water firsl be performed. erials such as wth the air. of the materia

erpendicular inplate

the acoustic iterial 1 the wa

resin tica Imped

Z (Kg1,442,83

nd pressures raterial 1 and th

ater and epoxya Transm

T’ (%132,55

nd reflected wtransmitted w

of 132.55%. T

osities in the med expressly fwn in figure 2the epoxy res

st (where the ud using a norm

water and epox

als to be taken

ncidence of th

impedance (2ater and materi

dancia acústica/m2.s)*106

respectively, whe epoxy resin

y resin mitancia ) 5

waves are in wave will be gThis we will re

Vol. 15, N

manufactured for this purpo2b. If there is psin, so we mus

ultrasonic beammal incidence p

oxy resin will b

into account t

he acoustic wa

2) and the paraial 2 the epoxy

a

we obtain the n will be mate

phase. Beinggreater than tepresent in fig

No. 1; 2021

composite ose and we porosity in st evaluate

m initially probe with

be studied,

to perform

ave on the

ameters of y resin, the

following erial 2.

g Z2>Z1, it that of the gure 7.

mas.ccsen

Figure 7 shown is the produwave (13Phase 2: EFor this slack of prof epoxy Table 4. A

Using forresults suTable 5. P

In this caindicates that T'<1 graphical

net.org

Figurshows the valthe transmitte

uct of multiply2.55%) whichEpoxy resin +

study procedurressure duringresin with air

Acoustic veloM

RA

rmulas 11 andummarized in Percentage of

Ma

ReAir

ase of searchina phase inverwhich implie

lly the results

re 7. Acousticlues of the inced wave (greeying the incidh gives intensi+ air re, it will be o

g the forming pr, the first willcity and impe

Material

esina epoxy 8Aire d 12 of the refltable 5.

f reflected andaterial

sina epoxy 85re ng for porosityrsion of the rees that the tranobtained.

Mod

c pressures in tcident (blue) aen) in the epoxdent power (20ities of 65.1 d

operated in a sprocess, it will be material 1

edance for epoVeCL

8552 2534

ected and tran

d transmitted wReR’

552 -99

y in the sampleflected wave nsmitted wave

dern Applied Sc

39

the case of refand reflected (xy resin matri00 dB) by the

dB and 265.1 d

similar way asl appear in the

1 and air will boxy resin and aelocidad acúst

L (m/s) 573,591435 40 nsmitted acous

wave for epoxeflectancia (%) 9,97%

le, it has beenwith respect t

e is smaller th

cience

flection in wa(orange) wavex. The reflect

e percentage odB respectivel

s in phase 1. Ie epoxy resin, be material 2.air tica Imped

Z (Kg2,830,0004

stic pressures

y resin and airTransT’ (%0,028

n obtained thatto the inciden

han the inciden

ater and epoxye pressures in med and transm

of reflected (32ly.

f there is poroso we must ev

dancia acústica/m2.s)*106

41 respectively, w

r mitancia

%) %

t Z2<Z1, so thnt wave. As Z2nt wave. Figu

Vol. 15, N

y resin medium I (wa

mitted intensiti2.55%) and tr

osity in the pavaluate the co

a

we obtain the

hat R' is negati2<Z1, it is also

ure 8 below w

No. 1; 2021

ater). Also ies will be ransmitted

anel due to mbination

following

ive, which o obtained ill express

mas.ccsen

Figure 8 sI (epoxy rintensitieand transthe incideconsideraintensity total (ther2.2.2 InspIf there itherefore,matrix, thcarbon fib

Figure 9.

For the stthe first pand for thquantity delaminaPhase 1: WThe data transmittadB. This

net.org

shows the valresin). It also s will be the pmitted wave (ent wave is eqation, the reflof the transmire is almost nopection for Deis a delaminat, it will be takhe carbon fibeber. The schem

Distribution o

tudy of delamiphase the watehe third and lasof wave that

ation. Water + epoxobtained for tance of 132.55value will be

Figure 8. Acoues of the preshows the tran

product of mu(0.028%). To qual to the inteected intensititted wave is po attenuation)elaminations tion in the mken into acco

er and finally ime can be see

of layers in the

ination identifer and epoxy mst one, the cart crosses the

xy resin the inspection5%, which tranthe one that in

Mod

oustic pressureessures of the insmitted waveltiplying the indetermine the

ensity of the wty will be 265practically zer, so that the p

aterial, it willount that the uit will meet then in the follow

e study of delathe fla

fication, it wilmatrix will bebon fibre and carbon fiber

n for porosity nslates into a rn the study of

dern Applied Sc

40

es in the case incident wavee (green) in aincident powere intensities, iwave transmitt5.02 dB and ro, however, t

porosity can be

l occur betweultrasonic beahe delaminatiowing figure 9.

aminations wiat surface of thll be studied ine selected, forthe air (possib

r is the great

will be valid reflected inten

f the second ph

cience

of epoxy resine (blue) and rer (possible por (200 dB) by t should be tated in the prevthe transmitte

the intensity oe easily identi

een layers of am will pass ton (if any) bef.

ith perpendicuhe plate n three phasesthe second th

ble delaminatiter possible t

for this phasensity of 65.1 dhase enters as

n and porosityeflected wave rosity). The rethe percentag

aken into accovious phase, ieed intensity wf the reflectedfied.

carbon fiber hrough the w

fore passing th

ular incidence

s each with a she epoxy matrion). It is interthing to be a

e, with a reflecB and a transmthe incident w

Vol. 15, N

y (orange) in theflected and trge of reflectedount that the ine 265.1 dB. Mwill be 0.0742d wave is prac

used as reinfwater layer, thhrough the nex

of the acousti

set of two matrix and the carresting above aable to detect

ctance of 32.5mitted intensitwave value.

No. 1; 2021

he medium ransmitted

d (99.97%) ntensity of

Making this 2 dB. The ctically the

forcement, e polymer xt layer of

ic wave on

terials. For rbon fibre, all that the t the later

55% and a ty of 265.1

mas.ccsen

Phase 2: ETaking inand acousTable 6. A

Using equTable 7. Table 7. P

In this caBeing Z2>that of theof its impfigure 10

Figure 10I (epoxy intensitieand transbeneficiaPhase 3: CIt remainspossible dtransmitteTable 8. A

net.org

Epoxy resin +nto account eqstic impedancAcoustic velo

Ma

ReFib

uations 11 and

Percentage of M

RFi

se you have to>Z1, it indicate incident wavpedance Z, th.

Figure 10. A0 shows the varesin). Also ss will be the p

smitted wave al if what is soCarbon fiber +s to be studieddelamination ed previously,Acoustic velo

M

FiA

+ carbon fiber quations 1 andce of the two mcity and impeaterial

esina epoxy 85bra de carbonod 12, the reflec

f reflected andMaterial

esina epoxy 8ibra de carbono Z2>Z1 so R' tes that T'>1, wve. Because ce transmitted

Acoustic pressualues of the preshown is the

product of mul(179.34%) whught is that th+ air (delamind how the acou(presence of , that is, the oncity and impe

Material

ibra de carbonAire

Mod

d 2, and the vamaterials listededance for epo

VelCL

552 257o AS4 137cted and transm

d transmitted wReR’

8552 79no AS4 is positive, whwhich impliesarbon fiber hawave will inc

ures in the delessures of the transmitted wltiplying the inhich gives int

he acoustic wanation): ustic wave woair). It shouldne with an int

edance of carbVeCL

no AS4 1334

dern Applied Sc

41

alues of the and in Table 6 hoxy resin and clocidad acústi(m/s)

73,591435 731,25 mitted acousti

wave for epoxeflectancia ’ (%) 9,34%

which indicatess that the ampas a greater cacrease its pow

lamination stuincident wave

wave (green) incident powertensities of 21ave reaches th

ould react whend be noted agatensity of 475.bon fiber and pelocidad acúst

L (m/s) 3731,25 40

cience

nalyzed materhave been detecarbon fiberica Impe

Z (Kg2,8324,57

ic pressures ha

y resin and caTransmT’ (%179,34

s that the incidplitude of the tapacity to tranwer to 179.34%

udy case for epe (blue) and rein the carbon r (265.1 dB) by10.33 dB and e possible del

n passing throain, that the in.43 dB. possible delamtica Imped

Z (Kg24,570,0004

rials referenceermined.

dancia acústicg/m2.s)*106

7 ave been deter

arbon fiber mitancia ) 4%

dent and reflectransmitted wa

nsmit sound, a%. We are go

poxy resin andeflected wave fiber. The re

y the percenta475.43 dB reamination.

ough the carboncident wave

mination (air) dancia acústica/m2.s)*106

41

Vol. 15, N

ed in Table 1,

ca

rmined and are

cted waves arewave will be gras indicated byoing to repres

d carbon fiber(orange) in th

eflected and trage of reflecteespectively. Th

on fiber and reis the one cal

a

No. 1; 2021

the speed

e shown in

e in phase. reater than y the value ent this in

r he medium ransmitted d (79.34%) his data is

aching the lculated as

mas.ccsen

Table 9. P

In this cawhich indobtained express g

Figure 11medium Iand transmreflected respectivehowever,identify adelaminaattenuatioit should 3. ResultThis studidentifiedUltrasounentire voldetectionis accurataccess byprocess omakes poobservingto undertadisadvantcomplicaof a coupsystem ansamples.

net.org

Percentage of M

FiA

ase of searchindicates a phasthat T'<1 wh

graphically the

Figure 11. Ac1 shows the vaI carbon fibermitted intensi(99.99%) an

ely. The refle the transmitte

a signal of 0.0ation in this tyon of the signabe considered

ts dy of defect idd in this type onds, as a NDTlume of the sa

n of very smalltely determiney a surface to of signal generossible an adeqg the test, theyake a great amtages, such a

ated to inspect,pling fluid, hend the determ

f reflected andMaterial

ibra de carbonAire

ng for delaminse inversion ohich implies the results obtain

coustic pressualues of the pr). The transmities will be thnd transmittedcted intensityed one is very0156 dB that wype of sampleal that had beed to work with

dentification iof material usiT technique, hample but withl defects (depeed, as well as th

perform the ration and recquate documey have a high vmount of induas requiring , the passage o

eterogeneities mination of ce

Mod

d transmitted wReR’

no AS4 -9

nations in the of the reflectehat the transmned.

res in the delaressures of th

mitted wave (ghe product of md wave (0.00

is practicallyy small. Therefwould be a qu

es with the sigen emitted is ph higher frequ

in composite ing the techniqhas a high penh the ability toending on the he estimation test, the resul

ception is contentation of it, tversatility sinstrial problemhighly qualif

of acoustic enevery close to

ertain characte

dern Applied Sc

42

wave for carboeflectancia ’ (%) 9,99%

e sample, it haed wave with mitted wave i

amination studhe incident wagreen) is also multiplying th033%) which y the same as tfore, 0.0033%uite low signagnal used of 1

practically totauencies and sou

materials hasque of ultrasonetration powo test greater tquality of theof size, orient

lt of the inspetrolled electrothere are no ri

nce the ultrasoums (Miyano et fied personneergy from the the surface meristics of the

cience

on fiber and pTransmT’ (%0,0033

as been obtainrespect to the

is smaller tha

dy case for carave (blue coloshown in the

he incident powgives intens

the one that h% of the dB wilal. Probably, i1 MHz frequeal. Therefore, iund intensitie

s shown how ound immersiower because inthicknesses, hamaterial anal

tation, shape aection is instanonically, the tesks for the opund test has aal., 1986; Wa

el, small, irr probe to the s

may not be dedefects requi

ossible delammitancia ) 3%

ned that Z2<Z1e incident wavan the inciden

rbon fiber andr) and reflectepossible delamwer (475.43 dities of 475.3

has come fromll be transmittit would not bency and 200 in order to idens.

porosities andon. n the test provas a high sensyzed), the pos

and nature of hntaneous (reaechnique of imerator or for thgreat amount

an et al., 2016egular, roughsample and vicetected and thires the use o

Vol. 15, N

mination

1, so that R' isve. As Z2<Z1

nt wave. Figu

d delaminationed wave (oran

amination. ThedB) by the per38 dB and 0

m the carbon fted, so the detebe possible to

dB intensity,entify this type

d delaminatio

vides informatsitivity which sition of internheterogeneity,al time) since mage capture the people whot of resources 6). They also hh or thin samceversa requir

he calibration of standard or

No. 1; 2021

s negative, , it is also re 11 will

n nge) in the e reflected centage of

0.0156 dB fiber layer, ector must identify a

, since the e of defect,

ons can be

tion of the allows the nal defects only need the whole of the test o could be that allow

have some mples are res the use of the test

r reference

mas.ccsenet.org Modern Applied Science Vol. 15, No. 1; 2021

43

Calculations of the percentages of acoustic intensities of reflected and transmitted waves allow more or less easily to determine if it is possible to detect defects (porosity and delaminations) in a material (composite). It will be taken into account that for the inspection of delaminations, an almost complete attenuation of the transmitted wave is produced, so in practice, for the search of this type of defects, it should be taken into account to work with higher frequencies, as well as acoustic intensities. 4. Discussion Unidirectional components are not commonly used in aerospace structural components due to their high anisotropy of mechanical properties. Normally multidirectional laminates are used which are optimal for different load conditions (Miyano et al,,1986; Marsh, 2002; DobrzaÉski, 2002). Even so, it is necessary to study unidirectional laminates since this allows us to know how the material behaves in the direction of the fibers, since the sheet presents maximum properties in this direction and minimum properties in the transversal direction. The tested laminate presents the highest resistance and elastic module and the lowest possible thermal expansion coefficient in one direction. The study of the use of the physical phenomena of interaction of the ultrasonic wave in the composite material is an excellent tool to identify defects by ultrasound in this type of materials. This will help the beginners who are initiated in the application of this type of methodologies (especially the ultrasounds) in the laboratories to be able to adjust in a theoretical way the measurement devices, one of the most pronounced difficulties in this type of measurements. This will also provide an approximate idea of the frequency and intensity of work for certain types of materials and defects, reducing preparation times and costs of testing. Acknowledgements The authors wish to thank CEU San Pablo University Foundation for the funds dedicated to the Project Ref. USP CEU-CP20V12 provided by CEU San Pablo University. References Alonso, M., & Finn, E. (1987). Física. Vol II (Campos y Ondas). Ed. Addison Wesley,

Iberoamericana,Wilmington. ASM International Handbook Committee. (1989). Engineered Materials Handbook, 1, Composites. Baker, A. A., Jones, R., & Callinan, R. J. (1985). Damage tolerance od graphite/epoxy composites. Composite

Structures, 4(1), 15-44. https://doi.org/10.1016/0263-8223(85)90018-2 Barbero, E. J. (1999). Introduce to Composite Materials Design. Taylord & Francis, Inc. Bishop, S. M. (1985). The mechanical performance and impact behavior of carbon fibre reinforced PEEK.

Composite Structures, 3, 295-318. https://doi.org/10.1016/0263-8223(85)90059-5 Cantwell, W. J., Curtis, P., & Morton, J. (1986). An assesment of the impact performance of CERP reinforced with

high-strain carbon fibres. Composite Science Technology, 25(2), 133-148. https://doi.org/10.1016/0266-3538(86)90039-4

DobrzaĔski, L. A. (2002). Fundamentals of Materials Science and Physical Metallurgy. WNT, Warszawa (in Polish).

Dunegan, H. L., & Tetelman. (1971). Acoustic Emission. Research and Development, 22(5), 20-24. Gray, S., Ganchev, S., Qaddoumi, N., Beauregard, G., & Radford, D. (1995). Porosity level estimation in

polymerer composites using microwaves. Applied Microwave Nondestructive Testing laboratory, Electrical Engineering Department, Colorado State University, Ft. Collins. Material Evaluation, 404-408.

Harvey, A. (1963). Microwave Engineering. Academic Press. Heru, S. B., Komotori, J., Shimizu, M., & Miyano, Y. (1997). Effects of the fiber content on the longitudinal

tensile fracture behavior of uni-directional carbon/epoxy composites. Journal of Materials Processing Technology, 67, 89-93. https://doi.org/10.1016/S0924-0136(96)02824-5

Hervfeld, K. F., & Litovitis, T. A. (1959). Absorption and Dispersion of Ultrasonic Waves, 535. Huang, Y. D., Liu, L., Zhang, Z. Q., & Wan, Y. (1998). On-line monitoring of resin content for film

impregnation process. Composites Science and Technology, 58, 1531-1534. https://doi.org/10.1016/S0266-3538(97)00185-1

mas.ccsenet.org Modern Applied Science Vol. 15, No. 1; 2021

44

Huang, Y. D., Liu, L., Qui, J. H., & Shao, L. (2002). Influence of ultrasonic tretment on the characteristics of epoxy resin and the interfacial property of its carbon fiber composites. Composites Science and Technology, 62(16), 2153-2159. https://doi.org/10.1016/S0266-3538(02)00148-3

ImieliĔska, K., Castaings, M., Wojtyra, R., Haras, J., Le Clezio, E., & Hosten, B. (2004). Air-coupled ultrasonic C-scan technique in impact response testing of carbon fibre and hybrid: glass, carbon and Kevlar/epoxy composites. Journal of Materials Processing Technology, 157-158, 513-522. https://doi.org/10.1016/j.jmatprotec.2004.07.143

INASMET. (1998). Materiales compuestos. Tecnologías de producción. Centro tecnológico de materiales Inasmet, San Sebastián, España.

Kapranos, P., & Priestner, R. (1987). NDE of difusión bonds. Metals and Materials, 194-198. KINO, G. S. (1979). Acoustic imaging for nondestructive evaluation. IEEE, 67(4), 510-525.

https://doi.org/10.1109/PROC.1979.11280 Marsh, G. (2002). Finding flaws in composites. Reinforced Plastics, 46(12), 22-28.

https://doi.org/10.1016/S0034-3617(02)80227-2 Maslov, K., Kim, Y. R., Kinra, K. V., & Pagano, J. N. (2000). A new technique for the ultrasonic detection of

internal transverse cracks in carbon-fibre/bismaleimide composite laminates. Composite Science and Technology, 60, 2185-2190. https://doi.org/10.1016/S0266-3538(00)00013-0

Miyano, Y., Kanemitsu, M., Kunio, T., & Kuhn, H. A. (1986). Role of matrix on fracture strengths of unidirectional CFRP. Journal of Composite Materials, 20, 520-538. https://doi.org/10.1177/002199838602000602

Miyano, Y., McMurray, M. K., Kitade, N., Nakada, M., & Mohri, M. (1994). Role of matrix resino n the flexure static behavior of unidirectional pitch-based carbon fiber laminates. Advanced Composite Materials, 4(2), 87-99. https://doi.org/10.1163/156855194X00222

Mouritz, A. P., Townsend, C., & Shah Khan, M. Z. (2000). Nondestructive detection of fatigue damage in thick composites by pulse-echo ultrasonics. Composites Science and Technology, 60, 23-32. https://doi.org/10.1016/S0266-3538(99)00094-9

Mundry, E. (1968). Ultrasonidos. Trad. Ed. 1961. INTI. Argentina, 124. Nesvijski, G. (2000). Some aspects of ultrasonic testing of composites. Composite Structures, 48, 151-155.

https://doi.org/10.1016/S0263-8223(99)00088-4 Ochelski, S. (2004). Experimental methods in construction composites mechanics. WNT, Warszawa (in Polish). Padmanabhan, K., & Kishore, E. (1995). Failure behaviour of carbon fibre/epoxy composites in pìn-ended

buckling and vending test. Composites, 26(3), 201-206. https://doi.org/10.1016/0010-4361(95)91383-G Pollock, A. A. (1987). Acoustic Emission Inspection. Metals Handbook, 17. Non Destructive Evaluation and

Quality Contro, 278-294. Potel, C., Chotard, T., Belleval, J. F., & Benzeggagh, M. (1998). Characterization of composite material by

ultrasonic methods. Composites Part B, 159-169. https://doi.org/10.1016/S1359-8368(97)00006-1 Rámirez López, Fc., Del Ojo, G., Fernández Soler, M. A., Valdecantos, C., Alonso, A., & De los Ríos, J. M. (1996).

Métodos de Ensayos No Destructivos. Instituto de Técnica Aeroespacial (INTA). Ramírez, F., & Col. A. (1982). Introducción a los métodos de ensayos no destructivos de control de calidad de los

materiales, 3ª edición, INTA, Madrid, Spain. Richardson, E. G. (1952). Ultrasonic Physics. Elsevier Publishing Companu, Amsterdam, London, New York.

https://doi.org/10.1063/1.3067396 Rojek, M., Stabik, J., & Wróbel, G. (2005). Ultrasonic methods in diagnostics of epoxy-glass composites.

Journal of Materials Processing Technology, 162-163. https://doi.org/10.1016/j.jmatprotec.2005.02.069 Sanglier, G., Del Ojo, G., De Miguel, J., & Peñaranda, J. A. (2003). Estudio mediante Ensayos No Destructivos de

la cubierta (“Radomo”) de un sistema experimental de radar de apertura sintética. 10º Congreso Nacional de E.N.D. Asociación Española de Ensayos No Destructivos. Cartagena, España.

mas.ccsenet.org Modern Applied Science Vol. 15, No. 1; 2021

45

Scarponi, C., & Briotti, G. (2000). Ultrasonic technique for the evaluation of delamination on CFRP, GFRP, KFRP composite materials. Composites, Part B 31, 237-243. https://doi.org/10.1016/S1359-8368(99)00076-1

Silk, M. G. (1982). Defect detection and sizing in metals using ultrasound. International Metal Rev., 1, 28-50. https://doi.org/10.1179/imr.1982.27.1.28 Vaccaro, C., & Akers, D. (1996). Damage assesment in a SMC composite by means of ultrasonic techniques.

Review of Progress in Quantitative Nondestructive Evaluation, 13. Plenum Press, New York. Valdecantos, C., Miguel, J., & Gallego, E. (1995). Ensayo por ultrasonidos de materiales compuestos. Técnicas de

impulso eco. 5º Congreso español de técnicas END. Sevilla, Spain. Wan, H., Ning, F., Hu, Y., Fernando, PKSC, & Pei, Z. J. (2016). Surface grinding of carbon fiber-reinforced

plastic composites using rotatory ultrasonic machining: Effects of tool variables. Advanced in Mechanical Engineering, 8(9), 1-14. https://doi.org/10.1177/1687814016670284

Wróbel, G., & Wierzbicki, A´. (2005). Ultrasounds in diagnosis of strength changes in laminates put in agening conditions. 3rd Scientific Conference on Materials, Mechanical and Manufacturing Engineering, 575-580.

Wróbel, G., Wierzbicki, A’., & Pawlak, S. (2005). Ultrasonic quality evaluation method for polyester glass laminated materials. 11th International Scientific Conference on Contemporary Achievements in Mechanics, Manufacturing and Materials Science. Gliwice - Zakopane, 1040-1044.

Copyrights Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).