ass article

A

Gb

Ca

b

a

ARRAA

KGPUIOF

1

rsaoabFy

mtuthadaci

h0

ARTICLE IN PRESSG ModelPSUSC-30590; No. of Pages 7

Applied Surface Science xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Applied Surface Science

jou rn al h om ep age: www.elsev ier .com/ locate /apsusc

lass fibres reinforced polyester composites degradation monitoringy surface analysis

atalin Croitorua, Silvia Patachiab,∗, Adina Papanceab, Liana Baltesa, Mircea Tiereana

“Transilvania” University of Brasov, Materials Engineering and Welding Department, Eroilor 29 Str., 500036 Brasov, Romania“Transilvania” University of Brasov, Product Design Environment and Mechatronics Department, Eroilor 29 Str., 500036 Brasov, Romania

r t i c l e i n f o

rticle history:eceived 2 April 2015eceived in revised form 29 May 2015ccepted 12 June 2015vailable online xxx

a b s t r a c t

The paper presents a novel method for quantification of the modifications that occur on the surfaceof different types of gel-coated glass fibre-reinforced polyester composites under artificial UV-ageing at254 nm. The method implies the adsorption of an ionic dye, namely methylene blue, on the UV-aged com-posite, and computing the CIELab colour space parameters from the photographic image of the colouredcomposite’s surface. The method significantly enhances the colour differences between the irradiatedcomposites and the reference, in contrast with the non-coloured ones. The colour modifications that

eywords:lass fibresolyesterV-ageing

mage analysisptical microscopyTIR spectroscopy

occur represent a good indicative of the surface degradation, alteration of surface hydrophily and rough-ness of the composite and are in good correlation with the ATR-FTIR spectroscopy and optical microscopyresults. The proposed method is easier, faster and cheaper than the traditional ones.

© 2015 Elsevier B.V. All rights reserved.

. Introduction

Glass fibre reinforced polyester composites (GFPCs) are cur-ently used in a plethora of applications such as constructiontructures, automotive covers, boat hulls, blades for wind turbinesnd so forth. Thus, studies regarding the influence of UV radiationn the GFPCs structure and properties under prolonged exposurere of outmost importance. Usually, these studies could be assessedy complex structural analysis, such as scanning probe microscopy,TIR spectroscopy, correlated with colorimetric or mechanical anal-sis, which are expensive and time consuming [1–4].

The aim of this study is to present a simple and time efficientethod for assessing the chemical modifications that occur on

he surface of different types of GFPCs exposed to artificial ageingnder 254 nm UV radiation, namely photographic image analysis ofhe surface of the composite material. The image analysis methodas been successfully applied in our previous studies in order tossess the roughness profile of wood materials, UV or electron beamegradation of wood veneers [5,6] as well as the adsorption kinetic

Please cite this article in press as: C. Croitoru, et al., Glass fibres reinanalysis, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.201

nd equilibrium of different types of dyes on poly(vinyl alcohol)ryogels [7,8]. Also, degradation of synthetic polymers or compos-tes has been studied by CIELab method, based on chromophoric

∗ Corresponding author. Tel.: +40 741649792.E-mail address: [email protected] (S. Patachia).

ttp://dx.doi.org/10.1016/j.apsusc.2015.06.070169-4332/© 2015 Elsevier B.V. All rights reserved.

group formation during the degrading process that leads to mate-rials yellowing or browning [9a].

Even though the CIELab method has been already used for GFR-PCs materials to study their degradation [9b], the novelty of thiswork consists in the following two aspects: (1) characterization ofthe sample’s colour by analysing the photographic image of thesample, using a suitable software and avoiding expansive spec-trophotometers use and (2) increase the sensitivity of the coloristicmethod by enhancing the modifications in the CIELab colour spaceparameters of the surface by using an ionic or polar dye adsorptionon the composite surface.

The novelty of this work consists in enhancing the modifi-cations in the CIELab colour space parameters of the surface ofdifferent types of composites when submitted to UV-irradiation,through an ionic or polar dye adsorption on the composite surface.The method relies on the principle that UV irradiation could pro-mote degradation on the surface of the composites, which leadsto the formation of polar groups. The polar groups are responsi-ble for a higher methylene blue adsorbed amount, which leads toan intensification of the stain colour, thus providing informationabout the performance of the material (testing of UV-protectivecoatings, correlation with the water adsorption values). Methylene

forced polyester composites degradation monitoring by surface5.06.070

blue adsorption from aqueous solutions has been widely used indeterminations regarding the oxidation degree for cellulosic mate-rials or for surface area determinations of different oxide materials,calcium carbonate [10], graphite, activated carbons, yeast [11,12],

dx.doi.org/10.1016/j.apsusc.2015.06.070


http://www.sciencedirect.com/science/journal/01694332

http://www.elsevier.com/locate/apsusc

mailto:[email protected]


ING ModelA

2 rface

eala

reb

scolawet

2

2

a

oaCc

ect

icam3

rapta

2

2

(1st2pn

i

ARTICLEPSUSC-30590; No. of Pages 7

C. Croitoru et al. / Applied Su

tc. Also, staining with this dye is frequently used in medicinend microbiology in order to enhance the visibility of the cellu-ar components or to highlight possible dysfunctionalities (such asbnormal growth) [13].

Up to this date, there are no studies in the reference literatureegarding the possibility of monitoring the degradation of GFPCsxposed to UV radiation, by quantifying the amount of methylenelue adsorbed on the material’s surface, through CIELab method.

Studies regarding the influence of radiation with a wavelengthmaller than 300 nm on the structure and properties of polyesteromposites have not been extensively reported up to date. Mostf the studies from the reference literature report the use of pro-onged UV-A and UV-B irradiation on several unsaturated and/orromatic polyester matrices [14,15]. Our approach, in using a lowavelength UV radiation (254 nm) decreases the necessary time in

valuation of the surface changes and offers permanent informa-ion regarding the efficiency of the protective coating.

. Materials and methods

.1. Materials

In this study, two different types of commercial GFPCs with rednd white acryl coatings of 0.2 mm thickness have been used.

The polymer matrix of the composites is composed of anrtophtalic-based resin (ENDYNE H 68372TA) and the reinforcinggent consists in a glass-fibre mat roving (E-type, Owens-Corningomposites LLC, U.S.A.). The average fibre weight fraction of theomposites has been 32%.

The red-coated composite consists in a layer of glass-fibre rovingmbedded between two layers of ortophtalic resin and the white-oated composite contains uniformly distributed glass fibres intohe resin.

For the image analysis, water adsorption, methylene blue stain-ng and FTIR spectroscopy the composites have been cut intoircular specimens of 30 mm diameter (the white coated samples)nd 20 mm diameter (the red samples) respectively. The compositeaterial average thickness was 2.50 mm for the white samples and

.30 mm for the red samples.All of the specimens were washed with ethanol in order to

emove the contaminants from the surface, dried at 105 ◦C for 4 hnd then conditioned for a week at 24 ◦C and 54% relative humidityrior to analysis. The average humidity at equilibrium for the twoypes of conditioned composites was 0.150% for the white samplesnd 0.206% for the red samples.

.2. Methods

.2.1. Accelerated ageing of GFPCsThe conditioned GFPCs have been introduced in an UV-irradiator

Bio-Link 254, Viber-Lourimat) and exposed for a total time of0 h to a UV radiation of 254 nm, having the irradiance valueet at 120 mJ/cm2. During the irradiation experiments the rela-ive humidity in the chamber was 55 ± 5% and the temperature4 ± 5 ◦C. Two sets of samples from each types of obtained com-osite have been exposed to UV on the coated and respectivelyon-coated side.

The notation of the samples in the tests performed is the follow-ng:

I: initial non-irradiated sample;


R: sample with red coating;W: sample with white coating;F: coated side of the sample;B: uncoated side of the sample;

PRESSScience xxx (2015) xxx–xxx

UVF: UV-irradiated on the coating;UVB: UV-irradiated on the back (not protected by coating);MB at the end of the codification: samples immersed in methyleneblue solution.

2.2.2. Water adsorption testsThe conditioned GFPCs were immersed into closed recipients

containing 10 mL of distilled water, and their mass has been deter-mined at precise time intervals during a 24 h interval.

The relative mass gain at equilibrium (�meq) of the samplesduring water storage was calculated using Eq. (1) [16]:

�meq = (meq − mt=0)mt=0

× 100 (1)

where meq is the mass of the composite (initial and irradiated onvarious sides) at equilibrium of water sorption and mt=0 is the massof the composite before water immersion (at t = 0).

2.2.3. Methylene blue adsorption on the composite’s surfaceIn order to determine the possible structural modifications

that occur on the surface of the samples during UV irradiation(polar groups formation) the GFPCs samples have been introducedin 10 mL of 200 mg/L aqueous methylene blue solution for 1 h,removed from the liquid and dried for 24 h at room temperature.Then, the pictures of coloured dried samples have been taken andused for the image analysis interpretation, according to Section2.2.4.

2.2.4. Image analysisColour changes on GFPCs surfaces due to UV-irradiation were

analysed using an alternative technique to those extensively usedup to this date. The novelty is the using of photographic imageanalysis, instead of a photocolorimeter. In our previous paperswe have demonstrated that this technique offers good relativeresults on UV-irradiated wood, in agreement with experimentaldata obtained by other analysis methods, such as FTIR spectroscopy[5,6].

Initial photographic images of the irradiated samples and refer-ence, as well as of the samples after MB sorption have been takenwith the help of a Sony DSC110 digital camera (3072 × 2034 pixelsresolution), under the same lighting conditions.

The individual images of the samples have been loaded in AdobePhotoshop and the L*, a*, b* parameters using the CIELAB (8 bit)channel were determined in twenty points for each specimen, andthe average value was used in further interpretations.

L* represents the lightness and varies from 100 (white) to 0(black) while a* and b* represent chromaticity indexes: +a* red, −a*

green, +b* yellow, −b* blue.The colour differences have been calculated using Eqs. (2)–(4)

and the total colour difference parameter �E* has been calculatedfrom Eq. (5) for each irradiated sample before and after MB sorption[17]:

�L∗ = L∗2 − L∗

1 (2)

�a∗ = a∗2 − a∗

1 (3)

�b∗ = b∗2 − b∗

1 (4)

�E∗ =√

�L∗2 + �a∗2 + �b∗2 (5)

where subscript 1 denotes the values obtained for the reference andsubscript 2 denotes the values after UV irradiation on the coatedand uncoated side respectively.


Positive values of �a* describe a red shift, negative values of�a* a green shift, while positive values of �b* represent a yellowshift and negative values of �b* a blue shift for the colour of theirradiated samples, in comparison to the reference.


IN PRESSG ModelA

rface Science xxx (2015) xxx–xxx 3

2

CU

2

almcbu

2

Ber

2

ufDd

tip

pw

�

wo�t

co

fe

3

tum

ucma

i

mlot

ples) and UV-aged samples treated with methylene blue presentedin Fig. 2 and Table 2 clearly indicate that surface of the samplesis more polar, owing to an increase in the blue colouration of thesamples after irradiation, on both sides.

Table 1Colour parameters differences for irradiated samples irradiated on various sides.

Irradiation Photo position �L* �a* �b* �E*

UVF

RF −1 1 −2 2.44RB −5 −13 4 14.49WF 0 0 0 0WB 2 −9 13 15.93



.2.5. Optical microscopy imagingThe optical microscopy images have been performed with a

arl-Zeiss Jena metallographic microscope, equipped with a digitalSB image acquisition camera, at 500× magnification.

.2.6. Determination of the static friction coefficientThe static friction coefficients (�s) of the samples (reference

nd UV-irradiated samples) on both sides, against metal (stain-ess steel), have been determined by the inclined plane tribometric

ethod, according to the reference literature [18]. Briefly, theonditioned specimens have been placed onto the surface of the tri-ometer (at null inclination). The inclination has then been variedntil the samples started to slide down the tribometer surface.

.2.7. FTIR spectroscopyThe ATR-FTIR spectra of the composites were obtained with a

ruker-Vertex 70 Fourier transform infrared (FTIR) spectrometer,quipped with an attenuated total reflectance (ATR) device with aesolution of 4 cm−1 in the 4000–600 cm−1 interval.

.2.8. Surface energy determinationsContact angle measurements of UV-aged samples and reference

sing distilled water and glycerol as reference liquids were per-ormed at 25 ◦C with an OCA System 20 goniometer, provided byata Physics Co., Ltd. 5 drops of test liquid, 4 �l in volume wereeposited onto the coated surface of the same GFPC.

The surface energy of the samples was calculated according tohe Wu method, with the help of the instrument software. Accord-ng to this method, the surface free energy is divided into a polarart and a disperse part.

According to this approach, the surface energy (�) is decom-osed into a Lifshitz-van der Waals (�d) dispersive component, asell as into a polar component, �p, according to Eq. (6) [19]:

= �s + �l − 4 · �dl

· �ds

�dl

+ �ds

− 4 · �pl

· �ps

�pl

+ �ps

(6)

here �dl

and �pl

represent the dispersive and the polar componentsf the test liquid(s), determined from the reference literature andds and �p

s represent the dispersive and polar components of theested surface.

The initial contact angle �0 at the beginning of the wetting pro-ess (at t = 0) distilled water and glycerol was used in the calculationf surface energy.

The relative error of the surface energy determinations was 1%or the overall, dispersive and polar components of the surfacenergy.

. Results and discussion

By analysing the samples from Fig. 1 it could be observed that inhe case of all the irradiated samples, delamination occurs on thencoated side, and this delamination and hence the roughness isore intense in the case of the white sample.Also, it could be observed that roughness increase on the

ncoated side occurs also in the case of irradiating the sample on theoated side, meaning the UV radiation is able to penetrate into theass of the material, promoting several structural modifications,

lso evidenced from FTIR spectroscopy analysis (Figs. 5 and 6).Table 1 illustrates the change in the colour parameters for the

rradiated samples, by comparing to the reference.As it can be seen from Table 1, the UV irradiation generally deter-


ines an overall decrease of the L* parameter, associated with theightness. This decrease could be correlated with the degradationf the material. Possible cromophore groups formation and � elec-rons conjugation (O H, C O, C C) on the surface sample, under

Fig. 1. Photographic images of the irradiated GFPCs and reference.

UV irradiation, determined the darkening of the material colour.More intense degradation of material means higher number of cro-mophores that lead to more intense and darker colour. In CIELABsystem this situation is described by L* parameter decrease.

The most pronounced decrease of the lightness parameter isregistered for the uncoated side of the samples exposed to UV, forwhich also the strongest delamination occurred. UV radiations alsodetermine on the opposite side of the sample a decrease in thelightness parameter, probably due to structural rearrangements ofthe material.

When irradiating the samples on the coating, a total colourmodification (�E*) increase of the coating is observed for the redsample, while in the case of the white-coated sample the colouris maintained. On the opposite face of the samples irradiated onthe coating, significant colour modifications occur, towards green(more pronounced for the red sample) and towards yellow for thewhite sample. Also, the total colour modification of the white sam-ple on the uncoated side, opposed to irradiation, is the highest,which could serve as an additional information that white sampleis more prone to degradation than the red sample.

When irradiating the samples on their uncoated sides, a morepronounced modification in �L* occurs, while for the oppositeside of the irradiation, minimal colour modifications occur, morenotably for the white sample, towards blue.

The photographic images of the reference (non-irradiated sam-


UVB

RF 2 −2 0 2.82RB −6 1 0 6.08WF −1 −1 −2 2.44WB −8 0 −3 8.54


ARTICLE IN PRESSG ModelAPSUSC-30590; No. of Pages 7

4 C. Croitoru et al. / Applied Surface Science xxx (2015) xxx–xxx

Table 2Colour parameters difference of the initial and irradiated samples immersed in MB in comparison with the (1) initial non-irradiated samples and (2) initial non-irradiatedsamples immersed in MB.

Sample Photo position �L* �a* �b* �E*

1 2 1 2 1 2 1 2

Non-irradiated

RF-MB −13

–

−24

–

−28

–

39.10

–RB-MB −33 −18 −44 57.87WF-MB −5 −17 −13 21.97WB-MB −25 −16 −53 60.74

UVF

RF-MB −18 −5 −39 −15 −35 −7 55.40 17.29RB-MB −29 4 −19 −1 −50 −6 60.84 7.28WF-MB −13 −8 −38 −21 −26 −13 47.84 25.96WB-MB −26 −1 −22 −6 −48 5 58.85 7.87

RF-MB −15 −2 −30 −6 −30 −2 45 6.63

tts

dsU

ct

biiiis

UVBRB-MB −36 −3 −14WF-MB −1 4 −16WB-MB −34 −9 −14

The differential increase in polarity of the samples after irradia-ion proves useful in enhancing several colour differences betweenhe irradiated samples and the reference, thus increasing the sen-itivity of the proposed image analysis method.

After irradiation, the colour modifications that occur on the irra-iated side are almost double by comparing with the non-irradiatedample. The most important colour modification occurs for theVB-WB-MB sample.

In case of the unstained samples, possible light reflection on theoated surface due to the glossiness determines small variations ofhe colour parameters on both white and red samples.

The adsorbed MB is able to increase the colour differencesetween the reference and the red sample irradiated on the coat-

ng with 600% and respectively 270% in the case of the samplerradiated on the unprotected side. The information regarding the


ncreased susceptibility of the white sample to UV degradations also maintained in the case of MB staining. Also, by using MBtaining technique it is possible to increase the colour differences

Fig. 2. Reference and irradiated samples stained with MB.

4 −54 −10 66.39 11.181 −8 5 17.91 6.482 −61 −8 71.22 12.20

between the two faces of the samples irradiated on one side, espe-cially for the white sample.

Delamination of the unprotected surface on irradiation for thewhite sample could be better observed from the optical microscopyimages, presented in Figs. 3 and 4. In the case of the initial referencesample, uniform embedding of the glass fibre rowing into the resincould be observed. In the case of the initial red sample, a higherinitial surface roughness could be observed, due to the presence ofsurface microdefects (pores, cracks, etc.).

Irradiation of the samples determines delamination of the glassfibres in the case of the white-coated samples and an increase in theporosity of the red-coated samples, as well as an overall increasein the colouration of the oxidized surface when immersed in MB.Delamination is also sustained by the values of the average rough-ness of the composite, determined by the image analysis method,as described in our previous work [5,6], as well as from the valuesof the friction coefficient, both presented in Table 3.

Assuming that static friction coefficient is direct proportional tothe roughness of the composite surface, it could be observed fromTable 3 that irradiated samples with the higher roughness (higherfriction coefficient) are the ones irradiated on the unprotectedside. From the data presented in Table 3, it could be concludedthat most affected by UV-irradiation are the white samples, atirradiation on the unprotected side. For all of the samples leach-ing of ionic compounds into distilled water has been observed,by monitoring the electrical conductivity of the storing water atdetermined time intervals, using a Consort C835 multiparameteranalyser. The irradiated samples eliminate more ionic compoundsinto the storing water. The white sample, irradiated on the unpro-tected side evidenced the higher percentual water uptake. Still thewater adsorption values remain at a low value (<1.2%) by comparing


to the classic hydrophyllic materials.The FTIR spectra of the initial and UV-irradiated samples on

both the coated and uncoated sides (Figs. 5 and 6) reveal several

Table 3Static friction coefficients, average roughness and percentual water uptake of theGFPCs.

Sample �s Ra Sample �meq (%)

IRF 0.1950 2.34IR 0.534IRB 0.1411 13.14

IWF 0.1719 4.52IW 0.775IWB 0.1773 27.32

UVF-RF 0.2024 2.46 UVF-R 0.739UVF-WF 0.1702 14.38 UVF-W 0.975UVB-RB 0.2058 16.11 UVB-R 0.807UVB-WB 0.1720 42.33 UVB-W 1.077



C. Croitoru et al. / Applied Surface Science xxx (2015) xxx–xxx 5

Fig. 3. Microscopic images of the unstained and MB-stained initial and UV-irradiated white samples on the unprotected side (500× magnification).

d UV-

it

ptsmvs

Fs

Fig. 4. Microscopic images of the unstained and MB-stained initial an

nformation regarding the structural modifications that occur onhe surface of the samples.

The composites analysed on the coated face (Figs. 5a and 6a)resent typical bands ascribed to acrylic resins. The weak absorp-ion from ∼3439 cm−1 and 1601–1639 cm−1 corresponds to O Htretching and bending vibration, and respectively to physic-sorbed


oisture [20]. The two broad bands in the 2854–2912 cm−1 inter-al, correspond to alkyl stretching modes and the sharp intensetretching vibration at 1721 cm−1 is ascribed to C O groups. The

ig. 5. (a) FTIR spectra of the initial and UV-exposed white samples for the coated side aide.

irradiated red samples on the unprotected side (500× magnification).

several distinct absorption bands from 1150 cm−1 to 1240 cm−1

can be attributed to the C–O–C stretching vibration modes from thealiphatic acrylate polyester. The bands from 1388 cm−1 to 724 cm−1

can be attributed to the �-methyl group vibrations. The band from987 cm−1 is the characteristic overtone absorption vibration ofacryl resins (fingerprint band), together with the band at 1082 cm−1


[21,22]. The band at 1441 cm−1 can be attributed to the bendingvibration of the C H bonds of CH3, whereas the band at 845 cm−1

could be ascribed to the inorganic filler from the coating [1].

nd (b) FTIR spectra of the initial and UV-exposed white samples for the uncoated



6 C. Croitoru et al. / Applied Surface Science xxx (2015) xxx–xxx

Fig. 6. (a) FTIR spectra of the initial and UV-exposed red samples for the coated side and (b) FTIR spectra of the initial and UV-exposed red samples for the uncoated side.

F ith MBU

so(tcmhfabgtaitmt(v

s8p[

oosoti

bands at 3445–3451 cm−1 and 1610 cm−1, attributed to the free-adsorbed water. The increase is more pronounced in the case ofthe white-coated composites. Also, a 3.4% increase of the car-bonyl absorption band intensity reported to the band centred at

Table 4Fitting parameters for the linear dependence between �E* and IOH respectively�p/�d .

Correlation type Samples analysisposition

Dependence: y = a + b·x

a b R2

ig. 7. Correlation between total colour modification for the composites stained wVF-RB; 6: UVF-WB; 7: UVB-RB; 8: UVB-WB.

Regarding the initial composites analysed on the uncoatedide (Figs. 5b and 6b), typical aromatic polyester bands could bebserved. The band centred at ∼3445 cm−1 (IWB) and ∼3451 cm−1

IRB) is related to the stretching vibration of O H groups [23]. Ashis stretching vibration is shifted to higher wavenumbers for thease of the IRB composite, a stronger interaction of the polyesteratrix with the glass fibre could be possible, accounting for the

igher UV resistance of the red-coated composite. The weak bandsrom 3075 cm−1 (IWB) and 3078 cm−1 (IRB) could be assigned to theromatic C H vibrations. The bands in the 2840–2960 cm−1 coulde ascribed to the C H stretching vibrations of the aliphatic alkylroups from the polyester matrix. A sharp intense band attributedo C O stretching vibration could be observed at 1721 cm−1 (IWB)nd 1728 cm−1 (IRB). The bands from 1593 to 1490 cm−1, weaker inntensity, could be assigned to different C H vibration modes fromhe aromatic ring and the bands located at 1423 and 1378 cm−1

ay correspond to the scissoring and rocking vibration modes ofhe alkyl groups. The strong sharp band occurring at 1268 cm−1

IWB) and respectively 1279 cm−1 (IRB) is attributed to the twistingibration mode of CH2 groups [24,25].

The bands at 1106 and 1065 cm−1 are attributed to C Otretching vibrations of the aromatic polyester and the bands from50 to 729 cm−1 could be attributed to the aromatic ring out oflane bending and to the mono-substituted aromatic ring stretch14].

After UV irradiation, several structural modifications occur.On the protective coating, an overall increase in the intensity

f the OH band could be observed, probably due to the formationf several polar compounds (alcohols, hydroperoxides) by chain


cission promoted by the 254 nm UV radiation. The formationf those polar compounds determines the overall increase ofhe moisture content of the composites, as determined from thentensity increase of the band at 1603 cm−1 as well as an increase in

and (a) IOH; (b) surface energy; 1: UVB-RF; 2: UVB-WF; 3: UVF-RF; 4: UVF-WF; 5:

the methylene blue adsorbed amount. An increase in the intensityof the band at 845 cm−1 ascribed to the inorganic filler could bedue to the overall structural modifications of the coating, namelythe quantitative reduction of the organic component and/or themodification of the IR absorption coefficient of the material.

The highest relative intensity modifications of the IR absorptionbands occur in the case of the white coating, in good agreementwith the colour image analysis based on the adsorbed MB amount(Fig. 7a and Table 4) and the higher ratio between the polar and thedispersive components of the surface energy (�p/�d) (Fig. 7b) [14].

When irradiating the samples on the unprotected side, it couldbe noted that several modifications occur on the coating, mainlyin increasing the ratio (IOH) between the absorption intensity cor-responding to the OH groups (I3445–3451) and the absorptionintensity of the band at centred 2900 cm−1 (I2900).

On the uncoated side, the UV irradiation determines the increasein the polarity of the surface, as determined from the higher rela-tive increase of the intensity corresponding to the OH absorption


FTIR-coloristic

Coated side 4.64 35.76 0.960Uncoated side 0.13 51.22 0.997

Surface energy-coloristic Coated side 2.11 66.00 0.973


ING ModelA

rface

2cssr

aitatoc

tstPhctati�il

4

oesfiaca

mtmftsa5t

mpm

A

gtt

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[



900 cm−1 could be noted in the case of the UVB-WB sample, inomparison with the initial sample, which could be due to theupplementary oxidation of the material, promoted by chain scis-ion and/or oxidation of the terminal part of the crosslinkedesin.

In the case of the UVB-RB sample, minimal increase of OHnd carbonyl absorption intensity could be observed (1.3–2%)n comparison with the IRB sample, reported to the band cen-red at 2900 cm−1. Minimal modifications of the aforementionedbsorption intensities occur in the case of the FTIR spectra ofhe UVF-RB composite, due to the higher UV-screening efficiencyf the red protective coating, in comparison with the whiteoating.

As it can be seen from Table 4, a linear dependence of �E* ofhe OH band relative intensity and of the �p/�d ratio for the MB-tained composites exists. Related to this aspect, the sensitivity ofhe coloristic method is higher for the uncoated side of the GFR-Cs, with respect to the increase of IOH, as demonstrated from theigher slope of the linear dependence in this case (Table 4). Thisould be due to the higher amount of adsorbed methylene blue onhe uncoated side, correlated with the higher surface roughnessnd chemical modifications that occur on this side. As the values ofhe contact angles (and hence the surface energy) is significantlynfluenced by surface roughness, so the dependence of �E* of thep/�d ratio has been discussed only for the coated side. The coat-

ng is more resistant to UV degradation, as demonstrated from theower values of the surface energies ratio.

. Conclusions

This study has been aimed at demonstrating the usefulnessf a colorimetric method based on CIELab colour space param-ters analysis by photographic image analysis in assessing theurface modifications that occur in different types of E-glassbre-reinforced ortophtalic polyester composites submitted toccelerate artificial UV-ageing. In order to test the general appli-ability of the proposed method, two composites with differentcryl-based coatings, of red and white colours have been used.

The innovative approach of the study is to enhance the colourodifications that occur on the surface of the composite by adsorp-

ion of an ionic dye, namely methylene blue. The total colourodifications promoted by the adsorption of this dye on the sur-

ace of the composite has been found to be directly proportionalo the relative intensity of the OH band, calculated from the FTIRpectra of the samples and also with the ratio between the polarnd the dispersive component of the surface energy, and are with–50% higher than in the case of the modifications correspondingo the non-treated surface.

Owing to the direct proportionality between the total colourodification parameter and the polarity of the analysed surface, the

roposed method can assess the degradation degree of a compositeaterial easier, faster and cheaper than the traditional ones.

cknowledgements


This article was supported by the Sectorial Operational Pro-ramme Human Resources Development (SOP HRD), financed fromhe European Social Fund and by the Romanian Government underhe project number POSDRU/159/1.5/S/134378.

[

PRESSScience xxx (2015) xxx–xxx 7

References

[1] D. Scalarone, M. Lazzari, O. Chiantore, Acrylic protective coatings modified withtitanium dioxide nanoparticles: comparative study of stability under irradia-tion, Polym. Degrad. Stabil. 97 (2012) 2136–2142.

[2] W. Harizi, S. Chaki, G. Bourse, M. Ourak, Mechanical damage characterizationof glass fiber-reinforced polymer laminates by ultrasonic maps, Compos. PartB – Eng. 70 (2015) 131–137.

[3] V. Cecen, Y. Seki, M. Sarikanat, I.H. Tavman, FTIR and SEM analysis of polyester-and epoxy-based composites manufactured by VARTM process, J. Appl. Polym.Sci. 108 (2008) 2163–2170.

[4] J.M. Ferreira, O.A.Z. Errajhi, M.O.W. Richardson, Thermogravimetric analysis ofaluminised E-glass fibre reinforced unsaturated polyester composites, Polym.Test 25 (2006) 1091–1094.

[5] S. Patachia, C. Croitoru, C. Friedrich, Effect of UV exposure on the surface chem-istry of wood veneers treated with ionic liquids, Appl. Surf. Sci. 258 (2012)6723–6729.

[6] C. Croitoru, S. Patachia, F. Doroftei, E. Parparita, C. Vasile, Ionic liquids influenceon the surface properties of electron beam irradiated wood, Appl. Surf. Sci. 314(2014) 956–966.

[7] R. Dobritoiu, S. Patachia, A study of dyes sorption on biobased cryogels, Appl.Surf. Sci. 285 (2013) 56–64.

[8] A. Papancea, S. Patachia, Characterization of dyes loaded PVA based hydrogelsthrough CIELAB method, Environ. Eng. Manage. J. 14 (2) (2015) 361–371.

[9] (a) A. Moldovan, S. Patachia, C. Vasile, R. Darie, E. Manaila, M. Tierean, Naturalfibres/polyolefins composites (I) UV and electron beam irradiation, J. BiobasedMater. Bio. 7 (2013) 58–79;(a) J.R. Correia, S. Cabral-Fonseca, F.A. Branco, J.G. Ferreira, M.I. Eusébio, M.P.Rodrigues, Durability of pultred glass-fiber-reinforced polyester profiles forstructural applications, Mech. Compos. Mater. 42/4 (2006) 325–338.

10] D.E. Chirkst, I.S. Krasotkin, O.V. Cheremisina, M.I. Streletskaya, M.V. Ivanov,Determination of the surface area of minerals by sorption of methylene blueand thermal desorption of argon, Russ. J. Appl. Chem. 76 (2003) 663–665.

11] C.A. Nunes, M.C. Guerreiro, Estimation of surface area and pore volume of acti-vated carbons by methylene blue and iodine numbers, Quim Nova 34 (2011),472-U309.

12] K. Painting, B. Kirsop, A quick method for estimating the percentage of viablecells in a yeast population, using methylene-blue staining, World J. Microbiol.Biotechnol. 6 (1990) 346–347.

13] A. Harari, R.S. Sippel, R. Goldstein, S. Aziz, W. Shen, J. Gosnell, Q.Y. Duh, O.H.Clark, Successful localization of recurrent thyroid cancer in reoperative necksurgery using ultrasound-guided methylene blue dye injection, J. Am. Coll. Surg.215 (2012) 555–561.

14] D.E. Moulzakis, H. Zoga, C. Galiotis, Accelerated environmental ageing study ofpolyester/glass fiber reinforced composites (GFRPCs), Compos. Part B – Eng. 39(2008) 467–475.

15] H. Gu, Degradation of glass fibre/polyester composites after ultraviolet radia-tion, Mater. Design 29 (2008) 1476–1479.

16] O.A.Z. Errajhi, J.R.F. Osborne, M.O.W. Richardson, H.N. Dhakal, Water absorp-tion characteristics of aluminised E-glass fibre reinforced unsaturated polyestercomposites, Compos. Struct. 71 (2005) 333–336.

17] M. Panek, L. Reinprecht, Colour stability and surface defects of naturally agedwood treated with transparent paints for exterior constructions, Wood Res. –Slovakia 59 (2014) 421–429.

18] S. Bobancu, R. Cozma, Instrument for measuring friction characteristics in aplane coupling, Proc. Ninth World Congress ToMM 4 (1995) 2935–2938.

19] A.P. Sirocic, Z. Hrnjak-Murgic, J. Jelencic, The surface energy as an indicatorof miscibility of SAN/EDPM polymer blends, J. Adhes. Sci. Technol. 27 (2013)2615–2628.

20] H. Miyazaki, Y. Teranishi, T. Ota, Fabrication of UV-opaque and visible-transparent composite film, Sol. Energ. Mater. Sol. C 90 (2006) 2640–2646.

21] D.J. Kang, D.H. Han, D.P. Kang, Fabrication and characterization of photocurableinorganic–organic hybrid materials using organically modified colloidal-silicananoparticles and acryl resin, J. Non-Cryst. Solids 355 (2009) 397–402.

22] C.I. Simionescu, A.A. Popa, E. Comanita, L. Ionescu, Poly(methyl metacry-late) macroazoinitiator – synthesis, characterization and its behavior in blockcopolymerization, Rev. Roum. Chim. 38 (1993) 1027–1030.

23] K. Rot, M. Huskic, M. Makarovic, T.L. Mlakar, M. Zigon, Interfacial effects inglass fibre composites as a function of unsaturated polyester resin composition,Compos. Part A – Appl. Sci. 32 (2001) 511–516.

24] C. Varga, N. Miskolczi, L. Bartha, G. Lipoczi, Improving the mechanical propertiesof glass-fibre-reinforced polyester composites by modification of fibre surface,


Mater. Design 31 (2010) 185–193.25] M.T. Isa, A.S. Ahmed, B.O. Aderemi, R.M. Taib, I.A. Mohammed-Dabo, Effect of

fiber type and combinations on the mechanical, physical and thermal stabilityproperties of polyester hybrid composites, Compos. Part B – Eng. 52 (2013)217–223.


http://refhub.elsevier.com/S0169-4332(15)01405-1/sbref0130























































































































































































































http://refhub.elsevier.com/S0169-4332(15)01405-1/sbref0170b


















































































































































































































































































































































































































































































































































ass article

Documents