This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Correlating the adhesion of an acrylic coating to the physico-mechanicalbehavior of a polypropylene substrate

A. Salimi a,n, S.M. Mirabedini a, M. Atai a, M. Mohseni b, M.R. Naimi-Jamal c

a Iran Polymer and Petrochemical Institute, P.O. Box 14965/115, Tehran, Iranb Department of Polymer Engineering and Color Technology, Amirkabir University of Technology, P.O. Box 15875-4413, Tehran, Iranc Department of Chemistry, Iran University of Science and Technology, P.O. Box 16846-11367, Tehran, Iran

a r t i c l e i n f o

Article history:

Accepted 12 January 2011Available online 21 January 2011

Keywords:

Polyolefins

Surface modification

Viscoelasticity

Adhesion by physical adsorption

a b s t r a c t

In this study, the adhesion strength of an acrylic coating onto a polypropylene (PP) based substrate was

studied. The adhesion strength of the PP containing various wt% of oxidized wax was found to be

dependent on the surface free energy of the substrate. To this end, the geometric mean and the acid–

base approaches were used to estimate the surface free energy, both of which shown to exist a direct

relation between surface hydrophilicity and the content of the oxidized wax. The viscoelastic behavior

of the adherend also contributed to the coating adhesion. The viscoelastic response of the blends was

investigated using dynamic mechanical thermal analysis (DMTA) and universal testing machine (UTM)

for the bulk properties of the substrate and a depth sensing indentation technique for its near-surface,

respectively. Bulk and near-surface moduli were decreased as a result of incorporation of wax additive

into PP matrix. These were attributed to the homogeneous distribution of low molecular weight wax

chains in the polymeric matrix and hence their impact on substrate integrity. The DMTA analysis

showed a single phase matrix for all blends. Using a linear trend line, the adhesion test results revealed

a good correlation to surface free energy calculations compared with the substrate modulus. The

weaker effect of substrate modulus on surface adhesion was ascribed to the substrate plasticity and/or

development of internal stresses in the coating layer. However, the glass transition temperature of the

coating suggested that the internal stresses in coating layer were responsible to relax prior to

conducting adhesion measurement.

& 2011 Elsevier Ltd. All rights reserved.

1. Introduction

The poor adhesion of coatings to polyolefins is a real concernin industry. Different approaches have been adopted to overcomethis difficulty. Due to the ease of application, the use of effectiveadditives such as chlorinated polyolefin (CPO) or polar waxes intoeither polymer substrate or coating formulation has beenattracted much attention [1–3]. The main properties of a sub-strate such as elastic modulus and elongation-at-break have beenreported to be influenced by the presence of such additives. In thecase of polyolefin/modified wax blends, the incompatibility of theadditive with the polymer matrix is responsible for the impair-ment of mechanical properties [4].

In polyolefin-based substrates the presence of modified waxeswas suggested to influence the wetting properties of the sur-face [5]. There are different mathematical approaches to estimatethe surface free energy of solid polymers. In geometric mean

theory, also known as Owens–Wendt–Rabel and Kaelble approach(OWRK approach) [2,6], the polar and dispersive components ofsurface free energy contribute to the total surface free energy.However, the acid–base theory involves the electron donor/acceptor interactions and London–van der Waals forces to calcu-late the components of surface free energy more closely withsurface chemical nature. The theory of acid–base, as proposed byvan Oss–Good–Chaunchury (VOGC approach) [7–9], has receivedsignificant supports from many researchers recently. In bothapproaches, different combinations of appropriate probe liquidsplay an important role in calculating the components of surfacefree energy [10].

In the study of coating adhesion, the potential effect ofsubstrate viscoelastic behavior has also been addressed. Ananalysis by Ryntz et al. [11] showed that the mold fillingparameters affecting the surface crystallinity are decisive inachieving a good coating adhesion to thermoplastic polyolefins.The effect of substrate on the coating adhesion can also beconsidered due to the existence of the internal stresses and alsobecause of the plastic deformation within the coatinglayer [12,13]. The internal stresses which appear during the

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/ijadhadh

International Journal of Adhesion & Adhesives

0143-7496/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.

doi:10.1016/j.ijadhadh.2011.01.003

n Corresponding author. Tel.: +98 021 44580000; fax: +98 021 44580021.

E-mail address: a.salimi@ippi.ac.ir (A. Salimi).

International Journal of Adhesion & Adhesives 31 (2011) 220–225

Author's personal copy

formation of polymeric coating, adversely affect its adhesionproperty [13]. However, the plastic deformation within coatinglayer, which appears during adhesion test increases the work ofadhesion. This phenomenon is an energy consuming process andhence increases the work of debonding [13,14]. The extent ofsubstrate viscoelasticity in underlying substrate determines therate of stress relaxation process in adherend during adhesiontest [15]. The energy dissipation, both within coating layer andpolymeric substrate, may be resulted in the higher adhesionstrength [14].

The substrate viscoelasticity can be evaluated by dynamicmechanical analysis. However, some recent approaches such asnano-indentation tests have shown to be useful in determiningthe elastic modulus of the uppermost layer of polymeric sub-strates. These techniques were used successfully to evaluate themodulus of elastomeric polymers [16–18]. However, the attrac-tive interactions at the tip–sample interface may give rise to somedeviations in exact determination of the surface elastic modu-lus [19]. By addressing these adhesive interactions using variouscontact models such as Johnson, Kendall, Roberts (JKR), thedeflection in elastic modulus could be resolved [19].

In our previous works, the dependence of the coating adhesionon the surface properties of PP/oxidized wax blends wasrevealed [5,20]. The studies on surface roughness showed adecrease in PP/oxidized wax blends hence no effect of mechanicalinterlocking on adhesion mechanism. The acidic or basic behaviorof the surface was left unsaid in detecting the electrostatic chargeover a pH range. Besides, due to the lack of considerable differencein surface topography after coating application, the increase inadhesion by diffusion was not observed. Therefore, the increase inthe surface adhesion in PP/oxidized wax blends arises mainly bythe physical adsorption and to some extent by other structuralproperties such as substrate modulus [5,20]. The present workaims at correlating substrate physical and mechanical properties tothe coating adhesion. To this end, the contribution of near-surfacemodulus and surface free energy in coating adhesion is discussed.

2. Experimental

2.1. Materials

A homopolymer PP supplied by Navid Zar Chimi (Tehran, Iran)was used as polymer matrix. The oxidized PP wax (OPPW) wasprepared via thermal oxidation of neat PP wax. At the end ofprocedure, the acid value reached to 13.5 mg KOH/g and themolecular weight decreased to ca. 1900 g/mol [5]. Various PP/OPPWblends of 2%, 6% and 10 wt% of OPPW content were prepared in a60 mL internal mixer (Haake SYS 90, USA) at 190 1C for 10 min at amixing speed of 60 rpm. The oxygen content of the PP/OPPW blendswas determined by elemental analysis to screen the changes in theamount of polar oxygenated groups in the blends.

Test specimens of 2 mm thick were prepared using a hydraulicpress at 190 1C for 5 min. The samples were first cut and thenwiped gently with acetone prior to coating application. Thespecimens were dipped for 2 s into an acrylic-based coating, ageneral high solid VOC primer (285–60 VOC) from Glasurit BASF

Coatings AG (Muenster, Germany). A typical homogenous dry filmof 50 mm thickness was obtained at ambient temperature forabout 15 min. The coated samples were left for 7 days at 23 1Cand 50% RH prior to testing.

2.2. Methods

A Kruss G2/G40 (Hamburg, Germany) contact angle measuringsystem was used to measure the contact angle of the blends usingdistilled water, formamide and diiodomethane as probe liquids.The surface tension and its components for the probe test liquidsare given in Table 1. The contact angle measurements wereperformed at about 23 1C with a drop volume of 5 mL at theimmediate time of drop insertion. The resulting contact angleswere then used for calculating the surface free energy of variousPP/OPPW blends using OWRK approach and VOGC approaches.Five replicates of each sample were tested and the average valueswith standard deviations were reported.

The bulk mechanical properties of the polymer substrate wereevaluated at ambient temperature with a 20 kN universal testingmachine (SMT-20, Santam, Iran) at a crosshead speed of 10 mm/min. According to ASTM D 638 standard practice, dog bone speci-mens were prepared by injection molding at 210 1C. The elasticmoduli E of the samples were determined from the slope of theinitial elastic region i.e., below 2% elongation.

The viscoelastic response and the bulk mechanical propertiesof polymeric substrates were further studied by DMTA using aTriton 2000 analyzer (Nottinghamshire, UK) under air atmo-sphere. Samples (32�20�2 mm3) were prepared in a minimolder (Dynisco polymer test, USA) at 210 1C. For DMTA study,the coating films were cast on the glassy substrate at ambienttemperature. The test temperature was increased from �100 to150 1C at a constant rate of 5 1C/min and a sinusoidal frequency of1 Hz. In all specimens, the modulus in the glassy region wasrecorded at �100 1C.

The viscoelastic response of the outermost substrate layer wasinvestigated by a depth sensing indentation system using a modelHysitron Inc (Minneapolis, USA). The surface of the samples wasindented using a Berkovich diamond tip. The procedure was aload–hold–unload cycle with a loading time of 30 s followed by a10 s holding time and an unloading time of 30 s. Successive testswere performed with the maximum force between 500 and600 mN. The data were reproduced in load–displacement curves.

The coating adhesion strength was evaluated using direct pull-off test method according to ASTM D 5179 standard. Fig. 1illustrates comprehensive details of the pull-off adhesion test.The dollies were glued on the coating’s surface using a two-partAraldite 2015 (Huntsman Advanced Materials, Germany) andthen left for at least 24 h to fully cure (step a). A slot was madearound the dolly through the film to eliminate the effect ofperipheral coating failure (step b). In the 20 kN universal testingmachine, the special grip was fixed (step c) and then pulled theadhered aluminum dolly at a speed of 10 mm/min normal to thesurface until the coating film was detached from PP substrate(step d). At least seven replicates were performed for each sampleand the average values were reported.

Table 1Surface tension and its components of probe test liquids at 23 1C [6,7].

Probe liquid gB (dyn/cm) gA (dyn/cm) gAB (dyn/cm) gLW (dyn/cm) gd (dyn/cm) gP (dyn/cm) gtot (dyn/cm)

Water 25.5 25.5 51 21.8 21.8 51 72.8Diiodomethane 0 �0 0 50.8 50.8 0 50.8Formamide 39.6 2.28 19 39.2 39.5 18.7 58.2

A. Salimi et al. / International Journal of Adhesion & Adhesives 31 (2011) 220–225 221

Author's personal copy

3. Results and discussion

3.1. Surface free energy measurements

The surface free energies of various PP/OPPW blends weremeasured using two common approaches in thermodynamicadsorption theory of adhesion, i.e., the geometric mean and theacid–base model. Accordingly the contact angles measurementsof probe liquids and their respective components were used ingeometric mean model (OWRK approach) to calculate the totalsurface free energy. According to this model (Eq. (1)) the polarcomponent gP and the dispersive component gd of solid surfacefree energy can be obtained [2,6,7]. Two test liquids in Table 1(distilled water and diiodomethane) were used for this approach:

gLð1þcosyÞ ¼ 2ffiffiffiffiffiffiffiffiffiffigd

SgdL

qþ2

ffiffiffiffiffiffiffiffiffiffigP

SgPL

qð1Þ

where S and L represent solid and liquid phases, respectively. Thetotal surface free energy gS is a summation of its polar anddispersive components. The polarity of polymer can be simplydefined as the ratio of the polar component of surface free energyto the total surface free energy of polymer, i.e., XP ¼ ðgP

S=gSÞ [6].In Fig. 2, the results of total surface free energy and the

respective polar and dispersive components for the neat PP andits blends with OPPW are shown as a function of blend OPPW

wt%. The numbers at each column denote the polarity of thepolymer for each composition.

The variations of gS and its components, gPS and gd

S reveal agradual increase by increasing the blend OPPW content. However,the increase in gP

S is more obvious. The gdS may be attributed

mainly to the dispersive forces at the interface. In other words,the higher value of gd

S may provide greater van der Waalsinteractions at the polymer/coating interface. Similarly increasinggP

S may also result in more polar bonding at the polymer/coatinginterface [6]. Based on the previous findings [21], the polarcomponent originates mainly from polar groups available at thesurface. OPPW is a low molecular weight polar polypropylene,which contains various oxygen-based functionalities such ascarbonyl and hydroxyl groups. The oxygen content of the blendvaries from 0.0014 wt% for 2 wt% OPPW to 0.0071 wt% for 10 wt%OPPW. As shown in Fig. 1, the polarity Xp for the neat PP is almostzero, due to the absence of polar groups, and increases more orless by an increase in OPPW content. The gP

S is a function of thedensity of polar groups available at the surface. Therefore, theenhancement in gP

S is an indication of increasing surface polarity.The more polar surface provides better surface wettability topolar liquids, and hence a greater adhesion at the polymer/coating interfaces.

Another approach used for measuring the surface free energywas the acid–base model (VOGC approach). According to Eq. (2),the total surface free energy is a contribution of acid–basecomponent gAB

S and Lifshitz–van der Waals componentgLW

S [2,6,8]. However, the acid–base component itself is derivedfrom acidic gA and basic gB parts:

gLð1þcosyÞ ¼ 2ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffigLW

S gLWL

qþ2

ffiffiffiffiffiffiffiffiffiffigA

SgBL

qþ2

ffiffiffiffiffiffiffiffiffiffigB

SgAL

qð2Þ

In this approach, all probe liquids were utilized to calculate thetotal surface free energy and its components. In Fig. 3, the resultsof total surface free energy and its components are shown for PPand its blends with OPPW. The ratio in parenthesis denotes theacid/base ratio for each sample, i.e., gA=gB.

As shown in Fig. 3, both gAB and gLW components increase withhigher OPPW content of the blend. However, the trend for gAB ismore pronounced. In this approach, the gLW, like the dispersivecomponent in OWRK method, corresponds to the non-polarbonding at the interface. The gAB component plays a moreimportant role in the interfacial adhesion [8,10]. This involvesthe electron donor–acceptor interactions at the two inter-faces [7,8]. Moreover, in determining gAB the ratio of acid/basealso indicates the hydrophilicity of the surface. In fact, higherbasic component is an indication of more surface hydrophili-city [7]. As shown in Fig. 3, in all samples the ratio of acid/base ishigher than unity, i.e., gB is lower than gA. However, the ratiotends to unity by increasing the blend OPPW content. This

Fig. 2. The variation of substrate surface free energy (OWRK method) by increas-

ing the blend’s OPPW content. The polar and dispersive components of surface

free energy are shown individually. XP is the polarity of polymer.

Fig. 3. The trend of surface free energy components (VOGC method) by increasing

the blend’s OPPW content. The ratio in parenthesis denotes gA=gB.

Fig. 1. Successive steps in pull-off adhesion test (a) the dolly was adhered on the

coating surface using adhesive, (b) the slot was made around the dolly using

special knife, (c) the grip is ready to pull the adhered dolly after clamping to it and

(d) the dolly was detached vertically with the maximum recorded load.

A. Salimi et al. / International Journal of Adhesion & Adhesives 31 (2011) 220–225222

Author's personal copy

indicates the higher proportion of gB and hence more hydrophi-licity of the surface as the blend oxygen content increases.

Briefly in both approaches the results of surface free energyindicate better surface hydrophilicity as the OPPW increases.Therefore, it is expected that the surface adhesion propertyincreases as the oxygen content of the blend increases. However,other structural properties playing influence in coating adhesionshould be also considered. Accordingly, the near-surface and thebulk moduli of underlying substrate and their impact on coatingadhesion have to be evaluated individually.

3.2. Viscoelastic properties

The indentation load–displacement curves for the neat PP andits blend specimens are given in Fig. 4. The maximum penetrationforce was set at 600 and 500 mN for the neat PP and its blendswith OPPW, respectively. However, in loading curves (first part)and at the same penetration force, the blends showed higherpenetration depth (displacement) compared to the neat PPsample. Besides, the penetration depth increased as the oxygencontent of the blends increased. Based on the analysis proposedby Oliver–Pharr (O&P) [16,17], the reduced modulus is directlyrelated to the slope of the upper portion of unloading curve (thirdpart). In Fig. 4 the curves for OPPW blends show less steep slopefor unloading curves by increasing the OPPW content. Moreover,the area of irreversible work, between loading and unloadingcurves, increased as the OPPW increased. This indicated higherplasticity by the increase in OPPW content [17]. The near-surfacemoduli were then calculated according to O&P procedure and asshown in Fig. 6 for a comparative purpose.

Briefly the data of indentation test state that the near-surfacemodulus decreases as the OPPW content increases. Due to theincrease in population of OPPW polar chains, the surface crystal-linity deteriorates and hence near-surface modulus decreases [5].The polar groups of OPPW chains inhibit the ease of chainordering during crystallization and hence a diminishing effect ofOPPW on blend crystallinity was observed [5]. Consequently thelower near-surface modulus results in more penetration in thesurface as shown in Fig. 4.

The viscoelastic properties of the blends were also evaluatedby DMTA. The storage modulus and loss factor curves for PP andits blends with OPPW are shown in Fig. 5.

As shown in Fig. 5, the increase in wax content resulted inbroader damping peak and lower glass transition temperature Tg.The maximum at the peak of loss factor curves may be used fordetermining Tg. Similar to that of the neat PP, all blends exhibiteda single broad relaxation peak (one Tg) in their loss factor curves.This corresponds to a single phase matrix and a relatively goodcompatibility of PP/OPPW system up to 10 wt% of OPPW [5]. Infact, the single phase may arise from a complete incorporation ofthe polar OPPW chains into the non-polar PP chains. There is alsoa tendency for the relaxation peaks to broaden by increasing theOPPW content. The broad relaxation peaks corresponds to highertoughening behavior compared to the neat PP. The shorter chainsprovide better chain flow and hence higher toughening for thematrix compared to that of longer chains [5].

The storage modulus values are in agreement with the tough-ening. The elastic moduli in the glassy state (about �100 1C) canbe used for showing the changes in the modulus of bulk materials.Another method for evaluation of elastic modulus is through themeasurements made by UTM. The results of elastic moduli of thebulk material in addition to those of near-surface modulus areshown in Fig. 6.

As shown in Fig. 6, the near-surface modulus is decreased byan increase in OPPW content in the blends. The same behaviorswere observed for the elastic moduli of the bulk material obtainedby DMTA and UTM. Therefore, the trend of moduli is almost thesame at near-surface and bulk. The declining trend of both near-surface and bulk moduli can be related to the uniform distribu-tion of OPPW chains in the polymeric matrix and hence theirconsistent role on sample mechanical properties. However, theOPPW and PP chains are not fully compatible. The effect of

Fig. 4. Load–depth curves for neat PP and its blends as function of OPPW content.

A schematic illustration of the load–displacement curve is shown on top left

corner.

Fig. 5. The storage modulus and loss factor curves for (a) neat PP (b) blend

containing 6 wt% OPPW and (c) blend containing 10 wt% OPPW.

Fig. 6. The relationship between near-surface modulus (nano-indentation) and

moduli of the bulk material (DMTA and UTM) with OPPW content.

A. Salimi et al. / International Journal of Adhesion & Adhesives 31 (2011) 220–225 223

Author's personal copy

incompatible additive on the polymer matrix originates mainlyfrom the degree of chemical similarity (solubility parameter) andthe additive amount. Regarding to our previous work [5] for highamount of OPPW, the adhesion strength of coating layerdecreased because of migration of low molecular weight OPPWto the surface and formation a weak boundary layer [5]. There-fore, it can be concluded that OPPW and PP are compatible tosome extent and the amount of additive is the determinant factorin the outcomes of incompatibility, e.g., coating adhesion.

3.3. Correlating surface free energy and modulus to adhesion

In addition to the surface free energy, the results of surfacemodulus were taken into account to study its relevance to thecoating adhesion. The values of pull-off adhesion test wereplotted against the near-surface modulus, surface free energies(two approaches) and their combination, individually. The inter-relation between the adhesion, substrate modulus and substratesurface free energy measurements based on the OWRK and VOGCapproaches are shown in Figs. 7 and 8, respectively.

As shown in Fig. 7, the adhesion is proportional to both surfacefree energy and near-surface modulus. However, the adhesion

correlates better with the surface free energy (R2¼0.97) com-

pared to near-surface modulus (R2¼0.87). The R2 values were

calculated using a linear trend line by which uses a linearequation (y¼mx+h) to calculate the least squares fit for a line,where m is the slope and h is the intercept. The R2 valuedetermines the reliability of the trend and accuracy of theforecast.

In Fig. 7, the correlation for the square root of the product ofmodulus and surface free energy did not show a good fitness(R2¼0.17). This might imply that it cannot well correlate to the

Kendall model for the critical pull-off force. As such the criticalpull-off force Pc is proportional to the surface modulus Es andcritical Griffith fracture energy per unit area Gc between the diskand coating film [22]:

Pc¼2pESGca3

1�n2

� �0:5

ð3Þ

where a is the radius of disk and n is the Poisson ratio of coatingfilm. Based on Dupre’s work of adhesion, G is twice that of thesurface free energy of the newly created surfaces [21]. Hence,Eq. (3) shows the equality of surface free energy and Es to the Pc.In Fig. 7, it seems that Eq. (3) has been verified experimentally fora specific coating with constant process variables. In Fig. 8, theadhesion correlates similarly to the square root of the product ofmodulus and surface free energy (R2

¼0.87) and near-surfacemodulus (R2

¼0.84) or surface free energy (R2¼0.87). In

both Figs. 7 and 8, there are better correlations between adhesionand the surface free energy measured even by OWRK and/orVOGC methods compared with near-surface modulus. While, theadhesion correlates better with surface free energy using OWRKmethod (R2

¼0.97) than VOGC (R2¼0.87).

In current coating/substrate system, the degree of influence ofsubstrate modulus on the surface adhesion can be explained bythe substrate viscoelasticity and/or development of internal stressin the coating layer [12,13]. Due to lower modulus in substratecontaining OPPW and hence higher plasticity during coatingdebonding, higher value in adhesion strength was reached. Thepractical adhesion comprises the surface free energy term, Wa inaddition to an energy absorbing term during failure [9]. Inethylene–polypropylene (EP copolymer) with soft phases, theincrease in adhesion can be explained by the increased plasticdeformation in which there is an energy dissipating process [15].Even the formation of EP nodules and their depth location belowthe surface are important parameters controlling the magnitudeof adhesion [15].

Another aspect of substrate modulus may be in developing thestress in coating layer. Due to the environmental conditions andtheir intrinsic properties, most organic coatings are practicallyunder stress. Internal stress during film formation is of greatimportance in loss of coating adhesion [12,13]. Perrara describedthe effect of coating stress via increasing the elastic work [14].Based on Stoney’s original analysis on the mechanics involved infreely supported beam method, a direct relationship between theinternal stress s and underlying substrate modulus Es has beenshown in Eq. (4) [12]:

s¼ Est2s

ð1�nsÞ6rtcð4Þ

where ts, ns and r are the thickness, Poisson’s ratio and radius ofcurvature of the underlying substrate, respectively. In a coating/substrate system with constant coating thickness tc, the Es

denotes directly the stress development in coating layer. There-fore, it seems in a coating/substrate system, the modulus ofunderlying substrate is a dominant factor in developing coatingstress. The lower substrate modulus gives a coating with a lower

Fig. 7. Adhesion values as a function of surface free energy (OWRK method), near-

surface modulus and the square root of the product of near-surface modulus and

surface free energy.

Fig. 8. Adhesion values as a function of surface free energy (VOGC method), near-

surface modulus and the square root of the product of near-surface modulus and

surface free energy.

A. Salimi et al. / International Journal of Adhesion & Adhesives 31 (2011) 220–225224

Author's personal copy

internal stress s developed within the coating layer. In fact, itfacilitates the coating to relax most of its stress.

It should be noted that internal stress in coating layer relaxesmore quickly as the temperature is raised. In other words, coatinglayer relaxes most of its stored elastic energy at temperaturehigher than its glass transition temperature Tg. Therefore, thediscussion for the employed coating is proceeded to find the Tg.Fig. 9 shows the DMTA curves of coating layer.

The coating layer was examined at the test conditions as forthe blend substrates. According to Fig. 9, the Tg of the coatinglayer was determined to be at about 45 1C, which is close to thetest temperature. Therefore, the coating layer is able to relax mostof its stress prior to and during the pull-off test. In the case ofapplied coating layer, there is no influence on the coating storedelastic energy and hence on its adhesion strength. The discussionis in agreement with the findings in Figs. 7 and 8 in which it hasbeen shown that the role of surface free energy is more importantthan the near-surface modulus to develop surface adhesion.

4. Conclusion

The incorporation of polar OPPW in the PP matrix caused asubstantial increase in both surface free energy and coatingadhesion of the blends. The reason is probably due to thepresence of polar groups on short polymer chain oxidized waxeswhich are located underneath the surface. The adhesion promot-ing effect of the oxidized waxes is suggested to work via goodsurface wetting, specific electron donor/acceptor interactions atthe interfacial region of coating layer with the outermost OPPWchains at the surface. In both methods of surface free energymeasurements, the surface hydrophilicity increases by the con-tent of oxidized wax. The DMTA analysis for all blends demon-strates the single phase morphology indicative of a single phasestructure. The simultaneous loss in modulus in bulk and near-surface corresponds to uniform distribution of OPPW chains inpolymeric matrix.

The near-surface modulus was used to find out the significanceof modulus on coating adhesion. There is higher plasticity for thesubstrate containing OPPW and hence higher energy dissipation

and practical work of adhesion during coating debonding. It isfound that the surface free energy in both approaches fits betterto the surface adhesion compared with the near-surface modulus.However, the surface free energy measured by VOGC method andsubstrate modulus showed equivalent role on adhesion proper-ties. Besides, the closeness of Tg of coating layer to the testtemperature brings about relaxation in most of internal stressesof the coating layer.

References

[1] Chodak I, Novak I. Surface modification of polypropylene by additives. In:Karger-Kocsis J, editor. Polypropylene: A–Z reference. Dordrecht: KluwerAcademic Publishers; 1999. p. 790–3.

[2] Awaja F, Gilbert M, Kelly G, Fox B, Pigram PJ. Adhesion of polymers. ProgPolym Sci 2009;34:948–68.

[3] Hailat M, Xiao HX, Frisch KC. Studies on the modification of thermoplasticpolyolefins (TPOs)—Part I: effect of various modifiers on the adhesion of atwo-component (2 K) polyurethane coating to the modified TPOs. J Elasto-mers Plast 2000;32:194–210.

[4] Novak I, Krupa I, Luyt AS. Modification of the polarity and adhesive propertiesof polyolefins through blending with maleic anhydride grafted Fischer–Tropsch paraffin wax. J Appl Polym Sci 2006;100:3069–74.

[5] Salimi A, Mirabedini. M, Atai M, Mohseni M. Studies on the mechanicalproperties and practical coating adhesion on PP modified by oxidized wax.J Adhes Sci Technol 2010;24:1113–29.

[6] Shimizu RN, Demarquette NR. Evaluation of surface energy of solid polymersusing different models. J Appl Polym Sci 2000;76:1831–45.

[7] van Oss CJ. Use of combined Lifshitz–van der Waals and Lewis acid–baseapproaches in determining the apolar and polar contributions to surface andinterfacial tension and free energies. J Adhes Sci Technol 2002;16:669–77.

[8] Chehimi MM. Harnessing acid–base interactions to improve adhesion. In:Mittal KL, Pizzi A, editors. Adhesion promotion techniques, technologicalapplications. New York: Marcel Dekker, Inc.; 1999. [Chapter 2].

[9] Packham DE. Surface energy, surface topography and adhesion. Int J AdhesAdhes 2003;23:437–48.

[10] McCafferty E. Acid–base effects in polymer adhesion at metal surfaces.J Adhes Sci Technol 2002;16:239–55.

[11] Ryntz RA, McNeight A, Ford A. How molding conditions affect adhesion ofcoatings to TPOs. Plast Eng 1996:35–8.

[12] Francis LF, McCormick AV, Vaessen DM, Payne JA. Development and mea-surement of stress in polymer coatings. J Mater Sci 2002;37:4717–31.

[13] Pines M, De Deurwaerder H. Effect of coating stress on adherence and oncorrosion prevention. Prog Org Coat 2001;43:18–24.

[14] Kohl JG, Singer IL, Schwarzer N, Yu VY. Effect of bond coat modulus on thedurability of silicone duplex coatings. Prog Org Coat 2006;56:220–6.

[15] Tomasetti E, Legras R, Henri-Mazeaud B, Nysten B. Plastic deformation inpolypropylene/(ethylene–propylene) copolymer blend during paint debond-ing. Polymer 2000;41:6597–602.

[16] Kim W-S, Yun I-H, Lee J-J, Jung H-T. Evaluation of mechanical interlock effecton adhesion strength of polymer–metal interfaces using micro-patternedsurface topography. Int J Adhes Adhes 2010;30:408–17.

[17] van Landingham MR, Villarrubia JS, Guthrie WF, Meyers JF. Nanoindentationof polymers: an overview. Macromol Symp 2001;167:15–43.

[18] Lee G, Kang SK, Kwon D. Characterization of elastic modulus and work ofadhesion in elastomeric polymers using microinstrumented indentationtechnique. Mater Sci Eng A 2008;496:494–500.

[19] Gupta S, Carrillo F, Li C, Pruitt L, Puttlitz C. Adhesive forces significantly affectelastic modulus determination of soft polymeric materials in nanoindenta-tion. Mater Lett 2007;61:448–51.

[20] Salimi A, Mirabedini M, Atai M, Mohseni M. Probing for adhesion mechanismin oxidized PP wax/pp blends. In: Proceedings of the 43rd IUPAC worldpolymer congress, Glasgow, UK, Macro 2010.

[21] Salimi A, Atai M, Mirabedini M, Mohseni M. Study of thermo-oxidativereactions of polypropylene wax in molten state. J Appl Polym Sci2009;111:2703–10.

[22] Brady Jr RF. A fracture mechanical analysis of fouling release from nontoxicantifouling coatings. Prog Org Coat 2001;43:188–92.

Fig. 9. The storage modulus and loss factor curves for the acrylic coating.

A. Salimi et al. / International Journal of Adhesion & Adhesives 31 (2011) 220–225 225