Progress in Organic Coatings 77 (2014) 247 256

Contents lists available at ScienceDirect

Progress in Organic Coatings

jou rn al hom ep age: www.elsev ier .com

Furan r ctivStructu riz

G. Riveroa Ecomateriales sidad 7600 Mar del Pb Ciencia e Inge s (INTEJuan B. Justo 43c Electroqumic EMA), Justo 4302, 760

a r t i c l

Article history:Received 5 JunReceived in reAccepted 27 SAvailable onlin

Keywords:Furan resinsPhenolic resinsOrganic protective coatingsNanoindentation

ient agous s wors, toxform

traditional phenolic resin. Physicochemical characteristics including chemical structure, surface polarityand glass transition temperature were evaluated by means of Fourier transform infrared spectroscopy,contact angle measurements and dynamicmechanical analysis, respectively. Nanomechanical and nan-otribological properties were assessed by depth sensing indentation techniques. As well, the corrosionresistance of the furan coating was determined by potentiodynamic polarization tests. The obtainedresults validate the furan resin as a feasible alternative to phenolics to protect aluminum.

1. Introdu

Phenolichave excellreactivity wthey have btrial applicaadhesives aapplicationtogether wsion. For exelectrical inthus prevenmedium [4,regulationssince the 8applicationproduction,products co

CorresponE-mail add

lfasce@.mdp.

0300-9440/$ http://dx.doi.o 2013 Elsevier B.V. All rights reserved.

ction

resins, synthesized from phenol and formaldehyde,ent chemical and thermal resistance, compatibility andith other resins. Due to these desirable properties,een widely used for over 100 years in many indus-tions that range from molding compounds to laminates,nd protective coatings [13]. Particularly, the latter

takes advantage of their minimal thermal expansionith their high resistance to abrasion, wear and corro-ample, they are used as protective interior varnishes,sulators layers and anticorrosion coatings for metals,ting the direct contact of metallic substrates with the5]. However, due to its high toxicity, there are rigorous

concerning the reduction of formaldehyde emissions0s [613]. The main restriction is pointed to indoors, but formaldehyde release can also occurs in every

usage, storage, transportation and deposition stage ofntaining residual formaldehyde [7].

ding author. Tel.: +54 223 4816600; fax: +54 0223 4810046.resses: grivero@.mdp.edu.ar (G. Rivero),edu.ar (L.A. Fasce), lbmanfre@.mdp.edu.ar (L.B. Manfredi).

Following the current tendencies to a minimization ofpetroleum-based materials, furfural is an appealing option toreplace formaldehyde in the resins formulation. Furfural can beeasily obtained from agriculture wastes; it is harmless to the ozonelayer and is not as toxic as formaldehyde is. Novolak resins preparedfrom furfural, phenol and formaldehyde have been used duringdecades for molding compounds, and many commercial patentshave been registered [1417]. Despite the great efforts to elimi-nate formaldehyde from formulations, up to our knowledge, thereis only one recent commercial register for molding resins based onfurfuryl alcohol, glyoxal and urea, without formaldehyde [18].

The reaction between phenol and furfural to give a furan resinwithout using formaldehyde in the initial formulation has beenrecently studied in our group [19,20]. In this work, the formu-lated furan resin was proposed as a potential protective coating foraluminum in replacement of the traditional phenolic resins. Theanticorrosion function of the coating may be strongly related to itsmechanical performance, thus directly inuenced by its chemicalstructure. Hence, an extensive experimental analysis was carriedout to evaluate the efciency of the proposed coating.

In order to characterize the mechanical performance of thincoatings deposited onto substrates, nanoindentation combinedwith nanoscratch tests appear as the most appropriate and accuratetechniques [2124]. In a nanoindentation test, a tip of well-dened

see front matter 2013 Elsevier B.V. All rights reserved.rg/10.1016/j.porgcoat.2013.09.015esins as replacement of phenolic proteral, mechanical and functional charactea, L.A. Fasceb, S.M. Cerc, L.B. Manfredia,

Instituto de Investigaciones en Ciencia y Tecnologa de Materiales (INTEMA), Univerlata, Argentinaniera de Polmeros Instituto de Investigaciones en Ciencia y Tecnologa de Materiale02, 7600 Mar del Plata, Argentinaa y Corrosin Instituto de Investigaciones en Ciencia y Tecnologa de Materiales (INT0 Mar del Plata, Argentina

e i n f o

e 2012vised form 29 April 2013eptember 2013e 27 October 2013

a b s t r a c t

Phenolic coatings are usually a convenand corrosion. Furan resins are analoby furfural in their formulation. In thiand used as an aluminum coating. Thuderivative was used instead. The per/ locate /porgcoat

e coatings:ation

Nacional de Mar del Plata, Juan B. Justo 4302,

MA), Universidad Nacional de Mar del Plata,

Universidad Nacional de Mar del Plata, Juan B.

nd economical way to protect metallic materials against wearto phenolics, as they are obtained by replacing formaldehydek, a furan resin based on furfural and phenol was synthesizedic emissions of formaldehyde were avoided, while a biobasedance of the proposed resin was compared with the one of a

248 G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256

geometry is driven into the material surface while the appliedforce and tip displacement data are collected. From nanoindenta-tion data, near surface elastic and plastic properties can be easilydetermined using the widespread method of analysis proposedby Oliver ato characterequires sutip is movedload is applare recordemine the frcoatings an[28]. In adding/substrainterface oc

The corrby potentiodensity genmetal sampor defects tcorrosion dto evaluate systems [30

In this wicochemicaof furan andout. In addaluminum w

2. Experim

2.1. Materi

A resolformaldehynol and a fowere placedity stirrer, aat 9.0 withallowed to tralized witreached. A 1prepared.

A furan pfurfural (Flurefrigerant,reaction me40 wt/vol%.30 min in aature was mstructural cviously repo110 C and as a catalysprepared.

Differentypes of reously polishalumina sulms of diffusing the pr(50 and 100

A two-scoated alumperature to

at this temperature to allow the release of water and oligomer bub-bles. In the second step, the temperature was raised up to 180 C(for furan coatings) and 190 C (for phenolic coatings) at 1 C/minand kept for 30 min to complete the curing.

eaftefor fumbe

ethod

Visco viscntongurared a

Fouri evonal g

cyclen mod ev400

ith thing (

DynaA exing r. Pri

the r relea

glasorreerage

Contatic cod usete

re. Drfacewas r

Scann an

thick waryofrse thted ster inerat

pation etric

Nanooindted ng Prlysisof thdentteriand Pharr [25]. This method has been already appliedrize different polymeric lms with success [26,27] butitably designed experiments. In a nanoscracth test, the

over the surface while a constant or progressive normalied. The tangential force and the normal displacementd. Nanoscratch tests have been effectively used to deter-iction coefcient and scratch resistance of polymericd to study the near surface deformation mechanismsition, the adhesion strength can be evaluated for coat-te systems, provided that coating debonding along thecurs in response to load [26,29].osion protective quality of coatings can be characterizeddynamic polarization tests that measure the currenterated after applying a cyclic potential to the coatedle. An electrical response implies the existence of poreshat allow the passage of ions, thus evidencing certainamage. This technique has been successfully employedthe corrosion process of several metal/polymer coating33].ork, a deep study concerning the comparison of phys-l characteristics and surface nanomechanical behavior

phenolic resins deposited onto aluminum was carriedition, the corrosive protection of the furan coating toas evaluated.

ental

als and sample preparation

-type phenolic prepolymer was prepared with ade to phenol molar ratio equal to 1.3. Samples of phe-rmaldehyde aqueous solution (37%, wt/wt) (Cicarelli)

in a stainless steel reactor equipped with a low veloc- thermometer and a reux condenser. The pH was kept

a solution of NaOH 40% (wt/wt) and the mixture wasreact for 2 h at 90 C. Thereafter, the mixture was neu-h a solution of boric acid until a pH value of 6.87.0 was0 wt% solution of phenolic prepolymer in acetone was

repolymer was synthesized from phenol (Anedra) andka). Phenol was molten into a reactor supplied with a

a thermometer and constant mechanical stirring. Thedia was adjusted using an aqueous solution of K2CO3

After heating up to 135 C, furfural was dropped during furfural to phenol molar ratio equal to 1. The temper-

aintained at 135 C for 4 h. A detailed chemical andharacterization of the furan prepolymer has been pre-rted [20]. The furan prepolymer was then heated up to

12 wt% of hexamethylenetetramine (HMTA) was addedt. A 10 wt% solution of furan prepolymer in acetone was

t samples of aluminum (6063-TG) coated with bothsins were made. The aluminum substrate was previ-ed with increasing grades of silicon carbide paper and

spensions with particle sizes up to 0.05 m. Polymericerent thickness were prepared by dip-coating processepolymer solutions and two different withdrawal rates

cm/min). In all cases, the immersion time was 4 s.tep thermal curing cycle was applied to both types ofinum plates. They were rst heated from room tem-

120 C at 1 C/min and the samples were kept for 30 min

HerF-100 The nu

2.2. M

2.2.1. The

in an Aa conmeasu

2.2.2. The

functiocuringmissioscanneof 600ized wstretch[34].

2.2.3. DM

a heatof 1 Hzcuringsilicon

Theature cThe av

2.2.4. Sta

methoGoniomSoftwathe sudrops

2.2.5. SEM

actual SEM

were cto expo

Coadiamewas op(0.2 minspeca geom

2.2.6. Nan

on coaScanni

Anaerties nanoining mar coated aluminum samples are identied as F-50 andran coatings and P-50 and P-100 for phenolic coatings.rs indicate the dipping withdrawal rate.

s

sity measurementsosity of the 10 wt% prepolymer solutions was measured

Paar Physica Rheometer (MCR 301-CC27) at 20 C, withtion of concentric cones. The stationary viscosity wasmong 1 and 100 rpm using 20 ml of solution sample.

er transform infrared analysis (FTIR)lution of characteristic bands associated to differentroups was followed by FTIR analysis through the whole. Spectra were obtained in a Mattson Genesis II, trans-de, equipped with a heating furnace. Samples were

ery 10 C from room temperature to 180 C in the range0 cm1. For comparison purposes, spectra were normal-e intensity of the band assigned to the C C benzene ring1595 cm1) whose value is expected to remain constant

mic mechanical analysis (DMA)periments were carried out in a Perkin Elmer DMA 7 atate of 10 C/min from 15 to 300 C and at a frequencysmatic specimens were cut from plaques obtained byesins between two glass slides previously treated withse agent (Siliar S.A., Argentina).s transition temperature (Tg) was taken as the temper-sponding to the maximum displayed in the tan curve.

results of three replicas were reported.

ct angle measurementsntact angle determination was made by the sessile droping a Ram Hart model 500 Advanced Contact Angler/Tensiometer equipped with DROPimage Advancedrops of 5 l of doubly distilled water were placed ontos of coated samples. The average contact angle of sixeported for each coating.

ing electronic microscopy (SEM) and Calotest

d Calotest experiments were used to determine theness of furan and phenolic coatings.s performed in a JEOL JSM-6460LV equipment. Samplesactured and coated with a 300 A gold layer previouslyeir lateral side. A typical micrograph is shown in Fig. 1.amples were ground by a rotating steel ball of 30 mm

a Calotest Compact unit from CSM Instruments. Ited at 2500 rpm for 2 min with an abrasive suspensionrticles). Coating thickness was determined by opticalof the abraded coating and substrate section applyingal relationship [35].

indentation and nanoscratch testsentation and nanoscratch experiments were carried outsamples in a Triboindenter Hysitron equipped with aobe Microscope module (SPM).

procedures for determining reliable mechanical prop-in polymeric lms deposited onto metal substrates byation are not yet standardized as for other engineer-ls. For the sake of simplicity and aiming to compare

G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256 249

Fig. 1. Coating thickness determination by SEM analysis in F-100 sample.

nanomechanical properties of furan and phenolic lms, experi-ments were accurately designed to apply the simple Oliver andPharr approach [25,36].

Nanoindentation tests were performed under load control con-ditions using a diamond Berkovich tip. Maximum loads (Pmax) werevaried from 0.1 to 9 mN to obtain properties at different indentationdepths aiming to analyze the possible inuence of the aluminumsubstrate [37,38].

For eaching period oand unloading curve [3using the st

Using thtic modulusfunction of to contact sof the mateated [27,41]slope of the

Two types of nanoscratch experiments were performed. In therst one, a diamond Berkovich tip was used to obtain the frictioncoefcient. A constant load of 100 N, a sliding speed of 0.33 m/s,and a lateral displacement of 10 m were applied. For each sample,10 experiments were carried out. The apparent friction coefcient() was calculated as the ratio of the measured tangential (Fx) andnormal (Fz) forces. The second type was intended to analyze the sur-face damage behavior during scratch deformation of furan coatings.A diamond spherical tip with 2 m radius of curvature was slid ata speed of 0.33 m/s while the load was linearly increased up to9 mN, which was the maximum force experimentally available.

2.2.7. Potentiodynamic polarizationPotentiodynamic polarization tests were carried out with a

standard three-electrode system to quantify the anticorrosive pro-tection of the furan coating. The specimens were connected to thealuminum either coated or uncoated as working electrode, a pureplatinum wire was used as the counter electrode and standardcalomel electrode (SCE) was used as the reference electrode. Poten-tiodynamic scans were performed in NaCl solution (0.15 mol/L)from 0.8 to 0.5 V at a scan rate of 1 mV/s and backwards. The polar-ization curves were measured in a Gamry Ref 600 electrochemicalunit (Gamry Instrument, USA) initially and, after 12 and 30 days ofimmersion in the solution at 20 C.

3. Results and discussion

3.1. Physicochemical characterization

syndehyion rs areatizeracteutedal stnds cross

rang loading condition, 5 indentations were made. A hold-f 15 s was applied at maximum load between loading

ing stages to minimize the creep effect on the unload-9,40]. Tip displacement was corrected by thermal driftandard Hysitron routine.e approach outlined by Oliver and Pharr, reduced elas-

(Er) and indentation hardness (H) were calculated as aindentation depth [25,36]. As well, the maximum loadtiffness-squared parameter (P/S2), which is a measurerial resistance to permanent deformation, was evalu-. The contact stiffness (S) was calculated from the initial

unloading curve.

Theformaldensatbridgeschem

ChasubstitchemicThe bato the lengthFig. 2. Scheme showing the condensation reactions that take placethesis of phenolic resins starts with the addition ofde to the phenol ortho and para positions. Then, con-eactions take place during curing in which methylene

formed and water and formaldehyde are released, asd in Fig. 2.ristic infrared spectra wavelengths corresponding to

benzene rings as well as methylene bridges in theructure of phenolic resins are well documented [4244].corresponding to methylene bridges, directly relatedlinking density, appear in the 14001500 cm1 wave-e of the infrared spectra, and their specic position

during curing of phenolic resins.

250 G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256

Fig. 3. Schem

can be relaparapara

lene bridgein furan rescrosslinked

As expespectra dueferences wein the furanbridges wassistently wiof bridges fof op andnolic resin, However, thresulted sim

The valuulus in the rfrom DMA values of bsimilar cheto crosslinkvery similar

The contples are incle showing the chemical structure of (a) phenolic and (b) furan resins.

ted to a particular position in the phenol ring; i.e.(pp), orthoortho(oo) and orthopara (op) methy-s, respectively [34]. Similarly, CH bridges are formedins during their curing reactions. The resulting highly

structure of both resins are shown in Fig. 3 [4,45].cted, furan and phenolic resins displayed similar FTIR

to their analogous chemical structure though slight dif-re observed. The three types of bridges were assigned

resin spectra (Fig. 4a), but the band associated with oo

not detected in the phenolic resin spectra (Fig. 4b) con-th previous reports [46,47]. The variation in the amountormed during curing is compared in Fig. 5. The amount

pp bridges was larger in the completely cured phe-but additional oo bridges appeared in the furan one.e nal amount of total bridges contained in both resinsilar.

es of glass transition temperature (Tg) and storage mod-ubbery state (Erubber) of phenolic and furan resins arisenexperiments are reported in Table 1. Tg and modulusoth resins were practically identical, evidencing theirmical structure. Moreover, as Erubber is directly related

density [48], the total amount of bridges appeared to be in both resins in agreement with FTIR results (Fig. 5).act angle measurements for both types of coated sam-uded in Table 1. The lower value measured for the furan

Fig. 4. Region of the FTIR spectra showing the characteristic bands correspondingto bridges in (a) furan and (b) phenolic resin. Some spectra obtained at differenttemperatures during curing are shown.

Fig. 5. Evolution of the bridges amount in phenolic and furan resins during curing.

G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256 251

Table 1Dynamic-mechanical properties and contact angle of the resins.

Resin Erubber (GPa) Tg (C) Contact angle (degree)

Phenolic 0.25 255 74.9 0.8Furan 0.24 252 66.8 2.5

Table 2Coating thickness values of furan and phenolic samples.

Sample P-50 P-100 F-50 F-100

Coating thickness(m)

0.24 0.08 0.29 0.08 2.14 0.10 2.62 0.04

coating displayed its greater hydrophilic character. The higherpolarity is a result of the polar groups in the furan coating surface,but it is also related with the interaction between these groupsand the hydroxyls of the aluminum substrate, as it will be laterdiscussed.

3.2. Coating thickness

Actual thickness values of furan and phenolic coatings mea-sured by SEM and veried by Calotest experiments are reported inTable 2. Phenolic lms were notably thinner than furan ones whenprepared uwithdrawalthe withdra

It is welseveral forcdrag [49,50the substratdrawal speelead to thic

The visclic prepolymrespectivelynesses can prepolymer

The withto control tis usually ewith x valuplotted acco

Fig. 6. Relatioand furan resin

oad vs. indentation depth curves for different maximum applied loads for (a)d (b) P-100 systems. The indentation response of the aluminum substrateed for comparison.

of x about 0.3 was observed, evidencing the same thicknessence with withdrawal speed. Hence, as differences in thick-e mainly due to the viscosity, similar values for both coatingse achieved by adjusting the solution concentration.

nomechanical properties

ical loaddepth curves obtained at different maximum loads for two of the studied systems are shown together

he substrate response in Fig. 7. The mechanical responsecoated systems was completely different from the one ofbstrate. While aluminum deformed plastically, the coat-s able to elastically recover deformation after the tip wased from the surface. Properties were determined as a func-

indentation depth and analyzed considering the inuenceunderlying substrate. The time-dependence of the indenta-sponse of the systems was clearly evident by the increase insing the same prepolymer solution concentration and rate. As expected, the coating thickness decreased aswal rate diminished for each type of resin.l known that the dip coating process is controlled byes such as gravitational, inertia, capillary and viscous]. The latter force moves the liquid solution upward withe and it is proportional to the liquid viscosity and with-d. So, larger solution viscosities and withdrawal speedsker coatings.osities of the 10% (w/w) solutions of furan and pheno-ers in acetone were 17.93 0.24 cP and 3.95 0.12 cP,. So, the great difference in the resulting coating thick-be directly related to the dissimilar viscosity of their

solutions.drawal speed, u0, is the most common parameter usedhe lm thickness, t. The relationship between t and u0xpressed as a power law mathematical form (t u0x)es ranging from 0 to 1 [51]. Fig. 6 shows thickness datarding to the mentioned correlation. For both coatings

Fig. 7. LF-100 anis includ

a valuedependness arcould b

3.3. Na

Typappliedwith tof the the suing waremovtion ofof the tion renship between coating thickness and withdrawal rate for phenolics. Additional data at 25 cm/min are also included.

indentationtion creep).curves pracchanged wcoatings (i.minum resploaddeptherally attribTherefore, aluminum s depth during the holding period at peak load (indenta- For the thickest coatings (i.e. F-100 in Fig. 7a), loadingtically overlapped while the shape of unloading curvesith increasing maximum applied load. For the thinnere. P-100 in Fig. 7b), the curves resembled the alu-onse especially at high indentation loads. For all cases,

curves did not exhibit discontinuities, which are gen-uted to failure events like coating detachment [5254].the adhesion of both furan and phenolic lms to theubstrate appeared to be very good.

252 G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256

Fig. 8. Contaccoatings on alu

The conpenetrationexpected totic modulusdeviation frtation deptmeasured iminum sub

Reduceddentation thickness-npendently dthe Er valuetation deptcombines thlying substrcoating [22iments for difference i

For furavery low insic elastic mdeterminedcompliant cthan 1 m,at depths lovalues of futhickness a

On the ccase of phe(about 0.2 range resulreliable indgenerally dtact area esvibrations. phenolic coant coatingconsiders thapparent m

E = Ec + (E

where E* reEs are the i

Reduced elastic modulus values as a function of thickness normalizedion depth for (a) furan coatings and (b) phenolic coatings deposited ontom substrate.

actual coating thickness and is a tting parameter. If Esn, the model tting provides the values of Ec and . The

matical form of Eq. (1) predicts a plateau value for low con-pths. Fitted model parameters are shown in Table 3. The eter increased while decreasing the coating thickness indi-a not straightforward dependence of apparent modulus witht stiffness vs. indentation depth for furan coatings and phenolicminum.

tact stiffness values, S, are plotted against maximum depth in Fig. 8. For a homogeneous material, S is

increase linearly with indentation depth since elas- does not vary with depth [38]. Data showed a positiveom linearity, which increased with increasing inden-h or reducing coating thickness. This means that thendentation response was inuenced by the stiff alu-strate.

elastic modulus values, Er, obtained from nanoin-experiments are shown in Fig. 9 as a function oformalized depth. A value of 72.5 2.8 GPa was inde-etermined for the aluminum substrate. As expected,

s of the coating/substrate systems increased with inden-h. These values represent an apparent property whiche mechanical properties of the coating and the under-ate, and are not an elastic modulus prole through the]. The dissimilar hmax/t range achieved in the exper-furan and phenolic coatings was due to the large

n their thicknesses (Table 2).n coatings, a plateau value was clearly observed atdentation depths (Fig. 9a). This value was the intrin-odulus of the furan coating. It coincided with the value

by applying the 10% rule, which establishes that foroating/stiff substrate systems and for coatings thicker

the elastic modulus of the coating can be evaluatedwer than 1/10 of the coating thickness [25,55]. The Er

Fig. 9. indentataluminu

is the is knowmathetact deparamcating ran coatings turned out to be independent on coatingnd near 10 GPa (Table 3).ontrary, a plateau value could not be observed for thenolic coatings (Fig. 9b). Because coatings were too thinm), the minimum experimentally achievable hmax/t

ted higher than 1/10 of the coating thickness. Obtainingentation curves at very low displacements (

G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256 253

Table 3Intrinsic nanomechanical properties of furan and phenolic lms.

Property Method F-50 F-100 P-50 P-100

Coating elasticmodulus (GPa)

10% rule 9.8 0.7 9.5 0.6 (Ec) Eq. (1) 10.2 0.1 (R2 = 0.91) 9.7 0.1 (R2 = 0.87) 12.4 1.2 (R2 = 0.93) 10.5 1.2 (R2 = 0.92)

Coating hardness (GPa) Average value 0.63 0.03 0.62 0.05 (Hc) Eq. (2) 0.57 0.04 (R2 = 0.82) 0.59 0.02 (R2 = 0.89)

hardness (Table 3). The intrinsic hardness of the furan coating wasabout a half of the aluminum hardness.

On the other hand, H values increased with increasing thickness-normalized depth for the case of phenolic coatings (Fig. 11b).Moreover, it is observed that for hmax/t larger than 1, H dataapproached the value of the aluminum (1.15 0.05 GPa). Aiming toobtain the intrinsic hardness of phenolic coatings, a simple modelproposed by Bhattacharya and Nix [59] was applied. The modelconsiders the combined effect of substrate and coating propertieson the apparent hardness as a function of indentation depth, as:

H = Hs + (Hc Hs) exp(

n (

h

t

)n)(2)

where H* represents the apparent hardness, Hc and Hs are theintrinsic properties of coating and substrate, is a tting parameterand n is equal to 2 for the case of soft coatings/hard substrates sys-tems [23]. Tof the phenthe furan co

The valuPmax/S2, i.e. plotted as aboth types oparameter ever, it cleamore resistThis result elastic and aluminum. ulus but ona given inda larger amto a lower ptypes of coaing that the

Fig. 10. ReducSolid lines sho

similar. Consistently, the H/Er2 ratio which is proportional to thePmax/S2 parameter, was almost identical for both types of coatings(Table 3).

3.4. Nanoscratch behavior

The average values of the apparent friction coefcient, app,measured at low loads were 0.35 and 0.38 for furan and phenoliccoatings, respectively. In these experiments, the scratch grooveswere imperceptible indicating that deformation was completelyrecovered after load removal, i.e. an elastic deformation modeprevailed. The app values reected the adhesion characteristicsbetween the coating and the diamond tip since the principal fric-tion mechanism under elastic contact conditions is adhesion. Thehe tted Hc values are reported in Table 3. The Hc valueolic coating was very close to the H value displayed byatings.es of the ratio of maximum load to stiffness squared,the materials resistance to plastic deformation [54], are

function of thickness-normalized depth in Fig. 12 forf coatings. Due to the inuence of the substrate, Pmax/S2

decreased as the indentation depth increased. How-rly emerged that furan and phenolic coatings resultedant to plastic deformation than aluminum substrate.could be explained by the benecial combination ofplastic properties of polymeric coatings in relation toPolymeric coatings showed about a 7-fold lower mod-ly 2-fold lower hardness than the substrate. Hence, atentation depth, the coating was able to accommodateount of elastic deformation than the substrate, leadingermanent damage. In addition, Pmax/S2 values of bothtings overlapped for the lowest hmax/t range, indicat-

mechanical protection conferred by both coatings wased elastic modulus as a function of thickness to contact depth ratio.w model ttings according to Eq. (1). The parameter is reported.

Fig. 11. Hardnfor (a) furan coThe dashed liness values as a function of thickness normalized indentation depthatings and (b) phenolic coatings deposited onto aluminum substrate.e represents the hardness of the aluminum substrate.

254 G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256

Fig. 12. Maximum load to the stiffness squared values as a function of thicknessnormalized indentation depth for furan and phenolic coatings. The dashed linerepresents the value determined for the aluminum substrate alone.

slightly lower app of the furan resin was consistent with its higherpolarity.

The damexperimentobtained foslope in thwas observform. The athe normalmation mescratch grocoating deftrack as wecess causedthe tip, thuwell, in SPMnation, demcoating andof a compliaif the shear of the syst

Fig. 13. Normplacement meloading.

Fig. 14. SPM image of the scratch groove left on the furan coating surface of F-100sample after ramping normal loading.

[62oatinesumnismss, wlic reiffers off

rrosi

ults sion um r imd a she coat reponsnces. For the coated sample, the current density decreasedage mechanism was investigated through nanoscratchs under increasing normal load. A typical responser furan coatings is shown in Fig. 13. An almost constante normal displacement vs. lateral displacement curveed indicating that the mechanical response was uni-pparent friction coefcient increased with increasing

force (or scratch distance) consistently with a defor-chanism of ductile ploughing [60]. SPM images of theoves revealed that during the sliding process the furanormed plastically and accumulated at both sides of thell as ahead of the crack tip (Fig. 14). The scratch pro-

shear-dominant stresses in the immediate vicinity ofs promoting shear yielding of the furan resin [61]. As

images there were no evidences of coating delami-onstrating that the adhesion strength between furan

aluminum was excellent. It is known that for the casent coating on a stiff substrate, delamination could occurstress along the interface exceeds the adhesion strengthem because the shear stress tends to dislocate the

coatingnolic c

In rmechahardnephenois no dcoating

3.5. Co

Resimmeralumin

Afteshoweity of tlm thcal resdiffereal displacement and friction coefcient as a function of lateral dis-asured for F-100 sample in Nanoscratch tests under ramping normal Fig. 15. Typic

aluminum sam]. A similar scratch behavior was observed for the phe-g.e, nanoscratch tests showed that coating deformation

was dominated by the yield stress. As the indentationhich is proportional to the yield stress, of furan and

sins was practically the same, it can be stated that thereence in the mechanical protection that both types ofered to the aluminum substrate.

on behavior

of potentiodynamic polarization tests at differenttimes are shown in Fig. 15 for uncoated and furan coatedspecimens.mersion in the NaCl solution, the uncoated aluminumtrong diminution of the current density in the vicin-rrosion potential, due to the development of a passivemained stable in the assayed period. The electrochemi-e of uncoated and furan coated specimens showed clearal potentiodynamic polarization curves of uncoated and furan coatedples.

G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256 255

Fig. 16. Changted aluminum(longitudinal sAl OH), with time is also inc

approximataluminum. substrate efollowed thformation oimproved w

As expectiodynamicnor furan cosuch defect

The gooaluminum ity among hydroxylateminum is ispontaneoulayer causinThe presencsamples imThere was oistics bandsexposure timthat the chtered durin

4. Conclus

A deep ccal and alumcompare thpolymer an

The coaconcentratithe large di

Both respractically tAs well, it wsurface nanequivalent.

The furaphenolic on

reected in a slight difference in their apparent friction coefcient(app).

Both types of coatings showed the same mechanical protectioninum. The scratch deformation behavior was dominated

yieldo evian coion be effeminucts i

lackon of

bet/hydas p

onal ion.

wled

horsnes al de

nces

Biederlysedogy 39Biedersol/ph

39 (2J. Songhenohnoloardzidizatiturialt streic resimaldeited Stmaldeanizaulatio002/9ent (hang,aldee of the relative amount of the bands corresponding to the hydroxyla- species measured at the wavelenghts: () 670 (stretch Al O), () 738tretch Al O), () 763 (vibrations Al O) y () 830 cm1(vibrationsimmersion time. Evolution of furan bridges amount during immersionluded.

ely one order of magnitude with respect to the uncoatedThis fact can be attributed to the diminution of thexposed area. However, after a long-time immersion, ite same trend as the uncoated material because of thef the passive lm, and the corrosion protection wasith the immersion time.ted, local corrosion spots were formed after the poten-

destructive tests. Nevertheless, neither delaminationating detachment was evidenced in the surrounding ofs.d adhesion between the organic furan coating and thesubstrate could be attributed to the excellent afn-the hydrophilic groups of the polymer (Fig. 3) and ad aluminum surface. It is well known that when alu-n contact with air a thin layer of aluminum oxide issly formed and then, ambient water is absorbed on thatg the hydroxylation of the aluminum oxide surface [63].e of such hydroxylated aluminum species in the coatedmersed in NaCl solution was veried by FTIR analysis.bserved an increase in the intensity of the character-

to alumby thewith n

Furcorrospositivsic aluor defe

Theadhesiafnityoxides

It wtraditicorros

Ackno

AuttigacioNacion

Refere

[1] M. ananol

[2] M. creogy

[3] H.-of pTec

[4] A. Gdar

[5] A. Sjoinnol

[6] ForUn

[7] ForOrg

[8] Reg[9] D. 2

liam[10] J. C

form

associated with Al O, Al OH [6367] with increasinge as shown in Fig. 16. On the other hand, it was probed

emical crosslinks of the furan coating remained unal-g the whole corrosion test, as demonstrated in Fig. 16.

ions

haracterization including physico-chemical, mechani-inum corrosion protection was performed aiming to

e performance of coatings based on a proposed furand a traditional phenolic resin.tings thickness obtained using the same prepolymeron solutions and dipping rates greatly differed due tosparity in furan and phenolic solutions viscosities.ins exhibited similar chemical structures and hencehe same dynamicalmechanical properties (Tg and E).as proved that, despite the difference in thickness, theiromechanical elastic (Ec) and plastic properties (H) are

n resin showed a greater hydrophilic character than thee, related to its larger content of polar groups. This was

1016.[11] G.E. Mye

Hamel (Ein ManufUSA, 198

[12] T. Salthamment, Ch

[13] M. SmidformaldeForestry

[14] X. Jie, Y. Jtional, 20

[15] H. Liu, B. (A), Inter

[16] Q. Pu, The[17] M. Toshik

131236, J[18] L.J. Min,

Composit2011.

[19] G. Riveroand its na

[20] G. Riveronanocom117 (201

[21] B. Bhushalms, Int stress of the polymers (ductile ploughing mechanism)dences of coating delamination.atings were able to efciently protect aluminum fromy conferring a stable barrier. In addition, a combiningct was displayed under severe conditions, as the intrin-m passive lm may be capable to block eventual poresn the polymeric lm.

of signs of coating detachment conrmed the excellent the coating with the substrate, probably due to the greatween hydrophilic groups of the furan polymer and theroxides evidenced in the aluminum surface.roved that the proposed furan resin could replace thephenolic resin to protect aluminum against scratch and

gements

gratefully acknowledge the Consejo Nacional de Inves-Cientcas y Tcnicas (PIP0014) and the Agencia

Promocin Cientca y Tecnolgica (PICT0682).

mann, K. Grob, Phenolic resins for can coatings: I. Phenol-based resole by GCMS, GC GC, NPLCGC and SEC, LWT Food Science and Tech-

(2006) 633646.mann, K. Grob, Phenolic resins for can coatings: II. Resoles based onenol mixtures or tert. butyl phenol, LWT Food Science and Technol-006) 647659., Z.-Z. Zhang, Study on the tribological and hydrophobic behaviors

lic coatings reinforced with PFW, PTFE and FEP, Surface and Coatingsgy 201 (2006) 10371044.ella, L. Pilato, A. Knop, Phenolic Resins: Chemistry, Applications, Stan-on, Safety and Ecology, Springer-Verlag, Berlin, Heidelberg, 2000.e, A. Vzquez, A. Cisilino, L.B. Manfredi, Enhancement of the adhesivength of the epoxyamine system via the addition of a resole-type phe-n, International Journal of Adhesion and Adhesives 27 (2007) 156164.hyde Standards for Composite Wood Products, Act VI, Congress of theates of America, WA, USA, 2010.hyde, Concise International Chemical Assessment, World Healthtion (WHO), Geneva, 2002.n CE No. 3093/94, in: E.C. Commission (Ed.) 2001/58/CE (2001).5/EC, Restriction of hazardous substances (RoHS), in: European Par-2000).

Z. Guo, R. Fortmann, H. Lao, Characterization and reduction ofhyde emissions from a low-VOC latex paint, Indoor Air 12 (2002)

rs, Advances in methods to reduce formaldehyde emission, in: M.P.d.), Composite Board Products for Furniture and Cabinets: Innovationsacture and Utilization, Forest Products Research Society, New York,9, pp. 5864.mer, S. Mentese, R. Marutzky, Formaldehyde in the indoor environ-

emical Reviews 110 (2010) 25362572.riakov, J. Sedliacik, J. Matyasovsky, P. Duchovic, Reduction ofhyde emission from plywood bonded with modied UF adhesive,and Wood Technology 76 (2011) 4448.un, Methd for Preparing Hot Box Furan Resin, 102153719A, Interna-11.Zhang, Preparation Method of Furan Resin for Casting, CN102161741national, 2011.rmal cracking resistant hard sand, B22C1/22, International, 2011.i, Y. Akira, J. Takashi, Binding Composition for Foundry Sand, 2011-apan, 2011.Y.J. Han, P.S. Hun, Furan Resin Composition, Furan Resin-Hardenerion and Foundry Sand Molds of the Same, KR20110049527 (A), Korea,

, V. Pettarin, A. Vzquez, L.B. Manfredi, Curing kinetics of a furan resinnocomposites, Thermochimica Acta 516 (2011) 7987., A. Vzquez, L.B. Manfredi, Synthesis and characterization ofposites based on a furan resin, Journal of Applied Polymer Science0) 16671673.n, X. Li, Nanomechanical characterisation of solid surfaces and thin

ernational Materials Reviews 48 (2003) 125164.

256 G. Rivero et al. / Progress in Organic Coatings 77 (2014) 247 256

[22] S. Bull, Nanoindentation of coatings: topical review, Journal of Physics D:Applied Physics 38 (2005) 393413.

[23] J. Malzbender, J.M.J. den Toonder, A.R. Balkenende, G. de With, Measuringmechanical properties of coatings: a methodology applied to nano-particle-lled solgel coatings on glass, Materials Science and Engineering R: Reports36 (2002) 47103.

[24] W. Shen, J. Sun, Z. Liu, W. Mao, J.D. Nordstrom, P.D. Ziemer, F.N. Jones, Meth-ods for studying the mechanical and tribological properties of hard and softcoatings with a nano-indenter, JCT Research 1 (2004) 117125.

[25] W.C. Oliver, G.M. Pharr, An improved technique for determining hardness andelastic modulus using load and displacement sensing indentation experiments,Journal of Materials Research 7 (1992) 15621584.

[26] R.A. DiFelice, J.G. Dillard, D. Yang, Chemical and nanomechanical properties ofplasma-polymerized acetylene on titanium and silicon, International Journalof Adhesion and Adhesives 25 (2005) 342351.

[27] L.A. Fasce, V. Costamagna, V. Pettarin, M. Strumia, P. Frontini, Mechanicalbehaviour of surface modied polypropylene lms with acrylic acid as graftingagent, Express Polymer Letters 2 (2008) 779790.

[28] R.R. Thridandapani, A. Mudaliar, Q. Yuan, R.D.K. Misra, Near surface defor-mation associated with the scratch in polypropyleneclay nanocomposite: amicroscopic study, Materials Science and Engineering: A 418 (2006) 292302.

[29] S. Bull, E. Berasetegui, An overview of the potential of quantitative coatingadhesion measurement by scratch testing, Tribology International 39 (2006)99114.

[30] V.J. Gelling, M.M. Wiest, D.E. Tallman, G.P. Bierwagen, G.G. Wallace,Electroactive-conducting polymers for corrosion control: 4. Studies of poly(3-octyl pyrrole) and poly(3-octadecyl pyrrole) on aluminum 2024-T3 alloy,Progress

[31] J. Ou, J. Wcorrosionon alumin

[32] S.S. Patharesin by sminum a

[33] M. Shabaof polyan3004 and3035304

[34] L.B. MaStructurehyde/phe

[35] SCM-Inst[36] ISO14577

and Mate[37] E. Le Bour

ization, V[38] R. Saha, W

mechanic[39] K. Geng, F

polymeri[40] A.H.W. N

indentati[41] D.L. Joslin

depth-se123126.

[42] T. Holopatative anJournal o

[43] L.B. Manfepoxy-phcured lm

[44] C. Rockniture of ph28 (2003

[45] K.J. Zeitsch, The Chemistry and Technology of Furfural and its Many By-Products, 1st ed., Elsevier Science, Amsterdam, The Netherlands, 2000.

[46] M.-F. Grenier-Loustalot, S. Larroque, D. Grande, P. Grenier, D. Bedel, Pheno-lic resins: 2. Inuence of catalyst type on reaction mechanisms and kinetics,Polymer 37 (1996) 13631369.

[47] M.-F. Grenier-Loustalot, S. Larroque, P. Grenier, J.-P. Leca, D. Bedel, Phenolicresins: 1. Mechanisms and kinetics of phenol and of the rst polycondensatestowards formaldehyde in solution, Polymer 35 (1994) 30463054.

[48] J. Pascault, H. Sautereau, R. Williams (Eds.), Thermosetting Polymers, MarcelDekker, New York, USA, 2002.

[49] L. Landau, B. Levich, Dragging of a liquid by a moving plate, Acta PhysicochimicaURSS 17 (1942) 4254.

[50] P. Schunk, A. Hurd, C. Brinker, Free-meniscus coating processes, in: S.F.Kistler, P.M. Schweizer (Eds.), Liquid Film Coating: Scientic Principles andTheir Technological Implications, Chapman & Hall, London, UK, 1997, pp.673708.

[51] P. Yimsiri, M.R. Mackley, Spin and dip coating of light-emitting polymer solu-tions: matching experiment with modelling, Chemical Engineering Science 61(2006) 34963505.

[52] S. Etienne-Calas, A. Duri, P. Etienne, Fracture study of organicinorganiccoatings using nanoindentation technique, Journal of Non-Crystalline Solids344 (2004) 6065.

[53] J. Malzbender, G. de With, The use of the indentation loading curve to detectfracture of coatings, Surface and Coatings Technology 137 (2001) 7276.

[54] J. Malzbender, G. de With, The use of the loading curve to assess soft coatings,Surface and Coatings Technology 127 (2000) 265272.

[55] A.C. Fischer-Cripps, A review of analysis methods for sub-micron indentationting, Vacuum 58 (2000) 569585.. Antu

the don te383ricotesing is, Me. Pele/soft

ence a. Bhatd witholids

Brisc inter126iang,tions(2001Shen, ting/sbe mian denrovingress u, Y. H

s on kahmium/rganiaveh,miniu. Undeerse

207in Organic Coatings 43 (2001) 149157.ang, J. Zhou, S. Liu, Y. Yu, X. Pang, S. Yang, Construction and study on

protective property of polydopamine-based 3-layer organic coatingsum substrate, Progress in Organic Coatings 68 (2010) 244247.

k, A. Sharma, A.S. Khanna, Value addition to waterborne polyurethaneilicone modication for developing high performance coating on alu-lloy, Progress in Organic Coatings 65 (2009) 206216.ni-Nooshabadi, S.M. Ghoreishi, M. Behpour, Direct electrosynthesisilinemontmorrilonite nanocomposite coatings on aluminum alloy

their corrosion protection performance, Corrosion Science 53 (2011)2.

nfredi, O. de la Osa, N. Galego Fernndez, A. Vzquez,properties relationship for resols with different formalde-nol molar ratio, Polymer 40 (1999) 38673875.ruments, Calotest Technical Features, 2004.-1, Metallic Materials Instrumented Indentation Test for Hardnessrials Parameter, International Standard, 2002.his, Indentation mechanics and its application to thin lm character-acuum 82 (2008) 13531359..D. Nix, Effects of the substrate on the determination of thin lm

al properties by nanoindentation, Acta Materialia 50 (2002) 2338.. Yang, T. Druffel, E.A. Grulke, Nanoindentation behavior of ultrathinc lms, Polymer 46 (2005) 1176811772.gan, B. Tanga, Viscoelastic effects during unloading in depth-sensingon, Journal of Materials Research 17 (2002) 26042610., W.C. Oliver, A new method for analyzing data from continuous

nsing microindentation tests, Journal of Materials Research 5 (1990)

inen, L. Alvila, J. Rainio, T.T. Pakkanen, IR spectroscopy as a quanti-d predictive analysis method of phenolformaldehyde resol resins,f Applied Polymer Science 69 (1998) 21752185.redi, M.J.L. Gins, G.J. Bentez, W.A. Egli, H. Rissone, A. Vzquez, Use ofenolic lacquers in food can coatings: characterization of lacquers ands, Journal of Applied Polymer Science 95 (2005) 14481458.

ak, T. Biernacka, M. Skarzynski, Some properties and chemical struc-enolic resins and their derivatives, Journal of Applied Polymer Science) 531542.

tes[56] J.M

Ontati831

[57] A. Tsenlm

[58] A.AlmSci

[59] A.Kateof S

[60] B.J.for121

[61] C. Xrela39

[62] W. coapro

[63] J. vImpPro

[64] P. Hcle

[65] M. minin O

[66] A. Ralu

[67] P.Rimm199nes, J.V. Fernandes, N.A. Sakharova, M.C. Oliveira, L.F. Menezes,etermination of the Youngs modulus of thin lms using inden-sts, International Journal of Solids and Structures 44 (2007)34.aux, G. Duarte, D. Chicot, E. Le Bourhis, E. Bemporad, J. Lesage, Depth-ndentation modeling for determination of Elastic modulus of thinchanics of Materials 42 (2010) 166174.gri, X. Huang, Nanoindentation on soft lm/hard substrate and hardsubstrate material systems with nite element analysis, Compositesnd Technology 68 (2008) 147155.tacharya, W.D. Nix, Analysis of elastic and plastic deformation associ-

indentation testing of thin lms on substrates, International Journaland Structures 24 (1988) 12871298.oe, Isolated contact stress deformations of polymers: the basispreting polymer tribology, Tribology International 31 (1998).

H.J. Sue, J. Chu, B. Coleman, Scratch behavior and material propertyhip in polymers, Journal of Polymer Science, Part B: Polymer Physics) 4759.B. Jiang, S.M. Gasworth, H. Mukamal, Study of tribological properties ofubstrate system in micrometer and nanometer scales with a scanningcroscope, Tribology International 34 (2001) 135142.

Brand, S. Van Gils, P.C.J. Beentjes, H. Terryn, V. Sivel, J.H.W. de Wit,g the adhesion between epoxy coatings and aluminium substrates,in Organic Coatings 51 (2004) 339350.uaming, Controlled coating of antimony-doped tin oxide nanoparti-olinite particles, Applied Clay Science 48 (2010) 368374.

an, D. Persson, C. Leygraf, In situ ATRFTIR studies of the alu-polymer interface upon exposure to water and electrolyte, Progressc Coatings 57 (2006) 7888.

Z.K. Tsameret, E. Grossman, Surface characterization of thin layers ofm oxide, Surface and Coatings Technology 88 (1997) 103111.rhill, A.N. Rider, Hydrated oxide lm growth on aluminium alloys

d in warm water, Surface and Coatings Technology 192 (2005).

Furan resins as replacement of phenolic protective coatings: Structural, mechanical and functional characterization1 Introduction2 Experimental2.1 Materials and sample preparation2.2 Methods2.2.1 Viscosity measurements2.2.2 Fourier transform infrared analysis (FTIR)2.2.3 Dynamic mechanical analysis (DMA)2.2.4 Contact angle measurements2.2.5 Scanning electronic microscopy (SEM) and Calotest2.2.6 Nanoindentation and nanoscratch tests2.2.7 Potentiodynamic polarization

3 Results and discussion3.1 Physicochemical characterization3.2 Coating thickness3.3 Nanomechanical properties3.4 Nanoscratch behavior3.5 Corrosion behavior

4 ConclusionsAcknowledgementsReferences