Short communication

Received: 25 July 2013 Revised: 21 August 2013 Accepted: 5 October 2013 Published online in Wiley Online Library: 19 November 2013

(wileyonlinelibrary.com) DOI 10.1002/sia.5345

Black wattle tannin as a zinc phosphatingcoating sealerRafael Silveira Peres,a* Eduardo Cassel,b Carlos Arthur Ferreiraa

and Denise Schermann Azambujac

This paper reports an investigation into the use of condensed black wattle tannin as an environmental friendly sealer afterzinc phosphating of carbon steel. In order to verify the sealer efficiency against corrosion, electrochemical impedancespectroscopy, potentiodynamic polarisation and salt spray tests were performed. To verify the pore modifications and phasecompositions, SEM, electron dispersive spectroscopy and X-ray diffraction analysis were carried out. The influence of thesealer on the adherence of finishing coatings was analysed. The results show that the tannin/sealer did not affect the finalcoating adhesion and improved the corrosion resistance of sealed samples. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: coating; phosphating; sealer; tannin; surface

* Correspondence to: R. Peres, Materials Engineering, Universidade Federal doRio Grande do Sul, Av. Bento Gonçalves 9500, 91501–970, Porto Alegre, Brazil.E-mail: rafael.s.peres@gmail.com

a Escola de Engenharia/PPGE3M, Laboratório de Materiais Poliméricos, UFRGS,Av. Bento Gonçalves 9500, 91501-970 Porto Alegre, Brazil

b Faculdade de Engenharia, Departamento de Engenharia Química, PUCRS, Av.Ipiranga 6681, 90619-900 Porto Alegre, Brazil

c Instituto de Química, Laboratório de Eletroquímica, UFRGS, Av. BentoGonçalves 9500, 90619-900, Porto Alegre, Brazil

1

Introduction

Phosphating is the treatment of ferrous and non-ferrous materialssuch as iron, steel, aluminium andmagnesium to form an insolublecrystalline phosphate layer on the surface.[1] This layer mainly im-proves corrosion resistance and paint adhesion on the substrate.[2]

All conventional phosphating processes are based on dilutedphosphoric acid solutions with the presence of heavy or alkalimetals ions.[3,4] The most common types of phosphate coatingsare based on zinc, iron and manganese.[3]

Because of phosphate crystal growth, phosphate coatingshave pores inherent to layer formation.[5] This fact decreasesthe corrosion resistance of the substrate, because unprotectedareas provide a path for corrosion attack.[6] An alternative toincrease corrosion resistance is sealing the pores once corrosionreactions are initiated at the coating/substrate interface.[4,7]

Chromatising is the most effective conversion coating treatmentused for sealing the pores of phosphating coatings.[4,8]This treat-ment improves corrosion resistance due to the deposition of insol-uble chromates in the pores and passivation of exposed metal.[4]

However, there are many environmental problems associated withCr+6,[9], and several alternatives have been proposed according to areview by Narayanan.[4] However, many of the new sealing treat-ments use expensive and toxic reagents. Xu et al.[10] used silicatesolutions to post-seal hot-dip galvanised steel. Susac et al.[11]

applied the organosilane bis-1,2-(triethoxysilyl)ethane to sealing a2024-T3 aluminium alloy, which had previously been coated byzinc phosphate. Banzeck et al.[12] used a treatment with niobiumammonium oxalate to passivate of the pores of carbon steel (SAE1010) after being phosphated in a zinc phosphating bath.

Tannins are natural and biodegradable polyphenolic com-pounds extracted from the bark, fruit and wood of several trees,including the black wattle tree (Acacia mearnsii De Wild.).[13] Theapplication of tannins in corrosion studies has been investigatedand their efficiency is controversial.[14–18] Tannins can convertactive rust into compounds that are more stable and corrosion-resistant.[18] Some authors have developed a tannin primer coatingthat exhibited excellent anticorrosive properties.[17,19,20] On the

Surf. Interface Anal. 2014, 46, 1–6

other hand, the presence of hydrolysable tannins inhibits theformation of magnetite.[15] Recent studies have shown that tanninscan be good corrosion inhibitors for steel in acidic media due to thechemisorption mechanism.[18,21–23] When Fe3+ ions react with tan-nins in aerated near neutral aqueous solution, a blue-black complexcalled ferric-tannate is formed.[24] However, the ability of the ferric-tannates to protect against corrosion has not been resolved,[23] es-pecially in long-term systems.[22] These contradictory results maybe related to the different types of tannins and different experi-mental conditions.[25]

The aim of this study was to evaluate the use of tannin extractedfrom the bark of the black wattle tree as a natural, non-toxic andnon-expensive long-term sealer of the zinc phosphating coating.Thus, this work intends to provide a specific and practice applica-tion of the black wattle tannin, evaluating the ferric-tannate behav-iour beneath the organic coating. To the best of our knowledge, theapplication of the tannin as pore sealer has not been published yet,because the works cited before are based manly in relative short-term inhibition studies and coatings formulation.

Experimental section

Material

All solutions and samples were prepared with analytical gradereagents. To prepare the zinc phosphating bath, zinc oxide

Copyright © 2013 John Wiley & Sons, Ltd.

R. S. Peres et al.

2

(Vetec, Brazil), nitric acid (Merck, Germany), nickel sulphate(Vetec, Brazil) and phosphoric acid (Merck, Germany) were used.Sodium sulphate (VETEC, Brazil) was employed in the preparationof the electrolyte solutions to electrochemical impedancespectroscopy (EIS) and potentiodynamic polarisation (PP) experi-ments. A two-component epoxy primer (Intergard 269; AkzoNobel, USA) was used as the finishing coating in adherence andsalt spray tests. The black wattle tannin sample used in this workwas supplied by TANAC (Montenegro, Brazil). The chemicalcomposition of the carbon steel samples is given in Table 1.

Sample preparation

Samples of carbon steel were prepared by degreasing with anacetone/chloroform mixture (1:1), abraded with silicon carbidepaper up to grade #1200, then degreasing and drying under ahot air stream. After the polishing and cleaning processes, thesamples were immersed in zinc phosphating solution (Table 2)at 80 °C for 10min. The samples were then rinsed thoroughlywith deionised water and dried under a hot air stream.

Sealing process

After the zinc phosphating treatment, the samples were im-mersed in a solution of 6 g L�1 of black wattle tannin for 40min.At this concentration, the pH value of the solution was close to6, which is optimal for the formation of ferric-tannate.[21,22] After40min, a blue-black residue was observed on the bulk of the bathand over the steel surface.

Surface analysis

The morphology of the phosphated steel surface was assessed by(SEM; Philips model XL30) coupled to an Edax energy dispersivespectrometer (EDS). All samples were coated with a thin-goldlayer prior to analysis.The phase identification of the phosphated sample was

investigated by X-ray diffraction analysis (XRD) with a Philipsdiffractometer (model X’Pert MPD) using Cu Kα radiation. A stepsize of 0.02° was used to create X-ray patterns to identify thematerial at a 2θ angle (4°–70°). Phases were identified by compar-ing with the Joint Committee on Powder Diffraction Standardsusing X’Pert HighScore software.

Table 1. Chemical composition of carbon steel sample

Chemical composition

Element C Mn P S Cu Cr

wt.% 0.103 0.46 0.013 0.096 0.01 0.18

Table 2. Composition of zinc phosphating bath.

Reagent Concentration (g L-1)

Phosphoric Acid (H3PO4) 17.5

Nitric Acid (HNO3) 20.0

Zinc Oxide (ZnO) 11.5

Nickel Sulphate (NiSO4) 0.25

wileyonlinelibrary.com/journal/sia Copyright © 2013 J

Electrochemical tests

Working electrodes were prepared from the phosphatisated steelsamples (25 × 15 × 1mm). An area of 1.0 cm2 was delimited bycold resin (epoxy) embedding.

A saturated calomel electrode (E = 0.241 V/NHE) was used asthe reference electrode, and all potentials are referred to it. Theauxiliary electrode was a Pt wire. The experiments were carriedout under naturally aerated conditions at 25 °C. The electrochem-ical measurements were performed using an AUTOLAB PGSTAT30 potentiostat (Echo Chemie, Netherlands) coupled to a fre-quency response analyser. The software used for the analysis ofimpedance spectra was NOVA 1.7 (Echo Chemie, Netherlands).The PP curves were recorded at a scan rate of 1mVs-1. Thecorrosion potentials (Ecorr) and the polarisation resistances (RP)were calculated at a scan rate of 0.1mVs-1 with potential rangeof Ecorr± 20mV. This method was based on the ASTM G59[26]

and ASTM G102[27] standards.The EIS measurements were performed in potentiostatic mode

at the open circuit potential. In this work, the open circuit potentialafter potential stabilisation is referred to as the corrosion potential,Ecorr. The amplitude of the EIS perturbation signal was 10mV, andthe frequency range studied was from 105 to 10�2Hz.

Salt spray and painting adhesion test

The carbon steel samples (100mm×50mm) were painted withan epoxy primer to a thickness of approximately 40μm. Salt spraytests were performed according to the standard ASTM B117.[28]

The dry paint film thickness was measured with a Byko-test7500 electromagnetic film thickness gauge (BYK Gardner,Germany). The mean value of six different areas of the steelsamples was used to calculate the final dry film thickness.

Adhesion measurements were made according to the ASTMD3359 standard test.[29] Eleven cuts (1mm of distance) weremade over the organic primer coating in two directions (perpen-dicular) in order to form squares. The number of the cuts and thedistance between them were based on the standard. The samplesurface was cleaned with a brush to eliminated residual coatinggenerated in the cut process. After that, a Scoth-880 (3M, USA)tape was attached over the cuts with the assistance of an eraser.The tape was pulled off with an angle near 180º in relation to thesample. The number of the detached primer coating wascompared with the standard, being classified in five degrees(5B to 0B). The highest degree (5B) corresponds to no removalof the primer, whereas the lowest degree corresponds to morethan 70% of primer removal.

Results and discussion

Characterisation of phases in the phosphate layer andmorphology

The phase compositions of the phosphate coating formed oncarbon steel before and after the sealing treatment wereanalysed by XRD as shown in Fig. 1. The phosphate coatingbefore the treatment with tannin (Fig. 1a) consisted mainly ofZn3(PO4)2 · 4H2O (hopeite) and Zn2Fe(PO4)2 · 4H2O (phosphophylite)phases, which are in agreement with the literature.[30] After thesealing process (Fig. 1b), the phosphate coating shows nosubstantial difference in the phase composition. Thus, thesealing bath (described before) does not affect the coating.

ohn Wiley & Sons, Ltd. Surf. Interface Anal. 2014, 46, 1–6

Figure 1. X-ray diffraction analysis patterns of the zinc phosphate coat-ing: (a) before and (b) after sealing.

Black wattle tannin as a zinc phosphating coating sealer

The amorphous nature of ferric-tannate could be confirmed[14,17]

once no additional peak was observed in the X-ray diffractionpatterns (Fig. 1).

The surfaces of the samples were analysed by SEM in backscat-tering electrons mode. The SEM micrograph and correspondingEDS for several areas in the unsealed samples are given in Fig. 2and the sealed samples are given in Fig. 3. The light areas showthe steel surface without phosphate crystals and the dark areascorrespond to the zinc phosphate crystals that formed duringthe phosphating process, as shown in the EDS spectra (Figs. 2and 3). Only the EDS spectra of the light areas in the sealedsample showed the presence of carbon (Fig. 3). Furthermore,the presence of a cracked layer in the pores (light areas) can beattributed ferric-tannate layer formation (Fig. 3).[31–33]

Figure 2. Micrograph and Edax energy dispersive spectrometer diagrams o

Surf. Interface Anal. 2014, 46, 1–6 Copyright © 2013 John W

Electrochemical tests

The anticorrosion efficiency of the sealed samples was evaluatedby PP and EIS. The PP curves of sealed and unsealed phosphatisedsteel samples were performed in aerated 0.1mol L-1 Na2SO4

solution after 4 h of immersion (Fig. 4).According to (Fig. 4), the anodic and cathodic current densities de-

creased when the phosphate steel was sealed. The presence of blackwattle tannin as a sealer shifted the Ecorr to a more positive value.

The corrosion current densities (jcorr) in Table 3 were calculatedfrom the extrapolation of the anodic and cathodic Tafel lines(Fig. 4).

According to Table 3, the polarisation resistance (Rp) showedthe highest value(18.85 kΩ.cm-2), and the jcorr value decreasedsignificantly in the sealed sample. This fact could be attributedto the ferric-tannate layer formed over the pores, evidencingthe protective character of this treatment.

To complement the PP data, EIS measurements wereperformed with the phosphatised steel samples in aerated0.1mol L-1 Na2SO4 after 4 h of immersion. The experimental andfitted EIS data are shown in (Figs. 5 and 6). According to Fig. 6,the Bode plots show two time constants whose maximum phaseangle was around �50°, attributed to the process under masstransport control, characteristic of phosphating layers.[8]

The equivalent circuit (Fig. 7) proposed by fitting the EISdiagrams is RS(CPE1R1)(CPE2R2), where Rs represents the ohmic re-sistance between the reference and the working electrode and R1and R2 represent the charge transfer resistance in physicalmeaning.[34] The capacitance was replaced by a constant phaseelement (CPE) that describes a non-ideal capacitor when thephase angle is different from �90°. The CPE impedance is attrib-uted to the distributed surface reactivity, surface heterogeneityand roughness of the current and potential distribution, whichin turn are related to the electrode geometry and the electrodeporosity.[35] The CPE impedance is given by equation 1:[34]

f an unsealed phosphatised carbon steel sample.

iley & Sons, Ltd. wileyonlinelibrary.com/journal/sia

3

Figure 3. Micrograph and Edax energy dispersive spectrometer diagrams of a sealed phosphatised carbon steel sample.

Figure 4. Potentiodynamic polarisation curves of unsealed (1) andsealed (2) phosphate steel sample immersed in aerated 0.1mol L�1

Na2SO4 solution for 4 h.

Table 3. Polarization parameters for the unsealed and sealed phos-phate steel samples immersed in aerated 0.1mol l�1 Na2SO4 solutionfor 4 h

Samples jcorr (μA cm-2) Ecorr (V) Rp (kΩ cm2)

Non-sealed 3.88 �0.65 1.458

Sealed 0.34 �0.59 18.85

Figure 5. Nyquist plots of unsealed (□) and sealed (●) phosphate steelsamples immersed in aerated 0.1mol L�1 Na2SO4 solution for 4 h. Thesolid black line (�) represents the fitting data.

R. S. Peres et al.

wileyonlinelibrary.com/journal/sia Copyright © 2013 J

4

1=ZCPE ¼ Q jωð Þn (1)

where ZCPE is the impedance and ω is the angular frequency.The parameters n and Q are independent of frequency. The CPErepresents a capacitor for n = 1, a resistor for n= 0 and a diffusionprocess for n= 0.5.

In Fig. 7, two time constants are presented: R1CPE1 related tothe inner conversion layer and R2CPE2 to the outer porous

ohn Wiley & Sons, Ltd. Surf. Interface Anal. 2014, 46, 1–6

Figure 6. Bode plots of unsealed (□) and sealed (●) phosphate steel samples immersed in aerated 0.1mol L�1 Na2SO4 solution for 4 h. The solid blackline (�) represents the fitting data.

Figure 7. Equivalent electric circuit proposed to simulate the experi-mental data of phosphated steel samples immersed in 0.1mol L-1 for 4 h.

Table 5. Comparison of the adhesion ratings of the primer coatingin the phosphatized samples

Samples Primer thickness (μm)Mean± SD* (n=6)

Adhesion rating(ASTM D3359)

Unphosphatized 41± 0.5 1B

Phosphatized 40± 0.5 4B

Sealed and phosphatized 42± 0.4 4B

* Standard deviation

Black wattle tannin as a zinc phosphating coating sealer

layer.[33,34] For these layers, the capacitance was replaced by aCPE due to the coating characteristics. In this case, the overallimpedance is given by the sum of the two layers correspondingto the impedance values and is characterised by a parallel combi-nation of the capacitance and resistance of each layer, asproposed in the literature.[8]

The fitting quality was evaluated based on error percentageassociated with each component, showing errors smaller than5%. The software used to simulate the EIS data was NOVA 1.7(Echo Chemie, Netherlands). The simulated data obtained fromthe fittings are given in Table 4.

According to Table 4, the sealing treatment with tanninpromotes the increase of the resistance and the decrease of thecapacitance improving the corrosion resistance of this treatment.These results are in agreement with the PP results (Table 3)showing the protective character of the formed layer.

Adhesion tests

Adhesion tests were performed to evaluate the influence of thesealer on the adherence of organic coatings on the phosphatisedsteel. The adhesion of the primer coating on the unsealed and

Table 4. Fitting parameters used to simulate the electrochemical impedanimmersed in aerated 0.1mol l�1 Na2SO4 solution for 4 h

Sample Equivalent circuit

Rs (Ω cm2) CPE1 (F cm-2 sn-1)

Non-Sealed Rs(R1Q1)(R2Q2) 15.36 2.71× 10-3

Sealed 17.21 7.43× 10-4

Surf. Interface Anal. 2014, 46, 1–6 Copyright © 2013 John W

sealed phosphatised samples was measured using a tape test.[29]

The results are summarised in Table 5.According to Table 5, the adhesion rating for sealed and

non-sealed phosphating coatings was 4B, equivalent to lessthan 5% primer coating removal. For the unphosphatisedsample, the adhesion rating was 1B, equivalent to 35%–65%primer coating removal. The results show that the addition ofthe sealer did not affect the adherence of paint to thephosphatised substrate.

Salt spray tests

To verify the influence of the sealer on the corrosion resistance ofthe coating/phosphating system, salt spray tests wereperformed.[28] Epoxy coatings on unphosphatised, phosphatisedand tannin/sealed phosphatised carbon steel samples were tested.Prior to salt spray tests, the organic coating was scratched to ex-pose the substrate. The duration of the salt spray tests was 672h.

Figure 8 show samples exposed to 672 h of salt spray. Evidentsigns of intense corrosive attack (Fig. 8a) were associated withthe unphosphated sample. The unsealed phosphatised sample(Fig. 8b) presented bubbles (corrosive attack) over the entire

ce spectroscopy plots for unsealed and sealed phosphate steel samples

Fitting parameters

n1 R1(kΩ cm2) CPE2 (F cm-2 sn-1 ) n2 R2 (kΩ cm2)

0.62 0.196 4.36× 10-4 0.65 6.895

0.39 1.230 1.30× 10-4 0.69 26.93

iley & Sons, Ltd. wileyonlinelibrary.com/journal/sia

5

Figure 8. Macrographs of the (a) unphosphated carbon steel, (b) unsealed phosphate and (c) tannin/sealed phosphate samples painted with epoxycoating and exposed to 672 h in salt spray test.

R. S. Peres et al.

6

surface. On the other hand, the tannin/sealed phosphatisedsamples presented bubbles only in the areas where the substratewas exposed (scratched area), showing an improvement incorrosion protection. In this case, the ferric-tannate used as sealershowed long-term protection against corrosion. The use of ferric-tannate layer concomitantly with organic coatings improves theoverall resistance against the corrosion.

Conclusions

The effect of black wattle tannin as a natural and non-toxic sealerafter phosphating treatment of carbon steel was evaluated in thisstudy. XRD confirmed that the ferric-tannate is amorphous andthe phosphate coating formed on carbon steel consisted mainlyof hopeite and phosphophylite. The SEM and EDS analysesshowed ferric-tannate layer formation on the exposed steelsurface (pore). Electrochemical and salt spray experimentsshowed an improvement in the corrosion resistance of the sealedphosphate coating. The adhesion tests indicate that the additionof the sealer did not affect the adherence of paint to thephosphatised substrate. Thus, black wattle tannin can be usedas an environmental friendly sealer in the zinc phosphating baththrough a simple and cheap process.

Acknowledgements

The authors thank the Brazilian government agency CNPq, whichprovided the financial support for this study and the scholarshipfor R.S Peres, the Microscopy Centre of the Pontifical CatholicUniversity (PUC) of Rio Grande do Sul and LACER and LACORfrom the Federal University of Rio Grande do Sul.

References[1] E. P. Banczek, P. R. P. Rodrigues, I. Costa, Surf. Coat. Technol. 2008,

202, 2008.[2] M. Sheng, Y. Wang, Q. Zhong, H. Wu, Q. Zhou, H. Lin, Surf. Coat.

Technol. 2011, 205, 3455.[3] ASM Handbook: Surface Engineering, American Society for Metals,

1994.[4] T. S. N. S. Narayanan, Rev. Adv. Mater. Sci. 2005, 9, 130.

wileyonlinelibrary.com/journal/sia Copyright © 2013 J

[5] E. P. Banczek, P. R. P. Rodrigues, I. Costa, Surf. Coat. Technol. 2009,203, 1213.

[6] D. Weng, P. Jokiel, A. Uebleis, H. Boehni, Surf. Coat. Technol. 1997,88, 147.

[7] J. Creus, H. Mazille, H. Idrissi, Surf. Coat. Technol. 2000, 130, 224.[8] S. M. Tamborim Takeuchi, D. S. Azambuja, A. M. Saliba-Silva, I. Costa,

Surf. Coat. Technol. 2006, 200, 6826.[9] B. Y. Johnson, J. Edington, M. J. O’Keefe, Mater. Sci. Eng. A 2003,

361, 225.[10] Y.-y. Xu, B.-l. Lin, Trans. Nonferrous Met. Soc. China 2007, 17, 1248.[11] D. Susac, C. W. Leung, X. Sun, K. C. Wong, K. A. R. Mitchell, Surf. Coat.

Technol. 2004, 187, 216.[12] E. P. Banczek, M. Terada, P. R. P. Rodrigues, I. Costa, Surf. Coat.

Technol. 2010, 205, 2503.[13] W. E. Hillis, in Black Wattle and its Utilisation (Eds: A. G. Brown, H. C. Ko),

Rural Industries Research and Development Corporation, Barton,1997, pp. 106.

[14] M. Favre, D. Landolt, Corros. Sci. 1993, 34, 1481.[15] M. Favre, D. Landolt, K. Hoffman, M. Stratmann, Corros. Sci. 1998,

40, 793.[16] J. C. Galván Jr, J. Simancas, M. Morcillo, J. M. Bastidas, E. Almeida, S.

Feliu, Electrochim. Acta 1992, 37, 1983.[17] G. Matamala, W. Smeltzer, G. Droguett, Corros. Sci. 2000, 42, 1351.[18] A. A. Rahim, E. Rocca, J. Steinmetz, M. J. Kassim, R. Adnan, M. Sani

Ibrahim, Corros. Sci. 2007, 49, 402.[19] G. Matamala, W. Smeltzer, G. Droguett, Corrosion 1994, 50, 270.[20] O. R. Pardini, J. I. Amalvy, A. R. Sarli, R. Romagnoli, V. F. Vetere, J. Coat.

Technol. 2001, 73, 99.[21] S. Martinez, Mater. Chem. Phys. 2003, 77, 97.[22] R. S. Peres, E. Cassel, D. S. Azambuja, ISRN Corros. 2012, 2012, 9.[23] A. A. Rahim, J. Kassim, Recent Pat. Mater. Sci. 2008, 1, 223.[24] J. Gust, J. Suwalski, Corrosion 1994, 50, 355.[25] O. Lahodny-Šarc, F. Kapor, Mater. Corros. 2002, 53, 264.[26] ASTM G59-97: Standard Test Method for Conducting

Potentiodynamic Polarization Resistance Measurements.[27] ASTM G102-89: Standard Practice for Calculation of Corrosion Rates

and Related Information from Electrochemical Measurements.[28] ASTM B117-11: Standard Practice for Operating Salt Spray (Fog)

Apparatus.[29] ASTM D 3359–02: Standard Test Methods for Measuring Adhesion

by Tape Test.[30] T. S. Narayanan, Rev. Adv. Mater. Sci 2005, 9, 130.[31] C. A. Barrero, L. M. Ocampo, C. E. Arroyave, Corros. Sci. 2001, 43, 1003.[32] T. K. Ross, R. A. Francis, Corros. Sci. 1978, 18, 351.[33] A. A. Rahim, E. Rocca, J. Steinmetz, M. Jain Kassim, Corros. Sci. 2008,

50, 1546.[34] G. W. Walter, Corros. Sci. 1986, 26, 681.[35] J. B. Jorcin, M. Orazem, N. Pébère, B. Tribollet, Electrochim. Acta 2006,

51, 1473.

ohn Wiley & Sons, Ltd. Surf. Interface Anal. 2014, 46, 1–6