7
Progress in Organic Coatings 49 (2004) 130–136 Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds Trinh Anh Truc a,b , Nadine Pébère a,, To Thi Xuan Hang b , Yves Hervaud c , Bernard Boutevin c a Centre Inter Universitaire de Recherche et d’Ingénierie des Matériaux (CIRIMAT), Laboratoire Interfaces et Matériaux, UMR CNRS 5085, ENSIACET 118, Route de Narbonne, 31077 Toulouse Cedex 04, France b Laboratory of Protective Coatings, Institute for Tropical Technology, Nghia Do, Cau Giay, Hanoi, Viet Nam c Laboratoire de Chimie Macromoléculaire, UMR CNRS 5076, ENSCM, 8 Rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France Received 25 March 2003; accepted 30 August 2003 Abstract This study is focussed on the use of organic phosphorous compounds for the improvement of the corrosion protection of a carbon steel by a UV-cured polyurethane coating. Two compounds were employed: one with a long hydrocarbon chain: tridecylphosphate (TDP) and the second one with an unsaturated hydrocarbon chain: methacryloxyethylphosphate (MOP). The compounds were used either for surface treatments before the application of the organic coating or added directly to the coating. Corrosion resistance of the coated steel was evaluated by electrochemical impedance spectroscopy. For the surface treatment, the treatment time was optimised for the mixture 5% TDP + 1.5% MOP. A treatment time of 60min led to the best protection. By incorporating the phosphorous compounds into the polyurethane coating, it was shown that TDP alone significantly improved the protective properties. © 2003 Elsevier B.V. All rights reserved. Keywords: Carbon steel; Surface treatment; UV-curing; Polyurethane coating; Electrochemical impedance spectroscopy 1. Introduction Anticorrosive pre-treatments before painting have been the subject of numerous investigations, particularly on car- bon steels and galvanised steels. Due to the problems of high toxicity associated with chromate treatment, differ- ent studies have been carried out to develop an environ- mentally more acceptable treatment. Organic phosphorous compounds are commonly used to inhibit the corrosion of carbon steel [1–4] and zinc [5] in aqueous solutions. Their use is relatively risk-free due to their low toxicity. In a previous work [6], surface treatments were carried out on a carbon steel in a solution containing a mixture of phosphorous compounds: tridecylphosphate (TDP) and methacryloxyethylphosphate (MOP). It was found that the presence of MOP in the TDP solution creates a synergistic effect for the formation of a film on the steel surface. For a solution containing 5% TDP, the optimum concentration of MOP giving the best protection is 1.5%. The film was porous and essentially identified as a Fe(TDP) n complex Corresponding author. Fax: +33-5-62-88-56-63. E-mail address: [email protected] (N. P´ eb` ere). with n = 1 or 2. The synergistic effect was explained by a competitive action of the two compounds on the carbon steel surface. At the beginning of immersion, the role of MOP is predominant and its aggressiveness generates the formation of ferrous ions which react with TDP to form the Fe(TDP) n complex. Then, the competition mechanism allows the growth of the film on the steel surface. However, the protective properties of the surface treatment were not durable and the treated carbon steel surface was degraded after twenty hours of immersion in a 0.1 M sodium chloride solution. In this study, two ways were chosen to improve the corro- sion resistance of the carbon steel treated with these organic phosphorous compounds: (i) by covering the pre-treated carbon steel by a UV-curable polyurethane aliphatic diacry- late coating (PU) or (ii) by incorporating the compounds in the PU coating. In the latter case, the compounds could play the role of corrosion inhibitors. The UV-curable resin was applied without solvent. This point is particularly in- teresting because it avoids the emission of volatile organic compounds during coating application. To evaluate the protective properties of the different systems, electrochemical impedance measurements were 0300-9440/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.porgcoat.2003.08.022

Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

Embed Size (px)

Citation preview

Page 1: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

Progress in Organic Coatings 49 (2004) 130–136

Evaluation of corrosion performance of a UV-cured polyurethanecoating in the presence of organic phosphorous compounds

Trinh Anh Truca,b, Nadine Pébèrea,∗, To Thi Xuan Hangb,Yves Hervaudc, Bernard Boutevinc

a Centre Inter Universitaire de Recherche et d’Ingénierie des Matériaux (CIRIMAT), Laboratoire Interfaces et Matériaux,UMR CNRS 5085, ENSIACET 118, Route de Narbonne, 31077 Toulouse Cedex 04, France

b Laboratory of Protective Coatings, Institute for Tropical Technology, Nghia Do, Cau Giay, Hanoi, Viet Namc Laboratoire de Chimie Macromoléculaire, UMR CNRS 5076, ENSCM, 8 Rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France

Received 25 March 2003; accepted 30 August 2003

Abstract

This study is focussed on the use of organic phosphorous compounds for the improvement of the corrosion protection of a carbonsteel by a UV-cured polyurethane coating. Two compounds were employed: one with a long hydrocarbon chain: tridecylphosphate (TDP)and the second one with an unsaturated hydrocarbon chain: methacryloxyethylphosphate (MOP). The compounds were used either forsurface treatments before the application of the organic coating or added directly to the coating. Corrosion resistance of the coated steelwas evaluated by electrochemical impedance spectroscopy. For the surface treatment, the treatment time was optimised for the mixture5% TDP+ 1.5% MOP. A treatment time of 60 min led to the best protection. By incorporating the phosphorous compounds into thepolyurethane coating, it was shown that TDP alone significantly improved the protective properties.© 2003 Elsevier B.V. All rights reserved.

Keywords: Carbon steel; Surface treatment; UV-curing; Polyurethane coating; Electrochemical impedance spectroscopy

1. Introduction

Anticorrosive pre-treatments before painting have beenthe subject of numerous investigations, particularly on car-bon steels and galvanised steels. Due to the problems ofhigh toxicity associated with chromate treatment, differ-ent studies have been carried out to develop an environ-mentally more acceptable treatment. Organic phosphorouscompounds are commonly used to inhibit the corrosion ofcarbon steel[1–4] and zinc[5] in aqueous solutions. Theiruse is relatively risk-free due to their low toxicity.

In a previous work[6], surface treatments were carriedout on a carbon steel in a solution containing a mixtureof phosphorous compounds: tridecylphosphate (TDP) andmethacryloxyethylphosphate (MOP). It was found that thepresence of MOP in the TDP solution creates a synergisticeffect for the formation of a film on the steel surface. Fora solution containing 5% TDP, the optimum concentrationof MOP giving the best protection is 1.5%. The film wasporous and essentially identified as a Fe(TDP)n complex

∗ Corresponding author. Fax:+33-5-62-88-56-63.E-mail address: [email protected] (N. Pebere).

with n = 1 or 2. The synergistic effect was explained bya competitive action of the two compounds on the carbonsteel surface. At the beginning of immersion, the role ofMOP is predominant and its aggressiveness generates theformation of ferrous ions which react with TDP to formthe Fe(TDP)n complex. Then, the competition mechanismallows the growth of the film on the steel surface. However,the protective properties of the surface treatment were notdurable and the treated carbon steel surface was degradedafter twenty hours of immersion in a 0.1 M sodium chloridesolution.

In this study, two ways were chosen to improve the corro-sion resistance of the carbon steel treated with these organicphosphorous compounds: (i) by covering the pre-treatedcarbon steel by a UV-curable polyurethane aliphatic diacry-late coating (PU) or (ii) by incorporating the compoundsin the PU coating. In the latter case, the compounds couldplay the role of corrosion inhibitors. The UV-curable resinwas applied without solvent. This point is particularly in-teresting because it avoids the emission of volatile organiccompounds during coating application.

To evaluate the protective properties of the differentsystems, electrochemical impedance measurements were

0300-9440/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.porgcoat.2003.08.022

Page 2: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136 131

carried out for various exposure times in a sodium chloridesolution. For the phosphorous compounds used in surfacetreatment, four treatment times were chosen: 15, 30, 60 and120 min. The results were compared with those obtainedwith a classical phosphatation treatment.

Finally, XPS analyses were carried out to identify thepresence of the phosphorous compounds at the carbon steelsurface after the removal of the organic coating.

2. Experimental

2.1. Materials

The sample selected for the study was an XC35 carbonsteel and had the following composition in wt.%: C= 0.35,Mn = 0.65, Si= 0.25, P= 0.035 and Fe to 100.

The test solution was a 0.1 M NaCl solution (reagentgrade) in contact with air.

TDP was supplied by Atochem. The second reactive usedwas a mixture of MOP, supplied by UCB chemicals. TDPand MOP were of technical grade and were used withoutfurther purification. The structures of the compounds arepresented inFig. 1.

The UV-curable polyurethane aliphatic diacrylateoligomer (EB270) was supplied by UCB chemicals. Todecrease the viscosity, two reactive diluents were used:hexanediol diacrylate (HDDA) and lauryl acrylate (LA)supplied by Aldrich. The different products were mixed inthe proportion 40/40/20 for EB270, HDDA and LA, re-spectively. The photoinitiator (3%) used was Irgacure 184supplied by Ciba.

The organic phosphorous compounds were incorporatedin the coating at 1% for MOP, 1% for TDP or 1% for themixture MOP+ TDP (TDP/MOP ratio was 3/1). The incor-poration of the organic compounds in the PU coating canmodify the microstructure of the polymeric network, partic-ularly in the case of MOP for which the methacrylate groupcan react with the acrylate group of the PU resin. The glasstransition temperature,Tg, was measured using differentialscanning calorimetry. The presence of the phosphorous

O 13R C H27

x = 1 (diacid) or 2 (monoacid)

O

(RO)x x P (OH)3-

CH2 C

CH3

C O (CH2)2

O

P (OH)3-x O

x

1 x 2

(a)

(b)

Fig. 1. Structure of the phosphorous compounds: (a) TDP and (b) MOP.

compounds did not change theTg value of the PU coating(−31C) showing that the addition of the compounds doesnot influence the polyurethane cross-linking.

2.2. Surface treatment

For all the experiments, the carbon steel samples(10 cm × 10 cm × 0.2 cm) were polished with an abra-sive paper from 80 to 1200 grade, degreased with ethanol,cleaned in water in an ultrasonic bath and then dried inwarm air. They were then immediately immersed in anethanol–water mixture (80–20%) containing the activecompounds. The concentrations of TDP and MOP were 5and 1.5%, respectively. The treatment solution parametershad been previously optimised[7–9]. The treatment timeswere chosen between 15 and 120 min. During the treatment,the solution was quiescent. The film formed was stablein air.

2.3. UV curing

The liquid formulation was uniformly applied to the steelsample to get, after curing, a 22–24m thick layer. A Fu-sion UV Curing System, Model LC6, was used. The curingwas accomplished by passing the coated sample under a1200 W-bulbH lamp with a conveyor speed of 15 m min−1.After that, IR analysis showed the disappearance of the char-acteristic bands of the acrylate bond at 1630 and 810 cm−1,indicating complete polymerisation.

2.4. Electrochemical impedance measurements

A classical three-electrode cell was used: the workingelectrode with an exposed area of 15 cm2, the saturatedcalomel reference electrode (SCE) and a platinum auxiliaryelectrode.

The electrochemical impedance measurements were per-formed using a Solartron 1250 frequency response analyserover a frequency range of 65 kHz to several mHz with fivepoints per decade using 30 mV peak-to-peak sinusoidal volt-age and a Solartron 1287 electrochemical interface. For eachsystem, three samples were tested in parallel.

2.5. XPS analysis

The XPS measurements were carried out on a VG Es-calab MKII. The steel samples covered with the PU coatingscontaining the phosphorous compounds, were immersed for20 days in the 0.1 M NaCl solution then, the coatings wereremoved in an ultrasonic bath. The carbon steel sampleswere rinsed with distilled water, dried in warm air andplaced in a vacuum chamber. The specimens were irradi-ated with a Mg K ray source. The X-ray power was 300 W.Angle-resolved measurements were made at a take-off angleθ = 90.

Page 3: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

132 T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136

3. Results and discussion

3.1. Carbon steel surface without pre-treatment coveredby the PU coating

Fig. 2 presents the impedance diagrams obtained forthe carbon steel covered with the PU coating for variousimmersion times in the 0.1 M NaCl solution. The dia-grams are characterised by two frequency domains. Thehigh-frequency (HF) part of the diagram is related to the or-ganic coating while the low-frequency (LF) part correspondsto the reactions occurring on the metal through defects andpores of the coating[10]. The HF loop progressively de-creases when the immersion time increases. After 21 daysof immersion, the impedance value was significantly lowershowing limited protection of the carbon steel. After immer-sion in the aggressive solution, the coating came unstuckindicating that it was poorly adherent to the substrate.

3.2. Carbon steel surface pre-treated by the organicphosphorous compounds and covered by a PU coating

The carbon steel was pre-treated in the solution contain-ing MOP+TDP for different times (15, 30, 60 and 120 min)and covered by the PU coating. To follow the degradationof the samples, we ran impedance measurements for variousimmersion times in the 0.1 M NaCl solution. As an exam-ple, the impedance diagrams obtained for the carbon steelpre-treated for 60 min and covered with the PU coating arepresented inFig. 3.

0

250

500

750

0 500 1000 1500

(a)

10 Hz

10 mHz

0

15

30

45

0 30 60 90

(b)

39.8 mHz15.8 Hz

1.58 kHz

0

1

2

3

0 2 4 6

(c)

10 mHz2.5 Hz2.5 kHz

Imag

inar

y P

art

/ kΩ

cm

2

Real Part / kΩcm2

Fig. 2. Electrochemical impedance diagrams obtained for the carbon steelwithout surface treatment covered by the PU coating for various immersiontimes in the 0.1 M NaCl solution: (a) 1 day, (b) 7 days and (c) 21 days.

0 100

5 104

1 105

1.5 105

0 100 1 105 2 105 3 105

10 mHz0.1 Hz1.58 Hz

0 100

2.5 104

5 104

7.5 104

0 100 5 104 1 105 1.5 105

1.58 Hz

0.1 Hz

0 100

4 103

8 103

1.2 104

0 100 8 103 1.6 104 2.4 104

0.39 Hz

10 mHz

15.8 Hz

Imag

inar

y P

art

/ kΩ

cm

2

Real Part / kΩ cm2

(a)

(b)

(c)

Fig. 3. Electrochemical impedance diagrams obtained for the carbon steelpre-treated for 60 min in the MOP+ TDP solution and covered by thePU coating for various immersion times in the 0.1 M NaCl solution: (a)1 day, (b) 7 days and (c) 21 days.

At the beginning of exposure (1 and 7 days), the LF partof the diagrams was not well defined. The HF part charac-terised the properties of the coating including the surfacetreatment. For a longer exposure time (21 days), the LF partbecame better defined. Whatever the immersion time, theimpedance values remained relatively high by comparisonwith those obtained for the system without treatment (Fig. 2).This result indicates that high protection was obtained withthe pre-treatment by the phosphorous compounds.

As in previous works[11–18], the values of the resistance,RHF, extracted from the HF loop were used to follow thedegradation of the different systems with immersion time inthe aggressive solution. The impedanceRHF is associatedwith ionic transport through the pores of the coating. Inthis study, whatever the system tested, the HF loop wasalways well defined. Thus, theRHF variations were used tocharacterise the protective properties of the PU coatings inthe presence of the surface treatment.

From the impedance diagrams obtained for the differenttreatment times, the values ofRHF as a function of expo-sure time in the aggressive solution are reported inFig. 4.For short treatment times (15 and 30 min), the resistancevalues decreased during the first 7 days of immersion. Theyremained stable for the 30 min treatment but increasedslightly from 7 to 21 days of immersion for the 15 min treat-ment. The fall ofRHF during the first few days of exposurewas attributed to the penetration of the electrolyte into the

Page 4: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136 133

100

101

102

103

104

105

106

0 5 10 15 20 25 30 35 40

RH

F/ k

Ω c

m2

Immersion time / days

Fig. 4.RHF versus immersion time in the 0.1 M NaCl solution for differenttreatment times of the carbon steel surface covered by the PU coating:(×) 15 min, () 30 min, () 60 min and () 120 min.

coating and/or to the increase of the number of defects. There-increase ofRHF was probably a result of defect blockageby corrosion product[13,19]. For the 60 min treatment, thevalues ofRHF were clearly higher than the values measuredwith the other treatment times. They decreased slowly withexposure time. After 35 days of exposure,RHF remainedhigh. This result shows an efficient protection of the carbonsteel by the PU coating with a 60 min treatment by theMOP + TDP mixture. For the 120 min treatment,RHF de-creased slightly during the first 7 days, and then decreasedmore rapidly. However, the values ofRHF were lower thanfor the 60 min treatment. This lower value ofRHF obtainedfor 120 min treatment can be explained by the significantthickness of the layer for this treatment time. Actually, itwas previously seen[6] that the layer thickness increasedprogressively until 120 min of treatment to reach a valueof about 10m. The presence of the thick layer due to thesurface treatment does not improve the global impedanceof the system.

FromFig. 4, it can be concluded that the 60 min treatmentled to the best protection of the carbon steel. After 35 days ofimmersion, no corrosion was visible on this coated sample.For the three other treatment times, it can be seen that after35 days of immersion in the aggressive solution, the valuesof the resistance were of the same order of magnitude andsome corrosion products appeared on the samples.

Electrochemical impedance measurements were also car-ried out for the carbon steel surface treated by classical phos-phatation and covered by the PU coating.Fig. 5 comparesthe variation ofRHF with immersion time in the NaCl so-lution for the carbon steel pre-treated by phosphatation andcovered by the PU coating and for the PU coating appliedto the carbon steel pre-treated by MOP+ TDP for 60 min.The decrease ofRHF for the phosphated system was signif-icant during the first 7 days of exposure butRHF remainedrelatively constant between 7 and 28 days of exposure. The

100

101

102

103

104

105

106

0 5 10 15 20 25 30 35 40

RH

F/ k

Ω c

m2

Immersion time / days

Fig. 5. RHF versus immersion time in the 0.1 M NaCl solution for: ()carbon steel pre-treated 60 min in the MOP+ TDP solution and coveredby the PU coating and () carbon steel pre-treated by phosphatation andcovered by the PU coating.

comparison of the two curves clearly shows superior pro-tection for the carbon steel pre-treated with MOP+ TDPfor 60 min and covered by PU coating compared to the clas-sical phosphatation for which, after 28 days of immersion,the sample surface was strongly corroded even though theorganic coating remained attached to the substrate.

3.3. Organic phosphorous compounds incorporatedinto the PU coating

In the first part of the study, we showed that the surfacetreatment of the carbon steel with MOP+ TDP stronglyimproved the protective properties of the PU coating. How-ever, this process presents two drawbacks: (i) the treatmenttime is relatively long and (ii) the adherence of the coating,estimated qualitatively, was less good than for the classicalphosphatation. Thus, in the second part of the study, theorganic phosphorous compounds were introduced in smallquantities (1 and 3%) into the coating. We have observed thatthe results were similar for the two concentrations and thus,only those obtained for the concentration of 1% are reportedhere.

Fig. 6presents the evolution of the corrosion potential as afunction of exposure time in the 0.1 M NaCl solution for thecarbon steel covered by the PU coating, by the PU coatingcontaining MOP, by the PU coating containing TDP and bythe PU coating containing the mixture MOP+TDP. For thePU coating with MOP,Ecorr remains stable with immersiontime at a value of−460 mV. For the PU coating containingTDP, during the first 3 days of immersion,Ecorr is shiftedin the cathodic direction (−500 mV), then it progressivelymoves towards and remains at anodic values (−150 mV af-ter 12 days). After 28 days of immersion,Ecorr is stronglyshifted towards the corrosion potential of steel. The an-odic potential values measured between 5 and 21 days of

Page 5: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

134 T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136

-800

-600

-400

-200

00 5 10 15 20 25 30

Co

rro

sio

n p

ote

nti

al /

mV

/SC

E

Immersion time / days

Fig. 6. Corrosion potential versus immersion time in the 0.1 M NaClsolution for the carbon steel covered by: () PU coating, () PU coatingcontaining MOP, () PU coating containing TDP, and () PU coatingcontaining the mixture MOP+ TDP.

immersion indicate passivation of the carbon steel surfacein the presence of TDP in the coating. In the presence ofboth MOP+ TDP in the PU coating, the evolution ofEcorrresembles that obtained in the presence of TDP alone butthe measured values are less anodic than in the case of TDPalone. In this case, competition between the MOP and TDPactions can be assumed with the dominant action of MOPwhich is to attack the carbon steel surface, in agreementwith the synergistic effect discussed previously[6].

For the PU coating alone, for the PU coating containingTDP and for the PU coating containing MOP+ TDP, it canbe observed that during the first 3 days of immersion, thecorrosion potential rapidly moves in the cathodic direction.This means that the coating is porous and allows penetrationof the electrolyte through the coating to reach the substrate.

The impedance diagrams were plotted to characterise thecorrosion resistance of the PU coating containing the phos-phorous compounds. As an example,Fig. 7 presents, forthe carbon steel covered by the PU coating containing TDP,the impedance diagrams obtained with immersion time inthe NaCl solution. Between 1 and 21 days of immersion,the impedance values remained high, which means that theprotective properties of the coating in the presence of TDPare satisfactory.

The variation ofRHF with immersion time for the car-bon steel covered by the PU coating with and without thephosphorous compounds is reported inFig. 8. For the PUcoating alone, the resistance values decreased quickly andlinearly with time showing the continuous degradation ofthe PU coating in the aggressive solution. For the PU coat-ing containing MOP and for the first days of immersion, theresistance value was lower than for the other systems, thenit fell progressively with time. For organic inhibitors used inorganic coatings, Braig[20] said that corrosion inhibition isonly possible when the reaction of the inhibitor with ferrous

0

1000

2000

3000

0 2000 4000 6000

25 Hz

1 Hz 10 mHz

0

1000

2000

3000

0 2000 4000 6000

15.8 Hz

0.63 Hz10 mHz

0

1000

2000

3000

0 2000 4000 6000

25 Hz

1 Hz 10 mHz

Imag

inar

y P

art

/ kΩ

cm

2

Real Part / kΩcm2

(a)

(b)

(c)

Fig. 7. Electrochemical impedance diagrams obtained for the carbon steelcovered with the PU coating containing TDP for various immersion timesin the 0.1 M NaCl solution: (a) 1 day, (b) 7 days and (c) 21 days.

ions occurs faster than rust formation, generating insolubleproduct. Otherwise a shift of the equilibrium creates theopposite effect: an acceleration of metal dissolution. In ourcase, the low resistance value and its decrease with time canbe explained by the aggressive action of MOP on the steelsurface[6]. However, it was observed that in the presenceof MOP in the PU coating, the film was more adherent bycomparison with the PU coating alone. It is known that in acoating, MOP plays the role of an adherence promoter[21].

10-2

10-1

100

101

102

103

104

105

0 5 10 15 20 25 30

RH

F/ k

Ω c

m2

Immersion time / days

Fig. 8. RHF versus immersion time in the 0.1 M NaCl solution for thecarbon steel covered by: () PU coating, () PU coating containingMOP, () PU coating containing TDP and () PU coating containingthe mixture MOP+ TDP.

Page 6: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136 135

For the PU coating containing TDP, the resistance de-creased during the first 5 days, and then increased, whichis unusual, after that it stayed relatively stable for 21 daysof immersion with a high resistance value compared to theother systems. This resistance stability revealed a high pro-tection of the coating in the presence of TDP. The decreaseof the resistance at the beginning of immersion was ac-companied by a shift ofEcorr towards more cathodic values(Fig. 6). This corresponds to the arrival of the aggressivespecies at the metal interface and to the corrosion of thecarbon steel. It was stopped by the presence of TDP onthe surface. This compound forms the insoluble and pro-tective species Fe(TDP)n [6]. Due to the presence of thislayer, the barrier properties of the coating were improved:the values of the corrosion potential became more anodic(Fig. 6) and the resistance increased. After 21 days of im-mersion, the aggressiveness of the NaCl solution led toa degradation of the system: the potential decreased andthe resistance fell abruptly. Whatever, the immersion time,the resistance remained high by comparison with the othersystems.

For the PU coating containing the mixture MOP+ TDP,RHF decreased abruptly during the first 3 days. Then, it re-mained stable for a long immersion times (between 3 and 21days). At the beginning of immersion this behaviour showsthe predominant action of MOP, which was essentially toattack the steel surface. Then, the stability of the resistancecan be attributed to the action of TDP, which reacts with theferrous ions to form the insoluble complex Fe(TDP)n.

From Figs. 6 and 8, it can be concluded that TDP orTDP + MOP incorporated into the PU coating improvedthe protection of the carbon steel. The efficiency of TDPalone was clearly evidenced but its action was limited in thepresence of MOP. After 28 days of immersion, all carbonsteel surfaces were corroded except for those covered by thePU coating containing TDP.

Depending on the organic phosphorous compounds in-corporated into the coating, different electrochemical resultswere obtained, revealing different corrosion performancesincluding adhesion. Thus, it is interesting to analyse theinterface between the carbon steel and the PU coating. XPSanalyses were carried out on the steel surface after the re-moval of the organic coating. As an example, the generalXPS spectrum of the interface steel/coating containing TDPis presented inFig. 9. Whatever the interface studied, phos-phorus, carbon, oxygen and iron were detected. The atomiccomposition of the different interfaces is reported inTable 1.A high percentage of oxygen and iron was noted. This in-dicates that the carbon steel surface was covered by an ironoxide/hydroxide layer[22]. The percentage of phosphoruswas high for the interface steel/coating containing MOP bycomparison with the interface steel/coating containing TDPand the mixture MOP+ TDP revealing an accumulationof MOP at the carbon steel surface. This result can be at-tributed to a fast migration of MOP through the PU coatingprobably during the application of the resin on the carbon

Fig. 9. XPS general spectrum of the steel/coating interface containingTDP.

steel surface before the cross-linking step. This hypothesiswas corroborated by the fact that theTg value is not mod-ified in the presence of MOP. The XPS analysis showedthat the atomic composition at the interface steel/PU coat-ing containing MOP+ TDP was close to that observed atthe interface steel/PU coating containing TDP alone. Thissuggests that the phosphorous compounds migrate untilthey reach the metal substrate. However, from impedancemeasurements, it was seen that the incorporation of bothcompounds MOP and TDP in the coating did not signifi-cantly improve corrosion protection as could be expectedfrom the results obtained previously[6]. This might be ex-plained by the fact that the concentrations of MOP and TDPat the interface were not optimised to lead to the synergis-tic effect. For surface treatments, it has been seen that theconcentration domain for which the synergistic effect is ob-served is very narrow: between 1/5 and 2/5 for MOP/TDP,respectively. It has been clearly demonstrated that for higherMOP concentrations, the synergistic effect was no longerobserved. In the coating, the migration of MOP could befaster than the migration of TDP, as suggested by XPS anal-ysis, so its aggressive action would be predominant at themetal surface. This suggests that the concentration of MOPshould be decreased in order to again find the synergisticeffect on incorporation of the organic compounds into thecoating.

So, it was shown that the corrosion protection affordedby the polyurethane coating containing 1% TDP is high.This result constitutes an advance in the research for newcompounds able to replace the conventional ones.

Table 1Atomic compositions for the interface steel/PU coating, PU coating con-taining MOP, PU coating containing TDP and PU coating containingMOP+ TDP

System P (%) C (%) O (%) Fe (%)

PU – 38 47 15PU containing MOP 9 25 48 18PU containing TDP 1 17 58 24PU containing MOP+ TDP 1 10 59 30

Page 7: Evaluation of corrosion performance of a UV-cured polyurethane coating in the presence of organic phosphorous compounds

136 T.A. Truc et al. / Progress in Organic Coatings 49 (2004) 130–136

4. Conclusion

From electrochemical impedance measurements, it wasfound that the treatment of carbon steel surfaces by aMOP + TDP mixture significantly improves the protectiveproperties of polyurethane coatings. A 60 min treatmentled to the best protection. Although the protection washigh, the treatment time was too long to be retained for anindustrial application. Thus, the phosphorous compoundswere introduced in the coating. The system containing TDPalone gave the highest protection. The presence of MOPin the coating did not improve the corrosion resistance ofsteel even when it was used in the mixture with TDP. Thecompounds migrate through the coating to react at the metalsurface but it was assumed that the concentration ratiosbetween MOP and TDP at the interface were not optimum.

The high performance of the coating containing TDPshows the feasibility of developing new formulations with-out toxic inhibitors.

Acknowledgements

The authors wish to express their gratitude to Dr. G.Chatainier for help with XPS measurements and dataevaluation.

References

[1] J.L. Fang, Y. Li, X.R. Ye, Z.W. Wang, Q. Liu, Corrosion 49 (1993)266.

[2] E. Kalman, B. Varhegyi, I. Bako, I. Felhosi, F.H. Karman, A. Shaban,J. Electrochem. Soc. 141 (1994) 3357.

[3] Y. Gonzalez, M.C. Lafont, N. Pébère, G. Chatainier, J. Roy, T.Bouissou, Corros. Sci. 37 (1995) 1823.

[4] Y. Gonzalez, M.C. Lafont, N. Pébère, F. Moran, J. Appl. Electrochem.26 (1996) 1259.

[5] B. Muller, I. Foster, Corros. Sci. 38 (1996) 1103.[6] T.A. Trinh, N. Pébère, X.H. To, B. Boutevin, Y. Hervaud, Corros.

Sci. 44 (2002) 2055.[7] X.H. To, N. Pébère, F. Dabosi, N. Pelaprat, B. Boutevin, J.P.

Parisi, Proceedings of the Ninth Forum sur les ImpédancesÉlectrochimiques, Paris, 1995, p. 115, Galvano-Organo 659 (1995)760.

[8] X.H. To, N. Pébère, F. Dabosi, N. Pelaprat, B. Boutevin, Y. Hervaud,Corros. Sci. 39 (1997) 1925.

[9] X.T. To, N. Pébère, N. Pelaprat, B. Boutevin, Y. Hervaud, Mater.Sci. Forum 289–292 (1998) 1193.

[10] L. Beaunier, I. Epelboin, J.C. Lestrade, H. Takenouti, Surf. Technol.4 (1976) 237.

[11] F. Mansfeld, M.W. Kendig, S. Tsai, Corrosion 38 (1982) 478.[12] M.W. Kendig, F. Mansfeld, S. Tsai, Corros. Sci. 23 (1983) 317.[13] N. Pébère, T. Picaud, M. Duprat, F. Dabosi, Corros. Sci. 29 (1989)

l073.[14] C. Corfias, N. Pébère, C. Lacabanne, Corros. Sci. 41 (1999)

1539.[15] C. Corfias, N. Pébère, C. Lacabanne, Corros. Sci. 42 (2000)

1337.[16] C. Le Pen, C. Lacabanne, N. Pébère, Prog. Org. Coat. 39 (2000) 167.[17] C. Le Pen, C. Lacabanne, N. Pébère, Prog. Org. Coat. 46 (2003)

77–83.[18] J. Kittel, Thèse de Doctorat de l’Université Paris VI, Paris, 2001.[19] J.F. Mc Intyre, H. Leidheiser, J. Electrochem. Soc. 133 (1986) 43.[20] A. Braig, Prog. Org. Coat. 34 (1998) 13.[21] B. Boutevin, B. Hamoui, J.P. Parisi, J. Appl. Polym. Sci. 52 (1994)

449.[22] N. Ochoa, G. Baril, F. Moran, N. Pébère, J. Appl. Electrochem. 32

(2002) 497.