Cs

AMa

b

c

d

e

a

ARRAA

KCSOSC

1

spum

asts

0d

Electrochimica Acta 54 (2009) 6472–6478

Contents lists available at ScienceDirect

Electrochimica Acta

journa l homepage: www.e lsev ier .com/ locate /e lec tac ta

athodic protection of carbon steel in natural seawater: Effect ofunlight radiation

lessandro Benedetti a,∗, Luca Magagnin b, Francesca Passaretti c, Elisabetta Chelossi d,arco Faimali d, Giampiero Montesperelli e

Istituto per l’Energetica e le Interfasi, IENI – CNR, Milano, via Roberto Cozzi 53 20125 Milano, ItalyDip. Chimica, Materiali e Ing. Chimica G. Natta, Politecnico di Milano, via Mancinelli 7, 20131 Milano, ItalyIstituto per l’Energetica e le Interfasi IENI – CNR, Lecco, c.so Promessi Sposi 29, 23900 Lecco, ItalyIstituto di Scienze Marine, ISMAR– CNR - Via De Marini 6, 16149, Genova, ItalyUniversità di Roma – Tor Vergata, Dipartimento di Scienze e Tecnologie Chimiche, Via della Ricerca Scientifica 00133, Roma, Italy

r t i c l e i n f o

rticle history:eceived 20 October 2008eceived in revised form 5 June 2009ccepted 6 June 2009vailable online 16 June 2009

eywords:athodic protectioneawaterxygen reductionunlight radiationalcareous deposit

a b s t r a c t

Cathodic protection of metals in seawater is known to be influenced by chemical–physical parametersaffecting cathodic processes (oxygen discharge, hydrogen evolution and calcareous deposit precipitation).In shallow seawater, these parameters are influenced by sunlight photoperiod and photosynthetic activity.The results presented here represent the first step in studies dedicated to cathodic protection in shallowphotic seawater. This paper reports on carbon steel protected at −850 mV vs. Ag/AgCl (oxygen limitingcurrent regime) in the presence of sunlight radiation but in the absence of biological and photosyntheticactivity, the role of which deserves future research. Comparison of results obtained by exposing electro-chemical cells to daylight cycles in both biologically inactivated natural seawater and in NaCl 3.5 wt.%solutions showed that sunlight affects current densities and that calcareous deposit interfere with light-currents effects. Sunlight radiation and induced heating of the solution have been separated, highlightingresults not otherwise obvious: (1) observed current waves concomitant with sunlight radiation dependfundamentally on solar radiation, (2) solar radiation can determine current enhancements from early to

late phases of aragonite crystal growth, (3) a three-day-old CaCO3 layer reduces but does not eliminatethe amplitude of the current waves. Theoretical calculations for oxygen limiting currents and additionalfield tests showed that sunlight, rather than bulk solution heating, is the main cause of daily currentenhancements. This was confirmed by polarizations performed at −850 and −1000 mV vs. Ag/AgCl (con-stant bulk temperature), during which the electrode was irradiated with artificial lighting. This test also

be thyer re

confirmed O2 discharge toin the oxygen diffusion la

. Introduction

The protection of metals in seawater is obtained by making thetructure work as a cathode with impressed current or by cou-ling to a less noble metal. The second method is preferentiallysed in seawater, with sacrificial anodes of aluminum-, zinc- oragnesium-based alloys.

Recently, protection techniques involving nontraditional

pproaches such as photo-electrochemistry [1,2] and marineediments [3] have been studied. Whatever the source of elec-rons, the current requirement for the protection of metallictructures in aerobic environments is controlled by oxygen com-

∗ Corresponding author. Tel.: +39 3336196808; fax: +39 2 66173321.E-mail address: alessandro.benedetti@cnr.it (A. Benedetti).

013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.oi:10.1016/j.electacta.2009.06.022

e cathodic process involved. A mechanism of radiation conversion to heatgion is proposed.

© 2009 Elsevier Ltd. All rights reserved.

ing into contact with the surfaces [4]. The oxygen dischargeoccurs via a two-electron mechanism with the formation ofH2O2 as an intermediate until about −900 mV vs. SCE; for highercathodic potentials, it occurs via a four-electron mechanismto form OH− directly [5]. At lower cathodic potentials, a four-electron reduction mechanism was obtained along with H2O2residual formation with a pre-reducing cathodic polarization[6].

The increase of OH− in seawater as a result of cathodic polar-ization shifts the HCO3

−/CO32− equilibrium towards the formation

of CO32−, inducing the precipitation of CaCO3 [7]. Furthermore, if

the pH value of 9.5 is reached through more cathodic polarization,

Mg(OH)2 starts to precipitate [8].

Calcareous deposit growth reduces oxygen diffusion towardsthe metal. As a consequence, the lower current requirement neces-sary for sustaining the protection potential represents an economicadvantage.

imica

pcf

sp

PtciMIc[

hd

ammicabocfc

cpiw

mracHTMcat[

dpo

tttteotdicmrb

A. Benedetti et al. / Electroch

The most protective deposits are obtained at more cathodicotentials [8–10] until the hydrogen evolution starts to mechani-ally compromise the deposited minerals to the point of mechanicalailure [11].

Even if the CaCO3 is supersaturated in seawater [12], variousubstances are a kinetic hindrance preventing CaCO3 spontaneousrecipitation: this process requires energy.

From an electrochemical point of view, the kinetic roles ofO4

3−, organic compounds, Mg2+ [8] and SO42− [13] were inves-

igated. Magnesium ions are able to block the precipitation ofalcite (the most stable CaCO3 phase) [14] by inducing morpholog-cal variations in its crystal geometry [15]. Concerning aragonite,

g2+ inhibits nucleation but not the following crystal growth [16].nvestigations have been carried out to better understand the pre-ipitation of Ca and Mg salts on cathodically protected metals17,18,19].

The protective characteristics of calcareous deposits depend onow the precipitation mechanism interacts with surrounding con-itions.

The effects of hydrodynamic regime [20,21], temperature [13,16]nd pressure [22,23] have been investigated. In natural environ-ents, a particular role is played by biology. At least since theid-seventies [24], it has been known that if no polarization is

mposed, the presence of biofilm strongly affects the electrochemi-al behavior of metals, inducing the ennoblement of active–passivelloys. The underlying mechanism of OCP ennoblement has still noteen fully explained. Consequently, the matter has been the objectf many investigations because of its significance in microbiologi-ally induced corrosion (MIC) [25–31]. Recently, Faimali et al. [32]ound a correlation among depolarization phenomena, bacterialolony growth and different applied potentials.

If a cathodic potential is applied, the interactions amongathodic currents, calcareous deposits and biofilms are highly com-lex. Important results have been achieved by highlighting bacterial

nteractions with cathodically protected pre-fouled metal [33] andith cathodically polarized naked metals [34,35].

To the best of our knowledge, cathodic protection in the environ-ent of photic shallow seawater has not yet been studied. Sunlight

adiation induces daily and seasonal cyclic modifications of physicalnd chemical parameters characterizing shallow seawater. How doathodic processes depending on sunlight radiation itself behave?ow do they behave in relation to induced thermal excursions?he photoperiod induces biologically mediated variation of O2, pH,g2+, PO4

3−, and SO42−, i.e., parameters affecting cathodic pro-

esses at the metal-seawater interface. Furthermore, biofilm, whosectivity can be influenced by sunlight [36], is able to transform ver-ical profiles of current velocity and dissolved oxygen saturation37].

How do the main cathodic processes behave under these con-itions, considering that what happens on a metal surface in theresence of calcareous deposits can be mediated by biological layersnce they colonize the calcareous deposits?

The off-shore oil industry is a very important field concerninghe application of cathodic protection techniques. Cathodic protec-ion in irradiated shallow seawater appears not to be a relevantopic for this field due to the negligible quantitative importance ofhe photic zone with respect to the whole vertical development ofxtractive plants. Nevertheless, in light of the limited informationn the photic zone effects, it would be interesting to characterizehis area and to evaluate its real contribution to CP. In shallow irra-iated seawater, the parameters affecting cathodic processes can be

nfluenced by (1) daily and seasonal radiation and induced heatingycles alone and (2) radiation and induced heating along with theediation of photobiology. Sunlight and temperature can affect the

ate of cathodic reactions, while biofouling can modify the oxygenalance at the electrode surface. In this case, the biolayer can play

Acta 54 (2009) 6472–6478 6473

both passive functions (a mere mechanical shield of O2 diffusingfrom the bulk) and active functions (local O2 balance modificationsthrough photosynthetic activity).

This paper presents field tests in which electrochemical cells areexposed to daylight cycles utilizing biologically inactivated naturalseawater. In this way, control conditions for CP in photic seawa-ter are investigated by focusing on how sunlight radiation affectscathodic seawater electrochemistry, under conditions where pho-tosynthetic processes would normally be stimulated but in theabsence of biological contributions. The role of biofouling will beinvestigated with further experiments.

Particular attention was paid to oxygen, since O2 is simultane-ously the main cathodic reactant at the potential considered anda key molecule in the aerobic environment. Focusing on oxygendischarge is important in order to determine reference conditionsbecause (1) hydrogen evolution as the dominant cathodic processcannot be excluded a priori in the range of CP potentials [38] and 2)physical-chemical parameters affect in different ways the involvedcathodic processes.

2. Experimental

Electrochemical tests were performed at the MARECO IENI-CNRmarine station, located in Bonassola, on the Ligurian Coast of theMediterranean Sea. Far from harbors and close to the nationalMarine Protected Area of “Cinque Terre,” the sea laboratory providesnatural seawater feeding tanks placed both inside the laboratoryand on the roof. Here, photosynthetic biology is allowed to developdue to the tanks’ exposure to the natural photoperiod.

Cathodic polarizations were performed on samples cut from� = 10 mm commercial carbon steel bars, with an exposed area of0.785 cm2. Surface finishing was performed with 120 grit paper.The anode was a platinum wire, and the reference was an Ag/AgClelectrode.

Field polarization tests were performed imposing for 70 h thepotential of −850 mV vs. Ag/AgCl for oxygen limiting discharge cur-rents (iL) as individuated in a dynamic scan (room temperature,1000 mV/h scan rate, explored potential range OCP/−1200 mV vs.Ag/AgCl), as shown in Fig. 7.

The apparatus for electrochemical investigations consisted ofAMEL mod. 2051 potentiostats connected to personal computerwith digital interfaces AMEL mod. 7800.

The experiments were performed in a 60 L tank filled with nat-ural seawater and in a 5 L cell containing NaCl 3.5 wt.%. Both tankswere placed on the laboratory roof and exposed to the photoperiod.All polarizations were performed in quiescent conditions.

The working electrode was placed at an immersion depth ofabout 15 cm, and its surface was turned in order to allow it to receivethe most intense sunlight at the zenith hour. The plane electroderesulted in a 45◦ angle with respect to the water plane. In order tomake the heating caused by the illumination over the NaCl 3.5 wt.%5 L cell comparable to that in the other tank, this cell was placedin a 60 L tank filled with seawater (thermostatic function). Directsunlight started around 8.00 a.m. and lasted until early afternoon(around 2.30 p.m.), ending because of the disappearance of the sunbehind a hill.

The natural seawater was biologically inactivated with anisothiazolin-based additive after the tank was filled, and no fil-tration was made. The treatment is the same as that employedfor corrosion resistance tests in seawater by Gusmano et al. [39].Following the supplier’s (Nalco Italiana) instructions, a concentra-

tion of 5 ppm was used to guarantee a biocide effect. The absenceof electrochemical effects induced by the biocide at the potentialinvestigated here was ascertained by statically polarizing a carbonsteel electrode at −850 mV vs. Ag/AgCl in NaCl 3.5% solution (acqui-sition rate 2 points/s). The test was performed as follows: once the

6 imica Acta 54 (2009) 6472–6478

clsfii4o

l“t

pcAest1ddswv

teJwtcsiutd

vwtpfsos

da

3

3

abh

tts

itN

3.2. Oxygen measurements

When sunlight radiation hits natural seawater, induced thermalexcursions affect oxygen behavior. In general, the dissolved oxygen

474 A. Benedetti et al. / Electroch

ircuit was closed, for the first 80 min the current was left to stabi-ize in quiescent conditions, and at the 80th minute, a mechanicaltirring of the solution was activated. After 40 min of stirring, a suf-cient amount of biocide solution was added in order to have 5 ppm

n the 1 L testing cell. Once the biocide was added, in the following0 min (from the 120th to the 160th minute), no current variationsccurred (data not shown).

The biocide treatment was done to avoid any contribution fromiving photobiology. The seawater thus treated is referred to asPINS” (Photosynthetic Inactive Natural Seawater) throughout theext.

The first session was performed in June. It consisted of a 70 holarization test exposing the NaCl 3.5% and PINS electrochemicalells to the photoperiod with an acquisition rate of 1point/10 min.second session was performed in October in order to separate the

ffects induced by sunlight radiation and heating of the bulk of theolution (acquisition rate 1point/4 min). Separation of T (tempera-ure) and light effects was accomplished by covering the tank for5 min twice a day between 8.00 a.m. and 14.00 p.m. for all threeays of the experiment. It was expected and practically verified thaturing the two sunlight interruptions, the bulk temperature of theolution at the electrode depth would remain constant due to highater thermal inertia. This allowed the attribution of all current

ariations to the temporary absence of radiation during coverage.The dissolved oxygen concentrations and concomitant tempera-

ures were manually measured with a Delta Ohm DO 9709 O2 probevery two hours from dawn to around midnight in the days of theune session. Because of the hydrodynamic perturbation connected

ith the modality in obtaining dissolved oxygen concentrations,hese O2 data were collected not directly in the electrochemicalell but in a control tank (named “PINS O2 control”) exposed to theame conditions. In order to compare the measurements obtainedn the “PINS O2 control” solution with the measurements in a nat-ral seawater environment, O2 concentrations were measured inhe sea area in front of the laboratory as well (calm wave, ∼10 cmepth, ∼5 meters off the shore).

Finally, laboratory tests were performed at −850 and −1000 mVs. Ag/AgCl with artificial light illumination. The acquisition rateas 1point/2 s. The two PAR (Photosynthetic Active Radiation) and

wo UVA lights, 9 W each, composing the irradiation unit werelaced at a distance of 5 cm above the electrode, whose surface

ormed a 45◦ angle with incident light (other angles were not con-idered here). Once the cell was closed, the lamps were switchedn at the 45th minute, and the irradiation lasted 5 min. The bulkolution temperature of 23 ◦C was not altered during illumination.

The morphology of calcareous deposits and properties obtaineduring polarization performed in June were investigated by SEMnd EDS.

. Results and discussion

.1. Field polarization data

Fig. 1 shows potentiostatic polarization data obtained in the PINSnd NaCl 3.5% tanks, both exposed to the natural photoperiod. Theulk temperatures of the solutions ranged from 18 ◦C (early dawnours) to 28 ◦C (early afternoon, before shade).

The highest current densities were obtained in the NaCl 3.5%ank, and they did not show any general downward trend. In con-rast, PINS current densities progressively decreased and reached

teady state values after 24/30 h.

Both curves exhibited current waves concomitant to sunlightllumination, but the amplitudes of the NaCl 3.5% waves were higherhan the amplitudes obtained in the PINS solution. Furthermore, theaCl 3.5% chronoamperometry data showed a wider dispersion in

Fig. 1. Cathodic current densities at −850 mV vs. Ag/AgCl in NaCl 3.5 wt.% and pho-tosynthetically inactive natural seawater (PINS) solutions. The electrochemical cellswere exposed to daylight cycles.

the hours where the electrode was not directly illuminated. Thisphenomenon will be an object of further investigations.

The decrease of current densities in natural seawater is relatedto the formation of a calcareous deposit as a consequence of alkalin-ity produced during cathodic oxygen discharge [7,8]. As previouslyinvestigated as previously shown [22], the comparison betweenresults obtained in only conductive solutions (NaCl 3.5%) and bio-logically inactivated natural seawater (PINS) solutions allows thequantification of the calcareous deposit’s effects. Under the condi-tions that we studied, the calcareous deposit provided an additionalresistance able to reduce current density to around 10 �A/cm2

(Fig. 1, PINS curve) during the dark hours once the steady statewas reached. In the absence of calcareous deposit and during thesame hours, the NaCl 3.5% currents were around 30 �A/cm2. Thecalcareous deposit obtained in the PINS solution was composed ofcauliflower-dendritic shaped aragonite crystals (Fig. 2).

Fig. 2. Calcareous deposit precipitated on carbon steel cathodically polarized for70 h at −850 mV vs. Ag/AgCl in photosynthetically inactive natural seawater (PINS)solution 4000 × Magnification. The electrochemical cell was exposed to daylightcycles.

A. Benedetti et al. / Electrochimica

Fs

cnd

gttpcfmpdde

ich

3

mst

[i(cm

odemTt(fsi

ig. 3. Bulk T and [O2] daily path as measured in “PINS O2 control” tank and in theea area in front of the laboratory (calm wave conditions).

oncentration depends positively on photosynthetic activity andegatively on temperature, which, in turn, positively affects oxygeniffusion in solutions.

In order to produce data allowing theoretical evaluations of oxy-en limiting current (iL), oxygen concentration [O2] and relatedemperatures T in the bulk of the solution were measured inhe “PINS O2 control” tank. T and [O2] measurements were alsoerformed in the seawater area in front of the laboratory foromparative purposes. As shown in Fig. 3, in the seawater inront of the laboratory, oxygen concentration exhibited a maxi-

um during the day and a minimum during the night due to theositive effect of photosynthetic activity prevailing over the [O2]ecrease induced by heating. In the “PINS O2 control” tank, theaily [O2] trend was inverted due to the presence only of thermalffects.

In biologically inactive seawater, though higher temperaturesnduce lower oxygen concentrations, the prevalence of the diffusionoefficient produces higher oxygen limiting currents. This aspectas been experimentally verified in previous work [14].

.3. Oxygen limiting current calculations

Oxygen concentrations in the “PINS O2 control” tank (effectivelyeasured) and oxygen concentrations in the PINS and NaCl 3.5%

eawater solutions were supposed to be equal, since the same solu-ions were exposed to the same environment.

Hence, calculations for iL were made starting from bulk T andO2] data measured in the “PINS O2 control” tank. The calculatedL values were compared with the experimental NaCl 3.5% curveFig. 1) in order to verify whether bulk [O2] and T modificationsould be considered the cause of observed daily current enhance-ents.

We then investigated the effects of solar radiation on dissolvedxygen. In the troposphere (0/10 km), molecular oxygen can absorbirectly in the region near IR and indirectly in natural waters vianergy transfer through the excitation of mediators like Chro-ophoric Dissolved Organic Matter (CDOM) and chlorophyll [40].

hese substances, absorbing in the UV–vis region, can promote

he formation of singlet oxygen (excited state) from triplet oxygenground state), making the molecule a reactive species able to trans-er its excess energy even to biological targets. Though O2 energetictatus can be affected by radiation absorbance, it is not believed tonfluence the dissolved oxygen concentration and diffusion (i.e., O2

Acta 54 (2009) 6472–6478 6475

limiting current parameters). The former depends on temperatureand salinity [41], and the latter depends on temperature [42].

3.3.1. Role of bulk solution heatingThe iL values were computed with a focus on bulk induced heat-

ing and related O2 concentration.The expression for oxygen limiting current is

iL = nFCDO2

ı(1)

where “n” is the number of involved electrons (n = 4), “F” is theFaraday constant (96500 C/mol e−), “DO2” is the oxygen diffusioncoefficient (m2/s), “C” is the oxygen concentration in the bulk(mg/L), and “ı” is the diffusion layer thickness for oxygen (m).

Eq. (1) allowed the calculation of iL with tabulated values forDO2 = f(T) and ı = f(T;[O2]) [42].

The reciprocal consistency of computed iL and NaCl 3.5%chronoamperometry data had to be ascertained since Eq. (1) pro-vides the fraction of oxygen discharge current density, while overallcathodic current density is i = iO2 + iH2, i.e., the sum of the oxygen(diffusive) and hydrogen (non-diffusive) terms.

In the absence of water turbulence, the iO2iH2

ratio depends on theapplied potential, pH and T.

At the potential investigated here (−850 mV vs. Ag/AgCl), dur-ing the photoperiod, the bulk solution temperature ranged between18/28 ◦C (Fig. 3). Therefore, temperature enhancements gave hydro-gen evolution a relative advantage with respect to oxygen dischargeduring the warmer hours, since the former is thermally activatedand the latter undergoes temperature-mediated opposite effects,as previously mentioned.

Comparison between the calculated iL and actual “i” shows thatboth these currents have to be considered both as iL. Hence, theiO2 > iH2 condition has to be satisfied during the warmer hours too.This aspect was verified as follows.

The bulk pH of the PINS solution is buffered at 8.2 [11], and thepH on the surface electrode can be assumed to be around 9: in ourdeposit, no brucite (precipitation pH∼=9.5 [8]) was found, and pH∼=9was reached at current densities and potentials typical of iL [11].Nevertheless, seawater is buffered due to the HCO3

−/CO32− system,

unlike the NaCl 3.5% solution. Some authors [43] have highlightedthe buffering effects of carbonated solutions (pH 8.3, seawater con-centrations) compared to non-buffered solutions. In the iL potentialwindow and non-buffered solutions, the cited authors found pH tobe slightly above 10. Therefore, under our experimental conditions,the interfacial pH in the NaCl 3.5% solution was considered to be9/10.

Hence, at 25 ◦C, with a slope of cathodic hydrogen overpoten-tial of −120 mV/decade, a hydrogen exchange current on iron of1 �A/cm2 [42], and a hydrogen equilibrium potential at pH 9÷10in a range of −800/−850 mV vs. Ag/AgCl, the hydrogen cathodiccurrent at −850 mV vs. Ag/AgCl was considered to be 1/10 �A/cm2.

This value for iH2 can be considered comparable to the value ofiH2 expected in field tests during warmer hours, when the relativeiH2 contribution is higher. Nevertheless, Fig. 1 shows that duringthis hour, iO2 + iH2∼=60 �A/cm2, leading us to assume the iH2 partto be negligible with respect to iO2.

Comparison between experimental NaCl 3.5% and calculated iLvalues suggested that bulk [O2] and T modifications induced bydirect sunlight cannot fully explain the measured current waves(see Fig. 4). An indirect verification can be approached by askingwhat temperature the bulk should possess to gain doubled cur-

rents (as for the NaCl 3.5% curve, Fig. 1) compared to bulk T and[O2] measured during early dawn hours.

Bulk T and [O2] experimental data provided an empirical O2 sol-ubility relation of [O2] = f(T), exhibiting a dependence law similarto reference data [44] with a weak quantitative difference (Fig. 5).

6476 A. Benedetti et al. / Electrochimica Acta 54 (2009) 6472–6478

Fc

I[ttin

dgdirf

p

i

Fm

ig. 4. a. Current densities at −850 mV vs. Ag/AgCl in NaCl 3.5% compared with iL

urve calculated from bulk T and [O2] data collected in “PINS O2 control” tank.

n conjunction with D/ı data tabulated at different temperatures42], it was possible to argue that to justify a doubled current gain,he bulk solution should possess much higher temperatures thanhe actual maximum measured during the zenith hour. Thus, it isndirectly confirmed that bulk T and [O2] modifications alone doot explain the observed daily current enhancements.

It is important to remember that in the presence of a calcareouseposit, oxygen access to free metal sites is controlled by the oxy-en diffusion layer and the thickness and porosity of the calcareouseposit. In particular, the calcareous deposit can become the limit-

ng factor for mass transport even irrespectively of hydrodynamicegimes, as the aragonite porosities can become the obligatory pathor O2 [20].

The limiting role played by calcareous deposit thickness andorosity is described in eq. (2) (eq. (1) modified) [45]:

L = nFCı

DO2+ t

pDO2

(2)

ig. 5. Comparison of oxygen solubility between literature reference and experi-ental data collected in “PINS O2 control” tank exposed to daylight cycles.

Fig. 6. Current densities at −850 mV vs. Ag/AgCl in NaCl 3.5 wt.% and PINS solutionsunder sunlight illumination with two artificial coverings lasting 15 min each. Thefirst wave (first day) and the third wave (third day) are presented here. The second(not shown) lies in the middle.

where “t” is the calcareous deposit thickness. The O2 diffusion coef-ficient “DO2” through the calcareous deposit must be weighted withmineral layer porosity “p” (p = Volumevoids

VolumeCaCO3layer). In our experiment,

aragonite layer thickness was assessed to be not larger than a few�m, based on rare visible uncovered metal sites and calcareousdeposits of generally regular topography (SEM observations). Eq.(2)), rewritten as iL = nFC×pDO2

pı+t, shows that calcareous deposit thick-

ness, especially if modest, is of less importance to iL than its porosity,as experimentally observed and shown in Figs. 1 and 2.

3.4. Separation of light and temperatures effects

3.4.1. Field testThe comparison between experimental NaCl 3.5% data and iL cal-

culated with bulk [O2] and T suggested that the heating induced bysolar radiation cannot full explain the observed trends of currentsunder the photoperiod.

A test was performed in order to verify the role of sunlight andinduced heating in the presence or absence of a calcareous deposit.For this purpose, the tank was covered twice a day (15 min eachcovering) for three days. Thus, interactions between sunlight andcurrent densities in the presence of a well-grown calcareous deposit(current wave registered during the 3rd day) and those in the pres-ence of an immature calcareous deposit (wave during the 1st day)were characterized. Fig. 6 shows how both in the absence (NaCl 3.5%solution) and in the presence (PINS solution) of a calcareous deposit,while sunlight was obscured the daily current bells lost the largestpart of their amplitude. Hence, the determinant role of solar radia-tion even in the presence of an aragonite layer was experimentallyverified.

3.4.2. Laboratory testThe protection potential investigated here was selected in order

to place attention on oxygen discharge, and iL computations wereperformed assuming oxygen involvement. Examining O2 and H2

branches belonging to the experimental cathodic polarization curve(see Fig. 7) and considering oxygen discharge to be the processinvolved, in the presence of illumination, the relative iO2 contri-bution to overall cathodic current decreases with the potential (seeFig. 7). This is due to the rapid increase of iH2 contribution to the

A. Benedetti et al. / Electrochimica

F1−

oat

iat−adebb

fiiTis

3r

at

0ieitTlfs

emril

[48].Analogously, the photons scattering in mineral layer architec-

ture towards polarized free metal sites would be converted intolocal heat with improved oxygen diffusion phenomena. In the first

ig. 7. Potentiodynamic curve on carbon steel in natural seawater (scan rate000 mV/h) and evaluations of light effects over oxygen discharge at −850 and1000 mV vs. AAC.

verall current density at more cathodic potentials. The points �nd � in Fig. 7 would belong to the cathodic curve under irradia-ion.

On the basis of these considerations, potentiostatic polar-zations at −850 and −1000 mV vs. Ag/AgCl were performed,nd current paths under artificial irradiation with constant bulkemperature were followed (Fig. 8a). Fig. 8b shows how at1000 mV vs. Ag/AgCl the current gain is less important thant −850 mV vs. Ag/AgCl. This result confirmed 1) that oxygenischarge is the involved cathodic process rather than hydrogenvolution and 2) that current enhancement under irradiation cane obtained in the iL potential window with constant T in theulk.

These observations suggest that, far from the bulk, inter-ace processes are involved. The radiation is probably convertednto additional heat at the metal/solution interface, induc-ng thermal effects and locally affecting O2 limiting current.he observed current plateau probably represents the largest

nfluencing effects induced by energy coming from the lightource.

.4.3. Interactions between calcareous deposit and sunlightadiation

Why are current enhancements still visible (even if of lowermplitude) in the presence of a mineral layer, and why does radia-ion still remain the main cause?

The specific heat capacities of iron and CaCO3 are respectively.4 and 0.8 kJ/K × kg [46], meaning that the same radiant energy

s converted into heat more efficiently in iron than in CaCO3. Nev-rtheless, metal and mineral deposits do not present themselvesn the same conditions against incident photons. More precisely,he first target that photons encounter is the aragonite deposit.his target has a surface that is 3D developed, more complex and

arger than the underlying surface offered by the metal (a 2D sur-ace and smaller overall surface area consisting of fragmented freeites).

Thus, following the metal radiation-to-heat conversion hypoth-sis, two mechanisms concerning the individual involvements of

etal and of CaCO3 can be imagined: (1) focusing on CaCO3, the

adiation hitting the calcareous deposit is converted into heat thats directly transferred into the oxygen diffusion layer, excludingight/metal interaction, and (2) focusing on metal, the carbonate

Acta 54 (2009) 6472–6478 6477

layer serves merely to let the photons scatter towards underlyingfree metal sites. Here, light/metal interaction takes place, inducingradiation conversion into heat. A mechanism can also be consideredinterlacing the previous two through balances of different contri-butions.

The possibility of photons being scattered by the calcareousdeposit towards free metal sites finds support in the literature. Ina recent work, Enriquez et al. [47] measured the absorption spec-tra of intact corals using the Caribbean scleratinian Porites branneri.Their research reports that laminae cut from the examined speciesabsorb light with efficiency up to five times greater than theirfreshly isolated algal symbionts. This observation was explainedby the increased absorption coefficients of photo-protective pig-ments when solar radiation and coral skeletons interact. Thishigher light absorbance, as reported in the cited reference, canbe explained in terms of multiple light scatterings taking placein the aragonite coral structure, since aragonite coral skeletonsexhibit low absorption and are efficient Lambertian scatterers

Fig. 8. (a) Oxygen current path in presence of artificial light (2 PAR lamps + 2 UVAlamps, 9 W/lamp, 5 cm distance) during potentiostatic polarization performed at−850 and −1000 mV vs. Ag/AgCl. (b) Values of data presented in (a) are normalizedwith reference to the value immediately preceding the lighting of the lamps.

6 imica

cese

lbintmvnli

ioe

ttcatcpcemwhwTg

4

vc

-

-

-

-

[[

[[[[[

[[[[[[[[

[[[

[

[[

[

[[

[

[

[[

478 A. Benedetti et al. / Electroch

ase, the target of the photon flux is an adapted photosensitivequipment, while in the second, is a cathodically polarized carbonteel electrode that probably undergoes scattered light mediatedffects.

Of course, hypotheses concerning the local conversion of sun-ight into heat with full, partial or no metal involvement need toe supported or falsified by additional experiments. Topics to be

nvestigated could be (1) the relationship between deposit thick-ess and light effect for time windows longer than 70 h (supposinghat a thicker deposit reduces the radiation effect), (2) measure-

ents with microprobes in order to assess near-metal T and [O2]ariations with respect to bulk, (3) light effects in quiescent andon-quiescent conditions in the absence and presence of a mineral

ayer, and (4) interference of the biolayer affecting aragonite/lightnteractions.

Moreover, all these aspects must be evaluated considering thatn open seawater the relationships among current density, calcare-us crystal growth and incident light are also a function of lightxtinction at depths >1 m. [49]

Finally, since no previous works regarding cathodic protec-ion in shallow irradiated seawater were found in the literature,he investigation of control conditions is here approached con-erning the photoperiod effects on cathodic protection in thebsence of a biologic components. In particular, attention was paido the simplest configuration in this scenario, i.e., oxygen dis-harge in irradiated but not biologically active seawater in theresence and absence of growing calcareous deposits in quies-ent conditions. Thus, other influences (photosynthetic biofouling,tc.) must be evaluated against this background. Further experi-ents about control conditions would include mainly wider timeindows and near-electrode investigations for a better compre-

ension of how light induces current enhancements (with orithout calcareous deposits) at O2 current limiting potentials.

hereafter, photosynthetic biolayer effects should be investi-ated.

. Conclusions

Carbon steel was cathodically protected in seawater at −850 mVs. Ag/AgCl (oxygen limiting currents) and by exposing electro-hemical cells to daylight cycles for three days.

In our experimental conditions, it was observed that:

sunlight illumination hitting the polarized electrode inducedcurrent enhancements both in the presence and in theabsence of a calcareous deposit. In the absence of a cal-careous deposit, the amplitude of the current waves waslarger.Oxygen limiting current calculations and field experimental veri-fication showed that bulk heating induced by sunlight irradiationdoes not explain the observed current waves; instead, the sunlightradiation itself plays the major role.Static polarizations performed at −850 and −1000 mV vs. Ag/AgClwith artificial illumination confirmed that radiation enhances therate of oxygen cathodic discharge when bulk temperature is con-stant.A mechanism is proposed: once radiation hits the electrodesurface, light conversion to heat occurs in the oxygen dif-fusion layer region, inducing temperature-related effects on

oxygen limiting currents. In the presence of aragonite, pho-tons scatter in the mineral layer’s porous structure, probablyeven towards underlying free metal sites, and are locally con-verted into heat as well, even if less efficiently (lower currentenhancements).

[[[

[

Acta 54 (2009) 6472–6478

Acknowledgements

The authors sincerely thank the following for their con-tributions: Alfonso Mollica (ISMAR-CNR), Marco Pini, GiordanoCarcano, Stefano Besseghini, Emilio Olzi (IENI-CNR, Lecco), RobertoMarangoni, Francesco Ghetti, Lorenzo Fulgentini (IBF-CNR, Pisa),Fabio Bolzoni (Dip. Chimica, Materiali e Ing. Chimica G. Natta,Politecnico di Milano), Gildo Storni, Giorgio Perboni and CristianiPierangela (CESI-Ricerche, Milano)

References

[1] H. Park, K.Y. Kym, W. Choi, J. Phis. Chem. B 106 (2002) 4775.[2] G.X. Shen, Y.C. Chen, C. Lin, Thin Solid Films 489 (2005) 130.[3] L.H. Orfei, S. Simison, J.P. Busalem, Environ. Sci. Technol. 40 (2006) 6473.[4] T. Foster, J.G. Moores, Corrosio/86, paper n. 294, Huston, Texas, 1986.[5] D. Gervasio, I. Song, J.H. Payer, J. Appl. Electrochem. 28 (1998) 979.[6] N. Le Bozec, C. Compere, M. L’Her, A. Laouenan, D. Osta, P. Marcus, Corrosion

Sci. 43 (2001) 765.[7] C. Barchiche, C. Deslouis, D. Festy, O. Gil, P. Refait, S. Touzain, B. Tribollet, Elec-

trochem. Acta 48 (2003) 1645.[8] W.H. Hartt, C.H. Culberson, S. Smith, Corrosion-NACE 40 (1984) 609.[9] W. Mao, W.H. Hartt, Corrosion/85, paper n. 317, Boston, Massachusetts, 1985.

[10] J.S. Luo, R.U. Lee, W.H. Hartt, S.W. Smith, Corrosion Sci. 47 (1991) 189.[11] G. Salvago, S. Maffi, L. Magagnin, A. Benedetti, S. Pasqualin, E. Olzi, Proceedings

of 13th International Offshore and Polar Engineering Conference, Honolulu,Hawaii, USA, 2003.

12] E.K. Chave, E. Suess, Limnol. Oceanogr. 4 (1970) 633.13] C. Barchiche, C. Deslouis, O. Gil, P. Efait, B. Tribollet, Electrochim. Acta 49 (2004)

2833.[14] A. Gutjar, H. Dabringhaus, R. Lacmann, J. Crystal Growth 158 (1996) 310.[15] Y. Zhang, R.A. Dawe, Chem. Geol 163 (2000) 129.[16] S.H. Lin, S.C. Dexter, Corrosion Sci., 44 (19886) 15.[17] C. Deslouis, D. Festy, O. Gil, G. Rius, S. Touzain, B. Tribollet, Electrochim. Acta 43

(1998) 1891.[18] C. Deslouis, D. Festy, O. Gil, V. Maillot, S. Touzain, B. Tribollet, Electrochim. Acta

45 (2000) 1837.[19] A. Neville, A.P. Morizot, J. Crystal Growth 243 (2002) 490.20] R.U. Lee, J.R. Ambrose, Corrosion/86, paper n. 292, Huston, Texas, 1986.21] K.E. Mentel, W.H. Hartt, T.Y. Chen, Corrosion Sci. 48 (1992) 489.22] S. Chen, W.H. Hartt, Corrosion 58 (2002) 38.23] S. Chen, W.H. Hartt, Corrosion 59 (2003) 721.24] A. Mollica, A. Trevis, Proceeding of 4th International Congress on Marine Cor-

rosion and Fouling, Antibes, Juan-Les-pins, France, 1976.25] H. Videla, Int. Biodeter. Biodegr. 34 (1994) 245.26] A. Mollica, Int. Biodeter. Biodegr. 29 (1992) 213.27] G. Salvago, L. Magagnin, Corrosion Sci. 57 (2001) 680.28] G. Salvago, L. Magagnin, Corrosion Sci. 57 (2001) 759.29] S. Motoda, Y. Suzuky, T. Shinohara, S. Tsujikawa, Corrosion Sci. 31 (1990) 515.30] W. Wang, J. Wang, X. Li, H. Hu, J. Wu, Mater. Corrosion 55 (2004) 30.31] B.J. Little, J.J. Lee, R.I. Ray, Corrosion Sci. 62 (2006) 1006.32] M. Faimali, E. Chelossi, F. Garaventa, C. Corrà, G. Greco, A. Mollica, Electrochim.

Acta 54 (2008) 148.33] S.C. Dexter, S.H. Lin, Int. Biodeter. Biodegr. 29 (1992) 231.34] J.P. Busalmen, S.R. de Sanchez, Appl. Environ. Microbiol. 71 (2005) 6235.35] S.G. Gomez de Saravia, M.F.L. de Mele, H.A. Videla, R.G.J. Edyvean, Biofouling 11

(1997) 1.36] L. Alonso-Saez, J.M. Gasol, T. Lefort, J. Hofer, R. Sommaruga, Appl. Environ.

Microbiol. 72 (2006) 5806.37] S. Nakano, A. Takeshita, T. Ohtsuka, D. Nakai, Limnology 7 (2006) 213.38] DNV – Recommended Practice RP-B401, Cathodic Protection Design, January

2005.39] G. Gusmano, G. Montesperelli, G. Forte, E. Olzi, A. Benedetti, Desalination 183

(2005) 187.40] R.G. Zepp, N.L. Wolfe, G.L. Baughman, R.C. Hollis, Nature 267 (1977) 421.41] APHA (American Public Health Association). Standard methods for the exami-

nation of water and wastewater, 18th edition, Washington, DC, 1992.42] P. Pedeferri, Corrosione e protezione dei materiali metallici, Polipress 2007 –

Politecnico di Milano, Piazza Leonardo da Vinci n.32, 20133 Milano.43] C. Deslouis, G. Frateur, G. Maurin, B. Tribollet, J. Appl. Electrochem. 27 (1997)

482.44] http://www.engineeringtoolbox.com/oxygen-solubility-water-d 841.html.45] J.R. Scully, H.P. Hack, D.G. Tipton, Corrosion/85, paper n. 214, Boston, Mas-

sachusetts, 1985.

46] http://www.engineeringtoolbox.com/specific-heat-solids-d 154.html.47] S. Enriquez, E.R. Mendez, R. Iglesias-Prieto, Limnol. Oceanogr. 5 (2005) 1025.48] W.J. Smith, Modern Optical Engineering: The Design of Optical Systems,

McGraw-Hill Inc., New York, 1966.49] GESAMP. The Sea-Surface Microlayer and its Role in Global Change. In GESAMP

Reports and Studies n. 59, WMO, 1995.