当前位置：文档库 › Mg-Zr-alloy-polymer-with-sodium-fluoride-inhibitor_2012_Surface-and-Coatin

Mg-Zr-alloy-polymer-with-sodium-fluoride-inhibitor_2012_Surface-and-Coatin

Mg –Zr alloy behavior in basic solutions and immobilization in Portland cement and Na-geopolymer with sodium ?uoride inhibitor

David Lambertin ?,Fabien Frizon,Florence Bart

CEA,DEN,DTCD,SPDE,F-30207Bagnols-sur-Cèze,France

a b s t r a c t

a r t i c l e i n f o Article history:

Received 10January 2012

Accepted in revised form 4May 2012Available online 11May 2012Keywords:Mg –Zr alloy

Electrochemistry Hydrogen evolution Corrosion inhibitor Encapsulation

The dismantling of uranium natural graphite gas nuclear reactor generates a large volume of fuel cladding.The fuel cladding materials are based on Mg –Zr alloy for UNGG.The dismantling strategy could be to encap-sulate these wastes into an ordinary Portland cement (OPC)or Na-geopolymer (alumino-silicate material)in a form suitable for storage.Corrosion behavior of Mg –Zr in OPC interstitial solution and activating solution of Na-geopolymer has been studied in the presence and absence of sodium ?uoride as corrosion inhibitor.Elec-trochemical methods have been used to determine the corrosion densities.Results show that the corrosion densities of Mg –Zr alloy in OPC solution are one order of magnitude more important than in activating solu-tion of Na-geopolymer and sodium ?uoride addition decreases corrosion densities in OPC interstitial solution.Hydrogen evolution of encapsulated Mg –Zr alloy has also been measured in both OPC and Na-geopolymer and results show that Na-geopolymer matrix appears to be an attractive binder in term of corrosion performance.

1.Introduction

The dismantling of gas cooled nuclear reactors,like UNGG (uranium natural graphite gas)in France or MAGNOX design in Great Britain,generates a large volume of fuel cladding.The fuel cladding materials are based on Mg –Zr alloy for UNGG [1]and Mg –Al alloy in Magnox:both claddings have been developed in the 1960s for corrosion and mechanical resistance at 500°C under CO 2atmosphere [2].

Since 1990,scrap metals from the MAGNOX reactor were packed with a mixture of Portland cement and blast furnace slag [3–5].The strategy of this formulation is to use a binder with a high pH,around 12.5,mixed with a minimum amount of water [4].Nevertheless,hy-drogen gas generation has been observed simultaneously with the for-mation of magnesium hydroxide on the surface of magnesium alloy,causing damage to the package [6].An alternative to the use of Portland cement has been proposed with the use of mineral geopolymer [3,7]or organic polymers [8]to limit the amount of water to avoid a hydrogen gas.Geopolymer materials present good ?uidity [9]that makes them compatible with industrial processes for embedding solids with com-plex form,such as fuel claddings.Concerning the conditioning of Mg –Zr alloy from UNGG reactors,the strategy could be relatively similar,i.e.an encapsulation of these wastes into a mineral binder,namely an ordinary Portland cement (OPC)or geopolymer (alumino-silicate ma-terial)in a form suitable for storage.However,due to the different

alloys composition in UNGG claddings,the corrosion behavior is expected to be relatively different from the MAGNOX materials.

On another hand,it is well known that ?uoride ion is a good corro-sion inhibitor for magnesium and its alloys [10–16].A study on pure Mg in alkaline ?uoride solutions [17]has been performed and pointed out that small additions of ?uoride (b 10?1M)decrease the corrosion rate.Then the corrosion densities increase with increasing ?uoride con-centration until limit content from where corrosion densities decrease sharply.Fluoride conversion coatings with Mg have been studied with XPS,SEM and XRD methods [11,12]and showed that a ?uoride-containing ?lm formed on the magnesium surface and more precisely KMgF 3or MgF 2compounds as a function of ?uoride concentration.

Therefore,it could be relevant to introduce some ?uoride in the en-capsulation materials in order to lower Mg-alloy corrosion and de-crease the hydrogen production in the mineral matrices.

Few studies have been published on corrosion behavior of fuel clad-ding based on Mg –Zr alloy [1,13,18].

The purpose of the present work is to study the corrosion behavior of Mg –Zr alloy using electrochemical methods and hydrogen evolution tests ?rstly in interstitial solution of OPC and activating solution of Na-geopolymer with additions of sodium ?uoride corrosion as inhibitor and secondly embedded in OPC paste and Na-geopolymer binders.2.Experimental

2.1.Samples,solutions and binders

The material used in the corrosion tests was as-cast Mg –Zr alloy in-active fuel cladding plates from UNGG reactor.The weight composition

Surface &Coatings Technology 206(2012)4567–4573

?Corresponding author.Tel.:+33466791480;fax:+33466797871.E-mail address:https://www.wendangku.net/doc/f911479459.html,mbertin@cea.fr (D.

10.1016/j.surfcoat.2012.05.008

Contents lists available at SciVerse ScienceDirect

Surface &Coatings Technology

j o u r n a l h o m e p a ge :w w w.e l s e v i e r.c o m/l o c a t e /s u r f c o a t

of alloy is Mg(99.49%),Zr0.49(%wt.)and Mn0.016(%).The prepara-tion and microstructure of Mg–Zr alloy for fuel cladding have been de-scribed by J.Blanchet in1960s[1,13].Recently,J.Li et al.[19]have shown that the grain size of the Mg–Zr alloy was signi?cantly reduced due to Zr addition and a?ne grain structure was achieved when the zirconium concentration was above0.4wt.%.

This material has been stored during a long time and a passive layer was formed at the surface,so the samples were cut in order to obtain a rectangular shape before being pickled in1M H2SO4solution for20s and rinsed with distilled water.

The alkaline solutions were prepared by dissolving alkali hydroxide NaOH,KOH or Ca(OH)2(Prolabo,Rectapur,98%)pellets,and amor-phous silica provided by Rhodia(Tixosil38)in Milli-Q water,following

the composition summarized in Table1.The chemical composition of the anhydrous binders determined by X-ray?uorescence(XRF)can be found in Table2and the formulations used for corrosion rate mea-surements are reported in Table1.The metakaolin used was purchased under the brand name of Pieri Premix MK from Grace Construction Products.

2.2.Electrochemical measurements

A classical three electrodes cell was used for the electrochemical in-vestigations.A platinum counter electrode and a saturated calomel ref-erence electrode(SCE)were used.The working electrode is made of Mg–Zr pieces partially recovered of epoxy resin for a well de?ned sur-face.The experiments were performed with a Princeton Applied Re-search Potentiostat VersaSTAT model.

The potentiodynamic polarization tests were performed after an open circuit potential(OCP)stabilization during24hours.Potentiodynamic polarization was performed starting at?0.4V/OCP until+0.7V/OCP. The scanning rate was1mV/s.

2.3.Hydrogen evolution test

Measurements of hydrogen evolution are initiated by placing Mg–Zr alloy in various alkaline solutions or binders in a?xed volume con-tainer under nitrogen at room temperature(20°C)with a70%of rela-tive humidity.The hydrogen analysis is undertaken by VARIAN model CP3800gas chromatography with Galaxie software and the time de-pendant results are used to determine corrosion rates.

In order to measure the hydrogen evolution,Mg–Zr alloy pieces were added in freshly made OPC or Na-geopolymer binder and hydro-gen analysis has been operated during90days.

2.4.Optical microscope and SEM observations

After immersion in alkaline solutions,sample surface morphologies were observed using scanning electron microscopy FEI Inspect S50.At 90days,binders with encapsulated Mg–Zr alloys were cut and ob-served with an optical microscope NIKON Eclipse LV100coupled with

a digital camera NIKON DS-Fi1.

2.5.Corrosion rates determinations

The fundamental measurement of the corrosion rate is the metal weight loss rate,ΔW,(mg/cm2/day)and can be converted to an aver-age corrosion rate,P W,(mm/year)using[20–22]:

P W?2:10ΔWe1TIn the overall corrosion reaction of pure Mg,one molecule of hydro-gen is evolved for each atom of corroded Mg.One mole(i.e.24.31g)of Mg metal corrodes for each mole of hydrogen gas produced.Therefore,

Table1

Chemical composition of alkaline solutions and formulation of OPC and geopolymer binders(g/L).

Products OPC interstitial

solution Activating solution

of geopolymer

OPC Geopolymer

Ca(OH)21–NaOH10370 KOH3–SiO2–333

CEM I52.5 PM ES(c)Water/ c=0.4

Metakaolin (MK)Water/ MK=0.81

SiO2333 NaOH370Table2

Chemical composition of CEMI and metakaolin(%).

CEMI52.5PM ES CP2(Lafarge,Le Teil factory)Metakaolin Pieri Premix

CaO67.41

SiO222.8454.4

Al2O3 2.738.4

Fe2O3 1.84 1.27

MgO0.81

Na2O0.14

K2O0.230.62

SO3 2.23

TiO2

1.6

Fig.1.The appearance of Mg–Zr alloy at24hours and2months after immersion in pore solution of OPC(1)and with sodium?uoride addition at1.25M(2),in activation solution of Na-geopolymer(3)and with sodium?uoride addition at1.25M(4).

4568 https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface&Coatings Technology206(2012)4567–4573

the hydrogen evolution rate,V H(mL/cm2/day),is related to the metal-lic weight loss rate,ΔW m(mg/cm2/day),using[23–25]:

ΔW?1:085V He2TThus the corresponding corrosion rate,P H(mm/year),is equal to: P H?2:279V He3TFor Mg corrosion,there is excellent agreement between the corro-sion rate measured by the weight loss rate and that evaluated from the hydrogen evolution rate[23].

In the Tafel extrapolation method for measuring the Mg corrosion rate,the corrosion current density,I corr(mA/cm2),is estimated by Tafel extrapolation of the cathodic branch of the polarization curve, and I corr is related to the average corrosion rate P i(mm/year)using [21,22,26,27]:

P i?22:85I corre4T3.Results and discussion

3.1.Immersion of Mg–Zr alloy in various basic solutions

Immersion tests were carried out in interstitial solution of OPC and activating solution of Na-geopolymer with addition of sodium?uoride at25°C.Fig.1presents the appearance of specimens before and after immersions during24hours and2months.On a macroscopic scale, the specimen surfaces at24hours are similar to uncorroded pieces. After2months,the in?uence of basic media composition is evidenced with the dark color surface specimen.Additions of sodium?uoride in solutions evidence a light color surface compared to without sodium ?uoride.

Fig.2shows the SEM images of representative typical areas speci-mens after immersion during2months in previous basic media.On Fig.2(a)a porous crystalline coating with needles and sheets is evidenced whereas Fig.2(b)exhibits small crystals with less porosity which can be attributed to the presence of sodium?uoride in the solu-tion.In the presence of silicates,the surface layer is less porous (Fig.2(c))and relatively dense.Addition of sodium?uoride induced a crystalline coating even more compact(Fig.2(d))[17].EDS analysis on Mg–Zr alloy immersed in alkaline solutions with sodium?uoride con?rms the presence of?uoride in the Mg–Zr surface coating?lm (Fig.3(b)and(d))as observed in previous studies with corrosion in-hibitor ef?ciency[11,12].EDS analysis does not con?rm the presence of?uoride compounds MgF2or NaMgF3on metal surface as identi?ed with XPS and XRD methods[11,12].

In order to determine corrosion behavior of immersed Mg–Zr alloy on the long run,hydrogen evolution V H(mL/cm2)has been obtained with analyses of hydrogen during2months(Fig.4).The evolution of hydrogen produced by corrosion of Mg–Zr alloy in OPC solution in-creases linearly over time whereas Mg–Zr in OPC solution with sodium ?uoride and in Na-geopolymer solution without NaF exhibits hydrogen evolutions which tend to stabilize.In Na-geopolymer solution with so-dium?uoride,hydrogen evolution of Mg–Zr is signi?cantly lower than in other solutions and does not evolve after12days.All these results on hydrogen evolution seem correlated with the previous observations of alloys morphologies that pointed out the signi?cance of the porosity of the coating layer.

Table3gives the corrosion rate P H after1,12and54days in solu-tions.It has been noticed that in all solutions,corrosion rates at1

and

Fig.2.SEM images of Mg–Zr alloy at2months after immersion in pore solution of OPC(a)and with sodium?uoride addition at1.25M(b),in activation solution of Na-geopolymer (c)and with sodium?uoride addition at1.25M(d).

4569

https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface&Coatings Technology206(2012)4567–4573

12days are in the same order.At 54days in activating solution of Na-geopolymer with sodium ?uoride,corrosion rate P H is lower than in OPC solution with and without sodium ?uoride or in Na-geopolymer solution without NaF.Therefore,corrosion performances of Mg –Zr in tested solutions evolve with time and measurements of hydrogen evo-lution during several weeks are necessary to conclude for long term behavior.

3.2.Electrochemical measurements

The corrosion properties of Mg –Zr alloy in representative solution of OPC and activation solution of Na-geopolymer with sodium ?uoride

addition have been evaluated by open circuit potential measurement and potentiodynamic polarization study.Fig.5shows the open circuit potential measurements for Mg –Zr alloy respectively in representative solution of OPC and activation solution of Na-geopolymer with addi-tion of sodium ?uoride during 24hours.The evolution of open circuit potential gives information on state of metal surface.A previous study [17]has pointed out the effect of ?uoride on corrosion rate de-creasing in alkaline solutions with pure Mg.Therefore,additions of so-dium ?uoride in OPC and activation solution of Na-geopolymer have been performed to evaluate the performance of sodium ?uoride inhibiting effect on Mg –Zr alloy.The open circuit potential for Mg –Zr in OPC solution trends to a more anodic value with ?uoride concentra-tion equal to 1.25M (Fig.5).Gulbrandsen et al.[17]have pointed out that the open circuit values give a trend of the corrosion density I corr .Indeed the corrosion current density decreases when the open circuit potential is more anodic.The OPC solution contains a mixture of NaOH,KOH and Ca(OH)2with a pH value equal to 13,and it is well known that in alkaline solution an anodizing ?lm appears on surface that is consisted of a magnesium hydroxide ?lms [17,28,29];therefore the open circuit E corr is more anodic with a corrosion resistance ?

lm.

Fig.3.EDS spectra of Mg –Zr alloy at 2months after immersion in pore solution of OPC (a)and with sodium ?uoride addition at 1.25M (b),in activation solution of Na-geopolymer (c)and with sodium ?uoride addition at 1.25M

(d).

Fig.4.Hydrogen evolution of Mg –Zr in OPC solution and activation solution of Na-geopolymer with addition of ?uoride as a function of time.

Table 3

P H for Mg –Zr alloy at 1,12and 54days evaluated from hydrogen evolution rate using Eq.(3).

P H (mm/year)P H-sol./P H-OPC

sol.

1day

12days 54days 1day 12days 54days OPC solution

1.510?2

210

1.1410?3

111OPC solution +[NaF]=1.25M 1.510?2 3.210?38.710?41

1.6

0.76Geopolymer solution 1.310?2310?37.910?40.86 1.50.7Geopolymer solution +[NaF]=1.25M b 10?2

110?3

210?4

b 0.660.5

0.17

4570 https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface &Coatings Technology 206(2012)4567–4573

The activation solution of Na-geopolymer contains a large amount of sodium hydroxide and sodium silicate with a pH value larger than 13.5:due to the higher pH value and the presence of silicates which are well known to be corrosion inhibitor [18,30,31],the potential of open circuit of Mg –Zr is more anodic value than in OPC solution.How-ever,unlike in OPC solution,addition of ?uoride in activation solution of Na-geopolymer gives a slightly more cathodic value open circuit.

Potentiodynamic polarization measurements have been carried out in OPC and activation solution of Na-geopolymer with ?uoride addition at 24hours (Fig.6).Corrosion current densities have been evaluated with the extrapolation of the cathodic branch provided a linear Tafel re-gion and the intersection with corrosion potential (E corr )[32].The eval-uated I corr values and corresponding P i are included in Table 4.The corrosion density of Mg –Zr alloy decreases in OPC solution with sodium ?uoride addition (Fig.6).In activation solution of Na-geopolymer,the corrosion current density of Mg –Zr is one order of magnitude less im-portant than in OPC solution and similar to OPC solution containing so-dium ?uoride.Addition of ?uoride in activation solution of Na-geopolymer induces a slightly increase of current density of Mg –Zr at 24hours.By comparison ratio of P H /P i at 1day (Table 5),it can be no-ticed that P H is always larger than P i .Therefore,P i obtained by

polarization curves with Tafel extrapolation is not a good approxima-tion for this measurement.This fact has been con ?rmed by Z.Shi [26]that examined Mg and Mg alloys corrosion rates evaluated by Tafel ex-trapolation and compared with those determined by weight loss and hydrogen evolution.This phenomenon that corrosion rate was not cor-rectly measured by P i has been attributed to the corroding area is differ-ent during solution immersion and measurement of the polarization curve [26]or a build up of surface corrosion products [33].

3.3.Encapsulation tests of Mg –Zr alloy in OPC and Na-geopolymer pastes Immobilizations of Mg –Zr alloy in OPC paste and in Na-geopolymer with sodium ?uoride addition have been performed.On Fig.7the pho-tographs and optical microscopy images of cut samples are reported.On a macroscopic scale in all cases,Mg –Zr samples are well encapsulat-ed in binders and cavities due to an important corrosion phenomenon are not evidenced on Mg –Zr and binders interface.However,some cracks are clearly evidenced for Mg –Zr alloy encapsulated in OPC paste with sodium ?uoride addition.These cracks have been attributed to an uncompleted cement Portland hydration due to precipitation of CaF 2during clinker hydration [34].Indeed it has been noticed that ad-dition of sodium ?uoride in OPC paste has an important effect on paste ?uidity.For Na-geopolymer binders,no cracks have been identi ?ed and additions of sodium ?uoride in Na-geopolymer did not cause effect on paste ?uidity due to absence of Ca 2+.After examination of the appear-ance of Mg –Zr encapsulated in various binders,the corrosion hydrogen evolutions have to be investigated.The determination of corrosion rates P H (mm/year)of immobilized Mg –Zr alloy have been obtained with analyses of hydrogen evolution V H (mL/cm 2)as a function of time (Fig.8).

In OPC mortar,the hydrogen evolution V H increases during 30days and seems to stabilize after 80days.In OPC,the hydrogen evolution is much more important than in other binders.Hydrogen evolutions of encapsulated Mg –Zr in OPC paste with sodium ?uoride and in Na-geopolymer without NaF are similar,so addition of sodium ?uoride in OPC is necessary to have similar corrosion performance than in Na-geopolymer (with or without sodium ?uoride).

However,hydrogen evolution results can be in ?uenced by binders porosity;indeed the porosity and the water content play a major role in the hydrogen diffusion in cement pastes [35].Recently hydrogen dif-fusion experiments [36,37]have shown on CEM I and Na-geopolymer pastes in 70%relative humidity have shown that hydrogen diffusion coef ?cient in Na-geopolymer is one order of magnitude more impor-tant than in CEMI paste.Concerning the pH effect,it is well

known

Fig.5.Open circuit evolution of Mg –Zr in OPC solution and in activation solution of Na-geopolymer with addition of ?uoride as a function of

time.

Fig.6.Potentiodynamic curves of Mg –Zr alloy in OPC solution and in activation solu-tion of Na-geopolymer with ?uoride addition solution at 24hours.

Table 4

I corr ,E corr and P i for Mg –Zr alloy at 24hours evaluated from polarization curves,open circuit and P i using Eq.(4).

I corr (mA/cm 2)

E corr (V/SCE)P i-24h

(mm/year)P i-sol./P i-OPC sol.OPC solution 5.510?4?1.51 1.2510?21OPC solution +[NaF]=1.25M 6.710?5?1.15 1.5310?30.12Geopolymer solution 5.910?5?1.28 1.3410?30.1Geopolymer solution +[NaF]=1.25M

210?4

?1.35

4.5710?3

0.36

Table 5

Ratios of P H /P i from the data of Tables 5and 6at 1day.

P H /P i

OPC solution

1.2OPC solution +[NaF]=1.25M 9.8Geopolymer solution

9.7Geopolymer solution +[NaF]=1.25M

2.1

4571

https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface &Coatings Technology 206(2012)4567–4573

that in Portland cement paste the ?nal pH of pore solution is comprised between 12.4and 13.5.Recently,R.Lloyd et al.have analyzed Na-geopolymer pore solution with various ?y ash contents and in all cases pH is higher than 13[38,39].Therefore,corrosion performance of Mg –Zr in these binders cannot be explained by the pH of the pore solution and hydrogen diffusion.The relative stability of the passiv-ation layer on the Mg alloys surface seems to be the more likely

explanation.

Fig.7.Macrograph images (left)and optical microscope images (right)of Mg –Zr alloy encapsulated in OPC paste (a,b)and with sodium ?uoride addition (c,d),in Na-geopolymer (e,f)and with sodium ?uoride addition (g,h).

4572 https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface &Coatings Technology 206(2012)4567–4573

Table 6gives ratios of P H-binder /P H-OPC determined in various binders;it can be noticed that short-term tests do not provide adequate predic-tion of long term behavior.4.Conclusions

?The corrosion rate P i of Mg –Zr alloy in OPC solution is one order of mag-nitude more important than in Na-geopolymer solution environment.?Corrosion rates using hydrogen evolution of Mg –Zr in binders are the following:

P H ?OPC >>P H ?GEO >P H ?OPC tNaF ? ?1:25

M ≈P H ?GEO tNaF ? ?1:5M

?Geopolymer Al:Na:Si:H 2O appears to be an attractive binder for Mg –Zr encapsulation.

?Unlike in Portland cement,?uoride addition in Na-geopolymer can be used as corrosion inhibitor without deteriorating the mechanical properties.

References

[1]R.Caillat,J.Hérenguel,M.Salesse,J.Stohr,J.Nucl.Mater.8(1963)1.[2]J.Pegoud,J.Guillaumix,https://www.wendangku.net/doc/f911479459.html,let,J.Nucl.Mater.41(1971)72.

[3]J.Morris,12th International Conference on Environmental Remediation and Ra-dioactive Waste Management,Liverpool,UK,2009.

[4]G.A.Fairhall,J.D.Palmer,Cem.Concr.Res.22(1992)293.[5]L.M.Spasova,M.I.Ojovan,J.Nucl.Mater.375(2008)347.[6] A.Setiadi,Adv.Appl.Ceram.105(2006)191.[7] A.Zosin,At.Energ.85(1998)510.

[8]G.Turner,15th International Symposium on the Packaging and Transportation of

Radioactive Materials,Miami,Florida,2007.

[9] A.Poulesquen,F.Frizon,https://www.wendangku.net/doc/f911479459.html,mbertin,J.Non-Cryst.Solids 357(2011)3565.[10]M.R.Bothwell,The Corrosion of Light Metals,Wiley,1967.[11]G.Song,D.St.John,Corros.Sci.46(2004)1381.

[12]M.D.Pereda,C.Alonso,L.Burgos-Asperilla,J.A.del Valle,O.A.Ruano,P.Perez,M.A.

Fernández Lorenzo de Mele,Acta Biomater.6(2010)1772.

[13]J.Blanchet,H.Coriou,L.Grall,M.Pelras,G.Plante,M.Salesse,J.Nucl.Mater.7

(1962)311.

[14]G.Williams,H.N.McMurray,R.Grace,Electrochim.Acta 55(2010)7824.

[15]P.M.Bradford,B.Case,G.Dearnaley,J.F.Turner,I.S.Woolsey,Corros.Sci.16(1976)

747.

[16] F.El-Taib Heakal,A.M.Fekry,M.Z.Fatayerji,Electrochim.Acta 54(2009)1545.[17] E.Gulbrandsen,J.Taft?,A.Olsen,Corros.Sci.34(1993)1423.[18]J.Blanchet,CEA-R-2815,1964.

[19]J.Li,G.Lorimer,J.Robson,B.Davis,Mater.Sci.Forum 488–489(2005)329.[20] D.A.Jones,Principles and Prevention of Corrosion,Maxwell-MacMillan,1992.[21]M.G.Fontana,Corrosion Engineering,McGraw Hill,1984.[22]H.H.Uhlig,Corrosion and Corrosion Control,Wiley,1973.[23]G.Song,A.Atrens,Adv.Eng.Mater.5(2003)837.

[24]M.-C.Zhao,M.Liu,G.Song,A.Atrens,Corros.Sci.50(2008)1939.[25]M.-C.Zhao,M.Liu,G.-L.Song,A.Atrens,Corros.Sci.50(2008)3168.[26]Z.Shi,M.Liu,A.Atrens,Corros.Sci.52(2010)579.[27]G.Song,Adv.Eng.Mater.7(2005)563.

[28]J.E.Gray,B.Luan,J.Alloys Compd.336(2002)88.

[29]Y.Mizutani,S.J.Kim,R.Ichino,M.Okido,Surf.Coat.Technol.169–170(2003)143.[30]X.Halloppeau,T.Beldjoudi,C.Fiaud,Corros.Rev.16(1998)27.

[31]Y.Ma,X.Nie,D.O.Northwood,H.Hu,Thin Solid Films 469–470(2004)472.[32] E.McCafferty,Corros.Sci.47(2005)3202.[33]Z.Shi,A.Atrens,Corros.Sci.53(2011)226.

[34] A.Macías,M.L.Escudero,Corros.Sci.36(1994)2169.[35] F.Frizon,C.Galle,J.Porous Media 12(2009)221.

[36] C.Boher,S.Lorente,F.Frizon,Transfert 2012Proceedings,Lille (France),2012,

p.222.

[37]T.H.Vu,F.Frizon,S.Lorente,J.Phys.D:Appl.Phys.42(2009).

[38]K.Andersson,B.Allard,M.Bengtsson,B.Magnusson,Cem.Concr.Res.19(1989)327.[39]R.R.Lloyd,J.L.Provis,J.S.J.van Deventer,Cem.Concr.Res.40(2010)

1386.

Fig.8.Hydrogen evolution (mL/cm 2)of Mg –Zr alloy encapsulated in OPC paste and Na-geopolymer with sodium ?uoride as a function of time.

Table 6

Corrosion rates P H (mm/year)of Mg –Zr alloy encapsulated in OPC and in geopolymer with ?uoride addition at room temperature.

P H P H-binder /P H-OPC 4days

29days 74days 4days 29days 74days OPC

3.510?2 2.9910?2 1.210?2111OPC with

[NaF]=1.25M 510?3610?4410?40.140.020.033Geopolymer

3.2810?3 1.510?310?30.090.050.08Geopolymer with [NaF]=1.25M

b 210?3

1.210?3

510?4

0.057

0.04

4573

https://www.wendangku.net/doc/f911479459.html,mbertin et al./Surface &Coatings Technology 206(2012)4567–4573