Downloaded from orbit.dtu.dk on: Dec 18, 2017 The durability of white Portland cement to chemical attack Nielsen, Erik Pram; Geiker, Mette Rica Publication date: 2004 Document Version Publisher's PDF, also known as Version of record Link back to DTU Orbit Citation (APA): Nielsen, E. P., & Geiker, M. R. (2004). The durability of white Portland cement to chemical attack. (BYG- Rapport; No. R-084). General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Erik Pram Nielsen The durability of white Portland cement to chemical attack DANMARKS TEKNISKE UNIVERSITET Report BYG·DTU R-084 2004 ISSN 1601-2917 ISBN 87-7877-147-1 SUMMARY In terms of volume, the severest conditions that reinforced concrete structures are subjected to usually involve some form of attack by chlorides. These include marine environments, roads and bridges exposed to de-icing salt, etc. The time it takes for the chlorides to reach the reinforcement is the main parameter used to predict the service life of the concrete structures. Much effort has been spent in understanding the processes resulting in the ingress of chlorides in concrete structures exposed to aggressive environments, and in predicting its service life. However, to date it has not been possible to accurately predict the time for chloride ions to reach the reinforcement of concrete structures without carrying out extensive and time consuming laboratory experiments to calibrate the service life models. It is generally agreed that when chlorides ingress into concrete, part of these become fixed to the cement paste hydrates through processes generally referred to as chloride binding. Furthermore, the rate at which these ions ingress into concrete is also known to be highly dependent on the available porosity. The aim of this study was to provide a more reliable model for predicting the ingress rate of chlorides in cement-based materials. A thermodynamic model for the phase equilibria in hydrated Portland cement based on application of the phase rule, was developed and then verified experimentally during this investigation. The experiments were conducted on three types of Portland cement with varying composition (one grey and two white Portland cements). This approach was extended to include the reactions resulting in the binding of chlorides and alkalis. The model’s accuracy to predict chloride binding in Portland cement pastes at any content of chloride, alkalis, sulfates and carbonate was verified experimentally, and found to be consistent with available data in the literature. Furthermore, the model was shown to be equally valid for the phase assemblage identified in 25 year old PC pastes exposed to NaCl. Chloride binding in Portland cement pastes is highly dependent on the content of alkalis; the higher content of alkalis the lower amount of bound chloride at any total content of chloride. Contrary to conventional wisdom, the content of alumina was shown to only play a relative minor role in the overall extent of chloride binding, as the major part of chloride is bound by the C-S-H phase. As alkalis are transported at a much slower rate into concrete than chlorides, low-alkali Portland cements should be used in aggressive environments containing chlorides. Furthermore, owing to the minor effect of alumina on chloride binding, low-alkali low-aluminate Portland cements could arguably be recommended for aggressive marine environments providing maximum protection to both chloride transport and sulfate attack. The increased chloride binding capacity obtained by reducing the alkali content of the cement also reduces the ratio of chloride to hydroxyl ions in the pore solution (i.e. ratios at 0.50-0.60 are typically given as the chloride threshold value for reinforcement corrosion) as long as the content of alkalis is lower than Na2Oeq < 0.30-0.35. This occurs even though the initial pH of the pore solution is of course also lower at these low alkali contents. The model was introduced into an existing finite element model for the ingress of chlorides into concrete accounting for its multi-component nature, and the composite theory was used for predicting the diffusivity of each ion based on the phase assemblage in the hydrated Portland cement paste. Excellent agreement was found between the chloride ingress and phase assemblage profiles predicted by the model, and those determined experimentally on w/p 0.45 Portland cement pastes exposed to 650 mM NaCl for 70 days. iii The durability of white PC to chemical attack iv RESUMÉ (IN DANISH) Klorider forbindes normalt med de værst tænkelige miljøer for placering af armerede betonkonstruktioner. Disse kan være havvandsmiljø, eller veje og brokonstruktioner udsat for tøsalte i vinterperioderne. Den tid der går indtil kloriderne når ind til armeringen, hvorefter armeringskorrosion kan påbegyndes, anvendes oftest som udtryk for betonkonstruktioners levetid. Man har i de seneste årtier haft stor fokus på at opnå forståelse for de mekanismer der resulterer i indtrængning af klorider i beton. Det egentlige formål er at bringe sig i stand til at skønne en armeret betonkonstruktions levetid så nøjagtigt som muligt. Eksisterende levetidsmodeller er dog endnu ikke i stand til at kunne forudsige en given betons levetid uden empirisk kalibrering til resultater fra omfattende, tidskrævende forsøg. Anvendeligheden af opnåede resultater er desværre også begrænset til præcis den valgte betonsammensætning. Derfor kan levetidsmodeller heller ikke anvendes til at udvælge den optimale bindersammensætning til et givet eksponeringsmiljø. Det er alment kendt, at en betydelig andel af de indtrængende klorider fastgøres til mineralerne i cementpastaen under indtrængningsforløbet. Denne proces er normalt benævnt kloridbinding. Derudover er der også generel accept for at hastigheden hvorved kloridioner trænger igennem beton er afhængig af cementpastaens porøsitet. Det overordnede mål i dette projekt er at udvikle en model til kvantificering af disse egenskaber på baggrund af den kemiske sammensætning af cementen. En termodynamisk model til at forudsige faseselskabet i hærdnet Portland cement er udviklet, med udgangspunkt i fasereglen. Modellen er verificeret på baggrund af forsøg på tre typer Portland cement med forskellig sammensætning (en grå og to hvide cementer). Modellen blev videreudviklet således at reaktioner af klorider og alkalier med hydrater i cementpastaen blev inkluderet. Modellens nøjagtighed til at forudsige kloridbinding i hærdet cementpasta til ethvert indhold af klorid, alkali, karbonat og sulfat, er bevist i rapporten. Modellens forudsigelser er i god overensstemmelse med resultater fra uafhængige forsøg i litteraturen, samt med resultater fra 25 år gamle cementpastaprøver eksponeret til NaCl. Kloridbinding i Portlandcementpastaer er meget afhængig af cementens indhold af alkalier; jo højere alkaliindhold desto lavere kloridbinding til ethvert totalt indhold af klorider. I modsætning til almen viden, har aluminium en beskeden indflydelse på kloridbinding fordi langt hovedparten af klorider bindes i C-S-H fasen. Alkalier trænger meget langsomt ind i beton i forhold til klorider, og derfor bør lav-alkali Portland cementer anvendes i kloridholdigt aggressivt eksponeringsmiljø. Givet den lave effekt af aluminiumindholdet på kloridbinding, bør lav-alkali, lav-aluminat Portland cementer anvendes i aggressive miljøer hvor både klorid og sulfat er til stede. Derved sikres sulfatbestandighed. Selv om den initiale pH-værdi i porevæsken reduceres ved at sænke alkali indholdet, så vil den øgede kloridbindingskapacitet i Portland cementer med Na2Oæqv. lavere end 0.30-0.35 vægt-% også reducere forholdet Cl/OH i porevæsken, der normalt forbindes med tærskelværdien for armeringskorrosion. Modellen blev introduceret i en eksisterende finit element model for diffusion i multikomponent væske, sammen med en kompositteori model for estimering af effektive diffusionskoefficienter. Modellens skøn på kloridindtrængning viste god overensstemmelse med de målte på v/p 0.45 cementpastaer (tre forskellige cementtyper) efter 70 døgns eksponering i 650 mM Na v The durability of white PC to chemical attack vi FOREWORD This thesis, revised according to minor suggestions by the Assessment Committee, was written by Erik Pram Nielsen and is the result of a three year project commenced in July 2001, in collaboration between Aalborg Portland A/S, Denmark, and the Department of Civil Engineering at the Technical University of Denmark (DTU). The Ph.D. candidate was employed by the enterprise during the entire project. The project was financed by Aalborg Portland A/S, with partial contribution by the Knud Hoejgaard Foundation for the traveling expenses for the two oversea visits; two months at NIST, Gaithersburg, USA, and two months at the Department of Civil Engineering at the University of Toronto,