Modified Hydrotalcites as Smart Additives for Improved

Modified Hydrotalcites as Smart Additives for Improved

Corrosion Protection of Reinforced Concrete

Modified Hydrotalcites as Smart Additives for Improved

Corrosion Protection of Reinforced Concrete

Proefschrift

ter verkrijging van de graad van doctor aan de Technische Universiteit Delft,

op gezag van de Rector Magnificus Prof. ir. K.C.A.M. Luyben, voorzitter van het College voor Promoties,

in het openbaar te verdedigen op 23 juni om 10.00 uur

door

Zhengxian YANG

Master of Science in Physical Chemistry, South-Central University for Nationalities, China

geboren te Huainan, Anhui Province, China

Dit proefschrift is goedgekeurd door de promotor:

Prof. dr. R.B. Polder

Samenstelling promotiecommissie:

Rector Magnificus, Technische Universiteit Delft, voorzitter

Prof. dr. R.B. Polder, Technische Universiteit Delft, promotor

Prof. dr. ir. K. van Breugel, Technische Universiteit Delft

Dr. J.M.C. Mol, Technische Universiteit Delft

Prof. dr. B. Elsener, Swiss Federal Institute of Technology in Zürich, Switzerland

Prof. dr. C. Andrade, Institute ‘Eduardo Torroja’ of Construction Science, Spain

Prof. dr. G.F. Peng, Beijing Jiaotong University, China

Dr. H.R. Fischer, TNO, The Netherlands

Prof. dr. ir. L.J. Sluys, Technische Universiteit Delft, Reservelid ISBN: 978-94-91909-25-2 Keywords: Modified hydrotalcites; Layered double hydroxides; Smart additives; Reinforced concrete; Durability; Service life; Corrosion; Chloride; Corrosion inhibitors; Amino acids Copyright © 2015 by Zhengxian Yang Email: [email protected] Printed by Haveka B.V. in The Netherlands All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted, in any form or by any means. Electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the author, except in the case of brief quotation embodied in critical reviews and certain other non-commercial uses permitted by copyright law. The author has put the greatest effort to publish reliable data and information. However, the possibility should not be excluded that it contains errors and imperfections. Any use of the data and results from this publication is entirely on the own responsibility of the user. The author disclaims any liability for damage which could result from that.

https://www.ethz.ch/

行，成于思，胜于言

Table of Contents

Chapter 1 General Introduction .......................................................................................................... 1

1.1 Background of this research project ....................................................................................................2

1.2 Objective and scope of this research ...................................................................................................4

1.3 Outline of this thesis .............................................................................................................................4

References ..............................................................................................................................................7

Chapter 2 Modified Hydrotalcites as Smart Additives for Improved Corrosion Protection of Reinforced

Concrete: A Literature Review ............................................................................................. 9

2.1 Introduction ....................................................................................................................................... 10

2.2 Corrosion of the steel in concrete ..................................................................................................... 10

2.2.1 Carbonation ................................................................................................................................ 12

2.2.2 Chloride induced corrosion ........................................................................................................ 13

2.3 Factors that affect corrosion of reinforcement ................................................................................. 15

2.3.1 Permeability of concrete ............................................................................................................ 15

2.3.2 Chloride binding ......................................................................................................................... 15

2.3.3 Concrete cover ........................................................................................................................... 16

2.3.4 Environmental conditions .......................................................................................................... 17

2.4 Corrosion preventive measures ........................................................................................................ 18

2.5 MHTs and their application in cementitious materials ..................................................................... 18

2.5.1 General aspect ............................................................................................................................ 18

2.5.2 Synthesis and characterization ................................................................................................... 20

2.5.3 Ion exchange of MHTs and its role in capturing chloride ........................................................... 21

2.5.4 Application in cementitious materials ........................................................................................ 22

2.6 Concluding remarks ........................................................................................................................... 25

References ........................................................................................................................................... 25

Chapter 3 Inhibition Performance Evaluation of Some Amino Acids against Steel Corrosion in Simulated

Concrete Pore Solution ..................................................................................................... 33

3.1 Introduction ....................................................................................................................................... 34

Table of Contents | II

3.2 Experimental...................................................................................................................................... 36

3.2.1 Materials ..................................................................................................................................... 36

3.2.2 Testing methods ......................................................................................................................... 36

3.3 Results and discussion ....................................................................................................................... 38

3.3.1 The effect of inhibitor addition on pH of the alkaline solution .................................................. 38

3.3.2 Open circuit potential (OCP) measurements ............................................................................. 39

3.3.3 Linear polarization resistance (LPR) measurements .................................................................. 43

3.4 Conclusions ........................................................................................................................................ 47

References ........................................................................................................................................... 48

Chapter 4 Synthesis and Characterization of Modified Hydrotalcites Using Selected Inhibitors as

Modifiers .......................................................................................................................... 51

4.1 Introduction ....................................................................................................................................... 52

4.2 Experimental...................................................................................................................................... 53

4.2.1 Materials ..................................................................................................................................... 53

4.2.2 Synthesis ..................................................................................................................................... 53

4.2.3 Characterization ......................................................................................................................... 54


4.3.1 X-ray diffraction analysis ............................................................................................................ 54

4.3.2 Infrared analysis ......................................................................................................................... 57

4.3.3 Thermal analysis ......................................................................................................................... 60

4.3.4 Intercalation efficiency under the calcination-rehydration condition ....................................... 63

4.4 Conclusions ........................................................................................................................................ 65

References ........................................................................................................................................... 65

Chapter 5 Anti-corrosion Performance Evaluation of Synthesized Modified Hydrotalcites in Simulated

Concrete Pore Solution ...................................................................................................... 69

5.1 Introduction ....................................................................................................................................... 70

5.2 Experimental...................................................................................................................................... 70

5.2.1 Materials ..................................................................................................................................... 70

5.2.2 Ion exchange of MHT with chlorides in simulated concrete pore solution ............................... 70

5.2.3 Anti-corrosion performance evaluation in simulated concrete pore solution .......................... 71


5.3.1 The role of MHT in capturing chlorides ...................................................................................... 72

5.3.2 Chloride exchange in simulated concrete pore solution ............................................................ 72

Table of contents | III

5.3.3 Anti-corrosion performance of the selected MHT ..................................................................... 78

5.4 Conclusion ......................................................................................................................................... 85

References ........................................................................................................................................... 86

Chapter 6 The Influence of Two Types of Modified Hydrotalcites on Chloride Ingress in Cement Mortar

......................................................................................................................................... 89

6.1 Introduction ....................................................................................................................................... 90

6.2 Experimental...................................................................................................................................... 91

6.2.1 Materials ..................................................................................................................................... 91

6.2.2 Sample preparation .................................................................................................................... 92

6.2.3 Testing methods ......................................................................................................................... 92


6.3.1 The effect of MHTs on workability of fresh mortar .................................................................... 95

6.3.2 The effect of MHTs on mechanical properties ........................................................................... 95

6.3.3 The effect of MHTs on porosity .................................................................................................. 98

6.3.4 The effect of MHTs on chloride penetration .............................................................................. 99

6.4 Conclusion ....................................................................................................................................... 105

References ......................................................................................................................................... 106

Chapter 7 The Anti-corrosion Performance of Two Types of Modified Hydrotalcites in Cement Mortar

with Embedded Steel ....................................................................................................... 109

7.1 Introduction ..................................................................................................................................... 110

7.2 Experimental.................................................................................................................................... 110

7.2.1 Materials ................................................................................................................................... 110

7.2.2 Sample preparation .................................................................................................................. 110

7.2.3 Anti-corrosion performance evaluation ................................................................................... 114

7.2.4 Chloride analysis ....................................................................................................................... 121

7.3 Results and discussion ..................................................................................................................... 122

7.3.1 Accelerated chloride migration test ......................................................................................... 122

7.3.2 Cyclic wetting-drying test ......................................................................................................... 130

7.3.3 Natural diffusion test ................................................................................................................ 136

7.3.4 The applied test methodology and the effect on chloride diffusion/migration and chloride

threshold ........................................................................................................................................... 139

7.3.5 Effect of MHT on time to corrosion initiation .......................................................................... 141

7.4 Conclusion ....................................................................................................................................... 144

Table of Contents | IV

References ......................................................................................................................................... 146

Chapter 8 Conclusions and Recommendations for Future Research .................................................. 149

8.1 General conclusions......................................................................................................................... 150

8.2 Industrial application potentials and valorization ........................................................................... 154

8.3 Recommendations for future research ........................................................................................... 154

Summary ....................................................................................................................................... 157

Samenvatting ................................................................................................................................. 160

Acknowledgements ........................................................................................................................ 163

Curriculum Vitae ............................................................................................................................ 165

List of Publications (selected) .......................................................................................................... 166

Chapter 1

General Introduction

2 | Chapter 1

1.1 Background of this research project

Reinforced concrete is the most widely used construction material across the world, due to its

relatively low cost and the excellent marriage between reinforcing steel and bulk concrete. Its

technical success has resulted from the complementary mechanical properties between

reinforcing steel (source of good tensile strength) and concrete (source of good compressive

strength) and their excellent physical and chemical compatibility. The similarity of concrete and

steel in thermal expansion contributes to eliminate large internal stresses due to differences in

thermal expansion or contraction [1]. In addition, the high alkalinity of the concrete pore solution

promotes corrosion protection by passivation of mild (reinforcing) steel [2]. All of these factors

have led to a large stock of structures which are expected to last very long; nowadays a service

life of 100 years is required for major structures, e.g. tunnels and bridges [3, 4]. However,

corrosion protection can be lost due to ingress of chloride ions, typically present in de-icing salts

and marine environment [5, 6]. Chloride ion transport by diffusion or capillary absorption in the

concrete pore system is relatively fast and chloride-induced corrosion has been recognized as a

main culprit to the durability of reinforced concrete [1]. A secondary issue is the carbonation of

concrete, which reduces the pH of the concrete pore solution to values where passivation

disappears [7]. Combined chloride ingress and carbonation increases the corrosion risk even more.

Both degradation processes threaten the initial passivation and thus concrete durability on the

time scale of 10 to 50 years. In all above-ground outdoor structures, plenty of oxygen and water

are available to promote relatively rapid corrosion. Corrosion products are much more

voluminous than the parent steel, causing tensile stresses in the concrete cover and subsequently,

its cracking and spalling on a relatively short term (< 10 years after corrosion initiation) [8, 9]. In

the midterm (10-20 years), steel cross section loss may cause insufficient tensile capacity and

thus may threat structural integrity and safety [10]. Consequently, signs of corrosion (rust

staining, concrete cracking) signal the need to repair and to reinstate corrosion protection, or to

even replace complete elements or structures. Such measures are at least laborious and time

consuming. Results are high unplanned maintenance costs (potentially up to the initial

construction costs), significant unavailability (out of service time) and waste of materials and

energy with associated emission of carbon dioxide. This problem is highly relevant for civil

engineering structures in the transport sector, such as bridges, tunnels, harbour quays and parking

structures [11]. Consequently, the construction industry is in need of improving the corrosion

protection of reinforced concrete structures, preferably by low-cost measures.

Modern service life design approach aims at providing sufficient concrete cover depth to the

reinforcing steel, while taking into account the resistance against chloride transport and the

critical chloride level at the steel for corrosion initiation. Traditional Standards oversimplify the

complexity of the mechanisms involved and provide insufficient performance in aggressive

environment. More advanced regulations [3, 4, 12] are based on modelling of chloride transport

General Introduction | 3

up to the critical corrosion initiating level and testing of materials under laboratory conditions, e.g.

by Rapid Chloride Migration testing [13, 14]. In particular the critical chloride content [15-18]

and the time-dependency of chloride diffusion [19] are not well understood. Consequently, high

uncertainties are still associated with long-term material behaviour and the influence of execution

variables that dominate concrete properties as produced on site. Presently available options for

improved corrosion protection are either too costly or too complicated; or insufficiently effective.

Stainless steel reinforcement is 5 or 10 times more expensive than reinforcing (carbon) steel [20,

21]. Cathodic prevention and protection may be effective but both are a special niche expertise

and are thus not applied on a wide scale [22, 23]. Coatings on the concrete surface do not last

long enough (10-20 years), which causes a maintenance cycle of its own [20]. Corrosion

inhibitors have been proposed but are generally not reliable in terms of long-term efficiency [24];

some are toxic, such as nitrites [25]. Thus, continuing research in the domain of materials science

is essentially needed in searching for more effective measures to improve the corrosion resistance

of reinforced concrete.

In the last two decades, more research interest has been attracted in developing new or

modified materials able to prevent corrosion initiation and slow down or even stop corrosion

propagation, as well as in understanding their underlying working mechanism. Among them,

modified hydrotalcites (MHTs) may represent a promising option for use in concrete as new type

of smart functional additives [26].

Recently, a study on the application of amino acid modified hydrotalcites in cementitious

materials has formed the basis of a joined TNO-AIDICO patent “corrosion inhibition of

reinforced concrete” (WO 2011/065825 A1) [27], that consists of hydrotalcites intercalated with

organic species, in particular eco-friendly amino acids, which can be directly applied (without

paint or polymeric carrier) as aqueous emulsion on a metallic substrate, or it can be mixed into

fresh concrete with the various components. However, its scale was relatively small and further

work was considered necessary by the applicants and their organisations.

Our preliminary work has shown that ion exchange occurs between free chloride ions in the

simulated concrete pore solution and anions intercalated in MHT reducing the free chloride

concentration which is equivalent to increased binding of chloride [28, 29]. Indications exist for

the natural occurrence of hydrotalcite in hydrated blast furnace slag cements [30-32], which are

known to bind more chloride ions than Portland cements. Increased binding would slow down

chloride transport. The preliminary work has also shown that certain organic anions with known

inhibitive properties could be intercalated, which then can be slowly released, possibly 'automatic'

upon arrival of chloride ions. Such inhibition increases the chloride threshold level for corrosion

initiation and/or reduces the subsequent corrosion rate. Less aggressive electrochemical potentials

have been observed in simulated concrete pore solution with MHT as compared to solutions

without MHT [29]. These results suggest that MHT has a high potential as active component in

concrete with corrosion protection properties that can be tailor made.

4 | Chapter 1

Hydrotalcite belongs to a large mineral group of naturally occurring Layered Double

Hydroxides (LDHs), in general formula [MII

1-x MIII

x (OH)2]x+

[(An-

x/n)]x-

·mH2O, where MII and

MIII

are di- and trivalent metal cations (MII: Mg

2+, Ca

2+, Zn

2+, Ni

2+, etc; M

III: Al

3+, Fe

3+, Ga

3+,

Co3+

, etc), and A

n- is an exchangeable interlayer anion with valence n. In this thesis, the

hydrotalcites particularly refer to Mg-Al hydrotalcites.

1.2 Objective and scope of this research

The research project aims at developing a new promising additive as an alternative approach

against chloride ingress into mortar (or concrete) and/or chloride induced corrosion based on

modified hydrotalcites (in particular those modified by eco-friendly amino acids) and

understanding and quantifying their effects in mortar (or concrete): (1) their interaction with

chloride ions and (2) their influence on steel corrosion. These effects should be measurable in

terms of reduced chloride diffusion rates and increased chloride threshold level for corrosion

initiation. The overall objective is to increase the tolerance of mortar/concrete structures with

regard to chloride induced corrosion and to increase their maintenance-free service life by

utilization of modified hydrotalcites. Potential applications are as (1) admixtures to fresh

mortar/concrete, (2) pre-casting treatment of reinforcing steel in new building and repair

situations and (3) as addition to repair mortars. This new additive when applied appropriately,

either incorporation of a small amount in bulk mortar/concrete or as a surface coating of the

reinforcing steel could prevent/delay the corrosion initiation. In addition, it is expected to have no

adverse side-effects on the properties of fresh and hardened mortar/concrete.

1.3 Outline of this thesis

As shown in Figure 1.1, this thesis is organized in 8 chapters. Chapter 1 gives the background,

motivation and the objective and scope of this research.

Chapter 2 presents a literature review of the existing knowledge with regard to synthesis and

characterisation methods of MHTs, ion exchange within the MHT structure as well as the

application of MHTs in cementitious materials. On top of that the mechanism of corrosion in

reinforced concrete, factors that affect corrosion of reinforcement and relevant corrosion

preventive measures are also briefly reviewed. This part is intended to be brief, with which many

textbooks [1, 10, 33-35] have explicitly dealt.

Chapter 3 evaluates the inhibition performance of some amino acids (in particular, glycine,

6-aminocaproic acid, 11-aminoundecanoic acid and p-aminobenzoic acid) against steel corrosion

in simulated concrete pore solution. The objective of this chapter is to select the most promising

amino acids with good inhibition performance as candidate modifiers for synthesis of MHTs.

Based on the results obtained from Chapter 3, six MHTs (with Mg/Al atomic ratios of 2.2

and 2.7) intercalated with nitrites and the selected amino acids inhibitors were synthesized in

Chapter 4 using the reconstruction method. They were characterized by means of X-ray powder


diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), Thermogravimetry (TG),

Differential scanning calorimetry (DSC) and relevant elemental analysis.

Chapter 5 investigates the ion exchange characteristics of the six synthesized MHTs and their

anti-corrosion performance in chloride-rich simulated concrete pore solution based on

electrochemical methods such as open circuit potential (OCP) and linear polarization resistance

(LPR). The objective of this chapter is to select the MHTs with best anti-corrosion performance

for use in both plain and reinforced mortar tests. Two MHTs, i.e., Mg(2)Al-NO2 and Mg(2)Al-

pAB were finally selected as the more promising MHT candidates for mortar test, which is the

main topic of the following two chapters.

Chapter 6 explores the influence of the two selected modified hydrotalcites on chloride

ingress into plain mortar, while Chapter 7 focuses on their anti-corrosion performance in mortar

with embedded steel. In Chapter 6, the effect of the two MHTs in plain mortar is studied by

workability test, strength test, porosity test, and rapid chloride migration and natural diffusion test.

In Chapter 7, the effect of the two MHTs on reinforcement corrosion is investigated by three

designated testing methods based on chloride exposure and OCP/LPR measurements with custom

designed reinforced mortar specimens:

a) An accelerated chloride migration test

b) A wetting-drying cyclic test

c) A natural diffusion test

Chapter 8 summarizes the results obtained in this research and gives the conclusions. Some

recommendations are given for future research.

6 | Chapter 1

Figure 1.1 Outline of the thesis.

Chapter 8

Conclusions and recommendations

Chapter 6

The influence of two types of

MHTs on chloride ingress in

cement mortar

Chapter 7

Anti-corrosion performance of two

types of MHTs in cement mortar

with embedded steel

Chapter 1

General introduction

Chapter 2

Literature review

Chapter 3

Inhibition performance

evaluation of some

amino acids against

steel corrosion in

simulated concrete

pore solution

Chapter 4

Synthesis and

characterization of

MHTs using selected

inhibitors as modifiers

Chapter 5

Anti-corrosion

performance evaluation

of synthesized MHTs

in simulated concrete

pore solution


References

[1] Bertolini L, Elsener B, Pedeferri P, Redaelli E, Polder RB. Corrosion of steel in concrete: prevention,

diagnosis, repair. Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA; 2013.

[2] Gouda V. Corrosion and corrosion inhibition of reinforcing steel: I. Immersed in alkaline solutions.

British Corrosion Journal. 1970;5(5):198-203.

[3] Fédération Internationale du Béton, Model code for service life design. Lausanne, Switzerland: fib Bull.

34; 2006.

[4] The European Union—Brite EuRam III, DuraCrete-Probabilistic Performance Based Durability Design

of Concrete Structures. Final Technical Report, Document BE95-1347/R17. CUR, Gouda, 2000.

[5] Polder R, Larbi J. Investigation of concrete exposed to North Sea water submersion for 16 years. Heron.

1995;40(1):31-56.

[6] Polder RB, Hug A. Penetration of chloride from de-icing salt into concrete from a 30 year old bridge.

Heron. 2000;45(2):109-24.

[7] Neville AM. Properties of concrete. Fourth edition. Harlow: Prentice Hall/Pearson; 2006.

[8] Shah V, Hookham C. Long-term aging of light water reactor concrete containments. Nuclear engineering

and design. 1998;185(1):51-81.

[9] Smith J, Virmani YP. Materials and methods for corrosion control of reinforced and prestressed concrete

structures in new construction (No. FHWA-RD-00-081). 2000.

[10] Bentur A, Berke N, Diamond S. Steel corrosion in concrete: fundamentals and civil engineering practice:

CRC Press; 1997.

[11] Gaal G. Prediction of deterioration of concrete bridges. PhD Thesis. Delft, Delft University of

Technology; 2004.

[12] Wegen G, Polder RB, Breugel KV. Guideline for service life design of structural concrete: A

performance based approach with regard to chloride induced corrosion. Heron. 2012;57(3):153-68.

[13] Luping T, Nilsson L-O. Rapid determination of the chloride diffusivity in concrete by applying an

electric field. ACI materials journal. 1993;89(1).

[14] NTBuild492. Concrete, mortar and cement-based repair materials: Chloride migration coefficient from

non-steady-state migration experiments. NordTest, Espoo. 1999.

[15] Polder RB. Critical chloride content for reinforced concrete and its relationship to concrete resistivity.

Materials and Corrosion. 2009;60(8):623-30.

[16] Alonso C, Andrade C, Castellote M, Castro P. Chloride threshold values to depassivate reinforcing bars

embedded in a standardized OPC mortar. Cement and Concrete Research. 2000;30(7):1047-55.

[17] Glass GK, Buenfeld NR. Chloride threshold levels for corrosion induced deterioration of steel. In:

Nilsson L, Ollivier J, editors. 1st RILEM International Workshop on Chloride Penetration into Concrete:

Rilem Publications SARL; 1995. p. 429-40.

[18] Angst U, Elsener B, Larsen CK, Vennesland Ø. Critical chloride content in reinforced concrete—a

review. Cement and Concrete Research. 2009;39(12):1122-38.

[19] Visser J, Polder R. Concrete binder performance evaluation in service life design. In: Kovler K, editor.

ConcreteLife'06-International RILEM-JCI Seminar on Concrete Durability and Service Life Planning:

Curing, Crack Control, Performance in Harsh Environments. Dead Sea, Israel: RILEM Publications

SARL; 2006. p. 330-40.

[20] Cigna R, Andrade C, Nürnberger U, Polder R, Weydert R, Seitz E. COST 521: Corrosion of steel in

reinforced concrete structures-final report. Luxembourg: European communities EUR20599;2002.

8 | Chapter 1

[21] Elsener B, Addari D, Coray S, Rossi A. Stainless steel reinforcing bars–reason for their high pitting

corrosion resistance. Materials and Corrosion. 2011;62(2):111-9.

[22] Pedeferri P. Cathodic protection and cathodic prevention. Construction and Building Materials.

1996;10(5):391-402.

[23] Polder R, Peelen W, Lollini F, Redaelli E, Bertolini L. Numerical design for cathodic protection systems

for concrete. Materials and Corrosion. 2009;60(2):130-6.

[24] Elsener B. Corrosion inhibitors for steel in concrete: state of the art report: Woodhead Pub Limited;

2001.

[25] Rosenberg A, Gaidis J. The mechanism of nitrite inhibition of chloride attack on reinforcing steel in

alkaline aqueous environments. Materials performance. 1979;18(11).

[26] Yang Z, Fischer H, Polder R. Modified hydrotalcites as a new emerging class of smart additive of

reinforced concrete for anticorrosion applications: A literature review. Materials and Corrosion.

2013;64(12):1066-74.

[27] Fischer HR, Adan O, Lloris Cormano JM, Lopez Tendero MJ. Corrosion inhibition of reinforced

concrete. WIPO Patent WO 2011/065825A1, 3 Jun 2011.

[28] Yang Z, Fischer H, Polder R. Synthesis and characterization of modified hydrotalcites and their ion

exchange characteristics in chloride-rich simulated concrete pore solution. Cement and Concrete

Composites. 2014;47:87-93.

[29] Yang Z, Fischer H, Cerezo J, Mol J, Polder R. Aminobenzoate modified MgAl hydrotalcites as a novel

smart additive of reinforced concrete for anticorrosion applications. Construction and Building Materials.

2013;47:1436-43.

[30] Roy DM, Sonnenthal E, Prave R. Hydrotalcite observed in mortars exposed to sulfate solutions. Cement

and Concrete Research. 1985;15(5):914-6.

[31] Wang S-D, Scrivener KL. Hydration products of alkali activated slag cement. Cement and Concrete

Research. 1995;25(3):561-71.

[32] Kayali O, Khan MSH, Sharfuddin Ahmed M. The role of hydrotalcite in chloride binding and corrosion

protection in concretes with ground granulated blast furnace slag. Cement and Concrete Composites.

2012;34(8):936-45.

[33] Poulsen E, Mejlbro L. Diffusion of chloride in concrete: theory and application: CRC Press; 2010.

[34] Tang L, Nilsson L-O, Basheer M. Resistance of concrete to chloride ingress. Testing and modelling:

Spon Press; 2012.

[35] Gjørv OE. Durability design of concrete structures in severe environments: CRC Press; 2014.

Chapter 2

Modified Hydrotalcites as Smart

Additives for Improved Corrosion

Protection of Reinforced Concrete: A

Literature Review

Part of the work described in this chapter has been published as: Yang, Z., Fischer, H., Polder, R.

Modified Hydrotalcites as A New Emerging Class of Smart Additive of Reinforced Concrete for

Anti-corrosion Applications: A Literature Review. Materials and Corrosion, 2013, 64(12):1066-

1074.

10 | Chapter 2

2.1 Introduction

Concrete is a porous and highly heterogeneous composite with features of various dimensions

ranging from nanometer-sized pores and calcium-silicate-hydrate (C-S-H) gel to micrometer-

sized air voids, millimeter-sized aggregate particles and to steel reinforcement that can be meters

in length. Concrete can consequently be penetrated by corrosive agents (e.g. certain chemical and

microbiological substances), liquids (e.g. water, in which various ions are dissolved) or gases (e.g.

oxygen and carbon dioxide present in the atmosphere) through capillary absorption, hydrostatic

pressure, or diffusion. Some of them promote corrosion of reinforcing steel. In addition,

disintegration of concrete exposed to freeze-thaw cycles in cold climates may also compromise

the protection of reinforcement. All of these factors potentially impose a serious threat on the

durability and serviceability of concrete structures, which accounts for large amounts of

unplanned repairs, associated costs, out of service time and waste of materials and energy [1-4].

Among those above-mentioned factors, the dominant aggressive external influence for civil

infrastructure, e.g. bridges and harbour structures, is the ingress of chloride ions, typically present

in de-icing salts and sea water [5-7].

Modified hydrotalcites (MTHs) represent a group of technologically promising materials for

addition to concrete to improve its durability in aggressive environment, owing to their low cost,

relative simplicity of preparation, and plenty of unique composition variables that may be

adopted [8, 9]. Up to date, a lot of academic work and commercial interest on MHTs have been

invested, but relatively few studies focus on cementitious materials, particularly in exploiting

their potential applications in corrosion protection of reinforced concrete structures. In this

chapter, the mechanism of corrosion in reinforced concrete, factors that affect corrosion of

reinforcement and relevant corrosion preventive measures are briefly introduced. In addition, the

existing knowledge with regard to synthesis and characterization methods of MHTs, ion

exchange within the MHT structure as well as the application of MHTs in cementitious materials

were reviewed. As a new emerging class of smart additive of reinforced concrete, MHTs are

expected to contribute to the effort of searching for effective measures to improve the durability

of reinforced concrete.

2.2 Corrosion of the steel in concrete

Steel in concrete is normally in a non-corroding and passive condition. During hydration of

cement, a highly alkaline pore solution (pH>12.5) develops, which facilitates the formation of a

passive oxide/hydroxide film on the surface of the steel. This protective film is only a few

nanometers thick and can effectively insulate the steel from the pore electrolyte so that the onset

of corrosion is delayed, allowing decades of relatively low maintenance [10]. However, this

protective film can be disrupted (i.e., depassivation) by the ingress of chlorides and carbon

MHTs as Smart Additives of Concrete | 11

dioxide from the atmosphere (i.e., carbonation) [5, 11]. Once corrosion has started, three main

consequences occur [12]: 1) corrosion of the reinforcement (either local pitting in cases of

chloride attack or uniform corrosion in the case of carbonation); 2) decrease of ductility due to

reduction of cross section of the reinforcing steel; 3) cracking and spalling of the concrete cover

due to build-up of voluminous corrosion products, which in turn foster the ingress of moisture,

oxygen and other aggressive agents into the concrete. Figure 2.1 schematically shows typical

corrosion damage as occurring to reinforcing steel. In increasing order, these consequences may

significantly compromise the structural integrity and safety. The development of corrosion in

reinforced concrete structures can be divided in two main stages according to Tuutti's corrosion

model as shown in Figure 2.2 [13, 14]. The first stage is the initiation of corrosion, in which the

reinforcement is passive but phenomena that can lead to loss of passivity, e. g. chloride

penetration into or carbonation of the concrete cover take place. The second stage is corrosion

propagation which is dependent on the availability of water and oxygen in the vicinity of the steel.

It starts when the steel is depassivated and may proceed until some form of failure associated

with internal and/or surface cracking and spalling of the concrete cover. The time before such

failure is often referred to the service life of the reinforced concrete element, which is determined

by the total duration of these two stages. It is worth pointing out that modern service life design

philosophy somehow only considers the initiation stage and neglects the propagation stage.

Figure 2.1 Typical corrosion damage to reinforced concrete.

12 | Chapter 2

2.2.1 Carbonation

The gradual ingress of carbon dioxide causes its reaction with the alkaline constituents of

concrete present in the pore solution (mainly as sodium and potassium hydroxides) and in the

solid hydration phases, reducing the pH of the concrete pore solution to a value (pH≈9.0) where

passivity of reinforcing steel is destroyed [15]. Calcium hydroxide, Ca(OH)2, is the main alkali in

cement hydration products that reacts most readily with CO2 to produce calcite (CaCO3). The pH

of the pore solution can be reduced to 8.3 [16], if all Ca(OH)2 has been depleted. The reaction

that takes place in aqueous solution is:

Ca(OH)2 + CO2 CaCO3 + H2O (2.1)

In normal practice, the corrosive impact of carbonation is limited and relatively easy to avoid [5].

For concrete with high cementitious material content and low w/c (<0.4), carbonation rates are

typically on the order of 1 mm per decade or less [17]; loss of passivity due to this cause within a

normal design life is generally not a concern. However, carbonation must be anticipated at

concrete cracks, where air essentially has direct access to the reinforcement, irrespective of

concrete cover and quality. In general, corrosion occurs more rapidly under conditions of

Figure 2.2 Schematic illustration of various stages involved in the development of reinforcement

corrosion in concrete [13, 14].


exposure to chlorides and chloride induced corrosion is of a more serious concern for reinforced

concrete in particular in a salt-laden environment.

2.2.2 Chloride induced corrosion

Soluble chlorides present in seawater, ground water or de-icing salts may enter concrete through

capillary absorption or diffusion. Chlorides may also be introduced into concrete by using

contaminated aggregates or mixing water in the production of concrete. Sometimes excessive

chlorides can be present in chemical admixtures e.g., when calcium chloride is added in fresh

concrete as a set accelerator, which was reported to cause premature structural problems related

to steel corrosion [18]. While the chloride ion (Cl-) has only a small influence on the pH of pore

solution, concentrations as low as 0.6 kg/m3 by weight of concrete have been projected to

compromise steel passivity [17]. According to the European standard EN 206, the maximum

allowed chloride contents are 0.2-0.4% chloride ions by mass of binder for reinforced and 0.1-

0.2% for prestressed concrete [19]. Chlorides can be chemically or physically bound, being

adsorbed to the hydrated cement paste. Numerous researchers have related corrosion risk to

chloride content. The chloride content in concrete is usually expressed in terms of the amount of

total chloride by mass of cement or cementitious binder. It should be realized that the total

chloride content is not really responsible for corrosion. The free chlorides instead of bound

chlorides are the only ones that can possibly destroy the passive film on the surface of the

Figure 2.3 A typical corrosion cell in a salt-contaminated reinforced concrete slab [17, 20].

14 | Chapter 2

reinforcing steel and therefore initiate corrosion. In addition, plenty of oxygen and water are

available in all above-ground outdoor structures to facilitate significant corrosion. Corrosion is

assumed to start when the concentration of chlorides at the embedded steel surface has reached a

certain so-called “chloride threshold” value or critical chloride content. Once the chloride

threshold has been exceeded, the local disruption of the passive film initiates corrosion cells

between the active corrosion zones (anode) and the surrounding areas that are still passive

(cathode). Figure 2.3 [17, 20] schematically shows the electro-chemical process and the main

reactions involved in chloride contaminated reinforced concrete. As can be seen, the chlorides

actually act as a catalyst in this process accelerating the corrosion. The chloride ions release from

the soluble complex of ferrous chloride by hydroxyl ions produced in cathodic reaction, so the

rust itself contains no chlorides. Depending on the condition of the steel surface and corrosion

degree, the composition of the rust can be varied. As shown in Figure 2.4, the volume of

produced rust can be six times larger than that of the original steel [21]. The resulting expansive

stress due to the increased volume is restricted inside the concrete. As corrosion propagates, it

could eventually cause cracking, spalling, or delamination of the concrete cover, which further

the accessibility of airborne carbon dioxide. It may need to be noted that although corrosion due

to carbonation proceeds at a much lower rate than that due to chloride ingress [22], the

combination of chloride ingress and carbonation will make the corrosion process more

complicated and the corrosion risk due to the combined effect is sometime higher than that of

either cause separately [2, 23].

Figure 2.4 Iron corrosion products and their relative volume [21].


2.3 Factors that affect corrosion of reinforcement

2.3.1 Permeability of concrete

Concrete, as a porous and highly heterogeneous composite, is subject to the ingress of various

ionic and molecular species from the environment. Excessive accumulation of certain species can

be a vitally deleterious factor affecting the service life of concrete structures. In general,

permeability indicates the property of concrete to allow substances to intrude the concrete and

attack the reinforcing steel resulting in corrosion. The penetration of gases, liquids or ions into

concrete takes place through pore spaces in the cement paste and paste-aggregate interfaces or

microcracks according to four basic mechanisms, namely: capillary suction, permeation (due to

pressure gradients), diffusion (due to concentration gradients), and migration (due to electrical

potential gradients) [5, 24]. Permeability is believed to be a decisive characteristic of concrete

durability, which is related to its microstructural properties, such as the size, amount, distribution,

tortuosity and connectivity of pores and microcracks [25-27]. The microstructure of concrete is

influenced by the water to cement ratio (w/c) of the concrete, the degree of cement hydration and

the inclusion of supplementary cementitious materials (SCMs) which serve to refine the pore

structure [28]. Although the bulk hardened cement paste influences the permeability significantly,

the influence of the paste-aggregate interface is also not negligible [29, 30]. Microcracking in the

paste-aggregate interface (also known as ITZ) can alter the connectivity of pores making

disconnected pores become connected and creating pathways for the flow of water and dissolved

ions. Substantial hydration, a relatively low w/c ratio and a well-developed ITZ, as well as

inclusion of SCMs (e.g., blast furnace slag, fly ash, silica fume) [31, 32], all contribute to reduce

the permeability of concrete and thus enhance its resistance to corrosion damage or other relevant

degradation issues.

2.3.2 Chloride binding

Chloride binding in concrete may be defined as the interaction between the porous matrix of

concrete and chloride ions which results in effective removal of chlorides from the pore solution

[33]. It is known that hardened cement paste has the ability to bind chlorides which makes

concrete itself the first natural barrier against chloride ingress, although this binding capacity to

some degree is limited [34, 35]. Effective binding will remove a part of chloride from the

transport process as well as alter the pore solution concentration and therefore the concentration

gradient driving chloride diffusion. The critical chloride content expressed by total mass able to

initiate corrosion of reinforcing steel will be increased accordingly for concrete with a high

chloride binding capacity. Binding of chlorides can take place through both chemical

combination and physical adsorption. Chlorides may interact with hydrated cement forming

different chloride bearing AFm-like (tetracalcium aluminate monosulfate) phases, such as

16 | Chapter 2

Friedel’s salt (3CaO•Al2O3•CaCl2•10H2O) or its iron analogue (3CaO•Fe2O3•CaCl2•10H2O)

and/or Kuzel’s salt (3CaO•Al2O3•½CaSO4•½CaCl2•12H2O) and solid solutions with other AFm-

like phases [36-39]. Chloride-ettringite has however been reported to occur only below 0°C and

usually at very high chloride concentrations [40, 41]. Chloride can also physically be adsorbed on

the surface of the C-S-H as well as in interlayer spaces and possibly be chemisorbed by C-S-H as

an oxychloride complex (i.e., 3CaO•CaCl2•xH2O) or substituted in the structure [39, 42-45].

Compared to the main binding product (i.e., Friedel’s salt), oxychlorides are highly soluble and

only stable at very high chloride concentrations [38]. From samples exposed to long term

submersion in sea water, it appeared that most of the chloride was bound either chemically or

physically to C-S-H, rather than by chemical binding in detectable crystalline compounds [6].

There are many factors associated with the constituents of the concrete affecting the chloride

binding capacity. These factors include cement type, curing temperature and age, pH of pore

solution, w/c ratio, and chloride concentration and so on. Based on previous work, Glass and his

co-workers found that the C3A (tricalcium aluminate) content in cement and the type and

proportion of cement replacement materials are the most important factors influencing the

binding capacity of concrete [46]. Cements containing high content of C3A and C4AF

(tetracalcium aluminoferrite) can make hardened cement pastes rich in AFm, which has the

ability to accommodate chlorides and could be able to bind relatively large amounts of chlorides

[47-49]. An increase of sulfate content in cement reduces the chloride binding capacity since

Friedel’s salt is not stable when excess sulfate ions are present, and tends to be converted to

ettringite [50]. However, it is worthy to be pointed out that the effect of external sulfates on

binding may constitute a difference between exposure to sea water (with significant sulfate) and

de-icing salts (usually pure chloride-bearing compounds). Chloride binding is significantly

influenced by the incorporation of SCMs, such as blast furnace slag, fly ash, silica fume and

natural pozzolans. Generally, blast furnace slag and fly ash increase chloride binding but silica

fume can decrease the chloride binding capacity [51-53]. It is also reported that a part of the

bound chloride could be released when the pH drops to values below 12 (e.g., carbonation or

sulfate attack) [54]. Although the protective passive film at the steel surface is still

thermodynamically stable at this pH, it might be assumed that all the chloride released will be

available to sustain local passive film breakdown.

2.3.3 Concrete cover

Corrosion of steel in reinforced concrete structures is considerably different from corrosion of

steel exposed to the atmosphere, as in the former case the steel is protected by the concrete cover,

which provides a good physical barrier protecting the steel from corrosion [55, 56]. This barrier

also limits the diffusion of oxygen that is necessary for sustaining corrosion. The thickness and

quality of concrete cover are very important factors in controlling the level of protection provided

to embedded steel. A test conducted on reinforced concrete showed that when the cover thickness


is increased from 30 to 40 mm, the rate of corrosion of the embedded steel could be reduced

about 91% after six cycles of wetting and drying [57]. Eurocode 2 [58] fixes minimum values of

the concrete cover ranging from 10 mm for a dry environment up to 55 mm for prestressing steel

in chloride-laden environments. Considering the construction variability met in practice,

Eurocode 2 suggests that these minimum values should be increased to obtain nominal values by

10 mm. However, the cover thickness cannot exceed certain limits for mechanical and practical

reasons. In particular a very thick cover may not work as well in defending against corrosion as

expected. In extreme cases, a thick unreinforced layer of concrete cover may facilitate the

formation of microcracks or even macrocracks due to tensile forces exerted by drying shrinkage

of the outer layer. In normal practice, the cover thickness shouldn’t be above 70 to 90 mm [5].

2.3.4 Environmental conditions

Corrosion itself is the consequence of environmental impacts. The environmental conditions

relate to numerous factors that are not always independent of each other. From a corrosion point

of view, these factors have simultaneous and complex synergistic effects connected to both the

macroclimate and to local microclimatic conditions acting on the concrete structure such as: the

source (internal or external) and type of chloride contamination, humidity, the temperature and

the frequency of wetting-drying or freezing-thawing cycles. The source of chloride can influence

both chloride binding and pH of the concrete pore solution [51, 59]. It was reported that a higher

chloride content can be tolerated when it is added as an additive to the concrete mix [60, 61]. In

terms of the type of the chloride salts, previous studies showed that the associated cation has a

strong influence on the diffusion rates and the chemical binding of the chloride [62]. It was found

that the diffusion coefficient of chloride combined with divalent cations is greater than that

combined with monovalent cations. In addition, Tuutti reported that the ratio of chemically bound

chloride to freely dissolved chloride for CaCl2 is approximately three times that for KCl [14].

Compared to arid or temperate climates, tropical or equatorial environment generally increases

the corrosion risk in concrete [63]. Reinforced concrete under chloride exposure experiencing

wetting-drying cycles could concentrate the chlorides at locations which have the highest average

moisture content [64]. Consequently, a significant increase in the local chloride concentration

results, which in turn leads to the occurrence of pitting. Freezing-thawing cycles can also impose

serious corrosion risk to the reinforced concrete in particular in cold regions where de-icing salts

are often used [65]. The European standard EN 206-1 [19] explicitly defines exposure

classifications based on environmental conditions to which the concrete is exposed and that result

in effects on concrete or the reinforcement or embedded metal. It should be noted that this

classification limits to average regional exposure conditions; local microclimatic conditions

however have not been considered.

18 | Chapter 2

2.4 Corrosion preventive measures

Prevention of reinforcement corrosion can be primarily achieved in the design stage by using

high quality concrete, sufficient cover, and optimizing concreting materials and mix design as

indicated in the European standard EN 206-1 and Eurocode 2 [19, 58]. Additional prevention

options are generally adopted when severe environmental conditions occur (e.g. marine

environment or the application of deicing salts), or on structures requiring very long service life,

as well as in rehabilitation. A number of corrosion preventive measures with their own

advantages and limitations have been applied separately or in a synergic manner either to existing

chloride contaminated structures or to new structures [5, 17, 66-71]:

1) Physical barrier systems such as applying waterproofing membranes, sealers and overlays

on concrete surfaces to retain and/or restrict subsequent chloride penetration; surface

coating of the reinforcing steel such as using epoxy-based coating to restrict oxygen’s

access to the surface of rebar.

2) Electrochemical techniques such as cathodic protection for polarizing the steel to a

potential at which new pitting is suppressed and existing pit growth is at least

substantially retarded, electrochemical chloride extraction (ECE) which aims at removing

chloride ions from the cover zone of chloride contaminated concrete and electrochemical

realkalisation (ER) which aims at restoring the alkalinity of carbonated concrete cover.

3) Use of corrosion resistant reinforcement such as galvanized steel, alloy claddings,

stainless steel and polymer-based reinforcing materials.

4) Use of corrosion inhibitors added to fresh concrete as an admixture, or applied to the

hardened concrete surface or used as a surface treatment on rebars before concreting.

5) Use of new methods and technologies such as nano-materials, self-healing, mineral

admixtures and multifunctional additives, etc.

2.5 MHTs and their application in cementitious materials

2.5.1 General aspect

Hydrotalcite is a natural mineral that was discovered in 1842 in Sweden and the first exact

formula, [Mg6Al2(OH)16]CO3•4H2O, was published in 1915 by Manasse [72]. Subsequently

hydrotalcite gave its name to a large mineral group of naturally occurring Layered Double

Hydroxides (LDHs), which are also known as hydrotalcite-like materials or are commonly

referred to as hydrotalcites. Although LDHs are available as naturally occurring minerals,

nowadays a great number of pure LDH compounds has been prepared and characterized in the

laboratory [73, 74]. Since all of these synthetic LDHs have the same structure as their parent

material hydrotalcite, they can also be termed modified hydrotalcites (MHTs), a term which is

used throughout this thesis. MHTs are structurally similar to the minerals Brucite (Mg(OH)2), and


Portlandite (Ca(OH)2), in which a central divalent metal cation is surrounded by six hydroxyl

groups in an octahedral configuration. These octahedral units form infinite, charge-neutral layers

by edge-sharing, with the hydroxyl ions sitting perpendicular to the plane of the layers. The layers

then stack on top of one another to form the three-dimensional structure. In MHTs, a fraction of

divalent cations is isomorphously substituted by trivalent cations, which generates a net positive

charge on the layers that necessitates the incorporation of charge-balancing anions in the

interlayer galleries. The remaining free space of the interlayer is occupied by water molecules via

hydrogen bonding. The most common anion found in naturally occurring hydrotalcites is

carbonate. In practice, there is no significant restriction to the nature of the interlayer charge-

balancing anions. The MHT structure can accommodate various cations in the hydroxide layers

with varying MII/M

III ratios as well as a wide range of anionic species in the interlayer regions.

This high variation allows the large variety of materials to be represented by a general formula

[MII

1-x MIII

x (OH)2]x+

[(An-

x/n)]x-

·mH2O, where MII and M

III are di- and trivalent metals

respectively, (MII: Mg

2+, Ca

2+, Zn

2+, Ni

2+, etc., M

III: Al

3+, Fe

3+, Ga

3+, Co

3+, etc,) and A

n- is an

exchangeable interlayer anion (CO32-

, SO42-

, Cl-, NO3

-, NO2

- and carboxylates, amino or

polyamino carboxylates, etc.) with valence n. The x value is in the range 0.20-0.33. A typical

structure of MHTs can be schematically shown as in Figure 2.5.

In cement chemistry, LDHs or hydrotalcite-like materials represent a group of phases that

form during cement hydration. The hydration products of tricalcium aluminate (C3A) and

tetracalcium aluminoferrite (C4AF) are hexagonal-layered materials denoted C2(A,F)H8,

Figure 2.5 Schematic representation of a typical carbonate hydrotalcite crystal structure (d-spacing

value d003 is the length of the interlayer space).

20 | Chapter 2

C4(A,F)H13, and C4(A,F)H19. These hydrates along with the AFm phases are recognized as

hydrotalcite-like compounds [75]. As mentioned above, it is well known that the stability of AFm

phases play a very important role in controlling the performance of concrete since chloride ions

could interact with hydrated cement forming chloroaluminate phases, such as Friedel’s (a

chloride-bearing AFm phase) salt and/or Kuzel’s (a chloride- and sulfate-bearing AFm phase) salt

and solid solutions with other AFm phases [39, 47]. Formation of these specific chloride-

containing salts is potentially a mechanism for retarding chloride diffusion and thus mitigating

chloride-induced corrosion.

2.5.2 Synthesis and characterization

A wide variety of cation combinations (e.g., MII, M

III) and different anions in the interlayer are

found in the structures of synthetic MHTs. It is suggested that only MII and M

III ions having an

ionic radius not too different from that of Mg2+

may be accommodated in the octahedral sites of

the brucite-like layers to form LDH compounds [74]. Cations that are too small, such as Be2+

or

too large such as Cd2+

will not be able to fit into the lattice, hence, other types of structures will

be formed. The most widely occurring compositional range corresponds approximately to an

MIII

/(MII+M

III) ratio of x between 0.20 and 0.33 as mentioned above. There are a number of

techniques that have been successfully applied to synthesize MTHs, among which three main

methods are frequently used. The most commonly method used is a simple coprecipitation of two

metal salts in alkaline solution at a constant pH value of about 10. The second one is based on the

classical ion exchange process in which the guest anions are exchanged with the anions in the

interlayer spaces of preformed LDHs to produce specific anion intercalated MTHs. The third

method is a lattice reconstruction after heating (calcination), which is based on the hydrotalcite-

like materials’ unique feature of “structural memory effect”, due to which the original structure is

reproduced after re-hydration. Therefore, this method is called the “calcination-rehydration”. In

addition, some other methods, such as sol-gel synthesis using ethanol and acetone solutions, and

a fast nucleation process followed by a separate ageing step at elevated temperatures have also

been reported. A common concern with all the methods is that in preparations of MHTs with

anions other than carbonate, it is always important to avoid contamination from ambient CO2

since the carbonate anion is readily intercalated and firmly held in the interlayer space.

Consequently, decarbonated and deionized water has been often used and exposure of the

reacting materials to the atmosphere needs to be carefully avoided during the preparation. More

details about the synthetic techniques for the preparation of MHTs can be found in the literature

[76, 77]. It should be pointed out that once formed, MHTs are stable in alkaline and neutral

solution although they are soluble at pH below 4.0 [78], which means that in alkaline and

carbonated concrete (pH around 9.0), MHT is a stable solid compound.

A variety of modern techniques has been employed to characterize synthesized MHTs.

Powder X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR) are


principally performed to check the structural characteristics of MHTs whilst other techniques

such as X-ray fluorescence (XRF) [79], X-ray absorption spectroscopy (XRS) [80] and electron

spin resonance (ESR) [81] as well as Raman spectroscopy [82] are less extensively employed

although reported. Energy-dispersive X-ray spectroscopy (EDX) [75], nuclear magnetic

resonance (NMR) [83, 84], atomic absorption spectroscopy [85], ion coupled plasma optical

emission spectrometer (ICP-OES) [86], ion chromatography (IC) [82] and Mössbauer

spectroscopy [79] as well as elemental analysis of carbon/sulfur, oxygen/nitrogen & hydrogen [82,

86] are also performed to determine the compositional formulae of MHTs. The thermal behavior

of MHTs are commonly studied using thermogravimetry (TG), differential scanning calorimetry

(DSC) and differential thermal analysis (DTA) [87, 88]. In certain cases, these techniques are

used in combination with a mass spectral analyzer to study the nature of the evolved gases during

the thermal treatment. Morphology and particle size are often checked by electron microscope

(e.g., SEM, TEM, FESEM) and particle size analyzer.

2.5.3 Ion exchange of MHTs and its role in capturing chloride

The key feature of the application of MHTs is their high anionic exchange capacity (2 to 4.5

millequivalents/g) which makes exchange of the interlayer ion by a wide range of organic or

inorganic anions versatile and easily achieved [89-91]. The anion-exchange capacity of MHTs is

inversely dependent on the layer charge density (i.e. the MII/M

III molar ratios in the brucite-like

sheets). Additionally, it is affected by the nature of the interlayer anions initially present.

Generally, the naturally occurring hydrotalcite, which has the chemical formula

[Mg6Al2(OH)16]CO3•4H2O, is regarded as the parent material of MHTs. The carbonate anion is

tenaciously held in the interlayer space, so ion-exchange reaction hardly occurs in carbonate form

hydrotalcite [92]. However, such materials are believed to have a great potential to be modified or

tailor-made. MHTs have greater affinities for multivalent anions relative to monovalent anions

[74, 76]. It was reported that the ion-exchange equilibrium constant tends to increase as the

diameter of the anion decreases [93]. Previous research [94] has also found that the interlayer

interactions can be direct or mediated through other species present in the interlayer region, both

by coulomb forces between the positively charged layers and the anions in the interlayer and by

hydrogen bonding between the hydroxyl groups of the layer with the anions and the water

molecules in the interlayer. In summary, MHTs have a very rich interlayer chemistry and can

participate in anion exchange reactions with great facility. For the promising use as a smart

functional additive for concrete against chloride attack, certain inorganic or organic anions with

known inhibitive properties may be intercalated in the structure of hydrotalcite. Then, an ion

exchange reaction in MHT can be triggered possibly 'automatic' upon arrival of external chloride

ions. The present chloride ions are captured and the intercalated inhibitive anions are

simultaneously released, which provide further inhibitory protection to the reinforcing steel. The

22 | Chapter 2

anion exchange process can be described in the following Eq. 2.2 [95] and also shown

schematically in Figure 2.6:

HT-Inh + Cl-(aq) HT-Cl + Inh

-(aq) (2.2)

where Inh- represents the intercalated inhibitive anions,; Cl

- is the free chloride ions present in

concrete pore solution. As such, MHT plays a dual-role against chloride-induced corrosion in

reinforced concrete acting as a chloride scavenger and providing corrosion inhibitors in parallel

as an internal inhibitor reservoir and protecting reinforcing steel from corrosion continuously.

2.5.4 Application in cementitious materials

Previous studies have demonstrated that several ion exchangers based on hydrotalcite dispersed

in polymeric coatings are strong corrosion inhibitors for the metallic substrates [95-106]. The

corrosion is hindered due to the release of an inhibitive anion that diffuses through the pore space

of the coating, and it is exchanged for chloride ions in the environment, which are “entrapped”

into the interlayer space of the unique layered structure of hydrotalcite [95-97, 107-109]. These

Figure 2.6 Working mechanism of MHTs in reinforced concrete exposed to external chloride load.


results suggest that MHT has a high potential as active component in concrete with corrosion

protection properties that can be tailor-made.

Sustainable development of concrete infrastructure continues to be of importance to the

construction industry. The use of supplementary materials or functional additives in concrete is

an integral component of these initiatives. Hydrotalcites or hydrotalcite-like phases have been

found in hydrated slag cements, which are known to be able to bind more chloride ions than pure

Portland cements [51, 52, 110]. Nonaka and Sato [111] patented a cement modifier that contains

hydrotalcite or hydrotalcite and blast furnace slag and provided a method capable of blending this

modifier into concrete/mortar without inhibiting the hydration reaction of cement and the method

for imparting required corrosion resistance and durability into the cement/mortar. The existence

of hydrotalcite-like phases such as Friedel’s salt or its iron analogue has been believed to be a

main contributor for chloride binding in cementitious materials and thus has enhanced the

corrosion resistance of reinforced concrete. The beneficial effects of Friedel’s salt on binding

chloride have indicated the possibilities for using other hydrotalcite-like compounds in concrete

as an effective chloride capturer or scavenger. Increased binding capacity would consequently

slow down chloride transport in concrete. Furthermore, MHTs are expected to be able to retain

bound chloride even in a neutral and high temperature environment, for example if the pH of the

pore solution drops due to carbonation. Raki et al. [75] demonstrated the promise of MHTs as

suitable hosts for intercalation of organic admixtures with the long-term view of controlling their

release rate in concrete by blending inorganic-organic nanocomposites (in small amounts) with

the cement. In their study, nitrobenzoic acid (NBA), naphthalene-2,6-disulfonic acid (26NS), and

naphthalene-2 sulfonic acid (2NS) salts, commonly used as admixtures in concrete production,

were intercalated through anion-exchange of nitrate in the host material, [Ca2Al(OH)6]NO3•nH2O.

It was suggested that potential future applications of these composite materials could be to

control the effect of admixtures on the kinetics of cement hydration by programming their

temporal release. In addition, in a recent patent [112], Raki and Beaudoin disclose that a

controlled release formulation for a cement-based composition can be produced by means of

hydrotalcites modified with different anions, which confer functions such as an accelerator, a set

retarder, and a superplasticizer when they are released. The patent also mentions the potential

corrosion inhibition function of this formulation but unfortunately no examples are supplied.

Ashida et al. [113] and Mihara et al. [114] disclose a cement admixture that comprises

hydrotalcite and other functional additives in which it is claimed that the hydrotalcite-based

admixture has excellent ability to capture chloride or carbonate ions, thus can simultaneously

prevent concrete deterioration caused by chloride and carbonation. Tatematsu et al. [115]

synthesized a hydrocalumite-like material (calcium-aluminum based MHTs) and added it into

cement mixtures as a salt adsorbent. Their results showed that once the admixed salt adsorbent

contacts with chloride ions it could adsorb them and release the intercalated nitrite anions in the

meantime. The released nitrite ions on the other hand could work as an efficient inhibitor for

chloride-induced corrosion. The corrosion inhibiting effect of the salt adsorbent on chloride-

24 | Chapter 2

induced corrosion was further confirmed by the experiments performed with a large-size

specimen. In another patent, Tatematsu et al. [116] also disclosed a hydrocalumite that contains

nitrite or nitrate ions that can be added directly to the concrete as a chloride ion scavenger to

prevent choride-induced corrosion. Kang et al. [117] patented a corrosion inhibitive repair

method for reinforced concrete structures using repair mortar containing nitrite-based

hydrocalumite and confirmed its effectiveness in inhibiting corrosion to some degrees. A cement

additive for inhibiting concrete deterioration was developed by Feng [118] and Tatematsu et al.

[119] with a mixture of an inorganic cationic exchanger: a calcium-substituted zeolite capable of

absorbing alkali ions (e.g., sodium, potassium) and an inorganic anionic exchanger:

hydrocalumite capable of exchanging anions (e.g., chlorides, nitrates, sulfates, etc.) that is

directly incorporated into mortar or concrete mixtures. The testing results from this specific

mixture indicated the potential of protecting concrete from deterioration by exchange of alkali

and chloride ions to mitigate alkali-aggregate reaction and corrosion of reinforcing steel under the

control of ion exchange mechanism. It is noted that corrosion inhibition systems based on this

type of cement additive refer to hydrotalcites intercalated with inorganic anions. However, the

possibilities of these kinds of corrosion inhibition systems using organic anions as the interlayer

inhibitive species are not discussed anyway. In another relevant patent, Kashima [120] describes

on one hand, the use of a carbonate hydrotalcite as a system able to incorporate chloride ions

from the environment in the hydrotalcite structure and on the other hand, polycarboxylic acid and

organoamine as corrosion inhibitors are also added to the cement-based materials to prevent

carbonation and corrosion of reinforcing steel due to the penetration of water, oxygen, etc., as

well as to adsorb and remove salt contained in the concrete. However, this work also does not

deal with the integral use of organic anions intercalated in the hydrotalcite structure although

those organic species are used as components of the concrete mixture. In view of the ease of

hosting of molecular anions in the interlayer space of MHTs, certain organic inhibitors which are

suitable for use in concrete could also be intercalated into the structures of MHTs. For the

envisaged use as a new additive to concrete, Fischer et al [121] recently patented a corrosion

inhibition system for reinforced concrete against chloride induced corrosion that consists of

MHTs intercalated with organic species, in particular eco-friendly amino acids, which can be

directly applied (without paint or polymeric carrier) as aqueous emulsion on a metallic substrate,

or it can be mixed into fresh concrete with the various components. The corrosion inhibition

system is believed to be able to work as a smart storage of active corrosion protective agent and

therefore can keep the chloride concentration below the threshold that would initiate corrosion of

reinforcing steel during the concrete service life. The experimental results from their work

demonstrate that the corrosion inhibition system based on some amino acids modified

hydrotalcites is at least as effective as those based on MHTs intercalated with inorganic inhibitors

in particular such as nitrites which are well known for their excellent corrosion inhibiting

performance in reinforced concrete system.


2.6 Concluding remarks

Corrosion-related durability issues are nowadays associated with high maintenance costs and

safety concerns, so any improvement in the design, production, construction, maintenance, and

materials performance of concrete could be crucial in improving the service life of concrete

structures, which in turn may have enormous social, economic and environmental benefits.

Modified hydrotalcites (MTHs) designed for chloride scavenging and release of inhibiting agents

represent a class of technologically promising materials owing to their low cost, relative

simplicity of preparation, and plenty of composition variables that may be adopted. Up to date, a

lot of academic work and commercial interest in MHTs has been invested, among others for

corrosion protection of metals, but not much of it has been directed towards cementitious

materials. In particular, research is needed for exploiting their potential applications in corrosion

protection of reinforced concrete. We are confident that future work on applications of new smart

functional concrete additives based on MHTs will expand rapidly and contribute greatly to the

effort of searching for effective measures to improve the durability of reinforced concrete.

References

[1] Yunovich M, Thompson NG. Corrosion of highway bridges: Economic impact and control

methodologies. Concrete International. 2003;25(1):52-7.

[2] Küter A. Management of Reinforcement Corrosion: A Thermodynamic Approach, PhD thesis,

Technical University of Denmark 2009.

[3] Arockiasamy M, Barbosa M. Evaluation of conventional repair techniques for concrete bridges-Final

Draft Report, WPI 0510847. Tallahassee, FL: Florida Department of Transportation; 2000.

[4] Page C. Degradation of reinforced concrete: Some lessons from research and practice. Materials and

Corrosion. 2012;63(12):1052-8.



[6] Polder R, Larbi J. Investigation of concrete exposed to North Sea water submersion for 16 years.

Heron. 1995;40(1):31-56.

[7] Polder RB, Hug A. Penetration of chloride from de-icing salt into concrete from a 30 year old bridge.

Heron. 2000;45(2):109-24.

[8] Costantino U, Ambrogi V, Nocchetti M, Perioli L. Hydrotalcite-like compounds: versatile layered

hosts of molecular anions with biological activity. Microporous and Mesoporous Materials.

2008;107(1):149-60.

[9] Cavani F, Trifirò F, Vaccari A. Hydrotalcite-type anionic clays: Preparation, properties and

applications. Catalysis today. 1991;11(2):173-301.

[10] Gaidis JM. Chemistry of corrosion inhibitors. Cement and Concrete Composites. 2004;26(3):181-9.

[11] Monticelli C, Frignani A, Balbo A, Zucchi F. Influence of two specific inhibitors on steel corrosion in

a synthetic solution simulating a carbonated concrete with chlorides. Materials and Corrosion.

2011;62(2):178-86.

[12] Andrade C, Alonso C, Molina F. Cover cracking as a function of bar corrosion: Part I-Experimental

test. Materials and Structures. 1993;26(8):453-64.

26 | Chapter 2

[13] Daigle L, Cusson D, Lounis Z. Extending service life of high performance concrete bridge decks with

internal curing. In: Tada aki Tanabe KS, Hirozo Mihashi, Ryoichi Sato, Kochi Maekawa, editor. Creep,

Shrinkage and Durability Mechanics of Concrete and Concrete Structures: Proceedings of the

CONCREEP 8 conference. Ise Shima, Japan2008. p. 51-3.

[14] Tuutti K. Corrosion of steel in concrete. Stockholm: Cement and Concrete Research Institute; 1982.


[16] Papadakis V, Fardis M, Vayenas C. Effect of composition, environmental factors and cement-lime

mortar coating on concrete carbonation. Materials and Structures. 1992;25(5):293-304.

[17] Hartt WH, Powers RG, Leroux V, Lysogorski DK. A critical literature review of high-performance

corrosion reinforcements in concrete bridge applications (No.FHWA-HRT-04-093). 2004.

[18] Pruckner F, Gjørv O. Effect of CaCl2 and NaCl additions on concrete corrosivity. Cement and

Concrete Research. 2004;34(7):1209-17.

[19] EN206-1 Concrete-Part 1: Specification, Performance, Production and Conformity. European

Committee for Standardization,2000.

[20] Shi X, Fay L, Yang Z, Nguyen TA, Liu Y. Corrosion of deicers to metals in transportation

infrastructure: Introduction and recent developments. Corrosion reviews. 2009;27(1-2):23-52.

[21] Marcotte TD. Characterization of Chloride-Induced Corrosion Products that form in Steel-Reinforced

Cementitious. PhD thesis. Waterloo, University of Waterloo; 2001.

[22] Basheer L, Kropp J, Cleland DJ. Assessment of the durability of concrete from its permeation

properties: a review. Construction and Building Materials. 2001;15(2):93-103.

[23] Moreno M, Morris W, Alvarez M, Duffó G. Corrosion of reinforcing steel in simulated concrete pore

solutions: effect of carbonation and chloride content. Corrosion Science. 2004;46(11):2681-99.

[24] Stanish K, Hooton R, Thomas M. Testing the Chloride Penetration Resistance of Concrete: A

Literature Review, FHWA contract DTFH61-97-R-00022 report. University of Toronto, Canada. 1997.

[25] Savas BZ. Effects of microstructure on durability of concrete, PhD thesis, North Carolina State

University; 1999.

[26] Glass G, Buenfeld N. Chloride‐induced corrosion of steel in concrete. Progress in Structural

Engineering and Materials. 2000;2(4):448-58.

[27] Yang Z, Shi X, Creighton AT, Peterson MM. Effect of styrene–butadiene rubber latex on the chloride

permeability and microstructure of Portland cement mortar. Construction and Building Materials.

2009;23(6):2283-90.

[28] Lothenbach B, Scrivener K, Hooton R. Supplementary cementitious materials. Cement and Concrete

Research. 2011;41(12):1244-56.

[29] Neville A. Suggestions of research areas likely to improve concrete. Concrete International.

1996;18(5):44-9.

[30] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. New York: Pearson Education, Inc; 2003.

[31] Mailvaganam NP. Repair and protection of concrete structures. Boca Raton: CRC Press; 1992.

[32] Hearn N, Hooton R, Nokken M. Pore structure, permeability, and penetration resistance

characteristics of concrete. In: Lamond J, Pielert J, editors. Significance of Tests and Properties of

Concrete and Concrete-Making Materials2006. p. 238-50.

[33] Lawler JS. Guidelines for concrete mixtures containing supplementary cementitious materials to

enhance durability of bridge decks: Transportation Research Board; 2007.

[34] Birnin-Yauri U, Glasser F. Friedel’s salt, Ca2 Al (OH)6 (Cl, OH)· 2H2O: its solid solutions and their

role in chloride binding. Cement and Concrete Research. 1998;28(12):1713-23.

[35] Rosenberg A, Hansson C, Andrade C. Mechanisms of corrosion of steel in concrete I. In: Skalny J,

editor. Materials science of concrete: American Ceramic Society; 1989. p. 285-314.


[36] Nielsen E, Herfort D, Geiker M, Hooton R. Effect of solid solutions of AFm phases on chloride

binding. Proceedings of the 11th International Congress on the Chemistry of Cement, Durban, South

Africa2003. p. 11-6.

[37] Glasser F, Kindness A, Stronach S. Stability and solubility relationships in AFm phases: Part I.

Chloride, sulfate and hydroxide. Cement and Concrete Research. 1999;29(6):861-6.

[38] Hobbs MY. Solubilities and ion exchange properties of solid solutions between the OH-, Cl

- and

CO32-

end members of the monocalcium aluminate hydrates. PhD thesis, University of Waterloo; 2001.

[39] Balonis M. The Influence of Inorganic Chemical Accelerators and Corrosion Inhibitors on the

Mineralogy of Hydrated Portland Cement Systmes, PhD thesis, University of Aberdeen Aberdeen, UK;

2010.

[40] Zibara H, Hooton D, Yamada K, Thomas M. Roles of cement mineral phases in chloride binding.

Cement Science and Concrete Technology. 2002;56:384-91.

[41] Ben‐Yair M. Studies on the Stability of Calcium Chloraluminate. Israel Journal of Chemistry.

1971;9(4):529-36.

[42] Beaudoin JJ, Ramachandran VS, Feldman RF. Interaction of chloride and CSH. Cement and Concrete

Research. 1990;20(6):875-83.

[43] Song S, Jennings HM. Pore solution chemistry of alkali-activated ground granulated blast-furnace

slag. Cement and Concrete Research. 1999;29(2):159-70.

[44] Luping T, Nilsson L-O. Chloride binding capacity and binding isotherms of OPC pastes and mortars.

Cement and Concrete Research. 1993;23(2):247-53.

[45] Hirao H, Yamada K, Takahashi H, Zibara H. Chloride binding of cement estimated by binding

isotherms of hydrates. Journal of Advanced Concrete Technology. 2005;3(1):77-84.

[46] Glass G, Hassanein N, Buenfeld N. Neural network modelling of chloride binding. Magazine of


[47] Csizmadia J, Balázs G, Tamás FD. Chloride ion binding capacity of aluminoferrites. Cement and


[48] Hussain SE, Al-Saadoun S. Effect of cement composition on chloride binding and corrosion of

reinforcing steel in concrete. Cement and Concrete Research. 1991;21(5):777-94.

[49] Rasheeduzzafar S, Hussain E, Al-Saadoun S. Effect of tricalcium aluminate content of cement on

chloride binding corrosion of reinforcing steel in concrete. ACI materials journal. 1993;89(1).

[50] Hussain SE, Al-Gahtani AS. Influence of sulfates on chloride binding in cements. Cement and


[51] Arya C, Buenfeld N, Newman J. Factors influencing chloride-binding in concrete. Cement and


[52] Dhir R, El-Mohr M, Dyer T. Chloride binding in GGBS concrete. Cement and Concrete Research.

1996;26(12):1767-73.

[53] Mangat P, Molloy B. Chloride binding in concrete containing PFA, GBS or silica fume under sea-

water exposure. Magazine of Concrete Research. 1995;47(171):129-41.

[54] Glass GK, Buenfeld NR. The presentation of the chloride threshold level for corrosion of steel in

concrete. Corrosion Science. 1997;39(5):1001-13.

[55] Reichling K, Raupach M, Broomfield J, Gulikers J, Nygaard P, Schneck U, et al. Local detailed

inspection methods regarding reinforcement corrosion of concrete structures. Materials and Corrosion.

2013;64(2):128-34.

[56] Angst U, Elsener B, Jamali A, Adey B. Concrete cover cracking owing to reinforcement corrosion–

theoretical considerations and practical experience. Materials and Corrosion. 2012;63(12):1069-77.

28 | Chapter 2

[57] Song Y-P, Song L-Y, Zhao G-F. Factors affecting corrosion and approaches for improving durability

of ocean reinforced concrete structures. Ocean engineering. 2004;31(5):779-89.

[58] prEN 1992-1-1, Eurocode 2: Design of concrete structures-Part 1: General rules and rules for

buildings European Committee for Standardization,2002.

[59] Tritthart J. Chloride binding in cement II. The influence of the hydroxide concentration in the pore

solution of hardened cement paste on chloride binding. Cement and Concrete Research.

1989;19(5):683-91.

[60] Hansson C, Frølund T, Markussen J. The effect of chloride cation type on the corrosion of steel in

concrete by chloride salts. Cement and Concrete Research. 1985;15(1):65-73.

[61] Ann K-Y, Jung H, Kim H, Kim S, Moon HY. Effect of calcium nitrite-based corrosion inhibitor in

preventing corrosion of embedded steel in concrete. Cement and Concrete Research. 2006;36(3):530-5.

[62] Bakker RF. On the cause of increased resistance of concrete made from blast-furnace cement to alkali

reaction and to sulfate corrosion. PhD thesis, RWTH-Aachen; 1980.

[63] Corvo F, Perez T, Dzib L, Martin Y, Castañeda A, Gonzalez E, et al. Outdoor–indoor corrosion of

metals in tropical coastal atmospheres. Corrosion Science. 2008;50(1):220-30.

[64] Tuutti K. Effect of cement type and different additions on service life. Proc Int Conf "Concrete 2000".

Dundee, Scotland, UK: E & FN Spon, Chapman & Hall, London; 2000. p. 1285-95.

[65] Shi X, Fay L, Peterson MM, Yang Z. Freeze–thaw damage and chemical change of a portland cement

concrete in the presence of diluted deicers. Materials and Structures. 2010;43(7):933-46.




1996;10(5):391-402.

[68] Page CL. Corrosion and protection of reinforcing steel in concrete. In: Page CL, Page MM, editors.

Durability of concrete and cement composites. Boca Raton, FL: Woodhead, Cambridge/CRC Press;

2007. p. 136-86.

[69] Polder R. Electrochemical techniques for corrosion protection and maintenance. In: Böhni H, editor.

Corrosion in Reinforced Concrete Structures, Woodhead Publishing, Cambridge, UK. Boca Raton, FL:

Woodhead, Cambridge/CRC Press; 2005. p. 215-41.

[70] Yang Z, Hollar J, He X, Shi X. A self-healing cementitious composite using oil core/silica gel shell

microcapsules. Cement and Concrete Composites. 2011;33(4):506-12.

[71] Han B, Yang Z, Shi X, Yu X. Transport properties of carbon-nanotube/cement composites. Journal of

materials engineering and performance. 2013;22(1):184-9.

[72] Manasse E. Idrotalcite e piroaurite. Atti Soc Toscana Sci Nat. 1915;24:92-105.

[73] Costa FR, Abdel-Goad M, Wagenknecht U, Heinrich G. Nanocomposites based on polyethylene and

Mg–Al layered double hydroxide. I. Synthesis and characterization. Polymer. 2005;46(12):4447-53.

[74] De Roy A, Forano C, Besse J. Layered double hydroxides: synthesis and post-synthesis modification.

In: Rives V, editor. Layered Double Hydroxides: Present and Future. New York: Nova Science

Publishers Inc.; 2001. p. 1-39.

[75] Raki L, Beaudoin J, Mitchell L. Layered double hydroxide-like materials: nanocomposites for use in

concrete. Cement and Concrete Research. 2004;34(9):1717-24.

[76] He J, Wei M, Li B, Kang Y, Evans DG, Duan X. Preparation of layered double hydroxides. Layered

double hydroxides: Springer; 2006. p. 89-119.

[77] Kanezaki E. Preparation of Layered Double Hydroxides. Interface Science and Technology.

2004;1:345-73.


[78] Kloprogge JT, Hickey L, Frost RL. The effects of synthesis pH and hydrothermal treatment on the

formation of zinc aluminum hydrotalcites. Journal of Solid State Chemistry. 2004;177(11):4047-57.

[79] Raki L, Rancourt DG, Detellier C. Preparation, characterization, and Mössbauer spectroscopy of

organic anion intercalated pyroaurite-like layered double hydroxides. Chemistry of Materials.

1995;7(1):221-4.

[80] Moggridge G, Parent P, Tourillon G. A NEXAFS study of the orientation of benzoate intercalated

into a layer double hydroxide. Physica B: Condensed Matter. 1995;208:269-70.

[81] Ukrainczyk L, Chibwe M, Pinnavaia T, Boyd S. ESR study of cobalt (II) tetrakis (N-methyl-4-

pyridiniumyl) porphyrin and cobalt (II) tetrasulfophthalocyanine intercalated in layered

aluminosilicates and a layered double hydroxide. The Journal of Physical Chemistry.

1994;98(10):2668-76.

[82] Rozov K, Berner U, Taviot-Gueho C, Leroux F, Renaudin G, Kulik D, et al. Synthesis and

characterization of the LDH hydrotalcite–pyroaurite solid-solution series. Cement and Concrete

Research. 2010;40(8):1248-54.

[83] Carlino S. The intercalation of carboxylic acids into layered double hydroxides: a critical evaluation

and review of the different methods. Solid State Ionics. 1997;98(1):73-84.

[84] Rey F, Fornés V, Rojo JM. Thermal decomposition of hydrotalcites. An infrared and nuclear

magnetic resonance spectroscopic study. Journal of the Chemical Society, Faraday Transactions.

1992;88(15):2233-8.

[85] Whilton NT, Vickers PJ, Mann S. Bioinorganic clays: synthesis and characterization of amino-

andpolyamino acid intercalated layered double hydroxides. Journal of Materials Chemistry.

1997;7(8):1623-9.

[86] Kang MR, Lim HM, Lee SC, Lee S-H, Kim KJ. Layered double hydroxide and its anion exchange

capacity. Advances in Technology of Materials and Materials Processing Journal. 2004;6:218-23.

[87] Kanezaki E, Kinugawa K, Ishikawa Y. Conformation of intercalated aromatic molecular anions

between layers of Mg/Al-and Zn/Al-hydrotalcites. Chemical Physics Letters. 1994;226(3):325-30.

[88] Yang Z, Fischer H, Polder R. Synthesis and characterization of modified hydrotalcites and their ion

exchange characteristics in chloride-rich simulated concrete pore solution. Cement and Concrete

Composites. 2014;47:87-93.

[89] Meyn M, Beneke K, Lagaly G. Anion-exchange reactions of layered double hydroxides. Inorganic

chemistry. 1990;29(26):5201-7.

[90] Van der Ven L, Van Gemert M, Batenburg L, Keern J, Gielgens L, Koster T, et al. On the action of

hydrotalcite-like clay materials as stabilizers in polyvinylchloride. Applied Clay Science.

2000;17(1):25-34.

[91] Shchukin DG, Zheludkevich M, Yasakau K, Lamaka S, Ferreira MGS, Möhwald H. Layer-by-Layer

Assembled Nanocontainers for Self-Healing Corrosion Protection. Advanced Materials.

2006;18(13):1672-8.

[92] Newman SP, Jones W. Layered double hydroxides as templates for the formation of supramolecular

structures. New York: Cambridge University Press; 2002.

[93] Miyata S. Anion-exchange properties of hydrotalcite-like compounds. Clays and Clay Minerals.

1983;31(4):305-11.

[94] Taylor H. Crystal structures of some double hydroxide minerals. Mineralogical Magazine.

1973;39(304):377-89.

[95] Buchheit RG, Guan H, Mahajanam S, Wong F. Active corrosion protection and corrosion sensing in

chromate-free organic coatings. Progress in Organic Coatings. 2003;47(3):174-82.

30 | Chapter 2

[96] Tedim J, Poznyak SK, Kuznetsova A, Raps D, Hack T, Zheludkevich ML, et al. Enhancement of

active corrosion protection via combination of inhibitor-loaded nanocontainers. ACS Appl Mater

Interfaces. 2010;2(5):1528-35.

[97] Williams G, McMurray HN. Anion-exchange inhibition of filiform corrosion on organic coated

AA2024-T3 aluminum alloy by hydrotalcite-like pigments. Electrochemical and Solid-State Letters.

2003;6(3):B9-B11.

[98] Sinko J, Kendig MW. Layered double hydroxides modified with a weak organic or inorganic acid; a

hydrotalcite modified with phosphates, chromates, molybdates, or 2,5-dimercapto-1,3,4-thiadiazole;

coil and aircraft primer. US Patent US7662312 B2, 16 Feb 2010.

[99] Miyata S. Coating with hydrotalcite solid solutions. US Patent US4761188 A, 2 Aug 1988.

[100] Miyazawa S, Morihira Y, Fujimori H. Mixture of hydrotalcite and binder. US Patent US6383270 B1,

7 May 2002.

[101] Gichuhi T, Novelli W. Synergistic corrosion inhibitor. US Patent US 2005/0235873 A1, 27 Oct

2005.

[102] Abe Y, Chiba M, Uchiyama H. Hydrotalcite powder comprising nitrite ions, method for preparation

thereof, and corrosion inhibitor composition comprising such. Japan Patent 2005336002, 8 Dec 2005.

[103] McMurray N, Worsley D. Anti-corrosion pigments. EP Patent EP1534787 A2, 1 Jun 2005.

[104] Hang TTX, Truc TA, Nam TH, Oanh VK, Jorcin J-B, Pébère N. Corrosion protection of carbon

steel by an epoxy resin containing organically modified clay. Surface and Coatings Technology.

2007;201(16):7408-15.

[105] Zheludkevich ML, Poznyak SK, Rodrigues LM, Raps D, Hack T, Dick LF, et al. Active protection

coatings with layered double hydroxide nanocontainers of corrosion inhibitor. Corrosion Science.

2010;52(2):602-11.

[106] Hang TTX, Truc TA, Duong NT, Vu PG, Hoang T. Preparation and characterization of

nanocontainers of corrosion inhibitor based on layered double hydroxides. Applied Clay Science.

2012;67-68:18-25.

[107] Tedim J, Kuznetsova A, Salak AN, Montemor F, Snihirova D, Pilz M, et al. Zn–Al layered double

hydroxides as chloride nanotraps in active protective coatings. Corrosion Science. 2012;55:1-4.

[108] Poznyak SK, Tedim J, Rodrigues LM, Salak AN, Zheludkevich ML, Dick LF, et al. Novel inorganic

host layered double hydroxides intercalated with guest organic inhibitors for anticorrosion applications.

ACS Appl Mater Interfaces. 2009;1(10):2353-62.

[109] Tedim J, Zheludkevich ML, Salak AN, Lisenkov A, Ferreira MGS. Nanostructured LDH-container

layer with active protection functionality. Journal of Materials Chemistry. 2011;21(39):15464-70.

[110] Kayali O, Khan MSH, Sharfuddin Ahmed M. The role of hydrotalcite in chloride binding and

corrosion protection in concretes with ground granulated blast furnace slag. Cement and Concrete

Composites. 2012;34(8):936-45.

[111] Nonaka S, Sato T. Cement modifier. Japan Patent JP2001089211, 3 Apr 2001.

[112] Raki L, Beaudoin J. Controlled release of chemical admixtures. US Patent US20070022916 A1, 1

Feb 2007.

[113] Morioka M, Ashida K, Handa M. Cement admixture and cement composition. Japan Patent

JP5262546A2, 12 Oct 1993.

[114] Mihara T, Morioka M. Cement admixture and cement composition. Japan Patent JP5330876A2, 14

Dec 1993.

[115] Tatematsu H, Sasaki T. Repair materials system for chloride-induced corrosion of reinforcing bars.

Cement and Concrete Composites. 2003;25(1):123-9.


[116] Tatematsu H, Nakamura T, Koshimizu H, Takatsu S. Chlorine ion scavenger. Japan Patent

JP04154648A2, 27 May 1992.

[117] Kang SP, Kim GD, Hong SY. The composition of repair mortar containing nitrite based

hydrotalcite(or hydrocalumite) and corrosion inhibitive repair method of reinforced concrete structure

using repair mortar described above. Korea Patent KR100515948B1, 12 Sep 2005.

[118] Feng N. Manufacturing method of concrete anticorrosion additive. China Patent CN1948207A, 18

Apr 2007.

[119] Tatematsu H, Nakamura T, Koshimizu H, Morishita T, Kotaki H. Cement additive for inhibiting

concrete deterioration. US Patent US5435846A, 25 Jul 1995.

[120] Kashima M. Cement mortar capable of preventing salt damage and rusting. Japan Patent

JP9142903A2, 3 Jun 1997.

[121] Fischer HR, Adan O, Lloris Cormano JM, Lopez Tendero MJ. Corrosion inhibition of reinforced

concrete. WIPO Patent WO 2011/065825A1, 3 Jun 2011.

Chapter 3

Inhibition Performance Evaluation of

Some Amino Acids against Steel

Corrosion in Simulated Concrete

Pore Solution

34 | Chapter 3

3.1 Introduction

As described in Chapter 2, chloride induced corrosion of reinforcing steel is one of the most

detrimental causes of premature failure for reinforced concrete structures. Commonly, corrosion

could be initiated when chloride concentration at the rebar surface exceeds a critical level,

typically as low as 0.4% by mass of cement for atmospherically exposed reinforced concrete

structures [1, 2]. The direct and indirect costs of reinforcement corrosion are substantial, as it

entails additional repair, rehabilitation, and monitoring activities to ensure the functionality and

aesthetics of concrete structures and components. Therefore it is crucial to design concrete to

resist environmental aggressiveness; and after corrosion initiation, to prevent or slow down

further deterioration. Among presently available corrosion preventive options (see Chapter 2),

corrosion inhibitors seem to be attractive owing to their low cost and the ease and flexibility of

application [3, 4].

Nowadays, a variety of inorganic salts and organic compounds with known inhibitive

properties have been produced and marketed for use in concrete [3, 5]. These corrosion inhibitors

could either be added to fresh concrete (or repair mortar) or applied to the surface of hardened

mortar or concrete [6]. Inorganic nitrite-based inhibitors have been proven to be the most

effective and widespread products available on the market. However, the risk of local corrosion

attack can be increased from using the nitrite-based inhibitors in the case of insufficient dosage.

In addition, critical concerns are the leaching out of nitrites over the long service life of a

concrete structure and subsequently the risk of their toxic effect posing on the surrounding

environment and health of human [7-9]. Hence, their application is restricted and is not allowed

especially for reinforced concrete structures permanently immersed in water owing to their

toxicity and solubility. Organic inhibitors are mainly based on mixtures of alkanolamines, amines

or amino-acids; some are based on emulsion of unsaturated fatty acid ester of an aliphatic

carboxylic acid and a saturated fatty acid [4, 9]. Experiments conducted in chloride-containing

alkaline solution showed that some organic inhibitors are effective only at high concentrations

(10%) due to the evaporation of the volatile constituent of the inhibitor from the alkaline solution

[10]. On the other hand, more and more concerns are also being raised regarding the

environmental impact of some common organic inhibitors. Increasing awareness of the health and

ecological risks has drawn much attention to amino acid-based inhibitors because they are

nontoxic, environmentally friendly, relatively cheap and easy to produce with higher purity. The

inhibition effect of amino acids against corrosion with respect to mild steel, stainless steel as well

as other metal substrates such as aluminum and copper has been investigated in acidic media [11-

16]. The results revealed high inhibition efficiency can be achieved when an appropriate type of

amino acid was used. Inhibition effect of amino acids along with some other organic inhibitors

was also studied for chloride induced corrosion on carbon steel in alkaline solutions [4, 17, 18]

Inhibition Performance Evaluation of Some Amino Acids | 35

and the results underlined the high inhibition efficiency by an 300 mV increase of pitting

potential when [Cl-]/[amino acid] molar ratio equals to 1.0 [4].

Although many organic and inorganic compounds have been tested as inhibitors against

chloride induced corrosion in reinforced mortar or concrete during the past decades [18, 19],

however, there are conflicting opinions about the reliability of these inhibitors for corrosion

protection in terms of long-term efficiency. A possible promising solution to overcome this

problem is the encapsulation/immobilization of desired inhibitors within the molecular structure

of a host compound. Then, the host compound can be added into concrete mixtures acting as a

kind of internal container or reservoir of the desired inhibitors. Owing to the unique fine tunable

molecular structure and high anion exchange capacity, modified hydrotalcites (MHTs) may have

the potential to be used to immobilize anionic inhibitors. For the envisaged use as an additive to

concrete against chloride attack, certain inhibitor anions with known inhibitive properties could

be intercalated in the structures of MHTs, which then can be slowly released in a controllable

way and provide a relatively long-term corrosion protection. The release of the intercalated

inhibitors is triggered by the arrival of chloride ions via anion exchange reactions. In this sense,

the MHTs can be envisaged as traps of chlorides. Therefore, the MHTs play a dural-role against

chloride-induced corrosion in reinforced concrete capturing chloride as a chloride scavenger and

providing of an intelligent release of corrosion inhibitors in parallel as an internal inhibitor

reservoir and protecting reinforcing steel from corrosion continuously.

In this chapter, four different types of sodium salts of amino acids (i.e., Glycine, 6-

aminocaproic acid, 11-aminoundecanoic acid, and p-aminobenzoic acid) were proposed as

potential candidates for the modification of hydrotalcite. The selection of the four

abovementioned amino acids is on the basis of the consideration of the possible effect of their

chemical structures on inhibition performance. It has been reported that the inhibiting behaviour

of the organic inhibitors is related to their chemical structures and type of the aggressive solutions

as well as the interaction with the surface of metal substrates [4, 20]. As shown in Table 3.1,

glycine, 6-aminocaproic acid and 11-aminoundecanoic acid represent the type of aliphatic amino

acids with short, medium and long carbon-chain length, respectively, while p-aminobenzoic acid

represents a group of aromatic amino acid. In addition, sodium nitrite was selected and used for a

comparison purpose due to its worldwide recognized inhibition performance in concrete. The

objective of this chapter is set to evaluate the inhibition performance of the five proposed

corrosion inhibitors in simulated concrete pore solution and select the inhibitors with good

performance as the modifiers for synthesis of MHTs, which is the main topic of Chapter 4.

36 | Chapter 3

Table 3.1 Chemical formulae and molecular structures of the four proposed amino acids.

Amino acid Chemical formula Molecular structure

Glycine NH2CH2COOH H2N CH2 C

O

OH

6-aminocaproic acid NH2C5H10COOH H2N

O

OH

11-aminoundecanoic acid NH2C10H20COOH H2N

O

OH

p-aminobenzoic acid NH2C6H4COOH C OHH2N

O

3.2 Experimental

3.2.1 Materials

NaOH, NaNO2, Glycine, and p-aminobenzoic acid were obtained from Sigma-Aldrich. 6-

aminocaproic acid and 11-aminoundecanoic acid were obtained from Acros organics. All

reagents are ACS grade and used as received without further purification. Boiled deionized water

was used for the preparation of aqueous solutions in order to minimize CO2 contamination. Steel

specimens used for corrosion inhibition evaluation were low-carbon steel (ASTM A36) coupons

with an exposed surface area of 100 mm2. Prior to immersion in the relevant testing solutions, the

steel specimens were ground successively using a series of silicon carbide emery papers of grades

320, 800, 1200, 2400 and 4000 in water. The steel was further cleaned with acetone under

ultrasonication and then dried.

3.2.2 Testing methods

All amino acids were pre-neutralized with a stoichiometric amount of NaOH to obtain their

corresponding sodium salts, which were denoted as Gly (Glycinate), pAB (p-aminobenzoate),

6ACA (6-aminocaproicate) and 11AUA (11-aminoundecanoate). The inhibition performance of

the four amino acids salts and sodium nitrite (NaNO2) was assessed based on open circuit

potential (OCP) and linear polarization resistance (LPR) measurements of the steel specimens in

a 0.1 M NaOH solution, which has been used in previous research to simulate the alkaline

concrete pore solution [18, 21-23]. The OCP and LPR measurements were carried out by using an

eight-channel Eco-Chemie Autolab-Potentostat PGSTAT20 under room temperature (RT) and

local laboratory environment. A conventional three-electrode electrochemical cell was employed


for measuring LPR with steel working electrode, a platinized titanium mesh as the counter

electrode and a saturated calomel electrode (SCE) as the reference electrode. A polarization

sweep from -20 to 20 mV relative to OCP was applied at a rate of 0.167 mV/s. Four groups of

testing solutions were prepared:

1) 0.1 M NaOH solution (as reference);

2) 0.1 M NaOH solution + 0.1 M inhibitors (as another reference);

3) 0.1 M NaOH solution + NaCl;

4) 0.1 M NaOH solution + 0.1 M inhibitors + NaCl.

It has to be noted that the NaCl was added stepwise as a source of Cl- contamination into the

testing solution when the steel electrode has reached the passivation state. The concentration of

NaCl in the testing solutions was progressively increased in several steps per every 24 h starting

from 0.05 M up to 0.4 M. Figure 3.1 shows the experimental setup and individual

electrochemical cell configuration. The test cells used are 200 ml plastic bottles in which 100 ml

testing solutions were filled. The three electrodes were stabilized through the lid of the bottle, on

which an additional hole was drilled for adding NaCl and to maintain the circulation of the air

during the test. NaCl crystals were first dissolved with part of the test solution with an extra

plastic cup. Then it was transferred back to the bulk solution in the test cell and mixed thoroughly

with a plastic pipette. There was no stirring application during the whole test period. The OCP of

the steel electrode in testing solution was recorded every 2 h, whereas LPR was recorded every 6

h up to 168 h. At least two or three specimens in same condition were employed to validate the

reproducibility of the results. All potentials are referred to the Saturated Calomel Electrode (SCE).

Figure 3.1 The experimental setup for OCP/LPR measurement: 8-channel Eco-Chemie Autolab-

Potentostat PGSTAT20 (left) and an electrochemical cell with three-electrode configuration (right).

38 | Chapter 3

3.3 Results and discussion

3.3.1 The effect of inhibitor addition on pH of the alkaline solution

From an electrochemical point of view, it is the potential of steel, Ecorr, relative to the pitting

potential, Epit, that determines whether corrosion will start or not. For the steel immersed in an

alkaline solution, the pitting potential depends on both environmental influences (e.g., chloride

content) and the properties of the metal. The open circuit potential of the passive steel, on the

other hand, only depends on the environmental conditions such as pH and oxygen content of the

solution [22, 24]. After a series of tests conducted in different alkaline media with various

chloride additions, Gouda [23] found that the cation of the chloride salts did not play a significant

role for breakdown of the passivity of the steel, and the pH being the dominating factor.

Furthermore, a linear relationship was found between pH and the logarithm of the critical

chloride concentration in the pH range of 11.75-13.50. A similar relationship was also obtained

by several other researchers in solutions containing mixed inhibitive and aggressive ions in near-

neutral media [25-27]. In this study, under the designated testing condition (i.e., aerated alkaline

stagnant solutions), apart from the inhibiting effect of the five proposed inhibitors, the effect of

possible carbonation by the airborne CO2 on pH of the alkaline solution was investigated at the

beginning and the end of the OCP/LPR measurements. Table 3.2 shows the pH values of the six

solutions that were measured before and after the 7 days (i.e. 168 h) testing period. It has to be

pointed out hereby that the measured pH of 0.1 M NaOH prepared in our laboratory is always

around 12.8 instead of the theoretical value of 13.0. As can be seen in Table 3.2, pH value at 0 h

(i.e. upon the preparation of the solution) for each testing solution does not show any evident

difference with that solution without chloride addition at 168 h, indicating that carbonation over

the entire testing period was negligible. In addition, the pH values of the 0.1 M NaOH solutions

admixed with the five inhibitors (each at 0.1 M) were found to drop by 0~ 0.1 unit compared to

that of the pure 0.1 M NaOH solution. On the other hand, when checked individually for each of

the testing solutions, the pH values were found to decrease slightly upon addition of NaCl. A

drop of 0.2~ 0.3 unit was measured at 168 h with the highest chloride addition level (0.4 M) for

all the test solutions likely due to a decrease in the activity coefficient of OH– when the solution

ionic strength increases [28]. In general, no to minor influence was found on pH of the alkaline

solution (i.e. 0.1 M NaOH) due to the addition of the five inhibitors (at 0.1 M).

Table 3.2 Measured pH of the solutions with/without chloride addition before and after the 7 days (168 h)

testing period (resolution of the pH meter is 0.001; all materials used in the testing solution have the same

concentration of 0.1 M).

Testing solution NaOH NaOH+NaNO2 NaOH+Gly NaOH+6ACA NaOH+11AUA NaOH+pAB

pH at 0 h 12.8 12.7 12.8 12.7 12.8 12.7

pH at 168 h (without Cl-) 12.7 12.6 12.7 12.7 12.7 12.7

pH at 168 h (0.4 M Cl-) 12.5 12.5 12.6 12.5 12.6 12.5


3.3.2 Open circuit potential (OCP) measurements

The OCP (also known as the corrosion potential, Ecorr) is the potential of the working electrode

relative to the reference electrode when no potential or current is applied to the cell. Although

anti-corrosion evaluation by means of OCP is not a standard test, prior studies have shown that

the OCP has many important applications and can provide qualitative information on the risk or

probability of corrosion [21, 29, 30]. In general, the evolution of OCP can be used to determine

the occurrence of corrosion initiation. In an alkaline chloride-containing solution, corrosion, in

particular localized corrosion, is initiated when the protective passive layer on the steel surface is

disrupted. Normally, corrosion of the steel in the simulated concrete pore solution is considered

not to be initiated if the OCP is equal to or more positive than -270 mV (SCE), a value which has

been widely adopted in the literature [29-31]. Consequently, the time point at which OCP drops

down to a value below -270 mV is considered as the point of corrosion initiation. The critical

chloride concentration (also known as chloride threshold, CT) at which the corrosion of the steel

electrode was initiated is then registered. Thus, the criterion of this series evaluation relates to the

immersion time registered for the particular inhibitor (i.e., the five proposed inhibitors) which can

cause the OCP to stay longest over a level of -270 mV with the highest NaCl concentration. It has

to be noted that due to the stepwise addition of chloride to the testing solutions in steps per every

24 h, the exact CT for corrosion initiation may not be derived directly from this series of tests.

However, the CT can be logically estimated as a value which is located in between two

consecutive chloride concentrations.

Figures 3.2-3.6 show the OCP evolution of the steel electrodes in solutions (0.1 M)

containing the five proposed inhibitors (i.e., Gly, 6ACA, 11AUA, pAB, and NaNO2) with and

without the addition of chlorides. For the chloride-containing solutions, the chloride was added

stepwise to increase the overall chloride concentration from 0.05 M to 0.4 M. The OCP was

recorded after 2 h immersion and was reported as the average value of at least two or three

specimens in the same condition (deviation is also depicted in the plot). The OCP evolution of the

steel in pure 0.1 M NaOH solution and in 0.1 M NaOH containing chloride (i.e., 0.1M NaOH +

Cl) was included as two references in all these figures. They were obtained respectively from the

same steel specimen. As can be seen from these figures, in cases of absence of chlorides, the

OCPs of all the specimens evolves similarly in time. Immediately after immersion, the OCPs of

all the freshly prepared steel electrodes reached a value in the range of -550 to -400 mV and

increased to a value around -270 mV after about 24-48 h immersion. Then the OCPs developed

positively to a relatively steady value around -200 mV. This clearly indicates that a stable passive

layer has formed on the steel surface in all the chloride-free alkaline solutions and is not

jeopardized by the presence of either nitrites or the amino acids based inhibitors.

For the steel specimens in chloride-containing solutions, various effects occurred.

(1) In pure NaOH (Figures 3.2-3.6) solution, and in NaOH solutions containing Gly (Figure

3.2) and 6ACA (Figure 3.3): the first addition of chloride at 0.05 M caused the OCPs of steel in

40 | Chapter 3

these solutions to drop at almost the same time by more than 100 mV compared to their respect

chloride-free solutions. Upon the addition of more Cl-, up to 0.4 M, the OCPs in both Gly and

6ACA containing solutions evolved similarly as in the case of a pure NaOH solution fluctuating

between -400 mV and -500 mV. Fluctuations of OCP in this range clearly suggest that sustained

active corrosion has been developed under these testing conditions. Compared to the pure NaOH,

Gly and 6ACA did not show any evident corrosion inhibiting effect on the steel at [Cl] = 0.05 M

at least under this applied testing regime. Thus, the CT for steel in 0.1 M NaOH solutions

containing 0.1 M Gly and 0.1 M 6ACA as well as in pure 0.1 M NaOH solutions was registered

as a value in between 0 and 0.05M.

(2) In NaOH solution containing pAB (Figure 3.4): the steel in pAB solution maintained its

passivation state with addition of chloride to a level of 0.1 M as indicated by more positive OCP

values than -270 mV. However, the addition of 0.2 M chloride caused a drop of the OCP to be -

272 mV at 102 h. This value dropped to be -366 mV at 120 h when more chloride was added

(concentration of 0.3 M) and further to values between -400 mV and -500 mV while reaching the

final chloride concentration of 0.4 M. This observation indicates that active corrosion has been

developed and the CT at which the steel could sustain its passivity in pAB solution should be

somewhere between 0.1 M and 0.2 M.

(3) In NaOH solution containing NaNO2 (Figure 3.5): The OCP of the steel in the NaNO2

solution stayed more positive than -270 mV until 120 h immersion, where chloride concentration

was lower than 0.3 M. However, with the presence of 0.3 M chloride, the OCP dropped rapidly

down to -340 mV at 126 h and started to fluctuate until 144 h, which indicated that pitting and

repassivation processes may have occurred in this period. With more chloride added (final

concentration of 0.4M), the OCP kept dropping and maintained a value at around -400 mV or

smaller suggesting the active corrosion has been developed. Thus, the CT of the steel in NaNO2

solution could be located somewhere between 0.2 M and 0.3 M.

(4) In NaOH solution containing 11AUA (Figure 3.6): In this case, 72 h was needed for the

steel electrode to reach the passivation state, which is longer than 24 h or 48 h for steel in pure

NaOH and the other four inhibitor solutions. The prolonged time of passivation in this case might

be coupled to the observation that part of the steel surface was covered by a film of 11AUA

suspension due to the poor solubility of 11AUA. The film formed on the steel surface may have

retarded the passivation progress. In addition, some flocculent precipitates was found during the

testing period, especially after 96 h immersion. This observation is possibly connected with a

reduction of solubility of 11AUA upon the addition of more chlorides. The OCP of the 11AUA

solution kept increasing slightly from -270 mV after the addition of 0.05 M chloride at 72 h

suggesting that the steel was still in passivated state. The addition 0.1 M chloride caused a drop of

the OCP from -260 mV at 96 h to -274 mV at 98 h and down to -350 mV at 120 h. The OCP kept

dropping when chloride concentration increased to 0.3 M and maintained a relatively stable

evolution afterwards, clearly indicating that active corrosion was present. Thus, the CT in

11AUA solution was registered to be a value between 0.05 M and 0.1 M.


Figure 3.2 OCP (Ecorr) evolution of steel in (0.1M Gly + 0.1M NaOH) solution and 0.1M NaOH

solution with/without stepwise increased chloride concentration.

Figure 3.3 OCP (Ecorr) evolution of steel in (0.1M 6ACA + 0.1M NaOH) solution and 0.1M NaOH


42 | Chapter 3

Figure 3.5 OCP (Ecorr) evolution of steel in (0.1M NaNO2 + 0.1M NaOH) solution and 0.1M NaOH


Figure 3.4 OCP (Ecorr) evolution of steel in (0.1M pAB + 0.1M NaOH) solution and 0.1M NaOH solution

with/without stepwise increased chloride concentration.


3.3.3 Linear polarization resistance (LPR) measurements

From a practical point of view, depassivation of steel means a development of active corrosion

that evolves with time. This situation cannot easily be detected by visual observation since the

appearance of colored oxides may take some time to appear or, on the opposite, its appearance

may not represent a steadily active corrosion [32]. The OCP measurement (the corrosion potential,

Ecorr) is usually employed to indicate the occurrence of depassivation. However, the indication

from OCP evolution might lead to misunderstanding in some cases either because the potential

shift may not be detected or because it may not really account for a significant activity. Such

uncertainties promote the need for use of the indications of the corrosion current as an additional

confirming parameter. The corrosion current (Icorr) can be calculated on the basis of polarization

resistance (Rp) obtained by LPR measurement. Icorr is inversely proportional to Rp according to

Stern-Geary equation (Eq. 3.1) [33]:

2.30( )

a ccorr

pa c

BIR

(3.1)

where βa and βc are the cathodic and anodic Tafel constants respectively. Although the value of

proportionality constant B may vary in a wide range from 8 mV to ∞ [34-36], a typical value of

26 mV is normally used for active corrosion and 52 mV for a passive state for steel bars in

concrete [37]. The corrosion current is often expressed as corrosion current density (icorr) when

Figure 3.6 OCP (Ecorr) evolution of steel in (0.1 M 11AUA + 0.1 M NaOH) solution and 0.1 M NaOH

solution with/without stepwise increased chloride concentration. Note: the steel in (0.1 M 11AUA + 0.1

M NaOH) solution needed more time (72 h) to passivate than in the other inhibitor solutions (24-48 h).

44 | Chapter 3

the surface area of the steel is concerned. As the criterion for active corrosion, an averaged

sustained corrosion current density higher than 0.1 µA/cm2 has been recommended for concrete

containing substantial moisture and oxygen [32, 38]. In addition, previous research based on

empirical observation reported that the measured corrosion current density in simulated concrete

pore solution and mortar was never below 0.1µA/cm2 when rust was observed [39]. However, it

is worth keeping in mind that the measured corrosion current is actually an average value over the

exposed steel surface area and the local current density inside a pit could be significantly higher

than that. Consequently, the measured value depends strongly on the number and the size of the

pits on the steel surface. To avoid such interpretation difficulties, an observable change in

corrosion current or current density over time rather than an absolute value might be a better

indication of the passive-active transition in the process of corrosion development. On the other

hand, a current density of 0.1 µA/cm2 was used in this research as an additional critical value for

determining the steel in a passive state, above which depassivation occurs.

Figures 3.7-3.11 present the corrosion current density (icorr) evolution of the steel in the

testing solutions containing 0.1 M inhibitors with and without chloride addition. Table 3.3

summarized corrosion initiation time and chloride threshold of the steel electrodes for all the test

solutions as derived from the OCP and LPR measurements. As can be seen from Figures 3.7-3.11,

the evolution of icorr of the steel in the five inhibitor solutions followed a similar trend but

differentiated at various levels of chloride addition. In particular, in the first 48 h (72 h for

11AUA) immersion (no chloride was added), icorr in all the solutions decreased progressively

reaching a value in the range of 0.01-0.07 µA/cm2 clearly indicating all the steel electrodes were

getting passivated during this period. With the first addition of chloride at 0.05 M, icorr increased

markedly for pure NaOH solution (Figures 3.7-3.11) as well as Gly (Figure 3.7) and 6ACA

(Figure 3.8) containing solutions by which a fluctuating increase followed when more chloride

was added. In the case of solution containing pAB (Figure 3.9), icorr started to slightly increase

from 0.08 (102 h) to 0.17 µA/cm2

(108 h) when 0.2 M chloride was added and maintained a

stable state afterwards. A significant increase of icorr occurred in pAB containing solution when

the chloride concentration was increased to 0.3 M and up to 0.4 M. A markedly increase of icorr

from 0.04 (120 h) to 0.35 µA/cm2 (126 h) was observed for the solution containing NaNO2

(Figure 10) with addition of 0.3 M chloride. The icorr increased up to be around 0.9 µA/cm2 when

0.4 M chloride was added. For solution containing 11AUA (Figure 11), although there is an

abrupt increase of icorr at 78 h upon the first addition of chloride at 0.05 M, the icorr increased

slightly from 0.08 (96 h) to 0.13 µA/cm2

(102 h) when 0.1 M chloride was added and went on to a

relatively stable state before a significant increase occurred at [Cl-] = 0.2 M and up to 0.3 M.

Overall, the icorr evolution of the steel based on LPR measurements is in good agreement with and

confirmed the results obtained from OCP measurements. On the other hand, it is interesting to

find that with more chloride added especially at the higher concentration levels, the steel in

solutions containing Gly, 6ACA, pAB and NaNO2 showed a higher icorr than that found for the

pure NaOH solution. This likely due to the insufficient inhibitor dosage in the solution relative to


the chloride content that has been added, which in turn may increase the risk of local corrosion

attack [5, 7]. On the contrary, the steel in 11AUA containing solution showed a lower icorr than

that in case of the pure NaOH solution with chloride concentration up to 0.3 M, which may be

attributed to the barrier effect resulting from the suspension film formed on the surface of the

steel due to the poor solubility of 11AUA.

Figure 3.7 Corrosion current density (icorr) evolution of steel in (0.1M Gly + 0.1M NaOH) solution and

0.1M NaOH solution with/without stepwise increased chloride concentration.

Figure 3.8 Corrosion current density (icorr) evolution of steel in (0.1M 6ACA + 0.1M NaOH) solution and


46 | Chapter 3

Figure 3.9 Corrosion current density (icorr) evolution of steel in (0.1M pAB + 0.1M NaOH) solution and


Figure 3.10 Corrosion current density (icorr) evolution of steel in (0.1M NaNO2 + 0.1M NaOH) solution

and 0.1M NaOH solution with/without stepwise increased chloride concentration.


Table 3.3 Time to corrosion initiation and chloride threshold of the steel in testing solutions derived from

OCP/LPR measurements (OCP was measured every 2 h and LPR was measured every 6 h from start of the

test until 168 h; all the materials used in the testing solution have a concentration of 0.1 M).

Testing Solution NaOH Gly+NaOH 6ACA+NaOH pAB+NaOH NaNO2+NaOH 11AUA+NaOH

Corrosion initiation time (h)

by OCP 52 52 56 102 126 98*

Corrosion initiation time (h)

by LPR 54 54 60 108 126 102

Chloride threshold (M)

by OCP and LPR 0-0.05 0-0.05 0-0.05 0.1-0.2 0.2-0.3 0.05-0.1

* Note: the steel in (0.1M 11AUA + 0.1M NaOH) solution needed more time (72 h) to passivate than in

the other inhibitor solutions (24-48 h).

3.4 Conclusions

Four different types of sodium salts of amino acids, which were denoted as Gly (Glycinate), pAB

(p-aminobenzoate), 6ACA (6-aminocaproicate) and 11AUA (11-aminoundecanoate) were

selected as potential candidates for the modification of hydrotalcite. Sodium nitrite was also

selected and used for the comparison purposes due to its well-recognized inhibition performance

in concrete. The anti-corrosion performance (in particular, chloride-induced corrosion) of the four

amino acids salts and sodium nitrite (NaNO2) was assessed based on the measurements of open

Figure 3.11 Corrosion current density (icorr) evolution of steel in (0.1M 11AUA + 0.1M NaOH) solution

and 0.1M NaOH solution with/without stepwise increased chloride concentration. Note: the steel in

(0.1M 11AUA + 0.1M NaOH) solution needed more time (72 h) to passivate than in the other inhibitor

solutions (24-48 h).

48 | Chapter 3

circuit potential (OCP) and linear polarization resistance (LPR) evolution of the steel specimens

in simulated concrete pore solution with the addition of chloride at concentrations varying from 0

to 0.4 M. The combined results from OCP and LPR measurements suggest that the anti-corrosion

performance of the five proposed inhibitors (0.1 M) in terms of the critical chloride concentration

(i.e., chloride threshold, CT) that they can sustain in simulated concrete pore solution is in the

order of:

NaNO2 (0.2-0.3M Cl

-) > pAB (0.1-0.2M Cl

-) > 11AUA (0.05-0.1M Cl

-) > Gly = 6ACA = NaOH

(0-0.05M Cl-)

Consequently, in view of the objective of this chapter, NaNO2, pAB and 11AUA were selected as

the most promising candidate modifiers for synthesis of MHT, which will be discussed in

Chapter 4.

References

[1] RILEM. Draft recommendation for repair strategies for concrete structures damaged by reinforcement

corrosion. Materials and Structures. 1994;27(171):415-36.

[2] Izquierdo D, Alonso C, Andrade C, Castellote M. Potentiostatic determination of chloride threshold

values for rebar depassivation: Experimental and statistical study. Electrochimica Acta.

2004;49(17):2731-9.


2001.

[4] Ormellese M, Lazzari L, Goidanich S, Fumagalli G, Brenna A. A study of organic substances as

inhibitors for chloride-induced corrosion in concrete. Corrosion Science. 2009;51(12):2959-68.

[5] Page C, Ngala V, Page M. Corrosion inhibitors in concrete repair systems. Magazine of Concrete

Research. 2000;52(1):25-38.

[6] Elsener B, Buchler M, Bohni H. Corrosion inhibitors for steel in concrete. In: Mietz J, Elsener B,

Polder R, editors. European Federation of Corrosion Publications (No 25), Papers from EUROCORR'

97. London, UK: IOM Communications Ltd; 1998. p. 54-69.


alkaline aqueous environments. Materials performance. 1979;18(11):45-8.



[9] Ormellese M, Berra M, Bolzoni F, Pastore T. Corrosion inhibitors for chlorides induced corrosion in

reinforced concrete structures. Cement and Concrete Research. 2006;36(3):536-47.

[10] Elsener B, Buchler M, Bohni H. Organic corrosion inhibitors for steel in concrete. In: Mietz J, Polder

R, Elsener B, editors. European Federation of Corrosion Publications (No 31), Papers from

EUROCORR' 99. London, UK: IOM Communications Ltd; 2000. p. 61-71.

[11] Yadav M, Kumar S, Gope L. Experimental and theoretical study on amino acid derivatives as eco-

friendly corrosion inhibitor on mild steel in hydrochloric acid solution. Journal of Adhesion Science

and Technology. 2014;28(11):1072-89.

[12] Muralidharan S, Babu BR, Iyer SV, Rengamani S. Influence of anions on the performance of isomers

of aminobenzoic acid on the corrosion inhibition and hydrogen permeation through mild steel in acidic

solutions. Journal of applied electrochemistry. 1996;26(3):291-6.


[13] El-Shafei A, Moussa M, El-Far A. Inhibitory effect of amino acids on Al pitting corrosion in 0.1 M

NaCl. Journal of applied electrochemistry. 1997;27(9):1075-8.

[14] Adamczyk L, Pietrusiak A, Bala H. Corrosion resistance of stainless steel covered by 4-aminobenzoic

acid films. Central European Journal of Chemistry. 2012;10(5):1657-68.

[15] Simonović AT, Petrović MB, Radovanović MB, Milić SM, Antonijević MM. Inhibition of copper

corrosion in acidic sulphate media by eco-friendly amino acid compound. Chemical Papers.

2014;68(3):362-71.

[16] Ashassi-Sorkhabi H, Majidi M, Seyyedi K. Investigation of inhibition effect of some amino acids

against steel corrosion in HCl solution. Applied surface science. 2004;225(1):176-85.

[17] Trabanelli G, Monticelli C, Grassi V, Frignani A. Electrochemical study on inhibitors of rebar

corrosion in carbonated concrete. Cement and Concrete Research. 2005;35(9):1804-13.

[18] Monticelli C, Frignani A, Trabanelli G. A study on corrosion inhibitors for concrete application.


[19] Paredes MA, Carvallo AA, Kessler R, Virmani YP, Sagüés AA. Corrosion Inhibitors in Concrete

(Second Interim Report No.FL/DOT/SMO/10-531). Florida Department of Transportation; 2010.

[20] Bazzi L, Kertit S, Hamdani M. Some organic compounds as inhibitors for the corrosion of aluminum

alloy 6063 in deaerated carbonate solution. Corrosion. 1995;51(11):811-7.



[22] Hausmann D. Steel corrosion in concrete--How does it occur? Materials protection. 1967;6(11):19-23.





[25] Matsuda S, Uhlig HH. Effect of pH, sulfates, and chlorides on behavior of sodium chromate and

nitrite as passivators for steel. Journal of the Electrochemical Society. 1964;111(2):156-61.

[26] Pryor M, Cohen M. The inhibition of the corrosion of iron by some anodic inhibitors. Journal of the

Electrochemical Society. 1953;100(5):203-15.

[27] Brasher DM, Reichenberg D, Mercer A. Comparative Study of Factors Influencing the Action of

Corrosion Inhibitors for Mild Steel in Neutral Solution: IV. Mechanism of Action of Mixed Inhibitive

and Aggressive Anions. British Corrosion Journal. 1968;3(3):144-50.

[28] Li L, Sagues A. Chloride corrosion threshold of reinforcing steel in alkaline solutions-Open-circuit

immersion tests. Corrosion. 2001;57(1):19-28.

[29] Cheng T-P, Lee J-T, Tsai W-T. Corrosion of reinforcements in artificial sea water and concentrated

sulfate solution. Cement and Concrete Research. 1990;20(2):243-52.

[30] Baweja D, Roper H, Sirivivatnanon V. Relationships between anodic polarisation and corrosion of

steel in concrete. Cement and Concrete Research. 1993;23(6):1418-30.

[31] Van Daveer JR. Techniques for evaluating reinforced concrete bridge decks. Journal of American

Concrete Institute. 1975;72(12):697-703.

[32] Alonso C, Andrade C, Castellote M, Castro P. Chloride threshold values to depassivate reinforcing

bars embedded in a standardized OPC mortar. Cement and Concrete Research. 2000;30(7):1047-55.

[33] Stern M, Geary AL. Electrochemical polarization I. A theoretical analysis of the shape of polarization

curves. Journal of the Electrochemical Society. 1957;104(1):56-63.

[34] Poursaee A. Potentiostatic transient technique, a simple approach to estimate the corrosion current

density and Stern–Geary constant of reinforcing steel in concrete. Cement and Concrete Research.

2010;40(9):1451-8.

50 | Chapter 3

[35] Song G. Theoretical analysis of the measurement of polarisation resistance in reinforced concrete.

Cement and Concrete Composites. 2000;22(6):407-15.

[36] Alonso C, Andrade C, Nóvoa X, Izquierdo M, Pérez M. Effect of protective oxide scales in the

macrogalvanic behaviour of concrete reinforcements. Corrosion Science. 1998;40(8):1379-89.

[37] Andrade C, Gonzalez J. Quantitative measurements of corrosion rate of reinforcing steels embedded

in concrete using polarization resistance measurements. Materials and Corrosion. 1978;29(8):515-9.

[38] Gonzalez J, Andrade C, Alonso C, Feliu S. Comparison of rates of general corrosion and maximum

pitting penetration on concrete embedded steel reinforcement. Cement and Concrete Research.

1995;25(2):257-64.

[39] Alonso C, Andrade C, Castellote M, Castro P. Reply to the discussion of the paper “Chloride

threshold values to depassivate reinforcing bars embedded in a standardized OPC mortar” by TU

Mohammed and H. Hamada. Cement and Concrete Research. 2001;31(5):839-40.

Chapter 4

Synthesis and Characterization of

Modified Hydrotalcites Using

Selected Inhibitors as Modifiers


Synthesis and Characterisation of Modified Hydrotalcites and Their Ion Exchange Characteristics

in Chloride-rich Simulated Concrete Pore Solution. Cement and Concrete Composites, 2014, 47:

87-93.

52 | Chapter 4

4.1 Introduction

As described in Chapter 2 and reviewed earlier [1-3], there are a number of techniques that have

been successfully applied for modification of hydrotalcites or hydrotalcite-like layered double

hydroxides (LDH). Among those, three main approaches are frequently adopted: (1) direct

synthesis by co-precipitation at constant pH [4-13]; (2) anion exchange of a precursor LDH [14-

20]; (3) reconstruction or rehydration of a calcined hydrotalcite in solution with desired

intercalating anions [21-25]. Figure 4.1 illustrates the typical experimental routes followed to

prepare the MHT using the three main synthesis methods. Based on the unique structural

recovery property or the so-called “memory effect” of calcined LDH, the reconstruction method

has been found to be more effective than the other two methods when large guest anions,

especially organic species with low affinity for LDH are needed to be intercalated into the

interlayer space [3, 21, 24, 26-29]. Another advantage of the reconstruction method is that the

incorporation of competing inorganic counter anions into the LDH could be avoided to a great

extent, although the inclusion of carbonates from atmospheric CO2 remains a problem. Therefore,

reconstruction is an important and economical method to synthesize MHTs with desired

inorganic or organic anions to fulfill specific application requirements. In this chapter, the

reconstruction method was employed to modify hydrotalcite with three different inhibitory agents,

i.e., NaNO2, pAB and 11AUA that were selected in Chapter 3. Figure 4.2 gives a schematic

diagram of the preparation of MHT using the reconstruction method.

Coprecipitation

Mg(NO3)2 , Al(NO3)3

+

Aqueous solution

containing guest anions

NaOH

pH≈10MgAl-Guest

Aging

°C60-70MgAl-Guest

(with improved cristallinity)

Anion exchange

MgAl-CO3Diluted HCl 0.1 M

pH≈ 5

MgAl-ClNaNO3 0.5 M

MgAl-NO3

Aqueous solution containing guest anions

MgAl-Guest

Reconstruction (memory effect)

MgAl-CO3

°C500

3-5hMgAlOx + CO2 + H2O

MgAlOx

Aqueous solution containing guest anions

MgAl-Guest

pH≈ 10-11

+ NaNO3

+ CO2 + H2O + NaCl

Figure 4.1 Typical experimental routes for the synthesis of MHT (atomic ratio of Al/(Mg+Al) is in the

range of 0.20-0.33).

Synthesis and Characterization of MHTs | 53

4.2 Experimental

4.2.1 Materials

NaOH, NaNO2, and p-aminobenzoic acid were obtained from Sigma-Aldrich. 11-

aminoundecanoic acid was obtained from Acros Organics. All reagents are ACS grade and used

as received without further purification. The hydrotalcites used in this study are two types of

commercially available products PURAL® MG 63 HT (the atomic ratio of Mg/Al is 2.2 and

denoted as Mg(2)Al-CO3) and PURAL® MG 70 HT (the atomic ratio of Mg/Al is 2.7 and

denoted as Mg(3)Al-CO3) provided by Sasol Germany GmbH. Boiled deionized water was used

throughout the experiment to minimize the possible contamination by airborne CO2.

4.2.2 Synthesis

The synthesis of MHT was carried out as follows. About 50 g of powder of either Mg(2)Al-CO3

or Mg(3)Al-CO3 was heated for 3 h at 500°C in a muffle furnace and then was cooled down in N2

flow to room temperature (RT). The obtained Mg-Al oxide precursor was transferred to a 2 l

glass reactor under the protection of N2. 500 ml warm water (50 °C) was subsequently added

under vigorous stirring for 30 min and then the mixture was cooled down again to RT.

Afterwards, 500 ml solution of 0.6 mol/l sodium salts of amino acids or 500 ml of 0.6 mol/l

sodium nitrite was added. The two types of amino acids used here are p-aminobenzoic acid and

11-aminoundecanoic acid. Both compounds were pre-neutralized with a stoichiometric amount of

NaOH to obtain their corresponding sodium salts, which were denoted as pAB (p-aminobenzoate)

and 11AUA (11-aminoundecanoate). It has to stress that in order to promote the intercalation, an

excess of guest anions (i.e., -pAB, -11AUA and -NO2) has to be included into the reaction

solution. Consequently, a molar ratio of 10/1 between guest anions and the CO32-

that is

originally present in hydrotalcites was adopted in the synthesis. In addition, 6 M NaOH or HNO3

was employed to adjust the pH of above mentioned solution to be around 10. To allow complete

Figure 4.2 Schematic diagram of the reconstruction process of calcined hydrotalcites.

54 | Chapter 4

reconstruction of the host layer structure, the resulting suspension was stirred continuously for 20

additional hours before subjecting to filtration. The modified products were collected and then

dried for 16-18 h at 105 °C under vacuum.

4.2.3 Characterization

X-ray powder diffraction (XRD) was performed on a Bruker D5005 diffractometer equipped with

Huber incident-beam monochromator and Braun PSD detector using Cu Kα radiation in the 2θ

region between 5 and 90°. The XRD patterns were interpreted mainly with respect to the position

of the basal peak (003), which depends on the distance between two adjacent metal hydroxide

layers in the MHT’s crystal lattice. Thermal analyses on powder samples were carried out using a

NETZSCH STA 449 F3 Jupiter®

simultaneous thermal analyzer TG/DSC under flowing Argon

(50 ml/min) at a heating rate of 10 K/min from 40 to 1100°C. FT-IR spectra were recorded using

a Perkin-Elmer Spectrum 100 Series spectrometer equipped with universal Attenuated Total

Reflection (ATR) unit over the wave-number range of 4000 to 600 cm-1

. The carbon content of

the two carbonate hydrotalcites (i.e., Mg(2)Al-CO3 and Mg(3)Al-CO3) was analyzed by a

Thermo Flash EA elemental analyzer. A Shimadzu TOC-VCPH total organic carbon analyzer and

a spectrophotometer (Spectroquant® Nova 60, Merck, Germany) were employed respectively to

analyze the content of intercalated organic anions and nitrite after dissolution of a known amount

of modified compound in diluted HCl solution. All measurements were conducted in duplicate.


4.3.1 X-ray diffraction analysis

XRD patterns of the two carbonate hydrotalcites (i.e., Mg(2)Al-CO3 and Mg(3)Al-CO3), their

calcinated form (i.e., Mg(2)Al-500C and Mg(3)Al-500C) and the modification derivatives are

shown in Figures 4.3 & 4.4. Before calcination, both Mg(2)Al-CO3 and Mg(3)Al-CO3 show

sharp and intense basal reflections at lower 2θ angles and less intense and asymmetric reflections

at higher angles, which represent the characteristic layered structure of a hydrotalcite-like

compound with high crystallinity [2]. After being calcined at 500°C for 3 h, all of the

characteristic reflections as observable before calcination disappeared indicating that the layered

hydrotalcite-like structure was destroyed and a mixed magnesium and aluminum oxide was

formed (i.e., Mg(2)Al-500C and Mg(3)Al-500C). However, complete reconstruction of the

layered structure is believed to take place after the modification process since similar but slightly

broader XRD patterns from the modification derivatives relative to the original carbonate

hydrotalcites have been observed. This suggests that the modification derivatives possess the

same basic structure as the carbonate hydrotalcites although the crystallinity is not as high as the

latter.


The XRD peaks of hydrotalcite-like compounds are generally indexed on the basis of a

hexagonal unit cell, the basal spacing of which can be calculated from the (003) position using

Bragg’s law. The gallery height of hydrotalcite-like compounds can be obtained by subtracting

hydroxide layer thickness of approximately 4.8Å from the basal spacing [30]. The gallery height

of MHT mostly depends on the size and orientation of the intercalated guest anion [3, 26] and

may also be affected by adsorbed water molecules especially in the surfactant modified

hydrotalcites [21, 30]. Table 4.1 gives the observed basal spacing d003 and the corresponding

gallery height for all the MHTs that have been synthesized. As expected, the position of the basal

reflections of all the modified products shifted to a higher d003 value caused by an expansion of

the gallery height due to the molecular size of the intercalates. The expanded gallery height

indicates that the carbonate anions in the unmodified pristine hydrotalcites were successfully

substituted by the guest anions (i.e., -NO2, -pAB, and -11AUA) following the calcination-

rehydration process as sketched in Figure 4.2. In addition, the observed basal spacing of the two

hydrotalcites and MHTs intercalated with the same guest anions was found to be nearly similar

irrespective of the two different Mg/Al atomic ratios (i.e., Mg/Al=2.2 and 2.7). On the other hand,

it is interesting to note that previous studies have reported that with lowering molar ratio of

Mg/Al, the basal spacing of hydrotalcites could slightly increase or decrease when intercalation

with different types of anions was performed [31-33]. The basal spacing values (Table 4.1)

Figure 4.3 XRD patterns of Mg(2)Al-CO3, its calcined form Mg(2)Al-500C and the modification

derivatives Mg(2)Al-NO2, Mg(2)Al-pAB and Mg(2)Al-11AUA.

56 | Chapter 4

obtained in this study for carbonate hydrotalcites and -pAB modified hydrotalcites are in

agreement with those previous research [28, 34], in which carbonate and -pAB anions were

described to orientate perpendicularly between the host mixed-metal hydroxide layer together

with interacting water molecules. For -NO2 modified hydrotalcites, the observed basal spacing of

7.8 Å is in accordance with the data reported in the literature [32, 35], in which -NO2 was

suggested to be lying flat and monolayer in the interlayer region with its plane parallel to the

hydroxide layer.

The XRD patterns of -11AUA modified hydrotalcites as shown in Figures 4.3 & 4.4, are

rather different with respect to those modified by -NO2 and -pAB which may suggest that the -

11AUA anion was intercalated in two particular orientations (e.g. in both perpendicular and tilted

position) within the interlayer gallery. These two possible configuration of -11AUA within the

interlayer gallery would consequently lead to the co-existence of two different phases of MHT-

11AUA. In this sense, the basal spacing d003 associated with these two phases was observed to be

16.4 Å and 10.8 Å respectively. Similar phenomena were often found and have been reported in

MHTs intercalated with some amino acids and other big organic molecules [21, 26-28, 36-38].

Furthermore, a relatively broader reflection peak appeared in the same angle position (2θ =11.6°)

in the case of the six synthesized MHTs with respect to peak (003) of the original unmodified

Figure 4.4 XRD patterns of Mg(3)Al-CO3, its calcined form Mg(3)Al-500C and the modification



hydrotalcites, suggesting the possibility of overlapping of both peaks in the diffractogram of

MHTs. This observation may also indicate that a certain amount of carbonate is still present in the

interlayer space of MHTs, likely resulting from the unavoidable CO2 contamination during the

rehydration process of the calcined hydrotalcites even under the protection of N2 flow [7, 14, 24].

Table 4.1 The basal spacing d003 and the corresponding 2θ position and gallery height obtained for

hydrotalcites and the modified derivatives.

MHT(Mg/Al=2&3) 2θ (deg) d003(Ǻ) Gallery height(Ǻ)

Mg(2)Al-CO3 11.6 7.6 2.8

Mg(2)Al-NO2 11.3 7.8 3.0

Mg(2)Al-pAB 5.9 15.0 10.2

Mg(2)Al-11AUA 5.4; 8.1 16.4; 10.8 11.6; 6.0

Mg(3)Al-CO3 11.5 7.7 2.9

Mg(3)Al-NO2 11.3 7.8 3.0

Mg(3)Al-pAB 6.0 15.0 10.2

Mg(3)Al-11AUA 5.4; 8.3 16.3; 10.7 11.5; 5.9

4.3.2 Infrared analysis

FT-IR Spectra of the two starting carbonate hydrotalcites and the six modified products are

shown in Figures 4.5 & 4.6. In general, the two starting (parent) hydrotalcites and their calcined

products (i.e., Mg(2)Al-500C and Mg(3)Al-500C) and the six MHTs intercalated with the same

guest anions displayed very similar FT-IR spectra despite of the two different Mg/Al atomic

ratios. Similar results were also found with respect to their XRD patterns. For the starting

materials (i.e., carbonate hydrotalcites), a broad band between 3500 and 3200 cm-1

is observed

representing the stretching vibrations of the hydrogen-bonded hydroxyl group of both hydroxide

layers and interlayer water molecules [21]. A weak shoulder at around 3500-3700 cm-1

counts for

the O-H stretching modes of weak hydrogen bonds from the physically adsorbed water. A weak

and broad shoulder at around 3000-3200 cm-1

is ascribed to the interaction between the interlayer

water and intercalated CO32-

anions, which involves mostly hydrogen bonding [2, 39, 40]. The

bending vibration of the interlayer H2O is also found in the broad bands around 1620 cm-1

. This

peak shifts to a higher wavenumber (1640 cm-1

) with a slight increase in intensity with increasing

Mg/Al ratio. The appearance of a highly intense band at around 1365 cm-1

without any distinct

shoulder is related to the asymmetric ν3 stretching mode of CO32-

and indicates that carbonate

ions occupy a highly symmetric site between the hydroxide layers [41, 42]. Moreover, several

58 | Chapter 4

characteristic bands appear below 1000 cm-1

can be attributed to metal-oxygen bond stretching

and various lattice vibrations associated with metal hydroxide layers.

Comparing the FT-IR spectra for the calcined products with those of the pristine

hydrotalcites (Figures 4.5 & 4.6), it can be concluded, that there was not a complete loss of

interlayer carbonate anions and bound water. This is indicated by the presence of ν3 stretching

mode band of CO32-

at around 1365 cm-1

and the O-H stretching band at 1740 cm-1

for interlayer

water molecules, although the intensity of both bands is much reduced. However, the XRD

patterns (Figures 4.3 & 4.4) demonstrate that calcination at 500°C destroys the crystal structure of

hydrotalcite. Similar observation has also been found in previous research [21, 32].

The FT-IR spectra of MHT include two sets of bands: one corresponds to the intercalated

anionic species and the other corresponds to the host hydrotalcites (Figures 4.5 & 4.6). The band

appearing below 1000 cm-1

comes from the various lattice vibration of the host metal hydroxide

layer. This observation supports the XRD results that the MHTs have a similar lattice structure of

the pristine unmodified hydrotalcites. The relatively weak and broad band at around 3500-3100

cm-1

comes mainly from O-H groups of the hydroxide layers and also from the N-H stretching in

the cases of -pAB and -11AUA modified products. The weak shoulder band observed at around

3500-3700 cm-1

for -pAB and -11AUA modified derivatives is assigned to the weak hydrogen

Figure 4.5 FTIR Spectra of Mg(2)Al-CO3, its calcined form Mg(2)Al-500C and the modification



bands from the physically adsorbed water. The occurrence of characteristic ν3 stretching bands

for CO32-

at around 1365 cm-1

in the spectra of all the six synthesized MHTs, suggests the co-

existence of CO32-

in the interlayer space. The less intensity or splitting of the ν3 band into a pair

(in the case of -11AUA modified products) is probably caused by the reduced symmetry of the

carbonate ions in the interlayer region [2, 43]. As mentioned in section 4.3.1, carbonate ions

retained due to the unavoidable CO2 contamination during the rehydration process of the calcined

hydrotalcites and/or to CO32-

simply remaining after calcination.

For nitrite modified hydrotalcites, the appearance of the characteristic band for -NO2 at

around 1241 cm-1

(in the case of Mg(2)Al-NO2) and at around 1238 cm-1

(in the case of Mg(3)Al-

NO2) reveals that the -NO2 has been successfully intercalated into the interlayer space through the

reconstruction process [32, 35, 44]. For products modified by -pAB (i.e., NH2C6H4COO-),

characteristic peaks present in the spectrum at around 1532 cm-1

and 1433 cm-1

correspond

respectively to the asymmetric and symmetric stretching vibrations associated with -COO-. In

addition, the other characteristic peaks are observed at around 1601 cm-1

for the N-H bending

mode, at around 1586 cm-1

for the aromatic C=C stretching mode and at around 1297 cm

-1, 1174

cm-1

and 1140 cm-1

for C-H bending vibrations of the benzene ring. The presence of these peaks

is comparable with previously reported data for -pAB modified hydrotalcite [14, 37, 45] and

indicates that -pAB has been successfully intercalated into the interlayer space of the modified

Figure 4.6 FTIR spectra of Mg(3)Al-CO3, its calcined form Mg(3)Al-500C and the modification


60 | Chapter 4

hydrotalcites. For -11AUA (i.e., NH2C10H20COO-) modified products, the adsorption bands at

around 1555 cm-1

and 1395 cm-1

are assigned to -COO- stretching vibrations. The bands

presenting at around 2926 cm-1

and 2852 cm-1

correspond respectively to the asymmetric and

symmetric stretching of alkyl group -CH2 in the carbon chain of the 11AUA molecules [46, 47],

whereas the band at around 1469 cm-1

is assigned to the bending mode of -CH2 [48]. On the other

hand, although the TG analysis (see discussion below) shows the presence of interlayer water in

MHTs, the FT-IR spectra of these MHTs do not show any observable bands at around 3000-3200

cm-1

, indicating the interaction between the interlayer water and intercalated CO32-

. Therefore, the

water molecules present in the MHT may not actively interact with CO32-

as in the parent

carbonate

hydrotalcites. In this regard, they might preferably fill the voids between the

intercalated guest anions (i.e., -NO2, -pAB and -11AUA) and the metal hydroxide layers.

Nevertheless, the comparison of the FT-IR spectra of the MHTs with the parent carbonate

hydrotalcites once again demonstrates the recovery of crystal structure and successful

intercalation of the three investigated guest anions.

4.3.3 Thermal analysis

Thermogravimetry (TG) and differential scanning calorimetry (DSC) were employed to

investigate thermal behaviors of the two carbonate hydrotalcites and the six modified derivatives.

The measured TG/DSC curves are shown in Figures 4.7 & 4.8 and the relevant results are

summarized in Table 4.2. For the two parent carbonate hydrotalcites, the thermal decomposition

proceeded in two main stages, which resemble the behavior of the same type of hydrotalcite

reported in the literature [49, 50]. In the first stage spanning from 40 to 300°C, the weight loss of

15.6% for Mg(2)Al-CO3 and 15.9% for Mg(3)Al-CO3 mainly corresponds to the elimination of

interlayer water molecules. The water adsorbed on the external surfaces of hydrotalcite can also

undergo desorption at this stage as evidenced by the weak endothermic peak at around 200°C

presented in the associated DSC thermograms. The observation of the desorption of adsorbed

surface water is comparable to the previously reported results [51]. A partial dehydroxylation (i.e.,

loss of structural water) of the metal hydroxide layers could also take place in this stage as

reported elsewhere [49]. In addition, the interlayer carbonates could also decompose in this stage

since it has been reported that the release of interlayer carbonates can start as early as 250°C [52].

In the second stage ranging from 300 to 1100°C, the weight loss of 28.4% for Mg(2)Al-CO3 and

28.8% for Mg(3)Al-CO3 is ascribed to a concomitant dehydroxylation of the metal hydroxide

layers and a decomposition of interlayer carbonates. It is often believed that there are two

separate decomposition reactions proceeding in the second weight loss process, as evidenced by

two endothermic peaks occurring in the associated DSC thermograms at this stage. The first of

the two peaks is attributed to the partial loss of hydroxyl group from the hydroxide layer and the

second one to the complete loss of hydroxyl group and the interlayer carbonates including a

collapse of the layered structures of hydrotalcite. It needs to point out that the complete loss of


interlayer carbonates and bound water could probably last up to 1100°C since the tenacious

retaining of carbonates and bound water after calcination of hydrotalcite for 3 h at 500°C has

been indicated by the FTIR spectra of calcined products (see Mg(2)Al-500C and Mg(3)Al-500C

in Figures 4.5 & 4.6).

As can be seen in Figures 4.7 & 4.8 and in Table 4.2, the modified derivatives show a similar

trend of mass loss as that of pristine hydrotalcite including two major stages. The corresponding

DSC thermograms exhibit two main features associated with the two major mass losses. Besides

the possibility of a partial dehydroxylation (i.e., loss of structural water) of the metal hydroxide

layers, the first weight loss is mainly attributed to the elimination of physisorbed and interlayer

water molecules. In addition, it is observed from the associated DSC thermograms (Figures 4.7 &

4.8) that the first decomposition stage for the modified derivatives shifts to lower temperatures

compared to the unmodified pristine hydrotalcite. As discussed in section 4.3.2, for pristine

hydrotalcite, the interlayer water molecules remain in close interaction with interlayer carbonate

anions and hydroxide layers through hydrogen bonding in a relatively constrained environment.

After modification, with the intercalation of new guest anions, such interactions turned to much

weaker or even disappeared. The second weight loss can be assigned to a concomitant

dehydroxylation (i.e., structural water) and decomposition of intercalated guest anions (i.e., -NO2.

-pAB and -11AUA). Furthermore, as can be seen from the associated DSC thermograms, the

Figure 4.7 Simultaneous TG/DSC analysis of hydrotalcite and its modified derivatives: a) Mg(2)Al-CO3;

b) Mg(2)Al-NO2; c) Mg(2)Al-pAB; d) Mg(2)Al-11AUA.

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Mass

ch

an

ge

(%)

Temperature/˚C

TG/DSC-Mg(2)Al-11AUA

Mass/% DSC/(µV/mg)

exo

d

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

40

45

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Heat

flow

(µ

V/m

g)

Mass

Ch

an

ge

(%)

Temperature/˚C

TG/DSC-Mg(2)Al-pAB

Mass/% DSC/(uV/mg) exo

c

-1.4

-1.2

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Ma

ss c

ha

ng

e (

%)

Temperature/˚C

TG/DSC-Mg(2)Al-CO3

Mass/% DSC/(µV/mg)exo

a

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Ma

ss c

ha

ng

e (%

)

Temperature/˚C

TG/DSC-Mg(2)Al-NO2

Mass/% DSC/(µV/mg)

exo

b

62 | Chapter 4

nature of the interlayer guest anion influences the thermal behavior markedly in the second stage

of the decomposition process, especially when big organic molecules (i.e.-11AUA) were

intercalated. Compared to -NO2 and -pAB modified products, the thermal decomposition of -

11AUA modified products begins at a lower temperature (242°C for Mg(2)Al-11AUA and 227°C

for Mg(3)Al-11AUA) in this stage and takes place in a more complex process, which is featured

by the presence of several endothermic peaks in the associated DSC thermograms (Figure 4.7d &

Figure 4.8d). The complex decomposition process in this stage could be likely owing to the

stepwise decomposition of the long hydrocarbon chains in -11AUA ions and the collapse of

layered structure of the host Mg-Al hydroxides. Similar to the pristine hydrotalcite, a further

condensation of hydroxyls included a collapse of the layered structures of hydrotalcite. A

complete loss of trace impurities such as carbonates and bound water may also occur in this stage.

On the other hand, the darkening of the -pAB and -11AUA modified products after the

thermogravimetric treatment further indicates the complete decomposition of -pAB and -11AUA

and the successful intercalation into the layer spaces of modified hydrotalcites.

Figure 4.8 Simultaneous TG/DSC analysis of hydrotalcite and its modified derivatives: a) Mg(3)Al-CO3; b)

Mg(3)Al-NO2; c) Mg(3)Al-pAB; d) Mg(3)Al-11AUA.

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Ma

ss c

ha

ng

e (%

)

Temperature/˚C

TG/DSC-Mg(3)Al-CO3


a

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Mass

ch

an

ge

(%)

Temperature/˚C

TG/DSC-Mg(3)Al-NO2

Mass/% DSC/(µV/mg) exob

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

40

45

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Mass

ch

an

ge

(%)

Temperature/˚C

TG/DSC-Mg(3)Al-pAB

Mass/% DSC/(uV/mg)exo

c

-1.4

-1.2

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

0 100 200 300 400 500 600 700 800 900 1000 1100

Hea

t fl

ow

(µ

V/m

g)

Ma

ss c

ha

ng

e (%

)

Temperature/˚C

TG/DSC-Mg(3)Al-11AUA


d


Table 4.2 The mass loss and the temperature range associated with the two major decomposition stages

obtained by simultaneous TG/DSC analyses for hydrotalcites and the modified derivatives.

MHT(Mg/Al=2&3)

1st decomposition stage 2nd decomposition stage

T (°C) Weight loss (%) T (°C) Weight loss (%)

Mg(2)Al-CO3 40-300 15.6 300-1100 28.4

Mg(2)Al-NO2 40-290 10.4 290-1100 30.3

Mg(2)Al-pAB 40-261 6.3 261-1100 48.9

Mg(2)Al-11AUA 40-216 5.3 216-1100 53.1

Mg(3)Al-CO3 40-300 15.9 300-1100 28.8

Mg(3)Al-NO2 40-289 11.0 289-1100 30.2

Mg(3)Al-pAB 40-255 10.9 255-1100 44.9

Mg(3)Al-11AUA 40-223 8.0 223-1100 48.6

4.3.4 Intercalation efficiency under the calcination-rehydration condition

The carbon content of the two starting carbonate hydrotalcites (i.e., Mg(2)Al-CO3 and Mg(3)Al-

CO3) was analyzed by a Thermo Flash EA elemental analyzer. The content of intercalated -pAB

and -11AUA anions was determined by total organic carbon (TOC) analysis; the -NO2 content

was analyzed photometrically by a Spectroquant®

NOVA60 spectrophotometer using a specific

nitrite test kit. The intercalation efficiency of the three inhibitive anions under the calcination-

rehydration condition was assessed by the intercalation degree (Id), which was calculated by the

following equations (Eq. 4.1 & 4.2):

100%i

mId

m (4.1)

23

2 23 3

23

2

2

HT co

inhco co

i inh

HT co

m mM

M mm M

m M

(4.2)

where m (% by mass of MHT) is the measured content of the anions that have been intercalated

into MHT via calcination-rehydration; mi (% by mass of MHT) is the maximum anion content

that can be intercalated to neutralize the positive charge of the host metal hydroxide layer, which

means all CO32-

ions in the starting carbonate hydrotalcites have been ideally substituted by the

inhibitive anions after calcination-rehydration (Figure 4.2). HTm (g) and 23co

m (% by mass of HT)

64 | Chapter 4

are the mass and carbonate content of the starting carbonate hydrotalcite. 23co

M (g/mol) and

inhM (g/mol) are the molecular weight of the CO32-

ions (i.e. 60 g/mol) and intercalated inhibitive

anions (i.e., 46 g/mol for -NO2, 136 g/mol for -pAB and 200 g/mol for -11AUA). Table 4.3 gives

carbonate content of the two starting hydrotalcites, measured anion content of the six synthesized

MHTs and the calculated intercalation degree. As can be seen from Table 4.3, the differences of

the intercalation degree between MHTs intercalated with the same anions are minor in spite of the

two different Mg/Al atomic ratios (i.e., Mg/Al =2 & 3). The largest difference in intercalation

degree is only 5.3% when comparing Mg(2)Al-11AUA with respect to Mg(3)Al-11AUA. In

general, the six MHTs presented a similar intercalation degree which is around 60%. In other

words, there was about 40% carbonates still presented in each of the six MHTs after the

calcination-hydration synthesis process. This high retaining degree of carbonates in MHT could

be ascribed to two possible reasons due to the high affinity of the carbonates to hydrotalcites and

the modified derivatives: (1) the unavoidable attraction of airborne CO2 during the rehydration

(i.e., intercalation) of the calcined hydrotalcites and post synthesis processes (see discussion in

XRD analysis); (2) the incomplete decomposition of interlayer carbonate anions in the

parent/starting hydrotalcites during the calcination process (i.e., heating for 3 h at 500°C), which

has been demonstrated by the FT-IR spectra of the calcined hydrotalcites as presented in section

4.3.2.

Table 4.3 The intercalated anion content of modified hydrotalcites and the corresponding intercalation

degree obtained from the calcination-rehydration synthesis procedure (carbonate hydrotalcites were

included as the starting materials; “-” means no applicable data).

MHT(Mg/Al=2&3) Anion content (by mass of

MHT %, in practice)

Maximum anion content

(by mass of MHT %,

assuming all carbonates in

starting hydrotalcites are

substituted)

Intercalation degree (%)

Mg(2)Al-CO3 12.4 - -

Mg(2)Al-NO2 11.1 19.0 58.4

Mg(2)Al-pAB 32.2 56.2 57.3

Mg(2)Al-11AUA 48.5 82.7 58.6

Mg(3)Al-CO3 11.4 - -

Mg(3)Al-NO2 10.5 17.5 60.0

Mg(3)Al-pAB 30.1 51.7 58.2

Mg(3)Al-11AUA 40.5 76.0 53.3


4.4 Conclusions

Six MHTs have been synthesized through the modification of two commercially available

carbonate hydrotalcites PURAL® MG 63 HT (the atomic ratio of Mg/Al is 2.2) and PURAL

® MG

70 HT (the atomic ratio of Mg/Al is 2.7) by three different inhibitor compounds, i.e., NaNO2,

pAB and 11AUA using the reconstruction method. The six synthesized MHTs were characterized

by means of XRD, FTIR and simultaneous TG/DSC. XRD analysis of both pristine and modified

hydrotalcites showed that the interlayer space (gallery height) was extended by the modification

and the modification derivatives possess the same basic crystal structure as the pristine

hydrotalcite, although the crystallinity is not as high as the pristine one. The unavoidable

presence of carbonates and some bound water in the interlayer space of MHTs was suggested by

XRD and further confirmed by FT-IR. The characteristic infrared bands for -NO2, -pAB and -

11AUA have been detected respectively in FT-IR spectra of their modification derivatives. The

simultaneous TG/DSC analyses showed that MHTs exhibited a similar trend of mass loss as that

of pristine hydrotalcite including two major decomposition stages, but the first decomposition

stage (the release of physisorbed and interlayer water) for the MHTs shifted to lower

temperatures compared to the unmodified pristine hydrotalcites. Compared to NaNO2 and pAB

modified hydrotalcites, the 11AUA modified hydrotalcite showed lowest thermal stability and a

more complex decomposition process because its decomposition began at a lower temperature in

the second stage and was featured by the presence of several endothermic peaks in the DSC

thermograms. The content of intercalated -pAB and -11AUA anions was further determined by

total organic carbon (TOC) analysis and the -NO2 content was analyzed photometrically by a

Spectroquant® NOVA60 spectrophotometer with the specific nitrite test kit. The intercalation

efficiency analysis revealed that there was about 40% carbonates still presented in each of the six

MHTs after the calcination-hydration synthesis process. Two possible reasons which could be

responsible for the high retaining degree of carbonates in MHT are: (1) the unavoidable attraction

of airborne CO2 during the rehydration (i.e., intercalation) of the calcined hydrotalcites and post

synthesis processes; (2) the incomplete decomposition of interlayer carbonate anions in the

parent/starting hydrotalcites during the calcination process (i.e., heating for 3 h at 500°C).

Nevertheless, the combined information from XRD, FT-IR, TG/DSC and elemental analysis

demonstrated that the three inhibitive anions, i.e., -NO2, -pAB and -11AUA have been

successfully intercalated into the interlayer space of the corresponding MHT although

carbonation contamination was found to be unavoidable.

References

[1] He J, Wei M, Li B, Kang Y, Evans DG, Duan X. Preparation of layered double hydroxides. Layered

double hydroxides: Springer; 2006. p. 89-119.

[2] Cavani F, Trifirò F, Vaccari A. Hydrotalcite-type anionic clays: Preparation, properties and

applications. Catalysis today. 1991;11(2):173-301.

66 | Chapter 4

[3] Carlino S. The intercalation of carboxylic acids into layered double hydroxides: a critical evaluation

and review of the different methods. Solid State Ionics. 1997;98(1):73-84.

[4] Aisawa S, Takahashi S, Ogasawara W, Umetsu Y, Narita E. Direct intercalation of amino acids into

layered double hydroxides by coprecipitation. Journal of Solid State Chemistry. 2001;162(1):52-62.

[5] Miyata S. The Syntheses of Hydrotalcite-Like Compounds and Their Structures and Physico-Chemical

Properties I: The Systems Mg2+

-Al3+

-NO3-

, Mg2+

-Al3+

-Cl-, Mg

2+-Al

3+-ClO4

-, Ni

2+-Al

3+-Cl

- and Zn

2+-

Al3+

-Cl. Clays and Clay Minerals. 1975;23:369-75.

[6] Constantino VR, Pinnavaia TJ. Basic properties of Mg-Al layered double hydroxides intercalated by

carbonate, hydroxide, chloride, and sulfate anions. Inorganic chemistry. 1995;34(4):883-92.

[7] Hibino T, Kobayashi M. Delamination of layered double hydroxides in water. Journal of Materials

Chemistry. 2005;15(6):653-6.

[8] Whilton NT, Vickers PJ, Mann S. Bioinorganic clays: synthesis and characterization of amino-

andpolyamino acid intercalated layered double hydroxides. Journal of Materials Chemistry.

1997;7(8):1623-9.

[9] Fudala A, Palinko I, Kiricsi I. Preparation and characterization of hybrid organic-inorganic composite

materials using the amphoteric property of amino acids: amino acid intercalated layered double

hydroxide and montmorillonite. Inorganic chemistry. 1999;38(21):4653-8.

[10] Coelho C, Stimpfling T, Leroux F, Verney V. Inorganic–Organic Hybrid Materials Based on Amino

Acid Modified Hydrotalcites Used as UV‐Absorber Fillers for Polybutylene Succinate. European

Journal of Inorganic Chemistry. 2012;2012(32):5252-8.

[11] Reinholdt MX, Kirkpatrick RJ. Experimental investigations of amino acid-layered double hydroxide

complexes: glutamate-hydrotalcite. Chemistry of Materials. 2006;18(10):2567-76.

[12] Aisawa S, Sasaki S, Takahashi S, Hirahara H, Nakayama H, Narita E. Intercalation of amino acids

and oligopeptides into Zn–Al layered double hydroxide by coprecipitation reaction. Journal of Physics

and Chemistry of Solids. 2006;67(5):920-5.

[13] Wei M, Shi Z, Evans DG, Duan X. Study on the intercalation and interlayer oxidation transformation

of L-cysteine in a confined region of layered double hydroxides. Journal of Materials Chemistry.

2006;16(21):2102-9.

[14] Perioli L, Ambrogi V, Bertini B, Ricci M, Nocchetti M, Latterini L, et al. Anionic clays for sunscreen

agent safe use: Photoprotection, photostability and prevention of their skin penetration. European

Journal of Pharmaceutics and Biopharmaceutics. 2006;62(2):185-93.

[15] Choy J-H, Kwak S-Y, Park J-S, Jeong Y-J, Portier J. Intercalative nanohybrids of nucleoside

monophosphates and DNA in layered metal hydroxide. Journal of the American Chemical Society.

1999;121(6):1399-400.

[16] Choy J-H, Kwak S-Y, Jeong Y-J, Park J-S. Inorganic layered double hydroxides as nonviral vectors.

Angewandte Chemie. 2000;39(22):4041-5.

[17] Wang J, Zhou J, Li Z, Liu Q, Yang P, Jing X, et al. Design of magnetic and fluorescent Mg-Al

layered double hydroxides by introducing Fe304 nanoparticles and Eu3+

ions for intercalation of

glycine. Materials Research Bulletin. 2010;45(5):640-5.

[18] Nakayama H, Hatakeyama A, Tsuhako M. Encapsulation of nucleotides and DNA into Mg–Al

layered double hydroxide. International Journal of Pharmaceutics. 2010;393(1):105-12.

[19] Zhao D, Bai Z, Zhao F. Preparation of Mg/Al-LDHs intercalated with dodecanoic acid and

investigation of its antiwear ability. Materials Research Bulletin. 2012;47(11):3670-5.

[20] Raki L, Beaudoin J, Mitchell L. Layered double hydroxide-like materials: nanocomposites for use in

concrete. Cement and Concrete Research. 2004;34(9):1717-24.


[21] Costa FR, Leuteritz A, Wagenknecht U, Jehnichen D, Haeussler L, Heinrich G. Intercalation of Mg–

Al layered double hydroxide by anionic surfactants: preparation and characterization. Applied Clay

Science. 2008;38(3):153-64.

[22] Williams G, McMurray HN. Anion-exchange inhibition of filiform corrosion on organic coated

AA2024-T3 aluminum alloy by hydrotalcite-like pigments. Electrochemical and Solid-State Letters.

2003;6(3):B9-B11.

[23] You Y, Zhao H, Vance GF. Adsorption of dicamba (3, 6-dichloro-2-methoxy benzoic acid) in

aqueous solution by calcined–layered double hydroxide. Applied Clay Science. 2002;21(5):217-26.

[24] Aisawa S, Kudo H, Hoshi T, Takahashi S, Hirahara H, Umetsu Y, et al. Intercalation behavior of

amino acids into Zn–Al-layered double hydroxide by calcination–rehydration reaction. Journal of

Solid State Chemistry. 2004;177(11):3987-94.

[25] Kameda T, Fubasami Y, Yoshioka T. Kinetics and equilibrium studies on the treatment of nitric acid

with Mg–Al oxide obtained by thermal decomposition of-intercalated Mg–Al layered double

hydroxide. Journal of Colloid and Interface Science. 2011;362(2):497-502.

[26] Newman SP, William J. Synthesis, characterization and applications of layered double hydroxides

containing organic guests. New Journal of Chemistry. 1998;22(2):105-15.

[27] Del Arco M, Cebadera E, Gutierrez S, Martin C, Montero M, Rives V, et al. Mg, Al layered double

hydroxides with intercalated indomethacin: synthesis, characterization, and pharmacological study.

Journal of Pharmaceutical Sciences. 2004;93(6):1649-58.

[28] Nakayama H, Wada N, Tsuhako M. Intercalation of amino acids and peptides into Mg–Al layered

double hydroxide by reconstruction method. International Journal of Pharmaceutics. 2004;269(2):469-

78.

[29] Chibwe K, Jones W. Intercalation of organic and inorganic anions into layered double hydroxides.

Journal of the Chemical Society, Chemical Communications. 1989(14):926-7.


chemistry. 1990;29(26):5201-7.

[31] Grover K, Komarneni S, Katsuki H. Uptake of arsenite by synthetic layered double hydroxides.

Water research. 2009;43(15):3884-90.

[32] Wan D, Liu H, Liu R, Qu J, Li S, Zhang J. Adsorption of nitrate and nitrite from aqueous solution

onto calcined (Mg–Al) hydrotalcite of different Mg/Al ratio. Chemical Engineering Journal.

2012;195:241-7.

[33] Sharma U, Tyagi B, Jasra RV. Synthesis and Characterization of Mg-Al-CO3 Layered Double

Hydroxide for CO2 Adsorption. Industrial & Engineering Chemistry Research. 2008;47(23):9588-95.

[34] Wang G-A, Wang C-C, Chen C-Y. Preparation and Characterization of Layered Double Hydroxides

– PMMA Nanocomposites by Solution Polymerization. Journal of Inorganic and Organometallic

Polymers and Materials. 2005;15(2):239-51.

[35] Thomas N, Pradeep Kumar G, Rajamathi M. Synthesis and intracrystalline oxidation of nitrite-

intercalated layered double hydroxides. Journal of Solid State Chemistry. 2009;182(3):592-6.

[36] Poznyak SK, Tedim J, Rodrigues LM, Salak AN, Zheludkevich ML, Dick LF, et al. Novel inorganic

host layered double hydroxides intercalated with guest organic inhibitors for anticorrosion applications.

ACS Appl Mater Interfaces. 2009;1(10):2353-62.

[37] Kameda T, Yamazaki T, Yoshioka T. Preparation and characterization of Mg–Al layered double

hydroxides intercalated with benzenesulfonate and benzenedisulfonate. Microporous and Mesoporous

Materials. 2008;114(1):410-5.

[38] Costantino U, Nocchetti M, Sisani M, Vivani R. Recent progress in the synthesis and application of

organically modified hydrotalcites. Z Kristallogr. 2009;224:273-81.

68 | Chapter 4

[39] Lin Y-J, Li D-Q, Evans DG, Duan X. Modulating effect of Mg-Al-CO3 layered double hydroxides on

the thermal stability of PVC resin. Polymer Degradation and Stability. 2005;88(2):286-93.

[40] Kloprogge JT, Wharton D, Hickey L, Frost RL. Infrared and Raman study of interlayer anions CO32−

,

NO3−, SO4

2− and ClO4

− in Mg/Al-hydrotalcite. American Mineralogist. 2002;87(5-6):623-9.

[41] Fuda K, Kudo N, Kawai S, Matsunaga T. Preparation of Zn/Ga-layered double hydroxide and its

thermal decomposition behavior. Chemistry Letters. 1993(5):777-80.

[42] Mokhtar M, Saleh TS, Basahel SN. Mg–Al hydrotalcites as efficient catalysts for aza-Michael

addition reaction: a green protocol. Journal of Molecular Catalysis A: Chemical. 2012;353:122-31.

[43] Bish D, Brindley G. A reinvestigation of takovite, a nickel aluminum hydroxy-carbonate of the

pyroaurite group. American Mineralogist. 1977;62(5-6):458-64.

[44] Palomares A, Prato J, Rey F, Corma A. Using the “memory effect” of hydrotalcites for improving the

catalytic reduction of nitrates in water. Journal of catalysis. 2004;221(1):62-6.

[45] Hsueh H-B, Chen C-Y. Preparation and properties of LDHs/polyimide nanocomposites. Polymer.

2003;44(4):1151-61.

[46] Choy J-H, Son Y-H. Intercalation of vitamer into LDH and their controlled release properties.

Bulletin-Korean chemical society. 2004;25(1):122-6.

[47] Han Y-S, Choy J-H. Exfoliation of layered perovskite, KCa2Nb3O10, into colloidal nanosheets by a

novel chemical process. Journal of Materials Chemistry. 2001;11(4):1277-82.

[48] Wang D-Y, Costa FR, Vyalikh A, Leuteritz A, Scheler U, Jehnichen D, et al. One-step synthesis of

organic LDH and its comparison with regeneration and anion exchange method. Chemistry of

Materials. 2009;21(19):4490-7.

[49] Rey F, Fornés V, Rojo JM. Thermal decomposition of hydrotalcites. An infrared and nuclear

magnetic resonance spectroscopic study. Journal of the Chemical Society, Faraday Transactions.

1992;88(15):2233-8.

[50] Bergaya F, Theng B, Lagaly G. Developments in clay science. Handbook of clay science. 2006;1.

[51] Yun SK, Pinnavaia TJ. Water content and particle texture of synthetic hydrotalcite-like layered

double hydroxides. Chemistry of Materials. 1995;7(2):348-54.

[52] Bera P, Rajamathi M, Hegde M, Kamath PV. Thermal behaviour of hydroxides, hydroxysalts and

hydrotalcites. Bulletin of Materials Science. 2000;23(2):141-5.

Chapter 5

Anti-corrosion Performance

Evaluation of Synthesized Modified

Hydrotalcites in Simulated Concrete

Pore Solution

Part of the work described in this chapter has been published as: Yang, Z., Fischer, H., Cerezo, J.,

Mol, J.M.C., Polder, R. Aminobenzoate Modified Mg-Al Hydrotalcites As a Novel Smart

Additive of Reinforced Concrete for Anticorrosion Applications. Construction and Building

Materials, 2013, 47, 1436-1443.

70 | Chapter 5

5.1 Introduction

Based on the results and discussion in Chapter 3, NaNO2, pAB and 11AUA were selected as the

most promising candidate modifiers for synthesis of MHTs. In Chapter 4, synthesis and

characterization of six MHTs (with two Mg/Al atomic ratios of 2 & 3) intercalated with the three

selected inhibitors have been described. This chapter describes the evaluation of the anti-

corrosion performance of the six synthesized MHTs in chloride containing simulated concrete

pore solution based on the open circuit potential (OCP) and linear polarization resistance (LPR)

measurements and chloride exchange experiments. The primary objective of this chapter is

therefore to select the MHTs with the best anti-corrosion performance for use in both plain and

reinforced mortar tests which will be discussed in the following Chapters 6 and 7.

5.2 Experimental

5.2.1 Materials

ACS grade NaOH and NaCl were obtained from Sigma-Aldrich. Six MHTs (with two Mg/Al

atomic ratios of 2 & 3) that were used in this study are the synthetic products from Chapter 4 (see

Table 4.3). Deionized water was used for the preparation of aqueous solutions. Steel electrodes

used for anti-corrosion performance evaluation were low-carbon steel (ASTM A36) coupons with

an exposed surface area of 100 mm2. Prior to immersion in the relevant testing solutions, the steel

electrodes were ground successively using a series of silicon carbide emery papers of grades 320,

800, 1200, 2400 and 4000 in water. The steel electrodes were further cleaned with acetone under

ultrasonication and then dried.

5.2.2 Ion exchange of MHT with chlorides in simulated concrete pore solution

The chloride exchange properties of the six synthesized MHTs were investigated in a mixed

solution containing 0.1 M NaOH and 0.5 M NaCl to simulate the concrete pore solution strongly

contaminated by chlorides. The schematic illustration of the procedure for chloride exchange

experiment is given in Figure 5.1. Typically, a volume of 20 ml mixed solution of 0.1 M NaOH

and 0.5 M NaCl was vigorously mixed with 0.5 g powder of MHTs in a screw capped glass

centrifuge tube. The suspension was then sealed to isolate it from the atmosphere and put in a

rotating device at room temperature for a predetermined period of exchange time to allow the

occurrence of ion exchange. The ion exchange time was set as 2 h, 6 h, 12 h, 24 h, 48 h, 72 h, 96

h and 120 h. Afterwards, the remaining solid at each exchange time point was separated from

bulk solution by centrifugation and washed thoroughly with deionized water. The separated solid

was collected and oven dried for 24 h at 105 °C under vacuum. Then, the chlorides in the solid as

well as the inhibitors released through the ion exchange in supernatant were respectively analyzed

Anti-corrosion Performance Evaluation of Synthesized MHTs | 71

photometrically by a Spectroquant® NOVA 60 spectrophotometer. Duplicates were measured

simultaneously to ensure the reliability of the test results. In addition, X-ray powder diffraction

(XRD) was employed to qualitatively analyze the crystal structure of MHT before and after

chloride exchange to further validate whether or not the ion exchange reaction indeed occurred

between MHTs and chlorides.

Figure 5.1 Schematic illustration of the procedure for chloride exchange experiment.

5.2.3 Anti-corrosion performance evaluation in simulated concrete pore solution

After the chloride exchange experiments, the anti-corrosion performance of the MHTs was

further evaluated by monitoring the OCP/LPR evolution of the steel in 0.1 M NaOH solution

contaminated by chlorides. Five solutions were prepared for this purpose:

1) 0.1 M NaOH (as control case) simulating the concrete pore solution;

2) 0.1 M NaOH + NaCl;

3) 0.1 M NaOH + 0.1 M inhibitor (i.e., pAB, NaNO2 or 11AUA) + NaCl;

4) 0.1 M NaOH + 0.1 M MHT;

5) 0.1 M NaOH + 0.1 M MHT + NaCl

here 0.1 M in “0.1 M MHT” represents the molar concentration of the inhibitor anions (-pAB, -

NO2 or -11AUA) which have been intercalated in MHT. NaCl was added as a source of Cl-

contamination into the testing solution after the steel electrode has reached the passivation state.

The NaCl concentration was progressively increased in several steps per every 24 h starting from

0.05 M up to 0.4 M. NaCl crystals were first dissolved with part of the test solution then mixed

72 | Chapter 5

thoroughly with the bulk solution inside the test cell. The corresponding readings of OCP every 2

h and LPR every 6 h were recorded up to 168 h. All potentials are referred to the Saturated

Calomel Electrode (SCE). More details about the OCP and LPR measurements and relevant

experimental setup can be found in Chapter 3. In order to study the effect of MHT’s addition on

alkalinity of the testing solution, the pH was measured at the start and the end of the OCP/LPR

testing period (i.e. 168 h) by a pH meter (Metrohm 827 pH lab). For all the tests, at least two or

three independent measurements per testing condition were performed to validate the

reproducibility of the results.


5.3.1 The role of MHT in capturing chlorides

The key feature of MHTs is their high anionic exchange capacity which makes exchange of the

interlayer ions by a wide range of organic or inorganic anions versatile and easily achieved [1-3].

For the envisioned use as a functional additive of reinforced concrete against chloride ingress, a

specific anion with known inhibitive properties could be intercalated into molecular structure of

the hydrotalcite through modification. Then, an ion exchange reaction in MHT can be triggered

upon arrival of the external chloride ions. The intercalated inhibitive anion is simultaneously

released and provides further inhibitory protection of the reinforcing steel as shown in Eq. 5.1 [4]:

MHT-Inh + Cl- MHT-Cl + Inh

- (5.1)

where -Inh represents the intercalated inhibitive ions in the molecular structure of MHT (i.e., -

pAB, -11AUA, and -NO2 anions in this research). Cl- is the free chloride ions present in concrete

pore solution. As such, MHT plays a dual role against chloride-induced corrosion in reinforced

concrete capturing chlorides as a chloride scavenger and providing corrosion inhibitors in parallel

as an internal inhibitor reservoir working continuously to protect reinforcing steel from corrosion.

5.3.2 Chloride exchange in simulated concrete pore solution

The chloride exchange properties of the six synthesized MHTs were investigated in 0.1 M NaOH

solution containing chloride ions. After the predetermined exchange time from 2-120 h, the MHT

solid was separated from the bulk solution. The amount of chlorides in remaining MHT solid and

the amount of released inhibitors in supernatant were subsequently analyzed. The chloride

exchange and inhibitor release ratios (in molar fraction) were calculated according to the

following equations (Eq. 5.2 & 5.3):

100%( )

b cl

MHT Inh Inh

m MChloride exchange ratio

m W M (5.2)


100%( )

Inh Ihn Inh

MHT Inh Inh MHT Inh

m M mInhibitor release ratio

m W M m W (5.3)

where mb (mg) is the mass of bound chloride detected in the remaining solid MHT; mMHT (g) is

the mass of dry solid MHT (i.e., MHT-pAB, MHT-11AUA or MHT-NO2); Mcl (g/mol) is the

molecular weight of chloride (i.e., 35.5); WInh (%) is original percentage content of the

intercalated inhibitive anion by mass of MHT (as shown in Table 4.3 of Chapter 4); MInh (g/mol)

is the molecular weight of the intercalated inhibitive anion (136 for -pAB, 200 for -11AUA and

46 for -NO2); mInh (mg) is the mass of the released inhibitive anion detected in supernatant (i.e., -

pAB, -11AUA or -NO2).

The profiles of chloride bound by MHT and corresponding inhibitors released from MHT are

given respectively in Figures 5.2 & 5.3, in which the chloride exchange and inhibitor release

ratios were plotted against the exchange time. XRD was employed to qualitatively analyze the

crystal structure of MHT before and after the chloride exchange experiments to further confirm

the occurrence of ion exchange between the intercalated inhibitive anions in MHT and free

chlorides in simulated concrete pore solution. The relevant XRD patterns are shown in Figures

5.4-5.6. As can be seen from Figures 5.2 & 5.3, all the six investigated MHTs show a rapid

Figure 5.2 Chloride binding profile of the six synthesized MHTs with Mg/Al atomic ratios of 2 & 3.

74 | Chapter 5

increase of concentration of the bound chloride and the released inhibitor during the first 2 h

followed by a slow and more sustained exchange until a period time of 120 h. The similarity of

the two profiles proves the occurrence of ion exchange of the intercalated inhibitive ions (i.e., -

pAB, -11AUA or -NO2,) with the chlorides. The rapid increase part of the curves in both profiles

is likely ascribed to the ion exchange starting at the edges or in the external part of the lamellar

structure of MHTs, where most of them are readily available to be exchanged. The following

slow and more sustained development of the both profiles is probably owing to the exchange of

the inhibitive ions which are located in the internal area of the lamellae and have to diffuse

through the MHT particles. It can be imaged that as small species (Cl-) exchange bigger inhibitive

ions (i.e., -pAB, -11AUA or -NO2), a consequent decease of the interlayer distance occurs. Thus

the initial exchange of chlorides with intercalated inhibitive ions in the external part of the

lamellar structure may probably lead to the formation of a phase boundary between internal zones

containing the bigger intercalated inhibitive ions and the external area in which inhibitive anions

have already been replaced by the smaller chlorides [5, 6].

As such, a smaller and bigger interlayer distance co-exist in the same crystal of MHT

resulting in a phase boundary, which may restrain the rate of the ion exchange. As the ion

exchange proceeds, the phase boundary moves towards the center of the crystal structure which

may in turn lead to the progressive reduction of the ion exchange rate. As an important structural

Figure 5.3 Release profile of the intercalated inhibitors from the six synthesized MHTs with Mg/Al

atomic ratios of 2 & 3.


parameter of MHT, the Mg/Al atomic ratio plays an important role in determining the ion

exchange capacity of the MHT although no direct correlation was reported between the two

parameters [7]. It can be observed from Figures 5.2 & 5.3, MHTs with Mg/Al atomic ratio of 2

show higher bound chloride and inhibitor release ratios than those with Mg/Al atomic ratio of 3

indicating a higher ion exchange capacity of MHT with Mg/Al atomic ratio of 2 with respect to

that of 3. This can possibly be ascribed to the higher charge density of MHT with Mg/Al atomic

ratio of 2 than that of 3 where fewer bivalent cations (Mg2+

) are isomorphously substituted by

trivalent cations (Al3+

) in the MHT molecular structure. The influence of the intercalated

inhibitive anion on chloride exchange/binding capacity is also evident. As shown in Figure 5.2,

the ratio of bound chloride varies with the type of intercalated anions increasing in the order

Mg(2)Al-11AUA< Mg(2)Al-NO2< Mg(2)Al-pAB in the group with Mg/Al atomic ratio of 2 and

Mg(3)Al-11AUA< Mg(3)Al-NO2< Mg(3)Al-pAB in the group with Mg/Al atomic ratio of 3. For

inhibitor release ratio, the same increasing sequence can be observed from Figure 5.3. This

observation is in accordance with previous research findings that the combination of the guest

(i.e., the intercalated anions as well as the chloride ions) and host (metal hydroxide layer) of the

MHT structure could exert a notable effect on the chloride exchange capacity [7, 8]. In addition,

the nature of the intercalated anions and their interactions in the interlayer region could also be

important factors governing the rate of ion exchange. Specifically, the larger interlayer space of

MHT-pAB (for both Mg/Al atomic ration of 2 & 3 with a d-spacing value of 15.0 Å) relative to

MHT-NO2 (for both Mg/Al atomic ration of 2 & 3 with a d-spacing value of 7.8 Å) provide more

accessibility and ion exchange sites for chloride ions to diffuse and replace the intercalated anions

of -pAB. That consequently contributes to both higher bound chloride ratio and inhibitor release

ratio of Mg(2)Al-pAB and Mg(3)Al-pAB compared respectively to Mg(2)Al-NO2 and Mg(3)Al-

NO2. For MHT-11AUA (for both Mg/Al atomic ration of 2 & 3 with a d-spacing value of 16.4 Å),

the poor solubility of the -11AUA might play a critical role in constraining the chloride exchange

with the intercalated anion of -11AUA, although MHT-11AUA also has a larger interlayer space

compared to MHT-NO2.

It is interesting to find that the inhibitor release ratios of all the six MHTs are higher than

their corresponding bound chloride ratios. For example, the bound chloride ratio at 120 h was

76.2% and 63.8% respectively for Mg(2)Al-pAB and Mg(3)Al-pAB, 65.2% and 55.2% for

Mg(2)Al-NO2 and Mg(3)Al-NO2, and 16.3% and 7.4% for Mg(2)Al-11AUA and Mg(3)Al-

11AUA, whilst their corresponding inhibitor release ratio at 120h was 92.6% and 81.2%, 78.2%

and 73.0%, and 28.0% and 17.0% respectively. In addition to the potential exchange of

intercalated inhibitive ions by CO32-

, which could be produced by airborne CO2 in the highly

alkaline solution, the differences between these inhibitor release and bound chloride ratios may

be caused by the co-existence of competitive OH- ions. Since the fact that the MHTs were

synthesized in alkaline solution (pH is around 10) over a reaction time of 24 h, the OH- may not

be expected to exchange the intercalated inhibitors from MHT [9]. However, previous research

[10-12] reported the possibility of decrease of solution pH (although limited) due to the

76 | Chapter 5

adsorption of OH- by hydrotacite-like compounds in a solution with very high alkalinity (pH 13-

14). In this research, although the synthesis of MHT was performed in a solution at pH around 10,

the chloride exchange experiment was conducted in a solution with higher alkalinity (i.e., 0.1 M

NaOH + 0.5 M NaCl; pH is around 13). That is to say, the higher solution alkalinity in the

chloride exchange experiment could cause a different OH- “equilibrium” concentration with

MHT compared to that in the condition of synthesis, which in turn may result in a partial

replacement of inhibitive ions by OH-. Moreover, some impurities of the inhibitive anions could

be taken up on the surface of the MHT crystal due to electrostatic attraction during the synthesis

and post synthesis processes. The liberation of these impurities might be another possibility

causing the difference between inhibitor release ratio and bound chloride ratio.

The crystal structure change of the six MHTs before and after chloride exchange confirmed

the occurrence of ion exchange between chlorides and the interlayer inhibitive anions (i.e., -NO2,

-pAB or -11AUA). As can be seen from Figures 5.4 & 5.5, the characteristic peaks in XRD

patterns of MHT-pAB (i.e., Mg(2)Al-pAB and Mg(3)Al-pAB) and MHT-NO2 (i.e., Mg(2)Al-

NO2 and Mg(3)Al-NO2) have been changed after chloride exchange. The d-spacing value of (003)

which correlates to the length of interlayer space of MHT was found changed respectively from

the original 15.0 Å (MHT-pAB) and 7.8 Å (MHT-NO2) to 7.7 Å, a value which is in good

agreement with reported basal spacing of chloride intercalated hydrotalcites [4, 13-15]. This

information clearly reveals that chloride exchange indeed occurred between chloride and the

intercalated inhibitive ions in both MHT-pAB and MHT-NO2. In addition, the sharper and more

symmetric diffraction peaks shown in XRD patterns of MHT-Cl (i.e., Mg(2)Al-Cl and Mg(3)Al-

Cl) suggest a typical layered structure with higher crystallinity resulting from the ion exchange

process. Regarding MHT-11AUA (Figure 5.6), however, differently from MHT-pAB and MHT-

NO2, the characteristic peaks in XRD patterns of MHT-11AUA, in particular, those peaks

between 5 and 25 (2θ/degree) in both Mg(2)Al-11AUA and Mg(3)Al-11AUA did not change

significantly after the 120 h chloride exchange experiment. This may indicate that the exchange

reaction was not really completed. This observation of incompleteness of exchange reaction

further validated the results of both much lower bound chloride ratio and inhibitor release ratio

(see Figures 5.2 & 5.3) relative to those of MHT-pAB and MHT-NO2.

Based on the combined results from chloride content analysis and XRD characterization,

MHTs with Mg/Al atomic ratio of 2 (i.e., Mg(2)Al-pAB, Mg(2)Al-NO2 and Mg(2)Al-11AUA)

were chosen to carry out the experiments in the following section (anti-corrosion performance

evaluation) because of their relatively higher chloride exchange capacity.


Figure 5.4 XRD patterns of MHT-pAB (Mg/Al = 2 (left) & 3 (right)) before (bottom) and after (top)

chloride exchange.

Figure 5.5 XRD patterns of MHT-NO2 (Mg/Al = 2 (left) & 3 (right)) before (bottom) and after (top)

chloride exchange.

Figure 5.6 XRD patterns of MHT-11AUA (Mg/Al = 2 (left) & 3 (right)) before (bottom) and after (top)

chloride exchange.

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

nsit

y/c

ou

nts

2Theta /degrees

(003), d=7.7Å

(003), d=15.0Å

Mg(2)Al-pAB

Mg(2)Al-Cl

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

ns

ity/c

ou

nts

2Theta/ degrees

Mg(3)Al-Cl

Mg(3)Al-pAB

(003), d=7.7Å

(003), d=15.0Å

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

ns

ity/c

ou

nts

2Theta /degrees

(003), d=7.7Å

(003), d=7.8Å

Mg(2)Al-Cl

Mg(2)Al-NO2

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

ns

ity/c

ou

nts

2Theta /degrees

(003), d=7.7Å

(003), d=7.8Å

Mg(3)Al-Cl

Mg(3)Al-NO2

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

nsit

y/c

ou

nts

2Theta /degrees

Mg(2)Al-Cl

Mg(2)Al-11AUA

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Inte

nsit

y/c

ou

nts

2Theta/ degrees

Mg(3)Al-Cl

Mg(3)Al-11AUA

78 | Chapter 5

5.3.3 Anti-corrosion performance of the selected MHT

5.3.3.1 The effect of MHT addition on pH of the alkaline solution

Previous studies have revealed that the pH value of the testing solution is a dominating factor

influencing the passivation of steel [16-18]. In this research, the effect of the three selected MHTs

(i.e., Mg(2)Al-NO2, Mg(2)Al-pAB, Mg(2)Al-11AUA) on pH of the alkaline solution including

the possible carbonation by the airborne CO2 was investigated under the designated testing

condition (i.e., aerated alkaline stagnant solutions). A similar investigation was also done with the

five proposed inhibitors in Chapter 3 (see Table 3.2). Table 5.1 gives the measured pH values of

the pure 0.1 M NaOH solution (data taken from Table 3.2 of Chapter 3) and three selected MHT

solutions at the start and the end of the 7 days (i.e. 168 h) OCP/LPR testing period. Comparing

the results presented in Table 5.1 with those shown in Table 3.2 and considering the discussion

that has made in section 3.3.1 (Chapter 3), similar conclusions can be drawn: (1) the carbonation

of the solutions containing the three MHTs over the entire testing period was ignorable; (2) no to

minor influence on pH of the alkaline solution (i.e., 0.1 M NaOH) was resulted from the addition

of the three MHTs (at 0.1 M). In this aspect, the differences in terms of either the addition of

MHTs or their counterpart inhibitors (i.e., Mg(2)Al-NO2 vs NaNO2; Mg(2)Al-pAB vs pAB;

Mg(2)Al-11AUA vs 11AUA) into the 0.1 M NaOH solution are negligible.

Table 5.1 Measured pH of the solutions with/without chloride addition during the 168 h testing period

(resolution of the pH meter is 0.001; all materials used in the testing solution have the same concentration

of 0.1 M).

Testing solution NaOH Mg(2)Al-NO2

+NaOH

Mg(2)Al-pAB

+NaOH

Mg(2)Al-11AUA

+NaOH

pH at 0 h 12.8 12.7 12.7 12.8

pH at 168 h (without Cl-) 12.7 12.6 12.6 12.7

pH at 168 h (0.4M Cl-) 12.5 12.5 12.4 12.5

5.3.3.2 Open circuit potential (OCP) measurements

The OCP (i.e., the corrosion potential, Ecorr) evolution of the steel electrode was studied to

evaluate the anti-corrosion performance of the three selected MHTs in simulated concrete pore

solution contaminated by chlorides. For OCP evolution, the same criterion of this series

evaluation as used in Chapter 3 was adopted. Consequently, the MHT that has the best anti-

corrosion performance would be the one which can stay over -270 mV with the highest NaCl

concentration. The critical chloride concentration (i.e., chloride threshold, CT) at which the

corrosion of the steel electrodes was initiated is then registered. Figures 5.7-5.9 give the OCP

evolution of the steel electrodes in the mixed solution of 0.1 M NaOH and 0.1 M MHT (i.e.,

Mg(2)Al-pAB, Mg(2)Al-NO2 or Mg(2)Al-11AUA) containing chlorides with stepwise increased

concentration ranging from 0.05 M up to 0.4 M. The OCP evolution of the steel in pure 0.1 M

NaOH solution and in its chloride containing solution (i.e., 0.1 M NaOH + Cl) was included as

reference cases in all these figures. They were obtained from the same steel specimens,


respectively. The OCP evolution in solutions containing 0.1 M inhibitor (i.e., pAB, NaNO2 or

11AUA as discussed in Chapter 3) was included for the comparison purpose to manifest the dual

role protecting function that MHT offers to the steel: capturing chlorides and simultaneously

releasing the intercalated inhibitors to further protect the steel. As can be seen from Figures 5.7-

5.9, when no chlorides were present in the solution, the OCPs of steel in all solutions containing

the three MHT evolved similarly with time as that in pure 0.1 M NaOH solution. Immediately

after immersion, OCPs of all the freshly prepared steel electrodes reached a value in the range of

-550 mV to -400 mV and increased to be around -270 mV after about 24-48 h (except for 72 h in

the case of 11AUA). Afterwards, they went on to become more positive in a relatively steady

state. This clearly indicates that a stable passive layer has formed on the steel surface in all the

chloride-free alkaline solutions, which is not jeopardized by the presence of the three MHTs as

that found in pure inhibitor solutions (see discussion in Chapter 3).

The OCP evolution changed when chlorides were added. With the addition of 0.05 M

chloride, the OCP of the steel electrode in pure 0.1 M NaOH solution dropped markedly from -

273 mV at 52 h by more than 100 mV within the following 24 h indicating active corrosion has

been developed. Since the first addition of chloride was at the concentration of 0.05 M, it was

realized that a chloride concentration lower than 0.05 M may also be possible to initiate the

corrosion of the steel in pure 0.1 M NaOH solution resulting in a markedly drop of OCP. Thus,

the CT for corrosion initial of the steel in the pure 0.1 M NaOH solution was logically registered

Figure 5.7 OCP (Ecorr) evolution of the steel in (0.1M Mg(2)Al-pAB + 0.1M NaOH), (0.1M pAB + 0.1M

NaOH) and 0.1M NaOH solutions with/without stepwise increased chloride concentration.

80 | Chapter 5

as a value between 0 and 0.05 M.

For the steel in solutions containing the three MHTs, the corrosion was initiated at various

time by the addition of chloride at various concentration levels. As can be seen from Figure 5.7,

the corrosion was initiated in solution containing Mg(2)Al-pAB with the addition of 0.3 M

chloride at 134 h. The OCP started dropping from -271 mV until -365 mV when 0.4 M chloride

was added. While for the solution containing pAB, the corrosion was found to be initiated at 102

h with the addition of 0.2 M chloride. The OCP started dropping from -272 mV until -366 mV

when 0.3 M chloride was added. Obviously, the CT shifted from somewhere between 0.1 M and

0.2 M (for pAB containing solution) to a higher value in the range of 0.2 M and 0.3 M (for

Mg(2)Al-pAB containing solution). The results clearly suggested that the ion exchange between

chlorides and pAB anions had occurred according to the earlier proposed ion exchange

mechanism (see section 5.3.1). The subsequently released pAB anions in Mg(2)Al-pAB

containing solution was believed to have shown some inhibiting effect and caused shifting of

corrosion initiation to a higher chloride concentration than in pAB containing solution. This

statement is supported by more positive OCP of the steel in solution containing Mg(2)Al-pAB

than in pAB containing solution as more chlorides were added. However, when the chloride

concentration increased up to 0.4 M after 144 h immersion, the OCPs in both cases exhibited a

similar trend maintaining a relatively stable evolution in an active corrosion state. This

observation is likely due to exhaustion of the ion exchange capacity of Mg(2)Al-pAB considering

Figure 5.8 OCP (Ecorr) evolution of the steel in (0.1M Mg(2)Al-NO2 + 0.1M NaOH), (0.1M NaNO2 +

0.1M NaOH) and 0.1M NaOH solutions with/without stepwise increased chloride concentration.


the maximum amount of 0.1 M pAB ions that could be exchanged by chlorides. On the other

hand, because most of pAB anions may have already been exchanged by chlorides, the inhibiting

effect from the released pAB anions seemed to be less effective or disappeared especially in the

late stage of the test, i.e., after 144 h immersion. As described previously, the anti-corrosion

performance improvement by Mg(2)Al-pAB can be ascribed to the dual role protecting function

of MHT against chloride attack.

The anti-corrosion performance improvement by Mg(2)Al-NO2 with respect to NaNO2, was

found to be not as significant as that of Mg(2)Al-pAB with respect to pAB. As can be seen from

Figure 5.8, the addition of 0.3 M chloride caused a markedly drop of OCP in solutions containing

NaNO2 or Mg(2)Al-NO2 clearly indicating corrosion of the steel has been initiated in both cases.

Specifically, the OCP dropped from -251 mV at 124 h to -340 mV at 126 h in NaNO2 containing

solution, while from -205 mV at 130 h to -279 mV at 132 h in Mg(2)Al-NO2 containing solution.

Consequently, the CT of the steel in the two solutions was registered as a same value in the range

of 0.2-0.3 M. However, the time to corrosion initiation was extended from 124 h in NaNO2

containing solution to 132 h in Mg(2)Al-NO2 containing solution.

As can be seen from Figure 5.9, the addition of 0.1 M chloride caused a markedly drop of the

OCP in both Mg(2)Al-11AUA and 11AUA containing solutions at almost the same time,

Figure 5.9 OCP (Ecorr) evolution of the steel in (0.1M Mg(2)Al-11AUA + 0.1M NaOH), (0.1M 11AUA +

0.1M NaOH) and 0.1M NaOH solutions with/without stepwise increased chloride concentration. Note: the

steel in (0.1M 11AUA + 0.1M NaOH) and (0.1M Mg(2)Al-11AUA + 0.1M NaOH) solutions needed

more time (72 h) to passivate than in the other inhibitor solutions (24-48 h).

82 | Chapter 5

indicating the corrosion of the steel has been initiated in the two solutions. Accordingly, a same

value of the CT in the range of 0.05-0.1 M was registered for these two solutions. With more

chloride addition until 0.3 M, OCPs of steel in these two solutions exhibited a quite similar

evolution as that in pure 0.1 M NaOH solution. The coincidence of the OCP drop at the same

chloride concentration and the similarity in OCP evolution with higher chloride concentrations

suggested that no evident improvement in anti-corrosion performance could have been made by

Mg(2)Al-11AUA relative to 11AUA, likely owing to the relatively low chloride exchange

capacity of Mg(2)Al-11AUA (see Figure 5.2 & 5.3 and corresponding discussion in section

5.3.2). Obviously, for Mg(2)Al-NO2 or Mg(2)Al-11AUA, the benefit of the abovementioned dual

function working mechanism of MHT was not fully fulfilled in terms of the inability for raising

the CT at least under the assigned testing condition of this research. Table 5.2 gives an overview

of corrosion initiation time and chloride threshold of the steel in all the test solutions derived

from the OCP and LPR measurements (see also discussion below).

5.3.3.3 Linear polarization resistance (LPR) measurements

The corrosion current density (icorr) evolution obtained by LPR measurements further confirmed

the OCP results and the results are shown in Figures 5.10-5.12 and in Table 5.2. As the Ecorr

evolution (Figures 5.7-5.9), the icorr evolution of the steel in pure 0.1 M NaOH solution and 0.1 M

inhibitor solutions was also included respectively as reference cases and for comparison purpose

in this series figures. In addition, the same criterion as used in Chapter 3 was adopted: an

observable change in current density over time was recognized as the indication of passive-active

transition in the process of corrosion development. On the other hand, a current density 0.1

µA/cm2 was used as an additional critical value for determining the steel in a passive state, above

which depassivation occurs. As can be seen from Figures 5.10-5.12, the icorr decreased

progressively reaching a value in the range of 0.01-0.07 µA/cm2 after the first 48 h immersion (72

h for 11AUA) indicating all the steel has been passivated. As expected, when no chloride was

added, the icorr in solutions containing the three MHTs showed a similar evolution trend as in pure

0.1 M NaOH solution in which the steel maintained the passive state until the end of the test, i.e.,

up to 168 h.

The addition of 0.05 M NaCl into the pure 0.1 M NaOH solution caused a considerable

increase in icorr from 0.02 µA/cm2 at the 48 h to 0.15 µA/cm

2 at 54 h. The icorr kept increasing to

be 0.4 µA/cm2 within the following 24 h suggesting active corrosion was developed. The icorr for

Mg(2)Al-pAB containing solution started increasing to be 0.2 µA/cm2 at 132 h when 0.3 M

chloride was added, while the icorr for pAB containing solution increased to be 0.17 µA/cm2 at

108 h with the addition of a lower chloride concentration at [Cl] = 0.2 M (Figure 5.10). For

solution containing Mg(2)Al-NO2 (Figure 5.11), the icorr increased to be 0.31 µA/cm2 at 132 h

with the addition of 0.3 M chloride, whereas, the icorr increased to be 0.35 µA/cm2 at 126 h when

0.3 M chloride was added into NaNO2 containing solution. In the case of Mg(2)Al-11AUA

containing solution (Figure 5.12), the icorr started increasing to be 0.28 µA/cm2 at 108 h with the


addition of 0.1 M chloride, while the icorr in 11AUA containing solution increased to be 0.13

µA/cm2 at 102 h when the same amount of chloride was added.

The high current density (>0.1µA/cm2) detected in solutions containing the three MHTs as

well as in solutions containing their counterpart inhibitors clearly revealed depassivation occurred

respectively at the detecting time points by the addition of chloride at the critical concentrations

(see results in Table 5.2). On the other hand, it is interesting to find that the steel in Mg(2)Al-

pAB containing solution exhibited a lower corrosion rate (lower icorr values) than in pAB

containing solution with addition of more chlorides, in particular after the corrosion has been

initiated. The solutions containing Mg(2)Al-NO2 or Mg(2)Al-11AUA, on the contrary, showed a

higher corrosion rate (higher icorr values) than their counterparts, i.e., NaNO2 or 11AUA

containing solution. Taking account of a certain amount of chlorides that have been bound via ion

exchange reactions, this observation is probably related to the limited amount of released

inhibitive anions, i.e., -NO2 or -11AUA, which in turn may result in a high corrosion risk. It has

been reported that inadequate dosages of inhibitors may increase the risk of intensified pitting

corrosion on passive steel in a chlorides contaminated alkaline solution [19-21]. Based on the

above investigation and the results presented in Table 5.2, the anti-corrosion performance of the

three selected MHTs in terms of the critical chloride concentration for initiating corrosion in

simulated concrete pore solution (i.e., 0.1 M NaOH) can be established:

Mg(2)Al-NO2 ≥ Mg(2)Al-pAB > Mg(2)Al-11AUA

Figure 5.10 Corrosion current density (icorr) evolution of the steel in (0.1M Mg(2)Al-pAB + 0.1M NaOH),

(0.1M pAB + 0.1M NaOH) and 0.1M NaOH solutions with/without stepwise increased chloride concentration.

84 | Chapter 5

Figure 5.11 Corrosion current density (icorr) evolution of the steel in (0.1M Mg(2)Al-NO2 + 0.1M NaOH),

(0.1M NaNO2 + 0.1M NaOH) and 0.1M NaOH solutions with/without stepwise increased chloride

concentration.

Figure 5.12 Corrosion current density (icorr) evolution of the steel in (0.1M Mg(2)Al-11AUA + 0.1M

NaOH), (0.1M 11AUA + 0.1M NaOH) and 0.1M NaOH solutions with/without stepwise increased

chloride concentration. Note: the steel in (0.1M 11AUA + 0.1M NaOH) and (0.1M Mg(2)Al-11AUA +

0.1M NaOH) solutions needed more time (72 h) to passivate than in other inhibitor solutions (24-48 h).


Table 5.2 Time to corrosion initiation and chloride threshold of the steel in testing solutions derived from

OCP/LPR measurements (OCP was measured every 2 h and LPR was measured every 6 h from start of the

test until 168 h; all materials used in the testing solution have the same concentration of 0.1M ).

Testing Solution NaOH pAB

+NaOH

Mg(2)Al-pAB

+NaOH

NaNO2

+NaOH

Mg(2)Al-NO2

+NaOH

11AUA

+NaOH

Mg(2)Al-11AUA

+NaOH

Corrosion initiation time

(h) by OCP 52 102 134 126 132 98* 102*

Corrosion initiation time

(h) by LPR 54 108 132 126 132 102 108

Chloride threshold (M)

by OCP and LPR 0-0.05 0.1-0.2 0.2-0.3 0.2-0.3 0.2-0.3 0.05-0.1 0.05-0.1

* Note: the steel in (0.1M 11AUA + 0.1M NaOH) and (0.1M Mg(2)Al-11AUA + 0.1M NaOH) solutions

needed more time (72 h) to passivate than in the other solutions (24-48 h).

5.4 Conclusion

In this chapter, firstly, the chloride exchange properties of the six synthesized MHTs (with two

Mg/Al atomic ratios of 2 & 3) intercalated with three inhibitive anions (i.e., -pAB, -NO2, and -

11AUA) were investigated in a mixed solution containing 0.1 M NaOH and 0.5 M NaCl to

simulate a concrete pore solution strongly contaminated by chlorides. The chloride exchange

experiments were conducted at a predetermined period of time from 2-120 h. The profiles of

chloride bound by the six MHTs and corresponding inhibitors released from the six MHTs were

respectively analyzed by a Spectroquant® NOVA 60 spectrophotometer after the chloride

exchange experiments. XRD was employed to characterize the crystal structural change of these

MHTs before and after chloride exchange experiments. The analysis of the chloride binding and

inhibitor releasing profiles revealed that MHTs with Mg/Al atomic ratio of 2 show higher bound

chloride and inhibitor release ratios than those with Mg/Al atomic ratio of 3 indicating a higher

chloride exchange capacity of MHT with Mg/Al atomic ratio of 2 with respect to that of 3. In

addition, it was found that both bound chloride and inhibitor release ratios varied with the type of

intercalated anions increasing in the order Mg(2)Al-11AUA< Mg(2)Al-NO2< Mg(2)Al-pAB in

the group of MHTs with an Mg/Al atomic ratio of 2 and Mg(3)Al-11AUA< Mg(3)Al-NO2<

Mg(3)Al-pAB in the group of MHTs with an Mg/Al atomic ratio of 3. The XRD analysis

confirmed that ion exchange between free chloride ions in simulated pore solution and the

intercalated inhibitive anions in MHT-pAB and MHT-NO2 (with Mg/Al atomic ratios of 2 & 3)

indeed occurred. However, an incompleteness of chloride exchange was found in the case of

MHT-11AUA (i.e., both Mg(2)Al-11AUA and Mg(3)Al-11AUA).

Secondly, based on the results obtained from chloride exchange experiments and XRD

analysis, three MHTs with Mg/Al atomic ratio of 2 (i.e., Mg(2)Al-NO2, Mg(2)Al-pAB and

Mg(2)Al-11AUA) with higher chloride exchange capacity were selected to further evaluate their

anti-corrosion properties in simulated concrete pore solution (i.e., 0.1 M NaOH used in this

research). As a dominating factor that plays a significant role for the passivation of steel, the

effect of the three MHTs on the pH of simulated concrete pore solution was investigated. No to

minor influence was found on pH of the simulated concrete pore solution due to addition of the

86 | Chapter 5

three MHTs. The anti-corrosion evaluation of the three selected MHTs (at 0.1 M) was performed

by monitoring the OCP and LPR evolution of the steel in simulated concrete pore solution (0.1 M

NaOH) with addition of chloride at various concentration levels from 0.05 to 0.4 M. The results

showed that Mg(2)Al-pAB exhibited a improved corrosion inhibiting effect on the steel in terms

of the extended corrosion initiation time and higher chloride threshold compared to its

counterpart inhibitor (i.e., 0.1 M pAB). Moreover, the results obtained from OCP/LPR

measurements along with those from the chloride exchange experiments validated that besides

the ion exchange between free chloride ions and the intercalated inhibitive anions in MHT, the

simultaneously released inhibitive anions, in particular -pAB, were found to exhibit a notable

inhibiting effect and caused shifting of the corrosion initiation to a higher chloride concentration

level. This evidence further manifested the dual role protecting function that MHT (in particular,

Mg(2)Al-pAB in this research), offers to the steel: capturing chlorides and simultaneously

releasing the intercalated inhibitors to further protect the steel. For Mg(2)Al-NO2 or Mg(2)Al-

11AUA, however, the benefit of the abovementioned dual role function of MHT was not fully

fulfilled in terms of the inability for raising the CT at least under the assigned testing condition of

this research.

In summary, the anti-corrosion performance of the three selected MHTs at 0.1 M in

simulated concrete pore solution in terms of the critical chloride concentration that they can

sustain is in the order of:

Mg(2)Al-NO2 (0.2-0.3M Cl-) ≥ Mg(2)Al-pAB (0.2-0.3M Cl

-) > Mg(2)Al-11AUA (0.05-0.1M Cl

-)

Consequently, Mg(2)Al-NO2 and Mg(2)Al-pAB were finally selected as the more promising

MHT candidates to be brought into the following mortar test, which will be discussed in Chapters

6 & 7.

References


chemistry. 1990;29(26):5201-7.

[2] Van der Ven L, Van Gemert M, Batenburg L, Keern J, Gielgens L, Koster T, et al. On the action of

hydrotalcite-like clay materials as stabilizers in polyvinylchloride. Applied Clay Science.

2000;17(1):25-34.

[3] Costa FR, Abdel-Goad M, Wagenknecht U, Heinrich G. Nanocomposites based on polyethylene and

Mg–Al layered double hydroxide. I. Synthesis and characterization. Polymer. 2005;46(12):4447-53.

[4] Buchheit RG, Guan H, Mahajanam S, Wong F. Active corrosion protection and corrosion sensing in

chromate-free organic coatings. Progress in Organic Coatings. 2003;47(3):174-82.

[5] Costantino U, Ambrogi V, Nocchetti M, Perioli L. Hydrotalcite-like compounds: versatile layered

hosts of molecular anions with biological activity. Microporous and Mesoporous Materials.

2008;107(1):149-60.

[6] Ambrogi V, Fardella G, Grandolini G, Perioli L. Intercalation compounds of hydrotalcite-like anionic

clays with antiinflammatory agents—I. Intercalation and in vitro release of ibuprofen. International

Journal of Pharmaceutics. 2001;220(1):23-32.


[7] Khan AI, Ragavan A, Fong B, Markland C, O’Brien M, Dunbar TG, et al. Recent developments in the

use of layered double hydroxides as host materials for the storage and triggered release of functional

anions. Industrial & Engineering Chemistry Research. 2009;48(23):10196-205.

[8] Lv L, Sun P, Gu Z, Du H, Pang X, Tao X, et al. Removal of chloride ion from aqueous solution by

ZnAl-NO3 layered double hydroxides as anion-exchanger. Journal of Hazardous Materials.

2009;161(2):1444-9.

[9] Plank J, Zhimin D, Keller H, Hössle Fv, Seidl W. Fundamental mechanisms for polycarboxylate

intercalation into C3A hydrate phases and the role of sulfate present in cement. Cement and Concrete

Research. 2010;40(1):45-57.

[10] You Y, Vance GF, Zhao H. Selenium adsorption on Mg–Al and Zn–Al layered double hydroxides.

Applied Clay Science. 2001;20(1):13-25.

[11] bin Hussein MZ, Zainal Z, Yahaya AH, Foo DWV. Controlled release of a plant growth regulator, α-

naphthaleneacetate from the lamella of Zn–Al-layered double hydroxide nanocomposite. Journal of

Controlled Release. 2002;82(2):417-27.

[12] Hang TTX, Truc TA, Duong NT, Vu PG, Hoang T. Preparation and characterization of

nanocontainers of corrosion inhibitor based on layered double hydroxides. Applied Clay Science.

2012;67:18-25.

[13] Bothe Jr JV, Brown PW. PhreeqC modeling of Friedel's salt equilibria at 23±1 °C. Cement and


[14] Thomas M, Hooton R, Scott A, Zibara H. The effect of supplementary cementitious materials on

chloride binding in hardened cement paste. Cement and Concrete Research. 2012;42(1):1-7.

[15] Constantino VR, Pinnavaia TJ. Basic properties of Mg2+

1-xAl3+

x

layered double hydroxides

intercalated by carbonate, hydroxide, chloride, and sulfate anions. Inorganic Chemistry.

1995;34(4):883-92.

[16] Hausmann D. Steel corrosion in concrete--How does it occur? Materials protection. 1967;6(11):19-23.





[19] Page C, Ngala V, Page M. Corrosion inhibitors in concrete repair systems. Magazine of Concrete

Research. 2000;52(1):25-38.




alkaline aqueous environments. Materials performance. 1979;18(11):45-8.

Chapter 6

The Influence of Two Types of

Modified Hydrotalcites on Chloride

Ingress in Cement Mortar


Laboratory Investigation of the Influence of Two Types of Modified Hydrotalcites on Chloride

Ingress into Cement Mortar. Cement and Concrete Composites, 2015, 58: 105-113.

90 | Chapter 6

6.1 Introduction

As proposed by Tuutti [1] (also see Figure 2.2 in Chapter 2), the development of corrosion in

reinforced concrete consists of two different stages. The first stage (i.e., initiation phase) is

related to the penetration of the critical chloride concentration up to the surface of reinforcement.

The second stage (i.e., propagation phase) is related to reinforcement corrosion and concrete

cover cracking and spalling, which could consequently be associated with severe damage to the

concrete structures. Therefore, the best corrosion prevention strategy may need accordingly to

include two aspects: 1) to improve the chloride binding capacity of the concrete or other inherent

properties being able to impede chloride transport in concrete which would subsequently increase

the critical chloride concentration (i.e., chloride threshold) and result in a delayed corrosion

initiation of the reinforcing steel; 2) to slow down the corrosion propagation after corrosion has

initiated. Most traditionally available corrosion prevention approaches such as coatings on

reinforcing steel, concrete surface sealer, stainless steel reinforcement and cathodic protection [2-

5] focus on one of the aspects and less attention has intentionally been paid to the other. A

promising option that has been proposed in this thesis is to use the modified hydrotalcite (MHT)

as an alternative approach against chloride penetration and corrosion initiation in mortar (or

concrete) owing to its unique tunable molecular structure and high anionic exchange capacity [6-

8].

Hydrotalcite has been found in hydrated slag cements, which are known to be able to bind

more chloride ions than pure Portland cement [9-11]. More recently, Kayali et al. [12, 13]

demonstrated that hydrotalcite which comprises 53.94% of the crystalline phases in hardened

ground granulated blast furnace slag (GGBFS) paste is the main hydration product responsible for

the remarkable improvement in chloride binding and corrosion protection by concrete containing

GGBFS. In Chapter 5, preliminary studies conducted in chloride-rich simulated concrete pore

solution have shown that ion exchange indeed occurred between chlorides and intercalated

inhibitive ions of the MHT structure in addition to the inhibiting effect on the corrosion of steel

rendered by the simultaneous release of inhibitive anions. A distinctive feature of MHTs relative

to the other corrosion prevention approaches is their dual-role working mechanism: capturing

aggressive chlorides and simultaneously releasing inhibitive interlayer anions to further protect

the reinforcing steel from corrosion [14] (also see Figure 2.7 in Chapter 2). Therefore, once

applied as an active component in mortar (or concrete), the MHT can be envisioned as both

internal reservoir of corrosion inhibitors and trap of chlorides. As such the effects of MHT were

set to be investigated on both plain and reinforced mortars with a focus on their interaction with

chloride ions in plain mortar and in reinforced mortar mainly focusing on their inhibition

influence on corrosion of the reinforcing steel. In this Chapter, research was carried out and the

corresponding results were discussed to investigate the effect of MHT on chloride penetration in

The Influence of MHT on Chloride Ingress in Cement Mortar | 91

plain mortar. The inhibition influence of MHT on corrosion of the reinforcing steel will be

discussed in Chapter 7.

6.2 Experimental

6.2.1 Materials

CEM I 42.5N Portland cement in accordance with European standard EN 197-1 [15], CEN-

Standard sand (particle size: 0-2 mm) conforming to EN 196-1 [16] and deionized water were

used. Two types of MHTs that were used were synthetic Mg(2)Al-pAB and Mg(2)Al-NO2

(Mg/Al atomic ratio 2:1). They were synthesized as described in Chapter 4 and denoted as MHT-

pAB and MHT-NO2 respectively. Elemental analysis revealed that 11.1% NO2- and 32.2% -pAB

anions by mass of MHT were intercalated into their molecular structures which corresponds to

2.41 mmol (NO2-) and 2.38 mmol (-pAB) per gram of the MHT. The specific surfaces tested

according to the BET method were 0.83 m2/g, 31.8 m

2/g and 14.2 m

2/g for CEM I 42.5N, MHT-

NO2 and MHT-pAB respectively. The particle size distribution analyzed by laser granulometry

using isopropanol as dispersion medium is shown in Figure 6.1.

0

10

20

30

40

50

60

70

80

90

100

0.1 1 10 100 1000

Cu

mm

ula

tiv

e

vo

lum

e

(%)

Particle diameter (µm)

CEM I 42.5N

MHT-NO2

MHT-pAB

Figure 6.1 Particle size distribution of the used materials.

92 | Chapter 6

6.2.2 Sample preparation

The mortar samples were prepared with a constant (cement + MHT) content, a constant water-to-

(cement + MHT) mass ratio of 0.50, a constant sand-to-(cement + MHT) mass ratio of 3 and a

MHT-to-(cement + MHT) mass ratio of 0%, 5%, 10%, respectively. In fact, MHT was used to

replace cement. This means that the “true” water-cement (w/c) ratio was 0.50, 0.53 and 0.56

respectively for the three replacement levels. For both the reference (without MHT admixed) and

mortar specimens admixed with MHTs, three specimens were prepared to ensure sufficient

statistical reliability of test results. For the specimens with MHTs, the MHT powder was mixed

first with dry cement in a mixer, and then sand was added and stirred thoroughly at a low speed

for 30 s. Then deionized water was added into the mixture and stirred for about 90 s (low speed

for 30 s and high speed for 60 s) to achieve good workability. After mixing, the fresh mixtures

were cast in the specific molds (prisms and cylinders) and were carefully compacted on a

vibrating table to minimize the amount of entrapped air. All the specimens were demolded after

curing under room temperature (RT) and local lab environment for 24 h. Subsequently, they were

moved to a fog room at 23 ± 2 °C and over 95% relative humidity and cured for 2, 6 or 27

additional days before subjecting to any assigned test. Another series of mortar specimens were

prepared for Rapid Chloride Migration (RCM) test only. They were prepared by adding MHT

instead of replacing cement at 5 and 10% of cement mass following the same procedure described

above, keeping the w/c ratio of 0.50 and cement content constant.

6.2.3 Testing methods

6.2.3.1 Flow table test

The consistence of fresh mortars was measured by the flow table following the procedures

described in EN1015-3 [17]. The apparatus used in this method includes a flow table, a truncated

conical mould, a tamper and a caliper. After mixing, the fresh mortar mixture was immediately

introduced in two layers into the test mould which was lightly lubricated with low viscosity non-

resin mineral oil and placed centrally on the disc of the flow table. Each layer was compacted by

ten short strokes of the tamper to ensure uniform filling of the mould. The excess mortar on top of

the mould was skimmed off. The free area of disc was wiped clean and dry. After 15 s, the mould

was raised slowly and vertically to spread out the mortar on the disc by jolting the flow table 15

times at a constant frequency of one jolt per second. The diameter of the mortar mixture was then

measured in two directions perpendicular to each other with the result to the nearest mm. Each

mortar sample was tested for two times and the final consistence was determined by the mean

value of the two measurements.

6.2.3.2 Mechanical test

The mechanical properties, in particular, the flexural and compressive strength were investigated

at curing ages of 3, 7 and 28 days by a standard three-point bending and a compression test using


40 × 40 × 160 mm3 mortar prisms. The test was carried out in accordance with EN 196-1 [16].

The flexural strength was measured by applying the load vertically to the mortar prisms at the

rate of (50±10) N/s by a MACBEN (Servo plus evolution) materials testing equipment. The

samples used for measuring the compressive strength were from the prism halves from the

flexural strength test. The compression load was applied at a rate of (2400±200) N/s. At least

three replicates were performed for both tests.

6.2.3.3 Mercury intrusion porosimetry

The Mercury Intrusion Porosimetry (MIP) method was used to determine the porosity of mortar

specimens with/without MHTs by a “Micrometritics Poresizer 9500” mercury intrusion

porosimeter. The porosity measurement was carried out in two stages. The first stage was at low

pressure: from 0 to 0.0036 MPa. The second stage was at high pressure running from 0.0036 to

210 MPa and followed by an extrusion running from 210 to 0.14 MPa. The MIP samples were

taken from the intact parts of the mortar samples after the mechanical test. The samples were first

crushed by hammer into small pieces with dimensions of 1-2 cm3. Then these pieces were

immersed in liquid nitrogen for about 3 minutes to stop the hydration and stored in a vacuum

freeze-dryer at -28°C for more than two months until a stable mass loss of 0.01%/day was

reached before conducting the test. The quick freezing and drying process at low temperatures

allowed the remaining liquid solution transform into ice microcrystals and further be removed by

sublimation without significant damage to the microstructure.

6.2.3.4 Rapid Chloride Migration (RCM) test

The RCM tested was conducted following the procedures described in NordTest method NT

Build 492 [18] to determine the chloride migration coefficient of the mortar. After curing for 28

days, the 100 mm × 85 mm mortar cylinders were taken out from the fog room. Two slices of 15

mm and 20 mm were cut respectively from top and bottom of the specimen and consequently a

100 mm × 50 mm mortar slice was obtained as test specimen. The newly cut top surface was the

one to be exposed to the chloride solution. The applied voltage was selected based on the initial

current flowing through the test specimen at a voltage of 30 V. The higher is this initial current,

the lower is the applied voltage. In this study, voltages of 20 V and 15 V were selected

respectively for the reference and mortar specimens admixed with MHTs based on the obtained

initial current. The test was applied to three specimens per mixing design and the test duration in

these cases was always 24 h.

6.2.3.5 Natural diffusion test

The natural diffusion test was performed in accordance with NordTest method NT Build 443 [19]

on two mortar specimens per mixing design. At a curing age of 28 days, the 100 mm × 85 mm

mortar cylinders were taken out from the fog room. A 15 mm mortar slice was cut from top of the

sample and consequently a 100 mm × 70 mm mortar slice was obtained as the test specimen. The

surface that was newly cut was the one to be exposed to the chloride solution. After being

immersed in Ca(OH)2 solution for 3 days saturation, all the faces of the test specimens except the

94 | Chapter 6

one to be exposed were dried at RT to a stable white-dry condition and then given a thick epoxy

coating (3-5 mm) in order to allow the chloride to penetrate through the specimen in one direction

driven by the concentration gradient. When the coating had hardened, the test specimen was fully

saturated again with Ca(OH)2 for 3 days. Afterwards, the mortar specimens were immersed in a

container filled with 16.5% NaCl solution. The ratio between the exposed area and the volume of

NaCl solution was kept constant at 3 cylinders in 6 l solution. The container were covered tightly

by a lid and stored in laboratory condition for totally 68 days. This solution was used for 35 days

and then replaced by a new 16.5% NaCl solution. The container was shaken once a week during

the test period. After the exposure a total chloride profile was obtained.

6.2.3.6 Chloride profiles

After being immersed in 16.5% NaCl solution for 68 days, the specimens were taken out and

washed by tap water. After wiping off excess water from the surface, a profile grinder was used

to obtain mortar powders by precision grinding at 2 mm depth increments in layers parallel to the

exposed surface up to a depth of 25 mm for determination of the chloride profile. The grinding

was performed within a diameter approximately 25 mm less than the full diameter of the mortar

cylinders to avoid the risk of edge effects and potential disturbance from the epoxy coating. The

mortar powders that were collected were then dried in an oven at 105 °C for 24 h. The chloride

profile consisted of two analyses, the extraction of total chloride and the determination of cement

content. The chloride extraction was carried out following a similar procedure described in EN

14629 [20]. Typically, 2 g of dry powder was weighed into a 250 ml beaker in which 30 ml

deionized water and 10 ml 5 M nitric acid was added subsequently. The mixture was shaken

manually for 30 s followed by heating until boiling and boiled for about 4 min under continuous

magnetic stirring. After cooling down, the mixture was vacuum filtered using a Buchner filter

with a medium grade filter paper. The residue both in the beaker and on the filter paper was

rinsed several times with deionized water to ensure all the chlorides had been collected. The

filtrates along with the rinse water were then transferred from the filter flask into a volumetric

flask. The solution in the volumetric flask was filled up with deionized water to 100 ml and 5 ml

solution was pipetted into a clean plastic cup in which 45 ml deionized water was added for

dilution to give a pH value of the solution between 1.0 and 2.0. The chloride content in the

diluted solution was analyzed photometrically by a Spectroquant® NOVA 60 photometer with a

chloride test kit. On the other hand, the solid residue left on the filter paper was dried in an oven

at 105 °C for 24 h. The cement content was then calculated according to previously recommended

methods [21-23] by subtracting the weight of insoluble substances from the initial weight of the

oven-dried mortar powder and correcting for hydrated water assuming 18% of the acid soluble

cementitious mass was hydration water, in addition to subtraction of 5% or 10% MHTs from the

total acid soluble mass where relevant.



6.3.1 The effect of MHTs on workability of fresh mortar

Figure 6.2 illustrates the workability results from the flow table test. Compared to the reference

mortar, the flow values of all mortar specimens admixed with MHTs are reduced to a certain

degree and the flow value decreases with increasing MHT content. Although EN 1015-3 [17]

does not specify a lower or upper limit flow value permitting an optimum mixing of fresh mortar,

however, the workability of mortar admixed with MHT-pAB and MHT-NO2 in the two used

dosages was recognized as on an acceptable level for real application when judged from on-site

mixing experience.

139

127

124 125

122

105

110

115

120

125

130

135

140

145

Ref. MHT-pAB(5%) MHT-pAB(10%) MHT-NO2(5%) MHT-NO2(10%)

Flo

w t

ab

le v

alu

e /

mm

Figure 6.2 Workability obtained from flow table test (in accordance to EN1015-3).

6.3.2 The effect of MHTs on mechanical properties

Strength, in particular, compressive strength and flexural strength are two important engineering

parameters to ensure the quality of concrete in its hardened state. According to EN-197 [15], the

standard compressive strength of a CEM I 42.5N mortar at 28 days is at least 42.5 MPa and hence

this could be considered as a lower limit to check the quality of cementitious materials made from this

type of cement. In addition, concrete with low porosity usually has high compressive strength and

high resistance to the penetration of aggressive ions, such as chlorides. Therefore, some

96 | Chapter 6

researchers link compressive strength and diffusion coefficient data to the porosity of the concrete

in order to globally assess the effect of a certain material on penetration of chloride. On the other

hand, although compressive strength is mainly used as an important criterion by the concrete

industry for field quality assurance, a certain flexural strength is also needed to meet the

prescriptive requirement especially for the design of highway and airfield slabs whose quality

assurance procedures rely primarily on flexural testing [24, 25]. In this study, the effect of the

partial replacement by MHTs for cement on the development of compressive strength and

flexural strength was studied.

The results of the compressive strength and flexural strength at three test ages (i.e., 3d, 7d

and 28d) are shown respectively in Figures 6.3 & 6.4, where each value was averaged from the

results of three individual tests. As shown in Figures 6.3 & 6.4, compared to the reference

specimens (i.e., Ref.), the incorporation of MHTs results in observable strength reductions at all

test ages for both types of strength. Specifically, for compressive strength, the maximum

reduction of 10% at 3 days is from MHT-pAB (5%), 15.6% at 7 days from MHT-pAB (10%) and

17.2% at 28 days from MHT-pAB (10%); for flexural strength, the maximum reduction of 19.3%

at 3 days from MHT-pAB (10%), 20% at 7 days from MHT-NO2 (10%) and 21% at 28 days from

MHT-pAB (10%). However, it has to stress that all of the compressive strength values at 28 days

are higher than 42.5 MPa, a limit value specified in EN-197 [15] for such type of cement and all

27.9

42.2

52.8

25.1

38.1

47.5

25.4

35.6

43.7

26.5

38.9

47.5

25.6

36.5

45.3

0

10

20

30

40

50

60

3-day 7-day 28-day

Co

mp

ress

ive s

tren

gth

(M

Pa

)

Curing age of the mortar specimen

Ref.

MHT-pAB(5%)

MHT-pAB(10%)

MHT-NO2(5%)

MHT-NO2(10%)

Figure 6.3 Compressive strength of the mortar specimens with/without MHTs (measured in accordance with

EN196-1; the bars are in the order left to right as presented in the key).


of the flexural strength values are well in excess of 4.5 MPa, a minimum mean 28-day flexural

strength (third-point loading) for paving recommended by ACI Committee 325 [26]. Concrete is

inherently considered to be a brittle material even though it exhibits some minor ductile

properties and there is no quantitative method to express the brittleness of concrete. In this regard,

it would be expected that compressive strength and flexural strength are closely related, i.e., as

the compressive strength increases, the flexural strength increases as well but at a decreasing rate.

Figure 6.5 is a plot of the development of flexural strength against compressive strength at the

three test ages on a log-log scale. Although the ratio of the two types strength depends on a

number of mixture-specific factors [24], a similarly linear relationship is suggested for all the

mortar mixes with/without MHTs. This is generally in agreement with the reported linear

relationship between the two types of strength [24, 25, 27]. Therefore, summarizing all the results

from the strength test, one may conclude that no remarkably negative effect on the strength

development would result from the incorporation of MHTs in the mortar. Furthermore,

incorporation of MHTs meets or exceeds certain design expectations and is expected to provide

sufficient performance in terms of the mechanical properties. Among the samples admixed with

MHTs, the use of MHT-pAB at 5% dosage performed the best in general. Additionally, likely

due to the lower porosity (see Figure 6.6 and following discussion), the mortar samples admixed

with the two types of MHTs at 5% dosage presented higher compressive and flexural strength

values than those at 10% dosages at the three test ages.

5.7

7.5

8.9

5.5

6.8

8.1

4.6

6.57.0

5.3

6.3

7.4

5.0

6.0

7.2

0

1

2

3

4

5

6

7

8

9

10

3-day 7-day 28-day

Fle

xu

ral

stre

ng

th (

MP

a )

Curing age of the mortar specimen

Ref.

MHT-pAB(5%)

MHT-pAB(10%)

MHT-NO2(5%)

MHT-NO2(10%)

Figure 6.4 Flexural strength of the mortar specimens with/without MHTs (measured in accordance with

EN196-1; the bars are in the order left to right as presented in the key).

98 | Chapter 6

0.5

0.6

0.7

0.8

0.9

1

1.0 1.5 2.0

Lo

g F

lex

ura

l S

tre

ng

th

Log Compressive Strength

Ref.

MHT-pAB(5%)

MHT-pAB(10%)

MHT-NO2(5%)

MHT-NO2(10%)

Figure 6.5 logarithmic relationships between the flexural strength and the compressive strength of the

mortar specimens with/without MHTs at the three test ages (lower point value is from the results of 3d

strength and upper point value is from 28d strength for all the curves).

6.3.3 The effect of MHTs on porosity

The MIP test has been performed on mortar samples at curing ages of 3, 7 and 28 days. The

Washburn equation as shown below is used to calculate the diameter of pores intruded by

mercury at each pressure step.

4 cos

DP

(6.1)

where D is the pore diameter (m), is the surface tension of mercury (N/m), is the contact

angle between mercury and the solid materials and P is the applied pressure (Pa). The MIP test

procedure includes intrusion and extrusion. The surface tension value used here is 0.485 (N/m)

and the contact angle is 132° based on previous studies [28, 29]. According to the Washburn

equation, pore sizes ranging from about 350 µm to 0.007 µm can be detected. Figure 6.6 shows

the total porosity of all the mortar specimens in which each value is the average of two parallel

measurements. As can be seen from Figure 6.6, relative to the Ref. specimen, the total porosity of

the mortar is higher for all the three test ages due to the incorporation of the MHTs in both dosage

levels (5% and 10%) with the exception of MHT-NO2 (10%) whose porosity at 3 days, however,

presented a relatively lower value. The increased porosity indicated the incorporation of the two

types of MHTs made the mortar a little more porous which in turn may increase the risk for

chloride ion penetrating through the mortar matrix without considering the chloride binding


behavior. As also seen from Figures 6.3 & 6.4, the increased porosity has indeed led to some loss

of the strength. In addition, when the results from the strength and porosity test are linked, it is

found that the development of the strength and porosity of the mortar specimens admixed with

MHTs followed the same trend with the curing time (i.e., strength increased and porosity

decreased as hydration proceed) as those of reference specimens. This might indicate that the

incorporation of MHTs has no or minor influence on the cement hydration process. Furthermore,

one has to realize that differences in water-cement ratio between the specimens caused by the

partial replacement of cement by MHTs could be also responsible for certain parts of strength

reduction and changed porosity.


3d 17.2 17.8 18.5 18.2 16.1

7d 14.2 14.5 16.1 14.2 15.2

28d 12.4 13.2 13.7 13.0 13.9

0

2

4

6

8

10

12

14

16

18

20

Poro

sity

(%

)

Figure 6.6 Porosity of the mortar specimens admixed with/without MHTs by MIP.

6.3.4 The effect of MHTs on chloride penetration

6.3.4.1 RCM test

Once the tests were finished, all specimens were split axially into two pieces and sprayed with

0.1 M silver nitrate (AgNO3) solution immediately onto the freshly split mortar surface to

visualize the penetration depth of chloride. The average penetration depth of chloride was

recorded accordingly. The non-steady-state migration coefficient (DRCM) was then calculated by

the following equation (Eq. 6.2) proposed by NT Build 492 [18].

100 | Chapter 6

(273 )0.0239(273 )

0.0238( 2) 2

dRCM d

T LxT LD x

U t U

(6.2)

where DRCM is the non-steady-state migration coefficient, ×10-12

m2/s; Xd is the average of the

chloride penetration depths, mm; T is average of the initial and final temperatures in the anolyte

solution °C; U is the applied voltage V; L is the thickness of the specimen, mm; t is the test

duration, h.

The chloride migration coefficient determined by the RCM test has been widely used as one

of the important parameters for resistance of cementitious materials to chloride penetration.

Figure 6.7 shows the RCM test results averaged from three test specimens combined with the

porosity results from MIP test at a curing age of 28 days. As can be seen from Figure 6.7, mortar

specimen of MHT-pAB(5%) presented the lowest RCM coefficient (DRCM) among all specimens

that were tested and on the other hand, when compared with the Ref. specimens presented a

relatively higher porosity. It is worth pointing out that the chloride transport and the

corresponding migration coefficient obtained in the RCM test are affected by the tortuosity factor

of the mortar but tortuosity on the other hand is a function of porosity [30, 31]. It has been found

that in many cases the tortuosity decreases with increasing porosity [32, 33]. In that sense, it

could be logically deduced that the mortar specimens with higher porosity would normally give a

higher value of diffusion coefficient if no active chloride binding could occur. This deduction is

further justified by recent work of Andrade and Bujak [34], in which they claimed that the reverse

Figure 6.7 Chloride diffusion coefficient obtained from RCM test (bars and numbers) and porosity at 28

days obtained from MIP test (squares and line) for MHT’s replacing cement and varying w/c.

15.1 14.8

19.2

17.6

20.8

11

12

13

14

15

0

5

10

15

20

25


Po

rosi

ty a

t 2

8 d

ay

s (%

)

DR

CM

( ×

10

-12

m2/s

)

DRCM

Porosity


relationship between porosity and chloride migration coefficients (Dns) of mortars with mineral

additions confirms the important role of binding. Therefore, we may attribute the co-occurrence

of lower DRCM and higher porosity of mortar admixed with MHT-pAB(5%) to an increased

chloride binding in the matrix. The increased chloride binding in the MHT-pAB(5%) specimens

could be traced back to the high anionic exchange capacity of MHT which makes exchange of the

intercalated -pAB anions by the intruded chlorides easily achieved (see Eq. 2.2 in Chapter 2). For

the other MHT samples, the higher porosity may dominate the chloride migration process by the

externally imposed electrical field although ion exchange reactions probably have occurred as

well.

From the above discussion, it is realized that the replacement of cement by the two MHTs in

mortar could have two opposing effects on chloride transport: (1) increasing permeability; (2)

enhancing chloride binding capacity. The increase of permeability was likely owing to the

increased w/c ratio i.e., 0.53 for 5% replacement and 0.56 for 10% replacement which

consequently resulted in increased porosity, compared to the reference specimen with a lower w/c

of 0.50. In order to further verify the enhanced chloride binding capacity of MHT specimens, an

additional series of mortar specimens was prepared by addition of the two MHTs, keeping the

w/c ratio of 0.5 and cement content constant. After 28 days curing, they were subjected to RCM

test and the results are given in Figure 6.8. As can be seen, except for MHT-NO2(10% add.), the

RCM coefficients of mortar specimens with added MHT were reduced markedly compared to the

Ref, being 21.9%, the highest percentage of decrease for MHT-pAB(5% add.), 20.5% for MHT-

pAB(10% add.) and 6.6% for MHT-NO2(5% add.). In addition, when compared with the results

presented in Figure 6.7, it can be found that RCM coefficients of mortar specimens which were

prepared by addition of the two types of MHTs at both dosages were lower than those of mortars

in which MHT was incorporated as partial replacement of cement. Therefore, the combined

information obtained from the two series of RCM tests as shown in Figures 6.7 & 6.8 suggested

that the effect on chloride binding enhancement due to the incorporation by cement replacement

of the two specific MHTs may be partly compensated by the increased permeability resulting

from the increased w/c ratio. The compensation was found to be more manifest when a higher

amount of cement was replaced by MHT, which could be indicated by the higher reduction rate

of RCM coefficient being 37.5% (12.0 vs 19.2) for MHT-pAB(10%) with respect to 20.3% (11.8

vs 14.8) for MHT-pAB(5%) and 26.0% (15.4 vs 20.8) for MHT-NO2(10%) with respect to 19.9%

(14.1 vs 17.6) for MHT- NO2 (5%). The effect of MHT-NO2 in terms of chloride retaining

seemed more beneficial when incorporated by addition than by replacing cement. For the two

specific mixing dosages (Figures 6.7 & 6.8), only MHT-NO2 added at 5%, i.e., MHT-NO2 (5%

add.), showed lower RCM coefficients than Ref, whereas the mortar specimens with 10% MHT-

NO2 presented higher RCM coefficients than Ref. in both series of RCM tests (i.e., MHT was

incorporated by addition or by replacing cement). In addition, it was also found that MHT-pAB

(10%) presented higher RCM coefficients with respect to Ref. (Figure 6.7). These findings may

102 | Chapter 6

suggest that there is a threshold level for both MHT-NO2 and MHT-pAB above which they could

be ineffective or even detrimental in terms of hindering the chloride transport in mortar.

6.3.4.2 Natural diffusion test

The transport of chloride into concrete is usually described by Fick's second law (Eq. 6.3) for

unidirectional diffusion into a semi-infinite homogeneous medium. Fick's second law counts for a

non-steady-state condition under which the flow of ions is not constant and the ion concentration

changes with time. However, the diffusion coefficient is assumed to be constant.

2

2

c cD

t x

(6.3)

In this condition, the mathematical solution to this problem is the error function (Eq. 6.4):

( , ) ( ) ( )4

s s i

app

xC x t C C C erf

D t

(6.4)

where C(x, t) is the chloride concentration (mass%) at depth x and time t, Ci the initial chloride

concentration (mass%), Cs the chloride concentration at the surface (mass%), Dapp the non-

steady-state diffusion coefficient also known as the apparent chloride transport coefficient (m2/s),

x the distance from the exposed surface (m) and t the exposure time (s).

Figure 6.8 Chloride diffusion coefficients of mortar with addition of MHT-pAB and MHT-NO2 at 28 days

obtained from RCM test (all specimens with constant w/c ratio and cement content).

15.1

11.812.0

14.1

15.4

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

Ref. MHT-pAB(5% add.) MHT-pAB(10% add.) MHT-NO2(5% add.) MHT-NO2(10% add.)

DR

CM

(×1

0-1

2 m

2/S

)


After finishing the natural diffusion test, the values for surface chloride concentration (Cs)

and apparent chloride transport coefficient (Dapp) were determined by fitting the error function

equation to the measured chloride profile by applying a nonlinear regression analysis in

accordance with the least squares method. The measured initial chloride concentration Ci in Eq.

6.4 was almost equivalent to 0% relative to the mass of cement and thus was not taken account

into further calculations. The measured chloride profiles in which total chloride concentration has

been averaged from two specimens as well as the best fitting curves are shown in Figure 6.9 (the

statistical variation is minor and not shown in the figure). The fitted results of Cs and Dapp are

given in Table 6.1. It should be pointed out that since Dapp is coupled with Cs in the curve-fitting,

the value of Dapp alone may not reflect the actual resistance of the mortar to the chloride

penetration. Therefore, the synergy of Cs and Dapp should be taken into account in order to

facilitate the interpretation and comparison of the results. In addition, as mentioned above, Cs

along with Dapp has been assumed to be constant during the curve-fitting using Eq. 6.4. However,

it was observed in practice that the surface chloride concentration is also time-dependent in

submerged zone of marine concrete structures especially for short term immersion [35, 36]. On

the other hand, it is recognized that surface chloride content is one of the major parameters often

used to estimate the chloride binding capacity of cementitious materials and consequently for

service life prediction [35, 37, 38]. Besides the fitted value of Cs, the total chloride content in the

surface zone (i.e., C(0-1), the chloride content in the first mm from the exposed surface of mortar;

data shown in Table 6.1) was also included in our discussion as an expression of the chloride

binding effect of the two MHTs.

As can be observed from Figure 6.9, all the MHTs specimens except MHT-pAB (5%)

showed higher total chloride concentrations in the profiles of all the layers than those of reference

specimens indicating relatively weak resistance to chloride penetration. For MHT-pAB(5%), the

total chloride concentrations were higher over the first 6 mm depth, while beyond 6 mm they

were lower than those of the reference mortar indicating a stronger resistance to chloride

penetration. In addition, as can be seen from Table 1, all the MHT specimens exhibited higher

values of both Cs and C(0-1) compared to the Ref. specimen. The increased chloride

concentration of C(0-1) along with the increased Cs clearly suggested a higher chloride binding

capacity due to the incorporation of MHT. This is supported by mathematical modeling work of

Glass and Buenfeld [39], who showed that an increase in the binding capacity especially nearer to

the surface resulted in an increase in total surface chloride concentration. This is probably owing

to the fact that an increased binding capacity allows more bound chloride to accumulate in the

surface of the specimens, thereby increasing the total chloride content [38, 39]. For MHT

specimens, the total chloride concentrations in both Cs and C(0-1) increased with increasing

dosages for both types of MHT most likely due to more chloride has been bound in the outer

layers by a larger amount of MHTs (i.e.,10% relative to 5%). However, the higher porosity may

have dominated the chloride transport process in higher depths producing a higher value of Dapp

when higher dosages of MHTs were incorporated, in spite of any possible chloride exchange or

104 | Chapter 6

binding behavior that may proceed during the test. In addition, mortar admixed with MHT-pAB

presented higher surface chloride concentrations and lower Dapp values at the two respective

dosage levels than mortar admixed with MHT-NO2. Considering the relatively small differences

between the porosity of MHT-pAB and MHT-NO2 specimens at 28 days, this may suggest that

MHT-pAB had higher chloride exchange capacity and thereby more chloride was bound in MHT-

pAB specimens than in MHT-NO2 specimens. Nevertheless, the combined information from

Figure 6.9 and Table 6.1 confirms the higher chloride binding capacity of the mortar due to the

incorporation of the two MHTs, hence in MHT specimens more chlorides were bound at the

surface compared to reference specimens. In addition, MHT-pAB(5%) possessed a notably

improved chloride diffusion resistance compared to all the other specimens indicated by the

relatively higher Cs, C(0-1) as well as lower Dapp values.

0

1

2

3

4

5

6

0 2 4 6 8 10 12 14 16 18 20 22 24 26

Ch

lorid

e con

cen

trati

on

(m

ass%

of

cem

en

t)

Depth below exposed surface (mm)

Rfe.-fitting MHT-pAB(5%)-fitting

MHT-pAB(10%)-fitting MHT-NO2(5%)-fitting

MHT-NO2(10%)-fitting Rfe.

MHT-pAB(5%) MHT-pAB(10%)

MHT-NO2(5%) MHT-NO2(10%)

Figure 6.9 Chloride profiles after 68 days exposure to a 16.5% NaCl solution and the best fitting error

function curves.

Table 6.1 Total surface chloride concentration and apparent chloride diffusion coefficient obtained by

fitting the error function equation to the measured chloride profiles and the measured total chloride content

at the depth of 0-1mm distance from the exposed surface.

Sample ID Ref. MHT-pAB(5%) MHT-pAB(10%) MHT-NO2(5%) MHT-NO2(10%)

Cs (mass% of cement) 4.7 6.0 6.3 5.3 5.5

C(0-1) (mass% of cement) 4.4 6.7 7.1 5.9 6.8

Dapp (×10-12 m2/s) 8.6 4.8 15.5 11.3 19.5


6.4 Conclusion

In this Chapter, two types of modified hydrotalcites (MHT-pAB and MHT-NO2) were

incorporated into cement mortars with two dosage levels replacing 5% and 10% by mass of

cement. The mortar samples were prepared with a constant (cement + MHT) content, a constant

water-to-(cement + MHT) mass ratio of 0.50 and a constant sand-to-(cement + MHT) mass ratio

of 3. The mortar properties such as workability, compressive and flexural strength, porosity, and

its rapid chloride migration and natural diffusion coefficients were investigated to shed light on

the effect of the two types of MHTs on chloride penetration in cement mortars and to study the

possible correlations between these properties.

The workability, strength and porosity tests revealed that no remarkably negative influence

on the strength development of hardened mortar and the consistence of fresh mortar would result

from the incorporation of the two specific MHTs in mortar, although the porosity to a certain

degree was increased when compared to the reference specimen. The co-occurrence of lower

RCM coefficient and higher porosity of MHT-pAB(5%) mortar suggested increased binding and

a beneficial effect of MHT-pAB in hindering the chloride transport. In addition, the relatively

higher surface chloride concentration in Cs as well as lower value of Dapp obtained from natural

diffusion test supported the finding observed from RCM test that MHT-pAB(5%) specimen

possessed a notably improved chloride diffusion resistance compared to reference specimens.

Furthermore, mortar incorporated with MHT-pAB presented lower RCM coefficient and higher

chloride concentrations in both Cs and C (0-1) as well as lower value of Dapp at the two respective

dosage levels than those of MHT-NO2 specimen. Considering the relatively small differences in

28-days porosity between MHT-pAB and MHT-NO2 specimens at the two respective replacing

dosages, this may suggest that MHT-pAB had a higher chloride exchange capacity and thereby

more chloride was bound in MHT-pAB specimens than in MHT-NO2 specimens. Moreover, it

was found that RCM coefficients of mortar specimens which were prepared by addition of the

two types of MHTs at both dosages (at constant w/c ratio) were lower than those of mortars in

which MHT was incorporated as partial replacement of cement. This evidence suggested that the

effect on chloride binding enhancement due to the incorporation by cement replacement of the

two specific MHTs may be partly compensated by the increased permeability resulting from the

increased w/c ratio. The compensation was found to be more manifest when a higher amount of

cement (in particular 10% vs 5% in this research) was replaced by MHT.

Nonetheless, the increased chloride binding and the improved chloride diffusion resistance of

MHT-pAB(5%) specimen validated that MHT-pAB could be a promising alternative for chloride

scavenging in mortar when an appropriate mixing dosage is adopted. This would subsequently

results in slower chloride transport, thus delaying corrosion initiation of the reinforcing steel.

However, one may need to realize that there may exist a threshold level for both MHT-NO2 and

MHT-pAB above which they could be ineffective or even detrimental in terms of hindering the

chloride transport. In this sense, more research is needed to optimise the incorporation dosages of

106 | Chapter 6

MHT-pAB in order to maximize its beneficial effect on mitigating chloride-induced damage to

concrete structures.

References

[1] Tuutti K. Corrosion of steel in concrete. Stockholm: Cement and Concrete Research Institute; 1982.






2001.


1996;10(5):391-402.

[6] Leroux F, Taviot-Gueho C. Fine tuning between organic and inorganic host structure: new trends in

layered double hydroxide hybrid assemblies. Journal of Materials Chemistry. 2005;15(35-36):3628-42.

[7] Khan AI, O’Hare D. Intercalation chemistry of layered double hydroxides: recent developments and

applications. Journal of Materials Chemistry. 2002;12(11):3191-8.

[8] Wang Y, Zhang D. Layered double hydroxides as a nanocontainer for encapsulating marine natural

product antifoulant: Intercalation and tunable controlled release of cinnamate. Materials Research

Bulletin. 2015;63:205-10.

[9] Dhir R, El-Mohr M, Dyer T. Chloride binding in GGBS concrete. Cement and Concrete Research.

1996;26(12):1767-73.

[10] Arya C, Xu Y. Effect of cement type on chloride binding and corrosion of steel in concrete. Cement

and Concrete Research. 1995;25(4):893-902.

[11] Wang S-D, Scrivener KL. Hydration products of alkali activated slag cement. Cement and Concrete

Research. 1995;25(3):561-71.

[12] Kayali O, Khan MSH, Sharfuddin Ahmed M. The role of hydrotalcite in chloride binding and

corrosion protection in concretes with ground granulated blast furnace slag. Cement and Concrete

Composites. 2012;34(8):936-45.

[13] Kayali O, Ahmed M, Khan M. Friedel’s salt and hydrotalcite–layered double hydroxides and the

protection against chloride induced corrosion. Civil and Environmental Research. 2013;5:111-7.

[14] Yang Z, Fischer H, Polder R. Modified hydrotalcites as a new emerging class of smart additive of

reinforced concrete for anticorrosion applications: A literature review. Materials and Corrosion.

2013;64(12):1066-74.

[15] EN197-1 Cement-Part 1: Composition, Specifications and Conformity Criteria for Common Cements.

European Committee for Standardization,2000.

[16] EN196-1 Methods for testing cement-Part 1: Determination of strength. European Committee for

Standardization,1994.

[17] EN 1015-3 Methods of test for mortar for masonry- Part 3: Determination of consistence of fresh

mortar (by flow table). European Committee for Standardization,1999.

[18] NTBuild492. Concrete, mortar and cement-based repair materials: Chloride migration coefficient

from non-steady-state migration experiments. NordTest, Espoo. 1999.

[19] NTBuild443. Concrete hardened: accelerated chloride penetration. NordTest, Espoo. 1995.


[20] EN14629 Products and systems for the protection and repair of concrete structures-Test methods-

determination of chloride content in hardened concrete. European Committee for

Standardization,2007.

[21] Castellote M, Andrade C. Round-Robin test on chloride analysis in concrete—Part I: Analysis of total

chloride content. Materials and Structures. 2001;34(9):532-49.

[22] Glass GK, Buenfeld NR. The determination of chloride binding relationships. In: Nilsson L, Ollivier J,

editors. 1st RILEM International Workshop on Chloride Penetration into Concrete1995.

[23] Gulikers JJW, Polder RB, Vries Jd. Recommendation for determining the chloride content in

hardened cement concrete (in Dutch). RWS Bouwspeurwerkrapport BSW 96-01: Ministry of

Transport; 1996.


[25] Lane DS. Evaluation of concrete characteristics for rigid pavements. VTRC 98-R24. Charlottesville:

Virginia transportation research council; 1998.

[26] ACI Committee 325. Guide for construction of concrete pavements and concrete bases. ACI 3259R-

91. Farmington Hills, Mich: American concrete institute; 1991.

[27] Raphael JM. Tensile strength of concrete. ACI Journal Proceedings: ACI; 1984.

[28] Cook RA, Hover KC. Experiments on the contact angle between mercury and hardened cement paste.


[29] Ellison AH, Klemm R, Schwartz AM, Grubb L, Petrash DA. Contact angles of mercury on various

surfaces and the effect of temperature. Journal of Chemical and Engineering Data. 1967;12(4):607-9.

[30] Samson E, Marchand J, Snyder K. Calculation of ionic diffusion coefficients on the basis of

migration test results. Materials and Structures. 2003;36(3):156-65.

[31] Ahmad S, Azad AK, Loughlin KF. Effect of the key mixture parameters on tortuosity and

permeability of concrete. Journal of Advanced Concrete Technology. 2012;10(3):86-94.

[32] Li L, Page C. Finite element modelling of chloride removal from concrete by an electrochemical

method. Corrosion Science. 2000;42(12):2145-65.

[33] Sun G, Zhang Y, Sun W, Liu Z, Wang C. Multi-scale prediction of the effective chloride diffusion

coefficient of concrete. Construction and Building Materials. 2011;25(10):3820-31.

[34] Andrade C, Buják R. Effects of some mineral additions to Portland cement on reinforcement

corrosion. Cement and Concrete Research. 2013;53:59-67.

[35] Tang L. CHLORTEST–Resistance of concrete to chloride ingress-From laboratory tests to in-field

performance-Guideline for practical Use of Methods for Testing the Resistance of Concrete to

Chloride Ingress. EU Funded Research Project under 5FP GROWTH Programme,2005.

[36] Nilsson L-O. Prediction models for chloride ingress and corrosion initiation in concrete structures.

Chalmers university of technology, Building Materials; 2001.

[37] Ehlen MA, Thomas MD, Bentz EC. Life-365 Service Life Prediction Model™ Version 2.0. Concrete

International. 2009;31(5):41-6.

[38] Ann K, Ahn J, Ryou J. The importance of chloride content at the concrete surface in assessing the

time to corrosion of steel in concrete structures. Construction and Building Materials. 2009;23(1):239-

45.

[39] Glass G, Buenfeld N. The influence of chloride binding on the chloride induced corrosion risk in

reinforced concrete. Corrosion Science. 2000;42(2):329-44.

Chapter 7

The Anti-corrosion Performance of

Two Types of Modified Hydrotalcites

in Cement Mortar with Embedded

Steel

110 | Chapter 7

7.1 Introduction

In this chapter, research is reported that was carried out to investigate the anti-corrosion

performance of two selected MHTs, i.e., Mg(2)Al-pAB and Mg(2)Al-NO2, in cement mortar with

embedded steel. To do this, three test methods based on chloride exposure and OCP/LPR

measurements were adopted:

1. An accelerated chloride migration test

2. A wetting-drying cyclic test

3. A natural diffusion test

The primary objective of this chapter is to give a picture of the effect of the two MHTs on critical

chloride concentration (also known as chloride threshold, CT) responsible for corrosion initiation

of the reinforcing steel, on chloride penetration in mortar and their possible correlations.

Moreover, it is expected to validate the promising use of novel MHT composites with selected

intercalated inhibitive anions as a new type of smart additive for reinforced concrete to combat

chloride-induced corrosion.

7.2 Experimental

7.2.1 Materials

CEM I 42.5N cement, CEN-Standard sand (particle size: 0-2 mm) and deionized water were used

for preparing mortar specimens. MHTs that were used are Mg(2)Al-pAB and Mg(2)Al-NO2

which are denoted as MHT-pAB and MHT-NO2 respectively. Hydrochloric acid (HCl) and

Urotropine from Sigma-Aldrich were used as received without further purification. A chemical

cleaner solution (1:1 diluted HCl + 3 g/l urotropine) was prepared by mixing 1000 ml H2O, 1000

ml HCl and 6 g urotropine. Reinforcing steel used in the experiment was low-carbon steel

(B500A) bars with a nominal diameter of 8 mm. Other relevant electrode materials are: AISI 304

type stainless steel mesh with a wire thickness of 0.5 mm and mesh width of 1.6 mm, platinized

titanium mesh with a wire thickness of 1.2 mm and mesh width of about 2.0 mm, copper plate

with a thickness of 0.8 mm, and REF401 type (Radiometer Analytical) Saturated Calomel

Electrode (SCE).

7.2.2 Sample preparation

7.2.2.1 Pre-treatment of the reinforcing steel bars

Reinforcing steel bars were cut into 120 mm long pieces from a longer bar. The newly cut ends

were slightly ground to get rid of sharp parts. A screw hole with a diameter of 3 mm was

subsequently drilled in one end of each bar for facilitating wire connection during the

electrochemical test (Figure 7.1). Before being embedded in mortar, the bars were pretreated to

The Anti-corrosion Performance of MHT in Cement Mortars with Embedded Steel | 111

remove any rust to get uniform surface condition. Typically, the bars were immersed in newly

prepared chemical cleaner solution for 5 min and then placed in ultrasonic bath for 2-3 min.

Afterwards, the bar surface was cleaned with a plastic bristle brush under tap water to remove

residual oxides. The bars were then degreased with acetone and rinsed with distilled water before

being dried using a hairdryer. Figure 7.1 shows a typical view of the steel bars before and after

the cleaning treatment.

Figure 7.1 Steel bars before and after cleaning (left); a screw hole with a diameter of 3 mm drilled in one

end of bar (right).

7.2.2.2 Preparation of reinforced mortar specimens

To investigate the anti-corrosion effect of MHT in mortar, MHT was incorporated in two

different ways: (1) as one of the mixing components in mortar to replace 5% or 10% cement mass

Figure 7.2 MHT as one of the mixing components in mortar: Mg(2)Al-pAB (MHT-pAB) and Mg(2)Al-

NO2 (MHT-NO2) used for mixing (left); MHT mixed with cement and sand in the bowl of a mixer (right).

112 | Chapter 7

(Figure 7.2); (2) as a surface coating of the reinforcing steel (Figure 7.3). For (1), the same

mixing design, mixing procedure and curing regime as described in Chapter 6 for preparing plain

mortar specimens was adopted (see section 6.2.2, sample preparation). The MHT’s effect on the

properties of fresh and hardened mortar were tested and can be found in Chapter 6. For (2), a

cement paste was pre-mixed with water-to-(cement + MHT) mass ratio of 0.4 and MHT to

replace 20% cement mass. Reference cement paste was also prepared without adding MHT. Then

the cement paste was uniformly applied over the surface of the steel bars with approx. 1.5-2.0

mm thickness. The coated bars were subsequently hung in a fog room and cured for 24 h to

achieve curing of the coating before being embedded into mortar.

Figure 7.3 Steel bar coated with a layer of cement paste admixed with MHTs (approx. 1.5-2.0 mm thick):

fresh state (left) and dry state after curing in a fog room for 24 h (right).

Figure 7.4 Schematic illustration of reinforced mortar specimen.

Mortar specimens were prisms of 40 × 40 × 110 mm3 in which a pretreated 120 mm long

steel bar was embedded at a cover depth of 10 mm. The bar was isolated from the air/mortar

interface by means of insulating heat shrink tubing (Farnell, Netherlands). The geometric length


and surface area of the steel bar that were finally exposed to mortar were 60 mm and 1560 mm2

respectively. Figure 7.4 schematically illustrates the reinforced mortar specimen. Eight groups of

reinforced mortar specimens were prepared in accordance with the above described two ways of

application:

1) Bar in mortar without addition of MHT (Ref.);

2) Bar coated with cement paste without addition of MHT and without addition in mortar

(Ref.(coating));

3) Bar in mortar with 5% replacement of MHT-pAB (MHT-pAB(5%));

4) Bar in mortar with 10% replacement of MHT-pAB (MHT-pAB(10%));

5) Bar coated with cement paste containing MHT-pAB and without addition in mortar

(MHT-pAB(coating));

6) Bar in mortar with 5% replacement of MHT-NO2 (MHT-NO2(5%));

7) Bar in mortar with 10% replacement of MHT-NO2 (MHT-NO2(10%));

8) Bar coated with cement paste containing MHT-NO2 without addition in mortar (MHT-

NO2(coating)).

After casting, the mortar prisms with the embedded bars were covered with plastic film and left

in the lab for 24 h (Figure 7.5). All the specimens were demolded on the next day and moved to a

fog room curing for 27 additional days. The specimens were taken out of the curing room at 28

days. After wiping off excess water from the surface, the specimens were left at room

temperature for 3 h to obtain a surface-dry condition and a PVC-pond (about 50 mm in depth)

was glued to the top surface of the specimens. All surfaces of specimen were carefully coated

with silicone sealant (Bison®

silicone kit) to prevent water evaporation during the test, except for

Figure 7.5 The configuration of the steel bars in the customized molds (left); reinforced mortar specimens

cast for corrosion test (right).

114 | Chapter 7

the top surface that would be exposed to sodium chloride solution and the bottom surface that

would be in contact with the electrode. When the silicone sealant had completely hardened after

about 24 h, the specimens were transferred to a plastic tray. Tap water was added into the tray in

order to keep the bottom surface of the specimens in a wet condition. The PVC-ponds of the

specimens were filled with a saturated Ca(OH)2 solution and covered by plastic film for 3 days

(Figure 7.6). Then the Ca(OH)2 solution was removed and the PVC-ponds were rinsed using tap

water to get rid of any precipitates before being filled by the test solution.

Figure 7.6 Prismatic reinforced mortar specimens with PVC-pond glued on the top surface and the

vertical surfaces coated with silicone sealant (left); ponds filled with saturated Ca(OH)2 solution in a

water-containing plastic tray (right).

7.2.3 Anti-corrosion performance evaluation

The anti-corrosion performance of two MHTs was evaluated by exposure of mortar specimens

with embedded bars to external chloride solution through three specific test methods based on

open circuit potential (OCP) and linear polarization resistance (LPR) measurements:

1. An accelerated chloride migration test

2. A wetting-drying cyclic test

3. A natural diffusion test

More technical details relevant to OCP and LPR measurements can be found in Chapter 3. Unless

otherwise stated, all potentials used in this chapter are referred to Saturated Calomel Electrode

(SCE). Figure 7.7 shows the custom designed experimental setup for the three series of tests

(specimens under accelerated migration test were connected to a power supply, while no power

supply connection for specimens under cyclic wetting-drying and natural diffusion test). The

specimens with ponds filled with chloride solution were set on a stiff plastic grid (about 20 mm


height) in a plastic tray. Between specimens and the plastic grid were sponge cloth and stainless

steel mesh. A sufficient amount of distilled water was poured into the plastic tray to maintain

moist condition of the sponge cloth throughout the tests. Once all the relevant electrodes were set

in place, the PVC-pond of the specimen was covered by parafilm to prevent evaporation of the

chloride solution. A more detailed description related to the experimental setup for each of the

three test methods will be discussed in the following sections.

Figure 7.7 Experimental setup for anti-corrosion performance evaluation of MHT (Note: specimens under

accelerated migration test were connected to a power supply; no power supply connection for specimens

under cyclic wetting-drying and natural diffusion test).

7.2.3.1 Accelerated chloride migration test under an external electrical potential

The principle of the accelerated chloride migration test is to apply an external electrical potential

across the mortar specimen forcing the chloride ions to migrate through the specimen. For this

test, a modified accelerated migration test setup based on Andrade’s integral corrosion test

method [1-4] was employed. Figure 7.8 schematically illustrates the electrode arrangement and

116 | Chapter 7

cross-sectional view of a reinforced mortar specimen in the accelerated chloride migration test.

The copper plate acting as cathode was submerged in chloride solution, while stainless steel mesh

acting as anode was placed underneath the specimen. The stainless steel mesh was also used as

the counter electrode (CE) for LPR measurement. A water-saturated sponge was sandwiched

between the stainless steel mesh and the specimen to ensure good electrical contact. A potential

drop of 6 V was applied by means of a power supply in order to accelerate the migration of the

chloride ions through the body of mortar towards the rebar. The voltage of 6 V was selected

based on the initial current flowing through the specimen. An initial current ranging from 1 to 10

mA was recognized as necessary for an optimal test condition [4]. A lower or higher voltage

would either possibly retard the test or raise the temperature and consequently vary the chloride

migration rate. The ponding solution was a mixed solution of 0.6 M NaCl and 0.4 M CuCl2 [4, 5].

The choice of CuCl2 solution as one of the chloride sources combined with the copper plate as

cathode (Cu/CuCl2) aims to avoid the production of OH- in the catholyte (ponding solution) and

thus prevent the negative effect of OH- on chloride migration. Normally, the main cathodic

reactions under the applied electrical field is the conversion of water into hydrogen gas and OH-

(Eq. 7.1), although the oxygen reduction may also occur (Eq. 7.2):

- -2 2

2H O + 2e H + 2OH (7.1)

- -2 2 O + 2H O + 4e 4OH (7.2)

The production of OH- in the catholyte would result in a decrease of the transference number of

Cl- in the mortar, which consequently slows down its penetration rate and even further possibly

influences the chloride threshold for initiating corrosion [2, 6]. When Cu/CuCl2 is used, the

copper reduction (Eq. 7.3) replaces OH- production as the main cathodic reaction due to the

higher reduction potential of copper relative to that of water [7].

2+ - Cu + 2e Cu (7.3)

In order to monitor the corrosion state of the embedded rebar, OCP (i.e., corrosion potential,

Ecorr) and LPR measurements were carried out using Solartron analytical SI1287 potentiostat

coupled with an eight-channel 1281 multiplexer. Due to the large amount of testing specimens,

the eight-channel multiplexer was expanded to 64 channels to achieve a more flexible and

automatic measurement by a custom designed computer program. As can be seen from Figure

7.8, a two reference-electrode system was adopted for OCP measurement in this specific

experimental setup. RE1, i.e., Saturated Calomel Electrode (SCE) was used for manually

measuring the OCP, while the platinized titanium mesh (RE2) was used as another reference

electrode for automatically measuring the OCP through the custom designed computer program.


Figure 7.8 Schematic illustration and cross-sectional view of the experimental setup for accelerated

chloride migration test. WE (working electrode): Reinforcing steel bar; CE (counter electrode): Stainless

steel mesh and also used as the Anode with copper plate as the Cathode; RE1 (reference electrode 1):

Saturated calomel electrode; RE2 (Reference electrode 2): Platinized titanium mesh. Once corrosion was

detected, a slice with 30 mm width was cut off from the left side of prism for chloride analysis.

118 | Chapter 7

When a certain voltage is applied to a reinforced mortar/concrete specimen, polarization of

the reinforcing steel occurs. This polarization may in turn cause the passive film that formed on

the reinforcing steel to become unstable and vulnerable to be destroyed upon the arrival of

aggressive species (e.g. chloride ions). In order to avoid/minimize such a polarization effect,

some pretests were performed by applying 6 volts potential to a reinforced mortar specimen using

a power supply. Figure 7.9 gives a typical potential evolution curve of the rebar by switching

on/off the power supply. As can be seen, when the power supply is switched on, the potential

increases immediately and the rebar is polarized. However, when the power supply is switched

off, the potential of the rebar drops back and recovers the initial value in about 45-90 min. This

pretest suggested that the OCP and LPR measurements could be done in a near-natural condition

with limited polarization effects after switching off the power supply for 45-90 min.

Figure 7.9 Typical potential evolution of the rebar embedded in mortar by switching on/off the power

supply at six volts.

In the real test, the OCP and LPR were measured after first switching off the power supply

and waiting for 45-90 minutes. The OCP was manually recorded via SCE, two or three times per

day after a 45-90 min waiting period as well as automatically recorded via RE2 (i.e., platinized

titanium mesh) three time per day exactly after waiting for 90 min. A conventional three-

electrode electrochemical system was employed for LPR measurement with the steel bar as

working electrode (WE), stainless steel mesh as the counter electrode (CE) and platinized

titanium mesh (RE2) as the reference electrode. The LPR measurement was performed once per

day upon finishing the first OCP measurement after switching off the power supply for 90 min. It

was automatically controlled and recorded by the custom designed computer program.

As explored in Chapter 3, the corrosion current density obtained from LPR measurement is

actually an average value over the whole exposed steel surface area and the local current inside a


pit could be significantly higher. Consequently, the measured value depends strongly on the

number and the size of the pits on the steel surface and the method has limited sensitivity for

early (small) pits [8, 9]. The occurrence of the first pit may be missed by LPR measurement. To

avoid interpretation difficulties, an observable shift in corrosion potential (Ecorr) as well as

corrosion current density (icorr) over time were set as criteria to identify the moment of the

depassivation of the embedded steel bars. As a quantitative measure for shifting criteria, the

empirical boundary values of Ecorr = -350 mV (SCE) and icorr = 0.1 µA/cm2 were adopted [3, 10].

That is, if Ecorr (SCE, i.e., RE1 used as the reference electrode) is found to be more negative than

-350 mV and/or icorr higher than 0.1 µA/cm2, it would then be considered that corrosion has been

initiated and active corrosion is developing. The time-to-corrosion initiation obtained through

both criteria are reported. For OCP measurement using the platinized titanium mesh (Ti, RE2) as

the reference electrode, however, no well-established boundary value of Ecorr is applicable and the

results were viewed as additional information to OCPs measured by SCE.

Once corrosion initiation (i.e., depassivation) was detected, the specimens were disassembled

from the experimental setup. Then, different procedures were followed for MHT in bulk

application and in coating application. For specimens in which MHT was incorporated as one of

the mixing components (bulk), a slice with 30 mm width was cut off from the left side of mortar

prism (see Figure 7.8 and Figure 7.11 later). This 30 mm mortar slice was used for chloride

threshold (CT) analysis as will be discussed below in section 7.2.4. The newly cut surface of the

mortar prism was immediately sealed using Bison® silicone kit and the pond on the top of the

prism was re-glued in which chloride solution was refilled. Then, the mortar prism was placed

back in the experimental setup in order to continuously monitor the corrosion process after

depassivation without applying a potential until around 30 days. For specimens in which MHT

was applied as a surface coating to the rebar, two specimens were broken upon depassivation

from which CT at the rebar depth was analyzed (see section 7.2.4). The other three or four

specimens were left for continued monitoring of the corrosion process after depassivation without

applying a potential.

7.2.3.2 Cyclic wetting-drying and natural diffusion test

The cyclic wetting-drying test was performed by one day of wetting followed by six days drying

under normal laboratory conditions. The OCP and LPR measurements were conducted during the

wetting cycles. For the natural diffusion test, the OCP and LPR were measured on a weekly basis

and sometimes more or less frequently upon the progress of the corrosion state of the steel bars.

Figure 7.10 gives a schematic illustration of the electrode arrangement and cross-sectional view

of the reinforced mortar specimen under both the cyclic wetting-drying and natural diffusion test.

All abbreviations of the relevant electrodes shown in Figure 7.10 have the same meanings and

served for the same purpose as those shown in Figure 7.8 for the accelerated chloride migration

test. However, the power supply was obviously not needed for these two tests. The ponding

solution used for these two tests was 16.5% NaCl solution in accordance with NordTest method

NT Build 443 [11].

120 | Chapter 7

The time to corrosion initiation (depassivation) of the embedded rebar under the wetting-

drying cycles and natural diffusion condition was determined by the same criteria as used for the

accelerated migration test. Once depassivation was detected, the same handling procedure for

mortar specimen as in the accelerated migration test was implemented. The CT at rebar depth of

corroding specimens was subsequently analyzed in a similar fashion. In addition, chloride

profiles of non-corroding specimens from both wetting-drying and natural diffusion test were

analyzed after the 30 weeks test period (see section 7.2.4).

Figure 7.10 Schematic illustration and cross-sectional view of the experimental setup for cyclic wetting-

drying and natural diffusion test (all the abbreviations of the relevant electrodes have the same meaning

and served for the same purpose as those shown in Figure 7.8.


7.2.4 Chloride analysis

Mortar specimens for chloride analysis were divided into two groups: (1) corroding specimens;

(2) non-corroding specimens. For corroding specimens, in particular from accelerated migration

test and wetting-drying test (no corroding specimens found in natural diffusion test in 30 weeks,

see discussion below), once the depassivation was detected, the chloride content at rebar depth

was analyzed to investigate the effect of MHT on chloride threshold (CT). For non-corroding

specimens, in particular after 30 weeks wetting-drying test and natural diffusion test, the chloride

content at three depths (see sampling below) was analyzed and chloride profiles were obtained

accordingly.

Figure 7.11 Schematic illustration of sampling area for chloride threshold (CT) analysis in mortar prism

with an embedded steel bar: (A) MHT incorporated in the bulk mortar; (B) MHT incorporated in the

coating on the rebar.

122 | Chapter 7

7.2.4.1 Sampling

Depending on the way that MHT was applied, different sampling methods were employed. As

shown in Figure 7.11A, for specimens in which MHT was incorporated as one of the mixing

components (App. #1, bulk), a slice with 30 mm width was cut off from the mortar prism for CT

analysis (for corroding specimens) and chloride profiling analysis (for non-corroding specimens).

This 30 mm slice was thought to have the same chloride penetration depth as that in the main part

of the mortar prism and therefore was used to represent the main part of the prism for these

analyses [12]. In this way, the main part of the prism could be kept for continued monitoring the

corrosion process after corrosion initiation. For CT analysis (Figure 7.11A), a small amount of

mortar powder (about 2 g) at the rebar depth (7.5-12.5 mm) was collected by dry drilling

(BOSCH drill; diameter: 5 mm) from the 15 mm thick part of the 30 mm slice that was originally

close to the rebar. Considering that in the “coating” specimens the material composition of mortar

at the interfacial zone of the embedded steel bar (containing MHT) is different from the rest part

of the mortar (without MHT), the specimens in which MHTs were applied as coating of the rebar

(App. #2, coating) were broken upon depassivation and a small amount of mortar powder (about

2 g) was drilled directly from the interfacial zone at the rebar depth (7.5-12.5 mm) (Figure

7.11B).

To analyze the chloride profile of the non-corroding specimens, the same sampling methods

with respect to the 30 mm slice (App. #1, bulk) and the broken specimens (App. #2, coating)

were used, while mortar powders were collected at three depths (about 2 g for each) below the

exposed surface: 2-7 mm (top part), 7.5-12.5 mm (middle part) and 27.5-32.5 mm (bottom part).

7.2.4.2 Analysis

The total (acid-soluble) chloride content of the mortar powders were determined following the

procedure described in Chapter 6 (see section of 6.2.3.6).


7.3.1 Accelerated chloride migration test

The OCP evolution of steel rebar embedded in the abovementioned eight groups of mortar

obtained via SCE and Titanium reference electrode during the accelerated chloride migration test

are shown respectively in Figures 7.12 &7.13, where each value has been averaged from the

results of three to four parallel specimens. The time scale of the x-axis in the two figures is the

test running time which includes the “power on” time and the “waiting time” (i.e., the power

supply was switched off during OCP/LPR measurements) before depassivation was detected and

the continued time after depassivation in the later stage when the specimens were disconnected

from the power supply. As can be seen from the two figures, the OCP (i.e., Ecorr) evolution

obtained from the two different reference electrodes follows the same trend, which suggests that

the use of platinized titanium mesh as one of the reference electrodes is feasible in this research.


It is interesting to note that in the later stage of the test after the depassivation was detected and

after the specimen was disconnected from the power supply for about 2 to 5 days, the OCPs of all

the specimens, except for MHT-NO2(10%), started to increase and went on to become more

positive than the empirical passive-active “boundary” value of Ecorr = -350 mV. This observation

indicated that the steel bars could have been experiencing repassivation. In addition, the marked

drop of OCPs and the obvious value-regain before and after switching off the power supply when

depassivation had been detected may further suggest that the corrosion occurring in the early

stage of test by the accelerated chloride attack was likely very un-uniform. In consequence, the

steel bars may possibly be able to repassivate after the accelerated migration process. In the case

of MHT-NO2(10%), the OCP drop in the early stage may be caused by more severe corrosion

than the other specimens, so it could be very hard to repassivate in the later stage when the power

supply was switched off. The evolution of the corrosion current density (icorr) obtained by LPR

measurements as shown in Figure 7.14 further confirmed the observation from the OCP

measurements.

Figure 7.12 The OCP/Ecorr (average of 3-4 specimens) evolution obtained using SCE as reference electrode

for rebars embedded in mortar specimens in the accelerated chloride migration test (Note: the time scale of

the x-axis is the real test running time which includes the “power on” time and the “waiting time” in the

early stage before depassivation was detected and the continued time after depassivation in the later stage

when the specimens were disconnected from the power supply).

124 | Chapter 7

Figure 7.13 The OCP/Ecorr (average of 3-4 specimens) evolution obtained using the platinized titanium

mesh as reference electrode for rebars embedded in mortar specimens in the accelerated chloride

migration test (Note: the time scale of the x-axis is the real test running time which includes the “power

on” time and the “waiting time” in the early stage before depassivation was detected and the continued

time after depassivation in the later stage when the specimens were disconnected from the power supply).

Figure 7.14 The corrosion current density/icorr (average of 3-4 specimens) evolution for rebars embedded

in mortar specimens in the accelerated chloride migration test (Note: the time scale of the x-axis is the

real test running time which includes the “power on” time and the “waiting time” in the early stage before

the depassivation and later the “power off” time after depassivation had been detected).


After finishing the accelerated migration test, the mortar samples were broken and the

corrosion pits were verified by visual inspection. Figure 7.15 presents examples of pitting

corrosion found on the repassivated steel bar and the corroding bar from MHT-NO2(10%)

samples. More and severe corrosion pits were found on the steel bar from MHT-NO2(10%)

Figure 7.15 Examples of the pits found on the repassivated bars (Ref., top) and depasssivated bars

(MHT-NO2(10%), bottom) after the accelerated chloride migration test.

126 | Chapter 7

samples. As for the other steel bar around which clear rust was found embedded in the mortar

wall, the corrosion occurring in the pits was probably slowly vanishing due to repassivation after

switching off the power supply. Commonly, hydroxyl ions exhibit higher mobility than other ions

(in particular, Cl- in this case) in aqueous solution [13]. Therefore, it can be imagined that without

external forces (electrical potential in this case), for the same corrosion pit during the same time

range, more hydroxyl ions (from pore solution) than chloride ions can enter into the pit. The

higher mobility of hydroxyl ions also results in a reduction of the mobility of chloride ions at

these sites [2, 6, 14, 15]. Consequently, a high local alkalinity and high OH-/Cl

- ratio could be

reached in the pit and that may subsequently result in repassivation. In addition, for mortar

specimens incorporated with MHTs (either in the bulk or in the coating), the inhibitors (i.e.,-pAB

and -NO2) that were released from MHTs via the chloride exchange reaction could also enter into

the pit and contribute to the repassivation through their inhibiting nature. For MHT-NO2(10%),

the higher porosity (as tested in Chapter 6) may be the dominating factor controlling the chloride

migration and/or diffusion process and consequently caused the severe pitting which did not

allow for repassivation.

Figures 7.16 & 7.17 respectively give the representative curves of Ecorr and icorr evolution

over the first 120 “power on” hours (the “waiting time” during the test has been deducted and not

included in the x-axis) during which depassivation was detected (i.e., an observable shift of Ecorr

below -350 mV and/or icorr above 0.1 µA/cm2). Table 7.1 gives the average depassivation time of

the eight groups of mortar specimens obtained from OCP and LPR measurements.

Figure 7.16 The OCP/Ecorr evolution for rebars embedded in mortar specimens during the early stage of the

accelerated chloride migration test; the “waiting time/power off” has been deducted and is not included in

the time scale of the x-axis.


Table 7.1 Depassivation time (in hour) of the reinforcing steel in the accelerated chloride migration test

obtained from OCP and LPR measurements (numbers in the parentheses are statistical variation).

Sample group Ref. Ref.

(coating)

MHT-pAB

(5%)

MHT-pAB

(10%)

MHT-pAB

(coating)

MHT-NO2

(5%)

MHT-NO2

(10%)

MHT-NO2

(coating)

Depassivation time (h)

by OCP 72 (±3) 73 (±3) 87 (±3) 80 (±15) 100 (±10) 80 (±7) 65 (±8) 87 (±3)

Depassivation time (h)

by LPR 80 (±7) 78 (±8) 101 (±8) 78 (±8) 118 (±8) 90 (±8) 84 (±14) 92 (±8)

Generally, the chloride transport can be related to the combined effects of the porosity and of

the amount of cementitious phases as well as other specific mixing components (MHT, in this

case) in the mortar/concrete matrix which are available to react with chloride hindering its

ingress. The relationship of porosity at 28 days (data was taken from Chapter 6) and the

depssivation time (td) obtained from OCP and LPR measurements is given in Figure 7.18 in

which the porosity of mortar specimens with coated rebars was accounted to be the same as that

of reference specimens (i.e., Ref.). As can be seen from Figure 7.18, compared to Ref., MHT-

pAB(5%) and MHT-NO2(5%) presented a relatively higher td and higher porosity, while

Ref.(coating) gave almost the same values. On the other hand, both MHT-pAB(coating) and

MHT-NO2(coating) showed higher td values than Ref. while they have the same porosity. These

observation may indicate the important role of MHT’s active chloride binding capacity in

slowing down the chloride ingress. It has to be noted that both MHT-pAB(10%) and MHT-

NO2(10%) respectively exhibited lower values of td than MHT-pAB(5%) and MHT-NO2(5%) in

Figure 7.17 The corrosion current density/icorr evolution for rebars embedded in mortar specimens

during the early stage of the accelerated chloride migration test (Note: the “waiting time/power off” has

been deducted and is not included in the time scale of the x-axis).

128 | Chapter 7

spite of the fact that more chloride could be bound by the larger amount of MHT, which

consequently would release more inhibitor anions (i.e, -pAB and -NO2) via chloride exchange

reaction. In this regard, it seems that the inhibiting effect of the simultaneously released inhibitors

has less influence on td comparing to the impacts of porosity and chloride binding. A possible

reason for this could be ascribed to the fact that the released anionic inhibitors were also

migrating with the chloride ions under the externally imposed electrical field. However, the

released inhibitors may have more influence on the chloride threshold as discussed below.

Figure 7.19 gives chloride threshold (CT) values of the eight groups of reinforced mortar

specimens obtained from the accelerated chloride migration test. The values correspond to mean

values obtained from two or three specimens and are expressed in total chloride content by

weight of cement. As can be seen from Figure 7.19, the CT values were found to be in the range

of 0.93-1.55 % (by cement mass) being the highest for MHT-pAB(coating) and lowest for Ref.:

MHT-pAB(coating) > MHT-NO2(coating) > MHT-pAB(10%) > MHT-NO2(10%) > MHT-

pAB(5%) > MHT-NO2(5%) > Ref.(coating) > Ref.

Compared to the Ref, the CT of MHT specimens was increased to a higher level, being 66.7%,

the highest percentage of increase for MHT-pAB(coating) and 28% the lowest for MHT-

NO2(5%). Considering the proposed dual function working mechanism of MHT (see section

2.5.3 of Chapter 2), the increased CT level resulted from the incorporation of MHT could be

ascribed to the combination effect of active chloride binding behavior of MHT and the inhibiting

Figure 7.18 The depassivation time obtained from accelerated migration test (columns) and porosity at

28 days obtained from MIP test (squares and line).


effect of the simultaneously released inhibitors. In addition, if MHTs incorporated in the same

percentage (i.e. 5%, 10% in mixing and 20% in coating), specimens incorporated with MHT-pAB

exhibited higher CT values than those with MHT-NO2 likely owing to the higher chloride binding

capacity of the MHT-pAB, which in turn resulted in less amount of -NO2 released from MHT

compared to the -pAB. Therefore, less amount of bound chloride as well as less amount of

released inhibitor may be the reasons responsible for the lower CT in the case of MHT-NO2.

Moreover, regardless of the two application ways, a relatively higher mixing percentage of both

MHTs, that is to say 10% relative to 5% in mortar and also 20% in the coating relative to 10% in

mortar, was found to increase the CT as well. In addition to the higher chloride binding capacity

resulting from the higher mixing amount of MHT, the relatively higher concentration of

inhibitors (i.e, -pAB and -NO2) which were released from MHT was believed to have contributed

to raise the CT level. This further supported the important effect of MHT as an internal reservoir

for inhibitive anions for preventing chloride-induced corrosion in mortar or concrete. The

beneficial effect of using corrosion inhibitors in bulk concrete or steel coating in terms of the

increased CT level was also studied in the previous research [16-18]. For Ref.(coating), in which

the rebar was coated with a layer of pure cement paste, the formation of a more dense and cement

rich layer at the steel-paste interface relative to the steel-mortar interface of Ref. was believed to

be the main reason for the increase of CT [14, 19, 20], although the increase percentage of CT

being 7.5% (0.93 vs 1.00) was not as high as those of MHT specimens.

Figure 7.19 Measured chloride threshold values from the accelerated chloride migration test.

130 | Chapter 7

Furthermore, it was found that MHT-pAB(coating) and MHT-NO2(coating) exhibited higher

CT values as well as the longer depassivation time (td) relative to the other specimens (except for

MHT-pAB(5%), which presented higher td than MHT-NO2(coating)). On the contrary, MHT-

NO2(10%) exhibited a relatively higher CT than MHT-pAB(5%), MHT-NO2(5%), Ref.(coating)

and Ref., but the shortest td among all the specimens. On the other hand, when comparing

individually, it was found that the incorporation of a higher percentage (10% relative to 5% in

these cases) of either MHT-pAB or MHT-NO2 on one hand increased the corresponding CT

values, but on the other hand, decreased td values to a certain degree. Besides the porosity effect

(Figure 7.18), a lower td of the specimens incorporated with a higher percentage of the two MHTs

was likely due to a higher chloride migration coefficient and a higher w/c ratio as described in

Chapter 6.

Nevertheless, based on the results and discussion above, one may need to realize that the td

and the accompanying CT can be dependent on a number of factors determined by the application

methods and mixing dosage of MHT, as well as the properties of MHT such as the type and its

chloride binding capacity and inhibiting effect conferred by simultaneously released inhibitors. In

addition, the mortar/concrete properties such as porosity and microstructure, the condition of

interfacial zone in the close vicinity of the reinforcing steel and external environment including

temperature, moisture content, oxygen concentration and accessibility, and other artificial

impacts (electrical force in this case) could also have their effect on td and/or CT. Therefore, it

may be concluded that no definite relationship could be established between the CT and the td

suitable for all the specimens. However, the td and CT obtained from the test may be helpful in

practice to give a perspective on corrosion initiation for mortar or concrete incorporated with

MHT.

7.3.2 Cyclic wetting-drying test

The OCP and LPR measurements were performed every week during the wetting cycles and the

recorded evolution of Ecorr and icorr are illustrated respectively in Figures 7.20 & 7.21. It is worth

pointing out that since the Ecorr evolution obtained from the platinized Ti reference electrode

followed the same trend as that of SCE, as observed in the accelerated chloride migration test

(see Figures 7.12 & 7.13), the Ecorr evolution results via platinized Ti reference electrode was not

included in the following discussion.


Figure 7.20 The OCP/Ecorr evolution for rebars embedded in mortar specimens during 30 weekly cyclic

wetting-drying test.

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr

/mV

(vs

SC

E)

Time (week)

Ecorr /OCP-MHT-pAB (5%)

MHT-pAB(5%)-s1

MHT-pAB(5%)-s2

MHT-pAB(5%)-s3

-600

-500

-400

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr

/mV

(v

s S

CE

)

Time (week)

Ecorr /OCP-MHT-pAB (10%)

MHT-pAB(10%)-s1

MHT-pAB(10%)-s2

MHT-pAB(10%)-s3

-500

-400

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr

/mV

(vs

SC

E)

Time (week)

Ecorr /OCP-Ref.

Ref.-s1

Ref.-s2

Ref.-s3

-700

-600

-500

-400

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr

/mV

(v

s S

CE

)

Time (week)

Ecorr /OCP-Ref.(coating)

Ref.(coating)-s1

Ref.(coating)-s2

Ref.(coating)-s3

Ref.(coating)-s4

-600

-500

-400

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr /m

V (

vs

SC

E)

Time (week)

Ecorr /OCP-MHT-NO2 (5%)

MHT-NO2(5%)-s1

MHT-NO2(5%)-s2

MHT-NO2(5%)-s3

MHT-NO2(5%)-s4

-600

-500

-400

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr

/m

V (

vs

SC

E)

Time (week)

Ecorr /OCP-MHT-NO2 (10%)

MHT-NO2(10%)-s1

MHT-NO2(10%)-s2

MHT-NO2(10%)-s3

MHT-NO2(10%)-s4

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr /

mV

(v

s S

CE

)

Time (week)

Ecorr/OCP-MHT-NO2 (coating)

MHT-NO2(coating)-s1

MHT-NO2(coating)-s1

MHT-NO2(coating)-s1

-300

-200

-100

0

0 5 10 15 20 25 30

Eco

rr /m

V (

vs

SC

E)

Time (week)

Ecorr /OCP-MHT-pAB (coating)

MHT-pAB (coating)-s1



132 | Chapter 7

Figure 7.21 The corrosion current density/icorr evolution for rebars embedded in mortar specimens during

30 weekly cyclic wetting-drying test.

0

0.1

0.2

0.3

0.4

0.5

0.6

0 5 10 15 20 25 30

i co

rr/µ

A.c

m-2

Time (week)

icorr - Ref.

Ref.-s1

Ref.-s2

Ref.-s3

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

i co

rr

/µA

.cm

-2

Time (week)

icorr - Ref.(coating)

Ref.(coating)-s1

Ref.(coating)-s2

Ref.(coating)-s3

Ref.(coating)-s4

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

i co

rr

/µA

.cm

-2

Time (week)

icorr - MHT-pAB (5%)

MHT-pAB(5%)-s1

MHT-pAB(5%)-s2

MHT-pAB(5%)-s3

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

i co

rr /µ

A.c

m-2

Time (week)

icorr - MHT-pAB (10%)

MHT-pAB(10%)-s1

MHT-pAB(10%)-s2

MHT-pAB(10%)-s3

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 5 10 15 20 25 30

i co

rr/µ

A.c

m-2

Time (week)

icorr - MHT-NO2 (5%)

MHT-NO2(5%)-s1

MHT-NO2(5%)-s2

MHT-NO2(5%)-s3

MHT-NO2(5%)-s4

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 5 10 15 20 25 30

i co

rr /µ

A.c

m-2

Time (week)

icorr - MHT-NO2 (10%)

MHT-NO2(10%)-s1

MHT-NO2(10%)-s2

MHT-NO2(10%)-s3

MHT-NO2(10%)-s4

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

i co

rr /µ

A.c

m-2

Time (week)

icorr - MHT-pAB (coating)

MHT-pAB(coating)-s1

MHT-pAB(coating)-s2

MHT-pAB(coating)-s3

0

0.1

0.2

0.3

0.4

0.5

0 5 10 15 20 25 30

i co

rr/µ

A.c

m-2

Time (week)

icorr - MHT-NO2 (coating)

MHT-NO2(coating)-s1

MHT-NO2(coating)-s2

MHT-NO2(coating)-s3


As can be seen from Figures 7.20 & 7.21, only few of the rebars embedded in mortar

specimens of Ref., Ref.(coating), MHT-pAB(10%), MHT-NO2(5%) and MHT-NO2(10%)

showed active corrosion (Ecorr ˂ -350 mV and icorr > 0.1 µA/cm2) during the 30 weekly wetting-

drying cycles. Specifically, the depassivation had been detected at the 26th week for one out of

the three Ref. specimens (Ref.-s3), at the 27th week for two out of five Ref.(coating) specimens

(Ref.(coating)-s2 and the other one was broken for CT analysis, so data was not shown in the

figure), at the 22nd week for one out of three MHT-pAB(10%) specimens (MHT-pAB(10%)-s2),

at the 29th week for one out of the four MHT-NO2(5%) specimens (MHT-NO2(5%)-s4) and at

the 18th and 20th week for two out of the four MHT-NO2(10%) specimens (MHT-NO2(10%)-s1

and MHT-NO2(10%)-s2, respectively). On the other hand, all of the rebars embedded in MHT-

pAB(5%), MHT-pAB(coating) and MHT-NO2(coating) mortar specimens maintained their

passive state and no active corrosion was detected yet after 30 weekly wetting-drying cycles.

Table 7.2 gives an overview of the time to depassivation and probability of corrosion of the steel

bars embedded in the eight groups of reinforced mortars during 30 weekly wetting-drying cycles.

The probability of corrosion is the percentage of corroding specimens out of total number of

specimens of each group. Obviously, MHT-pAB(5%), MHT-pAB(coating) and MHT-

NO2(coating) performed best among all the specimens on the basis of both depassivation time

and probability of corrosion.

Table 7.2 Depassivation time and probability of corrosion of the steel bars embedded in mortar specimens

during 30 weekly wetting-drying cycles (“-” means no depassivation detected).


(coating)

MHT-pAB

(5%)

MHT-pAB

(10%)

MHT-pAB

(coating)

MHT-NO2

(5%)

MHT-NO2

(10%)

MHT-NO2

(coating)

Depassivation time

(week) 26 27 - 22 - 29 18, 20 -

Probability of corrosion

(%) 33 40 0 33 0 25 50 0

The following order of the time to depassivation can be established in terms of the first

appearance of active corrosion (Table 7.2):

MHT-NO2(10%) < MHT-pAB(10%) < Ref. ≤ Ref.(coating) < MHT-NO2(5%) < MHT- pAB(5%)

~ MHT- NO2(coating) ~ MHT-pAB(coating) (no corrosion detected)

In general, the order of the time to depassivation established from the wetting-drying test is

consistent with that obtained from accelerated chloride migration test except for MHT-pAB(10%)

which performed similarly as Ref. and Ref.(coating) in the accelerated migration test (Table 7.1).

Although the general anti-corrosion behavior can be explained in terms of the combined

importance of the porosity, chloride binding capacity as well as the inhibiting effect from the

released inhibitors of MHT, it has to be emphasized that the order listed above is only based on

the first appearance of active corrosion and not for all specimens which were involved in the test.

That is, in other words, if all the specimens would have been found to be corroding, then the

134 | Chapter 7

statistical sense of time to depassivation may also need to be taken into account, which might

influence the results.

Measured chloride threshold (CT) values for corroding specimens at their respective

depassivation time during the 30 weekly wetting-drying cycles are presented in Figure 7.22. As

found in the accelerated chloride migration test, the incorporation of MHT increased the CT level

independent of whether the corrosion was detected earlier (in the cases of MHT-NO2(10%) and

MHT-pAB(10%)) or later (in the case of MHT-NO2(5%)) than Ref. In particular, the CT values

of MHT-pAB(10%), MHT-NO2(10%) and MHT-NO2(5%) were increased respectively by 67.4%

(3.18 vs 1.90), 64.2% (3.12 vs 1.90), and 36.4% (2.59 vs 1.90) relative to Ref. In addition, the CT

was also found increased due to the application of pure cement paste coating, i.e., Ref(coating),

but the percentage of increase being 14.7% (1.90 vs 2.18) relative to Ref. is less than those of

MHT specimens.

Figure 7.22 Measured chloride threshold values for corroding specimens at the depassivation time during

wetting-drying cyclic test.

Chloride penetration profiles of the non-corroding specimens after 30 weekly cyclic wetting-

drying cycles were analyzed and the results are shown in Figure 7.23 in which each value

corresponds to the mean value obtained from two or three specimens. All the specimens were

broken after 30 weeks for visual inspection in order to confirm the occurrence of corrosion.

Figure 7.24 shows an example of pitting corrosion detected in wetting-drying test.


Figure 7.23 Measured chloride profiles (average of 2-3 specimen) in 16.5% NaCl solution after 30 weekly

wetting-drying cycles.

Figure 7.24 An example of pits (from Ref.) developed during 30 weekly cyclic wetting-drying test.

136 | Chapter 7

As a further indicator of the relative rates of chloride ingress into the eight groups of mortar

specimens, the measured chloride profiles have been fitted to Eq.7.4 to yield surface chloride

contents (Cs) and non-steady-state diffusion coefficients (also known as the apparent chloride

transport coefficient, Dapp) from the error function solution of Fick’s second law, which applies to

unidirectional diffusion into a semi-infinite homogeneous medium of constant Dapp,

( , ) ( ) ( )4

s s i

app

xC x t C C C erf

D t

(7.4)

where C (x, t) is the chloride concentration at depth x below the exposed surface after time t. Ci is

the initial chloride concentration (mass%) and counted as 0% as described in Chapter 6. Cs is

often used as one of the major parameters to estimate the chloride binding capacity of the

specimens with higher values implying higher chloride binding [21, 22]. Obviously, the

assumptions involved in applying this equation cannot be rigorously justified for specimens under

wetting-drying condition [12, 23]. In fact, the diffusion coefficients obtained from wetting-drying

cycles may be regarded as pseudo-diffusion coefficients. The resulting Cs and Dapp values are

presented in Table 7.3. As can be seen from Table 7.3, compared to Ref., all specimens with

MHT in the mortar matrix exhibited higher Cs values, which confirms the important role of MHT

in terms of higher chloride binding capacity. The calculated diffusion coefficient was found to be

the lowest for MHT-pAB (5%) and the highest for MHT-NO2(10%). The coating specimens, as

expected, showed similar values of both Cs and Dapp as that of Ref., presumably due to the same

material composition in bulk mortar, hence, almost the same pathway for chloride transport

before arriving to the rebar depth. In this regard, the coating seems to have more influence on

raising the chloride threshold as discussed in the accelerated migration test. Nonetheless,

although assuming diffusion transport for wetting-drying is theoretically incorrect, the resulting

Cs and Dapp seem to be realistic and meaningful.

Table 7.3 Surface chloride content and chloride diffusion coefficient obtained by fitting the error function

equation (Eq.7.4) to the measured chloride profiles after 30 weekly wetting-drying cycles.


(coating)

MHT-pAB

(5%)

MHT-pAB

(10%)

MHT-pAB

(coating)

MHT-NO2

(5%)

MHT-NO2

(10%)

MHT-NO2

(coating)

Cs ( % by mass of cement) 5.8 5.8 8.6 7.6 5.8 7.3 7.0 5.8

Dapp (×10-12

m2/s) 3.3 3.2 1.8 4.5 3.3 3.0 4.8 3.2

7.3.3 Natural diffusion test

The Ecorr and icorr evolution of the rebars embedded in mortar specimens undergoing the natural

diffusion test are given respectively in Figures 7.25 & 7.26. Each value of both Ecorr and icorr

shown in the two figures has been averaged from the results of three or four parallel specimens.

Similarly, as observed in the accelerated migration test as well as in the wetting-drying cyclic

test, the Ecorr evolution recorded via the platinized Ti reference electrode follows the same trend


as that of SCE and is not shown. As can be seen from Figures 7.25 & 7.26, during the 30 weeks

test period, no active corrosion (Ecorr ˂ -350 mV and icorr > 0.1 µA/cm2) had been detected in any

of the specimens.

The average chloride penetration profiles of the eight groups of mortar specimens after 30

weeks natural diffusion are shown in Figure 7.27. Table 7.4 gives the results of surface chloride

contents (Cs) and apparent diffusion coefficient (Dapp) that were derived by fitting the error

function equation, Eq.7.4 to the measured chloride profiles. It is interesting to find that although

the experimental condition of natural diffusion is different from that of wetting-drying, the former

representing true diffusion conditions, the resulting chloride profiles as well as the Cs and Dapp

values calculated from the profiles followed the same pattern. In addition, for all specimens after

the 30 weeks test, the Cs and Dapp obtained from natural diffusion exhibited lower values than

those obtained from wetting-drying cycles, but by only about 10%. It has to be noticed that MHT-

pAB(5%) in both natural diffusion and wetting-drying test presented the highest Cs value along

with the lowest Dapp value relative to all other specimens indicating a more active chloride

binding and a stronger resistance to chloride penetration as well. More details about the

comparison of the three test methods (i.e., migration, cyclic wetting-drying and natural diffusion)

are discussed in the following section.

Figure 7.25 The OCP/Ecorr evolution for rebars embedded in mortar specimens during 30 weeks natural

diffusion test.

138 | Chapter 7

Figure 7.26 The corrosion current density/icorr evolution for rebars embedded in mortar specimens during

30 weeks natural diffusion test.

Figure 7.27 Measured chloride profiles in 16.5% NaCl solution after 30 weeks natural diffusion test.


Table 7.4 Surface chloride content and chloride diffusion coefficient obtained by fitting the error function

equation (Eq. 4) to the measured chloride profiles after 30 weeks natural diffusion test.


(coating)

MHT-pAB

(5%)

MHT-pAB

(10%)

MHT-pAB

(coating)

MHT-NO2

(5%)

MHT-NO2

(10%)

MHT-NO2

(coating)

Cs ( % by mass of cement) 5.3 5.3 8.1 7.0 5.4 6.8 6.5 5.4

Dapp (×10-12

m2/s) 3.0 3.0 1.6 4.0 3.0 2.8 4.4 3.0

7.3.4 The applied test methodology and the effect on chloride diffusion/migration

and chloride threshold

Based on the results and discussion above, the modified accelerated migration test is promising. It

is simple and can provide a useful means of classifying the effect of new corrosion mitigating

admixtures in a relatively short term. In addition to information obtained on chloride threshold

(CT) as well as the time to corrosion initiation, the corrosion process after corrosion initiation can

also be continuously monitored. Therefore, it gives more insight in terms of reinforcement

corrosion than testing only chloride penetration. Compared to the accelerated migration test, the

natural diffusion test, on the other hand, is closer to real-world scenarios in normal

circumstances, however, it is time consuming often requiring months or years to obtain the results

(no depassivation has been detected yet after 30 weeks in this research). Regarding the cyclic

wetting-drying test, which may be considered more natural than the accelerated migration test, it

turned out to be relatively slow since corrosion was only detected in few specimens during the 30

weeks testing period. It was found that there was good correspondence in terms of surface

chloride contents and diffusion coefficients between natural diffusion and wetting-drying test for

the non-corroding specimens with/without MHT (see Tables 7.3 & 7.4). The slightly higher

surface chloride contents and diffusion coefficients obtained from wetting-drying may be

ascribed to the different transport mechanisms of chloride ions in mortar related to these two

conditions. It may be worth pointing out that the rate of chloride transport by absorption

(especially in the case of wetting-drying) is greater than that of chloride transport by diffusion,

thus, more chloride enters into bulk mortar by absorption than by diffusion. This effect could be

more manifest in the near-surface zone of mortar. The consequence is that chloride tends to

accumulate in the surface zone during wetting-drying cycles, which in turn increases the surface

chloride content. On the other hand, during wetting, chloride solution penetrates a layer of

mortar; during drying, the evaporation front moves inward and takes some of the chlorides with

it. Additionally, chlorides may have been pushed inwards due to the carbonation which liberates

weakly bound chlorides as has been found elsewhere [12]. This may in turn have changed the

distribution of chloride ions and consequently increases the diffusion coefficient. Furthermore, it

has to be noted that the diffusion coefficient obtained from either wetting-drying or natural

diffusion test may possibly decrease with exposure/testing time because of factors such as

140 | Chapter 7

continuing cement hydration, chloride exchange reactions involved in MHT specimens and

precipitation of insoluble salts [24].

Bearing in mind that CT could be dependent on a number of parameters as discussed in

section 7.3.1, varying the testing method could also give a different CT value. Rearranging Figure

7.19 and Figure 7.22, comparing the influence of testing methods on CT (in terms of corroding

specimens), results into Figure 7.28, in which the chloride contents at rebar depth of the non-

corroding specimens from both wetting-drying and natural diffusion test after 30 weeks were also

included. For all corroding specimens, including the reference specimens and MHT specimens,

the CT values obtained from wetting-drying test were found to be more than twice as high as

those obtained from the accelerated migration test. In the accelerated migration test, some of the

negatively charged inhibitors ions (i.e., -pAB and -NO2) in the vicinity of rebar, which are

released via the chloride exchange reactions, could also migrate away from the rebar surface

under the electric field before getting interacted with the passive film. On the opposite, the

released inhibitors ions, which are located in the vicinity of the rebar in more natural conditions

(i.e., wetting-drying and natural diffusion), can be retained and have more chance and space

exerting their inhibiting effect to a larger degree. In this respect, the CT may be increased

markedly in conditions of wetting-drying and natural diffusion than the electrically accelerated

migration. On the other hand, one may realize that the higher rate of chloride transport as well as

released anionic inhibitor caused by the externally imposed electrical field may consequently

result in less binding [25, 26] and less inhibition than those of wetting-drying and natural

diffusion.

0.931.00

1.24 1.36

1.55

1.191.34

1.44

1.90

2.18

3.18

2.59

3.12

1.851.93

1.71

3.12

1.98

2.47

3.08

1.95

1.76 1.81

1.50

2.93

1.87

2.39

2.89

1.84

0

0.5

1

1.5

2

2.5

3

3.5

Ref. Ref.(coating) MHT-pAB(5%) MHT-pAB(10%) MHT-pAB (coating) MHT-NO2(5%) MHT-NO2(10%) MHT-NO2(coating)

Ch

lori

de

con

ten

t (m

ass

% o

f ce

men

t)

C_M C_W/D NC_W/D NC_D

Figure 7.28 Chloride contents at rebar depth of corroding specimens from migration test (C_M) and

cyclic wetting-drying test (C_W/D), and of non-corroding specimens from cyclic wetting-drying

(NC_W/D) and natural diffusion test (NC_D) after 30 weeks.


The chloride contents at rebar depth of the non-corroding specimens from both wetting-

drying and natural diffusion test after 30 weeks were also analyzed. As shown in Figure 7.28, for

the wetting-drying test, the chloride contents of the non-corroding specimens after 30 weeks test

were found to be lower than those of the corroding specimens (if any), but the differences in most

cases are not significant. In the cases of MHT-pAB(10%) and MHT-NO2(10%), some specimens,

however, presented a bit higher chloride contents than the CT values detected earlier from

corroding ones. This seems understandable since the non-corroding specimens experienced more

wetting-drying cycles and the longer testing time may account for the increase of the chloride

content at the same depth in mortar (i.e., rebar depth). On the other hand, it could also imply that

a possible variation of the CT may exist if all the specimens would have been finally corroding

and analyzed [27]. In terms of the natural diffusion (no corrosion detected yet), after 30 weeks

test, as expected, the chloride contents at the rebar depth in mortar specimens are lower than

those of non-corroding specimens from wetting-drying test, but the differences are not very

remarkable. In addition, the diffusion coefficients of the same group of specimens obtained from

30 weeks wetting-drying cycles are found to be comparable with those of obtained from natural

diffusion. This finding, to a certain degree, is in agreement with the work of Arya et al [28], who

found that after the first cycle of wetting-drying, chloride penetration away from the near-surface

zone of concrete is largely diffusion controlled. Furthermore, Arya et al [28] found that diffusion

coefficients after 24 weeks total immersion and 24 wetting-drying cycles are broadly similar.

Nevertheless, it is widely believed that corrosion of the reinforcement is most likely to be boosted

in wetting-drying conditions since oxygen, an essential element for corrosion, is also more

readily available.

7.3.5 Effect of MHT on time to corrosion initiation

As discussed above, the incorporation of the two MHTs, in particular, MHT-pAB (5%, bulk

application) and the coating application for both MHT-pAB and MHT-NO2 resulted in an

increased chloride threshold (CT) and postponed corrosion initiation compared to the reference

specimens. As the focus of this study is on MHT modified with amino acids, only MHT-pAB, in

both applications, i.e., MHT-pAB(5% bulk) and MHT-pAB(coating), will be explored in the

following discussion. Aiming to give a picture of the effect of MHT on time to corrosion

initiation of concrete structures in chloride contaminated environment, a simplification of the

DuraCrete [29, 30] modeling approach was adopted, by which only deterministic calculations

were made.

7.3.5.1 The DuraCrete model

The DuraCrete model [29] for chloride induced corrosion is based on the concept of chloride

penetration by diffusion and initiation of corrosion when the chloride content at the surface of the

reinforcing steel has reached a critical value (i.e., CT). Corrosion initiation due to chloride ingress

142 | Chapter 7

is specified as the limit state for service life design, while the propagation phase of the corrosion

is neglected. According to the DuraCrete model, the time-development of chloride profiles can be

approximated as following (Eq. 7.5):

( , ) ( ) ( )4 ( )

s s i

xC x t C C C erf

kD t t

(7.5)

where C(x, t), Cs, Ci, x and t have the same meaning and unit as used in Eq. 7.4. k is the

environmental coefficient by which D(t) is multiplied to obtain the chloride diffusivity in a real

structure. D(t) is the chloride diffusion coefficient, which is a function of time since the rate of

chloride penetration into the mortar/concrete decreases with time. This is due to factors such as

continuing cement hydration as described in section 7.3.4. The time dependency of D(t) is given

by Eq. 7.6 :

00( )

nt

D t Dt

(7.6)

where D0 is the diffusion coefficient at reference time t0; n is the ageing coefficient (0 < n < 1).

7.3.5.2 Application of the model

Following the methodology described above, the parameters in Equations 7.5 & 7.6 were defined

based on the experimental results that were obtained in the previously described tests. For

Figure 7.29 Calculated chloride profiles of the mortar specimens with/without MHT-pAB with a given

cover depth of 50 mm at their respective chloride threshold (CT, by mass of cement) and time-to-corrosion

initiation (TTC, as shown in the key).


example, Cs values from the wetting-drying test reported in Table 7.3 were taken into account for

each mix. Ci was 0% as measured in Chapter 6. The chloride threshold (CT) values were taken

from the migration tests as reported in Figure 7.19. The k-values were set to be 1.0 for simplicity.

D0 is the corresponding value of Dapp from wetting-drying test as reported in Table 7.3 and the

reference time t0 is counted as 105 days (15 weeks with respect to 30 weeks wetting-drying

cycles). The ageing coefficient n is set as 0.37 referring to the DuraCrete specified value in “tidal

and splash” conditions.

By inserting all the above mentioned values into Equations 7.5 & 7.6, Figures 7.29 & 7.30

give chloride profiles of mortar specimens with/without MHT-pAB, which were calculated

respectively for a given cover depth of 50 mm and a given time-to-corrosion initiation (TTC) of

50 years when CT was reached at the cover depth. Table 7.5 gives an overview of the results

calculated using the DuraCrete model with input from the experiments. As can be seen from

Table 7.5, with a given cover depth of 50 mm, TTC is significantly increased due to the

incorporation of the MHT-pAB in both applications. The highest increase, which was a more than

double TTC (+112.5%) with respect to reference mortar, was obtained by incorporation of 5%

MHT-pAB in the bulk. Second best was 20% MHT-pAB in cement paste as a coating on the

rebar (almost double TTC). On the other hand, if TTC is fixed at 50 years, the lowest cover depth

(i.e., 42 mm) can be required by incorporation of 5% MHT-pAB in the bulk, which is 13 mm less

Figure 7.30 Calculated chloride profiles of the mortar specimens with/without MHT-pAB at a given time-

to-corrosion initiation of 50 years with their respective cover depth (lines and arrows show chloride

threshold (CT, by mass of cement) values of MHT-pAB(5%) and Ref. at cover depth).

144 | Chapter 7

than the reference case (55 mm cover required). It has to be noted that the performance

improvement due to the incorporation of MHT-pAB in terms of the two application ways (i.e., in

the bulk mortar or as a coating on the steel) is almost the same (TTC: 85 years vs. 78 years or

cover depth: 42 mm vs. 43 mm). Considering the ease of execution and time cost (coated steel

needs (at least) one day of curing before casting can take place), the incorporation of MHT-pAB

in the bulk mortar seems more attractive from a practical point of view although only little

amount of MHT-pAB is needed for coating application. In addition, if further improvement is

desired, a combination of the two application ways may be worth considering.

Finally, it has to be pointed out that the calculations are deterministic and consequently yield

mean outcomes, i.e., time-to-corrosion initiation (TTC) or cover depth in our cases. In

consequence, the probability in terms of corrosion initiation at that point in time and space is 50%.

Such a high probability, however, would be unacceptable for service life design in practice. To

address this issue, various DuraCrete reports and later documents [31-33] provide information on

how to obtain lower than 50% probabilities. In our study, the DuraCrete model for chloride

transport was mainly used for comparing “options”. Further detailed probabilistic calculations

were considered out of the present scope.

Table 7.5 Input data and calculated results of time-to-corrosion initiation (TTC) at a given cover depth of

50 mm and the cover depth at a give TTC of 50 years based on DuraCrete model for mortar specimens

with/without MHT-pAB.

Sample group Ref. Ref.(coating) MHT-pAB(5%) MHT-pAB(coating)

Cs (% by mass of cement) 5.8 5.8 8.6 5.8

D0 (×10-12

m2/s) 3.3 3.2 1.8 3.3

CT (% by mass of cement) 0.93 1.00 1.24 1.55

k 1.0 1.0 1.0 1.0

t0 = 105 days/15weeks (year) 0.29 0.29 0.29 0.29

n 0.37 0.37 0.37 0.37

TTC (year, cover depth = 50 mm) 40 43 85 78

Cover depth (mm, TTC = 50 year) 55 52 42 43

7.4 Conclusion

Two MHTs (i.e., MHT-pAB and MHT-NO2) were incorporated into mortar specimens with

embedded steel and were applied in two different ways: (1) as one of the mixing components in

bulk mortar; (2) as a surface coating of the reinforcing steel. Three test methods including

electrically accelerated chloride migration, cyclic wetting-drying and natural diffusion test based

on chloride exposure were adopted to investigate the anti-corrosion performance of two MHTs.

The results obtained from the accelerated chloride migration test revealed that either the

incorporation of MHT-pAB to replace 5% cement weight as one of the mixing components in

mortar or MHT (in the cases of both MHT-pAB and MHT-NO2) applied as a surface coating of


the reinforcing steel produced a notably extended time to corrosion initiation. The chloride

threshold (CT) of MHT specimens, which was expressed in total chloride percent by weight of

cement, was found to increase to a higher level based on the results obtained from the accelerated

migration test. The effect was 66.7%, the highest percentage of increase for MHT-pAB(coating)

and 28% the lowest for MHT-NO2(5%) relative to the reference specimens. In addition, it was

found that MHT-pAB(coating) and MHT-NO2(coating) exhibited higher CT values as well as

longer time to depassivation (except for MHT-pAB(5%) in this case) relative to the other

specimens. Furthermore, if the MHT added in the same percentage (i.e. 5%, 10% in bulk and

20% in coating), specimens with MHT-pAB exhibited higher CT values than those with MHT-

NO2. When comparing individually, the incorporation of MHT in a larger amount (10% MHT

relative to 5% MHT) of either MHT-pAB or MHT-NO2 in mortar increased the corresponding

CT values, but on the other hand shortened the time to depassivation. Although the porosity of

the mortar may play a dominating role in determining the chloride transport in mortar (in

particular for MHT-NO2(10%)), the general improved anti-corrosion performance due to the

incorporation of MHT (especially in the cases of 5% MHT-pAB in mortar or in coating for both

MHT-pAB and MHT-NO2) could be ascribed to the combined effect of active chloride binding

by MHT and inhibition by the simultaneously released inhibitors.

Only few specimens showed active corrosion during 30 weekly wetting-drying cycles. For

the corroding specimens, the following order of time to depassivation was established based on

the first appearance of active corrosion:

MHT-NO2(10%) < MHT-pAB(10%) < Ref. ≤ Ref.(coating) < MHT-NO2(5%) < MHT-pAB(5%)

~ MHT-NO2(coating) ~ MHT-pAB(coating) (no corrosion detected)

The CT values of the corroding specimens obtained from wetting-drying test were found to be

more than twice as high as those from accelerated migration test. This is likely due to two effects:

(1) the synergic effect of MHT in terms of chloride binding and corrosion inhibition of the

simultaneously released inhibitors could be more manifest in wetting-drying cycles; (2) the

higher ion transport rate including chloride ions as well as released anionic inhibitor caused by

the externally imposed electrical field may consequently result in less chloride binding and less

inhibition under electrically accelerated condition than under wetting-drying condition.

No corrosion was detected after 30 weeks natural diffusion test. There was good

correspondence in terms of surface chloride contents (Cs) and diffusion coefficients (Dapp)

between natural diffusion and wetting-drying data for the non-corroding specimens with/without

MHT, although these two parameters for the same group of specimens obtained from natural

diffusion were a little smaller than those from wetting-drying test. Chloride profile analysis of the

non-corroding specimens from both natural diffusion and wetting-drying test revealed that MHT-

pAB(5%) presented the highest Cs value along with the lowest Dapp value relative to all other

specimens in both tests indicating a more active chloride binding and a stronger resistance to

chloride penetration.

146 | Chapter 7

To summarize, experimental results presented in this chapter served to validate that the two

investigated MHTs could be promising alternatives for preventing chloride-induced corrosion

when an appropriate mixing dosage is adopted and applied in a proper way, i.e., either

incorporation of a small amount (in particular, MHT-pAB to replace 5% weight of cement) in the

bulk mortar or as a surface coating on the reinforcing steel (MHT-pAB or MHT-NO2 to replace

20% weigh of cement in paste). In particular, the effect of MHT-pAB on time-to-corrosion

initiation of reinforcing steel in concrete structures in chloride contaminated environment is

further estimated based on the DuraCrete chloride transport model, which shows an important

improvement. With regard to the applied testing methodology, the modified accelerated

migration test was found to be promising. It is relatively simple and can provide a useful means

of classifying the effect of new corrosion mitigating admixtures in a relatively short time.

References

[1] Andrade C, Rebolledo N. Accelerated evaluation of corrosion inhibition by means of the integral

corrosion test. Internation Conference on Concrete Repair, Rehabilitation and Retrofitting III, Cape

Town, South Africa: CRC Press; 2012. p. 364-8.

[2] Castellote M, Andrade C, Alonso C. Accelerated simultaneous determination of the chloride

depassivation threshold and of the non-stationary diffusion coefficient values. Corrosion Science.

2002;44(11):2409-24.

[3] Andrade C, Buják R. Effects of some mineral additions to Portland cement on reinforcement corrosion.

Cement and Concrete Research. 2013;53:59-67.

[4] Spanish Standard UNE 83992–2 EX. Durability of concrete. Test methods. Chloride penetration tests

on concrete. Part 2: Integral accelerated method. Madrid, Spain: AENOR; 2012.

[5] Castellote M, Andrade C, Alonso C. Measurement of the steady and non-steady-state chloride

diffusion coefficients in a migration test by means of monitoring the conductivity in the anolyte

chamber. Comparison with natural diffusion tests. Cement and Concrete Research. 2001;31(10):1411-

20.

[6] Castellote M, Andrade C, Alonso C. Modelling of the processes during steady-state migration tests:

Quantification of transference numbers. Materials and Structures. 1999;32(3):180-6.

[7] Bard AJ, Parsons R, Jordan J. Standard potentials in aqueous solution. New York: Marcel Dekker;

1985.

[8] Gonzalez J, Andrade C, Alonso C, Feliu S. Comparison of rates of general corrosion and maximum

pitting penetration on concrete embedded steel reinforcement. Cement and Concrete Research.

1995;25(2):257-64.



[10] Alonso C, Andrade C, Castellote M, Castro P. Chloride threshold values to depassivate reinforcing

bars embedded in a standardized OPC mortar. Cement and Concrete Research. 2000;30(7):1047-55.

[11] NTBuild443. Concrete hardened: accelerated chloride penetration. NordTest, Espoo. 1995.

[12] Polder RB, Peelen WH. Characterisation of chloride transport and reinforcement corrosion in

concrete under cyclic wetting and drying by electrical resistivity. Cement and Concrete Composites.

2002;24(5):427-35.

[13] Atkins P. Physical chemistry. 5th ed. Oxford: Oxford University Press; 1994.


[14] Yonezawa T, Ashworth V, Procter R. Pore solution composition and chloride effects on the corrosion

of steel in concrete. Corrosion. 1988;44(7):489-99.

[15] Page C, Short N, El Tarras A. Diffusion of chloride ions in hardened cement pastes. Cement and


[16] Glass GK, Roberts A, Davison N. Achieving high chloride threshold levels on steel in concrete.

CORROSION 2004, NACE, Paper No 04332. 2004.

[17] Ann KY, Song H-W. Chloride threshold level for corrosion of steel in concrete. Corrosion Science.

2007;49(11):4113-33.

[18] Vedalakshmi R, Kumar K, Raju V, Rengaswamy N. Effect of prior damage on the performance of

cement based coatings on rebar: macrocell corrosion studies. Cement and Concrete Composites.

2000;22(6):417-21.

[19] Page C, Treadaway K. Aspects of the electrochemistry of steel in concrete. Nature. 1982;297:109-15.

[20] Glass GK, Buenfeld NR. Chloride threshold levels for corrosion induced deterioration of steel. In:

Nilsson L, Ollivier J, editors. 1st RILEM International Workshop on Chloride Penetration into

Concrete: Rilem Publications SARL; 1995. p. 429-40.

[21] Glass G, Buenfeld N. The influence of chloride binding on the chloride induced corrosion risk in

reinforced concrete. Corrosion Science. 2000;42(2):329-44.

[22] Ann K, Ahn J, Ryou J. The importance of chloride content at the concrete surface in assessing the

time to corrosion of steel in concrete structures. Construction and Building Materials. 2009;23(1):239-

45.

[23] Crank J. The mathematics of diffusion. 2nd ed. Oxford: Oxford University Press; 1975.

[24] Phurkhao P, Kassir M. Note on chloride-induced corrosion of reinforced concrete bridge decks.

Journal of engineering mechanics. 2005;131(1):97-9.

[25] Andrade C, Castellote M, Alonso C, González C. Non-steady-state chloride diffusion coefficients

obtained from migration and natural diffusion tests. Part I: Comparison between several methods of

calculation. Materials and Structures. 2000;33(1):21-8.

[26] Castellote M, Andrade C, Alonso C. Chloride-binding isotherms in concrete submitted to non-steady-

state migration experiments. Cement and Concrete Research. 1999;29(11):1799-806.

[27] Angst UM, Polder R. Spatial variability of chloride in concrete within homogeneously exposed areas.

Cement and Concrete Research. 2014;56:40-51.

[28] Arya C, Vassie P, Bioubakhsh S. Modelling chloride penetration in concrete subjected to cyclic

wetting and drying. Magazine of Concrete Research. 2014;66(7):364-76.

[29] The European Union—Brite EuRam III, DuraCrete-Probabilistic Performance Based Durability

Design of Concrete Structures. Final Technical Report, Document BE95-1347/R17. CUR, Gouda,

2000.

[30] Siemes T, Edvardsen C. Duracrete: service life design for concrete structures. In: Lacasse MA,

Vanier DJ, editors. Proc of 8DCMC, 8th International Conference on Durability of Building Materials

and Components. Ottawa: NRC Research Press; 1999. p. 1343–56.

[31] Wegen G, Polder RB, Breugel KV. Guideline for service life design of structural concrete: A

performance based approach with regard to chloride induced corrosion. Heron. 2012;57(3):153-68.

[32] Fédération Internationale du Béton, Model code for service life design. Lausanne, Switzerland: fib

Bull. 34; 2006.

[33] Polder RB, De Rooij MR. Durability of marine concrete structures: field investigations and modelling.

Heron. 2005;50(3):133-53.

Chapter 8

Conclusions and Recommendations

for Future Research

150 | Chapter 8

8.1 General conclusions

This chapter summarizes the main outcomes obtained from this research project and gives some

recommendations for further research, which probably could be initiated in the future based on

the experimental findings of this thesis.

The primary objective of this research project is to develop a new promising additive for

mortar and/or concrete as an alternative approach against chloride induced corrosion based on

modified hydrotalcites (in particular those modified by eco-friendly amino acids) and to

investigate their effects in mortar: (1) their interaction with chloride ions and (2) their influence

on corrosion of reinforcing steel. To this end, the research was divided into four stages with

associated tasks. Accordingly, the conclusions were drawn:

1. Evaluation of the inhibition performance of the five proposed corrosion inhibitors in

chloride-rich simulated concrete pore solution to select those with the best performance

for synthesis of MHTs.

The five proposed corrosion inhibitors are sodium nitrite and sodium salts of four types of amino

acids, namely, Glycine (Gly), 6-aminocaproic acid (6ACA), 11-aminoundecanoic acid (11AUA),

and p-aminobenzoic acid (pAB). Sodium nitrite was chosen for comparison purposes due to its

well-established inhibition performance in concrete. The anti-corrosion performance of the five

inhibitors was assessed based on open circuit potential (OCP) and linear polarization resistance

Figure 8.1 The highest chloride concentrations that the steel could sustain in simulated concrete pore

solutions containing respectively the five proposed inhibitors (i.e. NaNO2, pAB, 11AUA, 6ACA, and Gly,

each at 0.1 M). Note: the results were obtained based on OCP/LPR measurements; 0.1M NaOH as

simulated concrete pore solution was included for comparison purpose.

Conclusions and Recommentdations | 151

(LPR) evolution of (carbon) steel in simulated concrete pore solution with the stepwise addition

of chloride at concentrations varying from 0 to 0.4 M. Figure 8.1 illustrates the highest chloride

concentrations that the steel could sustain in simulated concrete pore solutions containing

respectively the five proposed inhibitors (each at 0.1 M).

Consequently, NaNO2, pAB and 11AUA were selected as the most promising candidate

modifiers for synthesis of MHT.

2. Synthesis of MHT using the selected inhibitors as modifiers and characterization by

means of XRD, FTIR & TGA/DSC and relevant elemental analysis.

Six MHTs were synthesized through the modification of two commercially available carbonate

hydrotalcites PURAL® MG 63 HT (Mg/Al atomic ratio 2.2) and PURAL

® MG 70 HT (Mg/Al

atomic ratio 2.7) with the three selected inhibitors from stage 1 using the calcination-rehydration

method. The combined information from relevant characterization, i.e., elemental analysis, XRD,

FT-IR and TG/DSC revealed that the carbonate ions in the two commercial hydrotalcites have

been successfully substituted by the three inhibitor anions, i.e., -NO2, -pAB and -11AUA.

3. Anti-corrosion performance evaluation of the six synthesized MHTs (two Mg/Al atomic

ratios of 2.2 and 2.7, which were denoted as 2 and 3 respectively) in simulated concrete

pore solution to select the MHTs with the best performance for use in mortar test.

The anti-corrosion performance of the six synthesized MHTs was evaluated based on the chloride

binding and inhibitor release kinetics and the OCP/LPR evolution of steel in chloride-containing

simulated concrete pore solution. Firstly, the chloride binding and inhibitor release kinetics of the

six MHTs were investigated in a mixed solution containing 0.1 M NaOH and 0.5 M NaCl. The

resulting profiles of chloride bound by the six MHTs and simultaneous release of inhibitors from

the six MHTs revealed that if having the same interlayer anion, MHT with an Mg/Al atomic ratio

of 2 showed a higher chloride exchange capacity than that with an Mg/Al atomic ratio of 3.

Secondly, the anti-corrosion performance of the three MHTs with higher chloride exchange

capacity, namely, Mg(2)Al-NO2, Mg(2)Al-pAB and Mg(2)Al-11AUA was further evaluated in

simulated concrete pore solution with the stepwise addition of chloride at concentrations varying

from 0 to 0.4 M. The main findings are: (1) No to minor influence was found on pH of the

simulated concrete pore solution due to addition of the three MHTs; (2) The OCP/LPR results

suggested that besides the ion exchange between free chloride ions and the intercalated inhibitor

anions in MHT, the simultaneously released inhibitor anions, in particular -pAB, exhibited a

notable inhibiting effect and caused a shift of the corrosion initiation of steel to a higher chloride

concentration level compared to its counterpart pure inhibitor (i.e., 0.1 M pAB). Figure 8.2 shows

the highest chloride concentrations that the steel could sustain in simulated concrete pore solution

containing respectively the three selected MHTs (each at 0.1 M).

Based on the results, Mg(2)Al-NO2 and Mg(2)Al-pAB were selected for use in mortar test.

152 | Chapter 8

4. Application of the two selected MHTs (i.e., Mg(2)Al-pAB and Mg(2)Al-NO2) in plain

and reinforced mortar to assess their effect on mortar properties (workability, porosity,

strength development), chloride transport and corrosion of reinforcing steel.

In plain mortar, the two MHTs were incorporated with two dosage levels replacing 5% and 10%

mass of cement with varying w/c ratios. Some conclusions are: (1) no remarkably negative

influence on the strength development of hardened mortar and on the consistence of fresh mortar

would result from the incorporation of the two MHTs, although the porosity to a certain degree

was increased compared to the reference specimen; (2) differences in w/c ratio between mortar

specimens caused by partial replacement of cement by the two MHTs could be also responsible

for lower strength and changed porosity; (3) At the same mixing dosage (either 5% or 10%),

Mg(2)Al-pAB presented a higher chloride binding capacity than Mg(2)Al-NO2, and more

chlorides were consequently bound in Mg(2)Al-pAB specimens than in Mg(2)Al-NO2 specimens;

(4) the active chloride binding behavior and the improved chloride diffusion resistance of

Mg(2)Al-pAB(5%) mortar specimen validated that MHT-pAB could be a promising alternative

for chloride scavenging in mortar when an appropriate mixing dosage is adopted; (5) there may

exist a threshold level for both Mg(2)Al-NO2 and Mg(2)Al-pAB above which they could be

ineffective or even detrimental in terms of hindering the chloride transport. Furthermore, an

additional series of mortar specimens were prepared for Rapid Chloride Migration (RCM) test

only. They were prepared by adding MHT instead of replacing cement at 5 and 10% of cement

Figure 8.2 The highest chloride concentrations that the steel could sustain in simulated concrete pore

solutions containing respectively Mg(2)Al-NO2, Mg(2)Al-pAB and Mg(2)Al-11AUA (each at 0.1 M).

Note: the results were obtained based on OCP/LPR measurements; 0.1 M NaOH as simulated concrete

pore solution was included for comparison purpose.


mass, keeping the w/c ratio of 0.5 and cement content constant. The results confirmed the higher

chloride penetration resistance, in particular of MHT-pAB at 5% addition, than reference mortar.

In reinforced mortar, the two MHTs were applied in two different ways: (1) as one of the

mixing components in mortar at two dosage levels replacing 5% and 10% mass of cement with

varying w/c ratios; (2) as a surface coating on the reinforcing steel in a cement paste replacing

20% of the cement mass. Three test methods including electrically accelerated chloride

migration, cyclic wetting-drying and natural chloride diffusion test based on chloride exposure

were adopted to custom designed mortar specimen with an embedded reinforcing steel bar.

The results obtained from the accelerated chloride migration test revealed that: either the

incorporation of MHT-pAB to replace 5% cement weight as one of the mixing components in

mortar or the MHT (for both MHT-pAB and MHT-NO2) applied as a surface coating of the

reinforcing steel produced a notably extended time to corrosion initiation. The delayed corrosion

initiation has been ascribed to the two synergic effects of MHT: (1) chloride binding effect to

slow down chloride transport in the body of mortar; (2) synergy of chloride binding and

inhibiting effect (via simultaneously released inhibitors from MHT) to increase the chloride

threshold of the reinforcing steel. In addition, the release of inhibitors from MHT via the chloride

exchange reaction was found to have more influence on the chloride threshold than on time to

corrosion initiation of the reinforcing steel.

Only few specimens showed active corrosion during 30 weekly wetting-drying cycles. No

corrosion was detected for MHT-pAB(5%), MHT-pAB(coating) and MHT-NO2(coating)

specimens, which therefore showed lowest probability of corrosion. The chloride threshold values

of the corroding specimens obtained from the wetting-drying test were more than twice as high as

those from the accelerated migration test. Two possible reasons have been proposed to explain

the higher chloride threshold values from wetting-drying test: (1) the synergic effect of MHT in

terms of chloride binding and corrosion inhibition of the simultaneously released inhibitors could

be more manifest in wetting-drying cycles; (2) the higher ion transport rate including chloride

ions as well as released anionic inhibitor caused by the externally imposed electrical field may

consequently result in less chloride binding and less inhibition under electrically accelerated

conditions than under wetting-drying conditions.

No corrosion was detected after 30 weeks natural diffusion test. Chloride profile analysis of

the non-corroding specimens from natural diffusion as well as wetting-drying test revealed that

MHT-pAB(5%) presented the highest surface chloride content (Cs) along with the lowest chloride

diffusion coefficient (Dapp) relative to all other specimens, indicating a more active chloride

binding and a stronger resistance to chloride penetration.

Compared to cyclic wetting-drying and natural chloride diffusion test, the modified

accelerated migration test seems to be promising. It is simple and can provide a useful means of

classifying the effect of new corrosion mitigating admixtures in a relatively short term. The

effects of MHT, in particular MHT-pAB, on service life of structures in chloride contaminated

154 | Chapter 8

environment is estimated based on DuraCrete modelling approach, which shows a significant

improvement.

In summary, the research and experimental results presented in this thesis met the

expectations and goals formulated at the start of the project. This is to say, a new type of smart

concrete additive based on amino acid modified hydrotalcites aiming to combat chloride-induced

corrosion has been developed and documented. This new additive on one hand, did not show a

remarkably negative influence on the mortar properties (i.e., workability in fresh state, porosity

and strength development of hardened mortar); on the other hand, when an appropriate mixing

dosage is adopted and applied in a proper way, either incorporation of a small amount (in

particular, MHT-pAB to replace 5% weight of cement) in bulk mortar or as a surface coating of

the reinforcing steel (MHT-pAB to replace 20% weight of cement in paste) does result in delayed

corrosion initiation and increased chloride threshold. As such, by using such a material, a longer

service life of reinforced mortar/concrete structures can be expected.

8.2 Industrial application potentials and valorization

Kisuma Chemicals BV (a main hydrotalcite producer/provider in the world market) and CRH

(one of the worldwide leading cement and concrete companies) have expressed their interest for

further cooperation. Kisuma Chemicals BV has made its first trial for producing aminobenzoate

modified hydrotalcite (MHT-pAB) and a follow-up industrial application project is currently

under discussion.

8.3 Recommendations for future research

In this project, six synthetic MHTs were developed. Two of them (i.e., Mg(2)Al-pAB and

Mg(2)Al-NO2) were selected and applied in reinforced mortar specimens in two different ways

(i.e., as a mixing component and as a steel surface coating). In terms of the type of MHT, mixing

dosages and the specific application ways, three advices could be made on next steps in future

research:

1. Use of MHT-pAB instead of MHT-NO2 because of the potentially toxic/carcinogenic

effect of nitrite pollution on the environment.

2. Optimize the incorporation dosages (either as cement replacement or as an additive) that

would be used in the two application ways (either in mortar mix or in the coating) in order

to maximize the beneficial effect of MHT on corrosion resistance of reinforcement.

3. Consider applying MHT-pAB in a synergistic way (i.e. mixing in bulk mortar combined

with the coating application) with the optimum dosages to fortify the beneficial effect of

MHT-pAB for improved corrosion protection of reinforced concrete.

In addition, the MHTs, in particular, amino acids modified Mg-Al based hydrotalcites (MHT-

pAB, MHT-Gly, MHT-6ACA and MHT-11AUA) as well as MHT-NO2 that were investigated in


this thesis belong to a large mineral family of Layered Double Hydroxides (LDHs), in general

formula [MII

1-x MIII

x (OH)2]x+

[(An-

x/n)]x-

·mH2O. Therefore, the synthesis of more kinds of MHTs

with various combination of different kinds of metal cations (i.e., MII such as Ca

2+, Mg

2+, Zn

2+,

Fe2+

and Ni2+

; MIII

such as Al

3+, Fe

3+, Cr

3+ and Ga

3+) and inhibitive interlayer anions (i.e., A

n-,

such as a wide range of inorganic or organic agents) and further optimisation of the synthesized

MHT to obtain the best advantages may be also an interesting part for future research. The

optimisation can include the screening of appropriate MHT members that are more compatible

with cementitious materials, the choice of an optimum incorporation dosage and an proper way

for applying the MHT, either as cement replacement or by addition and/or as a surface coating of

reinforcement.

Moreover, thanks to the unique characteristic of the layered molecular structure and high

anionic exchange capacity, hydrotalcite has a potential to be modified or tailor-made by a wide

range of inorganic or organic anionic agents as a new type of smart additive with desired

functions for cement-based materials (paste, mortar and concrete). This is to say, the specific dual

function of the MHT (i.e., capturing chloride ions and simultaneously releasing intercalated

inhibitors upon the arrival of chlorides) that was explored in this project is not the only beneficial

way that MHT can offer as a functional additive. Therefore, the scope of application for MHT

with various combination of different kinds of host metal hydroxides and interlayer anions in

cement-based materials could be significantly expanded. For example, a controlled release

formulation based on MHT can be made by encapsulation/immobilization of a desired functional

compound within the layered molecular structure of hydrotalcites. The produced MHT in this

aspect can be treated as an internal reservoir for this specific functional compound, which then

can be slowly released in a controllable way through an external stimulus. This functional

compound could be a superplasticizer, a shrinkage reducer, an ASR inhibiting agent, an air-

entraining agent, a pore solution viscosity adjuster, a setting accelerator/retarder and probably

other any concrete property adjusters. Compared to the condition where these admixtures are

used alone, such controlled release formulation of MHT is expected to have improved relevant

properties and provide the desired function at the time that is needed. On top of that, due to the

relatively high affinity of MHT to carbonates and sulphates, the MHT could be an effective trap

of carbonates and sulphates in the prevention of carbonation and/or sulphate attack on cement-

based materials.

Corrosion-related durability issues of concrete are normally associated with high costs and

safety concerns, as they entail additional repair, rehabilitation, and monitoring activities to ensure

the functionality and aesthetics of concrete structures and components. Hence, any improvement

in the structural design, materials selection, production, construction, and maintenance may help

in improving the service life of concrete structures, which in turn may have enormous social,

economic and environmental benefits. In this sense, the direct and indirect savings by using a

small amount of MHT, for example 5% MHT-pAB addition by mass of cement in the mortar

bulk or as a cementitious coating with 20% MHT-pAB applied to the reinforcement (as studied in

156 | Chapter 8

this research) could be substantial. Furthermore, efforts to improve the MHT (e.g. by optimizing

its synthesis and chemical composition) and/or to produce new types of MHTs with lower cost

are highly recommended for future research.

Another recommendation for further research relates to modeling approaches. In this thesis,

the effect of MHT, in particular MHT-pAB, on service life of structures in chloride contaminated

environment is estimated by a simplified application of the DuraCrete model. However, based on

the experimental results that have been obtained, a more advanced model for predicting the

service life of concrete incorporated with a certain kind of MHT that has been tailor-made with

the desired specific function could possibly be framed.

Summary

Corrosion of reinforcing steel is a major culprit to durability and serviceability of concrete

structures. This problem is highly relevant for civil engineering structures in the transport sector,

such as bridges, tunnels, harbour quays and parking structures. The dominant aggressive external

influence is the chloride load from de-icing salts or sea water, penetrating the concrete and

destroying the natural (high pH) passivation of the steel. The direct and indirect costs of

reinforcement corrosion are substantial, as it entails additional repair, rehabilitation, and

monitoring activities to ensure the safety, functionality and aesthetics of concrete structures and

components. In addition, many repairs have a short working life, necessitating repeated repairs

within the use life. Consequently, the construction industry is in need of improving the corrosion

protection of reinforcing steel, preferably by low-cost measures. Presently available corrosion

preventive measures are either too costly or technically too complicated to be applied on a wide

scale. Stainless steel reinforcement is 5 or 10 times more expensive than reinforcing (carbon)

steel. Cathodic prevention and protection may be effective but both are a special niche expertise

and are thus not applied on a wide scale. Coatings on the concrete surface normally do not last

long enough (10-20 years), which causes a maintenance cycle of its own. Corrosion inhibitors

seem to be attractive owing to their low cost and the ease and flexibility of application. However,

there are conflicting opinions about the reliability of the inhibitors for corrosion protection in

concrete in terms of long-term efficiency; some are toxic, such as nitrites. A possible promising

solution to overcome this problem is the encapsulation/immobilization of desired inhibitors

within the molecular structure of a host compound. The immobilized inhibitor then can be slowly

released in a controllable way by an external stimulus (e.g. chloride ions) and therefore provide a

relatively long-term corrosion protection.

Owing to the unique fine tunable molecular structure and high anionic exchange capacity,

modified hydrotalcites (MHTs) have the potential to be used for the immobilization of a desired

inhibitor. Hydrotalcite is one representative of large mineral group of Layered Double

Hydroxides (LDHs), in general formula [MII

1-x MIII

x (OH)2]x+

[(An-

x/n)]x-

·mH2O, where MII and

MIII

are di- and trivalent metals respectively, and An-

is an interlayer charge-balancing anion with

valence n. The x value is in the range 0.20-0.33. Although the most common anion found in

naturally occurring hydrotalcites is carbonate, in practice however, there is no significant

restriction to the nature of the interlayer charge-balancing anions. The MHT structure can

accommodate various cations in the hydroxide layers with varying MII/M

III ratios as well as a

wide range of anionic species in the interlayer regions.

Within the MHT family, a class of materials with emerging importance is that constituted by

MHTs intercalated with organic species. In addition, increasing awareness of the health and

ecological risks has drawn much attention to amino acid-based inhibitors because they are

nontoxic, environmentally friendly, relatively cheap and easy to produce with higher purity.

158 | Summary

Therefore, the marriage of the two kinds of materials is expected to not only offer an improved

inhibiting effect than using the inhibitor alone but also to impose less impact on environment.

Recently a study on the application of amino acid modified hydrotalcites in cementitious

materials has formed the basis of a patent (WIPO Patent, WO 2011/065825 A1). However, its

scale was relatively small and further work was considered necessary by the applicants and their

organisations.

In this research, four different types of sodium salts of amino acids (i.e., Glycine, 6-

aminocaproic acid, 11-aminoundecanoic acid, and p-aminobenzoic acid) were proposed as

potential candidates for the modification of hydrotalcite. Sodium nitrite was also chosen as a

modification candidate for comparison purposes due to its well-recognized inhibition

performance in concrete. Based on the anti-corrosion performance evaluation in chloride

contaminated simulated concrete pore solution (Chapter 3), sodium nitrite, sodium salt of p-

aminobenzoic acid (pAB) and sodium salt of 11-aminoundecanoic acid (11AUA) were selected

as the most promising candidate modifiers for synthesis of MHT. Subsequently, six MHTs (two

Mg/Al atomic ratios of 2.2 and 2.7, which were denoted as 2 and 3 respectively) were synthesized

through the modification of two commercially available carbonate Mg-Al hydrotalcites PURAL®

MG 63 HT (Mg/Al atomic ratio 2.2) and PURAL® MG 70 HT (Mg/Al atomic ratio is 2.7) by

NaNO2, pAB and 11AUA (Chapter 4). They were characterized by means of X-ray powder

diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), Thermogravimetry (TG),

Differential scanning calorimetry (DSC) and relevant elemental analysis.

The ion exchange characteristics of the six synthesized MHTs and their anti-corrosion

performance were investigated in chloride-rich simulated concrete pore solution (Chapter 5). The

results showed that ion exchange occurred between free chloride ions in the simulated concrete

pore solution and the inhibitive anions intercalated in MHT, thereby reducing the free chloride

concentration which is equivalent to increased binding of chloride in mortar/concrete. Moreover,

the simultaneously released anions, in particular -pAB, were found to exhibit a notable inhibiting

effect and caused shifting of the corrosion initiation of steel to a higher chloride concentration

level. This evidence manifested the dual role protecting function that MHT (in particular,

Mg(2)Al-pAB) offers to the steel: capturing chlorides as a chloride scavenger and providing of a

beneficial release of corrosion inhibitors in parallel as an inhibitor reservoir further protecting

reinforcing steel from corrosion.

The effects of two MHTs, i.e., Mg(2)Al-NO2 and Mg(2)Al-pAB, were investigated in both

plain and reinforced mortar specimens with a focus on their interaction with chloride ions in plain

mortar (Chapter 6) and in reinforced mortar, mainly focusing on their inhibition influence on

corrosion of the reinforcing steel (Chapter 7). In plain mortar, the two MHTs were incorporated at

two dosage levels replacing 5% and 10% mass of cement. A testing programme including

workability test, strength test, porosity test, and rapid chloride migration and diffusion test was

employed to investigate the effect of the two MHTs on chloride penetration in mortar. The results

indicated that the incorporation of Mg(2)Al-pAB at 5% dosage in mortar produced a notably

improved chloride diffusion resistance with no remarkably negative influence on the development

of mechanical strength and workability of fresh mortar, which therefore validated that the

Summary | 159

Mg(2)Al-pAB could be a promising alternative in hindering the chloride transport in mortar when

an appropriate mixing dosage is adopted.

In reinforced mortar, the two MHTs were applied in two different ways: (1) as one of the

mixing components in bulk mortar at two dosage levels replacing 5% and 10% mass of cement;

(2) as a surface coating on the reinforcing steel in a cement paste replacing 20% of the cement

mass. Three test methods including electrically accelerated chloride migration, cyclic wetting-

drying and natural chloride diffusion test based on chloride exposure were adopted to custom

designed reinforced mortar specimen. Although no corrosion was detected after 30 weeks natural

diffusion testing, the results obtained from accelerated chloride migration and cyclic wetting-

drying test revealed that when an appropriate mixing dosage is adopted and applied in a proper

way, the application of MHT either incorporation of a small amount (in particular, Mg(2)Al-pAB

to replace 5% weight of cement) in mortar or as a surface coating of the reinforcing steel

(Mg(2)Al-pAB or Mg(2)Al-NO2 to replace 20% weight of cement in paste) resulted in delayed

corrosion initiation and increased chloride threshold responsible for initiating corrosion. The

effects on service life of structures in chloride contaminated environment is estimated, which

shows a significant improvement.

In general, the research work presented in this thesis met the expectations and goals

formulated at the start of the project. As the first exploration on a wider scale into the application

of MHT in cementitious materials for corrosion protection purposes, a new type of smart concrete

additive based on amino acid modified hydrotalcites (in particular Mg(2)Al-pAB) aiming to

combat chloride-induced corrosion has been developed and documented. The results

demonstrated that by using such a material, a longer service life of reinforced mortar/concrete

structures can be expected. While realizing that more research is still needed for maximizing the

beneficial effect of MHT as a functional additive of cementitious materials, some

recommendations for further research are given in the last chapter of this thesis (Chapter 8).

MHT has a very rich interlayer chemistry and can participate in anion exchange reactions

with great facility. Therefore, the scope of application for MHT with combination of different

kinds of host metal hydroxides and various interlayer anions with desired specific function in

cement-based materials could be significantly expanded. For example, a controlled release

formulation based on MHT can be made by encapsulation/immobilization of a desired functional

compound within the layered molecular structure of hydrotalcites. This functional compound

could be a superplasticizer, a shrinkage reducer, an ASR inhibiting compound, an air-entraining

agent, a pore solution viscosity adjuster, a setting accelerator/retarder and probably other any

concrete property adjusters. In this respect, we are confident that future work on applications of

new types of smart functional concrete additives based on MHTs will expand rapidly and

contribute greatly to the effort of searching for effective measures to improve the durability or

other properties of reinforced concrete.

Samenvatting Corrosie van wapeningsstaal is een grote boosdoener voor duurzaamheid en bruikbaarheid van

betonconstructies. Dit probleem is zeer relevant voor civieltechnische constructies in de

transportsector, zoals bruggen, tunnels, havenkades en parkeerconstructies. De dominante

agressieve invloed van buitenaf is de chloridebelasting van dooizouten of zeewater, dat het beton

penetreert en de natuurlijke (hoge pH) passivering van het staal vernietigt. De directe en indirecte

kosten van wapeningscorrosie zijn aanzienlijk, vanwege de noodzaak tot reparatie, renovatie en

om veiligheid, functionaliteit en esthetiek van betonconstructies en componenten te garanderen.

Daarnaast hebben veel reparaties een korte levensduur, waardoor herhaalde reparaties binnen de

levensduur noodzakelijk zijn. Bijgevolg is er in de bouwsector behoefte aan verbetering van de

bescherming tegen corrosie van wapeningsstaal, bij voorkeur door goedkope maatregelen. Thans

beschikbare preventieve maatregelen tegen corrosie zijn ofwel te duur of technisch te

ingewikkeld om op grote schaal te worden toegepast. Roestvaststalen wapening is 5 tot 10 keer

duurder dan (koolstof) wapeningsstaal. Kathodische preventie en bescherming kunnen effectief

zijn, maar zijn een speciale deskundigheid en derhalve worden zij niet toegepast op grote schaal.

Coatings op het betonoppervlak gaan normaliter niet lang genoeg mee (10-20 jaar), wat een eigen

onderhoudscyclus veroorzaakt. Corrosie-inhibitoren lijken aantrekkelijk vanwege hun lage kosten

en het gemak en flexibiliteit van toepassing. Er zijn tegenstrijdige meningen over de

betrouwbaarheid van inhibitoren voor corrosiebescherming in beton in termen van langdurige

effectiviteit; sommige zijn giftig, zoals nitrieten. Een mogelijke veelbelovende oplossing voor dit

probleem is de inkapseling / immobilisatie van inhibitoren binnen de moleculaire structuur van

een gastheerverbinding. De geïmmobiliseerde inhibitor kan dan langzaam op een controleerbare

wijze worden vrijgegeven door een externe stimulus (bijvoorbeeld chloride ionen) en derhalve

een relatief lange termijn corrosiebescherming leveren.

Door de unieke fijn afstembare moleculaire structuur en hoge anionische

uitwisselingscapaciteit, hebben gemodificeerde hydrotalcieten (Modified Hydrotalcites, MHTs)

de potentie om te worden gebruikt voor het immobiliseren van een inhibitor. Hydrotalciet is een

vertegenwoordiger van de grote minerale groep Gelaagde Dubbele Hydroxides (Layered Double

Hydroxides, LDHs), in de algemene formule [MII

1-x MIII

x (OH)2]x+

[(An-

x/n)]x-

·mH2O, waarbij MII

en MIII

twee- en driewaardige metalen zijn en An-

een tussenlaag lading balancerend anion is met

valentie n. De x-waarde ligt tussen 0,20-0,33. Hoewel carbonaat het meest voorkomende anion is

in natuurlijk voorkomende hydrotalcieten, is er in de praktijk geen significante beperking op de

aard van deze anionen. De MHT structuur kan diverse kationen herbergen in de hydroxide lagen

met variërende MII/M

III verhoudingen alsmede diverse soorten anionen in de tussenlaag.

Binnen de MHT-familie is een klasse van materialen met toenemende betekenis

vertegenwoordigd door MHTs met ingebouwde organische ionen. Daarnaast is door toenemend

bewustzijn van gezondheids- en milieurisico’s veel aandacht voor op aminozuur gebaseerde

inhibitoren, omdat ze niet-toxisch, milieuvriendelijk, relatief goedkoop en eenvoudig te

Samenvatting | 161

produceren zijn met hoge zuiverheid. Daarom biedt de combinatie van de twee soorten materialen

naar verwachting niet alleen een verbeterd inhibitie-effect dan bij het gebruik van alleen de

inhibitor, maar ook minder impact op het milieu.

Onlangs heeft een studie over de toepassing van aminozuur gemodificeerde hydrotalcieten in

cementgebonden materialen de basis gevormd van een octrooi (WIPO Patent, WO 2011/065825

A1). De schaalgrootte was echter relatief klein en verder werk werd door de aanvragers en hun

organisaties noodzakelijk geacht.

In dit onderzoek werden vier verschillende natriumzouten van aminozuren (i.e. glycine, 6-

aminocapronzuur, 11-aminoundecaanzuur en p-aminobenzoëzuur) voorgesteld als mogelijke

kandidaten voor modificatie van (inbouw in) hydrotalciet. Natriumnitriet werd ook gekozen als

een modificatiekandidaat voor vergelijkingsdoeleinden vanwege de erkend goede

inhibitieprestaties van nitriet in beton. Gebaseerd op de anticorrosie-prestatiebeoordeling in met

chloride verontreinigde gesimuleerd betonporievloeistof (hoofdstuk 3), werden natriumnitriet

(NaNO2) en de natriumzouten van p-aminobenzoëzuur (pAB) en 11-aminoundecaanzuur

(11AUA) geselecteerd als meest veelbelovende inbouwkandidaten voor de synthese van MHT.

Vervolgens werden zes MHTs (met twee Mg/Al atomaire verhoudingen, namelijk 2,2 en 2,7, die

respectievelijk aangegeven werden met 2 en 3) gesynthetiseerd door de modificatie van twee

commercieel beschikbare carbonaat Mg-Al hydrotalcieten PURAL®

MG 63 HT (Mg/Al

atoomverhouding 2,2) en PURAL® MG 70 HT (Mg/Al atoomverhouding 2,7) door NaNO2, pAB

en 11AUA (Hoofdstuk 4). Ze werden gekarakteriseerd door middel van röntgendiffractie (X-ray

powder diffraction, XRD), Fourier transformatie infrarood spectroscopie (FTIR),

thermogravimetrie (TG), differentiële scanning calorimetrie (DSC) en relevante elementanalyse.

De ionenuitwisselingseigenschappen van de zes gesynthetiseerde MHTs en hun anti-corrosie

prestaties werden onderzocht in chloride-rijke gesimuleerde betonporievloeistof (hoofdstuk 5).

De resultaten toonden dat ionenuitwisseling plaatsvond tussen vrije chloride-ionen in de

gesimuleerde betonporievloeistof en de ingevoegde inhibitor-anionen in MHT, waardoor de vrije

chlorideconcentratie verminderde, wat overeenkomt met verhoogde binding van chloride in

mortel/beton. Bovendien bleken de gelijktijdig vrijgegeven anionen, met name -pAB, een

aanzienlijk inhibiterend effect te vertonen, wat een verschuiving veroorzaakte van de corrosie-

initiatie van staal naar een hogere chlorideconcentratie. Dit bewijs manifesteerde de dubbele

beschermende rol die MHT (met name Mg(2)Al-pAB) biedt aan het staal: binden van chloriden

en afgeven van corrosieremmers, zodat wapeningsstaal beter wordt beschermd tegen corrosie.

De effecten van twee MHTs, Mg(2)Al-NO2 en Mg(2)Al-pAB, werden onderzocht in mortel

zonder en met ingebed betonstaal, met de nadruk op hun interactie met chloride-ionen in de

mortelfase (hoofdstuk 6) en voor gewapende mortel voornamelijk gericht op hun inhibiterende

invloed op corrosie van het wapeningsstaal (hoofdstuk 7). In mortel werden de twee MHTs

opgenomen in twee doseringen, waarbij cement werd vervangen door 5% en 10% (massa) MHT.

Een programma met beproeving van de verwerkbaarheid, sterkte, porositeit, chloridemigratie en

chloridediffusie werd uitgevoerd om het effect van de twee MHTs in mortel onderzoeken. De

resultaten gaven aan dathet inmengen van Mg(2)Al-pAb als 5% dosering in mortel een

aanzienlijk verbeterde chloride-diffusieweerstand produceerde zonder opvallend negatieve

162 | Samenvatting

invloed op de ontwikkeling van mechanische sterkte en verwerkbaarheid van verse specie.

Hierdoor werd bevestigd dat Mg(2)Al-pAB een veelbelovend alternatief zou kunnen zijn om

chloridetransport in mortel te belemmeren bij een geschikte mengseldosering.

In gewapende mortel werden de twee MHTs op twee manieren gebruikt: (1) als een van de

componenten in bulkmortel op twee doseringsniveaus, ter vervanging van 5% en 10% van het

cementgewicht; (2) als een laag cementpasta op het wapeningsstaal, waarbij 20% van het

cementgewicht vervangen werd. Drie testmethoden, waaronder elektrisch versnelde

chloridemigratie, nat-droogcycli met chlorideoplossing en natuurlijke diffusie op basis van

permanente blootstelling aan chlorideoplossing werden aangepast aan speciaal ontworpen

proefstukken. Hoewel er geen corrosie werd waargenomen na 30 weken testen onder natuurlijke

diffusie, bleek er uit de resultaten van versnelde chloridemigratie en cyclische nat-droogproeven

dat bij een geschikte mengseldosering, de toepassing van MHT als kleine toevoeging (met name

Mg(2)Al-pAB als vervanging tot 5% van cementgewicht) in mortel of als een dunne laag op het

wapeningsstaal (met 20% Mg(2)Al-pAB of Mg(2)Al-NO2 op cementgewicht in pasta) resulteerde

in vertraagde corrosieinitiatie en verhoogde kritische chlorideconcentratie voor het initiëren van

corrosie. De effecten op de levensduur van constructies in een met chloride verontreinigde

omgeving zijn geschat, wat een significante verbetering laat zien.

In het algemeen heeft het onderzoek beschreven in dit proefschrift voldaan aan de

verwachtingen en doelstellingen geformuleerd bij het begin van het project. Als eerste verkenning

op grotere schaal voor de toepassing van MHT in cementgebonden materialen ter

corrosiebescherming, werd een nieuw type slim betonadditief ontwikkeld en gedocumenteerd op

basis van met aminozuur gemodificeerde hydrotalcieten (in het bijzonder Mg(2)Al-pAB) ter

bestrijding van chloridegeïnduceerde corrosie. De resultaten toonden aan dat met een dergelijk

materiaal een langere levensduur van gewapende betonconstructies kan worden verwacht. Met

het besef dat er meer onderzoek nodig is om het gunstige effect van MHT te optimaliseren als een

functionele toevoeging van cementgebonden materialen, zijn aanbevelingen voor verder

onderzoek gegeven in het laatste hoofdstuk van dit proefschrift (hoofdstuk 8).

MHT heeft een zeer rijke tussenlaagchemie en uitwisseling van anionen vindt gemakkelijk

plaats. Daarom kon het toepassingsgebied aanzienlijk worden uitgebreid voor MHT met een

combinatie van verschillende soorten gastheer-metaalhydroxiden en diverse tussenlaag-anionen

met de gewenste specifieke functie in cementgebonden materialen. Formuleringen zijn mogelijk

voor gecontroleerde afgifte gebaseerd op hydrotalcieten door inkapseling/immobilisatie van een

gewenste functionele verbinding in de gelaagde moleculaire structuur. Deze functionele

verbinding kan zijn een superplastificeerder, een krimpreduceerder, een alkali silica reactie

(ASR) hinderende verbinding, een luchtbelvormer, een poriewater viscositeit aanpassend

materiaal, een bindtijdversneller/bindtijdvertrager en waarschijnlijk andere stoffen die

betoneigenschappen beinvloeden. In dit opzicht zijn we ervan overtuigd dat toekomstig werk

betreffende toepassingen van nieuwe types van slimme functionele additieven voor beton op

basis van MHTs zal groeien en een belangrijke bijdrage zal leveren aan de inspanning van het

zoeken naar effectieve maatregelen om de duurzaamheid of andere eigenschappen van gewapend

beton te verbeteren.

Acknowledgements

The research described in this thesis was carried out under project number M81.609337 in the

framework of the Research Program of the Materials innovation institute (M2i) in The

Netherlands (www.m2i.nl). Execution and supervision of this project were facilitated by the

Section of Materials and Environment of the Faculty Civil Engineering and Geosciences at Delft

University of Technology. First of all, I would like to thank M2i for the financial support of this

project. My special thanks are dedicated to: Gitty Bouman, Monica Reulink, and Irina Bruckner,

the HR officers and Derk Bol, the program manager from M2i for their supports during my PhD

research.

Netherlands Organisation for Applied Scientific Research (TNO) as the industrial partner

supported this project and has been involved in this research. The project benefited greatly from

the collaboration and valuable discussions with Dr. Hartmut Fischer at TNO Materials

Performance, Eindhoven. Dr. Fischer has transfused his knowledge and enthusiasm for research

into this project. His strong support, kind help and encouragement are gratefully acknowledged.

I have always believed that the success of a PhD project is the combination efforts from both

the candidate and the supervisor(s). In this respect, I would like to express my sincere gratitude

and deepest appreciation to my promoter and supervisor Prof. Rob Polder for giving me the

opportunity to pursue my doctoral degree in The Netherlands, for his guidance and endless

support throughout all stages of this project, and for establishing contact to leading researchers in

our research field. Our many discussions always made me motivated and got the needed

confidence in the research outcomes. This work would not have achieved significant progress

without his effective supervision. His positive attitude and his trust have been one of the biggest

driving forces to me for more challenging endeavours. As a great mentor, he has always inspired

me to think independently and make decisions on my own from the very beginning of this project.

This has elevated me to develop a critical attitude towards my own research and evolve into a

more independent researcher than would have been otherwise possible.

I would like to thank Prof. Klaas van Breugel, the head of the Section of Materials and

Environment for providing me the chance to work in the M&E and all the support during these

years. Prof. Erik Schlangen, the head of M&E’s Microlab, is greatly acknowledged for sharing

his profound knowledge and the valuable discussions on the concrete mechanical properties and

related tests.

My special thanks goes to Assoc. Prof. Dr. Arjan Mol from the Corrosion group of

Department of Materials Science and Engineering in the Faculty of Mechanical, Maritime and

Materials Engineering. I am very much grateful for his willingness to be a help and collaboration

on part of this research and providing the lab space and electrochemical equipment for corrosion

test. Without his valuable contribution and expertise, the completion of this project could have

http://www.m2i.nl/

164 | Acknowledgements

not been possible. As an additional outcome of this project, our friendly collaboration had

resulted in two successful master’s projects in Faculty of 3mE. I enjoyed working as the first-line

supervisor and one of the graduation examination committee members of the two master students,

Jun Liu and Min Mao.

I would like to express my sincere gratitude to Prof. Carmen Andrade for her kind help and

hospitality during my stay in the Institute ‘Eduardo Torroja’ of Construction Science (IETcc-

CSIC), Spain. I benefited a lot from the insightful and valuable discussions on electrochemistry

and the integral corrosion test with Prof. Carmen Andrade. I really appreciate all her efforts and

instruction.

Gerrit Nagtegaal, Arjan Thijssen and John van den Berg from M&E, Agnieszka Kooijman

and Jose Cerezo Palacios from Corrosion lab of Faculty 3mE and Hans Beijersbergen van

Henegouwe from TNO are especially thanked for their indispensable technical assistance during

my lab work. Kees van Beek, Ron van Leeuwen and Marten van der Meer from DEMO, CiTG

are highly appreciated for their help in the design of multichannel electrochemical testing setup.

The conversations and many constructive discussions in particular with Kees van Beek are

unforgettable and worthy to be specially acknowledged.

Office time counts for one of the most important parts of a PhD student’s life. In this respect,

I was extremely lucky to share my office with my good friends, Rafid Al-Khoury for the first

year, and Frans van der Meer for the last three years of my PhD. I really enjoyed our daily

conversations in all aspects of life and science. I have been trying to understand Dutch culture

and language during the past four years, but my skills were far below the level to translate the

propositions and summary of this thesis. Many thanks to Frans van der Meer, Renee Mors, Nynke

Verhulst and of course Rob for helping me out on these parts.

I would like to thank all the colleagues in the Section of M&E for creating a nice working

environment and the friendship. It was a great pleasure to work with all of you. Many thanks go

to our secretaries Nynke Verhulst, Melanie Holtzapffel, and Claudia Baltussen for their kind help

on all kinds of documents and various daily issues. I would like to extend my special thanks to all

members of our Chinese committee in M&E, in particular Dr. Guang Ye and also to Dr. Xueyan

Liu from the Section of Pavement Engineering for your direct and indirect helps and the

enjoyable talks during the daily life. It was really memorable to meet you all in The Netherlands.

I am forever indebted to my parents for their endless love and encouragement. My deep

heartfelt gratitude goes to my sisters and brothers and their families for their encouragement,

generous support and always being there for me especially during my years’ stay abroad.

Last but certainly not least, I wish to express my gratitude to my wife Jinrui and our lovely

daughter Jessica (Tiantian) for their love, support and patience, and for giving me the things that

are so much more important in life than whatever else.

Zhengxian Yang

Delft, March 2015

Curriculum Vitae

Zhengxian Yang was born in Huainan, Anhui Province, China. He received his Bachelor’s degree

in Applied Chemistry and Master’s degree in Physical Chemistry both from China. In 2005, he

was appointed by Fuyang Normal College (FNC) as an assistant professor/lecturer in the

Department of Chemistry. At FNC, his job responsibilities involved teaching several courses such

as “Physical Chemistry” and “Applied Electrochemistry” at the undergraduate level, advising

undergraduate research assistants and developing collaborative and independent research projects.

Due to the dynamic expansion of his research initiatives, he was serving as a Principal

Investigator or Co-Investigator for several grants and contracts from university sources and

government departments. In 2007, he moved to the United States and joined Montana State

University-Bozeman (MSU) working as a Research Assistant/Associate in Western

Transportation Institute (WTI) at the College of Civil Engineering. Later, he was promoted to be

a Research Scientist due to his impressive performance. In WTI, he was actively involved in

several research projects, which aimed to improve the sustainability and durability of cementitous

materials, especially in the area of reinforcement corrosion, self-healing concrete, polymer

modified concrete and the application of new functional materials in cementitious composites

such as nanotubes and nanoclay for improving the transport properties and synthetic

microcapsules along with microfibers for self-healing cracking in concrete. He also worked on

the innovative use of recycled and waste materials such as the application of high volume fly ash,

waste carpet and paper/cellulose fibers in concrete.

As one of important steps in his research endeavor, he came to the Netherland in October

2010 working with Materials innovation institute (M2i) and started his PhD project under the

supervision of Prof. R.B Polder in Section of Materials and Environment of the Faculty Civil

Engineering and Geosciences at Delft University of Technology. The outcomes of the project

have attracted industrial attention. Kisuma Chemicals BV (a main hydrotalcites producer/supplier

in world market) and CRH (one of the worldwide leading cement and concrete companies) have

expressed their interests for further cooperation. He has published about 40 papers in peer-

reviewed journals and international conferences. In 2013, he was awarded the prestigious

“National Award for Outstanding Self-financed Chinese Students Study Abroad” by China

Ministry of Education for his impressive overall performance during PhD study.

List of Publications (selected)

Refereed journal papers

1. Yang, Z., Fischer, H., Polder, R. Laboratory Investigation of the Influence of Two Types of

Modified Hydrotalcites on Chloride Ingress into Cement Mortar. Cement and Concrete

Composites, 2015, 58: 105-113.

2. Yang, Z., Fischer, H., Polder, R. The Effect of Modified Hydrotalcites on Mechanical

Properties and Chloride Penetration Resistance in Cement Mortar. Key Engineering Materials,

2015(629-630):156-16.

3. Yang, Z., Fischer, H., Polder, R. Synthesis and Characterisation of Modified Hydrotalcites

and Their Ion Exchange Characteristics in Chloride-rich Simulated Concrete Pore Solution.

Cement and Concrete Composites, 2014, 47: 87-93.

4. Yang, Z., Fischer, H., Cerezo, J., Mol, J.M.C., Polder, R. Aminobenzoate Modified Mg-Al

Hydrotalcites As a Novel Smart Additive of Reinforced Concrete for Anticorrosion

Applications. Construction and Building Materials, 2013, 47, 1436-1443.

5. Yang, Z., Fischer, H., Polder, R. Modified Hydrotalcites as A New Emerging Class of Smart

Additive of Reinforced Concrete for Anti-corrosion Applications: A Literature Review.

Materials and Corrosion, 2013, 64(12):1066-1074.

6. Yang, Z., Hollar, J., He, X., Shi, X. A Self-healing Cementitious Composite Using Oil

Core/Silica Gel Shell Microcapsules. Cement and Concrete Composites, 2011, 33(4): 506-

512.

7. Yang, Z., Hollar, J., Shi, X. Surface-sulfonated Polystyrene Microspheres Improve Crack

Resistance of Carbon Microfiber-reinforced Portland Cement Mortar. Journal of Materials

Science, 2010, 45(13): 3497-3505.

8. Yang, Z., Hollar, J., Shi, X., He, X. Laboratory Assessment of A Self-healing Cementitious

Composite. Transportation Research Record: Journal of the Transportation Research Board,

2010, 2142: 9-17.

9. Yang, Z., Shi, X., Creighton, A. T., Peterson, M. M. Effect of Styrene-Butadiene Rubber

Latex on the Chloride Permeability and Microstructure of Portland Cement Mortars.

Construction and Building Materials, 2009, 23(6): 2283-2290.

10. Han, B., Yang, Z*., Shi, X, Yu, X. Transport Properties of Carbon-Nanotube/Cement

Composite. Journal of Materials Engineering and Performance, (co-first author and equal

contribution), 2013, 22(1): 184-189.

11. Shi, X., Yang, Z., Liu, Y., Cross, D. Strength and Corrosion Properties of Portland Cement

Mortar and Concrete with Mineral Admixtures. Construction and Building Materials, 2011,

25(8): 3245-3256.

12. Shi, X., Yang, Z., Nguyen, T.A., Suo, Z., Avci, R., Song, S. An Electrochemical and

Microstructural Characterization of Steel-Mortar Admixed with Corrosion Inhibitors. Science

in China, Series E: Technological Sciences, 2009, 52(1): 52-66.

List of Publications | 167

13. Shi, X., Fay, L., Yang, Z., Nguyen, T.A., Liu, Y. Corrosion of Deicers to Metals in

Transportation Infrastructure: Introduction and Recent Developments. Corrosion Reviews

2009, 27(1-2): 23-52.

14. Shi, X., Fay, L., Peterson, M.M., Yang, Z. Freeze-thaw Damage and Chemical Change of a

Portland Cement Concrete In the Presence of Diluted Deicers. Materials and Structures 2010,

43(7):933-946.

15. Shi, X., Fortune, K., Fay, L., Smithlin, R., Yang, Z., Cross, D, Wu, J. Longevity of Corrosion

Inhibitors and Performance of Anti-icing Productss after Pavement Application: A case study.

Cold Regions Science and Technology, 2012, (83-84): 89-97.

16. Shi, X., Akin, M., Pan. T., Fay, L., Liu, Y., Yang, Z. Deicer Impacts on Pavement Materials:

Introduction and Recent Developments. The Open Civil Engineering Journal, 2009, 3: 16-27.

Manuscripts under internal review

1. Yang, Z., Polder, R., Mol, J.M.C., Andrade, C. The Effect of Two Types of Modified Mg-Al

Hydrotalcites on Reinforcement Corrosion in Cement Mortar. Cement and Concrete Research,

2015.

2. Yang, Z., Cerezo, J., Mol, J.M.C., Fischer, H., Polder, R. Modified Hydrotalcites as New

Smart Additives of Reinforced Concrete for Anticorrosion Applications-preparation,

characterization and an Assessment in Alkaline Chloride Solution. Materials and Corrosion,

2015.

3. Yang, Z., Liu, J., Mol, J.M.C., Polder, R. Investigation of Inhibition Effect of Some Amino

Acids against Steel Corrosion in Chloride-containing Alkaline Solution. Journal of Applied

Electrochemistry, 2015.

4. Yang, Z., Mao, M., Mol, J.M.C., Polder, R. An Electrochemical Investigation of the

Inhibition Effect of Two Modified Hydrotalcites in Simulated Concrete Pore Solution.

Journal of Electrochemistry Society, 2015.

5. Yang, Z., Ouyang, X., Li, H., Cheng, G., Shi, X., Polymer Modification for Improved

Durability of Portland Cement Concrete: a literature review. Construction and Building

Materials, 2015.

Conference proceedings/Oral presentations in international conferences

1. Yang, Z., Fischer, H., Polder, R. The Effect of Modified Hydrotalcites on Mechanical

Properties and Chloride Penetration Resistance in Cement Mortar. The 10th International

Symposium on High Performance Concrete–Innovation & Utilization. September 16-18, 2014,

Beijing, China. (Invited lecture and the paper was selected as one of the best papers by

the conference).

2. Yang, Z., Fischer, H., Polder, R. The Application of Modified Hydrotalcites as Chloride

Scavengers and Inhibitor Release Agents in Cement Mortars. Concrete Solutions 2014, 5th

International Conference on Concrete Repair. September 1-3, 2014, Belfast, UK.

3. Yang, Z., Fischer, H., Cerezo, J., Mol, J.M.C., Polder, R. Modified Hydrotalcites as New

Smart Additive of Reinforced Concrete for Anticorrosion Applications-An Assessment in

Chloride-containing Alkaline Solutions. European Corrosion Congress (EUROCORR).

September 1 - 5, 2013, Estoril, Portugal.

http://www.freundpublishing.com/JOURNALS/materials_science_and_engineering.htm

168 | List of Publications

4. Yang, Z., Fischer, H., Polder, R. The Application of Modified Hydrotalcites as Chloride

Scavenger in Cement Mortars. Proceedings 6th International RILEM PhD Student Workshop

on Durability of Reinforced Concrete. July 4-5, 2013, Delft, The Netherlands. (This paper

was selected by the workshop for publication in Springer’s book).

5. Yang, Z., Fischer, H., Polder, R. Aminobenzoate Modified Hydrotalcites as a Novel Smart

Additive of Reinforced Concrete for Anticorrosion Applications. Proceedings of the 3rd

International Conference on Concrete Repair, Rehabilitation and Retrofitting. September 3-5,

2012, Cape Town, South Africa.

6. Yang, Z., Fischer, H., Polder, R. A New Smart Additive of Reinforced Concrete Based on

Modified Hydrotalcites: Preparation, Characterization and Anticorrosion Applications.

Proceedings of XXI International Materials Research Congress. August 12-17, 2012, Cancún,

Mexico.

7. Yang, Z., Fischer, H., Polder, R. A New Additive for Improved Corrosion Protection of

Reinforced Concrete by Modified Hydrotalcites, Proceedings of 1st International Congress on

Durability of Concrete (ICDC). June 18-21, 2012, Trondheim, Norway. (This paper was

selected as one of the best papers by the conference).

8. Yang, Z., Fischer, H., Polder, R. A Feasibility Study of a New Application of Modified

Hydrotalcites in Corrosion Protection of Reinforced Concrete. Proceedings of the 2nd

International Conference on Microstructure Related Durability of Cementitious Composites.

April 11-13, 2012, Amsterdam, Netherlands.

9. Yang, Z., Fischer, H., Polder, R. Synthesis of Modified Hydrotalcites and Preliminary

Evaluation of Their Corrosion Protection Effectiveness for Reinforced Concrete. Proceedings

5th International RILEM PhD Student Workshop on Durability of Reinforced Concrete.

February 9-10, 2012, Helsinki/Espoo, Finland.

10. Yang, Z., Hollar, J., Shi, X. Chloride Permeability and Crack Resistance of Cement

Composite Incorporating Pulp Microfibers and Nanoclay. Abstract submitted to the 2011

NACE International Annual Corrosion Conference and Exposition, March 13-17, 2011,

Houston, TX, USA.

11. Yang, Z., Fischer, H., Polder, R. Possibilities for Improving Corrosion Protection of

Reinforced Concrete by Modified Hydrotalcites – A Literature Review. Proceedings of 4th

International RILEM PhD Workshop on Durability of Reinforced Concrete. November 19-20,

2010, Madrid, Spain. (This paper was selected by the workshop for publication in

Springer’s book).

12. Yang, Z., He, X., Shi, X., Peterson, M.M. A Self-healing Coating System Using Passive

Smart Microparticles for the Corrosion Protection of Galvanized Steel. Abstract submitted to

the 2010 NACE International Annual Corrosion Conference and Exposition, March 14-18,

2010, San Antonio, TX, USA.

13. Yang, Z., Hollar, J., Shi, X. Laboratory Assessment of a Self-healing Cementitious

Composite. The First International Conference on Nanotechnology in Cement and Concrete,

May 5-7, 2010, Irvine, CA, USA.

14. Yang, Z., Hollar, J., He, X., Shi, X. Feasibility Investigation of Self-healing Cementitious

Composite Using Oil Core/Silica Gel Shell Passive Smart Microcapsules. Proceedings of the

Second International Conference on Smart Materials and Nanotechnology in Engineering

(SMN2009). July 8-11, 2009, Weihai, China.

List of Publications | 169

15. Yang, Z., Shi, X., Creighton, A. T., Peterson, M. M. Effect of Styrene-Butadiene Rubber

Latex on the Chloride Permeability and Microstructure of Portland Cement Mortars.

Proceedings of International Materials Research Conference (IMRC), June 9-12, 2008,

Chongqing, China.

16. Shi, X., Nguyen, T.A., Yang, Z. An Electrochemical and Microstructural Characterization of

Steel-Mortar Admixed with Corrosion Inhibitors. Proceedings of International Materials

Research Conference (IMRC), June 9-12, 2008, Chongqing, China.

Documents

Modified Hydrotalcites as Smart Additives for Improved