6/7/2012

1

Recent Advances in ASR Test Methods and Understanding

Mitigation Mechanisms, Part 2

ACI Spring 2012 ConventionMarch 18 – 21, Dallas, TX

Karen Scrivener graduated from University of Cambridge in 1979 in Materials Science. She went on to do a PhD on “The Microstructural Development during the Hydration of Portland Cement” at Imperial College, London completed in 1984. She remained at Imperial College until 1995 as Royal Society Research Fellow and then lecturer, heading the

Cement and Concrete Group in the Department of Materials. In 1995 she joined the Central Research Laboratories of Lafarge near Lyon in France as Head of research on Calcium Aluminate cements and expert of concrete durability in general. In March 2001 she was appointed as Professor and Head of the Laboratory of Building Materials, Department of Materials at the Swiss Federal Institute of Technology at Lausanne (EPFL, Ecole Polytechnique Fédérale de Lausanne), Switzerland. She created and is co-ordinating NANOCEM – the industrial-academic research network on cement and concrete which brings together 15 industrial and 24 academic partners. She is Editor-in-Chief of the leading academic journal in the field – Cement and Concrete Research.

Théodore ChappexKaren Scrivener

Understanding the role ofSupplementary Cementitious Materials

in mitigating Alkali Silica Reaction

4 From J. Hideker

Alkali Silica Reaction

Reaction between:

Alkali

Water

Amorphous silica

Long‐term expansion reaction present in certain concretes

Alkali Silica Reaction

6 From J. Hideker

Alkali Silica Reaction

6/7/2012

2

Amorphous SiO2:

Expansive gel:

Alkali Silica Reaction Effect of blended pastes on ASR expansion

SCMs are effective in reducing deleterious ASR (empirical additions):

Silicon and Aluminium addition are involved in the reduction of expansion

Aluminium rich SCMs are more efficient against ASR

The exact mechanism by which it happens is unclear!

5%

15%SFQ

15%MK

10%SFQ

10%MK

OPCMortar bars in 0.6M NaOHAt 38°C

Presentation outline

Influence of Si and Al onalkali fixation in C‐S‐H

11

Systems studied

Experimental systems:

MK

OPC

Q - Filler

SF

OPC

Si

Al

95, 90, 85%w

5, 10, 15%w

7.65, …%w

7.35, …%w

Paste sample

Pore solution

Piston

EDS Pore solution extraction TGA

C-S-H EDS analysis

13

The Si/Ca increase with increasing substitution for both systems (MK and SF‐Q)

The Al/Ca is constant for SFQ at all substitutions levels and increase with MK substitution

Pastes can  be compared in term of pore solution concentration

300 days:

100

150

200

250

300

350

400

450

K [m

mol/l]

OPC

5SFQ

5MK

10SFQ

10MK

15SFQ

15MK

0

50

100

150

0 100 200 300 400 500

Na [m

mol/l]

me [day]

5MK

5SFQ

10MK

10SSFQ

15SFQ

15MK

Pore solution analysis28 – 90 ‐ 300 days:

‐ Lowering of pH is confirmed

‐ No improve of fixation is observed up to 2 years in Al rich systems

Al doesn’t increase the alkali fixation capacity of C‐S‐H in blended pastes!

Another phenomena is involved to control ASR in presence of Al!

6/7/2012

3

Due to difficulties to quantify all the phases in presence in the blends (XRD/EDS/TGA), a comparative study was done, using the TGA water loss between 30 and 150°C, corresponding to the main C-S-H water loss.

C-S-H fixation capacity estimation

(Amount of free alkalis)

=

(Batch water) – (Total bounded water(800°C))

X

(Alkali concentration from the pore solution)

(Bound alkalis)

=

(Total alkalis of the system)__

(Amount of free alkalis)

Comparatively, the fixation capacity of SFQ and MK are similar. Aluminium has no influence on the fixation capacity of alkalis

C-S-H fixation capacity estimation: comparative

New approach:  focus on the  aggregates

90 days: 300 days:

Presentation outline

18

The effect of aluminium on silica dissolution

Pore solution composition of MK pastes

MK systems provide aluminium ions in the pore solution – A peak of aluminium appears during the first 90 days

19

0

0.5

1

1.5

2

2.5

3

3.5

0 100 200 300 400 500 600

Al [mmol/l]

me [day]

OPC

15MK

10MK

5MK

Silica dissolution in presence of aluminium ions

« Metal ions, particularly Al, reduce the rate of dissolution of amorphous silica»  [Lewin, 1961]

“only 1.35 Al atoms/nm2 was adsorbed (less than one atom layer), and this was sufficient to minimize the solubility of silica”

[Iler, 1973]

20

New approach: Focus on the aggregates

Marine chemistry and geology:

Prepared frustules (amorphous silica) in sea water with and without Al: [Koning, 2007]

Multiscale approach

Concrete / Mortar Reactive aggregates& pore solution

Amorphous silica& pore solution

Is this mechanism present in concrete systems?High alkalinity – Ca saturated

Multi scale approach:

6/7/2012

4

Concrete / Mortar

simulated pore solutionNo leaching of alkalis, good 

reproducibility0.6 M NaOH

Al(OH)3 saturated

Mortar barIllsee aggregate

System at 38°CTo accelerate ASR

Mortar expansion test

MK

OPC

Q - Filler

SF

OPC

Si

Al

95, 90, 85%w

5, 10, 15%w

7.65, …%w

7.35, …%w

23

Mortar expansion test

5%

15%SFQ

15%MK

10%SFQ

10%MK

OPC

The Al rich systems (MK) behave better than Al free sytems (SFQ)

The decrease in expansion goes from 10 to 30% by increasing SCM substitution

Concrete / Mortar Reactive aggregates& pore solution

The effect of Al on silica dissolution

Solutions:

‐Al ions concentrations corresponding to blended pastes(0mM / 4mM / 10mM / 30mM / saturated)

‐ 38°C / 60°C

‐ Portlandite saturated

‐0.6 M NaOH

Aggregate:

‐ Reactive Swiss Alps aggregates

Reactive aggregates test

25

Reacted fraction analysis (SEM‐Image analysis):

Percentage of gel pocket surface in function of total aggregate surface

[Ben Haha & al.]

26 27

6/7/2012

5

28

There is a clearinfluence of 

aluminium ions on aggregates gel formation!

Concrete / Mortar Reactive aggregates& pore solution

Amorphous silica& pore solution

Conc. Al(OH)4

‐ 0.6/0.2 M NaOH XPS/SEM analyses

0 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

1.5 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

2.5 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

5 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

10 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

15 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

20 mM CH sat/CH free – 20°C 14 / 28 / 60 / 90 / 120 days

CaCa

0.2/0.6 M Na

0.2/0.6 M Na

X mM Al

How does Al incorporate on SiO2 in paste pore solution conditions?

Fundamental study on silica

XPS on fused silica

Quantification

Binding typeSi-Al - Ca incorp.Structural energy

barrierBond strength

definition

How does Al incorporate on SiO2 in paste pore solution conditions?‐fct of alkalinity‐Ca saturated

Fundamental study on silica

30

SEM on fused silica

Surface stateprecipitates -Dissolution

CaCa

0.2/0.6 M Na

0.2/0.6 M Na

X mM Al

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.0 5.0 10.0 15.0 20.0 25.0

Al/Si

incorp

[%at]

Al in solu on [mM]

60d 0.2 60d 0.6

M NaM Na

Aluminium incorporation

31

Aluminium is incorporated on the surface of silica

Higher incorporationat lower pH (at 28 / 60 / 120 days)

+ 25%

Ca

0.2 / 0.6 M Na

X mM Al

Proposed Mechanism

Mechanism of dissolution of amorphous silica is similar to quartzif considered in term of chemical bounding

Dissolution of Q3 species make appears a Q2 surface, more reactive Q2 group retreat (step retreat like)

[Dove, 2008]

Proposed Mechanism

Al

Al(OH)-4 co-adsorb with Na+

deprotonated silanol groups are associated with Na+ sorbed in the outer-sphere plane

Al is incorporated in the silica framework

The Al ion stabilize the dissociated form of terminal silanol

[Labrid & Duquerroix 1991, Bickmore 2005]

[Koning, 2007]

[Trombetta, 2000]

[Dove, 2008]

6/7/2012

6

The aim of this test is to provide the same initial surface state for all the treatments

Differences on the 4 sizes were visible only after 90 days

Step 1:

Pre treatment:

30 mM Al / 0.2M Na

solution

90 days

The effect of aluminium on silica dissolution –« 4 sides test »

The aim of this test is to provide the same initial surface state for all the treatments

Differences on the 4 sizes were visible only after 90 days

Step 2:

Post treatment:28 /

60 / 90 / 120 days

0.6M Na solution

Step 1:

Pre treatment:

30 mM Al / 0.2M Na

solution

90 days

The effect of aluminium on silica dissolution –« 4 sides test »

Step 2:

Post treatment:28 /

60 / 90 / 120 days

0.6M Na solution

Step 1:

Pre treatment:

30 mM Al / 0.2M Na

solution

90 days

90 days post

treatment: Conclusion

37

It was showed that the aluminium present in the C-S-H of blended pastes doesn’t

decreases the alkalinity of the system

Al incorporation and effect was successfully showed for concrete condition

The mechanism controlling silica dissolution in presence of Al was partially explained and

confirmed

Al is incorporated in the silica framework and probably stabilizes the step retreat

dissolution of the Q2 species

The alkalinity has an influence on the aluminium incorporated fraction. The lower

alkalinity present a higher aluminium incorporation

Aluminium effect has to be combined with an alkalinity reduction to be

more efficient (visible on expansion curves?)

Approximation of dissolution rate? prediction of gel formation rate expansion!

38

Thank you