CIRIA C530 London 2001

On-site curing of concrete –

microstructure and durability

N R Buenfeld PhD MSc DIC CEng MICE MICT

R Yang PhD MSc

Century House, Telford Avenue, 6 Storey’s Gate, Westminster,

Crowthorne, Berks RG45 6YS London SW1P 3AU
TELEPHONE 01344 466007 TELEPHONE 020 7222 8891
FAX 01344 466008 FAX 020 7222 1708
EMAIL [email protected] EMAIL [email protected]
WEBSITE www.concrete.org.uk WEBSITE www.ciria.org.uk
 Summary

This book describes the second phase of an investigation into the influence of practical
on-site curing on the durability of concrete.

Scanning electron microscopy was used to evaluate the influence of di fferent curing
regimes and environmental exposures on the porosity and degree of hydration in the
“ curing affect ed zone” (CAZ), and their influence on durability and abrasion resistance.

The viability of the method was established.

On-site curing of concrete – microstructure and durability

Buenfeld, N R and Yang, R

Construction Industry Research and Information Association

CIRIA C530 © CIRIA 2001 ISBN 0 86017 530 8

Keywords
Curing, concrete, durability, abrasion resistance, carbonation, environmental exposure

Reader interest Classification

Design, specification, AVAILABILITY Unrestricted
construction and
CONTENT Investigation into practical on-site
supervising engineers
curing and durability
involved in building and
civil engineering works STATUS Committee-guided
USER Construction professionals

Published by CIRIA, 6 Storey’s Gate, Westminster, London SW1P 3AU. All rights
reserved. No part of this publication may be reproduced or transmitted in any form or by
any means, including photocopying and recording, without the written permission of the
copyright-holder, application for which should be addressed to the publisher. Such
written permission must also be obtained before any part of this publication is stored in
a retrieval system of any nature.

2 CIRIA C530
 Acknowledgements

The project leading to this publication is part of a collaborative programme between

CIRIA and the Concrete Society entitled Concrete techniques – site operations. It was
carried out under contract by The Imperial College of Science Technology and Medicine
and Taywood Engineering .

The report was written by Dr N. R. Buenfeld and Dr R. Yang of the Department of Civil
and Environmental Engineering at Imperial College.

The project was carried out and the report prepared under the guidance of the following
Steering Group members:
Mr R A McClelland (chair) Alfred McAlpine Construction Ltd
Dr P Bamforth Taywood Engineering
Mr S Crompton Ready Mix Concrete (UK) Ltd
Mr R Cather Arup Research and Development
Dr C Clear Civil and Marine Slag Cement Ltd
Mr D S Leek Mott MacDonald, Special Services Division
Mr J Frearson Rugby Cement (formerly Messrs Sandbergs)
Mr G P Hammersley Building Research Establishment Ltd
Dr T Harrison Quarry Products Association
Dr C Hopkins Wardell Armstrong
Mr N Loudon Highways Agency
Dr E Kay Sir William Halcrow and Partners
Mr L H McCurrich FOSROC International Technology
Mr M Messham Sir Alexander Gibb & Partners
Dr J B Newman Imperial College of Science Technology
and Medicine
Mr P Owens Quality Ash Association
Mr P F Pallet Training Consultant
Dr W Price Sandberg Consulting Engineers
Mr R Roberts Concrete Advisory Service
Mr G P Tilley Gifford and Partners
Mr P Titman Edmund Nuttall
Mr C Turton Design Group Partnership
Mr M Walker The Concrete Society
Mr P Woodhead Department of the Environment, Transport
and the Regions.

CIRIA C530 3
 The Steering Group delegated review of the detailed development to the following
specialist Technical Committee:
Mr R Cather (chair) Arup Research & Development
Mr R Baldwin Mott MacDonald
Mr G Hammersley Building Research Establishment Ltd
Mr A Harrison Rugby Cement
Dr C Hopkins Wardell Armstrong
Mr L Parrott British Cement Association
Mr P Titman Edmund Nuttall.

CIRIA’s research manager for this project was Dr B W Staynes.

Contributions made to this work by the following are gratefully acknowledged: Dr Colin
Hills, Emeka Agbasi, Roy Baxter and Eleanor Tipping at Imperial College who helped
with the SEM method development described in Section 5.1.2; Dr Phil Bamforth, David
Cullen and Andrew Pearce of Taywood Engineering; and Dr Mamoud Sadegzadeh of
Aston Materials Services Ltd who carried out the abrasion testing.

This project was funded by the Department of the Environment, Transport and the
Regions, the Concrete Society and CIRIA’s Core programme sponsors.

CIRIA, the Concrete Society and the authors gratefully acknowledge the support of
these funding organis ations, and the technical help and advice provided by the members
of the steering group and by the many other individuals who were consulted.
Contributions do not imply that individual funders necess arily endorse all views
expressed in published outputs.

4 CIRIA C530
 Executive summary

Curing is widely perceived as being an important factor in achieving durable concrete

structures. This seems reasonable in that curing allows hydration of cement to continue
which is expected to reduce capillary porosity, thereby strengthening the concrete and
increasing its resistance to penetration by aggressive agents such as chloride, sulphate
and carbon dioxide. However, direct evidence that the levels of curing achi eved on-site
increas es durability is scarce.

This report describes the second phase of an investigation into the influence of practical
on-site curing on the durability of concrete. The first phase was a literature revi ew
(CIRIA PR49 by Hammersley et al, 1997), which arrived at limited conclusions
concerning the effects of site curing on durability. The review also concluded that it was
extremely unlikely that any currently established laboratory or site test method could
form the basis of the assessment of on-site curing over the range of environm ental
exposures. The Project steering group for this second phase concluded that the
assessment of curing effectiveness and its influence on durability perform ance could be
most benefici ally determined by considering the microstructure of the concrete in the
curing affected zone (CAZ). This is the zone of concrete near to and including its outer
surface, which can be directly affected by active external curing measures.

The objective of this second phase of the project was primarily to establish the viability
of a method for investigating the link between practical curing methods and the micro-
structure of concrete in the CAZ. A further objective was to investigate the relationship
between microstructure in the CAZ and durability performance.

How should concrete microstructure be charact erised? The limited literature suggested
that, under some conditions, the CAZ is limited to a depth of only a few millimetres. It
was therefore decided that a microscopy-based technique could be appropriate in that it
would allow differences over small distances to be examined. To enable these to be
quantifi ed necessitated image analysis. The approach adopted to characterise concrete
microstructure was therefore to obtain images of the concrete using scanning electron
microscopy (SEM) and then to convert these images into microstructural parameters
using computer-based image analysis.

In a “ backscattered electron image” produced by the SEM (see Section 5.1.1 and
Appendix A2 for a background to SEM), polished surfaces of concrete show at least five
distinct phases according to their brightness (greyscale).

1. Porosity.
2. Calcium silicate hydrate gel.
3. Aggregat e.
4. Calcium hydroxide.
5. Anhydrous cement.

CIRIA C530 5
 The two phases of most interest were porosity (becaus e of its link with durability) and
anhydrous cement (closely linked to degree of hydration, the parameter most directly
characterising the degree of curing).

Preliminary work was undertaken to develop the SEM method and to demonstrate that
differences in porosity and anhydrous cement content could be identified and quanti fied
(see Section 5.1.2). This involved determining the optimal magnification (´ 500) and
other SEM and image analyser variables, developing methods of automatically
collecting images at predefined positions on a specimen, separating the aggregate from
the cement paste by image analysis, and determining the number of images required to
produce statistically acceptable results.

The method was then applied in a large test programme executed over a 12-month
period. Ten concrete test panels, measuring 1 m ´ 0.5 m ´ 0.15 m, of four di fferent
concretes (two of them including PFA or GGBS) were cured in up to four different
ways and were subject ed to nine months of natural exposure (outdoors, unsheltered
or sheltered) in north-west London. The curing regimes were chosen to represent the
range of regimes applied on construction sites and involved stripping formwork at day 1
and then:
· no additional curing
· immediate application of a curing membrane
· one day of moist curing (wet hessian sealed in polythene sheet)
· seven days of moist curing (wet hessian sealed in polythene sheet).

Porosity and anhydrous cement content were measured at depths of 0.5 mm, 3.5 mm,
6.5 mm and 20.5 mm into the concrete shortly after curing (at an age of seven days) and
again after the nine-month exposure period. This allowed the link between practical
curing methods and the microstructure of concret e in the CAZ to be investigated. After
nine months of exposure the panels were also tested for abrasion resistance and
carbonation depth, to allow the relationship between microstructure in the CAZ and
some aspects of durability performance to be investigated. Abrasion resistance and
carbonation depth were select ed as durability performance measures – not because they
are necessarily the mechanisms of great est concern to concrete structures, but because
the first phase of this project highlighted that they are most directly dependent on curing
and are also the least complex. Half of the panels have been returned to the exposure site
for long-term exposure.

The combination of backscattered electron imaging of polished surfaces in the SEM and
greyscale image analysis was confirmed to be a useful method for assessing the effects
of curing, being able to quantify microstructural gradients in coarse porosity and
anhydrous cement content in the cement paste fraction of concrete. The technique only
identifies the larger capillary pores (s ay > 0.5 µm) and does not have the resolution to
pick up the finer capillary pores, nor any of the gel pores. Consequently the porosity
determined by the technique represents only a small fraction of the total porosity
present. In seven-day old OPC concretes, around 42 per cent of the theoretical capillary
porosity (calculated based on the measured degree of hydration and free wat er/cement
ratio using the method described in Neville, 1995) was resolved at a free wat er/cement
ratio of 0.70, reducing to around 23 per cent at a free wat er/cement ratio of 0.55.
However, it is the coarser capillary pores that have the major influence on durability.

It was found that at early ages (eg seven days ) curing regime can have a significant
effect on the pore structure of concrete very near to the cured surface; poor curing
results in higher porosity paste near the surface in relation to the bulk concrete.

6 CIRIA C530
 An unexpected finding was that differences in curing that resulted in porosity profiles
had very little effect on anhydrous cement content. A possible explanation is that the
ambient humidity (around 70–80 per cent relative humidity, see Appendix A1) exper-
ienced by panels provided with no additional curing after one day in formwork was
suffi cient to empty larger capillary pores some distance from cement grains, but not the
finer capillary pores closer to the cement grains. Consequently hydration continued, but
near the concrete surface the resulting hydration products were formed away from the
larger capillary pores detected in the SEM.

Microstructural benefits gained from curing were found to diminish with long-term
exposure, with the effects of nine months of UK exposure (outdoors unsheltered or
sheltered) having a greater influence on microstructure than initial curing. This resulted
from both further hydration of the cement and carbonation of the concrete and made it
impossible to identify a CAZ depth by nine months.

Exposure environment (sheltered or unsheltered) also had more effect on carbonation

depth than initial curing. Nevertheless, both abrasion resistance (Chapter 3) and carbon-
ation depth (Chapter 4) were confirm ed as being dependent on curing and correlat ed
well with porosity measured by SEM (at nine months) in the zone affected (surface few
millimetres). Lower porosity was associated with increased abrasion resistance (Figure
6.2) and reduced carbonation depth (Figure 6.5). The correlations were good, despite
being based on four di fferent concretes. No correlation was observed between abrasion
resistance or carbonation depth and anhydrous cement content when the data from the
four concretes tested were considered together. However, when each concrete was
viewed on its own, the expected trend of increasing abrasion resistance and reducing
carbonation depth with reducing anhydrous cement content was usually apparent.

Recommendations are made for further res earch. These fall into two categories:
· those related to test method development
· those concerned with quantifying the effects of curing on concrete microstructure/
durability.

This latter group is ultimately aimed at establishing the most efficient way of producing
durable concrete on site, with particular attention to curing. It would allow guidelines to
be established for design/construction which could include “ trade-offs” between binder
type, water/cement ratio, curing regime and cover depth for di fferent exposure
environments. It will enable cover rules and standards to be revised on the basis of
improved knowledge and understanding.

Thus six conclusions are drawn.

1. Backscattered electron imaging of polished surfaces in the SEM combined with

greyscale image analysis is a useful method for assessing the effects of curing.
2. At early ages (eg seven days) curing regime can have a significant effect on the
pore structure of concrete very near to the cured surface; poor curing results in
higher porosity paste near the surface in relation to the bulk concrete.
3. Microstructural benefits gained from curing diminish with exposure, with the
effects of nine months of UK exposure (outdoors unsheltered or sheltered) having a
greater influence on microstructure than curing, such that the full extent of the CAZ
could not be identified by nine months.

CIRIA C530 7
 4. Both abrasion resistance and carbonation depth were dependent on curing and
correlated well with porosity measured by SEM (at nine months) in the zone
affected, ie in the surface few millimetres. Lower porosity was associated with
increas ed abrasion resistance and reduced carbonation depth.
5. The results confirm the importance of curing surfaces likely to be exposed to
abrasion early in their lives.
6. Further work is required to establish guidelines concerning the most efficient way
of producing durable concrete on site, taking account of “ trade-offs ” between
choice of curing regime and selection of binder type, water/cement ratio and cover
depth depending on the exposure environment.

8 CIRIA C530
 Contents

Summary........................................................................................................ 2

Acknowledgements .......................................................................................... 3

Executive summary .......................................................................................... 5

List of figures .................................................................................................11

List of tables ..................................................................................................12

Abbreviations .................................................................................................12

1 Introduction..............................................................................................13
1.1 Background ...................................................................................13
1.2 Objectives .....................................................................................13
1.3 Approach.......................................................................................13

2 Panels ......................................................................................................17
2.1 Concretes ......................................................................................17
2.2 Panel details...................................................................................18
2.3 Curing ..........................................................................................18
2.4 Seven-day sampling.........................................................................18
2.5 Exposure .......................................................................................19
2.6 Nine-month sampling.......................................................................20

3 Abrasion resistance....................................................................................21
3.1 Method .........................................................................................21
3.2 Results..........................................................................................21
3.3 Discussion.....................................................................................21

4 Carbonation depth.....................................................................................23
4.1 Method .........................................................................................23
4.2 Results..........................................................................................23
4.3 Discussion.....................................................................................23

5 Scanning electron microscopy (SEM)...........................................................25

5.1 Method .........................................................................................25
5.1.1 Brief background to SEM and image analysis........................25
5.1.2 Method development ........................................................25
5.1.3 Specimen preparation........................................................30
5.1.4 Effect of SEM sample storage conditions on results................30
5.2 Results..........................................................................................31
5.3 Discussion.....................................................................................44
5.3.1 Effect of concret e type on bulk results..................................44
5.3.2 Effect of curing on seven-day results....................................45
5.3.3 Effect of exposure on nine-month results ..............................46
5.3.4 Effect of curing on nine-month results..................................49
5.3.5 Correlation between porosity and anhydrous cement content ....49

CIRIA C530 9
 6 Correlations between abrasion, carbonation and SEM results.........................53

7 Conclusions ..............................................................................................59
7.1 Effects of practical on-site curing regimes on concrete microstructure .......59
7.2 Relationship between concrete microstructure and durability...................59
7.3 Effects of curing on durability ...........................................................60
7.4 SEM as a method for assessing the effects of curing...............................60

8 Recommendations for further research ........................................................63

8.1 Test method development .................................................................63
8.2 Effects of curing on concrete microstructure/durability...........................63

Appendix A1 – Meteorological data....................................................................65

Appendix A2 – Background to SEM and image analysis.........................................71

References .....................................................................................................73

10 CIRIA C530
 LIST OF FIGURES

1.1 Flow chart showing main activities of test programme .............................. 15

2.1 Test panel in formwork ....................................................................... 18
2.2 Sheltered exposure of panels ................................................................ 19
2.3 Unsheltered exposure of panels ............................................................ 19
3.1 Location of tests on a panel surface ....................................................... 21
5.1 Determination of cement paste porosity and anhydrous cement content ........ 27
5.2 Successive averages of porosity vs. number of fields ................................ 29
5.3 Porosity and anhydrous cement content of cement paste vs. depth
for seven-day old Concrete A specimens ................................................ 32
5.4 Porosity and anhydrous cement content of cement paste vs. depth
for seven-day old Concrete B specimens ................................................ 33
5.5 Porosity and anhydrous cement content of cement paste vs. depth
for seven-day old Concrete C specimens ................................................ 34
5.6 Porosity and anhydrous cement content of cement paste vs. depth
for seven-day old Concrete D specimens ................................................ 35
5.7 Porosity and anhydrous cement content of cement paste vs. depth
for nine-month old Concrete A specimens .............................................. 36
5.8 Porosity and anhydrous cement content of cement paste vs. depth
for nine-month old Concrete B specimens .............................................. 37
5.9 Porosity and anhydrous cement content of cement paste vs. depth
for nine-month old Concrete C specimens .............................................. 38
5.10 Porosity and anhydrous cement content of cement paste vs. depth
for nine-month old Concrete D specimens .............................................. 39
5.11 Porosity of the cement paste at 0.5 mm depth as a function of
concrete type, curing regime and exposure history ................................... 40
5.12 Porosity of the cement paste at 3.5 mm depth as a function of
concrete type, curing regime and exposure history ................................... 40
5.13 Porosity of the cement paste at 6.5 mm depth as a function of
concrete type, curing regime and exposure history ................................... 41
5.14 Porosity of the cement paste at 20.5 mm depth as a function of
concrete type, curing regime and exposure history ................................... 41
5.15 Anhydrous cement content of the cement paste at 0.5 mm depth
as a function of concrete type, curing regime and exposure history.............. 42
5.16 Anhydrous cement content of the cement paste at 3.5 mm depth
as a function of concrete type, curing regime and exposure history.............. 42
5.17 Anhydrous cement content of the cement paste at 6.5 mm depth
as a function of concrete type, curing regime and exposure history.............. 43
5.18 Anhydrous cement content of the cement paste at 20.5 mm depth
as a function of concrete type, curing regime and exposure history.............. 43
5.19 Backscattered electron images of Concrete A1 after nine months of
sheltered exposure) in the (a) uncarbonated, (b) carbonating and
(c) carbonated zones .......................................................................... 48
5.20 Porosity vs. anhydrous cement content of the cement paste at seven days,
as a function of depth ......................................................................... 50
5.21 Porosity vs. anhydrous cement content of the cement paste at nine months,
as a function of depth ......................................................................... 51

CIRIA C530 11
 6.1 Depth of carbonation vs. depth of abrasion as a function of concret e type ..... 53
6.2 Depth of abrasion at nine months vs. porosity of the cement paste
at nine months (at four depth increm ents) as a function of concrete type....... 54
6.3 Depth of abrasion at nine months vs. porosity of the cement paste
at seven days (at four depth increments) as a function of concrete type......... 54
6.4 Depth of abrasion at nine months vs. anhydrous cement content
of the cem ent paste at nine months (at four depth increments) as a
function of concret e type..................................................................... 55
6.5 Depth of carbonation at nine months vs. porosity of the cement paste
at nine months (at four depth increm ents) as a function of concrete type....... 56
6.6 Depth of carbonation at nine months vs. porosity of the cement paste
at seven days (at four depth increments) as a function of concrete type......... 56
6.7 Depth of carbonation at nine months vs. anhydrous cement content
of the cem ent paste at nine months (at four depth increments) as a
function of concret e type..................................................................... 57
A1.1 Temperature of sheltered exposure environment ...................................... 66
A1.2 Relative humidity of sheltered exposure environment ............................... 67
A1.3 Temperature of unsheltered exposure environment ................................... 68
A1.4 Rainfall for unsheltered exposure environment ........................................ 69
A1.5 Relative humidity of unsheltered exposure environment ............................ 70

LIST OF TABLES

1.1 Combinations of variables to be examined.............................................. 14

2.1 Concrete mix and panel details ............................................................. 17
3.1 Abrasion and carbonation depth test results............................................. 22
5.1 Exceptions to expected trends .............................................................. 47
A1.1 Early temperature and relative humidity of sheltered exposure environment .. 65

ABBREVIATIONS

BSE Backscattered electron

CAZ Curing affected zone
CSH Calcium silicate hydrate
GGBS Ground granulated blast furnace slag
OPC Ordinary Portland cement
PFA Pulverised fuel ash
SEM Scanning electron microscope/microscopy
s Sheltered
u Unsheltered

12 CIRIA C530
1 Introduction

1.1 BACKGROUND

Phase I of the project “The influence of practical on-site curing with special reference to
durability and abrasion resistance” reported on a desk review of published information
on the subject (CIRIA PR49, Hammersley et al, 1997). The majority of the research had
investigated the link between curing conditions and short-term, durability-related
perform ance for several concretes and exposure condition combinations. Whilst often
providing good data for the concrete and exposure tested, such data can be di fficult to
extrapolate to other situations.

In addition to arriving at conclusions on the effects of practical curing (comparable to

typical on-site curing) on concrete, in so far as was possible, it was intended that Phase I
should identify potentially useful test methods for assessing the effectiveness and extent
of on-site curing. Such methods would then be developed further during an essentially
laboratory based Phase II programme.

However, in Phase I it was concluded that it was extremely unlikely that any currently
available conventional test method could form the basis of assessing on-site curing over
the range of environmental exposures envisaged. The Project steering group for this
phase concluded that the assessment of curing effectiveness and its influence on
durability performance could be most benefici ally determined by considering the micro-
structure in the curing affected zone (CAZ). This is the zone of concrete near to and
including its outer surface, which can be directly affect ed by active curing measures
(CIRIA PR49, ibid).

1.2 OBJ ECTIVES

The objectives of Phase II of the project (covered by this report) were to establish the
viability of a method for investigating:
· the link between practical curing methods and the microstructure of concrete
in the CAZ
· the relationship between microstructure in the CAZ and durability perform ance.

This was to take into account practical construction technology and real-li fe timescales
and was intended to help to develop a fram ework for assessing the relationship between
practical curing methods and exposure perform ance.

1.3 APPROACH

The approach adopted to characterise concrete microstructure was to obtain high

magnifi cation images of the concrete using scanning electron microscopy (SEM) and
then to use image analysis to covert these into microstructural paramet ers (essentially
porosity and anhydrous cement). Preliminary work was undertaken to develop the SEM
method and to demonstrate that differences in microstructural features could be
identified and quanti fied (see Section 5.1.2).

CIRIA C530 13
 The method was then applied in a large test programme; the main steps are summarised
in the flow chart shown in Figure 1.1. Concrete test panels of four different concretes
were cured in different ways and were then subjected to natural exposure (outdoors
unsheltered or sheltered). Aft er nine months the panels were tested for abrasion
resistance and carbonation depth. Abrasion resistance and carbonation depth were
selected as durability measures – not because they are necess arily the mechanisms of
greatest concern, but because they are known to be dependent on curing. Porosity and
anhydrous cement content profiles into the concrete were measured, both shortly after
curing (at an age of seven days) and aft er the nine-month exposure period. Concretes,
curing regimes and the corresponding tests are summarised in Table 1.1.

Table 1.1 Co mbinations of variables to be examined

Concrete G rade Curing* Exposure' SEM at Abrasion Carbonation

7 days testing and SEM at
9 months
1 u 3 3
A
2 u 3
3 u 3
OP C 3
35 4 u 3
3 3
1 s 3
3 3
2 s 3
3 3
3 s 3
3 3
3
4 s
1 u 3
B
20 4 u 3
1 s 3 3 3
OP C 3
4 s 3 3
1 u 3
C
35 4 u 3
1 s 3 3 3
25% P FA 3
4 s 3 3
1 u 3
D
35 4 u 3
1 s 3 3 3
65% 3
4 s 3 3
GGBS
* Curing regime: 1 = no additional curing
2 = immediate application of a curing membrane
3 = 1 day moist curing (wet hessian sealed in poly thene sheet)
4 = 7 day s moist curing (wet hessian sealed in poly thene sheet)
' Expos ure: u = unsheltered
s = sheltered

It can be seen that 20 combinations of variables are involved, allowing the effects of
three cem ent types, two concrete grades, four curing regimes and two exposure environ-
ments to be assessed. Half of the panels (the outdoors unsheltered panels) have been
returned to the exposure site for long-term exposure and may be used in future res earch.

The project was undertaken by Imperial College, who carried out all of the SEM work,
with Taywood Engineering Ltd. (TEL) as subcontractors responsible for mix develop-
ment, panel manufacture, curing, exposure and abrasion and carbonation depth testing.

14 CIRIA C530
 Age
Panels
cast …………………………….0

Curing
regime

Panels sampled ……...…………………..……………………. 7 days

for SEM

Panels moved to ……………… 14 days

exposure locations

SEM specimens
prepared

SEM specimens
stored for 6 months

SEM
Panels sampled .………….. 9 months
for SEM

Abrasion SEM specimens

testing prepared

Carbonation
testing SEM

Figure 1.1 Flow chart showing main activities of test programme

CIRIA C530 15