Sustainability in Structural Concrete Design

Structural Engineering Documents

20

Sustainability in Structural Concrete Design

Jorge de Brito

Rui Vasco Silva

(Editors)

International Association for Bridge and Structural Engineering (IABSE)

Copyright © 2024 by

International Association for Bridge and Structural Engineering

All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher.

ISBN:

978-3-85748-201-4 (print)

978-3-85748-202-1 (ePub)

978-3-85748-203-8 (eBook)

DOI: https://doi.org/10.2749/sed020

Publisher

IABSE

Jungholzstrasse 28

8050 Zürich

Switzerland

Phone: Int. +41-43-443 9765

E-mail: [email protected]

Web: www.iabse.org

This book was produced in cooperation with Structurae, Dresdener Str. 110, Berlin, Germany (https://structurae.net).

Copyediting: Hilka Rogers

Layout & typesetting: Florian Hawemann

About the Editors

Jorge de Brito

Full Professor of Civil Engineering at Instituto Superior Técnico, University of Lisbon, Portugal. He has authored over 700 research articles in peer-reviewed journals and 12 books, including Sustainable Construction Materials: Recycled Aggregates. He is a member of 15 editorial boards of international journals and a member of international associations like IABSE, CIB, FIB, RILEM and IABMAS.

Rui Vasco Silva

PhD researcher in Civil Engineering at CERIS, IST-ID, Lisbon, Portugal. He has participated in several research projects and is now the Principal Investigator of a project on the carbon capture of recycled alkali-activated concrete. He authored over 50 publications in peer-reviewed journals, 15 book chapters, and 2 international books all concerning sustainable construction and eco-efficient concrete.

Preface

This Structural Engineering Document has the purpose of providing a clear insight into the most recent developments in academia and industry concerning the practices leading to the sustainability of structural concrete. It presents a wide range of information from low-carbon and recycled aggregate-containing structures to up-to-date developments on alkali-activated concrete, all intending to enlighten the reader on how to improve the sustainability indicators of these systems by using technologies with a high enough readiness level and thus more likely to be used in practice.

The three opening chapters offer a broad perception of what can be achieved by implementing a sustainable design for structural concrete from different perspectives. In the first chapter, Vanderley, Quattrone, Abrão, and Pillegi provide notions on a multiscale approach for low-carbon structures. This is followed by the chapter of Silvestre, which offers considerable insight into the life cycle assessment of sustainable concrete containing different waste materials. Kurda, de Brito, Ahmed, and Hafez deliver information on the use of various types of waste materials in concrete.

The large scale of waste coming from construction and demolition activities is absolutely undeniable. It corresponds to the largest percentage of waste coming from all economic sectors and, within the scope of circularity promotion and decreased carbon emissions-related goals of the United Nations 2030 Sustainable Development Goals, it should be dealt with accordingly. The tremendous worldwide research effort put forward towards the reincorporation of construction and demolition waste has demonstrated the high technical and economic feasibility of using recycled aggregates to produce structural concrete. For these reasons, the current Structural Engineering Document focuses extensively on this topic in chapters four to eleven. González-Fonteboa, Seara-Paz, and Martínez-Abella provide a crucial review of the short- and long-term performance of structural recycled aggregate concrete. Tošić, Kurama, and Torrenti then present interesting notions on the effect of recycled aggregates on the long-term deformation of concrete. Furthermore, Etxeberria goes into detail on several durability-related aspects affected by the use of recycled aggregates from construction and demolition waste. Sainz-Aja and Thomas deal with the understudied, albeit fundamental, fatigue phenomenon occurring on recycled aggregate concrete subjected to cyclic loading. After that, Silva and de Brito offer empirical practical rules on how to design recycled aggregate concrete structures, whereas Pacheco and de Brito tackle the matter of design based on reliability analysis. Poon, Shen, Jiang, and Xuan discuss how total recycling of concrete can be achieved by resorting to accelerated carbonation. Xiao, Cheng, Pan, and Fang describe an interesting new case study on large-scale structural applications of recycled aggregate concrete at the Shandong Shanghe Industrial Base.

Finally, given the considerable advancement of other low-impact technologies, the three last chapters focus on the latest findings on: the use of alternative aggregates for concrete production, namely those from electric arc furnace slag; alkali-activated concrete, which normally uses reactive industrial byproducts; and the potentially low impacts associated with concrete made with fibre-reinforced elements. Faleschini, Zanini, and Ortega-Lopez highlight the use of electric arc furnace slag in concrete production. Puertas and Mejía de Gutiérrez then provide recent in-depth insights on the development of alkali-activated materials. Firmo, Rosa, Correia, D’Antino, and Matos subsequently show how the corrosion of reinforcement can become an outdated aspect by using glass-fibre-reinforced elements in structural concrete.

The total amount of time and effort by all parties in putting this document together is tremendous. The Editors would like to thank all authors for their contributions and for allowing this project to come to fruition. Finally, the Editors also deeply acknowledge the remarkable work of the external reviewers, Professor Mark Richardson and Professor Yury Villagrán-Zaccardi, as well as the IABSE’s Chief Reviewer from the Editorial Board, Ann Schumacher.

Jorge de Brito and Rui Vasco Silva

September 2023

Table of Contents

Chapter 1 – Notes for a Multiscale Approach for Low Carbon Concrete Structures

1.1 Introduction

1.1.1 The Implication of Net Zero CO2 Target of the Cement Industry to Concrete Producers

1.1.2 Carbon Sources in Concrete and CO2 Footprint Measuring.

1.1.3 Objective

1.2 Fundamentals of Designing Low-Carbon Concretes

1.2.1 Low Carbon Concrete and Clinker Replacement

1.2.2 Defining Low Carbon Concrete

1.3 Engineering the Cement Paste for a Low Carbon Concrete

1.3.1 Procurement and Characterization of Raw Materials

1.3.2 Dispersant Selection and Dosage

1.4 Engineering Particle Size Distribution

1.4.1 Modelling Viscosity

1.5 Measuring Mixing Water Demand for Desirable Rheological Behaviour

1.6 Defining the Optimum Particle Size Distribution

1.7 Modelling Compressive Strength of Paste Mixtures

1.8 Selecting a Paste with Minimum Carbon Footprint for Each Application

1.9 Developing a Low Carbon Concrete Formulation

1.10 Results From Practical Applications

1.11 Final Remarks

1.12 Acknowledgements

1.13 References

Chapter 2 – Life Cycle Assessment in Support of the Sustainability of Current and Alternative Types of Structural Concrete

2.1. Introduction

2.2. Environmental Life Cycle Assessment of Concrete

2.2.1. Life Cycle Inventory (LCI)

LCA Databases for Raw Materials and Construction Products

2.2.2. Life Cycle Impact Assessment and Interpretation

2.2.3. Life Cycle Assessment of Traditional and Alternative Raw Materials for Concrete

Cement

Aggregates

Transportation of Raw Materials

2.2.4. Life Cycle Assessment of the Use Stage

2.2.5. Life Cycle Assessment of the End-of-Life Stage

2.3. Economic and Environmental Life Cycle Assessment of Concrete

2.4. Life Cycle Assessment-Driven Structural Design of Concrete

2.5. Conclusion

2.6 References

Chapter 3 – Optimizing the Performance of Green Concrete Using Nonconventional Recycled Materials

3.1. Introduction

3.2. Nonconventional Aggregates and Standards

3.3. Agricultural Wastes and Aquaculture Farming as Aggregates

3.4. Industrial Wastes Used as Aggregates

3.5. Municipal Wastes as Aggregates

3.5.1. Municipal Solid Waste Incinerator Bottom Ash

3.5.2. Sewage Sludge Ash

3.6. Insulating Aggregates

3.6.1. Tire-Rubber Aggregates

3.6.2. Plastic Waste Aggregate

3.6.3. Glass Waste Aggregates

3.7. Other Types of Aggregates

3.8. Conclusion

3.9 Acknowledgements

3.10 References

Chapter 4 – Short- and Long-Term Performance of Structural Recycled Aggregate Concrete

4.1. Introduction and Objectives

4.2. Short-Term Behaviour

4.2.1. Flexural Performance

4.2.2. Shear Performance

4.3. Long-Term Performance

4.3.1. Performance Under Sustained Load

4.3.2. Performance After Sustained Loading: Recovery and Failure Behaviour

4.4. Conclusions

4.5 Acknowledgements

4.6 References

Chapter 5 – Long-Term Deformation of Structural Recycled Aggregate Concrete

5.1. Introduction

5.2. Shrinkage of Recycled Aggregate Concrete

5.2.1. Experimental Results on the Shrinkage of Recycled Aggregate Concrete

5.2.2. Modelling of Recycled Aggregate Concrete Shrinkage

5.3. Creep of Recycled Aggregate Concrete

5.3.1. Experimental Results on RAC Creep

5.3.2. Modelling of the Creep of Recycled Aggregate Concrete

5.4. Long-Term Deflections of Recycled Aggregate Concrete Structures

5.4.1. Experimental Results on Deflections of Recycled Aggregate Concrete Structures

5.4.2. Modelling of Deflection Behaviour of Recycled Aggregate Concrete Structures

5.5. Conclusions

5.6 Acknowledgements

5.7 References

Chapter 6 – Durability of Recycled Aggregate Concrete

6.1. Introduction

6.2. Carbonation Resistance

6.3. Chloride Ion Penetration

6.4. Permeability

6.5. Alkali-Silica Reaction

6.6. Freeze-Thaw Resistance

6.7. Sulphate Attack

6.8. Conclusions

6.9 References

Chapter 7 – Fatigue of Recycled Aggregate Concrete

7.1. Introduction

7.2. Fatigue Test Methodologies

7.2.1. Fatigue Limit and Fatigue Life

7.2.2. Strength Range

7.2.3. Wöhler Curve or S-N Curve

7.2.4. Wave Type and Frequency

7.2.5. Test Specimens

7.3. Fatigue Behaviour of Recycled Aggregate Concrete Under Compression and Bending Cyclic Loadings

7.3.1. Strain Versus Cycles

7.3.2. Fatigue Limit Versus Compressive Strength

7.4. Fatigue Failure Micro-Mechanisms by Micro-CT Analysis

7.5. Resonance Fatigue Behaviour of Recycled Concrete

7.6. Effect of Temperature on Recycled Concrete

7.7. Fatigue for Railway Superstructure Applications

7.8. Conclusions

7.9 References

Chapter 8 – Structural Design of Recycled Aggregate Concrete

8.1. Introduction

8.2. Methodology

8.2.1. Certified Recycled Aggregates

8.2.2. Structural Design Methodology

8.2.3. Building Design

8.3. Results and Discussion

8.4. Concluding Remarks

8.5 References

Chapter 9 – Design Guidelines for Recycled Aggregate Concrete Based on Reliability Analysis

9.1. Introduction: Partial Factor Formats and Code Calibration

9.2. Recycled Aggregate Concrete

9.2.1. Coarse Recycled Aggregates Produced From Concrete Waste

9.2.2. Properties of Concrete Made With Coarse Recycled Concrete Aggregates

9.2.3. Uncertainty in Resistance Models and Recycled Aggregate Concrete

9.3. Methodology for Calibration of Design Equations

9.3.1. Code Calibration

9.3.2. Design Equations for Ultimate Limit State Recycled Aggregate Concrete Design

9.3.3. Deemed-to-Satisfy Provisions for Concrete Cover Design of Recycled Aggregate Concrete

9.4. Conclusion

9.5 Acknowledgements

9.6 References

Chapter 10 – Total Recycling of Concrete Waste Using Accelerated Carbonation

10.1. Introduction

10.2. Carbonation and Application of Coarse Recycled Concrete Aggregate

10.2.1. Physical and Microstructural Properties of Carbonated Coarse Recycled Concrete Aggregate

10.2.2. Microstructural Properties of Carbonated Coarse Recycled Concrete Aggregate

10.3. Application of Carbonated Coarse Recycled Concrete Aggregate in Recycled Aggregate Concrete

10.3.1. Compressive Strength of Recycled Aggregate Concrete

10.3.2. Microhardness of Recycled Aggregate Concrete

10.3.3. Elastic Modulus of Recycled Aggregate Concrete

10.3.4. Bulk Electrical Conductivity of Recycled Aggregate Concrete

10.3.5. Water Absorption of R Recycled Aggregate Concrete AC

10.3.6. Chloride Penetration Resistance of Recycled Aggregate Concrete

10.3.7. Gas Permeability of Recycled Aggregate Concrete

10.3.8. Carbonation Resistance of Recycled Aggregate Concrete

10.3.9. Relation Between Water Absorption and Permeability of RAC

10.4. Carbonation and Recycling of Fine Recycled Concrete Aggregate

10.4.1. Physical and Microstructural Compositions of Carbonated Fine Recycled Concrete Aggregate

10.4.2. Mineralogical Compositions and Chemical Reactivity of the Carbonated Fine Recycled Concrete Aggregate

10.4.3. Influence of Carbonated Fine Recycled Concrete Aggregate on the Properties of Recycled Aggregate Mortar

10.4.4. Bond Strength Between Recycled Aggregates and New Cement Mortar

10.5. Carbonation and Recycling of Recycled Concrete Fine Powder (RCFs, <150 um)

10.5.1. Phase Assemblance and Microstructure Evolutions of Carbonated Recycled Concrete Fines

Calcium Phase Evolution in Carbonation Products

Silicon Phase Evolution in Carbonated RCFs

Aluminium Phase Evolution in Carbonated RCFs

Microstructure Development

Potential Application of Carbonated RCFs as High-Reactive SCMs

10.5.2. Synthesis of Amorphous Nano-Silica from Recycled Concrete Fines

Design of Two-Step Carbonation Process for Synthesizing Amorphous Nano-Silica

Compositions of Carbonation Products

Microstructure of Carbonation Products

CO2 Consumption and Yields of Carbonation Process

10.6. Summary

10.7 Acknowledgements

10.8 References

Chapter 11 – Large-Scale Structural Applications of Recycled Aggregates Concrete – Shandong Shanghe Industrial Base

11.1. Introduction

11.2. Design Summary

11.3. Application of Recycled Concrete

11.3.1. Mix Proportion Design

11.3.2. Construction Process

11.4. Application Monitoring of Recycled Concrete

11.4.1. Dynamic Monitoring

11.4.2. Settlement Monitoring and Creep Monitoring

11.4.3. Deflection Monitoring for Beams

11.4.4. Shrinkage Monitoring of Recycled Concrete

11.4.5. Crack and Deflection Monitoring for Slab

11.5. Summary and Prospect

11.6 Acknowledgements

11.7 References

Chapter 12 – Electric Arc Furnace Slag Aggregates in Concrete

12.1. Introduction

12.2. Steelmaking Industry and Slags Formation

12.3. Electric Arc Furnace Slag Concrete

12.3.1. Characteristics of Electric Arc Furnace Slag: Physical Properties, Chemical Composition, and Mineralogy

12.3.2. Leaching Potential of Electric Arc Furnace Slag

12.3.3. Electric Arc Furnace Slag Concrete: Workability, Mechanical Properties

12.3.4. Electric Arc Furnace Slag Concrete: Probabilistic Models

12.3.5. Electric Arc Furnace Slag Concrete: Durability Properties

12.3.6. Electric Arc Furnace Slag Concrete: Environmental Impacts

12.3.7. Electric Arc Furnace Slag Concrete: Special Applications

12.4. Structural Applications

12.5. Conclusions

12.6 Acknowledgements

12.7 References

Chapter 13 – Alkali-Activated Concrete Performance

13.1. Introduction

13.2. What are Alkali-Activated Cements and Concretes (ACC)?

13.2.1. Definition and Brief History

13.2.2. Precursors and Alkaline Activators

13.2.3. Alkaline Cement and Concrete Preparation: Nature of the Main Activation Products

13.3. Fresh and Hardened Alkali-Activated Concrete (AAC) Performance

13.4. Alkali-Activated Concrete Durability

13.4.1. Sulphate Resistance

13.4.2. Resistance to Acid Media

13.4.3. Carbonation Resistance and Reinforcement Corrosion

13.4.4. Chloride Resistance and Reinforcement Corrosion

13.4.5. Heat and Fire Resistance

13.5. Final remarks

13.6 References

Chapter 14 – Fibre-Reinforced Polymer Reinforcement Towards More Durable and Sustainable Concrete Structures

14.1. Introductory Remarks

14.2. Fibre-Reinforced Polymer Bars for Reinforced Concrete Structures

14.2.1. Constituent Materials

14.2.2. Geometries and Manufacturing

14.2.3. Mechanical and Physical Properties

14.2.4. Bond to Concrete

14.2.5. Applications

14.3. Structural Behaviour and Design Principles of Reinforced Concrete Members Internally Reinforced with Fibre-Reinforced Polymer Bars

14.3.1. Bending

14.3.2. Shear

14.3.3. Available Design Guidelines/Codes

14.4. Critical Topics for FRP-RC Structures

14.4.1. Fire Behaviour

14.4.2. Seismic Behaviour

14.4.3. Durability

14.5. Sustainability of FRP Reinforcement

14.5.1. Overview

14.5.2. LCA Studies

14.5.3. Future Trends

14.6. Concluding Remarks

14.7 References

Chapter 1

Notes for a Multiscale Approach for Low Carbon Concrete Structures

Vanderley M. John, Marco Quattrone, Pedro C.R.A. Abrão, Markus S. Rebmann, Rafael G. Pilleggi

CEMtec National Institute for Advanced Eco-efficient Cementitious Technologies, Sustainable Construction Innovation Centre CICS, Polytechnic School, University of São Paulo

1.1 Introduction

The only construction with zero environmental impact is the one that is never made. Unfortunately, we do not build to prevent environmental impacts, but to serve humankind. The hypothetical zero-impact construction cannot improve the comfort, health, longevity, meaningfulness, or enjoyment of human life. While it is undeniable that certain buildings and facilities may have a marginal or even negative impact on the quality of human life, it is crucial to urgently expand the built environment in developing countries to enable impoverished citizens to lead fulfilling lives.

Therefore, the decision of whether to build or not becomes the most consequential one when it comes to the sustainability of the built environment. Having decided that a new construction (or a retrofit) is a human need, the challenge is how to minimize negative environmental impacts without compromising the ability of the built environment to serve humanity.

The increased frequency of extreme weather events is the already visible effect of global warming, with environmental stresses over the built environment frequently above the ones it was designed for. News often report that prolonged droughts make the natural environment prone to fire or made water reservoirs that supply drinking water or hydroelectricity too small to cope with the current uncertain environment. Extreme heat waves introduce thermal stresses above design. Rains heavier than expected cause urban floods and landslides. Stronger winds present an increasing challenge to the structural stability of roofs and façades. Carbon emissions are progressively becoming a cost, not only due to increasing taxation [1, 2] but also because products and activities with high CO2 footprints are perceived as risky, driving investment elsewhere as society moves towards a low-carbon economy. The buildup of the resilience of the built environment to cope with the inevitable effects of climate change will require more construction, which at the same time must not contribute to the carbon footprint.

Mitigating construction CO2 emissions requires deploying solutions that are affordable and technically viable with dramatically lower greenhouse gas (GHG) emissions. This cannot be achieved if we keep acting by promoting innovation at the material scale only. The challenge requires a coordinated multiscale action: material scale, construction scale and planning scale.

1.1.1 The Implication of Net Zero CO2 Target of the Cement Industry to Concrete Producers

The Global Cement and Concrete Association (GCCA) recently disclosed that the cement industry has set a goal (Fig. 1.1) of achieving Net Zero CO2 emissions by 2050 [3]. The average clinker content is expected to decrease to 0.52 (kg/kg) by 2050, compared to today’s average of 0.63 among GCCA members, resulting in a modest 9 % reduction in emissions, which we believe is a pessimistic view, vis-à-vis the current developments of low carbon solutions. Because it implies that clinker will make 52 % of the cement composition in the best-case scenario using renewable energy sources. It would still be an average of ~500 kg CO2/t of clinker (or 265 kg CO2/t of cement) from limestone decomposition, which needs to be captured and used or stored to ensure zero net emissions [4]. Carbon capture and storage or utilization (CCS/CCU) is expected to provide 36 % of the total CO2 mitigation but it will come with a steep cost between ~US$50–100/t of CO2. Considering the current clinker cost of approximately US$30/t and carbon capture cost of US$100/t of CO2, the taxation or carbon capture expense will add US$88/t to clinker, resulting in a significant 200 % cost increase. A similar situation can be observed for the second most important structural material: low-carbon steel will be more costly. Unfortunately, there is no other material that is low carbon, affordable, and scalable rapidly in order to replace steel and concrete completely.

These cost increases of essential materials will have social consequences for the humans living in the developing global south, which needs to build to provide a decent and safe built environment and lacks the extra money to spare. Therefore, if the mitigation action stays confined to the materials industry scale, as it has been so far, housing and infrastructure will become more expensive.

This cost inflation, however, will motivate engineers and real estate investors to look for savings. Delivering the much-needed built environment with fewer materials, dematerialization is an obvious but not easy option. Indeed, the cement industry expects that clients will act causing a 30 % reduction of cement demand in comparison with the reference scenario, incidentally, further reducing CO2 emissions. Concrete and mortar producers facing increase of cement cost are expected to reduce cement content in their formulation, equivalent to 11 % of the total CO2 mitigation. However, these reductions will not be enough to totally offset cement cost increase. Therefore, real estate developers and designers are expected to reduce by 21 % the concrete needed to fulfil the clients’ demand for the built environment.

Thus, affordable low-carbon cement-based construction will be a result of a multiscale, multistakeholder approach. At material scale, innovation will result in cement with progressively less clinker. At the scale of the concrete production, concrete with less and less cement. And at design and construction scale, engineers will seek for technical solutions and designs that dematerialize construction, reducing the amount of concrete (and other cement-based materials) required and, in consequence, reducing construction costs in order to offset the cost impact of low-carbon processes.

Fig. 1.1 GCCA CO2 emission targets and the contribution of each strategy to the total mitigation [3].

1.1.2 Carbon Sources in Concrete and CO2 Footprint Measuring

The GHG emissions from cement-based materials are mostly CO2, primarily due to fossil fuel oxidation in cement kilns, which can be almost neutralized by the use of renewable fuels, and limestone decomposition in the clinker process, which is not possible to avoid. The clinker footprint varies between ~860–900 kg CO2/t depending on the energy matrix, energy efficiency, and, to some extent, the CO2 intensity of the raw materials. The CO2 footprint of cement (kgCO2/t of cement) is reduced by partially replacing clinker with supplementary cementitious materials (SCMs) that have lower CO2 footprints (Table 1.1). Consequently, blending clinker with SCMs leads to an average direct CO2 emission of 635 kg CO2/t among the company’s members of the GCCA – Getting the Numbers Right (GNR).

Aggregates usually possess a low carbon footprint. Depending on the energy matrix and specific details of the process, a crushed natural aggregate typically has a CO2 footprint of 1–7 kg CO2/t. Even when considering that the mass of aggregates is about 5–7 times higher than cement, under typical conditions, the difference in the contribution of cement and aggregates to the concrete’s CO2 footprint is around one order of magnitude.

Table 1.1 Typical emission factors for main cementitious materials (kg CO2/t). The actual values vary widely in function of energy matrix, materials details (e.g., humidity of clay), plant detail, and operation standards. Emissions from secondary materials, such as fly ash and blast furnace slag can be higher if part of the CO2 emissions from the process is allocated to them, which depends on rules. Data from [5]

However, the CO2 footprint of aggregates can increase when sophisticated processing is required. For instance, the artificial drying of aggregates can significantly impact energy demand and the CO2 footprint, although it typically remains below 10 kg CO2/t [6]. Marine dredged sand and gravel tend to have higher CO2 footprints, estimated between 30–40 kg CO2/t in the UK [7].

Transportation distance and mode also play a role. The Intergovernmental Panel on Climate Change (IPCC) estimates that the CO2 footprint for road transportation varies between 200 g CO2/t-km for the largest trucks to 500 g CO2/t-km for medium trucks [8]. However, train and ship transportation result in substantially lower CO2 footprints. The CO2 footprint of aggregates heavily depends on transportation distances, but aggregates are widely available and affordable, leading to relatively short transportation distances. On the other hand, cement can be transported over longer distances, yet it constitutes less than 15 % of the concrete’s total mass.

Regarding the processing, transportation, and placement of concrete, they require relatively little energy compared to the raw materials. However, the thermal curing of concrete requires thermal energy, and depending on the amount and heat source, it can significantly increase the CO2 footprint of concrete. Thermal curing is often inefficient because it requires heating the aggregates, which constitute approximately 80 % of the mass. Therefore, in most of the cementitious materials, the cement used in the formulation is the main source of CO2 emissions.

1.1.3 Objective

This chapter focuses on the facts that arise after the decision to build with concrete has been made. Its main objective is to reduce the carbon footprint of concrete by optimizing the different components (paste, mortar, and coarse aggregates) that constitute the material, with particular emphasis on the paste. For this reason, this chapter refers to the multiscale approach. Discussing how to minimize carbon emissions by improving design and construction practices are certainly essential, but outside of the scope of this chapter.

1.2 Fundamentals of Designing Low-Carbon Concretes

1.2.1 Low Carbon Concrete and Clinker Replacement

The literature on sustainable construction usually associates low carbon concrete when a large fraction of the cement (or clinker) is replaced by secondary (residues) mineral additions. It is certain that replacing clinker by low-carbon materials will reduce the CO2 footprint of the cement (kgCO2/t of cement). However, the CO2 footprint of the concrete depends on the footprint of cement but also on the amount of cement in the mixture. The amount of cement required to achieve the required strength over time and other relevant performances of a particular concrete depends on several factors, including the fineness, shape, specific surface area and other properties of aggregates used in the mix.

SCMs have generally lower reactivity than the clinker, and in some cases their reactivity can be close to zero, especially for fillers. When blending SCMs with clinker, there is a reduction in the water chemically bound by the cement (g/g of cement) at all ages of interest. Depending on the market, sometimes a cement with little or almost no SCMs can have lower combined water. This can result from a cement particle size distribution (PSD) that includes particles coarser than 30 µm, which do not fully react at 28 days [17], eventually combined with low-quality clinker. In commercial cements, this lower reactivity leads to a reduction in the cement strength class [9], which is a proxy of the chemically combined water at 28 days when measured at a constant water/cement ratio. The lower the cement strength class, the higher the amount of cement needed to achieve a given concrete design compressive strength (Fig. 1.2). It’s worth noticing that most Environmental Product Declarations (EPD) currently refer only to cement types, and do not mention the cement strength class.

The amount of cement needed also depends on the mixing water demand, which is also influenced by cement characteristics such as, specific surface area or particle size distribution. These characteristics vary significatively, particularly for blended cements [10] influencing the actual amount of cement needed to deliver a concrete with a required strength [9].

Therefore SCMs can have a significant effect on the water demand for workability and the combined water content at 28 days [11], influencing the cement consumption for a given concrete. Some SCMs may decrease the water demand: strength may be achieved with less cement. Others may increase it, and more cement will be required.

As a result, the CO2 footprint of concrete is not solely determined by the clinker factor or the low CO2 footprint of the cement for the amount of a given cement required to formulate a concrete is a decisive factor in the equation.

Fig. 1.2 (Left) theoretical influence on cement strength class on the relative cement consumption (kg/m³) for various design strength, all other factors remaining the same. (Right) actual frequency distribution of cement intensity and carbon intensity from literature. Figures from [9] when measured with constant w/c ratio, cement strength class is a proxy for combined water at 28 days.

1.2.2 Defining Low Carbon Concrete

Low carbon concrete is a concrete that has lower CO2 footprint compared to the current industrial standards in a given region at the present time. Therefore, a valid representative baseline of the CO2 footprint of concrete is needed for every region. As the industry progresses towards lower CO2 emissions over time, this baseline will naturally evolve.

Usually, the functional unit for the CO2 footprint of concrete is kg/m³. However, concrete with different design strengths will have a different performance. For that reason, we recommend including the compressive strength at 28 days in the functional unit, resulting in a unit of kg⋅m-3⋅MPa-1 [12]. Some authors have suggested incorporating durability-related parameters (such as chloride diffusion and carbonation rate) into the functional unit. However, for environmental assessment, the crucial factor is the service life of the structure, which is highly dependent on local conditions. For example, the relevance of chloride diffusion would be negligible in chloride-free environments. Similarly, a high carbonation rate may speed up the CO2 capture process, aiding in global warming reduction, but it might not affect the service life of reinforced concrete exposed to an indoor and dry environment or in the absence of steel reinforcement. Thus, while these parameters are important for specific applications, they might not be helpful in establishing the benchmark needed to guide the industry towards low carbon concrete production.

In the past, we produced a benchmark of binder intensity (kg⋅m-3⋅MPa-1, at 28 days) and utilized default values for the CO2 footprint from clinker to establish a benchmark from literature [12]. This benchmark was then used to access more recent research on low-carbon concretes [13] (Fig. 1.3). The data revealed that all concretes with SCMs exhibit a minimum CO2 footprint of approximately 2 kg CO2⋅m-3⋅MPa-1 for compressive strengths above 40 MPa. As the compressive strength decreases to 20 MPa, the CO2 footprint increases exponentially, reaching around 4 kg CO2⋅m-3⋅MPa-1, including when clinker is replaced by inert fillers. For ordinary Portland cement (OPC) without admixtures, the best available laboratory technology resulted in a CO2 footprint of 4 kg CO2⋅m-3⋅MPa-1 for concretes with a compressive strength above 50 MPa. This value increases exponentially to 10 kg CO2⋅m-3⋅MPa-1 for those with a compressive strength of 20 MPa. We must keep in mind that laboratory conditions are controlled and allow for the selection of optimum raw materials, excellent processing machines and casting procedures, aspects that most of the time are not reproducible at industrial scale. In other words, laboratory results have a technology readiness level (TRL) up to 5 [14] showcasing the potential of a technical solution, but real-world application may require further considerations.

Fig. 1.3 Benchmark of CO2 footprint from laboratory results (graph) for pure Portland cement (top white area) and blended cement (yellow), where colour dots are experimental concretes were cement is replaced by a high volume of filler (from [13]). Superimposed shapes: the blue shape is EPD derived data of Australian ready-mix concretes [15–17]; the yellow square with a red line border is from a LCA inventory of a ready-mix concrete from Brazil [18].

For a technology contributing to mitigation efforts, it must reach the market, become an innovation, and achieve production scale-up, eventually displacing high CO2 emitting existing solutions. This process requires transforming the TRL from its initial stages of 3–5 in the laboratory to a TRL 9 market-ready solution. This transition demands considerable time and a substantially larger investment in Research, development & innovation, design and construction of industrial facilities. Additionally, it requires the development of an industrial-scale supply chain, overcoming regulation and standardization barriers, and initiating production, which will progressively scale up over time [14]. In our case, scaling up low-carbon concretes will probably require a long-time learning process.

To access what constitutes a low-carbon ready-mix concrete solution from a market perspective, it is becoming possible to produce a benchmark by combining information from EPDs from various countries, as we did with data from Australia [15–17] and from national inventories such as the one from Brazil [18]. From these limited data (Fig. 1.3) it is possible to estimate that the best available (commercial) technology of ready-mix concrete currently results in a CO2 footprint of around 5–6 kg CO2⋅m-3⋅MPa-1. As research progresses, it is expected that new innovative low-carbon concrete solutions with even lower CO2 footprints than the current best available technology (BAT) will enter the market. Achieving the best laboratory results involves reducing the CO2 footprint by a factor ranging between 2 and 6. This transformation requires long-term investment, a continuous learning process, and certainly a different supply chain.

In the market, there are concretes available with lower CO2 footprints, and some are even carbon-neutral (zero CO2 footprint). However, this is often achieved through the vendor acquiring carbon offset credits from the market to compensate for the emissions produced during the concrete production. These credits come from officially certified projects that have reduced GHG emissions compared to the current market practice. These projects include initiatives that destroy methane from landfill, ensure native forest preservation, generation of renewable energy, and even energy efficiency projects.

1.3 Engineering the Cement Paste for a Low Carbon Concrete

¹

The cement standards are dedicated, among other things, to control the amount of chemical hydration reaction the cement will produce over time. This reaction is typically measured by the minimum cement strength at different ages or the cement strength class at 28 days, with a constant water-to-cement ratio of ~0.5. The higher the strength, the greater the fraction of cement that has reacted with the water. A 42.5 MPa cement will have a 28-days combined water around 0,22 g/g, while a 32.5 MPa cement will have 0.18 g/g. The industry manages the strength by a combination of the amount and type of SCMs and adjusting grinding fineness.

The concrete strength is a function of capillary porosity produced by the water that has not reacted with the cement. The volume of hydrated products – represented by the combined water – needed to deliver a paste with a given strength depends on the volume of mixing water, which is determined by the desired rheological behaviour. When the mixing water is reduced, the amount of hydrated products needed is also reduced.

Therefore, engineering the paste can reduce the mixing water demand and compensate for dilution caused by fillers or other SCMs with low reactivity. The use of dispersants (or superplasticizers) alone may allow to reduce 10 % to 30 % of mixing water demand [19]. This reduction in water content subsequently leads to less combined water and less cement being required, resulting in a lower CO2 footprint for the concrete.

When the fines – cement and SCMs – are fully dispersed, it becomes possible to engineer the particle size distribution, leading to a further reduction in the mixing water demand, allowing further dilution of cement. Research data indicate that it is feasible to achieve up to 70 % reduction in mixing water by combining particle size optimization and full dispersion of fines, enabling dilution of the Portland cement by inert fillers without changing the strength, meanwhile generating a workable product [20–22].

The particle agglomeration destroys the measured particle size distribution of the fines. Therefore, particle packing optimization can only be achieved (or discussed) in formulations with fully dispersed fines. However, in the usual market, concrete formulations rarely have fully dispersed fines in the paste, due to the cost of admixtures, concerns about segregation, the need to meet minimum cement content standards, as well as cement paste volume to ensure adequate maximum paste thickness (MPT) [20], which depends on the volume and morphology of aggregates.

Our strategy, therefore, includes: (a) select a good, reactive commercial cement; (b) select fillers or other SCMs with various particle size distributions, including ultrafine particles (D90 < 5 µm); (c) select a dispersant admixture, and for all fines – cement, fillers and SCMs – determine the optimum admixture dosage to ensure full dispersion of the fines; (d) develop a mixture between cement and SCMs to minimize the mixing water demand of the paste; (e) adjust the binder content to desired mechanical properties; (f) engineer aggregates particle size distribution to minimize concrete paste volume.

1.3.1 Procurement and Characterization of Raw Materials

Selecting a reactive (high strength class) Portland cement preferably with a low surface area is recommended. To dilute this cement at least two fillers are required: (a) an ultrafine, performance filler with D90 < 5 µm; (b) a dilution filler, which has the particle size distribution and surface area as close as possible of the original cement. Varying the proportion between the dilution filler and cement allows adjusting the chemical reactivity of the paste without significantly affecting the rheology, since both materials have similar PSD and surface area. The ultrafine improves the packing and reduces the mixing water demand. Choosing a reactive high-strength cement type CEM I is advantageous if you have access to excellent fillers.

The particle size distribution can be determined by laser diffraction, while the specific surface area should preferentially be measured by Brunauer–Emmett–Teller (BET) gas adsorption. The ratio between the BET specific surface area and the surface area estimated from the laser diffraction particle size distribution, assuming all particles are perfect spheres, represents the shape factor. A shape factor approaching 1 indicates that the particles are rounded, as observed in silica fume and some fly ashes (Table 1.2). Thermogravimetric analysis (TGA) can be used to assess the potential of cement and reactive SCMs to react with water.

Fig. 1.4 Particle size distribution from cement and the dilution filler Q1 and the performance Q2. From [23]

1.3.2 Dispersant Selection and Dosage

Dispersants (or superplasticizers) are the simplest and more practical way to achieve up to 30 % of water reduction while maintaining similar workability [19], allowing a reduction of the combined water (or binder content) for a same given strength.

Superplasticizers interact with the chemical reactions of the cement [24], which will reduce dispersion progressively and even delay hydration affecting initial strength, as shown in Fig. 1.5. The addition of 1 wt% of a superplasticizer increased the induction period from around 2 hours (OPC-0.5-noSP) to 6 hours (OPC-0.37-1wt%SP), consequently extending the final setting time as well. Thus, achieving low-carbon cement relies on effective dispersion, requiring the use of an admixture compatible with the chosen cement. Compatibility may be tested by heat of hydration and workability loss over time as shown by [25, 26].

Fig. 1.5 Heat flow at 23°C of an CPV (similar to CEM I) with 0.5 w/c without superplasticizer and with 0.37 w/c and 1 wt% of a superplasticizer (polycarboxylate ether – ADVA CAST 527, GCP).

To measure the dispersant saturation content of cement compositions, different approaches can be used, the most common ones are the mini-slump test [27] and rotational rheometry [10, 23], where the dispersant is dosed in pastes formulated with low w/c (around 0.3–0.4) to minimize segregation during the test. Fig. 1.6 provides an example of the dispersant dosage of different Brazilian commercial cements, limestone fillers, and a calcined clay [10, 28]. The pastes were subjected to rotational rheometry (Rheometer MARS 60, HAAKE) using plate-plate geometry with a diameter of 35 mm and a gap of 1 mm. The shear programme used was the stepped flow test in two cycles of acceleration and deceleration, with the shear rate varying from 0 to 50 s-1. For each superplasticizer content, the yield stress was obtained at a shear rate equal to 0 s-1. In this case, the saturation dosage was set as the point prior to the stabilization of the yield stress. However, as done by [23], the saturation point could be set as the point prior to the stabilization of the apparent viscosity.

Fig. 1.6 Yield stress at 0s-1 from the upwards shear rate ramp as a function of the dispersant content for different Brazilian commercial cements, two limestone fillers and a calcined clay adapted from [10, 28].

From Fig. 1.6 it is observed that for the same water and dispersant content, the cement 16DE, 49DE (cement with diatomaceous earth) and the calcined clay