Geopolymer concrete in tunnelling

INTRODUCTION

What are geopolymer concretes and AACM?

Geopolymer concretes are considered a subset of alkali activated cementitious materials (AACM) containing Pulverised Fuel Ash (PFA) and Ground Granulated Blastfurnace Slag (GGBS).

Unlike Ordinary Portland Cement (OPC), which is a hydraulic binder that hydrates with water in standard concrete mixes, AACM requires the chemical inputs of a precursor plus an alkali activator that generally has a silicate or hydroxl basis.

Why do we need alternatives to OPC?

Production of OPC is estimated to contribute 8% of the world’s total CO2 emissions, equating to 3 billion tonnes of CO2/year (2016 figures), with 1 tonne of OPC resulting in 1 tonne of CO2. As such, it is a significant area of focus to reduce carbon on construction projects.

Wagners EFC

A geopolymer concrete called Earth Friendly Concrete® (EFC) has been developed by Australian building materials manufacturer/supplier Wagners EFC Pty Ltd over the past decade; it is effectively concrete as we know it, but without OPC.

The EFC binder content comprises approximately 75% GGBS and 25% fly ash, which are mixed with an alkali activator system that initiates the chemical reaction between the binder components. Other constituents such as aggregates, sand, and water are added in the same proportions as standard OPC-based concretes. This provides extremely low CO2 emissions and embodied energy as the binder has 80% less CO2 than an OPC binder.

So, 1m3 of the EFC is calculated to save around 200kg of CO2 compared to producing concrete with the conventional cementitious mixes. EFC has no calcium and, as this is what is attacked by sulphates (sulphate attack is generally attributed to the reaction of sulphate ions with calcium hydroxide and calcium aluminate hydrate to form gypsum and ettringite) and chlorides in conventional concretes, there are additional advantages of durability.

More than 100,000m3 of EFC has been poured in Australia with the first permanent works structural application being the construction of the Global Change Institute building at the University of Queensland, in 2012.

Figure 1 shows the CO2 emissions of a range of concrete binders, including EFC.

Alkali activated cements were developed in the 1930s and slag use increased several decades later with cement shortages. Research has increased considerably since the 1990s. Commercialisation has occurred with products no longer confined to laboratories.

EFC is currently available in the UK with strength grades up to C60/75. Initial trials and applications were conducted in early 2020 in London for temporary piling works. Further applications have been in temporary works for a number of projects, including the southern section of HS2 high-speed rail project where the Skanska/Costain/Strabag JV has installed approximately 3000m3 of C32/40 EFC in the temporary piling mat works at the Euston approaches section.

The first permanent works installation of EFC in the UK occurred on 30 November 2020, at the new Tesco petrol station at Canada Water, in London, where C32/40 EFC was installed in permanent piles by Keltbray Piling. Extensive testing of the EFC prior to the commencement of the project enabled the design engineers to be satisfied with the structural and durability performance of the material.

Compliance

Prescriptive specifications limit the potential use of AACM. Performance-based specifications are needed to allow their use through testing, which is allowed by EC2.

Extensive studies have been undertaken by approved test houses, independent experts and universities (including fire and durability testing), including: RMIT in Melbourne, Australia; Aachen University, in Germany; UNSW Australia; Aecom Australia; UQ Water Management Centre, in Australia; and, Dr James Aldred.

This testing and research has led to successful Deutsches Institut fur Bautechnik (DIBt) certification in Germany, which is aligned to EN:206 and can act as a foundation for wider European accreditation. Furthermore, this complies with the recently released Australian Standard TS199 2023 ‘Design of Geopolymer Concrete Structures’.

Properties and testing

Biogenic corrosion was tested at the University of Queensland through 12 months of exposure to a sewage environment and EFC performed significantly better than other concretes.

Chloride penetration tests showed similarly impressive results at UNSW Australia, with excellent chloride ion ingress results as shown in Table 1:

As mentioned, the mechanism for sulphate attack on concrete. This doesn’t occur with EFC (due to the lack of calcium) as is shown in Figure 2.

The different chemistry of EFC also provides a different temperature profile; EFC exhibits under half the temperature rise of comparable 65% slag mixes, which in turn can aid reduction of early age thermal cracking.

EFC production

EFC can be produced with existing plants, albeit they need to be clean from OPC; they simply need an activator dosing unit to be added. Otherwise, EFC can be pumped and placed in the same way as traditional concrete.

Use in Australia includes Toowoomba airport pavements where 50,000m3 has been poured, as well as use in wharves, multi-storey buildings, and a road bridge. In the UK, it has been used for temporary works on HS2, London Power Tunnels (LPT), and for permanent works in piling.

EFC USE IN TUNNELLING APPLICATIONS Accreditation & standards

For the DIBt approval process in Germany, full scale segments with EFC were cast, and sampling and testing undertaken. Data were submitted to DIBt, which in September 2019, wrote approved the components of the geopolymer “Wagners EFC binder” to produce concrete based on DIN EN 206-1 in conjunction with DIN 1045-2 for the following exposure classes: X0; XC1 to XC4; XF1 and XF3; XA1 to XA3; and,W0 and WF.

Wagners is currently working with BUI (Brameshuber + Uebachs Ingenieure GmbH), DIBt, and the Building Research Establishment (BRE) in the UK to obtain a European Accreditation Document that would enable the BRE to issue a similar compliance with BS EN 206-1. Accreditation is anticipated shortly. EFC complies with the requirements and recommendations of the UK PAS 8820 ‘Alkali Activated Cementitious Material’.

An Australian handbook entitled ‘Handbook for the design of geopolymer and alkali activator binder concrete structures’ is scheduled for release in 2023. It has been developed by a sub-committee of the Australian Standard for Concrete Structures.

Segments for Cross River Rail

Wagners won the concrete to supply segments to the Cross River Rail contract, in Brisbane, Australia, and used this as an opportunity to test EFC segments alongside those cast with normal OPC. The mix was a C50/60 concrete with 35kg/m3 of steel fibres and monofilament polypropylene (PP) fibres for fire resistance. See Figure 3.

Fire testing to RABT-ZTV curve at CSIRO, Australia

Concrete panels were placed in the furnace and subjected to the RABT-ZTV (Eureka) fire curve for hydrocarbon fires in rail tunnels.

The results for the steel fibre reinforced (SFR) EFC mix containing 1.5kg/m3 of PP fibres had similar magnitudes of surface loss resulting from the exposure to fire as the other Portland cement-based concretes.

After the fire tests, several cores were taken from different depths within the concrete panels. The residual compressive strengths were determined for the different depths within the panels.

The mean compressive strength of the EFC geopolymer concrete prior to fire testing was 55.0MPa. After fire testing, the lowest residual strength occurred at the hot face section and showed a decrease of 25.4%.

Beam Tests

Three beams of the EFC mix were tested in accordance with BS EN 14651:2005, ‘Test method for metallic fibre concrete – Measuring the flexural tensile strength (limit of proportionality (LOP), residual)’.

The results of the tests show that SFR EFC has similar magnitude LOP and residual flexural tensile strengths to those of a typical C50/60 OPC-based concrete containing Dramix 4D steel fibres.

Comparative results: EFC v production concrete

The comparative testing results of EFC and production concrete samples cast in May 2020 are summarised in Table 2. From the strength, beam and fire test results, it may be seen that EFC geopolymer concrete has comparable properties to those of the Portland cementbased concretes that were used to produce SFRC tunnel lining segments.

The above, combined with the improved resistance to exposure to chlorides and sulphates, give a compelling case for the use of EFC, regardless of carbon. Compared to OPC concrete, in terms of structural performance EPC is: 25MPa to 65MPa for all commercial grades; has 30% higher flexural tensile strength; low drying shrinkage – typically 350­-; similar modulus and Poisson’s ratio; and, high fire resistance. Also in comparison to OPC, in terms of durability it has: high acid resistance (sewer); high sulphate resistance; high chloride ingress resistance (marine), and low heat of reaction.

CO2 calculations for EFC

By adopting the embodied CO2 values contained in the EFC Carbon Footprint Declaration and the Mineral Products Association’s (MPA’s) published values, a carbon footprint table per tonne of binder material and binder content in various concrete mixtures has been developed by Wagners.

The table assumes that the embodied CO2 (ECO2) footprint of EFC in the UK is the same as in Australia. This is not considered unreasonable when assessing the supply chains and shipping distance of the binder and activator components. It is shown in Table 3.

A typical C50/60 concrete mix for tunnel lining segments in the UK would have a 420kg/m3 cementitious content containing 30% GGBS – i.e. a CEM IIB-S. From the table, it can be seen that the ECO2 of the binder content of 1m3 of this type of concrete would be 263kg/m3.

EFC with a similar binder content of 420kg/m3 would have a ECO2 value of 49kg/m3. This represents a reduction of approximately 214kg of embodied carbon in each cubic metre of concrete – a reduction of 81% compared with a concrete containing 30% GGBS replacement (a mix that some might claim is low carbon itself).

A typical railway tunnel of 7.50m i.d. and segmental lining of 350mm thickness would contain approximately 8.6m3 of concrete per linear metre of tunnel. By using EFC in the segments, a carbon footprint saving of 1.84 tonnes per linear metre, or 1,840 tonnes/km of tunnel, could be realised.

Since EFC requires a lower curing temperature than OPC-based concrete for early strength and demoulding, there are possible further savings to be made in both energy costs and carbon footprint. This can only be quantified by working with the segment manufacturer and discussing their heating regime in the curing chambers.

In conclusion, geopolymer concrete can be produced to meets all the hardened concrete design requirements of tunnel segments, has excellent spalling resistance, improved long term durability, and a lower carbon footprint.

LP-2 PROJECT

The London Power Tunnels, Phase 2 (LPT-2) project is a £1bn investment by National Grid (NG) to ensure a continued safe and secure supply of electricity to London, by rewiring the capital via deep underground tunnels to house high voltage cables. It is a vital investment in the electricity distribution infrastructure to help keep Londoners connected to reliable supplies.

The LPT scheme began in 2011 with Phase 1, which is complete and runs across North London from Hackney to Willesden, carrying the power cables and infrastructure to meet around 20% of the capital’s electricity demand; stretched out, the cables would run all the way around the M25 motorway ring, encircling London.

Works on LPT-2 began in Spring 2020, and will take around seven years to complete. It will run 32.5km of 3m-diameter tunnels and eight shafts across South London, constructed at an average depth of 30m below the road network. The tunnels run from Wimbledon to Crayford and the new power cables they will carry are to replace the present, end-of-life, oil-filled cables.

The Hochtief-Murphy JV (HMJV) was awarded Package 2 (Tunnels & Shafts) of LPT-2 in December 2019. HMJV took over the three drive sites within three months of award and a week after the covid lockdown was announced. The JV was later also awarded Package 5 (Headhouses and M&E).

Close collaboration between the client (NG), contractor (HMJV) and designer (Aecom) has ensured the programme has kept on track. Work was undertaken with four TBMs: one refurbished Lovat open face wedgeblock TBM, ‘Caroline’, in the West; and, three new Herrenknecht earth pressure balance (EPB) machines in the mixed ground of the central section and the Chalk of the East.

As of the end of March 2023 (time of the BTS talk) progress on Package 2 saw 100% of the easternmost and two westernmost drives complete, with the drive from New Cross to Kings Avenue close to (93%) being complete, and 61% of the drive from New Cross to Eltham finished.

Why reduce carbon?

Reducing carbon is the biggest challenge for ensuring the viability of infrastructure projects, other than funding.

It is fundamentally the right thing to do.

Doing so aligns to national government policies and the UN Sustainable Development Goals (UN SDGs). Most companies have policies to reduce carbon and the project client on LPT is particularly strong with a commitment to achieve carbon-neutral construction by 2025/26.

One of NG’s Working Groups is on Low Carbon Concrete and HMJVs proposal to use EFC aligned well to this strategic objective.

A breakdown of the carbon footprint of LPT-2’s Package 2 is shown in Figure 4 with the embodied carbon of materials and transportation of materials contributing 61% and 15%, respectively, of the project’s total.

Looking at materials alone, the pie-chart in Figure 5 shows 55% of the carbon footprint relates to concrete and a further 34% relates to other uses of cement.

HMJV and NG worked collaboratively from the tender stage on reducing carbon on the LPT-2 project. At tender, HMJV optimised the tunnel diameter and construction methodologies saving 30% of the baseline (50,000 tonne of CO2 equivalent (CO2e)).

Post award, shaft diameters were reduced, alignment changes made, temporary works and rebar were optimised, and hydrotreated vegetable oil (HVO) and hydrogen were used in some areas as fuel.

On the materials front, there was adoption of high GGBS and EFC, and tunnel arisings were re-used wherever possible. These measures saved a further 17% on the baseline (18,130 tCO2e). Elsewhere, there are several measures to reduce environmental impact, such as a new 400kV substation at the heart of the project in Bengeworth Road, Lambeth, being built using SF6-free, gas insulated switchgear technology, in a UK first.

EFC trials

In early 2021, HMJV sought approval to trial EFC with a goal for use in both temporary and permanent works.

This process commenced with a visit to Capital Concrete’s Silvertown Plant, in London, for an EFC batching demonstration.

A C32/40 mix was selected to trial and the Bengeworth Rd site was chosen to be the initial location. NG accepted the proposals in May 2021.

The testing proposed was:

Fresh Concrete – BS EN 12350:2019

• Slump Testing and Slump Retention
• Fresh Density
• Bleed Testing Hardened Testing – BS EN 12390:2019
• Compressive Strength
• Density (saturated/oven dry)
• Flexural Strength
• Tensile Splitting
• Autogenous Shrinkage – Modified ASTM C157
• Total Shrinkage – ASTM C157
• Hot Box – Temperature Rise/Temperature Differential

There was a challenge during the trials associated with the distance to the one Capital Concrete plant able to batch EFC at the time, with loads arriving 1hr 40 mins after batching. This may have contributed to some difficulty experienced with the workability and placement of the concrete mix on one occasion. The gang also slipped into existing habits at another trial, such as adding water to the surface to aid curing – which has a detrimental impact on EFC as it dilutes the surface alkali activator.

However, even with the logistical challenges every load arrived on site to be within the S4 slump range. Slump retention was good, with the slump class maintained for 2hrs after arrival. More trialling, usage and workshops with Wagners will help engineers and site supervisors gauge EFCs behaviour, and the use of the product at more plants will of course help.

Results were very positive overall. Compressive strength and tensile strengths were in excess of the design requirements and the teams became more familiar with the product with use.

Shrinkage testing was performed at BRE, in Watford, and yielded very positive results. Both total and autogenous shrinkage values were comfortably within the limits required by the Aecom design team: autogenous (sealed) at 62 microstrain compared to a max target of 250 microstrain; and, total (unsealed) at 297 microstrain compared to a maximum target of 550 microstrain.

For the Hot Box test, a 1.0m3 insulated box was constructed by the HMJV site team as per instruction from our subcontract partner, OTB Concrete. Thermocouples were installed in the box. The results backed up existing data regarding EFC’s low reaction temperature – the maximum temperature rise was only 13.9°C and a peak temperature was 39.9°C.

Combined with the minimal temperature differential, these data prove EFC and the use of AACM could be very beneficial for reducing or eliminating the risks associated with early age thermal cracking in large pours.

One key factor in the approval of EFC for LPT-2 project was the first-year durability results published from BRE, in October 2022. At the one-year point, the EFC samples in the test were outperforming traditional DC-4 concrete in the most onerous Class 5 sulphate conditions. This gave confidence to all concerned that the durability results as shown by Australian testing were valid.

The durability results in conjunction with the other trial results were combined into test reports and material approvals. At the end of 2022, HMJV gained approval for use of EFC in both temporary and permanent works from both the designer of the permanent works – Aecom – and the client – NG.

Next steps

HMJV is planning workshops with Wagners, OTB Concrete, and Capital Concrete, to educate staff and operatives on EFC before further use.

Temporary works pours are planned on several sites and permanent works opportunities are also being investigated. As long as shaft pumping tests at Kidbrooke for temporary works are successful, HMJV plan to undertake a 736m3 permanent works pour at the base of a shaft at Hurst, which will be the largest continuous cement-free concrete pour in the world.

[Post lecture note: the Hurst pour for permanent works took place on Earth Day, Saturday 22 April 2023. Through it, HMJV and NG saved 80 tonnes of CO2e compared to a typical ‘low carbon’ concrete (ie high GGBS replacement).

Further trials are also being undertaken by HMJV on LPT-2’s Package 5 for a C40/50 mix, following the same testing as previously undertaken, with the addition of a self-compacting test, which will be a J Ring Test to BS EN 12360-12.

CONCLUSIONS

Geopolymer concretes provide a great alternative to high replacement mixes by providing much higher CO2 savings and good performance. Cost, however, can be prohibitive and EFC is currently 30%-40% more expensive, albeit this will reduce with more UK usage.

Availability has been London-centric. Capital Concrete had one plant, now two with one more to come. However, with Wagners now making the activator in the UK it is possible to tankerise the activator and modify an existing plant for use, depending on volume required.

All staff need to be briefed on use, and designers and clients need confidence that requirements can be met and testing will do this. Collaboration across the project is essential to reduce carbon and this requires a culture of trust and cooperation, and the speakers would like to extend a big thanks to all parties involved.

Murphy and National Grid have also separately trialled Cemfree concrete, which is an alternative geopolymer concrete. It comes in bagged form and can be used anywhere in the UK.

From November 2020 to November 2022, Murphy were part of the LOCOWAG (Low Carbon Concrete in Aggressive Ground) Group. As part of this Murphy Ground Engineering (MGE) conducted two trials: at HP11 Austry – Barton Gas Pipe Diversion for Cadent; and, at Newman Street, Westminster. Both projects used Cemfree in piles.

Cemfree met the design requirements for these temporary works applications. Trials were successful for the proposed use but showed less strength gain than EFC with C28/35 being reached at 56 days, and durability test results are awaited. MGE won a sustainability award at the Ground Engineering Awards for this work. Based on this testing, Cemfree can certainly be recommended for temporary works applications but more test results would be needed to support a permanent use.

NG worked at its Deeside Centre for Innovation with Cemfree, BRE, Hanson, and Centrum Piling over a similar period. It arrived at very similar conclusions on strength and durability.

FINALLY

Everyone has a part to play in carbon reduction;

• Clients need to decide how much worth to place in lower carbon solutions
• Designers/Clients need to allow use of alternative materials through performance based, rather than prescriptive, specifications
• Designers need to take a pro-active approach to using test results and remember that EC2 allows testing
• Contractors need to push benefits and adopt where they can
• Everyone should raise awareness, work together, set targets, plan, and share data

Information sharing is important. HMJV and NG are thankful for historic data from both Aecom Australia and Byrne Brothers, and our experiences from LPT-2 will be made available to others.