The Healing Process

When cracks began forming in the concrete roof of the emergency rescue station beside Lake Galder in the Netherlands, engineers sprayed it down with a bacterial solution and let nature go to work. This spray-on success was the result of years of research by microbiologist Henk Jonkers and his colleagues from Delft University of Technology (TU Delft) in the Netherlands.

The TU Delft approach relies on placing Bacillus spores together with a food source in the concrete; if water gets into the concrete, the spores awaken and bacteria feed on the food, producing calcium carbonate which acts as a stable, inert filler. The spores and nutrients are introduced into the concrete mix within expanded clay particles 2–4mm wide.

The Bacillus bacteria are alkali loving; they must be. Concrete has a pH value of around 12, a hostile environment for most life forms. Jonkers first located the bacteria from samples collected in Russian and African soda lakes, but he has since discovered that the bacteria are common in the UK. ‘They are present in all kinds of dust and soils. You can find them anywhere,’ he says, which means they are safe. They also last – viable spores have been found in museum specimens 200 years old – which is important if they are to be used for structural applications.

Jonkers currently uses calcium lactate, along with nitrogen and phosphorus, as the food source for the growing bacteria. ‘It took me a long time to find a suitable food source for bacteria that could be included in the concrete mix,’ he explains. Many organic materials, such as glucose and sucrose, have a detrimental influence on concrete, but not so calcium lactate. ‘Add sugar and the concrete will not set,’ says Jonkers.

Calcium lactate is commercially produced as an ingredient for all kinds of foods; however, as an additive for concrete, it is expensive given the quantities required. ‘We are checking for cheaper food sources, but we haven’t come up with alternatives yet,’ explains Jonkers. 

So far, Jonkers and his colleagues have successfully applied their bacterial solution to visible hairline cracks up to 0.4mm wide on parking decks and thin shell roofs. Cracks like these can result in severe leakage problems and premature corrosion of embedded steel reinforcement in wet environments. Jonkers has recently joined forces with a European Union project that aims to test the self-healing concrete in sections of a Danish tunnel in two years’ time.

In the UK, a team engineers and scientists from Bath, Cambridge and Cardiff Universities are using a combination of microcapsules, microbes and polymer fibres to heal concrete. The UK project, named Materials for Life, which was launched in July 2013, also has several industrial partners and the plan is to have demonstrations of all three technologies up and running by the autumn of 2014.

The Cambridge engineers are developing microcapsules, which will crack or disintegrate and spill out their ‘healing agent’ in response to the earliest signs of damage – as concrete sets, tiny cracks are already forming.

‘These [microcapsules] are basically little bubbles which could be anything from a few microns to 200μm in size. They contain a healing agent, so a resin or glue, or something like sodium silicate as a liquid. Or they could contain bacteria,’ explains Abir Al-Tabbaa, who heads up the capsule studies at the University of Cambridge. The team plans to design a range of microcapsules, starting with simple solid shells that rupture as cracks spread to ones that respond to electrochemical signals. The latter could respond to erosion or entry of chloride ions, for example. This would trigger cargo release, preventing further degradation or chemical ingress. 

Microcapsules are produced and used in a wide range of sectors, from food, through cosmetics, to drugs, and there are different routes to their manufacture. Al-Tabbaa is talking to other relevant industries because they have experience in the production of microcapsules, shelf-life, properties and applications. The issues in the construction sector are, however, different, acknowledges Al-Tabbaa. Lower cost and long shelf-life will be critical to their success; the capsules could be called into action decades after the concrete sets, she explains. 

Meanwhile, researchers at Bath University plan to optimise Jonkers’ bacterial approach and make it ‘more feasible and more economical,’ explains Kevin Paine of the Centre for Innovative Construction Materials at the university. They are investigating three species of Bacillus, all taken from a German culture collection, as well as engaging with extremophile expert Michael Danson at the university to see if new suitable bacteria can be found.

‘One of the principle things we want to do is use a cheaper form of calcium salt [as the bacterial food source], maybe calcium acetate or calcium nitrate or calcium formate,’ says Paine. ‘So we need a bacterium that will be very good at converting that into calcite, but we also need a bacterium that will sporulate easily.’ Easy sporulation will help Paine and his colleagues produce the bacteria on an industrial scale, essential for concrete-based applications.

The Bath strategy also relies on success of the Cambridge team’s work to develop microcapsules. Currently, Jonkers and his colleagues impregnate clay particles with nutrients and bacterial spores and add this to the concrete. This light-weight aggregate, says Paine, reduces the strength of the concrete. The Bath researchers want to place the bacteria directly into the cement, sealed and protected in microcapsules. 

The self-healing mechanisms developed at Bath and Cambridge will overlap in the scale of cracks they work on: microcapsules work on nano/microscale cracks, so from 0.1μm –50μm, the bacteria heal at the micro/mesoscale, typically 50μm to 1mm. But neither of these systems can deal with large cracks. This is where the Cardiff University team comes into the picture – its approach can handle cracks of up to 2mm width.

Civil engineer Diane Gardner and her group at Cardiff are working on a technique that embeds shape memory polymer fibres into the concrete mixing stage. The polymers are stretched and frozen before adding to the concrete. When heated they shrink and revert to their original form. ‘By heating the polymer, we close cracks in the concrete and allow other healing mechanisms to occur,’ Gardner explains. The Cardiff group has used polyethylene terephthalate (PET) and hope to make fibres out of recycled PET. When heated above its glass transition temperature, the monomers start to realign, which causes the polymer to shrink. 

The group has tested its system on small beams with 2mm width cracks and demonstrated good healing. The next step is to find out how the polymer copes on a larger scale, and the team will experiment by running an electric wire through the concrete to apply heat. Gardner hopes that enough shrinkage of polymer can be achieved so that cracks close before damage occurs at the macro scale. ‘This would take it a step further and prevent damage even occurring,’ she says.

The combination of the three approaches – bacteria, microcapsules and polymer fibres –  has already attracted 20 industrial partners, including Mott Macdonald UK, Laing O’Rourke, Costain and National Grid.

The nuclear industry is interested in this self-healing cement for waste encapsulation and for structures that are difficult or impossible to inspect, as are the oil and gas industries, says Al-Tabbaa. Other uses could include soil-cement mixes that support construction projects such as dams. 

Paine believes that marine infrastructure could also benefit, though he notes that in this situation the bacteria would have to be alkali-loving and salt tolerant. This raises the prospect of multiple, specialised bacterial strains, depending on the application. ‘You would want bacteria that work in all situations, but we could end up with different bacteria for different environments. If you are using concrete in the marine environment, say for a wind turbine, you might need a different form of bacteria.’

Paine considers microencapsulation to be the biggest challenge facing the project. ‘You need a capsule, which will not break while the concrete is being mixed, and an awful lot of energy goes into a concrete mixer. Then when the concrete hardens they need to withstand the calcium silicate hydrate forming around them and not break at that point,’ he says, ‘but it will need to break and unload its cargo under cracks or strains later on....You could end up with microcapsules too strong or too weak. Getting that correct is going to be a real challenge.’

Meanwhile, Jonkers is also interested in using fibres for patch-repair cement, which could be added to damaged areas. Currently, synthetic fibres such as polyvinyl alcohol fibres are often used for repair, but they tend to peel off after five years. Fresh concrete is another option, but this, Jonkers explains, shrinks and detaches. He believes his [bacterial-based] self-healing concrete combined with fibres, could prove more useful.

So far, he has tested polymers, but he plans to test bio-based fibres in irrigation canals in Ecuador in summer 2014.

Spotting and repairing damage to these canals is labour intensive and they can be difficult to reach. ‘For this field study, we are going to produce concrete parts for irrigation canals using abacá [a banana species] fibre-based materials, combined with the bacteria-based self-healing [concrete],’ he explains. A cracked part of the irrigation canal will be replaced and monitored over time to see whether occurring cracks will indeed self-heal as expected. 

Finally, Portland cement, although widely used, is by no means the only cement being used today. Increasingly, blends with various additives are used. This complicates the solution for self-repair. It means any solution will have to be tested on the various materials, but Paine says this is where the industry partners will play a key role. ‘They can ask some of the real practical questions, which often academics don’t consider until much later.’ 

Cement denotes a category of materials that react with water to form an adhesive matrix.

The most widely used Portland cement, so called because it’s colour resembles the colour of stone found on the isle of Portland in Dorset, is made by turning together limestone and clay at a very high temperature, of ca 1500°C, and adding gypsum(CaSO₄.2H₂O), before grinding to a fine powder. This powder mainly comprises calcium silicates, with some calcium aluminates and calcium sulfate. When the powder is mixed with water, hydration and hydrolysis takes place, forming calcium silicate hydrates and calcium hydroxide, along with other by-products such as calcium sulfoaluminate hydrates. ‘These hydrates form slowly over time and create a complex gel which itself contains pores and between which there are pores,’ explains concrete expert Paine. As the gel hardens, it becomes less porous, a process that can go on for years.

Concrete then is a mixture of cement and aggregate. The aggregate is one of two types: fine, for example, sand or gravel; or coarse, for example, crushed rocks or stone ‘If you consider the aggregate is inert then all the interest is around the cement,’ says Paine, although aggregates are never truly inert and ‘all sorts of problems can occur when aggregate reacts with cement’. The complexity of the changes that can occur over the lifetime of concrete is thus a real challenge for any potential self-healing mechanism.