Ebook: Self Healing Materials

Soon after the landmark paper by Scott White and colleagues [1] demonstrating the concept of autonomous self healing in a man-made material, materials scientist in the Netherlands realised that the concept of self healing is not restricted to brittle polymers, but can be applied to all other classes of materials. This realisation was based on their conceptual insight that self healing requires the creation of ‘local, temporary mobility’ in an otherwise regular material [2]. In the Scott White concept, the material to be turned into a self healing grade was an existing regular high performance epoxy for aerospace applications. The ‘mobility’ was created by the use of a liquid healing agent (liquids being mobile by their very nature). The ‘local’ condition was met by encapsulating the liquid healing agent. The condition ‘temporary’ was also met as the liquid only was able to flow in the period between the crack opening the capsule and the onset of cross-linking upon contact with the catalytic particles dispersed in the matrix. As we all know, the concept of local temporary mobility also works in nature. If one's skin is cut, blood will temporarily be able to leave the veins and to flow into the cut. Once in direct contact with air, the blood will rapidly cross-link and the liquid blood becomes a flexible (non-flowing) gel which over a number of successive chemical reactions turns into new skin. Self healing of bones in a human body works in a similar way. First, there is a flow of electrolytes and proteins from the surrounding tissues to the fracture surface. Then, a bridging polymeric gel containing a high proportion of inorganic particles is formed, which in turn is transformed into new bone with a new microstructure to compensate for the loss in original optimised macrostructure. We also know from nature that healing does not occur in all cases: broken teeth just do not heal. To make our teeth function optimally they must be hard and sharp and not react to any food we may eat. Hence, in teeth the bone structure is very dense (not allowing space for microscopic flow) and chemically very stable (not allowing new bonds to form).

The above examples all deal with flow of a liquid-like substance over microscopic dimensions, but the concept of local temporary mobility is generic and can be applied at different length scales. In some self healing polymers that are capable of multiple healing, weaker (reversible) molecular bonds break sacrificially while keeping the polymer backbone intact. The dangling chain ends move around for a certain time in the vicinity of the fracture surface and then they recombine with dangling chains from the opposing fracture surface and lose their mobile state.

The challenge of creating self healing materials is not to create a totally new material (like CNTs and graphene being new materials) but to modify existing materials such that upon local fracture or scratching, they autonomously (or upon a slight stimulation) restore their prime functionality, be it mechanical strength and stiffness, water impermeability, corrosion protection or surface friction or any other functionality. Self healing concrete is essentially like any other concrete with all its good and bad properties, but with the added ability to heal cracks. Self healing creep steels are like any other creep steel with all their good and bad properties, but with the added ability to heal creep damage as it forms. Self healing organic coatings are like any other organic coating with their usual processing and properties characteristics, but have the ability to restore adhesion to the substrate and restore the corrosion protection of the underlying metallic substrate.

So, to make a material self healing, material designers only have to focus on how to create the three requirements local temporary mobility and to respect the ‘natural’ character of the material [3].

It is important to point out the fact that the three key concepts local temporary mobility already imply that healing will never be instantaneous, but will always take time. The healing ‘agent’ will have to ‘flow’ to the damage site, it has to interact with the crack faces and it has to ‘transform’ from a mobile state into a more solid state. This all will take time. The actual amount of time will depend on the nature of the material as well as the prevailing conditions. Healing of fractured bones takes typically 6 weeks as the deposition of the new organic-inorganic material in the crack is relatively slow. The net material transport rate is low and the conversion of the temporary flexible gel-bone structure into the strong, more rigid bone structure is a slow process too. In contrast, the healing of a cut in a human skin is relatively fast and typically occurs in 1 to 3 minutes. The short healing time is due to the rapid and abundant supply of blood through the veins in combination with a fast gelation of blood upon exposure to air. As we all know, the healing time also depends on the size of the cut. Small and superficial cracks take a shorter time to heal than deep and wide cuts. The healing time also depends on the prevailing conditions, just as in nature: while a healing time of 1 to 3 minutes for a skin cut is acceptable, healing of a cut in a tongue is a lot faster. The biological conditions in the mouth promote rapid and non-scar-forming repairs. The mobility aspect means that healing does not only depend on local chemistry but of course also on temperature. Routes which work well at room temperature may not work so well under arctic conditions, and routes which do not work so well at room temperature may work well at high temperatures. Finally, the healing time should always be seen in the context of the application. For restoring fluid containment after ballistic penetration a fast healing reaction in the sub second regime is required. For concrete buildings with an expected life time of 40 years or more, the healing may take longer and a healing time of weeks may be adequate.

In an ideal case, the self healing process would be truly autonomous, i.e. not requiring any human intervention and multiple successive healing events at the same damage location should be possible. Furthermore, ideally the new self healing material would have properties equal to those of the current non-self healing materials on the market. To make the wish list complete, the self healing material should have the same (if not lower) cost price as existing materials.

Let us first address the issue of reduction in property levels when turning a material into a self healing grade. In self healing materials we have to set aside a certain fraction of chemical bonds/atoms/molecules in the material in order to make it self healing. The mobile ‘healing’ bonds/atoms/molecules do not contribute to the principal mechanical or functional properties, and some reduction in initial properties is not to be avoided. However, in real applications it is not the material performance we want to maximise, but the performance of the product we want to make out of it. In designing a product constructed out of non-self healing materials, the designers and engineers apply a so-called ‘safety’ factor. This ‘safety’ factor takes into account the reduction in material performance due to the accumulation of unspecified damage during use. In case of an ideal self healing material, the damage is fully and perpetually healed and no safety factor is required. So, some reduction in initial properties is acceptable provided the loss in properties is less than that following from the application of a ‘safety’ factor [2]. There are numerous practical examples where self healing behaviour could lead to a product having an extended life time even if the initial material properties were lower:

– The Colosseum in the city of Rome would not be standing firm after more than 2000 years if the mortar used has not been self healing;

– A corrosion protective self healing polymeric coating would maintain the local corrosion protection even in case the coating was locally damaged by mechanical impact;

– Early failure in LED's would be prevented in case the thermal interface material connecting the die to the lamp structure would re-attach itself to both surfaces upon overheating due to local delamination;

– More slender concrete buildings could be constructed in case the concrete was self healing and moisture transport to the steel reinforcing structure was autonomously prevented;

– Pressure vessels in power stations would last longer in case self healing steels would resolve grain boundary pore formation and coalescence internally.

In all these cases it is not the performance of the undamaged material which is the limiting factor, but the response of the material and structure to unavoidable local damage.

When analysing the potential application fields for self healing materials, several fields of application can be identified. Most of them have a more secure performance of the product as a whole as the main driver for material change. The use of self healing materials is foreseen in applications where performance reliability after the occurrence of undetected minor damage has to be guaranteed (e.g. aircraft, ships, flood barrier constructions). Applications where a very long life is to be guaranteed under a spectrum of only partially known conditions (e.g. tunnels, bridges, conventional and nuclear reactors) are also of interest. While not always of structural relevance, in many applications a guaranteed high-quality outer surface (e.g. as in automotive exteriors and interiors, compact discs for data storage, glasses and optical devices) is highly appreciated and a factor making a crucial difference in the product appreciation. Finally, all applications where repair to local damage is costly and/or disruptive (such as sewage systems, sea-based wind turbine farms and motorways) would greatly benefit from self healing materials being available.

Finally, all new materials can only enter the market if the increase in performance outweighs the increase in cost. In this respect, self healing materials have an extremely high probability of entering the market. As stated earlier, self healing materials are essentially existing materials with an added functionality. In contrast to other novel materials such as diamond like coatings, room temperature superconductors, aramid fibres, fibre metal laminates carbon nanotubes and graphene, no new process and application technologies need to be developed. The processing of self healing polymer coatings and the processing of self healing concrete will in most cases require the same processing equipment as being used for their current non-self healing predecessors. The absence of a need to invest in totally new processing equipment will substantially reduce the introduction costs of the new self healing material grades. Of course, any material modifications to give new functionality will increase the material costs, as the complexity of the processing is likely to increase and the new modifications will generally not be as cheap as the base material. However, the increase in material cost in most cases can be kept at a relatively low level. Furthermore, in modern times, cost of a product, installation or piece of infrastructure at the moment of acquisition is no longer the dominant parameter. Material selection is increasingly made on the basis of a product life cycle analysis in which maintenance cost, life time extension, reliability issues related to early failure and end-of-live aspects play an equally important role as minimisation of the acquisition costs. A recent study on the pricing of a (very) low cost material (porous asphalt concrete for Dutch motorways) has shown that that a doubling of the bitumen cost to make it self healing would lead to a net profit of 20 MEuro per year if this would increase the life span of Dutch motorways by 50% [4]. Other studies (e.g. for self healing coatings for off-shore installations in the oil and gas industries) have shown an even much larger acceptable increase in cost price, provided the coating becomes truly self healing for a number of years.

Inspired by the scientific and technological challenges and the opportunities to create materials making products and installations last longer, at a lower maintenance cost and in a more reliable manner, a proposal for a programmatic study into the development of self healing materials was submitted in 2005 to the Dutch Ministry of Economic Affairs. The proposal was submitted by a consortium of academic experts working in the field of polymers, metals and concrete and several companies ranging in size from a multinational to a small company of less than 100 employees. The application was made in the context of the highly successful 25 year old Innovative Research Program (IOP) programmatic funding scheme of the Dutch Ministry of Economic Affairs. The aim of the proposal was to conduct pioneering research in the emerging field of self healing materials. The program differed from smaller programs outside the Netherlands by covering all material classes and not specifying the route to reach the self healing behaviour. In the proposal the consortium committed itself to transmit the key information on the new self healing material systems to the industry in such a manner that the new concepts would lead to new high-value products entering the market in record time.

The proposed IOP program Self Healing Materials was to be supervised by a Supervisory Board with members both from industry and academia. The research was to be conducted via projects each having one or more academic partners and at least two industrial partners per project. Professor Sybrand van der Zwaag (Delft University of Technology) was proposed as the director for this program. The total financial contribution requested amounted to 10 MEuro for a four year program with the option of an extension of the program for another four years and another financial contribution of the same magnitude.

The key element in the original IOP program proposal guiding the selection between the many high quality research projects submitted during the course of the program, was a table defining the distribution of the funds over the various material classes and types of activities. These percentages were the outcome of stimulating discussions during several preliminary workshops involving senior and junior representatives from companies, universities, the Ministry of Economic Affairs and institutes active in the field of materials science and engineering in the Netherlands.

As the table shows, a large share of the research funds (40%) was to be devoted to the field of self healing polymers. This was based on the notion that in comparison to other material classes, polymers have a much higher molecular mobility at room temperature and their processing conditions are generally milder. Opportunities were seen for both thermoset and thermoplastic polymers. The research was to focus both on extrinsic (i.e. the healing was to be realised via the addition of discrete encapsulated liquid healing agent containing entities) and intrinsic (i.e. the polymer molecule itself can heal the damage via reversible chemical or physical bonds) self healing approaches. The metals domain would receive a lower fraction of the funding (20%) as it was realised that self healing in metals would be difficult to realise. Due to the high melting point of metals the intrinsic mobility of atoms at room temperature would be very low. However, opportunities were seen for age hardenable alloys and creep resistant steels. Earlier work on self healing materials at the Delft Centre for Materials (Delft University of Technology) had identified some very promising leads to induce self healing in asphalt and concrete, so a relatively large fraction (25%) of the national budget was allocated to the domain of civil engineering materials. Finally, a substantial amount of funding (15%) was to go to the domain of fibre reinforced polymer matrix composites. The relatively high level of funding was based on the high added value of composite products. Furthermore, it was realised that the fibrous architecture of continuous fibre composites offered attractive opportunities to create long range transport of the healing agents. As to types of activity, most of the effort was to be devoted to the development of new concepts, yet it was widely agreed that substantial funding should go to the development of a theoretical basis and developing the computational framework. Much to the surprise of the Ministry and the senior academic administrators, the industry indicated that the financial commitment to application oriented research should be kept small. Intentionally, the fractional funding at cell level was left undefined, so as to give the Supervisory Board of the IOP the chance to build a coherent research program on the basis of the best and most innovative research proposals submitted.

At the mid-term point of the program a new priority table was defined, taking into account the lessons learned during the first four years of research.

Compared to Table 1, there were some significant changes in the grouping of materials to be studied as well as in the relative funding levels. It was found that the research on self healing composites had focussed almost exclusively on the healing of the polymer matrix. Hence, a new domain Polymers & composites was defined. The research on self healing in metals was found to be most successful in the field of high temperature metals (creep steels) and opportunities were seen to extend the field to high temperature ceramics. This led to the creation of a new domain Metals & ceramics, devoted to the healing of (bulk) metallic and ceramics. The domain of Civil engineering materials was left unchanged as it was found to function very well. While strictly speaking not a material class, Coatings were defined as an area of special attention. By their very nature coatings have an increased probability to become damaged due to external impact or mechanical loading. Furthermore, the presence of an unfractured substrate to keep the fractured coating surface in some form of topological registry offered an additional positive contributing factor. The inherent limited thickness of the coating presenting an additional constraint to the mass transport to the crack in comparison to healing of bulk materials. Research in the newly established coating domain was to focus on organic (polymer) coatings as well as on inorganic (metallic or ceramic) coatings. Finally, it was realised that there was an opportunity to open the field of self healing to Functional materials (i.e. materials which prime function is non-mechanical in nature, such a conduction of heat, emission of light or stimulation of an electrochemical reaction). To stimulate the research in this field, a new domain Functional materials was created and 15% of the budget was allocated to it. The apparent reduction in fractional funding for polymers and metals with respect to that in Table 1 was in part ‘administrative’ as both material classes were to be addressed in the domain Coatings. As to activities, more attention was to be given to the development of protocols for material testing and quantification of degree of self healing. Furthermore, in particular at the request of the participating industries, the research on theory and modelling was to be strengthened.

The IOP program financed 42 projects, involving 20 research groups located at 5 Dutch universities and no less than 68 companies. This book presents a highlight of the results obtained in the program. The chapters are structured along the domains defined in Table 2.

The first 9 chapters of this book all address research in the field of Polymers & composites. The section starts with a chapter on mechanochemistry of individual molecules in a polymer network and shows how fracture of a single molecule can trigger a catalytic reaction restoring the network again. The subsequent five chapters deal with various new polymer systems in which reversible chemical bonds are used to induce self healing behaviour. The new polymers are to find their way in the human body, or as robust elastomers for engineering and coating applications. These five chapters are followed by a chapter on the (only partially successful) creation of robust silica based capsules filled with liquid healing agents stable enough to withstand regular thermoplastic processing routes. The section ends with two chapters dealing with self healing in fibre reinforced composites. The first of these two chapters deals with a novel fibre concept for the storage of healing agents, while the other chapter deals with self healing in a composite structure involving the use of shape memory alloy wires to close the cracks formed.

The section Metals & ceramics contains 5 chapters. In the first chapter it is shown that under specific conditions dynamic precipitation at deformation induced defects can lead to a significant increase in the fatigue lifetime of aluminium alloys. The subsequent two chapters deal with research on extending the life time of steels by the autonomous filling of grain boundary cavities. It is demonstrated that both copper and gold are excellent alloying element to induce self healing in steels at temperatures as high as 550 0C. The final two chapters in this section focus on ceramics and deal with self healing behaviour at even higher temperatures. Significant advances in the wear resistance are reported.

While the previous sections contain chapters describe research on materials not yet ready to enter the market, the section Civil engineering materials contains 7 chapters dealing with research on both concrete and asphalt, spanning the entire range from lab scale experiments to real life field testing. The chapter also contains a numerical study on the rate of healing in bacterial concrete as a function of the microstructure and the crack dimensions. Models like these are crucial when re-designing and redimensioning real constructions on the basis of the self healing behaviour of the material. The two chapters on the prevention of frost damage in blast furnace cement and mortars describe more early stage work, but also they hold great promise for future applications.

In general, coatings are applied to provide various functionalities to different types of substrates and this diversity is reflected in the Coatings section. The section contains 6 chapters on the protection of wood by biofilms, self replenishing low friction polymeric coatings, self healing anti corrosion coatings on aluminium substrates and thermal barrier coatings on super alloys. Some of the work is still at its exploratory stage but the coatings for low friction, anti-corrosion and thermal protection are well on their way to commercialisation.

The book ends with a section on self healing Functional materials. Functional materials are materials which have different primary role than the handling of mechanical loads. The section contains only 3 chapters, yet describes truly pioneering research in the field of self healing in thermal interface materials for LED's and micro-electronic devices, layered organic light emitting diodes and catalytic nanoparticles in fuel cell membranes.

Finally, while this book shows the impressive range of results obtained purely in the context of this IOP Self Healing Materials program, the IOP program also initiated additional research together with Dutch Technological Top Institutes such as the Netherlands Institute for Metals Research and the Dutch Polymer Institute. It led to five EU programs with a strong Dutch participation. It also was responsible for the organisation of the first (ICSHM2009) and fourth (ICSHM2013) international conference on self healing materials and five international Summer Schools. The IOP program led to several new academic positions and new university courses on self healing materials, and many international collaborations. Finally, the IOP program led to new industrial products and markets and most importantly to vibrant academic-industrial networks.

We, the editors and authors of the 30 chapters, are convinced that this book will stimulate further development in this promising and highly attractive field of self healing materials. There are still many new concepts to be explored, existing concepts to be brought to a higher technology readiness level, legislation issues to be addressed and industrial experiences to be translated into improved versions of the first pioneering families of self healing materials before society can fully benefit from the full potential of self healing materials.

We trust the book will be a source of inspiration and reference for anyone with an interest in self healing materials for many years to come

Acknowledgements

The IOP Self Healing Materials program and this book would not have been possible without the contributions of many institutions and individuals. We acknowledge the crucial contribution of the Delft Centre for Materials (Delft University of Technology, Delft, The Netherlands) in opening the field and laying the conceptual foundation for this IOP program. During the entire program the Supervisory Board played a crucial role in guiding the development of the research and we thank its members Richard van de Hof (vice chairman), Jasper Michels (vice chairman), Ludo Aerts, Rinze Benedictus, Andre de Boer, Rolf van Benthem, Jan Bottema, Klaas van Breugel, Pieter Geurink, Jeff de Hosson, Gert-Jan Jongerden, Carel Kleemans, Cor Koning, Dick Koster, Henk Maatman, Jan Mijnsbergen, Ruth van de Moesdijk and Stephen Picken for their contributions and sustained commitment. The successive program coordinators Patrick van Veenendaal, Joris Vogelaar and Annette Steggerda-Batenburg made exceptional contributions to the smooth functioning of the program and solved many administrative issues. Finally, we thank all project leaders, industrial partners and the many postdocs and PhD students for their hard work and creative ideas. The excellent support from various departments of Rijksdienst voor Ondernemend Nederland (formerly known as AgentschapNL) during the execution of the IOP program is gratefully acknowledged too.

References

[1] S.R. White, N.R. Sottos, P.H. Geubelle, J.S. Moore, M.R. Kessler, S.R. Sriram, E.N. Brown & S. Wiswanathan, Nature 409 (2001) 794-797

[2] S. van der Zwaag, “Self Healing Materials: an alternative approach to 20 centuries of materials science”, S. van der Zwaag (editor); Springer, Dordrecht, The Netherlands (2007)

[3] S. van der Zwaag, N.H. van Dijk, H.M. Jonkers, S.D. Mookhoek & W.G. Sloof. Phil Trans Roy Soc A 367 (2009) 1689-1704

[4] “Self Healing Materials; concepts and applications”, NL Agency, The Hague, The Netherlands (reprint 2014)

Autonomously self-repairing and -healing processes will play a key role in the development of future ‘smart’ materials and devices. Here we present the development of a novel second generation Grubb's catalyst that can be activated through the application of mechanical force in solution as well as in the solid state. We show thorough kinetic analyses as well as extensive optimization for the improvement of the catalyst's activity and lifetime and have eventually translated the concept to the solid state. It could be successfully proven that mechanical activation initiates the ROMP to linear as well as cross-linked polymers in the solid state and by this we present for the first time the autonomous repair of a material that relies on the mechanochemical activation of catalysts on the molecular level. We are certain that this tailor-made prototypical catalyst-system establishes an important step for the implementation of self-healing materials in an everyday environment and are confident that this novel motif will stimulate future research on this field.

We have designed, prepared, and characterized a class of advanced intrinsic self-healing polymer nanocomposites where the self-assembled filler skeleton formed by nanoparticles is embedded in a continuous phase of a supramolecular elastomer. In intrinsic self-healing systems the ability for a material to repair itself is achieved by keeping the required mobility in the system while using nanoparticle reinforcement to achieve the overall stiffness and resilience. This means that it is indeed possible to design a material that reconciles both mobility and stiffness, which at first sight might seem a tall order. Unlike self-healing materials based on encapsulated reagents, solvents or adhesives, the intrinsic self-healing mechanism is engrained into the material itself, and self-healing can be repeated indefinitely, also at previously damaged locations. The reinforcing skeleton that interpenetrates through the polymer matrix not only prevents the supramolecular elastomer matrix from flowing, but also provides the nanocomposite with load-bearing capabilities. The supramolecular polymer we discuss here is boric acid modified poly(dimethylsiloxane), PBS. During the PBS synthesis boric acid cleaves silicone oil chains (PDMS) forming hydrogen bonding boric acid end-groups. These end-groups can subsequently also form covalent crosslinks via reversible boroxol rings. The supramolecular PBS elastomer is responsible for integrating the two-component nanocomposite as a whole by means of its reversible intermolecular H-bonding cross-links. By fine tuning the nanoparticle-molecule adhesion, the particle aspect ratio, and the particle network formation, we are able to obtain a load-resistant nanocomposite without prolonging the healing time.

Traditional polyetherimides (PEIs) are synthesized from an aromatic diamine and an aromatic dianhydride (e.g. 3,4'-oxydianiline and 4,4'-oxydiphtalic anhydride) leading to the imide linkage and outstanding chemical, thermal and mechanical properties yet with a complete absence of self-healing functionality. In this work we have replaced the traditional aromatic diamine by an aliphatic dimer diamine made from renewable resources. Such an approach led to the synthesis of a whole family of self-healing polymers capable of healing at room temperature or at elevated temperatures and ranging from relatively soft elastomers to rigid polymers depending on the composition. Here we report preliminary data showing the effect of the offset from the theoretical stoichiometric ratio on the general performance and healing behaviour of a room temperature healing polyetherimide. The systems with a slight excess of branched linear diamine and a post treatment of 12 hours at 150 °C showed the highest stress-strain properties (8 MPa and 500% elongation at break at 80 mm/min in tensile mode) with a recovery of properties close to 50% after 24 hours of pressure-less healing at room temperature. Lower initial properties yet higher healing efficiencies are obtained for shorter post annealing treatments.

Hydrogels are water swollen polymeric networks, capable of absorbing large amounts of water or biological fluids. The network is formed by cross-links between the polymeric constituents. These cross-links can either be physical or chemical. Chemical cross-linking methods have the advantage that the cross-linking density can easily be varied and with this also the mechanical properties of the final hydrogel. Physically cross-linked hydrogels are on the other hand adaptive and self-healing by nature. A combined chemically and physically cross-linked network could have highly interesting features from a self-healing materials point of view. In recent years, a number of chemical reactions have been utilized for hydrogel cross-linking. Since hydrogels hold great promise in a variety of biomedical applications, there is a need for novel cross-linking methods, especially those that are biocompatible. In this article, we describe the biocompatible cross-linking reaction SPAAC (strain-promoted azide-alkyne cycloaddition). In the SPAAC reaction, the highly selective reaction between a ring strained alkyne (BCN) and an azide simply occurs in water without any other additives. Poly(ethylene)glycol (PEG) hydrogels were successfully formed using SPAAC as the cross-linking method. We especially focussed on preparing soft hydrogels, mimicking the stiffness of soft body tissues. Confocal microscopy studies performed on the hydrogels revealed that gels containing the cell-adhesion peptide RGDS are a good substrate for cellular adherence. One of the disadvantages we encountered when using SPAAC was that it is hampered by slow reaction kinetics. Furthermore, an activatable cross-linking method is desired, especially in the field of injectable hydrogels. We therefore developed a novel cross-linking method which is both fast and activatable. This fast reaction between a catechol (DHPA) and BCN was called SPOCQ (strain-promoted oxidation-controlled cyclooctyne-1,2-quinone cycloaddition). The SPOCQ reaction only occurs upon oxidation of the DHPA (catechol) which can be performed both chemically and enzymatically. Both oxidation methods resulted in fast hydrogel formation. We showed that the SPOCQ and SPAAC reaction can be used in one pot, SPOCQ for the fast hydrogel formation and subsequently the SPAAC reaction for functionalization of these hydrogels. We also give an outlook on combining these chemical cross-linking methods with physical cross-linking by incorporation of the calcium-binding motif alendronic acid.

Colloidal gels are defined as continuous networks of aggregated particles dispersed in a liquid phase. This type of gels can be formed by self-assembly of particles as the colloidal building blocks. Owing to reversible non-covalent interparticle interactions between these building blocks, colloidal gels can display self-healing behavior upon destruction of the gel network. For instance, self-healing organic colloidal gels have been developed by self-assembly of oppositely charged gelatin nanoparticles, where the mechanical properties and self-healing capacity of the colloidal gels strongly depended on the size of the colloidal building blocks. Furthermore, in order to develop colloidal gels with enhanced elasticity, combinations of organic and inorganic building blocks were recently explored. Since inter-particle forces determine the final properties of colloidal gels, several surface functionalization strategies are being explored to vary the adhesion forces between particles aiming at optimized self-healing behavior.

Bone is a well known natural self-healing composite material, which consists of an organic component in the form of a collagen fibril-based hydrogel and an inorganic component in the form of calcium phosphate (CaP) crystals which reinforce the hydrogel.

In this project, an enzymatic method inspired by the self-healing properties of bone was applied to generate hydrogel-CaP composite biomaterials intended for applications in bone regeneration. This biomimetic approach involved the enzyme alkaline phosphatase (ALP), which mineralizes bone tissue with CaP. ALP was incorporated into several hydrogel types to induce mineralization, after which physicochemical and cell biological characterization was performed. Physicochemical characterization involved determination of the amount and type of CaP formed and its distribution within the hydrogel, changes in mechanical strength resulting form CaP formation and morphology of the CaP deposits. Cell biological characterization involved assessment of the adhesion, viability and proliferation of bone-forming cells (osteoblasts). The approach developed in this project is applicable to any kind of hydrogel.

As expected, mineralization led to an improvement in compressive strength. The amount and type of CaP formed was dependent on hydrogel. Generally speaking, predominantly calcium-deficient hydroxyapatite CDHA formed in hydrogels containing a large number of calcium-binding chemical groups, e.g. catechol/poly(ethylene glycol) (cPEG), gellan gum (GG), GG enriched with polydopamine, while a mixture of CDHA and amorphous CaP formed in other hydrogels, e.g. platelet-rich fibrin (PRF) collagen, oligo(poly(ethylene glycol) fumarate) (OPF) and chitosan. CaP formation was predominantly in the outer region of hydrogels. In the case of chitosan hydrogels, ALP not only caused mineralization, but also acceleration of hydrogel formation. In the case ofGG, CaP formation also significantly promoted osteoblast adhesion and proliferation. The results obtained led to four journal publications and one patent and pave the way for the application of the hydrogel-CaP composites as implant materials for bone regeneration.

We report the synthesis and properties of solvent-filled silica microcapsules to be used in thermoplastic extrinsic self-healing polymeric systems processed via conventional extrusion processing. The silica microcapsules were produced via hydrolysis and condensation of tetraethyl orthosilicate (TEOS) using an aqueous glycerol microemulsion as the template. The effect of the solvent/TEOS ratio on the dimensions and wall thickness of the microcapsules was studied. Pre-hydrolysis of TEOS was found to be essential to obtain isolated free-flowing silica microcapsules. Microcapsule compression loads of up to 350 mN were obtained. Full details of the work have been reported elsewhere [1].

This chapter summarizes the research on the development of compartmented fibres for self-healing fibre reinforced polymer composite applications. The compartmented fibre concept combines the advantages of both capsule based and hollow fibre based extrinsic healing strategies: the mesoscopic fibre geometry facilitates the spatial distribution of the healing agent in a fibre reinforced composite while the microscopic vacuoles within the fibre filled with a healing agent allow for very local and distributed healing events. Fibres containing the healing agent discretely distributed along the fibre can be obtained from an alginate emulsion containing the healing agent as the dispersed phase spun into a coagulation bath. The vacuole characteristics can be tuned by adjusting the emulsion and the spinning conditions. Fibres containing the self-healing agents have been embedded in both thermoplastics and thermosets and demonstrated the high potential of this alternative concept for self-healing composites.

Continuous fiber reinforced composite materials are susceptible to matrix cracking and delamination upon impact. Active and passive wires can be embedded within the composite material to support the healing behavior. Upon a local heating stimulus the wires, oriented mostly in the out-of-plane direction, can assist the buildup of compressive stresses to close matrix cracks and delaminations. Shape memory alloy wires, prestrained to contract upon heating, can actively close cracks in a damaged region during the heat treatment. A closing action can also be achieved in a passive sense, based on differences in thermal expansion of the composite material and the wires employed.

The effects of the wire type, fraction and distribution on the closing and healing behavior of composite materials were studied for a simplified model case. The requirements for optimal matrix healing have been determined. The results are strongly dependent on the thermo-mechanical properties of the composite material and the wires. Small amounts of shape memory alloy wires are already sufficient to generate adequate compressive stresses at the healing temperature. Prestraining of the wires is not a prerequisite. Passive wires, such as glass and aramid fibers, can be employed in a similar way. A stitch type of wire distribution is more effective than a woven type of distribution.

Experimental verification of the closing behavior of active and passive out-of-plane wires has been carried out through consolidation experiments of pre-impregnated composite layers. Active shape memory alloy wires were capable of generating sufficient compressive stresses in the out-of-plane direction to support closure and welding of separate composite layers. The fracture toughness obtained was comparable to that of a reference plate of the same material manufactured with a hot press. The active wires also coped better with size variations in the out-of-plane direction than passive wires based on glass and aramid fibers.

The option of extending the fatigue life of precipitation hardenable aluminium AA2024 alloys under high cycle fatigue loading conditions by dynamic precipitation in under-aged material has been explored. The fatigue life enhancement should be due to formation of clusters or precipitates at deformation induced damage sites which is to retard crack initiation. To this aim the fatigue lifetime and microstructural changes due to fatigue loading of peak aged and under-aged aluminium AA 2024 samples, containing low and high solute levels respectively, have been studied.

For fatigue testing at an R-value of −0.4 a substantial increase in fatigue life time for the under-aged material was observed at stress ranges at 265 MPa. Detailed microstructural investigations at (sub-)nanometer level using Positron Annihilation (PA), Thermo-Electric-Power (TEP) TEM revealed changes in the solute level and precipitate state during cyclic static loading in the plastic regime but no changes in solute level during fatigue loading were detected. Hence we did not succeed in correlating the improvement in high cycle fatigue life times for under-aged material to dynamic precipitation effects. More work is required to determine the effect of higher solute levels on the fatigue lifetime for a wider range of testing conditions.

This work is a study into the possibilities of self healing of early stage creep damage (i.e. grain boundary pores) in iron based model alloys by site-specific precipitation of substitutionally dissolved Cu atoms. The precipitation of copper is found to occur in the form of (a) spherical nanoscale precipitates within the matrix, (b) decoration of dislocations and/or (sub)grain boundaries and (c) precipitation at free creep cavity surfaces. Due to the comparable atomic size of Cu and Fe the copper precipitation is found to be only weakly site specific and does not have a strong preference for creep cavity surfaces. The addition B and N to the Fe-Cu alloys generally retards the Cu diffusion along dislocations and (sub)grain boundaries. This does not change the overall precipitation mechanism, but modifies the precipitation rate and distribution over the potential precipitation sites. No indications were found for an independent healing mechanism for creep damage by BN precipitation within creep cavities for the studied ferritic Fe-Cu alloys. The defect-induced precipitation of Cu was only weakly affected by the introduction of about 0.1 wt.% C in the alloy.

This chapter summarises the main findings of the work already published in the open literature.

Autonomous repair of creep damage can be achieved during high-temperature creep in iron based alloys by the addition of gold as a solute healing agent. The creep lifetime at a temperature of 550 °C is found to be extended significantly by the introduction of about 1 at.% of solute Au in the ferritic Fe matrix, when the alloy has been homogenized at elevated temperature and subsequently quenched. Combined electron microscopy techniques demonstrate that the improved creep properties result from the selective Au precipitation at the early-stage creep cavities, preferentially formed on grain boundaries oriented perpendicular to the applied stress. The selective precipitation of gold atoms at the free surface of a creep cavity results in pore filling, and, thereby, self healing of the creep damage. The large difference in atomic size between the Au and Fe strongly reduces the nucleation of precipitates in the matrix. As a result, the matrix acts as a reservoir for supersaturated solute until damage occurs. Grain boundaries and dislocations act as fast routes for solute gold transport from the matrix to the creep damage. The efficiency to heal creep damage is found to depend strongly on the applied stress. For lower stress levels filling fractions of up to 80% have been observed for the open-volume creep damage.

This chapter summarises the main findings of the work already published in the open literature.

Ceramics have important advantages in application due to their high hardness and stiffness, high temperature stability and wear resistance. However, often the coefficient of friction is too high in sliding contacts, resulting in high frictional losses. Therefore ceramic composites containing a soft, self lubricating phase have potential to be used in these applications as the presence of the soft phase as a second phase in the ceramic gives the possibility of gradual supply to the surface and additional control over the thickness of the soft surface layer.

In this research, a copper oxide doped zirconia composite (CuO-TZP) is used as a self-lubricating ceramic composite system. The frictional behaviour of this material has been experimentally studied at several temperature levels. Further, a model has been developed which includes the processes responsible for maintaining the soft third body layer at the interface. The developed model includes a source flow and wear flow in balance for a stable thin soft layer in the contact. The model can predict the thickness of the third body layer under several tribological conditions in the mild wear regime as well as the operational conditions at which a stable thin soft layer is formed.

It can be concluded from the research that the coefficient of friction of TZP can be significantly reduced by the presence of a thin Cu-rich layer on the surface. The tribological performance of CuO-TZP under dry sliding conditions strongly depends on the operational conditions, like velocity, contact pressure and temperature. The model shows that the supply of soft phase to the surface is mainly determined by soft phase concentration as well as the applied load while the removal is mostly influenced by the microgeometry of the counter surface.

Crack damage in so-called MAX phase metallo-ceramics, like Ti₃AlC₂, Ti₂AlC and Cr₂AlC, can be healed by high temperature oxidation. Then the crack gap is filled with mainly Al₂O₃, which restores the integrity of the component made out of these MAX phases. This oxidation induced self-healing is an attractive mechanism, since the MAX phase itself is part of the healing reaction and the healing agent is formed with oxygen from the ambient air. Full strength recovery can be realized when the crack gap is fully filled with oxide. Also crack damage can be healed multiple times even when re-cracking occurs at the same location in the MAX phase material. Self-healing of these MAX phases together with their unique combination properties makes them a prospective material for operating under extreme conditions at high temperatures.

Sodium monofluorophosphate, chemically known as Na₂PO₃F and from here on referred to as Na-MFP, has long been known as a surface applied corrosion inhibitor in the concrete industry. Over the past decade, Na-MFP has gained large interest as an inorganic self-healing agent, particularly on blast furnace slag cement (BFSC) concrete. BFSC is an important product of the cement industry, especially in Northern European countries. In the Netherlands it holds a market share of more than 50% and besides many technical advantages it is an environmentally friendly product. However, its carbonation rate is a huge drawback compared to ordinary Portland cement (OPC) performances and requires a large-scale industrial and feasible solution to keep BFSC equally attractive for the building industry.

Numerous experimental and theoretical studies, embedded within the framework of the innovative research program of self-healing materials (IOP-SHM), have investigated the influence of Na-MFP as self-healing agent on BFSC systems upon carbonation. Results of these studies have shown that Na-MFP has the potential to “heal” the microstructure of carbonation-attacked cementitious material surfaces rich in BFS. Under the presence of water Na-MFP reacts with the slag-bearing carbonated concrete matrix to form amorphous calcium phosphates (ACPs), which increase the mechanical quality of the material surface, decrease the pore volume, the capillary water uptake and decelerate the carbonation rate. As a consequence damage through micro coarsening of pore structure and related cracking can effectively be compensated and the frost salt scaling durability of the BFSC products can significantly be increased.

This chapter summarizes research results generated over the past couple of years and gives an outlook towards the development of an autogenous inorganic self-healing mechanism induced by Na-MFP in BFSC systems.

Damage due to salt crystallization is a ubiquitous problem in porous building materials. Crystallizing salts can cause severe damage to building materials. Especially lime-based mortars, which are often found in historic buildings, are prone to salt damage due to their pore size distribution and limited mechanical strength. Renovation costs for replacing or repairing materials affected by salt crystallization damage are considerable. Existing solutions, such as cement-based salt-resistant plasters with mixed-in water-repellent additives, generally have a low compatibility with the existing materials. Where the current methods are based on increasing the strength of the material or on limiting the ingress of (salt-laden) water into the porous material, our method is to tackle the problem at its base, by modifying the crystallization of the salts. By influencing the crystallization process with special chemicals, so called crystallization modifiers, salt crystallization on the surface of the material instead of in the pores or at a lower supersaturation level is favoured. Both are factors which limit crystallization damage. By mixing-in the crystallization modifiers in a lime-based mortar before its application, a smart self-healing (restoration) product can be made which will act at the moment the salts are expected to cause damage, i.e. at the onset of crystallization. Our proposed method could therefore result in a longer service-life of buildings and, consequently, in lower restoration costs.

In a sequential number of IOP funded research projects the suitability of very specific but otherwise harmless bacteria are tested for their ability to repair cracks significantly improving the durability of concrete structures. Such a bacterial repair mechanism would be beneficial for the economy and the environment at the same time, as concrete is worldwide the most applied building material. This new type of ‘bio-concrete’ or ‘bacteria-based concrete’ could make costly manual repair unnecessary resulting in minimizing the use of raw materials, as structures will last much longer. In nature a huge number of different varieties of bacteria occur and some of these are well adapted to artificial man-made environments such as concrete. From a human perspective concrete may seem an extreme environment as the material is dry and rock-solid. However, this does not apply to a specialized group of bacteria, the ‘extremophiles’, named after their habit to love extreme conditions. Some of these bacterial species are not only known to love extremely dry conditions, but also to be able to produce copious amounts of limestone particularly under alkaline conditions. Limestone is from a physico-chemical viewpoint a concrete compatible material and therefore ideally suitable for durable repair of cracks in concrete. In a series of TU Delft based research projects three types of bacteria-based self-healing concrete products have been developed: 1. Self-healing concrete, 2. Self-healing repair mortar, and 3. A spray-able liquid repair system. Whereas for the first two products a granular ‘healing agent’ in form of encapsulated bacterial spores and feed is required, encapsulation of healing agent ingredients is not needed in the last product. While the focus in the early stage of the research projects was particularly on isolation of suitable bacteria and development of various forms of healing agents, the emphasis in the later stages of the projects was on practical outdoors applications. Full scale applications in collaboration with stakeholder parties involving casting of self-healing concrete irrigation canal elements in Ecuador, patch repair for elimination of leaking cracks of concrete walls using self-healing repair mortar, and reduction of frost damage sensitivity and sealing of cracks in concrete parking decks by application of bacteria-based liquid repair system, proofed the functionality and market potential of these products.

The paper deals with several modeling approaches that are relevant for understanding the self-healing mechanisms in a class of civil engineering materials like blast furnace slag cement and concrete. Since the present paper is a compilation of modeling studies carried out by the authors, it does not aim at being mathematically rigorous but it is merely descriptive of nature. The approaches are used for the prediction of the likelihood for the onset of the self-healing mechanism, as well as for the prediction of the rate and capacity of the self-healing process. Although the approaches are applied to civil engineering materials, the mathematical nature of the treatment allows an easy generalization to other classes of materials.

Porous asphalt shows excellent performance in both noise reduction and water drainage. Although porous asphalt has these great qualities, its service life is much shorter (sometimes only half) compared to dense graded asphalt roads. Ravelling, which is the loss of aggregate particles from the surface layer, is the main damage mechanism of porous asphalt surface wearing courses. In this research, an induction healing approach (namely, activating the healing process of asphalt concrete through induction heating) was developed to enhance the durability of the porous asphalt roads. Steel fibres are added to a porous asphalt mixture to make it electrically conductive and suitable for induction heating. When micro cracks are expected to occur in the asphalt mastic of the pavement, the temperature of the mastic can be increased locally by induction heating of the steel fibres so that porous asphalt concrete can repair itself and close the cracks through the high temperature healing of the bitumen (diffusion and flow). The closure of micro cracks will prevent the formation of macro cracks. In such a way, ravelling can be avoided or delayed in the end.

To test the induction healing technology in a real porous asphalt road, a trial section was constructed on Dutch motorway A58 in December 2010. This trial section survived already five winters and is still in perfect shape. Experiments were done on cores drilled from the trial section and the results coincided with those on the laboratory made samples. A first heat treatment of the section with a large scale induction machine was performed in June 2014. Based on the laboratory experiments and field experiences it can be concluded that induction healing is a very good approach to enhance the durability of porous asphalt pavement. It is expected that the service life of porous asphalt can be doubled and will be similar to the service life of dense asphalt.

Asphalt concrete is one of the most common types of pavement surface materials used in the world. It is a porous material made at very high temperatures (~180 °C) that consists of a mixture of asphalt binder (bitumen), aggregate particles and air voids. After some years of use, the stiffness of asphalt concrete increases, its relaxation capacity decreases, the binder becomes more brittle, micro-cracks develop in it and cracking of the interface between aggregates and binder occurs. This mechanism is especially detrimental in porous asphalt where it leads to ravelling. Ravelling, which is the loss of aggregate particles from the surface layer, is the main damage mechanism of porous asphalt surface wearing courses. In the chapter “Unravelling of porous asphalt” in this book it is discussed that ravelling can be avoided or delayed by mixing steel wool fragments through the asphalt. These fragments make it possible to heat the asphalt with induction energy and by that close micro-cracks. In this project another approach is followed: the use of rejuvenators. These rejuvenators are encapsulated and mixed through the asphalt. These capsules break due to the constant fatigue loads and to the higher stiffness of the binder that happens when it oxidizes. Once the capsules are broken, a permanent deformation happens in the capsules and if the fatigue loads of traffic continue, they suffer a permanent deformation and the rejuvenator is released and softens the binder and makes it more flexible again. In the project the capsules are optimized and the process of rejuvenation is studied.

In the Netherlands, 4 million tons of reclaimed asphalt (RA) were produced per year according to the European Asphalt Pavement Association (EAPA). Over 90% of the total reclaimed asphalt was reused into new asphalt constructions. However, for the most used new porous asphalt surface layer on the Dutch motorway system, the allowable RA percentage is only 25%. With the increasing needs of high percentage (up to 100%) and high performance (surface-to-surface) recycling due to both economic and environmental issues, it is of high importance to investigate the possibility to recycle the high quality surface layer (porous asphalt, stone mastic asphalt, etc.) RA back to the surface layers. The ultimate goal is to realize a 100% surface-to-surface asphalt recycling. In this research, a high performance material system (namely, rejuvenator) was developed to regenerate the flexibility and self-healing capability into the RA, which are the most important elements guaranteeing the long service life of asphalt mixtures. Based on the knowledge on the chemical/mechanical responses on the rejuvenated and re-aged bituminous materials, together with the in-situ diffusion analysis with Computer Tomography (CT) scan and Fourier Transform Infrared Spectroscopy (FTIR) measurements, a rejuvenator with excellent rejuvenating capability was developed. To evaluate the rejuvenating capability of the developed rejuvenator in a real application, a trial production was performed with a highly ecologic asphalt recycling (HERA) system by our project partners with high RA percentages, and a trial section was then constructed in 2014 with success. Experiments carried out on the cores drilled from the section indicated good performance. Now it is important to monitor the test section over time in order to gather more useful real service data. Based on the laboratory and field trial section results, it is expected that the developed rejuvenating product can guarantee a high performance surface-to-surface recycling.