Comments

|

Comments This Article

CONCRETE: A challenge for modeling complexity

Dr. Klaas van Breugel

Abstract

For understanding concrete, it should be considered a system rather than a material. Looking at concrete as a system helps us to understand its complexity and its response to thermal and hygral loadings and opens the way for modifying existing concrete mixtures and designing new ones with predefined properties. Designing new concretes requires fundamental knowledge of the material’s performance at different scales and insight in different fields of science. Thus, a multiscale and multidisciplinary approach is believed to be a prerequisite for innovation in materials research, for understanding ageing of cement-based systems as well as their potential of self-healing. This paper concentrates on the potential and challenges for modelling and models to link different scales and disciplines. This will be illustrated with a few examples, i.e. fracture processes in cementitious materials, self-healing of concrete and the use of smart nanoparticles for mitigating the risk of rebar corrosion.

Introduction

Concrete is a complex material. A short review of the mechanical properties of concrete already reveals that the compressive strength of concrete depends on more than 25 parameters [1]. The dependency of concrete properties on so many parameters complicates the characterization of this material and the interpretation of test results. This holds for the mechanical properties, but even more for the time-dependent properties, like creep and relaxation [2, 3, and 4]. There is no doubt: Concrete is an extremely complex material on every scale of observation. Given the complexity of cement-based materials, one could wonder whether these materials should be qualified a material at all. In fact it is an artificial construct rather than a material. According to Kahn, an architect, concrete is a product of the mind (in [5]). As such concrete in not different from, for example, a skyscraper. The deflection of the skyscraper under wind load cannot be inferred from overall key characteristics of the building, like the mass per cubic meter, but can only be understood and predicted if we know the properties of the individual components from which the building was built and how these components are connected. Similarly, the performance of concrete under any type of loading cannot be inferred from its specific mass, but requires knowledge of the properties of its constituents and how these constituents are connected and work together in carrying applied loads.

Architecture

Looking at materials as a product of the mind is, in essence, not new. Philosophers have been dealing with the real world around them for many centuries, always trying to get answers on fundamental “why” questions. The background of the search for these answers has always been the hope that one day, after they had understood the nature of matter, they would be able to reshape reality according to their own wishes. As early as the 4th century BC the Creek philosopher Democritus assumed that matter consisted of very small inseparable entities, called atoms. These atoms were the basic building blocks of Democritus’ virtual world. The relevance of this world view for the daily practice was not very big in those days, if relevant at all. For solving practical problems not philosophers, but architects were consulted. What architects actually did was creating a virtual construct of well-defined components of which they knew, or assumed, their performance. In this respect it is worthwhile to remember the original definition of architecture. “Architecture is the art and science of designing and constructing buildings and other physical structures for human shelter or use”. A wider definition includes the design of the total built environment, from the macro level of how a building integrates in its surrounding context, to the micro level of architectural or construction details and of manufacturing at all conceivable scales.

a) Metropole (USA)

From these definitions we learn that architecture is an almost holistic concept. It aims at serving mankind by linking art and science. This combination of art and science is a prerequisite for innovation! It is the basis for designing all kind of systems and products for the benefit of the society and the environment, including the design of materials.

b) Apartment Block Montreal (Cn)

C) Apartment block Rotterdam (NL) , d) Sherical houses (NL)

Designing materials

System thinking

Like a skyscraper, also concrete can be considered as a system, made of basic building blocks. A system is, by definition, a set of interacting or interdependent components forming an integrated whole. The components can be defined at subsequent scales, ranging from the atomic scale to the macroscale. At all these subsequent scales systems exhibit a few common characteristics:

• Systems have structure;
• Systems demonstrate a certain behaviour;
• Systems have interconnectivity, for example of the solid phase or the pore phase;
• Systems exhibit certain functions, or sets of functions.

On top of these common characteristics a system may also change with elapse of time. This is typical for cement based systems in the early and later hydration phase and in the subsequent stages of the concrete’s lifetime.

Designing with Basic Building Blocks at different length scales

At the sup-macroscale town planners place their basic building blocks, i.e. skyscrapers and apartment blocks, in a rectangular, circular or free mesh (Fig. 1a). Streets and canals, as well as sewage systems and electrical cable networks constitute vital connective systems for proper functioning of a city. The town planner’s basic building blocks are designed by architects. For architects cubic boxes act as basic building blocks. In order to create specific aesthetic effects these cubes can be placed a bit shifted relative to each other (Fig. 1b) or rotated around the point of centroid (Fig. 1c). Alternatively, spherical basic building blocks might be considered (Fig. 1d). These examples illustrate that architecture is not just piling up of cubes of sphere, but t is a special kind of piling up.

Whereas metro poles, apartment blocks and houses are products of the mind at the macroscopic level ‘in foresight’, many materials are products of the mind at the micro- or nanostructure level, or even at the atomic and molecular level, ‘in hindsight’. The architecture of raw materials has been discovered after decades, or even ages of research. The more we become to know about the secrets of materials, the more we realize that we have just become able to intervene in the microstructure of materials in order to modify them and/or to create new materials with predefined properties. An early example of architecture of metals has been presented by Rosenhain in 1915 already [6]. Metal crystals were assumed to consist of clusters of cubic basic building blocks (Fig. 2a,b). The concept of cubic basic building blocks is found back in the well-known NIST model, developed in the eighties for simulating the evolution of the microstructure and hardening cement based systems (Fig. 2c) [7]. Virtual microstructures built up of spherical basic building blocks are created by models proposed by Navi et al. [8], Ishida et al. [9], van Breugel and Ye [10,11] (Fig. 2d) and Nothnagel [12]. In the latter micro structural models the cement grains are considered as spherical grains, which gradually grow with progress of the hydration process. The formation of inter particle contact will affect the rate of dissolution/reaction of the anhydrous cement grains. Hydration models, which explicitly allow for the effect of the formation of inter particle contacts on the rate of hydration, have been denoted integrated kinetic models [10].

a) Titanium alloy [6], b) Cartoon of simulated , c) Simulated microstructure , d) virtual microstructure

Titanium crystal structure [6] of cement paste [7] of cement paste [11]

Fig. 2: a) Microstructure of titanium alloy [6]; b) conceptual titanium crystal growth [6]; c) cement paste, NIST model [7]; d) cement paste simulated with HYMOSTRUC [11] 6

The growth of spherical particles and their mutual contacts is caused by the formation of reaction products, i.e. ettringite, calcium hydroxide (CH) and calcium silicate hydrates (CSH). The properties of the reaction products and the type of bonds between them are decisive for the performance of these virtual microstructures. After decades of studying the nature of reaction products and of conceptual modelling [13,14,15], researchers are now able to perform abinitio calculations of these products [16,17].

a. Virtual Microstructure , b. Contact Area , c. Strength vs Contact Area

Fig. 3: Concept of interparticle contact area (3b) and relationship between calculated contact area of a virtual microstructure (3a) and strength (3c) [19].

Bio-based self-healing of cracks

Mechanisms of bio-based healing

One of the preconditions of self-healing of cracks in concrete is the transport of matter to the crack. In a living organism transport of ingredients takes place via a vascular system. In plants and trees ingredients are transported via a network of pores. A porous material like concrete also has a pore system through which transport processes are possible, but a driving force that makes transport of healing species happening is not inherently present. Temperature, moisture, pressure or concentration gradients are needed to trigger transport of species. But still then it is not easy to transport sufficient matter to the spot, i.e. crack, where healing is needed. Promising in this respect is the concept of bacteria-based healing of cracks [32]. The idea is that after cracking mixed-in bacteria on fresh concrete crack surfaces are activated in the presence of water, and then start to multiply and precipitate minerals, such as calcium carbonate, and close the crack. The healing mechanism is presented schematically in Figure 7. Although conceptually the mechanisms are simple, a variety of problems have to be tackled, both from the science point of view and also from the perspective of modelling.

 

a. Virtual Microstructure , b. Delft Lattice Model , c. Lattice System in Tension

Fig. 4: Virtual 3D microstructure (left) as basis for generating a lattice mesh (right) for Delft Lattice Model analyses. Sample 100x100x100 μm3, loaded in tension [20,21].

Bacteria in high pH

A precondition for successful bacteria-based healing of cracks is that the bacteria can survive the high pH in the concrete of about 12 to 13. It was found that from a microbiological viewpoint the application of bacteria in concrete, or concrete as a habitat for specialized bacteria, is not odd at al. Although the concrete matrix may seem at first inhospitable for life because of its high alkalinity, comparable natural systems are known in which bacteria do thrive, even in a very dry environment. Inside rocks, even at a depth of more than 1 km within the earth crust, in deserts as well as in ultra-basic environments, active bacteria are found (see [31,32]). These desiccation- and/or alkali-resistant bacteria typically form spores, i.e. specialized cells able to resist high mechanically- and chemically induced stresses [33]. These spores have extremely long lifetimes: spores are known to be viable for up to 200 years [34].

Fig. 5: Cracking pattern in virtual microstructure of cement paste. w/c = 0.4. Degree of hydration is 44%, 69% and 90%, respectively. Samples 100x100x100 μm3 [20].

The bacteria to be used as self-healing agent in concrete should be fit for the job, i.e. they should be able to perform long-term effective crack sealing, preferably during the total life time of a structure. The principle mechanism of bacterial crack-healing is that the bacteria themselves act largely as a catalyst, and transform a precursor compound to a suitable filler material. The newly produced compounds, such as calcium carbonate-based mineral precipitates, than act as a type of bio-cement that seals or heals newly formed cracks. Thus for effective self healing, both bacteria and a bio-cement precursor compound should be integrated in the material matrix.

 

Fig. 6: Load-displacement diagram of cement paste samples loaded in tension [20].

Jonkers et al [35] found that when bacterial spores were directly added to the concrete mixture, their lifetime was limited to one to two months. The decrease in life-time of the bacterial spores from several decades when in dry state to only a few months when embedded in the concrete matrix was attributed to continuing cement hydration, resulting in matrix pore-diameter widths typically much smaller than the 1-μm sized bacterial spores.

Microstructure and materials properties

Evolution of strength of cement paste

The aforementioned integrated kinetic models have in common that they can generate a virtual microstructure. This microstructure is a system of (partly) hydrated cement particles connected by reaction products. The intensity with which the cement particles are connected has been assumed to correlate with the mechanical properties of the system. Fig. 3a shows a virtual microstructure of cement paste generated with the simulation model HYMOSTRUC3D. From this virtual microstructure the contact area between hydrating cement particles can be calculated (Fig. 3b). The size of these contact areas is supposed to reflect the intensity of the inter particle bonds. Fig. 3c shows how the calculated inter particle contact area correlates with the measured compressive strength of cement pastes. Results are presented for three cement pastes, made with w/c 0.35, 0.5 and 0.6, respectively [18,19]. Fig. 3c suggests that the observed correlation is independent of the water-cement ratio. Hence it seems that the hypothesis, that the contact area correlates with the mechanical properties, makes sense.

Simulation of cracking in an axially loaded virtual microstructure

Cement paste, while being formed, will exhibit hydration induced deformations. In mortar and concrete these

deformations will be restrained by aggregated particles. Hence, stresses will be generated in the cement paste and cracking may occur. Besides shrinkage-induced stresses, cement paste may also become stressed and cracked due to external mechanical loads. How cracks develop in a virtual microstructure has been simulated by using the Delft Lattice Model (DLM) [20]. This model was developed in the nineties for simulating fracture processes in heterogeneous systems like mortars and concrete at the mesoscale. Cement paste itself, however, is also a heterogeneous material, consisting of hydration products (gel) and unhydrated cement cores. Qian [21] has used the concept of the Delft Lattice Model for analyzing stress patterns and cracking of a virtual microstructure at the microscale. The cement grains were connected by beam elements, which form a three-dimensional network (Fig. 4a,b). The mechanical properties of the elements are a function of the size of the contact areas with which cement particles are connected. The virtual microstructure is then loaded in tension in a strain controlled virtual test (Fig.4c). If the stress in one of the lattices exceeds its strength this element will be “removed” from the system and a new equilibrium is established in a numerical iteration process.

With this model the effects of a variety of mixture parameters have been analyzed, like the fineness of the cement, the degree of hydration and the water-cement ratio [20,21]. In Fig. 5 the crack patterns are shown of a

cement paste sample made with a cement with Blaine value 420 m2/kg and water-cement ratio 0.4, loaded in

tension (deformation controlled) at a degree of hydration of 44%, 69% and 90%, respectively. The corresponding load-deformation diagram is shown in Fig. 6. At lower degree of hydration the cracks are spread throughout the volume of the paste, whereas at a higher degree of hydration cracking is more localized. Because in the latter case fewer micro cracks are formed the system responds more brittle. The more brittle response can also be inferred from the shape of the load-displacement curves in Fig. 6. Similar results were found in experimental studies of tensile properties of young age concrete by Dao et al [22]. Relatively high values of the fracture energy obtained for mixtures at early ages were suggested to reveal that cracking of young concrete involves a significant zone of plastic straining or micro cracking in the vicinity of crack tips. If this is correct, this phenomenon should be considered in the interpretation of experimental creep data for young concrete.

Modeling 'beyond cracking'

Damage prevention versus damage control

Concrete is a brittle material with microcracks throughout the body of material already prior to application of any external load. In concrete technology these cracks are not considered damage and are acceptable as long as a prevailing crack width criterion is not exceeded. The fact that in structural concrete cracks are acceptable does not mean that they are desirable. Cracks may jeopardize the resistance of concrete against ingress of harmful substances. The concrete may then deteriorate fast and the steel reinforcement is no longer adequately protected against corrosion. Cracks may also be undesirable from the functionality and aesthetic point of view. So, even though cracks are not problematic from the safety point for view, they are undesirable from the overall performance point of view. It would be good if cracks, if considered unavoidable because of the inherent brittleness of cement-based materials, could be healed by a built-in self-healing mechanism. Damage control should then not be restricted to control of the crack width – the common practice in concrete design -, but should also focus on healing of the cracks once they occur.

Autogenous self-healing and self-sealing

In most of the traditional concrete mixtures 20-30% of the cement is left un hydrated. The amount of un reacted cement is higher the coarser the cement and the lower the water/cement ratio of the mixture. If cracking of the concrete occurs, un reacted cement grains may become exposed to moisture penetrating the crack. In that case the hydration process may continue and hydration products may fill up and heal the crack. This inherent self-healing mechanism is known since long and known as autogenous healing. This inherent self-healing capacity of concrete makes it a material with a lot of ‘forgiveness’. Since long mankind has taken advantage of this peculiar property of concrete, even though it has never been designed on purpose to be a self-healing material. Neither the self-healing process itself, nor the required preconditions for making this process to happen are completely understood today. As a consequence the self-healing capacity of cement-based systems is considered a positive feature of concrete indeed, but too unreliable yet to take into account explicitly in the design of concrete structures. Only a few exceptions are known where designers explicitly count on self-healing of cracks, for example in the design of watertight cellars or reservoirs made of reinforced concrete [23].

Fig. 7: Scenario of crack-healing by concrete-immobilized bacteria. Bacteria on fresh crack surfaces become activated due to water ingression, start to multiply and precipitate minerals such as calcite (CaCO3) (after Jonkers [32])

Preconditions for the occurrence of self-healing in ordinary concretes are, apart from the presence of unhydrated cement and moisture, a limited crack width [24]. The smaller the cracks are, the higher the probability that the cracks will heal. A cement-based product that is designed for a small crack width is ECC (Engineered Cementitious Composites) developed by Li et al [25]. The main purpose for designing ECC was to make a ductile material that is able to make large excursions in the post-cracking phase. The use of small fibres ensures that on cracking the crack width remains very small, typically 50 μm. Because of this small crack width, ECC has a remarkable self-healing capacity, even recovery of the original strength after healing.

Autonomous self-healing of concrete

 

Fig. 8: Self healing admixture composed of expanded clay particles (left) loaded with bacterial spores and organic bio-mineral precursor compound (calcium lactate) [36].

Whereas autogenous self-healing can be considered an inherent feature of cement-based systems, autonomous self-healing is defined as a man-made self-healing mechanism. An often mentioned way to realize autonomous self-healing is by dispersing capsules containing either a cementitious or synthetic healing agent in the mixture. This encapsulation concept has been proposed by White et al [26] in 2001 for self-healing polymers and by Dry [27,28,29] in the early nineties of the past century for self healing concrete. On cracking the capsules may rupture, while releasing the healing agent into the crack. This is the most common concept, but not often used in concrete yet. Since concrete is a heterogeneous system by definition, adding capsules to the mixture does not significantly change the nature of the material. However, many modelling challenges are ahead, for example: what is the most suitable size, ll shape and number of capsules so as to ensure that a crack will pass through a capsule;

• The amount of healing agent required to fill a crack;
• The type of healing agent;
• The transport of the healing substance to ensure homogenous filling of the crack.
• The effect of the presence of capsules on the short- and long-term mechanical properties

Regarding the optimal shape of capsules, numerical simulations have revealed that capsules with a large aspect ratio, with cylindrical tubes as the extreme, are more effective than spherical capsules. Huang et al [30] studied that rate of filling of micro cracks by progressive hydration in a paste in which 30% of the cement was left un hydrated at the moment of cracking. Preliminary numerical simulations revealed that the self-healing efficiency of 10 μm wide water-saturated micro cracks after about 100 hours is over 40%, i.e. 40% of the crack was found to be filled with new reaction products.

Encapsulation of bacteria

With the aim to increase the lifetime and associated functionality of bacteria in the concrete, the effect of bacterial spores and simultaneously needed organic bio mineral precursor compound (calcium lactate), Jonkers

[36] tested encapsulation of these components in porous expanded clay particles (Figure 8). It was found that protection of the bacterial spores by encapsulating them in such particles before addition to the concrete mixture indeed substantially prolonged their life-time. After 6 months incorporation in concrete no loss of viability of the spores was observed, suggesting that their long-term viability as observed in dried state when not embedded in concrete is maintained.

Evidence of bacterial self-healing

Fig. 9: Light microscopic images of pre-cracked control (A) and bacterial (B) concrete specimen before (left) and after (right) healing.Healing after 2 weeks submersion in water.

In order to test the bacterial healing of cracks in concrete test specimens were prepared in which part of the dense aggregate was replaced by similarly sized expanded clay particles loaded with the biochemical self-healing agent (bacterial spores 1.7x105 g-1 expanded clay particles, corresponding to 5x107 spores dm-3 concrete, plus 5% w/w fraction calcium lactate, corresponding to 15g dm-3 concrete). The amount of lightweight aggregate represents 50% of the total aggregate volume. Control specimens had a similar aggregate composition, but these expanded clay particles were not loaded with the bio-chemical agent.

The self-healing capacity of pre-cracked concrete disks (10 cm diameter, 1.5 cm thickness), sawn from 56 days

water cured concrete cylinders, was tested by measuring the evolution of water transport through the disks and by taking light microscopic images before and after the permeability test. For determination of the permeability pre-cracked concrete disks were glued in an aluminium ring and mounted in a permeability setup. The generated crack width was 0.15 mm, running completely through the specimen. After cracking both sets (6 of each) of control and bacterial concrete specimens were submerged for two weeks in tap water at room temperature. Subsequently, permeability of all cracked specimens was quantified by recording of tap water percolation during a 24 hours period.

Comparison between bacterial and control specimens revealed a significant difference in permeability. While cracks of all six bacterial specimens were completely sealed, resulting in no measurable permeability (percolation of 0 ml water/h), only 2 out of six control specimens appeared perfectly healed. The four other control specimens featured permeability (water percolation) values between 0 and 2 ml/h. Microscopic examination of cracks at the water-exposed side of the slab revealed that in both control and bacterial specimen precipitation of calcium carbonate-based mineral precipitates occurred. However, in the control specimens precipitation largely occurred near the crack rim, leaving major parts of the crack unhealed, whereas efficient and complete healing of cracks occurred in bacterial specimen with mineral precipitation predominantly within the crack (Figure 9B). The most obvious reason for massive white precipitation of calcium carbonate near the crack rim of the control specimen (Figure 9A) is that concentration of both reactants calcium hydroxide and carbon dioxide are relatively high due to the opposing diffusion gradients of the respective reactants [35]. Calcium hydroxide diffuses away from the crack interior towards the overlying bulk water while carbon dioxide diffuses from the bulk water towards the crack interior where it is scavenged by high concentrations of calcium hydroxide.

Modelling of bio-based self-healing

There is no doubt that modelling of bio-based self healing processes is even more challenging than that of ‘traditional’ autogenous self-healing processes. It requires of knowledge of both inorganic and organic systems: a real multidisciplinary approach. Moreover, change of mind set is required, since a symbiosis of building materials and bacteria is now considered beneficial instead of detrimental for the concrete.

Smart nanoparticles for mitigating the risk of corrosion

The self-healing concept

Fig. 11: Effect of nanoparticles, micelles and vesicles,onwater permeability [42,43]

The (self-) healing of cracks and densification of the macrostructure cannot completely prevent reinforced concrete structures from the risk of rebar corrosion. Therefore, there still remains a demand for innovative concepts to mitigate the risk of corrosion. Recently the effect of nanoparticles on the resistance against corrosion has been investigated. Koleva et al [37,38,39] and Hu etal [40,41] observed significant micro structural changes after adding so-called micelles to plain mortar. These micelles are nano-size particles, about 50 nm in diameter, prepared from poly-ethylene oxide di-block polystyrene (PEO-b-PS) (Fig. 10). Even with a low concentration of micelles, 0.025 % by weight of dry cement, in a mortar with cement-to-sand ratio 1:3 and water-to-cement ratio 0.5, the porosity of both the bulk matrix and a steel-matrix interfacial transition zone (ITZ) decreased significantly [40]. The coefficient of water permeability was 3 orders of magnitude lower for the micelles-containing specimen compared to the micelles-free mortar [42,43] (see Fig. 11). This increase in density of the cement paste substantially contributes to the resistance against corrosion of the embedded steel.

A next step to even further mitigate the risk of rebar corrosion could be the use of smart nano particles, which have the potential to react to changes of the chemical environment by a sort of self-healing mechanism. Koleva and her co-workers [44,45] invested the effect adding PEO113-b-PS780 vesicles, which are particles similar to micelles but carrying an “active” compound, on the rate of corrosion of steel bars places in a simulated pore solution. The active compound was CaO. The hypothesis is that in the event of an aggressive external influence, i.e. carbon dioxide penetrating the material thus carbonating the matrix or Cl- penetration followed by localized corrosion on the steel surface, the “charged” vesicles will participate in a self-healing mechanism by releasing their core material. The released core material, i.e. CaO, will restore the alkalinity in the bulk matrix and repair the passive layer on the steel surface. Koleva et al [44] found that, in line with what was expected, the surface of a steel rebar placed in pore solutions with and without loaded vesicles exhibited a significant difference, even for a very low concentration of ”loaded” particles (4.9 10-4 g/l). When NaCl was added to the solution, as corrosion accelerator, the steel in vesicles-containing pore solution exhibited again superior performance compared to the vesicle-free solution. ESEM observation of the product layers on the steel surface after 7 days of conditioning revealed a more homogeneous and compact protective layer on the steel surface of the specimens conditioned in vesicle-containing solution.

Modeling aspects of nano-inspired reduction of corrosion activity

 

Fig. 10: Formation of frozen core-shell micelles from PEO113-b-PS70 di-block copolymer in aqueous media (Koleva et al [39])

A comprehensive model for corrosion of reinforcing bars in concrete is still missing. Adding smart nanoparticles toa concrete mixture in order to mitigate the risk of corrosion will not bring us closer to such a comprehensive model. However, the presented promising preliminary results of the experimental studies do constitute a challenge for both conceptual and numerical modelling in order to further optimize these smart mixtures and to reliably predict the long-term performance of these modified mixtures.

Challenges for modelling: Concluding remarks

Understanding concrete is a matter of understanding its details. This concerns the properties of its components

and how these components act together is a complex system. The better we understand how this system works, the more opportunities for innovative cement based systems will emerge. As illustrated in this paper, a prerequisite for innovation is that the required research is multiscale and multidisciplinary. This means that researchers have to cross borders! Very often these border-crossing initiatives will not immediately lead to spectacular innovations. A certain incubation period will be needed before unprecedented concepts are mature

and convincing enough to receive acceptance from a wider group of people. But with elapse of time more examples will emerge and demonstrate that innovation really benefits from border-crossing activities. In the early eighties many people could hardly believe that predicting stresses in hardening concrete would be possible with microstructure-based models. Today these models have found their way to structural designers and to building sites, where concrete curing control systems are actually using these models. The same holds for the development of microstructure-based models for evaluating transport and degradation processes in cement-based systems [46]. Even though these models still need further improvement, they have already substantially contributed to both the research community and the building industry.

Modelling is the vehicle for border-crossing transfer of knowledge and expertise. This modelling has once been

denoted the Alchemy of cement chemistry [47]. Alchemy was used there as the metaphor for the tedious type of labour people are only prepared to do as long as they believe that in the end their effort will pay off: gold! We know, however, that the alchemists did never reach that goal. But, as it has been remarked by Dijksterhuis [48], in the end these alchemists gave society something that was worth many times more than gold, viz. the modern chemistry, including lots of innovations! Similarly it might well be that the benefits of modelling activities will finally consist of lots of innovations beyond our imagination yet. But even today clearly defined fields of interest are known, where multiscale and multidisciplinary models are believed to be a precondition for new developments. These fields are:

• Materials design, for example self-ll healing materials;
• Virtual testing of material and structures;
• Service life predictions;
• The use of smart sensors for monitoring of structures;
• Studies for mitigating the ecological footprint of materials use and of the entire building process.
• Economic optimizations.

It has to be mentioned that the involvement of many disciplines, the prerequisite for success, is also the weak point in the whole story. Often researchers are primarily focused on short-term achievements within their own discipline. This attitude certainly constitutes one of the most serious constraints for innovation. Border-crossing research requires investment of time and energy in fields where researchers may feel uncomfortable. But without the vision of parties involved in innovation processes, and without their willingness to cooperate across the borders of their own discipline, badly needed innovations and new developments will remain hard to accomplish, if not an illusion.

Acknowledgement

The author wishes to express his thanks to Dr. D.A. Koleva, Prof. Dr. E. Schlangen, Dr. Ye Guang, Dr. Z. Qian, Dr. H.M. Jonkers and Dr. V. Wiktor for their contribution to this paper.

References

1. Breugel, K. van. "Concrete: A materials that barely deserves that qualification". Proc. 76. Int. RILEM Week, Aachen, Vol. 11, 2010, pp. 13-32
2. Kaman, A., Denarie, E., Sadouki, H., Bruhwiler, E., "UHPFRC tensile creep at early age". Materials and Structures, Vol. 42, No.1, 2009, 113-122.
3. Bazant, Z.P., Hauggaard, A.B., Baweja, S., "Microprestress - Solidification theory for aging and drying creep of concrete". In: Advances in Building Materials Science, Aadificatio Publishers, 2996, pp. 111-130
4. Meakawa, K., "Simulation of long-term time dependent mechanics of an existing PC bridge by full 3D multi-scale integrated FE analysis". Proc. CONMOD’10, 2010. 23
5. Mc Carter, R., Louis I. Kahn, Phaidon Press, London, UK. 2005, 512 p.
6. Rosenhain, W., "An introduction of the study of physical metallurgy", 2nd Edition, Constable 7 Company Ltd., London. 1915.
7. Bentz, D.P., "Three-dimensional computer simulation of Portland cement hydration and microstructure development", J. Am. Ceram. Soc., Vol. 80, 1, 1997, 3-21.
8. Navi, P., Pignat, C., "Simulation of effects of small inert grains on cement hydration and its contact surfaces", in 'The modelling of microstructure and its potential for studying transport properties and durability", Eds. Jennings et al., NATO ASI Series E, 304, 1996, 227-241.
9. Ishida, T., Mabrouk, R.T.S., Maekawa K., "An integrated computational framework for performance evaluation of cementitious materials and structures under various environmental actions". Proc. 6th Int. Conf. 'Creep, shrinkage and durability me-chanics of concrete and other quasi-brittle materials’. Ed. Ulm et.al., 2001, 511-516.
10. Breugel, K. van, "Simulation of hydration and formation of structure in hardening cement-based materials". PhD Thesis, Delft University of Technology, 1991, 295 p.
11. Ye, G., "The Microstructure and Permeability of Cementitious Materials". PhD Thesis, Delft University Press, Delft, 2003, 186 p.
12. Nothnagel, R., "Hydrations- und Strukturmodell fur Zementstein". TU Braunschweig. PhD Thesis. 2007.
13. Feldmann, R.F., Sereda, P.J., "A model for hydrated Portland cement paste as deduced from sorption-length change and mechanical properties". Materials & Structures, No. 6, 1968, pp. 509-519.
14. Wittmann, F.H., Properties of hardened cement paste, Proc. 7th Int. Conf. on the Chemistry of Cement, Sub-theme VI-2. 1980. 24
15. Jennings, H.M., Colloid model of C-S-H and implications to the problem of creep and shrinkage, Materials & Structures, Vol. 37, No. 265, 2004, pp. 59-70.
16. Dolado, J.S., Campillo, I., de Miguel, Y., Erkizia, E., Porro, A., Ayuela, A., Sanchez-Portal, D. "Ab-initio modelling as a tool for understanding C-S-H gel". In Y.R. de Miguel et al (eds) Proc. Sec. Int. Symposium on Nanotechnology in Construction, Bilbao, 2005, pp. 67-72.
17. Pellenq, R.J.M., Kushima, A., Shahsavari, R., van Vliet, K.J., Buehler, M.J., Yip, S., Ulm, F.J., « A realistic molecular model of cement hydrates". Proc. Nat. Academy of Sciences of the USA, Ed. Z.P. Bazant. 2009.
18. Sun, Z.H, Ye, G., Voigt, T., Shah, S.P., Breugel, K. van, "Early age properties of Portland cement pastes investigated with ultrasonic shear waves and numerical simulation". In (Ed.), Advances in concrete through science and engineering, Evanston, Illinois, USA, 2004, pp. 1-13.
19. Voigt, T., Ye, G., Sun, Z., Shah, S.P., Breugel, K. van, "Early age microstructure of Portland cement mortar investigated by ultrasonic shear waves and numerical simulation". Cement and Concrete Research, Vol. 35, No. 5, 2005, pp. 858-866.
20. Schlangen, E., Qian, Z, Ye, G., Breugel, K. van, "Materials research in progress: Simulation of cracking (in Dutch)". CEMENT, (Accepted for publication), No. 3, 2010.
21. Qian, Z., "3-D analysis of cement paste”. MSc Thesis, Delft University of Technology, 2008, 151 p.
22. Dao, V.T.N., Dux, P.F., Morris, P.H., “Tensile properties of early age concrete". ACI Materials Journal, Vol. 106, No. 6, 2009, pp. 483- 492.
23. Lohmeyer, G., "Wiesse Wannen - Einfach und Sicher". Beton-Verlag, Düsseldorf, 1994, 255 p.
24. Homma, D., Mihashi, H., Nishiwaki, T., "Self-Healing capacity of fibre reinforces cementitious composites". J. of Advanced Concrete Technology, Vol. 7, 2, 2009, pp. 217-228. 25
25. Li, V.C., Yang, E.H., "Self-healing in concrete materials". In Self Healing Materials. Ed. S. van der Zwaag, Springer, 2007, 161-193.
26. White, S.R., Sottos, N.R., Geubelle, P.H., Moore, J.S., Kessler, M.R., Sriram, S.R., Brown, E.N., Viswanathan, S., "Autonomic healing of polymer composites". Nature 409, 2001, pp 794-797.
27. Dry. C.M., "Procedures developed for self repair of polymer matrix composite materials in composite structures". (3). Elsevier. London, 1996, pp. 263-269.
28. Dry, C.M., "Matrix cracking, repair and filling using active and passive modes for smart timed releases of internal chemicals". Smart Materials & Structures, 3, 1994, pp. 118-123.
29. Dry, C.M., "Development of a self-repairing durable concrete". Natural Process Design, Icn. Minnesota, pp. 507-452.
30. Huang H., Ye, G., "Simulation of self-healing by further hydration in cementitious materials". Cement and Concrete Composites, Vol. 34, 4, 2012, pp. 460-467.
31. Breugel, K. van, "Self-healing material concepts as solution for aging infrastructure". 37th Conference on Our World in Concrete & Structures, 2012, Singapore
32. Jonkers, H.M., "Self-Healing Concrete: A Biological Approach". In: Self Healing Materials. Ed. S. van der Zwaag, Springer, 2007, pp. 195-204.
33. Sagripanti, J.L., Bonifacino, A., "A. Comparative sporicidal effects of liquid chemical agents". Applied Environmental Microbiology 62, 1991, pp. 545-551.
34. Schlegel, H.G., "General microbiology", 7th edition, Cambridge University Press, 1993.
35. Jonkers, H.M., Thijssen, A., Muyzer, G., Copuroglu, O., Schlangen, E., “Application of bacteria as self-healing agent for the development of sustainable concrete”. Ecological Engineering 36, 2, 2010, pp. 230-235.
36. Jonkers, H.M., “Bacteria-based self-healing concrete”. HERON Vol. 56, No. 1/2, 2011.
37. Koleva, D.A., Ye, G., Zhou, J., Petrov, P., Breugel, K. van, "Material properties of mortar specimens at early stage of hydration in the presence of polymeric nano-aggregates", Int. Conference on "Microstructure related Durability of Cementitious Composites, Nanjing, China, 13th-15th October, RILEM Publications SARL., 2008, pp. 161-168.
38. Koleva, D.A., Breugel, K. van, Ye, G., Zhou, J., Chamululu, G., Koenders, E.A.B., "Porosity and Permeability of Mortar Specimens Incorporating PEO113–b–PS218 Micelles", Special issue of ACI Materials Journal, SP267, 2009, pp. 101-110.
39. Koleva, D.A., Breugel, K. van, "Tailor-made Nano Materials for Corrosion Control and Improved Matrix Characteristics in Cementbased Systems”. Proc. 4th Int. Symp. on Nanotechnology in Construction, Crete, Greece, 2012.
40. Hu, J., Koleva, D.A., Breugel, K. van, "Effect of Admixed Micelles on the Microstructure Alterations of Reinforced Mortar Subjected to chloride Induced Corrosion". Procedia Engineering 14, 2011, pp. 344-352.
41. Hu, J., Koleva, D.A., Breugel, K. van, Mol, J.M.A., de Wit, J.H.W., "The Influence of Admixed Micelles on Corrosion Performance of Reinforced Mortar", the European Corrosion Congress (ERUOCORR), Moscow, Russia, 2010.
42. Koleva, D.A., Hu, J., Breugel, K. van, Milkova, V., Petrov, P., "The influence of tailored nano/micro polymeric aggregates on material properties of cement-based systems", J. of Intern. Sci. Publications: Materials, Methods & Technologies, volume 5, part 1, ISSN 1313- 2539, http://www.science-journals.eu., 2011, pp. 63-73.
43. Hu, J., Koleva, D.A., Breugel, K. van, "Microstructure analysis and global performance of mortar with tailored nano aggregates".  2^nd International Symposium on Service Life Design for Infrastructure (SLD2010), Delft, The Netherlands. RILEM Publications SARL 2010, PRO 70, 2, 2010, pp. 791-797. 27
44. Koleva, D.A., Hu, J., de Wit, J.H.W., Boshkov, N., Radeva, Ts., Milkova, V., Breugel, K. van, "Electrochemical Performance of Low-carbon Steel in Alkaline Model Solutions Containing Hybrid Aggregates". ECS Transactions, 28, 24, 2010, pp. 115-111.
45. Hu, J., Koleva, D.A., Breugel, K. van, "The Influence of PEO113-b-PS780 Vesicles on the Corrosion Performance of Carbon Steel in Simulated Pore Solution". ECS Transactions, 41, 16, 2012, pp. 1-9.
46. Ishida, T., Mabrouk, R.T.S., Maekawa K., "An integrated computational framework for performance evaluation of cementitious materials and structures under various environmental actions". Proc. 6th Int. Conf. Creep, shrinkage and durability mechanics of concrete and other quasi-brittle materials. Ed. Ulm et.al., 2001., pp. 511-516.
47. Breugel, K. van, "Modelling of Cement-Based Systems - The Alchemy of Cement Chemistry". Cement & Concrete Research, 34, 2004, pp. 1661-1668.
48. Dijksterhuis, E.J., "The mechanization of the world picture". London, Oxford University Press, 1969, 537 p.

About the Author

Current position: Full professor at the faculty of Civil Engineering and Geo sciences of Delft University of Technology. Expertise in design of concrete storage and protective systems, young concrete, durability issues and (numerical) modelling of cement-based materials. Member of several international organisations ACI, RILEM, IABSE and fib.

Dr. Klaas Van Breugel

Affiliation: Delft University of Technology, Netherland.

He has chaired several national and international research committees and was active in number of international European research projects. He is (co-)author of more than 500 conference and journal papers and editor of several books, reports and conference proceedings.

(This paper was presented during the RN Raikar Memorial International Conference & Dr. Suru Shah Symposium on ‘Advances In Science & Technology Of Concrete’ held on 20-21, December 2013 at Hotel Hyatt Regency, Mumbai, organized by India Chapter of American Concrete Institute and published in the compendium of Technical Papers brought out by  ICACI). Built Expressions were the official media partners of the event.