Nano-tailoring organo-mineral materials - Controlling strength and healing with organic molecules in mineral interfaces
Concrete is the largest volume manufactured solid on earth manufactured from abundant, inexpensive raw materials like limestone, sand and gravel. The most common cement in concrete is ordinary Portland cement (OPC) that undergoes chemical reaction, polymerization and crystallization when clinker is mixed with water. Portland cement clinker production contributes about 5% of global anthropogenic CO2 emissions, and about 60% of this CO2 is of "fossil" origin, i.e. contained in the limestone rock which is calcined during the manufacturing process1. Although the "fossil CO2" emitted during the manufacturing process is slowly re-absorbed by the hardened concretes produced from the cement, this process takes hundreds to thousands of years1. Researchers and startup companies have proposed a multitude of “green” or “low CO2” concretes with very different chemistry and mechanical properties1,. There are a number of challenges associated with exchanging OPC with a greener alternative - new microstructures, mechanical properties and durability have to be studied.
The most important challenge is to assure tensile strength of novel concrete construction elements. Because OPC holds a pH of about 12.5 it protects steel against corrosion and cheap steel may be used to add tensile strength1 to concrete that in itself is as brittle as glass. A multitude of fibre reinforced concretes exist and after asbestos was banned for health reasons most fibres in use are of organic origin. Tensile strength of high volume concrete is still a challenge because of the price of fibres and additional cement and because long fibres tend to fold together while the wet cement paste is mixed and poured. The construction industry makes extensive use of polymers to control the rheology of cement paste and thus influence its water demand, stability over static and dynamic segregation and ultimately the strength of the hardened material. Despite the long-time industrial practice, the interaction of polymer surface adsorption with the ongoing hydration reactions is not entirely clarified and a general lack of reproducibility is observed5. General conclusions about the performance of specific polymers or surfactants are difficult to draw, and each mix has to be individually tested before being used on the construction site5.The limited ductility of the hardened cement paste poses important issues of lack of tightness in deep oil well cementing for casing, zonal isolation and plug and abandonment. Recent important failures resulted in impressive blow-outs and oil spills. The search for cements with increased ductility or self-healing properties is a topic of increasing importance.
Common to the challenges mentioned here are the current widespread use but lack of fundamental understanding of the interactions of the organic material with the inorganic cement. This is also one of the challenges in engineering biocompatible cements for human implants.
Sedimentary rocks have many similarities with concrete in microstructure and strength. Carbonate sedimentary rocks, like limestone predate concrete as a construction material and a large fraction of Europe’s monuments and historic buildings are limestone. Salt crystallization in limestone and concrete causes considerable damage and the prevention or healing of such damage is of great concern to Europe’s cultural heritage. Many of the world’s largest oil reservoirs are carbonate rocks that are prone to creep, compaction and subsidence during oil production. Especially during injection of water or CO2 to enhance oil recovery such compaction causes well damage and leakage. The IPPC report on CO2 capture and storage identified CO2 injection for enhanced oil recovery (EOR) as an economically interesting large scale CO2 storage option. The Norwegian Petroleum Directorate has recently published a CO2 Storage Atlas for the Norwegian North Sea where CO2 injection for EOR is identified as an important storage opportunity. Detailed studies to evaluate challenges are called for and the importance of a stable supply of CO2 is stressed. The southern Norwegian shelf is close enough to the European continent to be part of a trans-European CO2 transport network. The southern Norwegian shelf chalk fields and particularly ongoing research on EOR by CO2 WAG (Water Alternating Gas) at Ekofisk is highlighted in the Storage Atlas11. However the risk of further compaction of the chalk reservoir resulting from CO2 injection needs to be assessed. CO2 storage in carbonate reservoirs thus requires a very high control of long-term strength of the reservoir rock. Recent research shows that the presence of organic molecules in chalk is key to understanding the strength of chalk reservoirs on geological time scales, thus also this societal challenge needs improved understanding of the strength of organo-mineral cementitious materials.
State of the art
Mineral grains are usually much stronger than cemented aggregates of mineral grains. The strength of aggregate materials, like concrete and carbonate rocks is controlled by grain arrangement (microstructure) and strength of the individual mineral grain contacts (interfaces). The mean values of the properties like subcritical crack growth speed, creep rates and yield stress depend on the “average microstructure” and “average grain interface strength”. However, concrete and sedimentary rocks are prototypical disordered materials, whose strength cannot be defined as a single property, but a number of different measures of how the material behaves when being deformed beyond elastic strain. The deformation processes feature non-trivial scaling requiring statistical models to rationalize the material response. Indeed, fluctuations in the properties show a large dynamic range, highlighting the non-trivial scaling of the complex deformation process. A number of models describing the transition between brittle and ductile behaviour in disordered materials have been proposed, setting a framework for experimental analysis. Non-elastic deformation is, for most solids, associated with permanent loss of strength. Materials that reform bonds of the same strength as the original bonds may retain full strength under shear or compression. For example a sand pile (dry or wet) has the same material strength irrespective of the deformation. Such materials are often termed self-healing materials (SHM). SHM often have a degree of ductility to avoid single, open cracks that cannot heal, but rather form lots of microcracks with sufficiently small aperture for the healing mechanism to be active
, 20. We have recently found that chalk is a natural SHM13. Chalk consists of tiny calcite crystals that are intrinsically very reactive. If these crystals could recrystallize, chalk would have a very high creep rate under its own load. However, the recrystallization is passivated by organic molecules. When a crack appears the crack surfaces consist of fresh unpassivated calcite that may react to heal the crack.
The increased resolution of X-ray tomography and focused ion beam (FIB) – scanning electron microscopy (SEM) allows the detailed characterization of the 3D nano- and microstructure and crack surfaces. These measurements can be complemented by numerical modelling of cement and porous rocks. The marriage of 3D imaging, image segmentation and numerical simulations of flow in the segmented volume is currently being integrated into a single workflow by commercial companies like FEI, Lithicon and Ingrain. However, reactive fluid flow and mechanical properties of the segmented microstructure are still beyond the standard methods.
The use of the visualization techniques mentioned above provides a unique approach to investigate the structures arising from self-assembly (SA) processes. SA provides a route to create unusual structures in new and ingenious ways by using objects of different shape. As a matter of fact SA is being employed to guide the mineralization of artificial bone and it is used by Nature to make organic-mineral bio-composite materials such as Abalone nacre. Nacre is a composite of aragonite (CaCO3) platelets glued together by biopolymers. It has a fracture toughness 1000 times higher than that of aragonite. Nacre is one of the most studied biological organic-mineral composite and many self-assembled materials mimicking nacre have been made.
The strength of any aggregate is a complex function of the collective properties determined by the microstructure and specific properties at individual grain contacts like surface energy, adhesion, friction, fracture, dissolution and growth. During the aggregation and cementation process the solid mineral surfaces are brought within a few nanometers of each other with a nano-confined water film in the interface. This aqueous film allows diffusion, and can sustain shear stresses as a solid would. The nano-confined liquid film mediates crystallization, dissolution, material transport, subcritical fracturing and fracture healing. The forces between the two solids may be attractive or repulsive depending on the thickness of the water layer, the pH, the ion concentration and the ion distribution of the solid surfaces. These forces may be predicted thermodynamically by assuming that the confined solution behaves as a bulk liquid. However, the behaviour of water under nm confinement can be very different from bulk, a fact that explains the failure of the DLVO theory. Recent experimental evidence shows, that nano-confinement induces nucleation and growth at bulk concentrations where no crystals should form. In addition, when the solid surfaces are reactive they alter the ion distribution at the surfaces creating a feedback mechanisms that may cause instabilities. Recent experiments of crystal growth in nano-confinement cannot be predicted with classical thermodynamic theory, showing the need for improved concepts, theories and models to describe nucleation and crystallization in confinement. Currently, we cannot predict if growing crystals in contact will attract (and the growth will form cement and strong adhesion) or if the surfaces will repel (and growth will push the crystals further apart). Both cementing growth and “force of crystallization” causing damage (for example salt damage) are observed in nature and in construction materials, but we cannot predict when one or the other will happen8.
In order to explain the processes in the nano-confined interface between growing or dissolving crystal surfaces one needs to understand crystal growth and dissolution. The explanation of growth/dissolution processes requires a good understanding of the dynamics of atomic steps at solid surfaces. This topic has been at the forefront of fundamental studies of crystal growth and dissolution for several decades. The atomic force microscope (AFM) became comparatively cheap in the 1990s and measurements of atomic step motion during dissolution and growth uncovered the mechanisms of growth/dissolution inhibitors (additives). It is now becoming clear, however that also at this scale there is intrinsic variability of rates and reactivity not explained by current models. Additives, organic molecules, modify the growth/dissolution process. Computer simulations provide a good approach to understand how additives operate at mineral interfaces. Indeed, the modelling of organic-mineral interfaces has attracted significant attention. For example ab initio quantum computations (Density Functional Theory (AI-DFT)) have been employed to quantify calcite surface energies. However simulating large systems is difficult, and molecular simulations (MS) with classical forcefields are better suited to explore the dynamics of solutions and surfaces under confinement. MS has been used to establish correlations between molecular adsorption and mineral topology  (e.g. atomic steps), as well as to compute interfacial free energies of mineral surfaces ,33. The direct simulation of nucleation events is difficult, given the high activation barriers associated to the process. A range of methods have been proposed to simulate these “rare events”. Metadynamics, provides an approach to simulate nucleation by forcing the system away from frequently visited states, hence enabling a route to overcome high energy barriers. Such methods offer great potential to understand the role of nano-confinement and organic-mineral interfaces on growth/dissolution processes.
Computer simulations can also be employed to quantify the mechanical properties of granular materials. However, this is a multiscale problem, spaning a huge range of length scales: from 10-9-10-6 m, characteristic of structure of matter and defects, to >10-4 m, characteristic of crystals and phases, passing through the length scales that are relevant to defects interaction, 10-6-10-3 m. Modelling all these length scales requires a concerted computational approach. Hence, in addition to the atomistic approaches discussed above a whole range of methods have been deployed to investigate the mesoscopic domain, 10-8-10-6 m (Monte Carlo, stochastic dynamics) and larger scales >10-6 m (finite elements, continuum hydrodynamics, Dissipative Particle Dynamics or Lattice Boltzmann). Although multiscale approaches have been employed to study macromolecules and their application to cementitious materials and organic-mineral interfaces is still in its infancy. There are a few studies that have considered generic healing mechanisms as well as cement microstructure and particle packing. These studies are promising and hihlight the great scope for innovation in computational studies of materials. NanoHeal will take full advantage of the excellent computational groups taking part in the consortium, which cover all the necessary length sales needed to understand the relationship between material strength, interfacial interactions and the role of organic additivies in releasing stress and promoting healing.
Overview and objectives
There is a current need to develop new cements with different chemistry, microstructure and mechanical properties. These cements should target current needs in industry, technology, health and conservation science. These scientific, industrial and societal challenges will be addressed through the following aims:
- to develop innovative probes and models for nanoscale processes that open novel perspectives in design and control of organo-mineral materials.
- to measure and improve the strength and durability of 1) new man-made cemented materials like “green concrete”, speciality cements in construction and oil and gas recovery, and biocompatible implants and 2) natural sedimentary rocks inside reservoirs and as construction materials
- to educate young interdisciplinary researchers at the intersectoral interface between fundamental science and European industry.
 E. Gartner, Cem. Concr. Res. 34, 1489 (2004). E. M. Gartner and D. E. Macphee, Cem. Concr. Res. 41, 736 (2011)
 J. L. Provis and J. S. J. van Deventer, editors, Alkali Activated Materials (Springer 2014).
 V. C. Li, J. Adv. Concr. Technol. 1, 215 (2003). M. Li and V. C. Li, Mater. Struct. 46, 405 (2012).
 R.J.Flatt, N.Martys and L.Bergström, MRS Bulletin, 29, 314 (2004). K.Yamada, S.Ogawa, S.Hanehara, Cem. Concr. Res. 31, 375 (2001).
 S. Le Roy-Delage, et al. SPE 128226, (2010). http://www.slideshare.net/Statoil/plug-abandonment.
 J. Anderson et al. , Deep Water - The Gulf Oil Disaster and the Future of Offshore Drilling - Report to the President (2011).
 G. W. Scherer, Cem. Concr. Res. 34, 1613 (2004). B. Lubelli and R. P. J. van Hees, J. Cultural Heritage 8, 223 (2007)
 J. T. Fredrich, et al, SPE Reservoir Eval. & Eng. 3, 348 (2000)
 B. Metz, O. Davidson, H. de Coninck, M. Loos, L. Meyer, and C. U. Press, IPCC Special Report on Carbon Dioxide Captue and Storage (Intergovernmental Panel on Climate Change, 2005).
 E. K. Halland, W. T. Johansen, and F. Riis, CO2 Storage Atlas Norwegian North Sea (Norwegian Petroleum Directorate, 2011).
 J. Morbee, J. Serpa, E. Tzimas, and P. O. of the E. Union, The Evolution of the Extent and the Investment Requirements of a Trans-European CO2 Transport Network (European Commission, 2010).
 T. Hassenkam, et al. Proc. Natl. Acad. Sci. 108, 8571 (2011). P. Japsen, D.K. Dysthe, et al. J. Geophys. Res. Earth 116, (2011).
 M. J. Alava, P. K. V. V Nukalaz, and S. Zapperi, Adv. Phys. 55, 349 (2006).
 L. I. Salminen, A. I. Tolvanen, and M. J. Alava, Phys. Rev. Lett. 89, 4 (2002). S. Santucci, et al. Phys. Rev. Lett. 93, 4 (2004).
 D. De Tommasi, G. Puglisi, and G. Saccomandi, Phys. Rev. Lett. 100, 4 (2008). S. Kale and M. Ostoja-Starzewski, Phys. Rev. Lett. 112, 5 (2014). C. B. Picallo, J. M. Lopez, S. Zapperi, and M. J. Alava, Phys. Rev. Lett. 105, 4 (2010).
 M. Wu, et al, Constr. Build. Mater. 28, 571 (2012). K. Van Tittelboom and N. De Belie, Materials 6, 2182 (2013).
 F. Tariq, R. Haswell, P. D. Lee, and D. W. McComb, Acta Mater. 59, 2109 (2011).
 E. Masoero, E. Del Gado, R. J.-M. Pellenq, F.-J. Ulm, and S. Yip, Phys. Rev. Lett. 109, 155503 (2012).
 B. Pokroy, S. H. Kang, L. Mahadevan, and J. Aizenberg, Science, 323, 237 (2009).
 J. D. Hartgerink, E. Beniash, and S. I. Stupp, Science, 294, 1684 (2001).
 A. Sellinger, P. M. Weiss, A. Nguyen, Y. Lu, R. A. Assink, W. Gong, and C. J. Brinker, Nature 394, 256 (1998).
 J. N. Israelachvili “Intermolecular and surface forces” 3rd ed. Associated Press, Waltham (2011)
 D. K. Dysthe and R. Wogelius, Chem. Geol., 230,175 (2006)
 J. Faraudo and F. Bresme, Phys. Rev. Lett., 92, 236102 (2004)
 R. M. Espinosa-Marzal, et al, Phys. Chem. Chem. Phys. 14, 6085 (2012).
 A. Royne and D. K. Dysthe, J. Cryst. Growth 346, 89 (2012).
 C. Misbah, O. Pierre-Louis, Y. Saito, Reviews of Modern Physics, 82 981 (2010), A. Pimpinelli and J. Villain, “Physics of crystal growth”, Cambridge Univ. Press, Cambridge (1998).
 H. H. Teng et al. Science 282, 724 (1998), E.A. Pachon-Rodriguez, A. Piednoir, and J. Colombani, Phys. Rev. Lett. 107, 146102 (2011)
 A. Luttge, R. S. Arvidson, and C. Fischer, Elements 9, 183 (2013).
 J.H. Harding, D. M. Duffy, M.K, Sushko, P. M. Rodger, D. Quigley and J.A. Elliott, Chem. Rev., 108, 4823 (2008).
 H. Sakuma, M.P. Andersson, K. Bechgaard, and S.L.S. Stipp, J. Phys. Chem. C, 118, 3078-3087 (2014);
 T.D. Perry, R.T. Cygan and R. Mitchell, Geochim. Cosmochim. Acta, 70, 3508 (2006)
 S. Elhadj, E.A. Salter, A. Wierzbicki, J.K. de Yoreo, N. Han, P.M. Dove, Cryst. Growth Des., 6, 197 (2006)
 S. Iglauer, M.S. Mathew and F. Bresme, J. Coll. Interf. Sci., 388, 405-414 (2012);
 A.F. Voter, F. Montalenti, T.C. Germann, Ann. Rev. Mater. Res., 32, 321 (2002)
 A. Laio and M. Parrinello, Proc. Natl. Acad. Sci. USA, 99, 12562 (2002), D. Quigley et al., J. Chem. Phys., 134, 044603 (2011).
 M.O. Steinhauser, Computational Multiscale Modeling of Fluids and Solids – Theory and Applications, Springer.
 J.J. de Pablo, Annu. Rev. Phys. Chem., 62, 555 (2011).
 A.C. Balazs, Materials Today, 10, 18, (2007). S. Tyagi, J.T. Lee, G.A. Buxton and A.C. Balazs, Macromolecules, 37, 9160 (2004).
 D.P. Bentz, J. Am. Ceram. Soc. 80, 3 (1997).
 H. He, P. Stroeven, M. Stroeven, L.J. Sluys, Computers and Concrete, 8, 677 (2011).