RICERCA ITALIANA     

Biomineral interfaces on carbonates and sulphides

Università degli Studi di Torino

Abstract

There are two parts to the program: the first one, concerning the experiments, is shared by two Units (Torino and Cagliari) and the second one, related to the interpretation of the experimental results, which will be common among all the three Units (Torino, Cagliari and Milano).

The general plan is to do, for experimental:

1a) - Nucleation and growth of CaCO3 calcite crystals (both from aqueous solution and gel) in the presence of specific inorganic impurities (Li+, Na+, Mg2+), carboxylic acids and oligosaccharides which are reckoned to be modifiers of the growth morphology, in order to select those morphologies which are alternately dominated by one of the four main forms found in biogenic natural calcites like {10.4}, {01.2},{00.1} and {11.0}.
1b) - Examination of their surface structures, before and after the adsorption of both inorganic ions and biologically relevant molecules (aspartic and glutamic acids, serine and threonine) with the aim of determining the nano-structures and the self-assembled adsorbed layers. The presence of steps on crystal faces, generated either by 2D growth islands or spiral, their roughness, the kinked or stepped structure of the faces will be stressed, down to the nano-scale level, by means of the atomic force microscopy, along with the presence of epitaxial layers formed by the shaping action of the surface adsorption during growth.

2a) - Biologically induced precipitation of Hydrozincite and co-precipitation of other hydrous carbonates. The coordinative environment of heavy metals ( Pb, Cd e Cu) co-precipitated with the Hydrozincite, will be investigated by X-ray absorption (XAS), anomalous diffraction and EPR ( for Cu only). This analysis shall concern different carbonates (ZnCO3 -Smithsonite, Zn5(CO3)2(OH)6 - Hydrozincite), both synthetic and natural (Rio Naracauli, Iglesiente, Sardegna). Synthetic samples will be prepared at varying metal concentration, both under bulk precipitation and under low kinetic regime.
2b) - Dissolution of PbS-Galena [(001)form] and FeAsS - Arsenopyrite in the presence of carbonate ions and organic anions present in soils. That will be carried out in oxygen saturated solutions (with and without organic molecules) in a continuous flow reactor. According to the obtained results, will be investigated as well the kinetics in the presence of complex organic molecules such as employed for the adsorption on calcite crystals. Galena and Arsenopyrite dissolution will be studied by means of innovative analytical techniques ( in situ-micro Raman and AFM, X-ray photoelectron spectroscopy).

Concerning the interpretation of experimental results, the following path is foreseen:

3a) - Determining the structural character of the investigated crystal faces along with the most probable profiles they can assume within the mother phase.
3b) - Calculating the adsorption energies of the adsorbed species with the aim of justify the observed morphological modifications.
3c) - Modelling the interfacial reaction kinetics and evaluating their activation energies, with the parallel aim of improving the current theories for nucleation and growth in 3D closed systems and for flat crystal surface dissolution.
3d) - Simulation and visualization of molecular self-assembling processes at the biomineral interfaces.

The three Units will continuously and closely share both the validation of the results and the comparison among the analytical interpretation, simulation and experimental data. <<<

Principal Investigator

Dino AQUILANO Università degli Studi di TORINO

Research Objectives

The aim of this research project consists in finding the nano-structures that forms at the interface between minerals and their mother phase in the presence of biologically relevant molecules, choosing those leading to self-assembled adsorbed phases and understanding the mechanisms forming them. The aim is of giving a contribution to the knowledge of the complex biomineralization phenomena. Since the general objective is quite elaborated, it should be better to deal with it at many levels.

The first level concerns the real knowledge of the crystal surfaces on which the biomolecular self – assembling is assumed to take place. Obtaining this preliminary step is not trivial, but fundamental. As a matter of fact, all these surfaces have to be known both experimentally and theoretically, before foreign species are adsorbed on them. Lack of one out of these two complementary aspects seriously compromises the quality of the work. In fact, if the observation is lacking the nano-scale configuration of the surface before the adsorption cannot be known and then the comparison with the same surface, after the adsorption is meaningless. On the contrary, if a preliminary theoretical knowledge of the intrinsic surface characteristics is missing ( its character, its reconstruction and relaxation, the most probable profile, the surface potential, the surface tension without adsorption), it would be hard to reasonably interpret the experimental data. Then, experiments could be deprived of their most fruitful aspect, namely the formation mechanism of the adsorbed phases, their influence on growth kinetics (or dissolution) and the consequent morphological modifications of the substrate crystal.
Coupling of both these conditions (theoretical and experimental knowledge of the surface) rarely is found in biomineralization literature, neither for relevant mineral species such as carbonates and sulphides. This can be more easily found in researches on crystal surfaces concerning either the solid state physical chemistry or the materials science. We reasonably think that this goal can be achieved by our three Units, because we shall integrate in our project the education and the experience of researchers in Earth Sciences, Chemistry Physics, Physics and Mathematics.

The second level, concerning the acquisition of the interface kinetics in the presence of adsorbed species ( both inorganic and organic), represents the key of the project: reaching this objective would result in sensible advance in the knowledge of biomineralization phenomena and mineral weathering in the presence of organic compounds. In our project, the experimental methods and the instrumental techniques will be not only on hand of the kinetic investigation on crystal populations, but on reaction mechanisms face-by-face.
Two out of our Units ( Torino Unit in particular) are experienced for a long time in face-by-face kinetics and knew the advantages in breaking free from the "averaged value" which is typical of crystals treated "as a whole". In fact, the "as a whole" approach in which the crystal is assumed to be a sphere can work, at least, to represent nucleation phenomena while growth (dissolution) mechanisms of the individual crystal forms are inevitably masked.
Using nano-scale techniques (such as the atomic force microscopy) shall allow us to build a sound and sophisticated experimental basis for face-by-face kinetics and for surface topology, as we did in the past with the aid of less advanced techniques.
Lastly, we shall consider not only the " famous "forms , such as the cleavage calcite rhombohedron or the {001} form of galena on which there is a large amount of papers, but we propose studying also the kinetics of the "unhandy" forms whose surface structure is either complex or ambiguous. It is no coincidence that these forms usually occur in biominerals we like to study and that, at least for calcite, these forms are the most frequently occurring in geological environment.

The last level, concerning the methods of interpretation of experimental results, is the most innovative in our project. Certainly, there is today a store of knowledge on the theories usually applied in both "crystal growth" and "surface science". This is largely proved by the results achieved in the Materials Science, especially in the researches on crystallization from vapour phase and melt. The situation is not so satisfying for solution growth, partly for the system is generally more complex and partly because biomineralization phenomena are promoted not only by deterministic aspects but also by other probabilistic events, such as the molecular self-organisation.
The entry in the Project of a Unit composed by mathematicians ( Milano) is fundamental to achieve this advanced goal. On the one hand they can improve our modelling or adapt the current theories for the interpretation of those phenomena which follow traditional paths ( morphological changes due to ionic impurity action during nucleation, growth, adsorption). On the other hand, when we are dealing with the understanding of self-assembling of foreign molecules on the mineral surfaces, new models are needed to solve the related probabilistic problem. The well grounded experience of this "ad hoc" Unit can give the research a push in the right direction. <<<

Timescale

24 months

National and international background

The meaning of the term "biomineralization" cannot be reduced to the only minerals produced by living organisms; actually it can be extended to all mineralised products having the structure of composite materials in which both inorganic and organic components coexist. Biomineral phases, having formed under controlled conditions, often have properties such as shape, size, crystallinity, isotopic and trace element content quite unlike its purely inorganically formed counterpart.
We will deal, in this research proposal, with the "biomineral interface", namely with those nano-structures that form in the transition zone between crystal surfaces of minerals (calcite, hydrous carbonates, metal sulphides) and its mother phases when organic molecules of biological interest are present in it. A wide reference frame has been produced in the last twenty years on biomineralization, as the reader may find in some fundamental reviews [1,20].
The main requirements to be fulfilled in order to build a biomineral interface lie in the capability of inorganic crystal surface to recognize the organic molecules which have to be adsorbed on it [1]. When this condition obtains, the thermodynamic interfacial properties are so affected that, for instance, the nucleation frequency of one out of the calcium carbonate polymorphs is enhanced with respect to that of another one [2].
Determining the crystal surface conditions gains in importance. Then, at the beginning of the nineties, new interest grew for the knowledge produced from the sixties by the crystal growers and special attention was paid to the changes that crystal surfaces undergo when growing from solution in the presence of both inorganic and organic impurities [3]. At the same time the first ex-situ and in-situ AFM observations on crystal surfaces came out.


Nano- structures and self- assembling on calcite

The cleavage {10.4} calcite rhombohedron was investigated in detail and, from one year to the next, the knowledge changed over from the overall crystal face kinetics to the measurement of the surface nano-structures (spiral steps and/or 2D embryos) spreading on its faces, determining its growth rate and changing their shape and rate according to the adsorbed impurities [4].
Interpreting the morphological and kinetic data drives into finding the reasons why a given { hkl } form is more stable than another one: then, the calculation of the surface energies of the most important forms of calcite (both in the presence of solvent and other specific impurities) became a subject of great interest, along with the simulation of the surface site energies [5].
The most advanced researches in biomineralization concern calcium carbonates: the oriented nucleation of calcite is found depending on the self-organization of organic monolayers ( SAMs – self assembled monolayers) pre-adsorbed on crystalline substrates [6-8]. At the same time, the number of reports on epitactic interface nano-structures increases [16] and the molecular growth mechanisms of the {10.4} form are better and better known [9,13,17]. The chirality of the adsorbed molecules enhances the growth anisotropy of the complementary spiral steps [18] while modelling the interaction between organic adsorbates and inorganic crystalline substrate explains their inhibitory effect on crystal growth [19].
Some fundamental problems have still to be solved, in spite of both quality and quantity of the obtained results:
- The study of the biomineral interfaces rarely goes beyond the {10.4} form. This crystallographic form certainly is the most popular and easy to investigate thanks to the uniqueness of its profile facing the growth (or dissolution) environment. Nevertheless, it is not the most important calcite form occurring in Nature, both in inorganic and biogenic crystals, for other forms overtake its occurrence frequency (namely, the {10.0}prism, the {21.4}scalenohedron, the steep {01.2} and the flat {01.8} rhombohedron, and the {00.1} pynacoid). Then, it should be very interesting to study how the adsorption of biomineralizing molecules could influence the growth kinetics of these forms.
- Among the just quoted forms, the only one which has not to be reconstructed is the {01.2}, even if it shows two not equi-probable profiles. The other {10.0}, {21.4} and {01.8} forms have stepped character, while {00.1} is a kinked one, all needing to be reconstructed before considering the interaction with the adsorbed molecules. The mere relaxation, just now considered in the
literature, is not realistic, if not ambiguous.
- The condition of the surfaces, i.e. their defect density, generally is not examined before the molecular adsorption. Cleavage surfaces are usually used, whilst for a better understanding of morphological changes, the crystal surfaces have to be "as grown", namely with their surface defects and nano-structures due to interplay of their growth mechanisms.
The practice gained by Torino Unit in the general field of crystal growth and especially in the calcium carbonate morphology is a sound background to face the above mentioned problems. With this aim we have already tackled the following steps:
- theoretical equilibrium and growth morphologies of the most stable polymorphs of calcium carbonates [10],
-the perfection of the nano-thick surfaces of calcite crystals nucleated and grown encompassing gas cavities in pure solution and in the presence of specific impurities [11],
-the competition between the growth kinetics of calcite forms having different characters (both from gel and solution growth), along with the finding of growth nano-structures at the solution/crystal interface [12,14,15].


Biomineralization and surface kinetics of hydrous carbonates and metal sulphides


Understanding mechanisms ruling the interaction between minerals and bio-sphere is central to predict both stability and reactivity of minerals. Such a knowledge is useful in environmental problems of heavy metal mobilisation and trapping. In fact, precipitation and dissolution of carbonate minerals partially regulate the fate and transport of anthropogenic pollutants, especially heavy metals, at both local and Earth scale.
Hydrocerussite and hydrozincite are typical biologically induced carbonate minerals [20]. In a dismantled mine area (Iglesiente – Sardegna, Italy) hydrozincite [Zn5(CO3)2(OH)6] was unexpectedly found in an iron-poor stream of nearly neutral pH where both photosynthetic bacterium (Scytonema) and microalga ( Chlorella) were found [21]. In fact, hydrozincite is expected to precipitate, from pure aqueous solution, only above solution pH about 10. This bio-mineral is able to mitigate the Zn pollution of surface waters. Since also other hydrous carbonates (Pb, Cd, Cu) co-precipitate with hydrozincite, also the content of other heavy metals is reduced.
Moreover, stability constants for complexes between metal and organic anions secreted by living organisms show high values, and the interaction between minerals and bio-sphere can be regarded as a way to mobilize heavy metals from polluted soils, that result in pollution of surface water after interaction with soils rich in metal sulphides. Mass exchange between soils and vegetation takes place at the interface between mineral surfaces and root exudates. Organic acids are the main biochemical component of root exudates [22] and can be found in significantly high concentrations in both hydrothermal and sedimentary environments.
Recently, the effect of organic ligands on mineral surfaces was investigated to gain insight into the complexity of organic activity at the mineral-water interface. A fairly large body of literature is available for silicate and oxide minerals, whilst little is known about the interaction between metal sulphide surfaces and organic ligands at Earth surface conditions.
Previous work undertaken by our group have shown that dissolution of (001) galena surface in contact with oxygen-saturated solutions takes place via the formation of nanometric phases that coat the surface [23]. These are, likely, native sulphur and secondarily phases with Pb-O type bonding [24]. As previously determined at the macroscopic scale, the activation energy for galena dissolution falls below 20 KJ/mol, at pH of 3, while it averages 45 KJ/mol at pH around 4-5. In the presence of dissolved organic groups (like acetate) apparent activation energy for galena dissolution in oxygen-saturated solutions decreases by about 10 KJ/mol [25]. Understanding and control of the elementary reactions involved in bio-mineralisation in carbonate bearing environment are poorly known [26], as well as elementary reactions ruling the interaction between carbonate ions and metal sulphide surfaces [27], both in presence and without organic matter.
Molecular simulations are a promising technique allowing one to investigate the separate components of the precipitation (dissolution) mechanism. Given the recent advances in calculation power, these technique allow one to achieve molecular scale results on molecular dynamics, that cannot be directly achieved by measurements.
"First principles methods" can tackle problems involving tenths of atoms (~ 200) and follow processes on time scales of some picoseconds. These short times and small systems provide still important insights into the first steps of the reactions monitoring several properties, like electron density, charge transfer, Wannier centers and Boys orbitals. To bridge the time and length scales
between micro- and mesoscopic pictures promising results have been achieved with the so called multiscale computer simulations.
An example is the QM/MM approach [28]. The system of interest is divided into two parts. The region of chemical interest, typically the binding sites of molecules or the active sites in enzymatic processes, is treated at the quantum mechanical level (QM-part) while the remaining domain (solvents included) is described via classical interaction potentials (MM-part). Thus, an interaction feedback between the larger classically- and quantum mechanically-treated subsystems is introduced in the calculation, increasing their predictive power.
Our group already applied molecular simulations to investigate the reactivity of mineral surfaces. First principle calculation [29] indicate that (001) galena surface undergoes relaxation for several atomic layers. Recent theoretical studies (Wright et al, 1999) have shown that the ideal clean surface of PbS is unreactive as well as the presence of vacancies in the surface layer does not provide a relevant surface reactivity. However, the investigation of water dissociation in proximity of surface nano-structures (steps) suggests the enhancement of reactivity.
Understanding of biological mineralization involves interdisciplinary studies at different scales, ranging from field studies to molecular scale studies [20]. As shown by recent molecular simulations [31] is actually possible to gain further insight into the mechanisms of minerals dissolution/precipitation with and without addition of dissolved chemical species secreted by living organisms.


The theoretical support of Mathematics to the elementary crystallization processes.

Nucleation and growth of crystals are often coupled with an external field (concentration, temperature, pressure, etc.) whose evolution depends on the growing crystal itself. Nucleation is usually random both in space and time, and hence the phase transition is mathematically modelled by a stochastic process, strongly coupled with the evolution of the underlying field. Sometimes, the coupling, together with a small diffusion coefficient, is responsible of the heterogeneity of the underlying field, that causes the anisotropic growth of the crystal.
This is the case of polymer crystallization, where the radial growth rate of the single crystals depends on the temperature field which, in turn, depends on the growing crystal through the release of latent heat, that causes a local increase of the temperature in the crystal surroundings.
The research group of the Department of Mathematics ( Milano University) has developed a strong competence on the analysis of crystal growth processes, in presence of heterogeneities in space and time. [32] sums up the recent work done in these last years by the research group on the Mathematics of Polymer Crystallization Processes, coordinated by prof. Capasso.
In [33] a simulation method for a "birth and spread" process coupled with an underlying field is presented, which realistically reproduce the crystallization process and is helpful both for avoiding real laboratory experiments and for the correct estimation of the typical parameters of the process; the estimators may be based on the geometric and morphologic characteristics of the obtained crystals.
Crystallization phenomena are often represented by multiple scales models; specifically with interacting particle systems on the microscopic scale and with "mean field" models on higher scales, by coupling the continuous fields (temperature, concentration, etc. ) whose evolution is typical of the macro scale, with the stochastic phenomena at the micro scale. Therefore the evolution of these systems can be represented by hybrid models, which describe at the macroscopic scale the evolution of the growing crystal [34-36]. These models lead to the formulation of suitable "laws of large numbers" that provide an average of the micro-scale stochastic phenomena, so that it is possible to formulate deterministic models that explain the mean evolution of the system at a higher scale. These models have been studied in [39].
The models previously presented have been studied specifically for polymer crystallization, where the underlying field is the temperature field; nevertheless they can be extended for other crystallization processes, e.g. when the growth is coupled with a surrounding supersaturation field .
The methods of parameter estimation that involve the use of geometrical models are based on the analysis of the (stochastic) geometrical characteristics of the obtained crystals (size and shape). Estimates can be obtained by applying to these geometries some typical techniques of Statistical Shape Analysis [40] and of Stochastic Geometry. The group of the Milano University has a long term experience in these topics [37, 38, 42]. <<<