| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Geological Sciences, University of Michigan, 2534 C.C. Little Building, 1100 N. University Avenue, Ann Arbor, Michigan 48109-1063, USA
** Division of Engineering and Applied Sciences, Harvard University, Pierce Hall, 29 Oxford Street, Cambridge, Massachusetts 02138, USA
*** Forschungszentrum Karlsruhe, Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany
**** Institut für Mineralogie, Universität Münster, Corrensstraße 24, D-48149, Germany
***** Departamento Cristalografía y Mineralogía, Universidad Complutense, 28040 Madrid, Spain
Corresponding author: E-mail address: ubecker{at}umich.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT SELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Subsequently, we will increase the level of complexity by investigating the interaction between organic molecules and mineral surfaces. As an example, we describe the influence of different organic growth inhibitors on the growth velocity of barite and use molecular simulations to identify where these organic molecules attack the surface to inhibit growth.
Nature provides a number of complex organic molecules, so-called siderophores that are secreted by plants to selectively extract Fe ions from the surrounding soil. The molecular simulations on siderophores are complemented by atomic force-distance measurements to mimic the interaction of these molecules with Fe and Al oxide surfaces. The combination of simulations and force-distance measurements allows us to evaluate initial complexation on metal oxide surfaces (which is different from metal complexation in solution), steric hindrances, the possibility to remove metal ions from oxide surfaces, and selectivity for removal of Fe3+ over Al3+.
Finally, we describe first attempts to find polypeptide sequences that may be used as precursors for biomineralization of calcite surfaces.
|
SELECTIVE ATTACHMENT OF MONOVALENT IONS TO POLAR STEPS ON SULFATES |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Two primary strategies can be pursued to solve the scale problem: (i) Crystal growth inhibitors are used to prevent scale precipitation. The attachment of these organic molecules to active growth sites disrupts the nucleation process and hinders the continuation of growth (Bosbach and others, 2002). (ii) Complexing agents are applied to dissolve barite precipitates. The latter has the disadvantage that large amounts of chelators are necessary because one chelator molecule is consumed for each dissolved cation (Coveney and others, 2000).
In order to understand barite scale formation and to improve scale treatment technologies, a sound understanding of the complex precipitation and dissolution processes is required. Although much work has been dedicated to study these processes from a macroscopic point of view (Dove and Czank, 1995), microscopic mechanisms such as two-dimensional nucleation, spiral growth, and etch pit formation have to be studied at a molecular level. Atomic Force Microscopy (AFM) can be used to characterize individual growth and dissolution mechanisms in situ. Details on barite growth and dissolution have been reported in numerous AFM studies (Archibald and others, 1997; Pina and others, 1998a, 1998b; Bosbach, 2002). Special attention has been paid to the effect of organic growth inhibitors (van der Leeden and others, 1995; Bosbach and others, 1998), organic chelators (Putnis and others, 1995), pH (Dunn and Yen, 1999), and temperature (Monnin and Galinier, 1988; Higgins and others, 1998). However, the influence of the ionic strength, specifically the role of monovalent cations temporarily attaching to surface steps, on the barite scale formation has been neglected in microscopic studies. However, the role of background electrolytes can be important because in offshore exploration, the injected seawater has high ionic strength (I=0.7). Only one study involving calcite has attempted this: Shiraki and others (2000) have determined the dissolution kinetics in 0.1 M NaCl solution in AFM experiments. Macroscopic experiments indicate that barite precipitation (He and others, 1995) and dissolution kinetics are affected by the presence of background electrolytes (Christy and Putnis, 1993). Here, we review AFM observations on the influence of ionic strength on growth and dissolution of barite with some comparative experiments with celestite (Becker and others, 2002). Furthermore, certain microtopographic features such as the morphology of etch pits and molecular islands can be clearly associated with the presence of specific background electrolytes. In highly saline solutions, we obtain very similar etch pit morphologies to those which have previously been attributed to specific interactions between organic molecules and certain sites on a mineral surface (Wang and others, 1999a, 1999b, 2000). In other words, it is crucial to understand possible interactions of single-atom ions with mineral surfaces, in particular their influence on growth rates and morphologies, before more complex organic-mineral interactions such as biomineralization are considered.
In recent years, an increasing number of molecular simulation studies were performed on the influence of foreign cations and inhibitor molecules on the growth and dissolution of mineral surfaces, particularly on carbonate (Parker and others, 1993; Nygren and others, 1998; de Leeuw and others, 1999; Gerbaud and others, 2000; Cooper and de Leeuw, 2002; de Leeuw and Cooper, 2004) and sulfate surfaces (Allan and others, 1993; Rohl and others, 1996; de Leeuw and Parker, 1997; Redfern and Parker, 1998; Risthaus and others, 2001; Becker and others, 2002). It was shown that a precise description of hydration energies in solution and at the surface is crucial for the analysis of the subtle but significant differences of the adsorption thermodynamics of different ions on specific surface sites (Sayle and others, 2000). The net adsorption energy to a specific surface site can be described in the following way (eq 1):
 | (1) |
Experimental and Theoretical Methods to Determine the Adsorption of Monovalent Ions on Barite Steps
The experimental and theoretical methods used here are described in more detail elsewhere (Pina and others, 1998a, 1998b, 2004; Risthaus and others, 2001; Becker and others, 2002). In brief, crystal growth and dissolution was measured as step advancement and retreat velocities in AFM fluid cell experiments on (001) barite and (001) celestite surfaces as a function of different supersaturation values S (with S = IAP/Ksp, IAP is the ion activity product in solution = acation x asulfate) and background electrolyte concentrations.
Calculations of sulfate surfaces and on the interaction between background electrolyte ions and specific sulfate surface features were performed using empirical potential force field models based on the sulfate potentials developed by Allan and others (1993). For all computations of step energies and adsorption energies, we applied two-dimensional periodic boundary conditions parallel to the surface. For computations involving surface relaxation, the uppermost two molecular layers (out of four) were allowed to relax. This thickness is sufficient because relaxation does not progress far into the bulk of what we tested using thickness-dependent surface relaxation calculations.
We define step energies as the energies that a surface step is energetically less favorable (per unit length of step; see also Becker and Gasharova, 2001) than the same number of atoms on a flat terrace. Step energies are calculated by comparing a relaxed step structure with a relaxed flat surface ("relaxed" case) or an unrelaxed step with a bulk terminated surface ("unrelaxed" case).
Bulk hydration energies of ions dissolved in solution were taken from experimental values from Rashin and Honig (1985) and references therein. Partial hydration energies of adsorbed surface species were calculated by performing molecular dynamics simulations with four water layers on the surface with and without the adsorbed species (see also Becker and others, 1998). Water potentials were tested to be in good agreement (less than 5% error) with the experimental hydration energies described in Rashin and Honig (1985) by applying the droplet method.
AFM Results of Monolayer Step Growth in the Presence of Background Electrolytes
The most dominant growth mechanism on barite and celestite surfaces is the attachment of ions from solution onto kink sites leading to the advancement of molecular steps. The growth kinetics of such steps was studied at various background electrolyte concentrations. However, in order to study this effect quantitatively, step growth rates were measured on barite (001) surfaces at constant supersaturation with increasing ionic strength. Due to the decrease of the mean activity coefficients of aqueous Ba2+ and SO42-, the concentration of Ba2+ and SO42- in solution had to be increased with increasing salinity in order to obtain constant IAPs for barite and celestite, respectively. In addition, the formation of NaSO4– and BaSO4 species in solution further reduces the free SO42- activity. Consequently, Ba2+ and SO42- concentrations have to be increased non-stoichiometrically with increasing NaCl background electrolyte concentration.
After injecting a moderately supersaturated BaSO4 solution (S = 19.6) with low background electrolyte concentration (ionic strength = 0.003) into the AFM fluid cell, two-dimensional islands with a height of one molecular BaSO4 layer form on a barite (001) surface as shown in figure 1A. Their lateral spreading results in a distinct circular sector-shaped morphology, which reflects the relative growth rates in different crystallographic directions. The morphology is controlled by straight step edges parallel to [120] and [1
0] (since both are symmetry equivalent, they are often referred to as 
) and a curved step edge between them. The orientation of these islands alternates in consecutive molecular layers due to the 21 screw axis parallel to [001]. Step edges parallel to form relatively stable periodic bond chains with a low kink site density. In contrast, the curved step edges that advance in the fast growth direction have a high kink site density. Therefore, the curved step edges are the most reactive parts of the barite (001) surface where step growth preferentially occurs. The growth velocity of these islands is highly anisotropic. Growth normal to the curved step edges is approximately 10 times faster than growth normal to the straight edges.
|
0], fig. 1B) of the sector-shaped islands becomes more obtuse.
|
at low ionic strength (Fig. 1C). This leads to spiral growth hillocks with relatively little lateral growth compared to two-dimensional islands. Also in the case of spiral growth, growth rates in different crystallographic directions change with increasing NaCl concentration resulting in more elongated growth spirals (Fig. 1D). Similar observations have been obtained on the (001) surface of isostructural celestite (Risthaus and others, 2001; Becker and others, 2002).
Dissolution: etch pit formation and monolayer step retreat.—
Similar solvent effects on monolayer step reactivity can also be obtained from in situ dissolution experiments. After exposing a freshly cleaved barite (001) surface to deionized water, relatively few etch pits form (up to 0.1 –0.2 pits/µm2 in 15 min., fig. 1E). Typically, the etch pit morphology is defined by monolayer step edges parallel to and to [010]. In consecutive molecular layers, they have opposite orientations, due to the 21 screw axis normal to the (001) surface. During etch pit formation, dissolution occurs via the retreat of monolayer steps. The lateral spreading rate of etch pits in pure deionized water is 0.08 ± 0.02 nm/s parallel to [010] and 0.03 ± 0.01 nm/s parallel to [100]. In high ionic strength solution (up to 1.0 M NaCl), the step reactivity and the etch pit morphology changes. The lateral spreading rate increases to 1.27 ± 0.03 nm/s parallel to [010] and 0.31 ± 0.03 nm/s parallel to [100]. Consequently, the etch pits are more elongated parallel to [010] and bounded by curved monolayer steps (Fig. 1F). Also, the initial number of etch pits is 1.4 –2.2 pits/µm2 after 5 min., about an order of magnitude higher than in pure water.
Computational Description of Interactions between Monovalent Ions in Solution and Divalent Mineral Surfaces
In order to interpret the influence of Na+ and Cl– in solution on the morphologies and rates of growth and dissolution of sulfate surfaces, one has to analyze the changes in the bulk solution, in the near surface region, and direct interactions between the background electrolyte ions and the surface. Most insight can be gained from molecular simulations to resolve the mechanisms of the latter processes.
Dissolution.—
When we try to develop a picture of these etch pits from a molecular point of view of a pure BaSO4 (001) surface, we obtain an etch pit as shown in figure 3. Figure 3 shows an etch pit that is bounded by step directions as in the experimental AFM image shown in figure 1E. The step directions are bounded by periodic bond chains (alternating Ba2+ and SO42- ions) with no dipole moment perpendicular to the step, both first indications for a relative stable step (step energy 0.31 eV/Å). In contrast, the third step direction || 
would be either bounded by SO42– or Ba2+ ions, resulting in a dipole moment perpendicular to the step and an unfavorable step energy (1.33 eV/Å). In addition to the difference in step energies, a polar step is more likely to be attacked by water molecules from solution due to its local charge. One way to solve the problem of a local dipole moment is to "construct" a jagged edge that would represent an etch pit as in figure 4. Figure 4 can be understood as if half of the ions along the previously polar step are removed, which removes the dipole moment perpendicular to the step. Therefore, the step energy (in vacuum) decreases to 0.63 eV/Å in the unrelaxed case and 0.50 eV/Å for the case where the step atoms are allowed to relax. If we stay in our framework of an etch pit that would just consist of a BaSO4 surface, such a step has a high density of kink sites (1.8 kink sites/nm) which makes it more likely for hydration and, thus, for dissolution to occur (Fig. 4).
|
|
steps (0.31 eV/Å). The energy budget of adsorption of Na+ or Cl– to the PBC-like step is much less favorable, such that we can conclude that adsorption of monovalent background ions takes place predominantly on steps that would be polar without the background electrolytes (Fig. 5). Thus, the increased relative stability of steps parallel [010] with respect to is due to the formation of PBC-like -Na-SO4-Na-SO4- or -Cl-Ba-Cl-Ba- chains along the step, which are less polar (and hence more stable) than terraces bounded by either SO42- or Ba2+. It has to be noted that the described adsorption pattern is not a static termination of the step but rather a temporary picture. The decoration of a polar step with Na+ will always be incomplete such that dissolution progresses for undersaturated BaSO4 solutions concurrent with the formation of thermodynamic equilibrium conditions between Na+ and Ba2+ on step sites or between Cl– and SO42-. In addition, even though the adsorption of Na+ (Cl–) to polar steps is more favorable than the adsorption of Ba2+ (SO42-), the divalent ions are energetically downhill on lattice sites of flat terraces or in the bulk.
|
Na+, SO42– Cl–). Further experimental studies will have to verify these findings.
|
|
 | (2) |
|
|
THE INTERACTION OF ORGANIC GROWTH INHIBITORS WITH MINERAL SURFACES |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Traditionally, the adsorption of molecules to mineral surfaces has been indirectly investigated from bulk experiments. Measurements of crystallization rates in the presence of inhibitors have been attributed to the adsorption of molecules to active growth sites on surfaces (Cabrera and Vermilyea, 1958; Sangwal, 1998). Thus, by experimental determination of adsorption isotherms, it is possible to interpret crystal growth behavior in the presence of a wide variety of inorganic and organic additives. The in situ observation of surfaces during growth using AFM in combination with molecular simulations helps us to understand the effect of inhibitors on the thermodynamics and kinetics of crystal growth at the molecular scale (Bosbach and others, 1996, 2002). Here, we describe the inhibiting effect of five polyphosphonic acids, hydroxyethylene diphosphonic acid (HEDP), nitro trimethyl phosphonic acid (NTMP), methylene diphosphonic acid (MDP), amino methylene phosphonic acid (AMP), and sodium phosphonobutane tricarboxylic acid (PBTC, fig. 9).
After the (i) qualitative description of the microtopography of growing surfaces in the presence of the inhibitors (changes in the shape of both growth steps and two-dimensional nuclei, nucleation density, et cetera); (ii) the dependence of growth rates of monomolecular steps on inhibitor concentration and adsorption isotherms was analyzed, which allows us to quantitatively compare the effectiveness of the inhibitors. Next, (iii) the molecular modeling and energy calculations of the interaction of the five phosphonic acids with the barite (001) face were used to study the role of different surface sites as potential adsorption sites.
Finally, in order to obtain a general view of the adsorption process and its inhibiting effectiveness, the microtopographic observations and the quantitative information provided by both adsorption isotherms and growth rates of monomolecular steps are discussed, together with molecular modeling of the inhibitor-crystal interface.
First we follow the same approach as described for the adsorption of monovalent ions to polar steps. For the interaction between HEDP and the barite surface, we derived empirical potentials from quantum mechanical calculations using molecular fragments (Fig. 10) and the program Gaussian (Frisch and others, 1998) as described in Becker and others (2002). One can compare the adsorption of one SO42- or two Cl– ions with the adsorption of H2-HEDP2- to the same steps. A sulfate ion in the "jagged step" picture (analogous to fig. 4 but with sulfate bounding) can either be replaced by two chloride ions or one H2-HEDP2- ion. Where substitution of SO42- is still a competitive process, the adsorption of H2-HEDP2- to either [100] or [010] steps or the adsorption to kink sites on steps is much stronger and thus irreversible. This explains why sites in the experimental images with H2-HEDP2- bonded to them are pinned during step advancement (Sangwal, 1998) and no dynamic exchange can take place that leads to the formation of straight, crystallographically oriented steps. Even though concentrations that are typically used for growth inhibitors such as HEDP are in the µmol range and, thus, adsorption occurs in form of single diphosphonate ions adsorbed to steps, the co-adsorption of H2-HEDP2- to neighboring sites stabilizes such an adsorption site even more due to the formation of hydrogen bonds between neighboring diphosphonate ions.
|
|
100 nm along the [100] direction, solutions with the same BaSO4 concentration but containing a range of concentrations (0.5 –60 µmol/l) of HEDP, NTMP, MDP, AMP and PBTC were passed over the crystal surfaces and the growth rates of monomolecular steps on barite (001) faces in the presence of inhibitor were measured.
Molecular simulations of in-vacuo adsorption energies were performed on a barite (001) cluster containing 2484 atoms, that is, 414 BaSO4 formula units (details can also be found in Pina and others, 2004). The cluster was bounded by periodic bond chains (
120
direction and its symmetry equivalents) such that there was no dipole moment perpendicular to the surface and none parallel to the edges of the cluster. This is important in order to avoid surface instabilities and to avoid long-range interactions between the cluster and the adsorbate that are caused by the cluster construction rather than the interaction of the growth inhibitor molecule with a specific surface site. For the calculation of in-vacuo adsorption energies, we used the species AMP–1, AMP2-, MDP2-, HEDP2-, NTMP2-, NTMP4-, PBTC2-, and PBTC4-; the change in hydration energies from the dissolved state to the adsorbed one was calculated subsequently. Molecular mechanics and molecular dynamics calculations with subsequent energy optimizations were used to find absolute absorption minima for each surface site. All calculations were performed using empirical force fields, derived by Allan and others (1993) for interactions within the barite crystal, so-called Universal Force Fields by Rappe and others (1992) for interactions within the inhibitor molecules, and by Becker and others (2002) for interactions between the phosphonates and the crystal surface.
The presence of inhibitor in solution reduces the growth rates and, especially in the case of HEDP and AMP, changes the morphology of the islands. In order to quantify the inhibiting effect of the HEDP, NTMP, MDP, AMP and PBTC molecules, we have measured growth rates along the [100] directions of two-dimensional island growth on the barite (001) face for each inhibitor and for different concentrations. We derived the normalized vi/v0 step velocities (where vi and v0 are the growth rates along [100] with and without inhibitor, respectively) as a function of inhibitor concentration. A rapid decrease in step advancement rates can be found for inhibitor concentrations lower than 10 µmol/l (Pina and others, 2004). After this initial decrease, the inhibiting effect of the phosphonic acids becomes weaker and for further increase in inhibitor concentration, the retardation of the step advancement reaches a plateau with little dependence between growth rate and inhibitor concentration. The extension and slope of the plateaus are different for each phosphonic acid. Only for PBTC, complete inhibition for molecule concentrations higher than 10 µmol/l is observed.
Derivation of Adsorption Isotherms from AFM Growth Experiments
In order to compare the effectiveness of the organic molecules, HEDP, NTMP, MDP, AMP and PBTC, we have represented the step rates and inhibitor concentrations using Langmuir diagrams. In such diagrams, the v0 /(v0–vi) parameter, calculated from growth rates of individual monomolecular steps, is plotted against the inverse of the inhibitor concentration, [Inh]–1. Table 1 shows the approximately linear relationship of v0 /(v0-vi) as a function of [Inh]–1 for the five inhibitors studied (linearity expressed by the respective correlation coefficients). This means that the mechanisms of inhibition within the error of experiment can be satisfactorily described by Langmuir adsorption isotherms:
 | (2) |
|
All inhibitor molecules gain energy by adsorbing to corner sites in a hydrated environment. It is interesting to note that the phosphonates (AMP, MDP, HEDP, and NTMP) prefer the configuration where Ba2+ is located at the obtuse corner whereas PBTC prefers to adsorb to Ba2+ at an acute corner.
For each molecule, there is also a kink site configuration for which adsorption is energetically downhill. Kink sites are likely the most important sites for crystal growth blocking. We found the most energetically favorable kink site/phosphonate bond for NTMP4-. However, it may be more valid to consider all 22 kink-site configurations that we evaluated. For PBTC4-, all adsorption energies in a hydrated environment for binding to any of the kink sites are negative (average is –1.36 eV) whereas this is only the case for about half (12 out of 22) of the adsorption events for NTMP4- (with an average of –0.86 eV). This ratio may even become worse if we consider that kink sites lose most of the hydration energy during adsorption at the site types that we consider. This indicates that whatever the type of kink site, it can be blocked by PBTC4-, but about half the kink sites can continue growing if NTMP4- is the chosen inhibitor.
The fact that the initial adsorption follows the behavior predicted by Langmuir’s model implies that the inhibiting effect of the organic molecules studied in this work on barite is due to the attachment of molecules on active growth sites. Our calculations indicate that these active sites are kink sites located along steps. Adsorption to corners is also exothermic, but on most growth islands, there are many more kink than corner sites. In all cases of kink site adsorption, the adsorption energies are higher than 0.4 eV (
40 kJ/mol). Such high adsorption energies mean that the adsorption has a chemical character, that is, the inhibitor-active site bonds are quite strong (typical energies for physical adsorption are lower than 40 kJ/mol while for a chemical adsorption values up to and even over a hundred kJ/mol – 0.4 to < 1eV/adsorbed inhibitor molecule –are expected; Brixner, 1967).
Other possible positions for adsorption, such as flat terraces or steps without kinks or defects provided positive adsorption energies, indicating that they are negligible as positions for crystal growth blocking. This is in agreement with the observed growth rate-inhibitor concentration plateaus for high concentration of inhibitor. Thus, once all kink sites along the steps on the barite (001) surface are occupied with inhibitor molecules, the effectiveness of the inhibitor reaches a maximum and a further increase in the inhibitor concentration does not lead to a significant decrease in growth rates because no more adsorption positions are available on the crystal surface. Only when the concentration of inhibitor in the solution is very high (above 10 µmol/l or 20 µmol/l), a layer of inhibitor molecules can be formed on the barite surface. However, since the adsorption energies of the inhibitor on terraces are positive, no adsorption can be expected. The observed covering must be, therefore, interpreted as a flocculation of inhibitor molecules on the barite (001) surfaces. Nevertheless, such a layer can prevent the barite growth units reaching the surfaces acting as "physical barrier" for the growth process and completely stop growth.
The strength of the adsorption of inhibitor molecules on kink sites can also be quantified through the affinity constants, k1, in equation (2) (see table 1). These constants are in the same range as those obtained by Amjad (1995) for the adsorption of phosphonates on calcium phosphate from bulk experiments (Amjad, 1995, reported k1 values of 8 · 10–6, 17 · 10–6 and 31.2 · 10–6 for PBTC, HEDP, and AMP, respectively). By considering both the ranking of our measured constants and the behavior of the inhibitors for high concentrations, our data suggest the following order of inhibiting effectiveness for the phosphonates: PBTC > NTMP > MDP > HEDP >>AMP. This ranking is consistent with the ranking obtained by Amjad (1995): PBTC >HEDP >AMP (Amjad, 1995, did not study the behavior of MDP and NTMP).
Another interesting aspect of the inhibiting phenomenon is the change in the shape of the islands (see last column in table 1). Our AFM observations indicate that, while in the presence of HEDP and AMP, barite islands become irregular after a short time, the other inhibitors tested do not promote such pronounced changes in island shape, which essentially remain with their typical circular sector shape. Since HEDP and AMP are less effective inhibitors, the change in island shape can be attributed to a weaker attachment of these molecules to the active site along steps that can result in a continuous adsorption and desorption of inhibitors on those positions. This would imply that growth positions would be no longer blocked, resulting in an isotropic irregular shape with time. In contrast, PBTC, NTMP and MDP are able to block kink sites along monomolecular steps in a more efficient (and permanent) way. As a consequence, growth is more strongly inhibited but the shape of the islands is preserved.
|
AZOTOBACTIN-GOETHITE/DIASPORE INTERACTIONS |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
In our previous work, we demonstrated that force measurements made using an Atomic Force Microscope (AFM) were useful in examining the siderophore-mineral interactions by providing direct molecular-level quantitative force data (Kendall and Hochella, 2003). Specifically, these data showed that the siderophore azotobactin exhibits a degree of mineral specificity that favors iron over aluminum oxides. Our molecular simulations of azotobactin-diaspore (
-AlOOH) and azotobactin-goethite (-FeOOH) interactions are designed to complement and more fully understand, at least in a qualitative framework, these experimental force measurements, while further evaluating the notion of surface reactivity as a component in siderophore-mediated mineral dissolution. Specific aims include determining which siderophore functional groups attach to the surface and characterizing the nature and geometry of these coordinations. This includes evaluating the question if there are any steric hindrances in forming a multi-bonded complex between the surface metal atom and the siderophore while the latter is adsorbed to the surface. In addition, these calculations were used to discriminate between relatively weak and strong linkages using energy-distance profiles and derived forces. Using these simulated forces, we make a qualitative comparison with experimental forces to see if the same distinction in goethite and diaspore force signatures observed using the AFM can be predicted by the model. Additional separate quantum mechanical calculations on azotobactin fragments bonded to a single Fe or Al atom were also completed to assess the source of specificity (for example, electron affinity) between chelating O atoms and Fe, and to complement the empirical force-field calculations on the whole system by deriving Femineral/Almineral-siderophore specific potentials. The effect of the interaction on the metal in the mineral structure was also examined with a specific question in mind: Is "dissolution" observed under any of the simulated, albeit simplified, conditions? For example, does azotobactin, in contact with the mineral surface in vacuum, or under hydrated conditions simulated with a constant dielectric field or with added water clusters, completely extract a cation from the oxide surface?
Molecular Mechanics Methodology of Siderophore-Oxide Interactions
Simulations were carried out using the Cerius2 computational package (Accelrys, Inc.). In order to develop a set of empirical potentials (for example, Buckingham potentials, Gale, 1998) that mimic the physico-chemical properties of goethite and its isostructural Al-equivalent diaspore, we used the program package GULP (Gale, 1998). The interactions within the siderophore molecule are simulated by using the Universal Force Field (UFF) (Rappe and others, 1992) in combination with Coulomb potentials derived by using the Qeq charge equilibration scheme (Rappe and others, 1992). We derived potentials for the interactions between the chelating functional groups of the siderophore and the metal ions on the surface (Fe, Al) using a quantum mechanical approach. Details on the computational methods can be found in Kendall and others (2005).
Hydrated environments of siderophore-metal fragments were modeled as a continuum of uniform dielectric constant using Tomasi’s Polarized Continuum Model (PCM) (Miertus and Tomasi, 1982) on quantum mechanically optimized vacuum structures. It should be noted that the PCM of solvation is somewhat inaccurate, especially for small ions. However, use of the model on larger organic fragments results in comparatively lower errors and, thus, provides a sufficient approximation. In addition, fragments of the siderophore with two functional groups and their Fe/Al chelates were calculated as a hydrated complex using a quantum mechanical approach (B3LYP, HF/DFT hybrid, Becke-3-parameter-Lee-Yang-Parr, Lee and others, 1988; Becke, 1993) with a 6-31g Pople-type basis set, by completing the hydration environment of the metal cation with four water molecules.
Methodology of Azotobactin and Azotobactin-Mineral Molecular Mechanics Simulations
Most of the molecular mechanics simulations were performed using vacuum conditions (dielectric constant 
= 1) to reduce computation time; however, solvated conditions in selected runs were simulated by increasing the dielectric constant ( = 80) or adding clusters of individual water molecules to the system. The azotobactin molecule was constructed based on the structure provided by Palanche and others (1999) (Fig. 12). To facilitate discussion of the siderophore structure, each O atom has been numbered according to figure 12. This numbering system is referred to when describing the O atoms that coordinate and release from each surface or the free Fe3+ ion. Coordination or bonding to the surface was defined as when the Fe/Al-O (siderophore oxygen) distance is less than 2.2 Å. For larger distances, the combination of Lennard-Jones and Coulomb potentials, as implemented by the Universal Force Field is in good agreement with our quantum mechanical results.
|
|
After the stretching of the molecule, and prior to simulating the movement of the molecule toward the mineral surface, the hydroxamate methyl carbon was released, and allowed to move freely during minimizations. The Au atom remained fixed and AFM extension (approach) or retraction motion was modeled by decreasing or increasing, respectively, the distance between the gold atom and the mineral surface. The molecule was advanced in 0.3 to 1.5 Å steps during extension and 0.5 to 1.5 Å steps upon retraction. Note the vertical spatial resolution of the piezoelectric stage used in the AFM force measurements (Kendall and Hochella, 2003) is 1 Å. An optimization was completed after each step and the overall energy of the system plus interatomic distances between siderophore O atoms and surface bound metals were recorded. In addition, molecular dynamics (MD) calculations were performed, where the system was ramped to a temperature of 300K followed by anneal dynamics (annealing cycles between 100K and 300K) and a subsequent energy minimization.
Results from the Simulation of Siderophore-Oxide Interactions
Under the modeling conditions, the simulated force/energy distance curves can be separated into a number of events during approach/retraction:
Simulated Approach Force Trace - Azotobactin and Goethite/Diaspore
Upon approach of the azotobactin molecule to both the goethite and diaspore from a Au-surface separation of 63 Å to 10Å, the siderophore O atoms coordinate with metals on the surface in a sequence that is consistent with their relative position on the peptide backbone: the C-terminal groups (for example, the homoserine (Hse) lactone and hydroxamate moieties) coordinate first followed by groups closer to the ring structure at the N-terminus. This is expected because each amino acid was aligned in order upon stretching the molecule to a maximum contour length prior to the surface approach. Although it was not fixed and atomic motion was permitted, the linker molecule remained extended and rigid throughout the simulation. This is due to the fairly rigid bond and dihedral angles within the aliphatic linker molecules. As a result, the linker angles away from the axis of the extension and retraction, serving as a guide to lay the azotobactin molecule along its length onto the surface. The extension trace for each mineral is similar, with the diaspore curve essentially shifted down in relative energy scale due to the lowered lattice energy associated with diaspore and differences in the definition of the metal-oxygen interaction potentials associated with each mineral (Fig. 14). All O atoms in both the peptide backbone and amino acid side chains had the potential to coordinate with metals in the surface. However, steric constraints imposed by the siderophore structure dictated which ones came close enough to ultimately interact. These included (in order of coordination to the surface): O1/O2 (hydroxamate), O4/O5 (Hse lactone), O3/O7 (backbone carbonyl O atom and Hse), O11 (Ser hydroxyl), O13/O14 (ß-hydroxyacid) (see also fig. 15). The slashes ("/") in the previous sentence indicate attachment of the two O atoms in the chelating group was approximately simultaneous.
|
|
35 Å Au-surface separation, the linker folds the azotobactin molecule back upon itself with the ß hydroxyacid (O13/O14) serving as a hinge point. While this behavior may be unique to the presence of the linker, and not necessarily reflective of a surface conformation typical in nature, it is interesting to note that a similar hydroxyacid hinge point position is observed in the proposed azotobactin-Fe(III) aqueous complex structure (see fig. 13). A second extension simulation was completed with a lateral component added to the downward motion. This was done to avoid generation of the hinge point and to stretch and lay the upper part of the molecule onto the mineral, thereby giving groups that are above the ß hydroxyacid an opportunity to interact with the surface. Additional coordination of groups closer to the Au atom (where the tip is) was, however, not observed. Aside from the unique structural constraints imposed for the purposes of simulating the AFM conditions, it is not surprising that some of the possible charge transfer groups, including the oxygens that participate in aqueous chelation of iron (that is, catechol), do not interact with the surface. A similar observation has been made when the siderophore desferrioxamine (DFO) interacts with goethite at circumneutral pH and ionic strengths similar to our experimental AFM conditions (Kraemer and others, 1999; Cocozza and others, 2002). DFO is a smaller ligand relative to azotobactin, and employs three hydroxamate groups linked by a methyl backbone to form a hexadentate aqueous complex with iron. However, when releasing iron from a solid form, multiple kinetic and adsorption experiments show that DFO has a pseudo-first order rate coefficient and an enthalpy/entropy relationship comparable to the single-hydroxamate ligand acetohydroxamic acid (aHA) (C2H5NO2) (Holmen and Casey, 1996; Kraemer and others, 1999; Cocozza and others, 2002). This suggests a similar goethite dissolution mechanism for DFO and aHA, and, thus implies only one or two of the three DFO hydroxamate groups interact with the surface (Cocozza and others, 2002). Holmen and Casey (1996) postulate a reduction in the ligand’s conformational freedom when proximal to the surface, the hydrophobicity of the DFO backbone, and the goethite structure all play a role in limiting the number of groups that can participate in surface reactions with the iron centers. As discussed above, our model also suggests that many of these factors are indeed relevant when considering the interaction of azotobactin with the surface.
Significant gains in energy are calculated upon the coordination of each azotobactin group (Fig. 14) with relatively large changes associated with the attachment of the hydroxyacid and the C-terminal Hse (lactone) (O4/O5). Relatively smaller gains in energy are associated with protonated groups attaching to the surface (for example, O7), suggesting a weaker, hydrogen bond component may be important in these interactions. The final energy of the goethite system at a Au-surface separation of 10.8 Å is –7547.4 kJ/mol, and the derivative of the energy profile with respect to distance showed a maximum force of 923 pN. This force peak, qualitatively equivalent to an experimental jump to contact force (for example, the force at which the AFM cantilever jumps into contact with the surface; see Kendall and Lower, 2004), is associated with the docking of the Hse lactone group in our model. It may be tenuous to compare these results directly to AFM data collected in solution; however, it is worth noting that the modeled force magnitude and distance at which this jump occurred (52 Å) compared favorably with averaged experimental AFM data values (at both pH 3.5 and pH 7) collected with relatively high spring constants (ks = 0.123 N/m). This was not the case with other data sets collected with softer (less stiff) cantilevers (ks 0.06 N/m) that were perhaps less adept at capturing the steep gradient generated by the Hse lactone jump in the energy profile (Cappella and Dietler, 1999; Kendall and Lower, 2004). Here the lower spring constant resulted in a longer jump to contact distance because the onset of the jump was already able to bend the cantilever.
Other factors should be considered when making this semi-quantitative comparison of modeled forces with AFM data. The simulated value is representative of a single azotobactin molecule interacting with the surface. Based on dimensional considerations and a Johnson-Kendall-Roberts model of adhesion, the capture of a single molecule interaction was not predicted for the prior AFM experiments (Kendall and Hochella, 2003). Also, the simulation was completed in a vacuum and the jump to contact energies and distances are expected to be lower in the presence of water due to charge shielding. An assessment of the additive effect of multiple interactions (or a reassessment of the number of interactions captured in the AFM data), together with a characterization of the influence of a solvated environment may help to simulate the AFM experiments more closely. The difference between such an approach and the current study would provide more information on solvation and hydrogen bonding interactions.
Simulated Azotobactin - Free Fe3+ Interaction
The bonding sequence during extension towards the goethite surface is different from the bonding order associated with the interaction of the azotobactin-linker construct with a single ferric iron atom. This emphasizes the different roles that each functional group may play depending if the siderophore encounters an Fe atom in a solid or aqueous form. As the free ferric iron is moved into the siderophore structure parallel to its extended peptide backbone axis, the following attachment and release sequence is revealed (O atoms separated by commas indicate two discrete events): O1 attach, O2 attach, O4/O5 attach, O7 attach, O5 release, O8/O9 attach, O4 release. Again, the hydroxamate (O1, O2) is shown to participate in the coordination. However, steric hindrances preclude attachment of the catechol and the ß-hydroxyacid groups (O13/O14), even though the ferric iron was advanced to a separation of 8 Å from the gold atom. The expected hexadentate (hydroxamate, catechol, hydroxyacid) coordination that is suggested for the stable azotobactin (Azb) -Fe(III)aq complex is not observed, probably as a result of the direction and geometry of the advance of the iron into the structure and the reduction in the degrees of conformational freedom imposed by the linkage. Note that this protocol was employed to facilitate comparison with experimental and simulated force experiments. The presence of the available uncoordinated chelating groups in the model containing free iron are, however, in agreement with the experimental AFM data on the azotobactin-goethite interaction in Fe-rich solutions. During AFM experiments, the azotobactin-goethite interaction was probed after successively adding soluble free Fe3+ ion (Kendall and Hochella, 2003). A sharp decrease in adhesion forces with increasing [Fe(III)aq] was followed by a plateau of residual
adhesion that existed in spite of sufficient iron to satisfy the chelating groups associated with all of the azotobactin molecules attached to the tip, cantilever and substrate (Kendall and Hochella, 2003). The source of the residual adhesion was hypothesized to be free chelating groups or other sidechain or backbone O atoms interacting with the surface; an idea that is now supported by the conformation suggested in the azotobactin-Fe(III) model.
Simulated Retraction Force Trace - Azotobactin and Goethite/Diaspore
Prior to retracting the azotobactin from the surface, the metals in each mineral lattice in close association (<2.2 Å away) with siderophore O atoms were identified (8 total for goethite and diaspore) and marked with distance monitors (these distance monitors have no direct influence on the calculations; rather they help to describe the metalmineral-Osiderophore interaction). These metal atoms were allowed to relax and a Morse potential was defined between the metal and the coordinating siderophore O atom. Because no additional short-range interaction of siderophore groups was observed at Au-surface separation distances of less than 36 Å during approach, simulation time was reduced by beginning the retractions at this point. Metals are removed from both minerals by the siderophore in the simulated retraction.
Goethite.— During retraction from the goethite, one Fe atom is completely removed from the structure by the terminal Hse group (O4/O5) and 4 others are displaced significantly from their equilibrium position on the surface (for example, Fe-Ogoethite distance exceeds >2.5Å, fig. 16) before snapping back to the lattice. With the exception of a sharp increase followed by a sharp decrease in the Fe atom displacement associated with ß hydroxyacid O14, the displacement magnitudes steadily increase as the overall accumulated stress is transferred from one functional group to the next (data not shown). This trend parallels the energy profile, which is characterized by a large initial buildup, followed by a steady rise in energy with periodic spikes correlating to the release of specific functional groups (some with a metal attached, fig. 15B). As expected, the order of release from the surface is close to the reverse of the extension attachment sequence, with differences in the order of release of the C-terminal groups. The release sequence is as follows: O13, O14, O11, O7/O3, O2, O1, O4/O5. Interestingly, the Hse OH (O7) and a component of a backbone carbonyl (O3) are coupled in their release, in spite of a separation of almost a nanometer in contour length (for example, the maximum end-to-end distance of the extended, linearized azotobactin molecule). Conversely, a lack of coupling is observed between the hydroxamate and hydroxyacid oxygen pairs separated by 1 to 2 Å.
|
The combination of quantum mechanical and empirical force-field simulations show that even though only two functional groups at a time can chelate a metal atom at the surface (in contrast to six as in the case of siderophore-metal complexation in solution), it is possible that despite the steric hindrance, Fe3+ ions can be removed from the goethite surface (Kendall and others, 2005).
Siderophore Affinity for Fe3+ over Al3+
The specificity and increased affinity of azotobactin for the Fe atom in goethite over Al in diaspore that is evident in force measurements and in the molecular mechanics simulations (for example, increased energy changes upon retraction) is corroborated by the binding energies calculated from first principles. With both siderophore analogs, the acetohydroxamic acid and the hydroxycarboxylic acid, binding energies are more negative for Fe3+ compared to Al3+. The absolute ab initio values show the same relationship as energies calculated from experimentally derived formation constants (Kf); however, the latter values are higher (less negative). This is, in part, because the ab initio calculations do not include an entropy component, which would drive the energies down to a less negative value.
Based on previous observations (Hider, 1984; Albrecht-Gary and Crumbliss, 1998; Kendall and Hochella, 2003), it was hypothesized that part of the observed specificity and increased affinity of azotobactin for Fe atoms was a result of iron’s increased electronegativity over Al. Relative comparison of the optimized Mulliken charge distributions from the Gaussian98 models of the Fe and Al ligand complexes supports this argument. The larger ionic radius of Fe and an electronegativity closer to that of oxygen results in an Fe atom with lower positive charge associated with less negative O atoms. In other words, the Fegoethite-Osiderophore bond has a more covalent character than the Aldiaspore-Osiderophore bond. Comparison of the hydration energies associated with the Fe3+ and Al3+ ions coupled with observations made during the ab initio runs present the possibility of surface hydration energies playing a role in the siderophore mineral interaction forces observed with the AFM. While these values have not been measured for goethite and diaspore, it is possible that the increased (more negative) hydration energy associated with the Al3+ ion (versus Fe3+) could correlate with an increased surface hydration energy for diaspore over goethite. Therefore, the lowered siderophore affinity observed for the diaspore surface could also reflect the extra energy required to remove water from the diaspore surface.
|
ADSORPTION OF OLIGOMERS TO STEP EDGES AS A PRECURSOR OF BIOMINERALIZATION |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
4) calcite surface. The lowest energy faces of pure calcite are the (104) family of faces. Our main objective is to find suitable orientation of amino-acid residues in these peptide chains, where these peptide chains align themselves parallel to the calcite surface. The stereochemical relation between the coordination environment of ions in specific crystal faces (Ca2+, CO32- in this case) and the arrangement of ligands (that is peptide residues, oligomers) around ions bound to the surface is a potential factor for organic nucleation and selectivity of biominerals. If the distance between the repeating residue units in the adsorbent matches the distance between the repeating units in the adsorbate surface, the adsorbent long chain polymer or oligomeric organic compound can lie parallel to the surface step. To start with, we have performed calculations of various sequences of small chain peptide residues along non-polar (periodic bond chains of alternating Ca2+ and CO32- ions with no dipole moment perpendicular to the step) and on polar surface steps (steps bounded by either Ca2+ or CO32- ions). Each peptide residue was composed of 3 amino acids. The maximum adsorption energy that we found was for (phe-leu-lys)4-, where the total adsorption energy to a non-polar calcite surface step is –1.071 eV [average of –0.357 eV/(amino acid unit), negative values = adsorption; positive = repulsion]. Adsorption energy values for other peptide chains vary between –0.1879 eV/(amino acid) to –0.2890 eV/(amino acid). Almost parallel alignment of these 3-amino acid peptide residues is observed when aligned with polar Ca2+-bounded surface steps where the dominant interaction between the negatively charged peptide backbone and Ca2+ along the surface step is of electrostatic nature. Adsorption energy values in this case range between –0.1733 eV/(amino acid) to –0.2602 eV/(amino acid). The most energetically favorable adsorption is obtained for the sequence (phe-leu-lys)4- with –0.2978 eV/amino acid.
Using the same concept, we have used longer peptide chains having 12 amino acids. Sequencing of amino acids has been obtained from encoded GPA, a calcium binding protein in the coccolithophorid Emiliania Huxleyi (Corstjens and others, 1998). Interaction of this 12-amino acid long peptide chain in both acidic (protonated) and alkaline (deprotonated) conditions with polar calcite steps for both cases (Ca2+ and CO3– bounding) has been studied. At alkaline conditions, the adsorption energy of the negatively charged peptide with the Ca2+-bounded step edge is –0.09824 eV/amino acid. When the peptide residue is neutral (acidic condition), the adsorption energy is –0.1978 eV/ amino acid residue (CO32- at step edge). In this long chain peptide, amino acid residues at the middle part of the chain are closer to the surface than the end-members giving a U shape to the peptide chain (Fig. 17A). Better parallel alignment of the peptide chain to the calcite surface is observed in the neutral peptide than for the –13 charged peptide residue. This can be owed to the fact that in the negatively charged long chain peptide, electrostatic interactions between charged side chains of the amino acids and the peptide backbone coupled with steric hindrance of large side chain prevents parallel alignment. It is also known that proteins can inhibit calcite growth and that ability is attributed to backbone flexibility of peptide chain (Gerbaud and others, 2000), which may be another reason why we see flexible, not parallel, peptide chains on these polar steps.
|
|
SUMMARY AND CONCLUSIONS |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The Influence of Monovalent Ions on the Growth/Dissolution Kinetics and Morphology of Divalent Crystals
Monovalent ions tend to stabilize previously polar steps on the surfaces of divalent crystals. Consequently, the morphologies of microscopic surface features such as growth islands and spirals, and etch pits are changed, ultimately influencing crystal habits.
Monovalent ions can increase both growth and dissolution rates by modifying the rate-controlling steps of growth and dissolution. For growth, this is the initialization of a new growth row. For dissolution, this is the formation of a kink site along the edge of an etch pit.
The latter point is another example that growth and dissolution are not necessarily reverse processes of each other.
Growth-Inhibitor-Sulfate Interactions
Direct AFM measurements of growth rates of barite steps in the presence of different concentrations of inhibitors allowed us to construct adsorption isotherms indicating a Langmuir-type of behavior for all inhibitors used. A linear relationship exists between v0 /(v0-vi) and [Inh]–1 for inhibitor concentrations lower than 10 µmol/l. Since such isotherms have been constructed from growth rates measured in situ on monomolecular barite steps, they represent a direct demonstration of the molecular basis of the Langmuir model for adsorption processes. Slopes of the adsorption isotherms were calculated from AFM experiments. Their values are similar to values obtained conducting classical bulk experiments. This
demonstrates that AFM can be used to obtain quantitative adsorption data.
Calculations conducted using molecular modeling methods showed that the five studied phosphonates are only effective as growth inhibitors by blocking kink sites along monomolecular steps. Adsorption positions on terraces cannot be considered as possible inhibition sites due to their positive adsorption energies. This is in agreement with AFM observations and measurements. Calculated adsorption energies of the five phosphonates on kink sites are in the order of hundreds of kJ/mol, indicating chemisorption.
From both the growth rate versus inhibitor concentration data and the calculated slope of the adsorption isotherms (that is, affinity constant), we can give the following ranking of inhibitor effectiveness: PBTC > NTMP > MDP > HEDP >>AMP. This ranking of inhibiting effectiveness is consistent with previous experimental works.
Siderophore-Oxide Interactions
Molecular and quantum mechanical calculations are effective in providing qualitative and quantitative information on the azotobactin-metal oxide interaction that is captured during force measurements on the same system. Computer simulations of the AFM extension force traces show that both backbone and sidechain azotobactin oxygens have the opportunity to interact with the diaspore and goethite mineral surfaces; however, steric considerations and constraints imposed by the presence of the mineral lattice dictate which ones ultimately coordinate. Similar functional groups are associated with each mineral surface after extension; however, upon retraction of the molecule from these surfaces, the energy profiles of these pull-off events for diaspore and goethite are distinct. Quantum mechanical calculations of Fe3+/Al3+ chelates including hydration energies provide evidence that the ligand-surface interaction energetically favors the iron oxide. Metal removal is predicted in the models completed in vacuum; and additional simulations, ones that employ a bidentate, mononuclear coordination (as opposed to a bidentate, binuclear coordination) or include water to act as a helper ligand confirm that siderophore metal removal via direct surface contact is possible in a natural system. In spite of the observed surface affinity, the ability of azotobactin to compete with smaller organic ligands (for example, oxalate) for sorption sites on the mineral remains to be determined. Future work could include sorption experiments coupled with force measurements to determine the effect of smaller ligands on azotobactin surface affinity and adhesion.
Steric hindrances limit the number of oxygens that coordinate with the surface to eight; however, for siderophore-oxide interactions in a natural system without the linkage to the AFM tip imposing limitations on molecular conformation, this number is expected to be higher. Azotobactin adsorption geometry is initially binuclear, but increased surface coordination (for example, a higher Osiderophore:Fe ratio associated with the surface complex) is likely upon displacement of the metal by the ligand or with changes in surface microtopography. One example of the latter is an increased coordination that results from siderophore oxygens accessing undercoordinated metals in surface sites associated with step edges. In both cases, increases in coordination result in a more stable configuration and stronger Fe-Osiderophore bond.
Upon release from the surface (that is, due to thermal motion), the Fe-Osiderophore bond(s) persist, allowing metal removal from the lattice. The energetics of the system is then minimized as the ligand reconfigures to enclose the iron in a multidentate coordination, and the defect site in the mineral lattice relaxes and becomes hydrated. A similar dissolution pathway is predicted for the diaspore; however, azotobactin specificity for the iron oxide surface over the aluminum oxide surface is observed in this study and in the AFM measurements (Kendall and Hochella, 2003). With both minerals present, this suggests the pathway will favor the iron oxide surface and will allow efficient azotobactin-mediated Fe(III) acquisition directly from the solid iron form.
Polypeptide-Calcite Interactions
Even though we are still beginning to evaluate the plethora of possibilities to let polypeptide chains interact with specific steps on calcite surfaces, we are starting to understand certain principles, for example steric hindrances caused by "bulky" sidechains. Future work will be directed to use adsorption energy calculations in a genetics-like way to find optimized polypeptide sequences for specific calcite (or other biomineral) faces and step directions.
|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTSELECTIVE ATTACHMENT OF...THE INTERACTION OF ORGANIC...AZOTOBACTIN-GOETHITE/DIASPORE...ADSORPTION OF OLIGOMERS TO...SUMMARY AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Albrecht-Gary A. M., and Crumbliss, A. L., 1998, Coordination chemistry of siderophores: Thermodynamics and kinetics of iron chelation and release, in Siegel, A., and Sigel, H., editors, Metal Ions in Biological Systems: Iron Transport and Storage in Microorganisms, Plants and Animals, v. 35: New York, Marcel Dekker, p. 239–327.
Allan N. L., Rohl, A. L., Gay, D. H., Catlow, C. R. A., Davey, R. J., and Mackrodt, W. C., 1993, Calculated bulk and surface properties of sulfates: Faraday Discussions, v. 95, p. 273–280.[CrossRef]
Amjad Z., editor, 1995, Mineral Scale Formation and Inhibition: New York, Plenum Press, 354 p.
Archibald D. D., Gaber, B. P., Hopwood, J. D., Mann, S., and Boland, T., 1997, Atomic force microscopy of synthetic barite microcrystals: Journal of Crystal Growth, v. 172, p. 231–248.[CrossRef]
Becke A. D., 1993, Density-functional thermochemistry. III. The role of exact exchange: Journal of Chemical Physics, v. 98, p. 5648–5652.[CrossRef]
Becker U., and Gasharova, B., 2001, AFM observations and simulations of jarosite growth at the molecular scale: probing the basis for the incorporation of foreign ions into jarosite as a storage mineral: Physics and Chemistry of Minerals, v. 28, p. 545–556.[GeoRef]
Becker U., Bosbach, D., and Pina, C., 1998, Dynamic simulation of crystal growth at the molecular scale: Toronto, Canada, International Mineralogical Association, 17th General Meeting, p. A92.
Becker U., Risthaus, P., Bosbach, D., and Putnis, A., 2002, Selective attachment of monovalent background electrolyte ions and growth inhibitors to polar steps on sulfates as studied by molecular simulations and AFM observations: Molecular Simulation, v. 28, p. 607–632.[CrossRef]
Black S. N., Bromley, L. A., Cottier, D., Davey, R. J., Dobbs, B., and Rout, J. E., 1991, Interactions at the organic inorganic interface - Binding motifs for phosphonates at the surface of barite crystals: Journal of the Chemical Society-Faraday Transactions, v. 87, p. 3409–3414.
Blount C. W., 1977, Barite solubilities and thermodynamic quantities up to 300°C and 1400 bars: American Mineralogist, v. 62, p. 942–957.[Abstract]
Bosbach D., 2002, Linking molecular scale barite precipitation mechanisms with macroscopic reaction rates, in Hellmann, R., and Wood, S. A., editors, Water-Rock Interactions, Ore Deposits, and Environmental Geochemistry: A Tribute to David A. Crerar: St. Louis, The Geochemical Society, Special Publication No. 7, p. 97–110.
Bosbach D., and Hochella, M. F., Jr., 1996, Gypsum growth in the presence of growth inhibitors: A scanning force microscopy study: Chemical Geology, v. 132, p. 227–236.[CrossRef][Web of Science][GeoRef]
Bosbach D., Hall, C., and Putnis, A., 1998, Mineral precipitation and dissolution in aqueous solution: in situ microscopic observations on barite (001) with atomic force microscopy: Chemical Geology, v. 151, p. 143–160.[CrossRef][Web of Science][GeoRef]
Bosbach D., Coveney, P. V., Griffin, J. L. W., Putnis, A., Risthaus, P., Stackhouse, S., and Whiting, A., 2002, The rational design, synthesis and demonstration of the recognition and binding of a diaza-dioxa-12-crown-4 diphosphonate macrocycle to all crystal growth faces of barium sulfate: Journal of the Chemical Society, Perkin Transactions 2, p. 1238–1245.
Brixner L. H., 1967, Segregation coefficients of rare earth niobates in CaMoO4: Journal of the Electrochemical Society, v. 114, p. 108–109.
Busenberg E., Plummer, L. N., and Parker, V. B., 1984, The solubility of strontianite (SrCO3) in CO2-H2O solutions between 2 and 91°C, the association constants of SrHCO3+ (aq) and SrCO3O (aq) between 5 and 80°C, and an evaluation of the thermodynamic properties of Sr2+ (aq) and SrCO3(cr) at 25°C and 1 atm total pressure: Geochimica et Cosmochimica Acta, v. 48, p. 2021–2035.[CrossRef][Web of Science][GeoRef]
Busing W. R., and Levy, H. A., 1958, A single crystal neutron diffraction study of diaspore, AlO(OH): Acta Crystallographica, v. 11, p. 798–803.[CrossRef][Web of Science][GeoRef]
Cabrera N., 1953, Macroscopic spirals and the dislocation theory of crystal growth: Journal of Chemical Physics, v. 21, p. 1111–1112.[CrossRef]
Cabrera N., and Vermilyea, D. A., 1958, The growth of crystals from solution, in Doremus, R. H., Roberts, B. W., and Turnbull, D., editors, Growth and Perfection of Crystals: London, Chapman and Hall, p. 393–410.
Cappella B., and Dietler, G., 1999, Force-distance curves by atomic force microscopy: Surface Science Reports, v. 34, p. 1–2.
Cheah S. F., Kraemer, S. M., Cervini-Silva, J., and Sposito, G., 2003, Steady-state dissolution kinetics of goethite in the presence of desferrioxamine B and oxalate ligands: implications for the microbial acquisition of iron: Chemical Geology, v. 198, p. 63–75.[GeoRef]
Christy A. G., and Putnis, A., 1993, The kinetics of barite dissolution and precipitation in water and sodium-chloride brines at 44-85°C: Geochimica et Cosmochimica Acta, v. 57, p. 2161–2168.[GeoRef]
Cocozza C., Tsao, C. C. G., Cheah, S. F., Kraemer, S. M., Raymond, K. N., Miano, T. M., and Sposito, G., 2002, Temperature dependence of goethite dissolution promoted by trihydroxamate siderophores: Geochimica et Cosmochimica Acta, v. 66, p. 431–438.[GeoRef]
Cooper T. G., and De Leeuw, N. H., 2002, Adsorption of methanoic acid onto the low-index surfaces of calcite and aragonite: Molecular Simulation, v. 28, p. 539–556.[CrossRef]
Cornell R. M., and Schwertmann, U., 1996, The Iron Oxides: Structure, Properties, Reactions, Occurrence and Uses: Zürich, VCH Verlagsgesellschaft, 573 p.
Corstjens P., van der Kooij, A., Linschooten, C., Brouwers, G. J., Westbroek, P., and de Vrind-de Jong, E. W., 1998, GPA, a calcium-binding protein in the coccolithophorid Emiliania Huxleyi (Prymnesiophyceae): Journal of Phycology, v. 34, p. 622–630.[CrossRef][Web of Science]
Coveney P. V., Davey, R. J., Griffin, J. L. W., and Whiting, A., 1998, Molecular design and testing of organophosphonates for inhibition of crystallisation of ettringite and cement hydration: Chemical Communications, v. 14, p. 1467–1468.[CrossRef]
Coveney P. V., Davey, R., Griffin, J. L. W., He, Y., Hamlin, J. D., Stackhouse, S., and Whiting, A., 2000, A new design strategy for molecular recognition in heterogeneous systems: A universal crystal-face growth inhibitor for barium sulfate: Journal of the American Chemical Society, v. 122, p. 11557–11558.[CrossRef]
Davey R. J., Black, S. N., Bromley, L. A., Cottier, D., Dobbs, B., and Rout, J. E., 1991, Molecular design based on recognition at inorganic surfaces: Nature, v. 353, p. 549–550.[CrossRef]
De Leeuw N. H., and Cooper, T. G., 2004, A computer modeling study of the inhibiting effect of organic adsorbates on calcite crystal growth: Crystal Growth and Design, v. 4, p. 123–133.[CrossRef]
De Leeuw N. H., and Parker, S. C., 1997, Atomistic simulation of the effect of molecular adsorption of water on the surface structure and energies of calcite surfaces: Journal of the Chemical Society-Faraday Transactions, v. 93, p. 467–475.
De Leeuw N. H., Parker, S. C., and Harding, J. H., 1999, Molecular dynamics simulation of crystal dissolution from calcite steps: Physical Review B, v. 60, p. 13792–13799.
Dove P. M., and Czank, C. A., 1995, Crystal-chemical controls on the dissolution kinetics of the isostructural sulfates - celestite, anglesite, and barite: Geochimica et Cosmochimica Acta, v. 59, p. 1907–1915.[GeoRef]
Dunn K., and Yen, T. F., 1999, Dissolution of barium sulfate scale deposits by chelating agents: Environmental Science and Technology, v. 33, p. 2821–2824.[CrossRef]
Frisch M. J., Tomasi, J., Pople, J., and Gaussian, I., 1998, Gaussian98: Pittsburgh, Pennsylvania.
Gale J., 1998, GULP (General Utility Lattice Program): London, The Royal Institution.
Gerbaud V., Pignol, D., Loret, E., Bertrand, J. A., Berland, Y., Fontecilla-Camps, J. C., Canselier, J. P., Gabas, N., and Verdier, J. M., 2000, Mechanism of calcite crystal growth inhibition by the N-terminal undecapeptide of lithostathine: Journal of Biological Chemistry, v. 275, p. 1057–1064.
Gualtieri A. F., and Venturelli, P., 1999, In situ study of the goethite-hematite phase transformation by real time synchrotron powder diffraction: American Mineralogist, v. 84, p. 895–904.[Abstract]
Hansen D. C., McCafferty, E., Lins, C. W., and Fitzpatrick, J. J., 1995, An FT-IR investigation of parabactin adsorbed onto aluminum: Applied Surface Science, v. 84, p. 85–90.[CrossRef]
Hay B. P., and Rustad, J. R., 1994, Structural criteria for the rational design of Selective Ligands - Extension of the MM3 force-field to aliphatic ether Complexes of the alkali and alkaline-Earth cations: Journal of the American Chemical Society, v. 116, p. 6316–6326.[CrossRef]
Hay B. P., Zhang, D. L., and Rustad, J. R., 1996, Structural criteria for the rational design of selective ligands. 2. Effect of alkyl substitution on metal ion complex stability with ligands bearing ethylene-bridged ether donors: Inorganic Chemistry, v. 35, p. 2650–2658.[Medline]
Hay B. P., Dixon, D. A., Vargas, R., Garza, J., and Raymond, K. N., 2001, Structural criteria for the rational design of selective ligands. 3. Quantitative structure-stability relationship for iron(III) complexation by tris-catecholamide siderophores: Inorganic Chemistry, v. 40, p. 3922–3935.[CrossRef][Medline]
He S. L., Oddo, J. E., and Tomson, M. B., 1995, The nucleation kinetics of barium-sulfate in NaCl solutions up to 6 M and 90°C: Journal of Colloid and Interface Science, v. 174, p. 319–326.[CrossRef]
Hersman L., Lloyd, T., and Sposito, G., 1995, Siderophore-promoted dissolution of hematite: Geochimica et Cosmochimica Acta, v. 59, p. 3327–3330.[CrossRef][Web of Science][GeoRef]
Hersman L., Maurice, P., and Sposito, G., 1996, Iron acquisition from hydrous Fe(III)-oxides by an aerobic Pseudomonas sp: Chemical Geology, v. 132, p. 25–31.[CrossRef][GeoRef]
Hider R. C., 1984, Siderophore mediated adsorption of iron, in Clarke, M. J., Ibers, J. A., Mingos, D. M. P., Palmer, G. A., Sadler, P. J., and Williams, R. J. P., editors, Structure and Bonding: Siderophores from Microorganisms and Plants, v. 58: Heidelberg, Springer-Verlag, p. 25–87.
Higgins S. R., Jordan, G., Eggleston, C. M., and Knauss, K. G., 1998, Dissolution kinetics of the barium sulfate (001) surface by hydrothermal atomic force microscopy: Langmuir, v. 14, p. 4967–4971.[CrossRef]
Holmen B. A., and Casey, W. H., 1996, Hydroxamate ligands, surface chemistry, and the mechanism of ligand-promoted dissolution of goethite -FeOOH(s): Geochimica et Cosmochimica Acta, v. 60, p. 4403–4416.[GeoRef]
Holmen B. A., Tejedor-Tejedor, M. I., and Casey, W. H., 1997, Hydroxamate complexes in solution and at the goethite-water interface: A cylindrical internal reflection Fourier transform infrared spectroscopy study: Langmuir, v. 13, p. 2197–2206.[CrossRef]
Jones F., Clegg, J., Oliveira, A., Rohl, A. L., Ogden, M. I., Parkinson, G. M., Fogg, A. M., and Reyhani, M. M., 2001, Anomalous behaviour within a systematic series of barium sulfate growth modifiers: CrystEngComm, v. 3(40), p. 165–167, DOI: 10.1039/b106729j.[CrossRef]
Jones F., Oliveira, A., Rohl, A. L., Parkinson, G. M., Ogden, M. I., and Reyhani, M. M., 2002, Investigation into the effect of phosphonate inhibitors on barium sulfate precipitation: Journal of Crystal Growth, v. 237, p. 424–429.[CrossRef]
Kalinowski B. E., Liermann, L. J., Brantley, S. L., Barnes, A., and Pantano, C. G., 2000, X-ray photoelectron evidence for bacteria-enhanced dissolution of hornblende: Geochimica et Cosmochimica Acta, v. 64, p. 1331–1343.[CrossRef][Web of Science][GeoRef]
Kendall T. A., and Hochella, M. F., Jr., 2003, Measurement and interpretation of molecular-level forces of interaction between the siderophore azotobactin and mineral surfaces: Geochimica et Cosmochimica Acta, v. 67, p. 3537–3546.[CrossRef][Web of Science][GeoRef]
Kendall T. A., and Lower, S. K., 2004, Forces between minerals and biological surfaces in aqueous solution: Advances in Agronomy, v. 82, p. 1–54.
Kendall T. A., Hochella, M. F., Jr., and Becker, U., 2005, Computational modeling of azotobactin-goethite/diaspore interactions: Applications to molecular force measurements and siderophore-mineral reactivity: Chemical Geology, v. 216, p. 17–35.[GeoRef]
Kraemer S. M., Cheah, S. F., Zapf, R., Xu, J. D., Raymond, K. N., and Sposito, G., 1999, Effect of hydroxamate siderophores on Fe release and Pb(II) adsorption by goethite: Geochimica et Cosmochimica Acta, v. 63, p. 3003–3008.[CrossRef][Web of Science][GeoRef]
Lee C. T., Yang, W. T., and Parr, R. G., 1988, Development of the Colle-Salvetti correlation-energy formula into a functional of the electron-density: Physical Review B, v. 37, p. 785–789.[Medline]
Li M., and Mann, S., 2000, Emergence of morphological complexity in BaSO4 fibers synthesized in AOT microemulsions: Langmuir, v. 16, p. 7088–7094.[CrossRef]
Liermann L. J., Kalinowski, B. E., Brantley, S. L., and Ferry, J. G., 2000, Role of bacterial siderophores in dissolution of hornblende: Geochimica et Cosmochimica Acta, v. 64, p. 587–602.[CrossRef][Web of Science][GeoRef]
Lumetta G. J., Rapko, B. M., Garza, P. A., Hay, B. P., Gilbertson, R. D., Weakley, T. J. R., and Hutchison, J. E., 2002, Deliberate design of ligand architecture yields dramatic enhancement of metal ion affinity: Journal of the American Chemical Society, v. 124, p. 5644–5645.[Medline]
Mann S., 2001, Biomineralization: Oxford, United Kingdom, Oxford University Press, 198 p.
McWhirter M. J., Bremer, P. J., Lamont, I. L., and McQuillan, A. J., 2003, Siderophore-mediated covalent bonding to metal (oxide) surfaces during biofilm initiation by Pseudomonas aeruginosa bacteria: Langmuir, v. 19, p. 3575–3577.[CrossRef]
Miertus S., and Tomasi, J., 1982, Approximate evaluations of the electrostatic free-energy and internal energy changes in solution processes: Chemical Physics, v. 65, p. 239–245.[CrossRef]
Monnin C., and Galinier, C., 1988, The solubility of celestite and barite in electrolyte-solutions and natural waters at 25°C - a thermodynamic study: Chemical Geology, v. 71, p. 283–296.[GeoRef]
Nielsen A. E., and Sohnel, O., 1971, Interfacial tensions electrolyte crystal-aqueous solution, from nucleation data: Journal of Crystal Growth, v. 11, p. 233–234.[CrossRef]
Nygren M. A., Gay, D. H., Catlow, C. R. A., Wilson, M. P., and Rohl, A. L., 1998, Incorporation of growth-inhibiting diphosphates into steps on the calcite cleavage plane surface: Journal of the Chemical Society-Faraday Transactions, v. 94, p. 3685–3693.
Palanche T., Marmolle, F., Abdallah, M. A., Shanzer, A., and Albrecht-Gary, A. M., 1999, Fluorescent siderophore-based chemosensors: iron(III) quantitative determinations: Journal of Biological Inorganic Chemistry, v. 4, p. 188–198.[Medline]
Parker S. C., Titiloye, J. O., and Watson, G. W., 1993, Molecular modeling of carbonate minerals - studies of growth and morphology: Philosophical Transactions of the Royal Society of London Series A-Mathematical Physical and Engineering Sciences, v. 344, p. 37–48.
Pina C. M., Becker, U., Risthaus, P., Bosbach, D., and Putnis, A., 1998a, Molecular-scale mechanisms of crystal growth in barite: Nature, v. 395, p. 483–486.[CrossRef]
Pina C. M., Bosbach, D., Prieto, M., and Putnis, A., 1998b, Microtopography of the barite (0 0 1) face during growth, AFM observations and PBC theory: Journal of Crystal Growth, v. 187, p. 119–125.[CrossRef]
Pina C. M., Putnis, C. V., Becker, U., Biswas, S., Carroll, E. C., Bosbach, D., and Putnis, A., 2004, An atomic force microscopy and molecular simulations study of the inhibition of barite growth by phosphonates: Surface Science, v. 553, p. 61–74.[CrossRef]
Putnis A., Putnis, C. V., and Paul, J. M., 1995, The efficiency of a DTPA-based solvent in the dissolution of barium sulfate scale deposits: San Antonio, Texas, Paper SPE 029094 Proceedings of the 1995 SPE International Symposium on Oilfield Chemistry, p. 773–785.
Qi L. M., Cölfen, H., and Antonietti, M., 2000, Control of barite morphology by double-hydrophilic block copolymers: Chemistry of Materials, v. 12, p. 2392–2403.[CrossRef]
Rappe A. K., Casewit, C. J., Colwell, K. S., Goddard, W. A., and Skiff, W. M., 1992, UFF, a full periodic-table force-field for molecular mechanics and molecular-dynamics simulations: Journal of the American Chemical Society, v. 114, p. 10024–10035.[CrossRef]
Rashin A. A., and Honig, B., 1985, Reevaluation of the Born model of ion hydration: Journal of Physical Chemistry, v. 89, p. 5588–5593.[CrossRef][Web of Science]
Reardon E. J., and Armstrong, D. K., 1987, Celestite (SrSO4(s)) solubility in water, seawater and NaCl solution: Geochimica et Cosmochimica Acta, v. 51, p. 63–72.[GeoRef]
Redfern S. E., and Parker, S. C., 1998, Atomistic simulation of the effects of calcium and strontium defects on the surface structure and stability of BaSO4: Journal of the Chemical Society-Faraday Transactions, v. 94, p. 1947–1952.
Risthaus P., Bosbach, D., Becker, U., and Putnis, A., 2001, Barite scale formation and dissolution at high ionic strength studied with atomic force microscopy: Colloids and Surfaces A-Physicochemical and Engineering Aspects, v. 191, p. 201–214.[CrossRef]
Rohl A. L., Gay, D. H., Davey, R. J., and Catlow, C. R. A., 1996, Interactions at the organic/inorganic interface: Molecular modeling of the interaction between diphosphonates and the surfaces of barite crystals: Journal of the American Chemical Society, v. 118, p. 642–648.[CrossRef]
Sangwal K., 1998, Growth kinetics and surface morphology of crystals grown from solutions: Recent observations and their interpretations: Progress in Crystal Growth and Characterization of Materials, v. 36, p. 163–248.[CrossRef]
Sayle D. C., Catlow, C. R. A., Harding, J. H., Healy, M. J. F., Maicaneanu, S. A., Parker, S. C., Slater, B., and Watson, G. W., 2000, Atomistic simulation methodologies for modelling the nucleation, growth and structure of interfaces: Journal of Materials Chemistry, v. 10, p. 1315–1324.[CrossRef]
Shiraki R., Rock, P. A., and Casey, W. H., 2000, Dissolution kinetics of calcite in 0.1 m NaCl solution at room temperature: An atomic force microscopic (AFM) study: Aquatic Geochemistry, v. 6, p. 87–108.[CrossRef][Web of Science][GeoRef]
Telford J. R., and Raymond, K. N., 1996, Siderophores: Comprehensive Supramolecular Chemistry, v. 1, p. 245–266.
Van der Leeden M. C., and van Rosmalen, G. M., 1995, Adsorption behavior of polyelectrolytes on barium-sulfate crystals: Journal of Colloid and Interface Science, v. 171, p. 142–149.[CrossRef]
Wang K. S., Resch, R., Dunn, K., Shuler, P., Tang, Y. C., Koel, B. E., and Yen, T. F., 1999a, Dissolution of the barite (001) surface by the chelating agent DTPA as studied with non-contact atomic force microscopy: Colloids and Surfaces A-Physicochemical and Engineering Aspects, v. 160, p. 217–227.[CrossRef]
Wang K. S., Resch, R., Koel, B. E., Shuler, P. J., Tang, Y. C., Chen, H. J., and Yen, T. F., 1999b, Study of the dissolution of the barium sulfate (001) surface with hydrochloric acid by atomic force microscopy: Journal of Colloid and Interface Science, v. 219, p. 212–215.[Medline]
Wang K. S., Resch, R., Dunn, K., Shuler, P., Tang, Y. C., Koel, B. E., and Yen, T. F., 2000, Scanning force microscopy study of etch pits formed during dissolution of a barite (001) surface in CDTA and EDTA solutions: Langmuir, v. 16, p. 649–655.[CrossRef]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
N Orn) and catechol moieties are known to chelate aqueous metals including Fe3+(aq) (modified from 
