Numerous publications report the existence of intracellular “Si” storage pools in diatoms representing intracellular concentrations of ca. 19–340 mM depending on the species. “Si” storage pools in diatom cells, if present, are supposed to accumulate “Si” for the production of new valves. The accumulated “Si” is then transported into the silicon deposition vesicle (SDV) where the new cell wall is synthesized. Interestingly, the reported concentrations of intracellular “Si” within the storage pool sometimes strongly exceed the solubility of monosilicic acid (ca. 2 mM pH <9). Various types of “Si” storage pools are discussed in the literature. It is usually assumed that “Si” species are stabilized by the association with some kind of organic material such as special proteins, thus forming a soluble silicic acid pools inside the cells. In an effort to mimic the above phenomenon, we have used a variety of neutral or cationic polymers that stabilize two soluble forms of “Si,” silicic and disilicic acids. These polymers include amine-terminated dendrimers, amine-containing linear polymers (with primary, secondary or tertiary amines), organic ammonium polymers, polyethylene glycol (PEG) neutral polymers, co-polymers (containing neutral and cationic parts) and phosphonium end-grafted PEG polymers. All the aforementioned polymeric entities affect the rate of silicic acid polycondensation and also the silica particle growth. Synergistic combinations of cationic and anionic polymers create in situ supramolecular assemblies that can also affect the condensation of silicic acid. Possible mechanisms for their effect on the condensation reaction are presented, with an eye towards their relevance to the “Si pools,” from a bioinspired/biomimetic point of view.
Biosilicification is the formation of siliceous materials in nature (Fig. 1). It also includes uptake, storage, transport and processing of “soluble” forms of silica by living organisms. The diatom  has been used as a model system for the study of biosilicification processes and as a source of inspiration for biomimetic approaches [2–7]. Other biosilica-forming organisms are protists (see for review ), sponges [8–12] and plants [13–15].
It has been suggested that several diatom species preferentially take up silicon (“Si”) as monosilicic acid [Si(OH)4] [16, 17] via special Silicon Transport proteins (SIT proteins) . However, the intracellular “Si” processing, transport, and transfer into the Silica Deposition Vesicle (SDV) are not well understood [19, 20]. Numerous papers refer to the existence of intracellular “Si” storage pools in diatoms representing intracellular concentrations of ca. 19–340 mM depending on the species [21–24].
Silicon storage pools in diatom cells, if present, are supposed to accumulate “Si” for the production of new valves. The accumulated “Si” is then transported into the SDV where the new cell wall is synthesized. Interestingly, the reported concentrations of intracellular “Si” within the storage pool sometimes strongly exceed the solubility of monosilicic acid (ca. 2 mM pH <9) . Various types of “Si” storage pools are discussed in the literature. It is usually assumed that “Si” species are stabilized by the association with some kind of organic material such as special proteins, thus forming a soluble silicic acid pools inside the cells. However, there is no evidence for the presence of “Si” inside the vesicles observed in diatoms yet. In contrast, “Si”-containing vesicles could be identified during the formation of the siliceous spicules of sponges .
The idea of silicic-acid stabilization by organic material has meanwhile lead to various in vitro investigations in order to characterize the influence of the biomolecules found in diatoms upon silicic acid. For example, Kinrade et al. discovered by 29Si NMR spectroscopy that carbohydrate-like molecules can covalently interact with silicate to form 5- and 6-coordinated stable silicon complexes. These observations may be relevant for the understanding of “Si” transport and intracellular stabilization because polysaccharides are indeed found to be attached to diatom biosilica . They may also imply that polysaccharide-like polymers (and “small” molecules) can affect “Si” transport and/or condensation [27–31].
The idea of polyamine-stabilized sols could recently be supported by in vitro studies of silicic acid condensation in the presence of 1-vinylimidazole . Furthermore, Annenkov et al. studied the interaction between poly(vinyl amine) and silicic acid in solution . The formed soluble poly(vinyl amine)/silica composite nanoparticles were discussed as a species relevant with respect to the aforementioned stable “silicon storage pools” in diatoms. Interestingly, amine-terminated polyaminoamide dendrimers [34–37] and other polymers [38–42] (at sufficiently low concentrations, e.g., 40–100 ppm) were previously shown to slow down silicic acid polycondensation.
It is, therefore, an important issue to gain further insight into the interaction between polymeric entities and silicic acid species in solution at a molecular level. The aim of the present paper is to compare the silicic acid stabilizing performance of a number of polymers (see Fig. 1).
Detailed experimental protocols have been described in previous publications [35, 36, 43–45]. However, we provide the essential information below.
Reagents, chemicals and materials. Sodium silicate Na2 SiO3·5H2 O was purchased from EM Science (Merck). Ammonium molybdate [(NH4)6 Mo7 O24·4H2 O] and oxalic acid (H2 C2 O4·2H2 O) were obtained from EM Science (Merck). Sodium hydroxide (NaOH) was purchased from Merck, and hydrochloric acid 37 % from Riedel de Haen. In-house deionized water from an ion-exchange resin was used for all experiments. This water was tested for molybdate-reactive silica and was found to contain negligible amounts. Acrodisc filters (0.45 μm and occasionally 0.20 μm) from Pall-Gelman Corporation were used.
Preparation of supersaturated sodium silicate solutions. A solution containing silicate (500 ppm as SiO2), was prepared by dissolving 4.4 g of Na2 SiO3·5H2 O in 2.5 L nanopure water (a non-glass container must be used), followed by overnight rigorous stirring. This solution contains exclusively molybdate-reactive silica in agreement with the literature. Stock solutions of the additives in water were 1 % w/v (10 000 ppm). The following solutions were prepared for the silicate spectrophotometric detection test: (a) 10 g of ammonium molybdate were dissolved in 100 mL water and its pH was adjusted between 7–8 with NaOH to avoid precipitation of ammonium molybdate. (b) HCl 1 + 1 is prepared by mixing one volume 37 % HCl with equal volume water. (c) A total of 8.75 g of oxalic acid was dissolved in 100 mL water. All solutions were kept in polyethylene containers (glass containers must be avoided in order to minimize SiO2 dissolution and silicate leaching into the test solutions).
Silicic acid polycondensation in the absence of polymers (“control” protocol). One hundred mL from the 500 ppm sodium silicate stock solution (see above) was placed in a polyethylene beaker which charged with a teflon-covered magnetic stirring bar. The pH of this solution was initially ∼11.8 and is subsequently adjusted to 7.00±0.1 by addition of HCl and NaOH (the change in the resulting volume was about 3 % and was taken into account for subsequent calculations). The beaker was then covered with plastic membrane and set aside without stirring. The solutions were checked for molybdate-reactive silica by the silicomolybdate method every 2 h for the first 12 h or after 24, 48, 72 h time intervals after the onset of the pH adjustment to 7.0. There must be strict time control in measuring molybdate-reactive silica, in order to avoid conversion of higher oligomers/colloidal silica to silicic and disilicic acids. Specifically, after the ammonium molybdate and HCl solutions were added to a working sample, a period of 10 min has to pass until the solution of oxalic acid is added to the same sample. Then, another 2-min period has to follow until the final measurement. All samples (control and in the presence of PEGs) were treated in precisely the same way. Separate experiments were performed in which the working solutions were stirred, but no difference in molybdate-reactive silica levels was found, compared to the quiescent solutions.
Silicic acid polycondensation in the presence of polymers. One hundred mL portions of the 500 ppm sodium silicate stock solution (see above) were placed in polyethylene containers and charged with teflon-covered magnetic stir bars. In each container, different volumes of polymers (from the prepared 10 000 ppm stock solutions) were added to achieve desirable polymer concentration. These ranged from 20–40–60–80–100 ppm and the added volumes were 200–400–600–800–1000 μL. After that, the same procedure as the “control” protocol was followed.
The use of polymeric inhibitors or dispersants to control silicic acid polycondensation is principally based on two approaches: inhibition and dispersion. Inhibition is defined as the prevention of silicic acid oligomerization or polymerization. As a result, silicic acid remains soluble and, therefore, formation of colloidal silica is prevented. Dispersion, on the other hand, is the prevention of particle agglomeration to form larger-size particles and the prevention of the adhesion of these particles onto surfaces. In this paper we will consider the “inhibition” approach.
|Polymer||Anionic component||Cationic component||Neutral component||Overall charge at pH 7||Nature of anionic component||Nature of cationic component||Nature of neutral component|
|PAMAM-1||Yes||Yes||Yes||+||–||Protonated primary and tertiary amine groups||Amide groups|
|PAMAM-2||No||Yes||No||+||–||Protonated primary and tertiary amine groups||Amide groups|
|PEI||No||Yes||No||+||–||∼25 % primary amines, ∼50 % secondary amines and ∼25 % amines||–|
|PPEI||Yes||Yes||No||0||Phosphonate groups||Protonated tertiary amine groups||–|
|PALAM||No||Yes||No||+||–||Protonated primary amine groups||–|
|PAMALAM||No||Yes||Yes||+||–||∼45 wt % diallyldimethylammonium groups||∼55 wt % acrylamide groups|
|PCH||Yes||Yes||Yes||+||Phosphonate groups||Protonated tertiary amine groups||Amide and alcohol groups (from the N-acetylglucosamine ring)|
|PEGP+||No||Yes||Yes||+||–||Phosphonium groups||Ether groups|
|PVP-DMAEM||No||Yes||Yes||+||–||Protonated secondary amine groups||Internal amide (from the vinylpyrrolidone group)|
Abbreviations: PAMAM, polyaminoamide dendrimers of generation 1 or 2; PEI, polyethyleneimine; PPEI, phosphonomethylated polyethyleneimine; PALAM, polyallylamine; PAMALAM, poly(acrylamide-co-diallyl-dimethylammonium chloride); PCH, phosphonomethylated chitin; PEG, polyethylene glycol; PEGP+, phosphonium-grafted polyethylene glycol; PVP, polyvinylpyrrolidone; PVP-DMAEM, poly(1-vinylpyrrolidone-co-2-dimethylaminoethyl methacrylate); PVA, polyvinyl alcohol.
The selected polymers show a variety of structural features. All contain some degree of cationic charge. Some (PAMAM-1, PAMAM-2, PEI, PALAM, PAMALAM) possess cationic charge exclusively. Others (PPEI, PCH) are zwitter-ionic, i.e., they have cationic and anionic charge on the polymer backbone. Some polymers possess positive charge by virtue of protonated amine groups (PAMAM-1, PAMAM-2, PEI, PALAM), others have a “pure” cationic charge due to a tertiary N group (PAMALAM). The two co-polymers (PVP-DMAEM and PAMALAM) possess a neutral moiety (pyrolidone and acrylamide, respectively), together with the cationic one.
These additives have been tested at varying concentration for their ability to stabilize molybdate-reactive silica (mono- and di-silicic acids). Below, silicic acid stabilization results after 24 h are presented at 40 ppm concentration for all polymers (Fig. 3).
It becomes apparent that all polymers (except PVA and PCH) show variable stabilization activity (higher silicic acid levels than the “control”). PAMAM-1 and PAMAM-2 (both have their surface amine groups protonated at pH 7) are very effective inhibitors at 40 ppm. The presence of protonated amine groups is not the only necessary condition for good inhibition. Notice that polymers PEI and PALAM (also having their amine groups protonated at pH 7) show substantially reduced performance. This could be explained by the fact that excessive cationic charge causes the polymeric additive to be entrapped and hence de-activated within the colloidal silica matrix. PAMALAM, which is a polymer that possesses a tertiary N group is a “mediocre” stabilizer. From the zwitter-ionic polymers (PPEI and PCH), PPEI is a very effective stabilizer. In this case, it appears that the negative charge (–PO3 H- for PPEI) “balances” the positive charge in such a way that the polymer continues to be active, but inhibitor entrapment and deactivation is stopped. For the PCH polymer, perhaps the anionic charge (due to –PO3 H-) is too excessive and the cationic charge (necessary for inhibition) is “neutralized.” The phosphonium polymer PEGP+-4000 exhibits excellent stabilization performance, stabilizing 370 ppm silicic acid. The neutral polymer PEG-10 000 is also a good stabilizer, keeping 345 ppm silicic acid soluble.
It is interesting to evaluate the effect of polymer concentration increase on the levels of silicic acid measured. Thus, the polymer concentration was doubled (an increase from 40 ppm to 80 ppm). The effect in stabilization activity is shown in Fig. 4.
PAMAM-1 maintains its performance, in contrast to PAMAM-2, whose activity is dramatically compromised (there is a drop in silicic acid level from 374 to 238 ppm). PEI also shows some minor reduction. PEGP+-4000 and PEGP-1000 remain constant. PALAM reduces its previous activity (a drop from 270 to 220 ppm). PPEI shows a minor increase (from 363 to 373 ppm). The polymers that substantially increase their activity are PAMALAM (from 225 to 300 ppm), PVP-DMAEM (from 220 to 370 ppm), PCH (from 180 to 250 ppm). Further dosage increase, however, caused no further enhancement (data not shown). For comparison reasons, Fig. 5 shows the concentration-dependence increase or decrease of silicic acid levels per polymer.
It is apparent that polymer concentration increase has detrimental effects on stabilization activity for certain polymers. It can be explained upon examination of the competing reactions taking place concurrently. They are the following.
Polymerization of silicic acid. This occurs through an SN 2-like mechanism that involves attack of a silicic acid monomer on a second silicic acid molecule. This pathway generates at first short-lived di-silicic acid, which in turn continues to polymerize in a random way to eventually yield colloidal silica particles.
Silicic acid stabilization by the cationic additive. This is the actual stabilization step and occurs presumably through cation-anion and hydrogen bonding cooperative interactions.
Flocculation between the polycationic polymerand the negatively charged colloidal silica particles (at pH 7) that are formed by the uninhibited silicic acid polymerization.
The cationic polymer is trapped within the colloidal silica matrix, based on process (c). This is demonstrated by the appearance of a light flocculent precipitate (or dispersion at times). Polymer entrapment causes its depletion from solution and its deactivation. Therefore, only a portion of the polymer is available to continue inhibition at much lower levels than initially added to the polymerization medium. Thus, soluble silicic acid levels continue to decrease because eventually there is not sufficient polymer to perform the stabilization. Polymer entrapment is directly proportional to cationic charge density. For example, PEI, PALAM and PAMAM-2 with high positive charge density creates composite precipitates with colloidal silica rapidly.
The data presented in Figs. 3–5 on PVA and PEG are intriguing. Both polymers are neutral at all “physiological” pH values, and, yet, PVA shows no stabilization activity at all. In contrast PEG-10 000 is an effective stabilizer of silicic acid. The only viable interaction of the PEG backbone with silicic acid is hydrogen bonding. It should be noted that PVA can also use its –OH moieties as hydrogen bond acceptors from the silanol groups. Apparently, such hydrogen bonds offer no advantage to silicic acid stabilization. In turn, ethereal oxygens from the PEG backbone can (and apparently are) form hydrogen bonds with the silanol groups of silicic acid. This is demonstrated schematically in Fig. 6.
In order to get further insight to the silicic acid stabilization mechanism by PEG oligomers/polymers, we have carried out the following analysis. It refers to PEG 10 000, but can be generalized for PEG polymers of various molecular weights. This analysis is based on the data shown in Table 2.
|Time (h)||Control||PEG-1500||PEG-2000||PEG-6000||PEG-10 000||PEG-12 000||PEG-20 000|
Every O atom of the PEG backbone can (theoretically) form a maximum of two H-bonds, as also shown in Fig. 6. This means that each PEG 10 000 molecule (containing 227 O atoms) can stabilize a maximum number of 454 molecules of silicic acid. By considering that 60 ppm (6 μM) of PEG 10 000 stabilize 177 ppm molybdate-reactive silica (370–193 = 177 ppm or 2.94 mM silicic acid above the “control” level) after 8 h, the molar ratio of Si:PEG is 490. This corresponds to virtually 100 % loading of the PEG 10 000 with silicic acid molecules, and the PEG polymer utilizes all its oxygen atoms to stabilize silicic acid at that concentration. By repeating the analysis for other PEGs at various concentrations, % loadings can be obtained for each experiment. The meaning of “% loading” refers to the% of O atoms on the PEG chain interacting with silicic acid molecules via hydrogen bonding. The calculation of “% loading” is based on the hypothesis that the theoretical maximum number of silicic acid molecules stabilized by a single PEG chain is double the number of oxygen atoms present on the PEG backbone. Again, the chemical basis for this hypothesis is that each O atom possesses two lone pairs, which can act as H-bond acceptors from two H-O-Si moieties.
It is given by the following simple equation:
These curves are shown in Fig. 7. Based on this analysis, the dependence of Si:O ratio on PEG concentration can be calculated, see Table 3. The maximum value of the Si:O ratio should not exceed the theoretical maximum of 2 in agreement with our observations.
|PEG 10 000 concentration (ppm)||PEG 10 000 concentration (μM)||“control” silicic acid (ppm)||silicic acid in presence of PEG (ppm)||silicic acid stabilized solely due to PEG 10 000 (ppm)||silicic acid stabilized solely due to PEG 10 000 (μM)||# of silicic acid molecules stabilized||% PEG 10 000 loadinga||Si:O ratio|
aIt is assumed that each PEG 10 000 molecule can stabilize a maximum of 454 molecules silicic acid (two Si per one O).
Amorphous silica precipitates were collected in the presence of PEGs after 8 h and 3 days, and studied by SEM, Fig. 8. A general observation is the tendency for particle aggregation upon prolonged silicic acid polymerization in the presence of PEG. Moreover, silica particles seem to possess the common spherical morphology. Lastly, with increasing PEG MW (i.e., increased chain length) there is a decrease in particle size and more pronounced aggregation, obviously due to PEG association with the silica particles as corroborated by the NMR and FT-IR results (not shown here). A systematic enhancement of aggregation upon PEG chain length increase was also observed by Vong et al. .
There are numerous open questions regarding pathways in nature for silicic acid transport, storage, polycondensation and finally biosilica production. Biopolymers play a profound role in every step of the process [47–50]. Our research focuses on bioinspired/biomimetic approaches to model the ability of living organisms (diatoms and sponges) to store silicic acid (or other forms of silica) prior to its polycondensation to produce amorphous silica . In this paper we have presented in a comparative way the ability of certain polymers (cationic and neutral) to stabilize silicic acid. These molecular machineries could be envisioned as direct analogs of natural systems to preconcentrate silicic acid prior to biosilica synthesis.
KDD thanks the EU for funding the Research Program SILICAMPS-153, under the ERA.NET-RUS Pilot Joint Call for Collaborative S&T projects.
 F. E. Round, R. M. Crawford, D. G. Mann. The Diatoms. Cambridge University Press, Cambridge (1990). Search in Google Scholar
 N. Nassif, J. Livage. Chem. Soc. Rev. 40, 849 (2011). Search in Google Scholar
 Z. Bao, E. M. Ernst, S. Yoo, K. H. Sandhage. Adv. Mater. 21, 474 (2009). Search in Google Scholar
 M. B. Dickerson, K. H. Sandhage, R. R. Naik. Chem. Rev. 108, 4935 (2008). Search in Google Scholar
 N. Kröger, K. H. Sandhage. MRS Bulletin35, 122 (2010). Search in Google Scholar
 R. Gordon, D. Losic, M. A. Tiffany, S. S. Nagy, F. A. S. Sterrenburg. TrendsBiotechnol.27, 116 (2009). Search in Google Scholar
 S. Neethirajan, R. Gordon, L. Wang. TrendsBiotechnol. 27, 461 (2009). Search in Google Scholar
 K. Shimizu, J. Cha, G. D. Stucky, D. E. Morse. Proc. Natl. Acad. Sci. USA95, 6234 (1998). Search in Google Scholar
 W. E. G. Müller, M. Rothenberger, A. Boreiko, W. Tremel, A. Reiber, H. C. Schröder. Cell. Tissue Res.321, 285 (2005). Search in Google Scholar
 H. C. Schröder, X. Wang, W. Tremel, H. Ushijima, W. E. G. Müller. Nat. Prod. Rep. 25, 455 (2008). Search in Google Scholar
 H. C. Schröder, D. Brandt, U. Schloßmacher, X. Wang, M. N. Tahir, W. Tremel, S. I. Belikov, W. E. G. Müller. Naturwissenschaften94, 339 (2007). Search in Google Scholar
 H. Ehrlich, R. Deutzmann, E. Brunner, E. Cappellini, H. Koon, C. Solazzo, Y. Yang, D. Ashford, J. Thomas-Oates, M. Lubeck, C. Baessmann, T. Langrock, R. Hoffmann, G. Wörheide, J. Reitner, P. Simon, M. Tsurkan, A. V. Ereskovsky, D. Kurek, V. V. Bazhenov, S. Hunoldt, M. Mertig, D. V. Vyalikh, S. L. Molodtsov, K. Kummer, H. Worch, V. Smetacek, M. J. Collins. Nature Chem. 2, 1084 (2010). Search in Google Scholar
 G. J. Raleigh. Soil Sci. 60, 133 (1945). Search in Google Scholar
 T.-H. Liou, S.-J. Wu. Ind. Eng.Chem. Res. 49, 8379 (2010). Search in Google Scholar
 H. A. Currie, C. C. Perry. Ann. Botany100, 1383 (2007). Search in Google Scholar
 J. C. Lewin. Plant Physiol. 30, 129 (1955). Search in Google Scholar
 Y. Del Amo, M. A. Brzezinski. J. Phycol. 35, 1162 (1999). Search in Google Scholar
 M. Hildebrand, B. E. Volcani, W. Gassmann, J. I. Schroeder. Nature385, 688 (1997). Search in Google Scholar
 M. Hildebrand. In Biomineralization – from Biology to Biotechnology and Medical Applications, E. Bäuerlein (ed.), p. 170, Wiley, Weinheim (2000). Search in Google Scholar
 M. Sumper, E. Brunner. ChemBioChem9, 1187 (2008). Search in Google Scholar
 D. Werner. Arch. Mikrobiol. 55, 278 (1966). Search in Google Scholar
 S. W. Chisholm, F. Azam, R. W. Eppley. Limnol. Oceanogr. 23, 518 (1978). Search in Google Scholar
 V. Martin-Jézéquel, M. Hildebrand, M. A. Brzezinski. J.Phycol. 36, 821 (2000). Search in Google Scholar
 V. Martin-Jézéquel, P. Lopez. Prog. Mol. Subcell. Biol.33, 99 (2003). Search in Google Scholar
 H. C. Schröder, F. Natalio, I. Shukoor, W. Tremel, U. Schloßmacher, X. Wang, W. E. G. Müller. J.Struct. Biol. 159, 325 (2007). Search in Google Scholar
 A. Chiovitti, R. E. Harper, A. Willis, A. Bacic, P. Mulvaney, R. Wetherbee. J. Phycol.41, 1154 (2005). Search in Google Scholar
 S. D. Kinrade, A.-M. E. Gillson, C. T. G. Knight. J. Chem. Soc. Dalton Trans. 307 (2002). Search in Google Scholar
 S. D. Kinrade, R. J. Hamilton, A. S. Schach, C. T. G. Knight. J. Chem. Soc., Dalton Trans. 961 (2001). Search in Google Scholar
 S. D. Kinrade, A. S. Schach, R. J. Hamilton, C. T. G. Knight. Chem. Commun. 1564 (2001). Search in Google Scholar
 S. D. Kinrade, R. J. Balec, A. S. Schach, J. P. Wang, C. T. G. Knight. Dalton Trans. 3241 (2004). Search in Google Scholar
 S. D. Kinrade, J. W. Del Nin, A. S. Schach, T. A. Sloan, K. L. Wilson, C. T. G. Knight. Science285, 1542 (1999). Search in Google Scholar
 V. V. Annenkov, E. N. Danilovtseva, V. A. Pal’shin, V. O. Aseyev, A. K. Petrov, A. S. Kozlov, S. V. Patwardhan, C. C. Perry. Biomacromolecules12, 1772 (2011). Search in Google Scholar
 E. N. Danilovtseva, V. A. Pal’shin, Y. V. Likhoshway, V. V. Annenkov. Adv. Sci. Lett. 4, 616 (2011). Search in Google Scholar
 E. Neofotistou, K. D. Demadis. Colloids Surf. A. 242, 213 (2004). Search in Google Scholar
 K. D. Demadis, E. Neofotistou. Chem. Mater. 19, 581 (2007). Search in Google Scholar
 E. Mavredaki, E. Neofotistou, K. D. Demadis. Ind.Eng. Chem. Res.44, 7019 (2005). Search in Google Scholar
 E. Neofotistou, K. D. Demadis, Int. J. Corros. Scale Inhib.3, 28 (2014). Search in Google Scholar
 K. D. Demadis. J. Chem. Technol. Biotechnol. 80, 630 (2005). Search in Google Scholar
 K. D. Demadis, A. Stathoulopoulou. Ind. Eng. Chem. Res.45, 4436 (2006). Search in Google Scholar
 K. D. Demadis, E. Neofotistou, E. Mavredaki, M. Tsiknakis, E.-M. Sarigiannidou, S. D. Katarachia. Desalination179, 281 (2005). Search in Google Scholar
 E. Mavredaki, A. Stathoulopoulou, E. Neofotistou, K. D. Demadis. Desalination210, 257 (2007). Search in Google Scholar
 K. D. Demadis, M. Preari. Des. Wat. Treat. in press (2014). DOI 10.1080/19443994.2014.927803. Search in Google Scholar
 E. Neofotistou, K. D. Demadis. Desalination167, 257 (2004). Search in Google Scholar
 M. Preari, K. Spinde, J. Lazic, E. Brunner, K. D. Demadis. J. Am. Chem. Soc.136, 4236 (2014). Search in Google Scholar
 K. Spinde, K. Pachis, I. Antonakaki, E. Brunner, K. D. Demadis. Chem. Mater.23, 4676 (2011). Search in Google Scholar
 M. S. W. Vong, N. Bazin, E. A. Sermon. J. Sol-Gel Sci. Technol.8, 499 (1997). Search in Google Scholar
 H. Ehrlich, K. D. Demadis, P. G. Koutsoukos, O. Pokrovsky. Chem. Rev.110, 4656 (2010), and references therein. Search in Google Scholar
 H. Ehrlich. In Encyclopedia of Geobiology. J. Reitner and V. Thiel (eds.), p. 796, Springer Verlag (2011). Search in Google Scholar
 H. C. W. Skinner, H. Ehrlich. In Treatise on Geochemistry. Vol. 10: Biogeochemistry, 2nd ed., K.K. Turekian, H.D. Holland (eds.), p. 105, Elsevier Science (2013). Search in Google Scholar
 E. Brunner, P. Richthammer, H. Ehrlich, S. Paasch, P. Simon, S. Ueberlein, K.-H. van Pee. Angew. Chem. Int. Ed.48, 9724 (2009). Search in Google Scholar
©2014 IUPAC & De Gruyter