Controlled morphogenesis of amorphous silica and its relevance to

American Mineralogist, Volume 97, pages 1381–1393, 2012

Controlled morphogenesis of amorphous silica and its relevance to biosilicification Jia-Yuan Shi,1 Qi-Zhi Yao,2 Xi-Ming Li,2 Gen-Tao Zhou,1,* and Sheng-Quan Fu3 CAS Key Laboratory of Crust-Mantle Materials and Environments, School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, P.R. China 2 School of Chemistry and Materials, University of Science and Technology of China, Hefei 230026, P.R. China 3 Hefei National Laboratory for Physical Sciences at Microscale, University of Science and Technology of China, Hefei 230026, P.R. China. 1

Abstract Biogenetic biosilica displays intricate patterns that are structured on a nanometer-to-micrometer scale. At the nanoscale, it involves the polymerization products of silica, apparently mediated by the interaction between different biomolecules with special functional groups. In this paper, using tetraethyl orthosilicate [TEOS, Si(OCH2CH3)4] as a silica source, phospholipid (PL) and dodecylamine (DA) were introduced as model organic additives to investigate their influence on the formation and morphology of silica in the mineralization process. Morphology, structure, and composition of the products were characterized using a range of techniques including FESEM, TEM, SAXRD, TG-DTA, solid-state 29Si NMR, FTIR, and nitrogen physisorption. The FESEM and TEM analyses demonstrate that increasing PL concentrations at constant DA content leads to the formation of siliceous elongated structures. Localized enlargement can also be observed during further growth of elongated structures, displaying some features of the earliest recognizable stage of valve development in diatoms. In addition, in the presence or absence of PL, a series of control experiments using ammonia instead of DA show that no elongated structures are obtained, suggesting that the formation of elongated silica structures results from the cooperative interactions between PL and DA molecules. Because both organic amines (e.g., long-chain polyamines, LCPA) and phospholipid membranes (e.g., silicalemma) are of special importance for biosilicification in diatoms and sponges, our results imply that phospholipids are involved in the formation of organic aggregates, and thus influence the amines-mediated silica deposition. As such, our results may provide a new insight into the mechanism of biosilicification. Keywords: Biosilica, phospholipids, organic amine, biomimetic mineralization, biosilicification

Introduction Biomineralization is a process by which living organisms produce organic/inorganic composites, often to harden or stiffen existing tissues (Lowenstam and Weiner 1989; Sigel et al. 2008). Approximately 80% of these organic/inorganic structures are crystalline (Weiner and Addadi 1997; Addadi et al. 2003). Silica (biosilica) is the second-most abundant constituent of biominerals after carbonate (Lowenstam 1981; Perry and Keeling-Tucker 2000), and biosilica always consists of glassy amorphous solid (Schröder et al. 2008). Biosilica has attracted much attention because of its unique morphologies and hierarchical structures, fascinating mechanical properties, and potential applications in many fields (e.g., Round et al. 1990; Sumper et al. 2003; Sumper and Brunner 2006; Losic et al. 2009). Understanding biomineralization requires elucidation of the underlying cellular and molecular biological processes. It is especially important to determine how organic molecules (such as proteins, polysaccharides, and lipids) can be involved in the formation of specific mineral substructures. Phospholipids, often in the form of bilayer structures, have been commonly involved in natural biomineralization processes (e.g., Lowenstam and Weiner 1989). In particular, phospholipids as important membrane constituents of biological vesicles are commonly involved * E-mail: [email protected] 0003-004X/12/0809–1381$05.00/DOI: http://dx.doi.org/10.2138/am.2012.4081

in delineating reaction compartments for the crystallization of biominerals (e.g., Gorby et al. 1988; Collier and Messersmith 2002; Bäuerlein 2003; Anderson et al. 2005). The isolation and structural determination of phospholipids from biosilicification organisms including sponges and diatoms have been reported (e.g., Djerassi and Lam 1991; Early et al. 1996; Genin et al. 2008; Ivanisevic et al. 2011). For example, the sponge cell membrane is unique in terms of its lipid diversity, and Ivanisevic et al. (2011) suggest that lysophospholipids could play an important role in sponge embryogenesis and morphogenesis. Phospholipid membrane was also found to surround hexactinellid spicules and delimit the confined space where silicification occurs (Uriz 2006). Silica deposition in diatoms is also a membrane phenomenon. It occurs within a specialized compartment known as the silica deposition vesicle (Sumper and Kröger 2004), whose membrane, called the silicalemma, consists of a typical lipid bilayer (Chiappino and Volcani 1977; Bäuerlein 2003). In recent decades, various organic and biological molecules have been successfully separated and identified from cell-wall extracts of diatoms. Organic amines (long-chain polyamines, LCPAs) extracted from diatoms can induce in vitro precipitation of spherical silica particles under physiological pH conditions (Kröger et al. 2000). However, the spheres obtained in these biomimetic experiments are not seen in the smooth, elongated girdle bands or raphe of diatoms (Reimann et al. 1965). This means that additional factors must come into play to shape biosilica (Davis and Hildebrand

1381

1382

SHI ET AL.: MORPHOGENESIS OF AMORPHOUS SILICA AND BIOSILICIFICATION

2008). Several previous works have revealed a close association of biosilica with one intralumenal side of the silica deposition vesicle during the formation of early-stage siliceous structures (Schmid and Schulz 1979; Schmid and Volcani 1983). Based on these investigations, Davis and Hildebrand (2008) suggested a model where silica polymerization determinants are anchored to the silicalemma by means of interactions with transmembrane proteins and the cytoskeleton. Recently, using the ion-abrasion SEM approach, Hildebrand et al. (2009) found that the silicalemma is tightly attached to silica in areas where silica was deposited. This indicates that membrane components of the silica deposition vesicle could become part of the silica structure. Thus, the examination of valve development in diatoms indicates that the membrane-bound silica deposition vesicle could provide, not only spatial constraints, but also distinct chemical influences for the silica polymerization (Schmid and Volcani 1983). In this study, phospholipid (PL) and dodecylamine (DA) were selected as model organic additives to influence the precipitation of silica in biomimetic silicification. Phospholipids, which have a hydrophilic head and two hydrophobic tails, are a major component of all plasma membranes (e.g., Palsdottir and Hunte 2004). Recent evidence suggests that simple silicon alkoxides, such as TEOS, may be reasonable model substrates for the in vivo precursors of biosilification (Brutchey and Morse 2008). TEOS has also been used in numerous enzymatically catalyzed silica synthesis in vitro (Cha et al. 1999; Zhou et al. 1999). Therefore, TEOS was selected as an inorganic Si precursor in all of our experiments. The goal of this study is to examine the effect of lipids on the development of silica morphology and to reveal the contribution of the interactions between amine and lipid to biogenic silica morphogenesis. Silica particles with different morphologies were obtained in the presence of PL and DA, and the morphological evolution of the deposited silica from spherical through test-tube-like to tadpole-looking features were also observed at different concentrations of PL. Since the amine-mediated silica deposition and intimate association between silicalemma (phospholipid) and biosilica are important features of diatom silicification, our biomimetic mineralization experiments may prove useful for a deeper understanding of biosilicification.

Experimental procedures All starting chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd. and used without further purification. Phosphatidylcholine (PL) is of biotech grade, whereas all other reagents, such as ethanol, DA, and TEOS, are of analytical grade. Deionized water was also used in the syntheses. For all experiments, glassware was cleaned with aqua regia (3:1 HCl/HNO3), rinsed thoroughly with deionized water, and oven-dried overnight before use. The silica syntheses were always carried out in 45 mL of solution with a volume ratio of water to ethanol of 1:2. A typical procedure was as follows: 0.19 g of PL (0.26 mmol) and 0.16 g of DA (0.86 mmol) were dissolved in 45 mL water-ethanol solution through ultrasonification and then stirred for about 5 min to obtain a clear solution. Then 0.1 mL of TEOS (0.45 mmol) was injected into the solution using a 100 μL syringe, and opaque floccules formed in the solution after 20 min. The mixture was continuously stirred at room temperature for an additional 6 h. The resultant particles were isolated by centrifugation (4000 g for 4 min), cleaned by four cycles of centrifugation/washing/redispersion in ethanol to remove the organic components possibly adsorbed on the surfaces of the mineralized product, and allowed to dry at room temperature for 1 day. To examine the microstructures of the mineralized products, some samples were further treated two times with 50 mL of HCl ethanol solution by stirring at 60 °C for 6 h. The HCl solution was prepared by adding 15 mL of conc. HCl to 120 mL EtOH. For other morphogenesis

of silica microstructures, the procedures were the same as above, except that some experimental parameters were changed. Comparative experiments were performed with ammonia (2 mL) instead of DA. The detailed experimental conditions and the corresponding morphologies of the products are listed in Table 1. Several analytical techniques were used to characterize the synthesized products. Field emission scanning electron microscopy (FESEM, JEOL JSM-6700F) was applied to investigate the size and morphology. The samples were coated with a thin film of gold. Transmission electron microscope (TEM) images were obtained on a JEM 2010 transmission electron microscope with an accelerating voltage of 200 kV. The TEM samples were prepared by depositing and evaporating a few drops of the sample suspension in ethanol on a copper grid. The powder X‑ray diffraction (XRD) patterns of the samples were recorded with a Japan Rigaku TTR-III X‑ray diffractometer equipped with graphite monochromatized CuKα irradiation (λ = 0.154056 nm), employing a scanning rate of 0.02 °/s in the 2θ range 0.8–10°. Infrared spectra were collected on a Nicolet 8700 FTIR spectrometer on KBr pellets. Thermogravimetric analysis (TGA) was carried out using a SDTQ 600 TG/DTA thermal analyzer (TA, U.S.A.) with a heating rate of 10 °C/ min from room temperature to 800 °C in a flow air atmosphere. 29Si MAS NMR spectra were recorded on a Bruker AV III 400 (WB) spectrometer at 79.49 MHz under single pulse mode with a 7 mm zirconia rotor, pulse width of 5.5 μs, and a relaxation delay of 120 s. The magnetic field was 9.395 T. The spinning rate was 5 kHz and a total number of 64 scans were recorded. N2-sorption isotherms of the samples were measured using a Micromeritics Tristar II 3020 M instrument at liquid-nitrogen temperature. To avoid changes of properties of organic components in silica structures different degasification temperatures of 300 and 100 °C were used for the samples with and without heating extraction, respectively. From the adsorption isotherm, the Barrett-Joyner-Halenda theory (BJH) was used to calculate the mesopore volume and its size distribution. Specific surface areas were calculated by the Brunauer-Emmett-Teller (BET) method in the relative pressure range of P/P0 = 0.05–0.3. Pore volumes were obtained from the volumes of N2 adsorbed at or close to P/P0 = 0.95.

Results FESEM images in Figure 1 display the size and morphology of silica particles synthesized with 20 mM DA and 5.68 mM PL (sample D-4 in Table 1). Numerous FESEM images of the product show a uniform size and regular shape of the synthesized particles. From the image of some half-open particles, the obtained product corresponds to hollow structures. The lowmagnification image of the sample shown in Figure 1a indicates that the length of the tube-like products ranges from several hundred nanometers up to ~1 μm, and tubes have an average outer diameter of ca. 300 nm. Further magnification (Fig. 1b) allows the observation of tubes with crinkled surfaces and wall thicknesses of about 50 nm. However, the FESEM images can only provide basic information on the morphological properties of the particles at a large scale. To observe the internal structures of the tube-like structures, TEM measurements were performed on sample D-4. Figures 1c–1f shows the TEM images of sample D-4 before and after Table 1. Experimental conditions and the resulting silica structures Sample DA (mol/L) PL (mmol/L) Ammonia (mL) pH Morphology D-0 0.02 0 − 11.5 no product D-1 0.02 0.60 − 11.5 perfect and cracked sphere D-2 0.02 2.09 − 11.5 cracked sphere D-3 0.02 3.89 − 11.5 disk-like D-4 0.02 5.68 − 11.5 test-tube-like D-5 0.02 7.47 − 11.5 tadpole-like A-1 − 0.60 2 11.5 no product A-2 − 2.09 2 11.5 solid sphere A-3 − 2.99 2 11.5 disk-like A-4 − 3.89 2 11.5 flake and hollow sphere A-5 − 4.78 2 11.5 hollow sphere A-6 − 5.68 2 11.5 hollow sphere A-7 − 7.47 2 11.5 irregular


Figure 1. (a and b) SEM images of sample D-4 synthesized with 0.02 M DA and 5.68 mM PL; (c–f) TEM images of the sample D-4 before (c and d) and after (e and f) extraction via a 60 °C heating in a HCl–ethanol solution.

1383

1384


60 °C extraction with a HCl–ethanol solution. A typical image of a synthesized tube without treatment by the HCl solution is presented in Figure 1c, and the test-tube-like structure is revealed in the image through contrast difference between the center and margin. Figure 1d depicts the local magnification of the tube wall. The lamellar feature is clearly visible after heating extraction, as shown in Figures 1e and 1f. This may be due to the removal of the organic components (Liu et al. 2008). Thus, organic components are incorporated in the siliceous structures and influence the inner structures of silica, indicating that an organic-inorganic composite has been formed under the current conditions. The existence of pores in the shells of the silica tubes is confirmed by the corresponding high-resolution TEM image (Fig. 1f) of the treated sample. Moreover, as indicated by arrows in Figures 1e and 1f, separated single layers are observed, which is another example of evidence of the multilayer structure. In summary, sample D-4 shows half-open tubes with crinkled, porous and multilayer walls. The porosity of such structures has been determined by SAXRD and nitrogen physisorption. The typical small-angle XRD patterns of sample D-4 before and after 60 °C extraction is presented in Figure 2. The SAXRD pattern of the untreated sample (solid curve) exhibits a strong, relatively broad reflection at 2.0 °2θ and a very weak broad shoulder in the region near 4.5 °2θ. It can be indexed as a lamellar phase with a d-spacing of 4.31 nm calculated from the Bragg equation (Cheng and Liu 2002; Li et al. 2004; Cheng et al. 2007). After treatment in a HCl–ethanol solution for 6 h, sample D-4 displays a SAXRD pattern (dashed curve) with single broad peak shifting to a lower 2θ position. Correspondingly, the pore-pore correlation distance increases from 43.1 to 53.8 Å. These results indicate the retention of the framework upon removal of organic components (Kim et al. 2000). It should also be pointed out that the second reflection near 4.5 °2θ disappears after heating extraction, which suggests that the porous structure is disturbed due to the removal of the organic components, though the silica laminas still remain (Fig. 1f). Thus, the SAXRD results are in agreement with the idea that the porous product is a layer-like aggregation. Nitrogen adsorption–desorption isotherms were also measured to determine the porosity of sample D-4 before and after 60 °C heating extraction. The isotherms are presented in Figure 3, they are similar to each other in overall shape, indicating that the porous structures have not been destroyed by the post extraction. Nevertheless, the amount of adsorbed nitrogen greatly increases after the extraction treatment (Fig. 3b), indicating an obvious increase in pore volume after the removal of the organic components. Both samples exhibit type IV adsorption characteristic according to IUPAC classification (Kruk and Jaroniec 2001; Qu et al. 2010), with a H4 hysteresis loop that closes around at a relative pressure of ca. 0.4. Furthermore, the type H4 loop is often associated with narrow slit-like pores (Sing et al. 1985). The isotherms also show a large increase at P/P0 = 0.45–0.5 due to the capillary condensation in the mesopores (Kao et al. 2007). Calculated from the adsorption branch of the nitrogen isotherm with the BJH method, an average pore size of the untreated and the extracted samples is 37.4 and 40.0 Å, respectively. The corresponding BET surface area and pore volume are 249 m2/g and 0.19 cm3/g, 572 m2/g and 0.37 cm3/g, respectively. The results above show that the pore size and volume of the

Figure 2. Representative low-angle X‑ray diffraction patterns of sample 4 before (dashed) and after (solid) a 60 °C heating extraction in a HCl–ethanol solution. The numbers near the diffraction peak are the pore–pore correlation distances in angstrom (Å).

Figure 3. N2 adsorption–desorption isotherms of sample 4 before (a) and after (b) a 60 °C heating extraction in a HCl–ethanol solution.

products increased after 60 °C extraction while the pore shape is maintained, as shown by the unchanged type of hysteresis loops. Thus, organic compounds enter the silicious porous structures, and the resulting structures correspond to an organic-inorganic composite. The formation of pores is probably favored by the co-deposition of organic components and silica. Figure 4a displays the infrared spectrum of sample D-4. It presents several characteristic vibration bands of the silica framework and adsorbed water. The broad and intense band with a maximum at 1066 cm–1 as well as the shoulder at 1200 cm–1 are characteristic of the antisymmetric stretching vibrational mode of the Si–O–Si siloxane bridges (Agger et al. 1998). The vibrational bands at ca. 800 and 457 cm–1 can be assigned to the symmetric stretching and bending modes of Si-O-Si, respectively


(Michaux et al. 2008). The adsorption peak belonging to the Si-O stretching vibration of Si–OH bond appears at 970 cm–1 (Michaux et al. 2004). The weak shoulder bands around 3649 cm–1 can be ascribed to mutually hydrogen-bonded Si–O–H stretching (Ji et al. 2007). The spectrum shows a peak around 3421 cm–1 corresponding to the hydroxyl groups (Venkatathri et al. 2008; Li et al. 2002), and the peak at 1636 cm–1 can be assigned to the δ(HOH) of physisorbed water (Zhao et al. 2008; Wang and Liu 2005). The presence of organic components can be proved by peaks belonging to the stretching vibrations of C–H bonds at 2853 and 2924 cm–1 (Venkatathri et al. 2008). At 1466 cm–1 the characteristic vibrations of C–C bonds are also observed (Cauda et al. 2009; Venkatathri et al. 2008). The spectrum shows a band at 1735 cm–1 corresponding to the carbonyl group of PL. Antisymmetric bending mode of the O-P-O bonds at 569 cm–1 corresponding to the phosphate group of PL is also observed (Sadasivan et al. 2005). However, the bands between 1175 and 1345 cm–1 belonging to the P-O stretch are masked to some extent by additional Si-O-Si framework vibrations from the mineralized particles at 1080 cm–1 (large band) and 1200 cm–1 (shoulder) (Seddon et al. 2002). Similarly, the bands around 3333 cm–1 corresponding to the NH2 stretch (Vidyadhar et al. 2003) are masked by hydroxyl groups vibrations at 3421 cm–1. On the other hand, a peak around 1540 cm–1 is observed; it can be ascribed to bending δ(NH) vibration modes, which is the characteristic band of DA (Chen et al. 2010). Therefore, we believe that PL and DA are intimately associated with the mineralized structures. In contrast, for the same sample, the removal of the organic templates upon HCl-ethanol extraction is confirmed by the fact that the peaks belonging to C–H, C=O and C–C bonds disappear (as highlighted by four dashed lines in Fig. 4b). Figure 5 shows the TG-DTA curves of the test-tube-like structures (sample D-4). The TG curve reveals a ~29.1% total weight loss from room temperature to 800 °C (Fig. 5b). An ~7.3% weight loss from room temperature to 120 °C and the

Figure 4. FTIR spectra for the sample D-4 before (a) and after (b) extraction with a HCl-ethanol solution.

1385

corresponding endothermal peak at 45 °C in the DTA curve indicate the evaporation of surface-adsorbed water and ethanol. The remaining 21.8% loss, however, cannot be accounted for purely by the loss of the organic components. In fact, the condensation of the silica network under the experimental conditions was incomplete, and further polymerization/condensation of the network, with the loss of the organic template, probably occurs near 200 °C (Meegan et al. 2004). The endothermic peak at 200 °C (vertical arrow in Fig. 5) and the significant exothermic peak at 330 °C (horizontal arrow in Fig. 5) in our DTA curve also confirm the occurrence of similar polymerization/condensation process with the loss and combustion of the organic template. The extra polymerization/condensation step during calcination at 800 °C overnight is further evidenced by 29Si magic-anglespinning (MAS) NMR of the silica particles (sample D-4) (Fig. 6). Prior to calcination, the nanotubes are composed of 8.6% of Q2 [(SiO)2Si(OH)2 at –91.05 ppm], 47.7% of Q3 [(SiO)3Si(OH) at –102.43 ppm], and 43.7% of Q4 [(SiO)4Si at –111.15 ppm], indicating a relatively open framework. After calcination the composition is 96.6% Q4 and 3.4% Q3 (Fig. 6b), demonstrating that the silica is present as a highly condensed network after calcination. Similar tests were repeated for another four times, and the results are listed in Table 2. After calcination, the “molar mass” of silica falls from 66.08 ± 0.36 to 60.36 ± 0.06. The mass loss due to the condensation of silanol groups on the TGA scale is merely 6.7%. Thus, the weight loss due to the combustion and desorption of organic components is about 15.1% (21.8 – 6.7%). Summarizing, TEM, SAXRD, and nitrogen adsorptiondesorption analyses demonstrate that the silicious samples correspond to porous structures, and that organic compounds are

Figure 5. DTA and TG curves of sample D-4.

1386


between the inside and the outside of the particles. The outer part displays wormhole-like mesopores, while a lamellar mesostructure is observed in the inner region. Moreover, no obvious opaque floccules are observed for syntheses in the absence of PL upon stirring for 6 h (sample D-0 in Table 1), indicating that the addition of PL facilitates the deposition of silica. When the PL concentration is increased to 2.09 mM (sample D-2), SEM analyses (Fig. 7c) show that the obtained particles keep the morphological properties of sample D-1, the diameters vary from ~200 to 500 nm, and cracks appear on the particle surfaces. The TEM image of sample D-2 further reveals that the lamellar meso-structure is radially arranged from the central to the outer part of the particle (Fig. 7d). Compared with sample D-1, the increase of the PL concentration in the mineralization system leads to looser lamellar meso-structures. A further increase of PL concentration to 3.89 mM (sample D-3, Fig. 7e) indicates no obvious change in the particle size relative to sample D-2. However, the structures of the particle central parts differ in samples D-3 and D-2: a depression of the central area of the particle occurs in sample D-3 (Fig. 7e) (marked by arrows). As a result, the morphology changes from solid spherical to disk-like, indicating a looser internal structure formed upon increase of the PL concentration. These features are further revealed in the TEM image (Fig. 7f) showing a central depression in spheres, as indicated by a lighter contrast at the

Figure 6. 29Si MAS NMR of sample D-4 before (a) and after an 800 °C overnight calcination and modeled fit to Voigt Q2, Q3, and Q4 peaks. Table 2. Values of Q2, Q3, Q4, and the corresponding molar masses of sample D-4 before and after calcining Q2 (%)

Q3 (%) Q4 (%) Untreated sample 1 8.6 47.8 43.6 2 5.9 52.9 41.2 3 4.6 55.8 39.6 4 5.7 61.9 32.3 5 6.3 50.7 43.0 Calcined sample 1 – 2.2 97.8 2 – 3.1 96.9 3 – 3.8 96.2 4 – 3.3 96.7 5 – 2.9 97.1

Molar mass (g/mol) 66.03 65.91 65.94 66.70 65.80 60.27 60.36 60.43 60.39 60.34

part of them and influence the inner structures of silica. FTIR and TG-DTA analyses confirm this conclusion. To understand the effect of PL on the morphogenesis of the siliceous structures, mineralization experiments with different concentrations of PL were carried out. Figure 7 shows the FESEM and TEM images of the mineralized products at the PL concentrations of 0.60, 2.09, 3.89, and 7.47 mM with a DA fixed at 20 mM. At a PL concentration of 0.60 mM (sample D-1 in Table 1), spheres with smooth surfaces are obtained, as shown in Figure 7a, and the particle diameter is not uniform: it ranges from 150 to 900 nm. TEM micrograph (Fig. 7b) reveals presence of the discrete and spherical particles with a disordered, wormlike pore arrangement. Some structural differences are observed

Figure 7 continues on next page


1387

◄Figure 7. FESEM (a, c, e, and g) and TEM (b, d, f, and h) images of the silica products formed with different concentrations of PL: previous page (a and b) 0.60 mM, sample D-1; (c and d) 2.09 mM, sample D-2; (e and f) 3.89 mM, sample D-3; (g and h) 7.47 mM, sample D-5.

center. Combining with the half-open tube-like structures in sample D-4 (Fig. 1), it is reasonable to consider that the disklike particles in sample D-3 (Figs. 7e and 7f) correspond to a transition from a solid spherical to a hollow tube morphology. Figures 7g and 7h depict the SEM and TEM micrographs of sample D-5 prepared with an initial PL concentration of 7.47 mM. Interestingly, the final silica product corresponds to tadpole-like hollow particles, whose length can reach several micrometers.

The outer diameters of the “big head” and the “long tail” are ca. 900 and 500 nm, respectively. Wrinkly surface is also a feature observed in sample D-5. The tadpole-like structure of sample D-5 can be considered as a morphological evolution of the testtube-like morphology of sample D-4. The elongation of the open end and multi-layer silica coating of the other end of the tubes occur when the PL concentration is up to 7.47 mM. In summary, morphologies of the products change from

1388


spherical through test-tube-like to tadpole-looking features with increasing concentration of PL. Besides, the addition of PL into the system facilitates the formation of the particles. We conclude that the PL concentration is a key parameter for the formation of different architectures of silica. As the dosage of PL gradually increases, the inner structure of the particles becomes more and more loose, and disk-like shapes (sample D-3) can be seen as the intermediate between solid spheres and hollow tubes. Figure 8 shows the SAXRD patterns of the extracted silica samples obtained in the 20 mM DA system with increasing concentrations of PL. The broadness and low intensity of the low 2θ reflection with a maximum between 2θ = 1.5° and 2θ = 2.5° indicate that there is no long-range order in the pore arrangement in the treated SiO2 particles (Hu et al. 2010). This is in accordance with the TEM observations (Fig. 7). Referring to literature (Tanev and Pinnavaia 1995; Bagshaw et al. 1995), however, the single peak pattern of this type of material indicates the presence of a uniform distribution of pore diameters in the mesoporous range. The pore–pore correlation distance of the calcined samples increases from 38.2 Å in sample D-1 with the smallest size of particles to 56.9 Å in sample D-5 with the largest size of particles. Therefore, increasing the dosage of PL increases the pore diameter of the obtained silica particles. Figure 9 provides N2 adsorption-desorption isotherms for the silica prepared at PL concentrations from 0.60 to 7.47 mM and the corresponding samples extracted at 60 °C for 6 h; the analytic data are also given in Table 3. The isotherms of these mesoporous samples exhibit typical type-IV isotherms with a well-defined reversible adsorption step at P/P0 of about 0.4. An obvious H4-type hysteresis loops can also be observed, which confirms the existence of lamellar structures in these samples (Karkamkar et al. 2004), as can be seen in the TEM images (e.g., Figs. 1e and 1f). Capillary condensation of liquid nitrogen in the framework-confined mesopores occurs at P/P0 = 0.4–0.6 (Park and Pinnavaia 2009) and the hysteresis loop can be explained by the presence of inter-particles pores (Liu et al. 2009). The smallest hysteresis loop is observed for the sample prepared with a PL concentration of 0.60 mM, which can be attributed to the relatively ordered assembly of the mesostructure under low-PL content (Liu et al. 2009). The pore diameters of these materials increase gradually with the increase of PL content; the enlargement of d-spacing from 38.2 to 56.9 Å is also suggested by the XRD measurements. These data indicate that the inner structures of these particles become more and more loose as the dosage of PL increases, which is in good agreement with the TEM results. To identify the influence of the interaction between amines and lipids in biomimetic silicification, new series of experiments using ammonia instead of DA were carried out. The detailed experimental conditions and the corresponding morphologies of the products are listed in Table 1. Very little product was obtained at the concentration of PL 0.60 mM (sample A-1 in Table 1), therefore detailed analysis and characterization could not be performed. At a concentration of 2.09 mM PL (sample A-2), spherical particles with smooth surfaces were obtained, as illustrated in Figure 10a. The TEM image (Fig. 10b) unveils a porous structure with a disordered, wormlike pore arrangement, which is similar to sample D-1 (Fig. 7b). At 2.99 mM PL, most silica structures (sample A-3) are disk-like with depressed middle

Figure 8. XRD patterns of the 60 °C extracted silica particles formed with different PL concentrations. The numbers near the diffraction peak are the pore–pore correlation distances in angstrom (Å).

Figure 9. N2 adsorption–desorption isotherms of the samples D-1, D-2, D-3, D-4, and D-5. For clarity, the isotherms of D-1, D-2, D-3, D-4, and D-5 are offset along Y-axis for 0, 200, 250, 400, and 500 cm3/g, respectively. Table 3. Textural parameters of various samples after heating extraction Sample d-spacing SBET (m2/g) Pore size Vp (cm3/g) Wall thickness (nm) (nm) (nm) D-1 3.82 899 2.81 0.56 1.01 D-2 4.40 532 3.51 0.41 0.89 D-3 4.58 495 3.85 0.48 0.73 D-4 5.38 564 4.00 0.37 1.38 D-5 5.69 551 4.44 0.42 1.25

parts (Fig. 10c, marked by arrows). The particles in sample A-3 (Figs. 10c and 10d) have a uniform morphology relative to particles in sample D-3 (Figs. 7e and 7f). A further increase of the PL concentration to 3.89 mM shows the coexistence of both sheet-like and spherical particles (Fig. 10e). With the TEM, the


silica spheres observed in the FESEM image can be identified as hollow particles with rough surfaces (Fig. 10f) whose outer mesoporous shells are characteristic of multilayer structures (marked by arrow). The sheet-like structure can also be observed in Figure 10f (marked by arrowhead), where no obvious contrast between the inner and outer parts of the particle can be seen. At a PL concentration of 4.78 mM, the particles (sample A-5) are mainly isolated hollow spheres with multilayer silica shells while the sheet-like structures are absent (Figs. 10g and 10h). Some hollow structures begin to collapse at PL concentration of 5.68 mM (Figs. 10i and 10j). Finally, an irregular morphology is observed when the concentration of PL reaches 7.47 mM in the system (not shown). In general, the morphology of the product changes from solid particles to hollow spheres with increasing concentration of PL. A similar trend is observed in the experiments using DA as a catalyst (from the sample D-1 to D-5). However, no hollow tubes formed when ammonia is used instead of DA. Thus, the interaction between the amine and lipid can be seen as a decisive factor for the morphology transformation from hollow spherical to tubular particles.

Discussion These results demonstrate that phospholipid influences the surface texture and inner structure of silica (e.g., Figs. 1 and 7). As is well known, monodisperse SiO2 solid spheres can be readily prepared by the Stöber process (Stöber et al. 1968). Moreover, sphere morphology of silica can be obtained during biomimetic silicification using organic molecules extracted from diatoms frustules, such as silaffins and long-chain polyamines (Kröger et al. 2000; Sumper et al. 2003). In terms of long-chain polyamines, if added to monosilicic acid solution, the polyamines can induce the rapid precipitation of silica and can control the size of silica spheres in vitro (Kröger et al. 2000). However, In addition to sphere morphology, some special structures have also been observed during the biosilicification in diatoms. At the earliest recognizable stage of valve development in N. pelliculosa, the earliest valve is an elongated siliceous structure approximately 6 μm long (Chiappino and Volcani 1977). In contrast, the morphologies of the samples obtained in our DA-PL experiments change from solid nanospheres to hollow microtubes as a function of the starting concentration of PL from 0.60 to 5.68 mM (Figs. 1 and 7). Moreover, the lengths of microtubes increase from 1 μm (Fig. 1) to 5 μm (Figs. 7g and 7h) as PL concentration increase from 5.68 to 7.47 mM. It indicates that the presence of PL in the silicification region favors the formation and continuous growth of siliceous elongated structures. In addition, an obvious enlarged head was also observed while increasing the PL concentration to 7.47 mM (e.g., Figs. 1, 7g, and 7h), which is reminiscent of the slightly enlarged area in the central region of “primary central band” (Zurzolo and Bowler 2001; Chiappino and Volcani 1977). Recently, X‑ray photoelectron spectroscopy (XPS, Tesson et al. 2008) and solid-state NMR (SSNMR, Tesson et al. 2009) studies were performed on diatom cells for analyzing the chemical composition of the diatom surface. The XPS analysis revealed a high concentration of lipids indicating their presence as a structural part of the cell wall in the form of carboxylic esters. The SSNMR study also demonstrated that lipids are tightly associated with silica, even after harsh chemi-

1389

cal treatment (Tesson et al. 2008, 2009). Therefore, these studies suggest that phospholipid molecules, which extensively exist in biological membrane (e.g., silicalemma) and have been found in biosilica of diatoms (Chiappino and Volcani 1977; Bäuerlein 2003), would contribute not only to the formation of elongated structures but also to the localized enlargement of siliceous band. Silica microtubes can be produced by a biomimetic approach in the presence of peptidic lipid (Ji et al. 2007), cationic lipid (Ji et al. 2004), and glycolipid (Ji et al. 2005). However, to the best of our knowledge, no report on the preparation of test-tube-like or tadpole-looking silica structures (Figs. 1, 7g, and 7h) in the presence of phospholipid and organic amine can be found, and the length of microtubes can be easily controlled by tuning the PL concentration. Moreover, a series of control experiments using ammonia instead of DA demonstrate that no elongated structures can be obtained (Fig. 10), even when the concentration of PL reaches 7.47 mM. In more specific terms, the silica particles obtained with increasing PL concentration without the addition of DA were just solid spheres (Figs. 10a and 10b), disk-like particles (Figs. 10c and 10d), sheet-like structures and hollow spheres (Figs. 10e–10j). This indicates that the elongated silica structures cannot be formed in the absence of DA (organic amine with a carbon chain). Recent efforts have also been made to prepare silica just in the presence of organic amine; however, the mineralized products are dominated by solid or hollow spheres rather than silica tubes (Tanev and Pinnavaia 1996; Lin and Chen 2005; Zhang et al. 2007). In combination with the results mentioned above, both organic amine and phospholipid are required. Therefore, the formation of the test-tube-like hollow silica microtubes can be attributed to the interactions between PL and DA molecules. In our opinion, Si-O− groups electrostatically interact with ammonium head groups from PL and/or DA molecules. Thus silica formation can be induced by organic molecules. As a result, the shape-controlled formation of silica is quite dependent upon the morphogenesis of organic aggregates. Spherical vesicles of phospholipids, which have been extensively studied (Lipowaky and Sackmann 1995), are the most common self-assembled structure, and siliceous hollow spheres have also been prepared in the presence of phospholipids (Bégu et al. 2003, 2004). In our study, hollow spheres have also been observed (sample A-6) when 5.68 mM of PL was used in the NH3-PL experiments (Figs. 10i and 10j). By using DA instead of ammonia, test-tube-like hollow silica microtubes were obtained with the same concentration of phospholipide. In this case, the dodecyl chains of DA molecules insert among the phospholipid hydrophobic chains, while the NH2 or NH3+ heads interact with P-O− phospholipid groups by hydrogen bonding and electrostatic interaction (Galarneau et al. 2010). In other words, DA molecules probably insert into phospholipids and give rise to the structural changes observed in the self-assembled organic structures. Eventually, silica microtubes (Fig. 1) rather than hollow spheres (Figs. 10i and 10j) are obtained. The ability of LCPAs to produce silica depends on the ability of these organic amines to self-assemble (in vitro) into aggregates in solution (Sumper et al. 2003). Multivalent anions such as phosphate, sulfate, or citrate ions may cross-link LCPAs through the establishment of hydrogen bonds and electrostatic interactions, which is critical for silica formation (Kröger et al.

1390


Figure 10. FESEM and TEM images of the silica products formed with different PL concentrations using ammonia as catalyst: (a and b) A-2; (c and d) A-3; (e and f) A-4; next page (g and h) A-5; and (i and j) A-6.


1391

See Figure 10 caption on previous page

2000). Silicic acid species absorb in and/or onto the aggregates of the organic amines and multivalent anions, then polymerize into silica (Sumper and Kröger 2004). As seen above, these two kinds of interactions can also be established between PL and DA molecules. Thus, it is possible that they are able to influence silicification process in a similar way. In recent literature, polyallylamine was also used to catalyze the polycondensation of silicic acid. In this case, silica precipitation took place only above a so-called “threshold value” of phosphate concentration (Brunner et al. 2004). In our experiments, while the added amount of organic amine (DA) is fixed to 20 mM, intact nanospheres (sample D-1) are obtained (Figs. 7a and 7b) at a PL concentration of 0.60 mM. However, no obvious opaque floccules are observed in the absence of PL, other conditions being constant. Therefore, the aggregates formed between phospholipids with P-O− groups and organic amine molecules with NH2 or NH3+ heads also facilitate the formation of silica particles (Gonçalves da Silva and Romão 2005). In this context, it is likely that phospholipids are involved in the formation of organic aggregates, and thus facilitate the silica formation. Hildebrand et al. (2009) also found that

the silicalemma is tightly attached to the biosilica. This implies a genetic relation between membrane components of the silica deposition vesicle and the formation of silica structures.

Acknowledgments This work was financially supported by the Natural Science Foundation of China (No. 41172049), the Chinese Ministry of Science and Technology (No. 2011CB808800), and the Knowledge Innovation Program of the Chinese Academy of Sciences, Grant No. KZCX2-YW-QN501.

References cited Addadi, L., Raz, S., and Weiner, S. (2003) Taking advantage of disorder: Amorphous calcium carbonate and its roles in biomineralization. Advanced Materials, 15, 959–970. Agger, J.R., Anderson, M.W., Pemble, M.E., Terasaki, O., and Nozue, Y. (1998) Growth of quantum-confined indium phosphide inside MCM-41. The Journal of Physical Chemistry B, 102, 3345–3353. Anderson, H.C., Garimella, R., and Tague, S.E. (2005) The role of matrix vesicles in growth plate development and biomineralization. Frontiers in Bioscience, 10, 822–837. Bagshaw, S.A., Prouzet, E., and Pinnavaia, T.J. (1995) Templating of mesoporous molecular sieves by nonionic polyethylene oxide surfactants. Science, 269, 1242–1244. Bäuerlein, E. (2003) Biomineralization of unicellular organisms: An unusual mem-

1392


brane biochemistry for the production of inorganic nano- and microstructures. Angewandte Chemie International Edition, 42, 614–641. Bégu, S., Durand, R., Lerner, D.A., Charnay, C., Tourné-Péteilh, C., and Devoisselle, J.M. (2003) Preparation and characterization of siliceous material using liposomes as template. Chemical Communications, 640–641. Bégu, S., Girod, S., Lerner, D.A., Jardiller, N., Tourné-Péteilh, C., and Devoisselle, J.M. (2004) Characterization of a phospholipid bilayer entrapped into nonporous silica nanospheres. Journal of Materials Chemistry, 14, 1316–1320. Brunner, E., Lutz, K., and Sumper, M. (2004) Biomimetic synthesis of silica nanospheres depends on the aggregation and phase separation of polyamines in aqueous solution. Physical Chemistry Chemical Physics, 6, 854–857. Brutchey, R.L. and Morse, D.E. (2008) Silicatein and the translation of its molecular mechanism of biosilicification into low temperature nanomaterial synthesis. Chemical Reviews, 108, 4915–4934. Cauda, V., Muhlstein, L., Onida, B., and Bein, T. (2009) Tuning drug uptake and release rates through different morphologies and pore diameters of confined mesoporous silica. Microporous and Mesoporous Materials, 118, 435–442. Cha, J.N., Shimizu, K., Zhou, Y., Christiansen, S.C., Chmelka, B.F., Stucky, G.D., and Morse, D.E. (1999) Silicatein filaments and subunits from a marine sponge direct the polymerization of silica and silicones in vitro. Proceedings of the National Academy of Sciences, 96, 361–365. Chen, D.Y., Jiang, M.J., Li, N.J., Gu, H.W., Xu, Q.F., Ge, J.F., Xia, X.W., and Lu, J.M. (2010) Modification of magnetic silica/iron oxide nanocomposites with fluorescent polymethacrylic acid for cancer targeting and drug delivery. Journal of Materials Chemistry, 20, 6422–6429. Cheng, G. and Liu, C. (2002) Preparation of lamellar mesoporous silica microspheres via SDS templates. Materials Chemistry and Physics, 77, 359–364. Cheng, X., Liu, S., Lu, L., Sui, X., Meynen, V., Cool, P., Vansant, E.F., and Jiang, J. (2007) Fast fabrication of hollow silica spheres with thermally stable nanoporous shells. Microporous and Mesoporous Materials, 98, 41–46. Chiappino, M.L. and Volcani, B.E. (1977) Studies on the biochemistry and fine structure of silica-shell formation in diatoms. Protoplasma, 93, 205–221. Collier, J.H. and Messersmith, P.B. (2002) Phospholipid strategies in biomineralization and biomaterial research. Annual Review of Materials Research, 31, 237–263. Davis, A.K. and Hildebrand, M. (2008) Molecular processes of biosilicification in diatoms. In H. Sigel and A. Sigel, Eds., Metal Ions in Life Sciences. Biomineralization. From Nature to Application, vol. 4, p. 255–294. Wiley, London. Djerassi, C. and Lam, W.K. (1991) Phospholipid studies of marine organisms. Part 25. Sponge phospholipids. Accounts of Chemical Research, 24, 69–75. Early, T.A., Kundrat, J.T., Schorp, T., and Glonek, T. (1996) Lake Michigan sponge phospholipid variations with habitat: A 31P nuclear magnetic resonance study. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 114, 77–89. Galarneau, A., Sartori, F., Cangiotti, M., Mineva, T., Di Renzo, F., and Ottaviani, M.F. (2010) Sponge mesoporous silica formation using disordered phospholipid bilayers as template. The Journal of Physical Chemistry B, 114, 2140–2152. Genin, E., Wielgosz-Collin, G., Njinkoué, J.M., Velosaotsy, N.E., Kornprobst, J.M., Gouygou, J.P., Vacelet, J., and Barnathan, G. (2008) New trends in phospholipid class composition of marine sponges. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 150, 427–431. Gonçalves da Silva, A.M. and Romão, R.I.S. (2005) Mixed monolayers involving DPPC, DODAB and oleic acid and their interaction with nicotinic acid at the air–water interface. Chemistry and Physics of Lipids, 137, 62–76. Gorby, Y.A., Beveridge, T.J., and Blakemore, R.P. (1988) Characterization of the bacterial magnetosome membrane. Journal of Bacteriology, 170, 834–841. Hildebrand, M., Kim, S., Shi, D., Scott, K., and Subramaniam, S. (2009) 3D imaging of diatoms with ion-abrasion scanning electron microscopy. Journal of Structural Biology, 166, 316–328. Hu, J.K., Shan, W., Zhang, W.J., Zhang, Y.H., and Tang, Y. (2010) Morphological diversity of dual meso-structured HMS and their transformation process. Microporous and Mesoporous Materials, 129, 210–219. Ivanisevic, J., Pérez, T., Ereskovsky, A.V., Barnathan, G., and Thomas, O.P. (2011) Lysophospholipids in the Mediterranean sponge Oscarella tuberculata: seasonal variability and putative biological role. Journal of Chemical Ecology, 37, 537–545. Ji, Q., Iwaura, R., Kogiso, M., Jung, J.H., Yoshida, K., and Shimizu, T. (2004) Direct sol-gel replication without catalyst in an aqueous gel system: From a lipid nanotube with a single bilayer wall to a uniform silica hollow cylinder with an ultrathin wall. Chemistry of Materials, 16, 250–254. Ji, Q., Kamiya, S., Jung, J.H., and Shimizu, T. (2005) Self-assembly of glycolipids on silica nanotube templates yielding hybrid nanotubes with concentric organic and inorganic layers. Journal of Materials Chemistry, 15, 743–748. Ji, Q., Iwaura, R., and Shimizu, T. (2007) Regulation of silica nanotube diameters: Sol-gel transcription using solvent-sensitive morphological change of peptidic lipid nanotubes as templates. Chemistry of Materials, 19, 1329–1334. Kao, H.M., Liao, Y.W., and Ting, C.C. (2007) Synthesis of cubic mesoporous silica SBA-1 with bulky headgroup surfactant cetyltripropylammonium bromide. Microporous and Mesoporous Materials, 98, 80–88.

Karkamkar, A.J., Kim, S.S., Mahanti, S.D., and Pinnavaia, T.J. (2004) Lamellar mesostructured silicas with chemically significant hierarchical morphologies. Advanced Functional Materials, 14, 507–512. Kim, S.S., Pauly, T.R., and Pinnavaia, T.J. (2000) Non-ionic surfactant assembly of wormhole silica molecular sieves from water soluble silicates. Chemical Communications, 835–836. Kröger, N., Deutzmann, R., Bergsdorf, C., and Sumper, M. (2000) Species-specific polyamines from diatoms control silica morphology. Proceedings of the National Academy of Sciences, 97, 14133–14138. Kruk, M. and Jaroniec, M. (2001) Gas adsorption characterization of ordered organic-inorganic nanocomposite materials. Chemistry of Materials, 13, 3169–3183. Li, Y.H., Xu, C.L., Wei, B.Q., Zhang, X.F., Zheng, M.X., Wu, D.H., and Ajayan, P.M. (2002) Self-organized ribbons of aligned carbon nanotubes. Chemistry of Materials, 14, 483–485. Li, J., Xu, Y., Wu, D., and Sun, Y. (2004) Hydrothermal synthesis of novel sandwich-like structured ZnS/octylamine hybrid nanosheets. Solid State Communications, 130, 619–622. Lin, H.Y. and Chen, Y.W. (2005) Preparation of spherical hexagonal mesoporous silica. Journal of Porous Materials, 12, 95–105. Lipowaky, R. and Sackmann, E. (1995) Structure and dynamics of membranes, 1052 p. Elsevier, Amsterdam. Liu, S., Rao, J.C., Sui, X., Cool, P., Vansant, E.F., Van Tendeloo, G., and Cheng, X. (2008) Preparation of hollow silica spheres with different mesostructures. Journal of Non-Crystalline Solids, 354, 826–830. Liu, X.B., Sun, H., Chen, Y., Yang, Y.H., and Borgna, A. (2009) Preparation of spherical large-particle MCM-41 with a broad particle-size distribution by a modified pseudomorphic transformation. Microporous and Mesoporous Materials, 121, 73–78. Losic, D., Mitchell, J.G., and Voelcker, N.H. (2009) Diatomaceous lessons in nanotechnology and advanced materials. Advanced Materials, 21, 2947–2958. Lowenstam, H.A. (1981) Minerals formed by organisms. Science, 211, 1126–1131. Lowenstam, H.A. and Weiner, S. (1989) On Biomineralization, 324 p. Oxford University Press, New York. Meegan, J.E., Aggeli, A., Boden, N., Brydson, R., Brown, A.P., Carrick, L., Brough, A.R., Hussain, A., and Ansell, R.J. (2004) Designed self-assembled betasheet peptide fibrils as templates for silica nanotubes. Advanced Functional Materials, 14, 31–37. Michaux, F., Carteret, C., Stébé, M.J., and Blin, J.L. (2004) A facile approach to load CdSe nanocrystallites into mesoporous SBA-15. Materials Chemistry and Physics, 88, 192–196. ——— (2008) Hydrothermal stability of mesostructured silica prepared using a nonionic fluorinated surfactant. Microporous and Mesoporous Materials, 116, 308–317. Palsdottir, H. and Hunte, C. (2004) Lipids in membrane protein structures. Biochimica et Biophysica Acta, 1666, 2–18. Park, I. and Pinnavaia, T.J. (2009) Large-pore mesoporous silica with threedimensional wormhole framework structures. Microporous and Mesoporous Materials, 118, 239–244. Perry, C.C. and Keeling-Tucker, T. (2000) Biosilicification: the role of the organic matrix in structure control. Journal of Biological Inorganic Chemistry, 5, 537–550. Qu, X.F., Yao, Q.Z., Zhou, G.T., Fu, S.Q., and Huang, J.L. (2010) Formation of hollow magnetite microspheres and their evolution into Durian-like architectures. Journal of Physical Chemistry C, 114, 8734–8740. Reimann, B.E.F., Lewin, J.C., and Volcani, B.E. (1965) Studies on the biochemistry and fine structure of silica shell formation in diatoms. I. The structure of the cell wall of Cylindrotheca fusiformis. Journal of Cell Biology, 24, 39–55. Round, F., Crawford, R., and Mann, D. (1990) The Diatoms, 745 p. Cambridge University Press, U.K. Sadasivan, S., Khushalani, D., and Mann, S. (2005) Synthesis of calcium phosphate nanofilaments in reverse micelles. Chemistry of Materials, 17, 2765–2770. Schmid, A.-M.M. and Schulz, D. (1979) Wall morphogenesis in diatoms: deposition of silica by cytoplasmic vesicles. Protoplasma, 100, 267–288. Schmid, A.-M.M. and Volcani, B.E. (1983) Wall morphogenesis in Coscinodiscus wailesii. I. Valve morphology and development of its architecture. Journal of Phycology, 19, 387–402. Schröder, H.C., Wang, X., Tremel, W., Ushijima, H., and Müller, W.E. (2008) Biofabrication of biosilica-glass by living organisms. Natural Product Reports, 25, 455–474. Seddon, A.M., Patel, H.M., Burkett, S.L., and Mann, S. (2002) Chiral templating of silica-lipid lamellar mesophase with helical tubular architecture. Angewandte Chemie International Edition, 41, 2988–2991. Sigel, A., Sigel, H. and Sigel, R.K.O. (Eds.) (2008) Biomineralization: From Nature to Application. 4. Metal Ions in Life Sciences, 700 p. Wiley, New York. Sing, K.S.W., Everett, D.H., Haul, R.A.W., Moscou, L., Pierotti, R.A., Rouquérol, J., and Siemieniewska, T. (1985) Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure and Applied Chemistry, 57, 603–619. Stöber, W., Fink, A., and Bohn, E. (1968) Controlled growth of monodisperse

SHI ET AL.: MORPHOGENESIS OF AMORPHOUS SILICA AND BIOSILICIFICATION silica spheres in the micron size range. Journal of Colloid and Interface Science, 26, 62–69. Sumper, M. and Brunner, E. (2006) Learning from diatoms: Nature’s tools for the production of nanostructured silica. Advanced Functional Materials, 16, 217–228. Sumper, M. and Kröger, N. (2004) Silica formation in diatoms: the function of longchain polyamines and silaffins. Journal of Materials Chemistry, 14, 2059–2065. Sumper, M., Lorenz, S., and Brunner, E. (2003) Biomimetic control of size in the polyamine-directed formation of silica nanospheres. Angewandte Chemie International Edition, 42, 5192–5195. Tanev, P.T. and Pinnavaia, T.J. (1995) A neutral templating route to mesoporous molecular sieves. Science, 267, 865–867. ——— (1996) Biomimetic templating of porous lamellar silicas by vesicular surfactant assemblies. Science, 271, 1267–1269. Tesson, B., Masse, S., Laurent, G., Maquet, J., Livage, J., Martin-Jezequel, V., and Coradin, T. (2008) Contribution of multi-nuclear solid state NMR to the characterization of the Thalassiosira pseudonana diatom cell wall. Analytical and Bioanalytical Chemistry, 390, 1889–1898. Tesson, B., Genet, M.J., Fernandez, V., Degand, S., Rouxhet, P.G., and MartinJézéquel, V. (2009) Surface chemical composition of diatoms. ChemBioChem, 10, 2011–2024. Uriz, M.J. (2006) Mineral skeletogenesis in sponges. Canadian Journal of Zoology, 84, 322–356. Venkatathri, N., Srivastava, R., Yun, D.S., and Yoo, J.W. (2008) Synthesis of a novel class of mesoporous hollow silica from organic templates. Microporous and Mesoporous Materials, 112, 147–152. Vidyadhar, A., Hanumantha Rao, K., and Chernyshova, I.V. (2003) Mechanisms

1393

of amine/feldspar interaction in the absence and presence of alcohols studied by spectroscopic methods. Colloids and Surfaces A, 214, 127–142. Wang, J.C., and Liu, Q. (2005) Structural change and characterization in nitrogenincorporated SBA15 oxynitride mesoporous materials via different thermal history. Microporous and Mesoporous Materials, 83, 225–232. Weiner, S., and Addadi, L. (1997) Design strategies in mineralized biological materials. Journal of Materials Chemistry, 7, 689–702. Zhang, L.X., Li, P.C., Liu, X.H., Du, L.W., and Wang, E.K. (2007) The effect of template phase on the structures of as-synthesized silica nanoparticles with fragile didodecyl-dimethylammonium bromide vesicles as templates. Advanced Materials, 19, 4279–4283. Zhao, Y.F., Qi, Y., Wei, Y.X., Zhang, Y.Y., Zhang, S.G., Yang, Y., and Liu, Z.M. (2008) Incorporation of Ag nanostructures into channels of nitrided mesoporous silica. Microporous and Mesoporous Materials, 111, 300–306. Zhou, Y., Shimizu, K., Cha, J.N., Stucky, G.D., and Morse, D.E. (1999) Efficient catalysis of polysiloxane synthesis by silicatein a requires specific hydroxyl and imidazole functionalities. Angewandte Chemie International Edition, 38, 779–782. Zurzolo, C. and Bowler, C. (2001) Exploring bioinorganic pattern formation in diatoms. A story of polarized trafficking. Plant Physiology, 127, 1339–1345.

Manuscript received December 18, 2011 Manuscript accepted May 18, 2012 Manuscript handled by Daniel Vielzeuf