Silica–Carbonate Biomorphs: Emergent Materials

• Silica–carbonate biomorphs are self-organized inorganic composites formed by co-precipitation of amorphous silica and alkaline-earth carbonates under high pH conditions, yielding diverse life-like structures.
• They exhibit intricate nanometric ordering with aligned carbonate crystallites embedded in a silica matrix, characterized using advanced texture tomography and X-ray diffraction techniques.
• Geometric growth theories unify their morphogenesis by linking edge-driven accretion, diffusion-limited kinetics, and pH-dependent silica chemistry to produce complex hierarchical forms.

Searching arXiv for recent and foundational papers on silica-carbonate biomorphs and related geometric growth theory. Silica–carbonate biomorphs are self-organized inorganic composite microstructures formed by co-precipitation of an alkaline-earth carbonate mineral—typically BaCO$_3$, SrCO$_3$, or CaCO$_3$—together with amorphous silica under alkaline conditions. They are composed of nanometric carbonate crystallites embedded in or separated by an amorphous silica matrix, and they self-organize into a multitude of shapes with a distinct long-range order of the carbonate nanocrystals. The term designates a class of abiotic materials that can generate thin-walled sheets, leaves, worms, single or double helices, corals, cups, and related life-like forms through geochemical self-organization rather than biological templating. Current research treats them simultaneously as model systems for emergent hierarchical materials, as probes of abiotic morphogenesis relevant to astrobiology and microfossil interpretation, and as structurally ordered platforms for optical, electronic, or magnetic functionalization (Kaplan et al., 2021, Frewein et al., 28 Aug 2025).

1. Composition, morphology, and status as emergent materials

Silica–carbonate biomorphs are composite precipitates in which the carbonate phase is not a single crystal. Instead, the structures are built from nanometric carbonate crystallites surrounded by amorphous silica. In silica-witherite biomorphs, the carbonate component is witherite, BaCO$_3$, and the nanocrystals are reported as rod-shaped witherite crystallites elongated along the crystallographic $c$-axis and assembled into a co-aligned, textured polycrystalline architecture. Local Rietveld line-profile analysis with a prolate ellipsoid model gave average crystallite sizes corresponding roughly to lengths of 20–40 nm and diameters of 10–20 nm. With reconstructed voxel sizes of 250–500 nm, each voxel contains on the order of $10^2$–$10^3$ crystallites (Frewein et al., 28 Aug 2025).

At the microscale, these nanocrystals collectively organize into elaborate morphologies. Reported examples include leaves, worms, single or double helices, corals, vaselike, coral-like, and helical precipitates, as well as curved sheets, scrolls, leaves, warped bowls, fan-like forms, and thin wall-like composite precipitates. The recurrence of such morphologies across carbonate–silica systems motivates the use of “biomorph” as a morphological term rather than as a claim about biological origin.

A common misconception is that biomorphs are simply mineral objects that visually imitate organisms while remaining structurally trivial. The current literature indicates the opposite. Their significance lies not only in external shape but also in internal organization: they possess a local fiber texture, distinct growth motifs, and crystallographic properties that vary between morphologies and within a single structure. A plausible implication is that the biomorph concept is best understood as a coupled problem of morphology, texture, and local reaction environment, not merely of silhouette or gross shape (Frewein et al., 28 Aug 2025).

2. Edge-driven growth as a geometric description of biomorph morphogenesis

A central theoretical description treats biomorph-like precipitates as thin-walled structures that grow by localized accretion at a moving edge. In this formulation, the wall is represented as a smooth two-dimensional surface $\vec{\mathbf{X}}(\sigma,t)$ embedded in $\mathbb{R}^3$, and the active rim is an evolving space curve whose temporal wake constitutes the deposited wall. The basic kinematic law is

$\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$

where $_3$0 is the local Lagrangian growth speed and $_3$1 is the in-surface growth direction. The tangent to the edge, the growth direction, and the surface normal form an orthonormal triad,

$_3$2

with $_3$3. The key geometric constraint is that $_3$4 lies in the tangent plane of the already formed surface, not along the surface normal $_3$5, so the deposited wall remains smooth (Kaplan et al., 2021).

The geometric state is encoded by six scalar fields: the local line metric $_3$6, geodesic curvature $_3$7, normal curvature $_3$8, geodesic torsion $_3$9, second normal curvature $_3$0, and growth speed $_3$1. These are defined by

$_3$2

$_3$3

In this framework, $_3$4 measures in-plane bending of the active rim, $_3$5 measures out-of-plane bending of the rim, $_3$6 measures twisting of the surface-adapted frame, and $_3$7 quantifies curling in the growth direction. For biomorphs, these variables furnish a compact description of lobes, petals, shells, helices, scrolls, and warped sheets without treating each category as a separate phenomenon.

The compatibility relations that ensure the moving edge leaves behind a valid smooth surface are the Codazzi–Mainardi equations and the Gauss equation. In the co-moving frame they yield

$_3$8

$_3$9

together with metric evolution,

$_3$0

and the Gauss relation

$_3$1

which leads to

$_3$2

These equations make explicit that in-plane curvature, out-of-plane curvature, torsion, curling, and local speed must coevolve consistently. This suggests a unifying morphogenetic language in which flowers, leaves, shells, vases, and helices correspond to different trajectories in the coupled dynamical system $_3$3 (Kaplan et al., 2021).

3. Constitutive closure, diffusion-limited behavior, and chemical interpretation

The geometric equations are not closed until $_3$4 and the curling dynamics are specified. One constitutive law introduced for the edge speed is a symmetry-based power series in geometric invariants,

$_3$5

with

$_3$6

The term $_3$7 acts like line tension along the rim and suppresses unstable outward kinks, analogous to Mullins–Sekerka stabilization in dendritic solidification. In the same theory, a deliberately simple closure for mean-curvature evolution is

$_3$8

For $_3$9, local nonuniformity in $c$0 is amplified; for $c$1, the surface evolves toward low mean curvature. In biomorph language, this controls whether walls become more rolled and folded or instead tend toward lower mean curvature while redistributing curvatures (Kaplan et al., 2021).

A more biomorph-specific earlier closure, reproduced in the same work for BaCO$c$2–SiO$c$3 coprecipitation, is

$c$4

Here the first term relaxes curling along the edge with diffusivity $c$5, and the second introduces a source term involving a coarse-grained bending parameter $c$6, assumed inversely proportional to local pH. The same model is interpreted באמצעות a Péclet number,

$c$7

where $c$8 is the time-dependent edge circumference. This directly links geometric morphogenesis to experimental control variables by making pH enter through $c$9 and transport competition enter through $10^2$0 (Kaplan et al., 2021).

The experimental motivation for such closures includes reported growth data with $10^2$1 and a growth instability increasing with $10^2$2, both interpreted as hallmarks of diffusion-limited growth. The same earlier formulation explained vaselike, coral-like, and helical precipitates and predicted pH-dependent sequential growth pathways that were later synthesized. The theory is therefore geometric and kinematic rather than chemically microscopic: it does not derive growth speed from explicit transport equations for ions, silica polymerization, carbonate speciation, surface reaction kinetics, elasticity, fracture, or grain orientation, but instead assumes that those effects are coarse-grained into constitutive parameters. The paper explicitly states that closure relations require input from physical chemistry and that resolving the scale separation and non-locality of species transport remains an open challenge. This suggests that geometric compatibility is necessary but not sufficient for a full mechanistic theory of biomorph formation (Kaplan et al., 2021).

4. Three-dimensional crystallographic texture and internal heterogeneity

A major recent advance is the use of X-ray texture tomography and X-ray diffraction computed tomography to map silica-witherite biomorphs in three dimensions and with submicrometric spatial resolution. Texture tomography reconstructs the local orientation distribution function in each voxel and yields the nematic director $10^2$3, representing the average preferred direction of the long crystal axis, together with the nematic order parameter $10^2$4, which measures alignment from $10^2$5 for no directional order to $10^2$6 for perfect alignment. The order tensor is defined as

$10^2$7

where $10^2$8 denotes orientation, $10^2$9 is the unit vector along the crystalline $10^3$0-axis, and the largest eigenvalue and corresponding eigenvector give $10^3$1 and $10^3$2, respectively. For these biomorphs, the local texture is generally a fiber texture, meaning crystallites share a common axis direction but are azimuthally random around that axis. Pole figures were plotted in multiples of a random distribution, and the ODFs were reconstructed using a hyperspherical harmonic model up to order 8 (Frewein et al., 28 Aug 2025).

Complementary XRD-CT reconstructs local powder diffraction patterns voxel by voxel. Rietveld refinement was used to extract crystalline material density or scale factor, lattice parameters and unit-cell volume $10^3$3, crystallite size, and anisotropic shape broadening. Peak profiles were fitted with a pseudo-Voigt model including isotropic and anisotropic size broadening, and crystallites were approximated as prolate ellipsoids with rotational symmetry around the major axis, again taking the $10^3$4-axis as the anisotropic direction. From this, particle volume $10^3$5 and anisotropy $10^3$6 were obtained, with $10^3$7 ranging from about 1.1 to 2.1 (Frewein et al., 28 Aug 2025).

The experiments were performed at ESRF ID13 with X-ray energy $10^3$8 keV, beam size $10^3$9 nm, a detector $\vec{\mathbf{X}}(\sigma,t)$0-range usable from 0.5–32 nm$\vec{\mathbf{X}}(\sigma,t)$1 and up to 40 nm$\vec{\mathbf{X}}(\sigma,t)$2 at the detector corners, step size 250–500 nm, about 260 projections over 10 tilt angles from $\vec{\mathbf{X}}(\sigma,t)$3 to $\vec{\mathbf{X}}(\sigma,t)$4, and exposure time 2 ms per pattern. The resulting conclusion is that silica-witherite biomorphs are not crystallographically uniform objects. Their texture, crystal size and shape, and lattice state vary both between different biomorph morphologies and within a single biomorph. A common misconception—that a given biomorph morphology corresponds to a single uniform crystallographic program—is therefore not supported by the tomographic evidence (Frewein et al., 28 Aug 2025).

5. Morphology-specific growth motifs: leaves, corals, helices, and worms

The leaf morphology is a sheet-like structure about 2 $\vec{\mathbf{X}}(\sigma,t)$5m thick and roughly 50 $\vec{\mathbf{X}}(\sigma,t)$6m long from a single nucleation center. Texture tomography shows that along the central axis crystallites are aligned with the growth direction, while toward the sides the director diverges and, at the outer edges, crystallites become aligned transverse or normal to the sheet. The order parameter is moderate, with about $\vec{\mathbf{X}}(\sigma,t)$7 in central and outer-edge domains, $\vec{\mathbf{X}}(\sigma,t)$8 in intermediate domains, and down to $\vec{\mathbf{X}}(\sigma,t)$9 in the terminal region where a kink and bending occur. The nucleus has larger particles, about $\mathbb{R}^3$0 nm$\mathbb{R}^3$1, and anisotropy up to 2.0, compared with about $\mathbb{R}^3$2 nm$\mathbb{R}^3$3 and anisotropy around 1.7 in the rest of the leaf. The unit-cell volume starts near 305.9 Å$\mathbb{R}^3$4 at the nucleation center, rises to about 306.5 Å$\mathbb{R}^3$5 across most of the sheet, and reaches 306.8 Å$\mathbb{R}^3$6 in the low-order terminus. Importantly, when the leaf bends, local nematic directors remain parallel to each other rather than bending with the macroscopic curvature. The same holds for the secondary curled sheet. The authors therefore identify a motif of bending by staggering or offset packing rather than continuous rotation of each crystallite with the sheet (Frewein et al., 28 Aug 2025).

The coral morphology consists of sheet-like and cone-like domains with radius about 20 $\mathbb{R}^3$7m growing approximately isotropically from a single point. It forms at higher Ba supersaturation and lower initial pH than the other morphologies. Two orientation motifs were identified: in cone-like domains, crystallites are oriented transverse to the radial growth direction; in sheet-like domains, crystallites are oriented along the growth direction. The order parameter is lower overall than in leaf or helix morphologies, roughly $\mathbb{R}^3$8–0.35, and some domains show higher order on the lower side facing the substrate or surface. Strong variation in crystallite density occurs, particularly at the bottoms of cone segments where crystalline density is higher; these regions were suggested to represent rapid BaCO$\mathbb{R}^3$9-rich precipitation with little or no silica co-precipitation. The coral accordingly exemplifies a fast-growth, high-supersaturation regime in which multiple motifs coexist (Frewein et al., 28 Aug 2025).

The double helix nucleates from a dense ellipsoidal center about 5 $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$0m in diameter, then develops asymmetrically into a hollow sheet-like oval structure on one side and a connecting sheet plus curling strands on the other, ultimately yielding a double helix. The helix diameter decreases from about 20 $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$1m initially to about 10 $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$2m when tightly wound. Texture mapping shows a developmental sequence: the lower part of the nucleus contains crystallites aligned parallel to the main helix axis; then the central region grows upward and crystal directions bend by about $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$3 to both sides; the two strands then grow downward with axes in transverse growth direction; as the helix tightens, crystallites become increasingly aligned with the local growth direction. Order parameters reach about 0.5 in relatively straight sheet-like parts, including side walls of the hollow structure and the connecting sheet between strands, but are lower in more complex curling regions. The lower part of the nucleus has higher order, around 0.5, and larger particles, about $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$4 nm$\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$5, whereas the upper part has smaller particles, about $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$6 nm$\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$7, and expanded lattice, about 307.2 Å$\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$8. This morphology makes the developmental sequence nucleation $\frac{d\vec{\mathbf{X}}}{dt}=\hat{\mathbf{n}}\,U,$9 sheet-like outgrowth $_3$00 curling strands $_3$01 tightened helix crystallographically explicit (Frewein et al., 28 Aug 2025).

The worm is a single helical or rolled structure built by growth along a central fiber, from which material is deposited outward in a helical manner. It has radial size about 10 $_3$02m and pitch about 5 $_3$03m. The director field is straight along the long axis at the center and bends toward the normal direction at larger radii, supporting a model in which a central core forms first and later layers are deposited outward while curling. The order parameter decreases from about $_3$04 in the central region to about $_3$05 near the outer surface. The strand shows a core-shell organization. The core contains small particles, about $_3$06 nm$_3$07, higher anisotropy up to 1.6, and unit-cell volume around 306.2 Å$_3$08; the shell contains larger particles, about $_3$09 nm$_3$10, lower anisotropy about 1.4–1.5, and a smaller unit-cell volume reported as about 303.8 Å$_3$11. The text notes that the shell value is notably lower than most other reported $_3$12 values and may reflect either a distinct local state or a likely typographic inconsistency, but the qualitative point is that core and shell differ markedly in lattice and crystal-growth state (Frewein et al., 28 Aug 2025).

Taken together, these morphology-specific observations yield recurring growth motifs: straight sheet growth with crystallites aligned with growth direction; bending by staggering, in which the sheet bends while local crystallites remain essentially parallel; curling along crystal axis or curved growth, in which local orientation follows structural curvature; and transverse orientation in curled structures. The suggestion that transverse orientation may act as a termination criterion for leaf morphology or trigger transition from straight growth to curling is explicitly presented as a mechanistic proposal rather than a direct in situ observation (Frewein et al., 28 Aug 2025).

6. Synthesis conditions, chemical regimes, significance, and unresolved issues

For silica-witherite biomorphs, the reported standard synthesis used equal volumes of 8.9 mM sodium metasilicate and 5 mM BaCl$_3$13 at initial $_3$14, followed by exposure to atmospheric CO$_3$15, which diffuses into solution and drives carbonate formation. Under these conditions, the reaction yields a mixture of morphologies including sheets or leaves and single and double helices. Coral-like structures required much higher barium concentration and lower initial pH, specifically BaCl$_3$16=250 mM and starting $_3$17. Corals formed faster, in about 1.5–2 h, whereas helicoidal structures took around 6 h (Frewein et al., 28 Aug 2025).

The crystallographic data are interpreted in terms of at least two chemically distinct growth regimes. In an early or high-pH regime, silicate is thought to be predominantly monomeric and coupling between silicate and precipitating carbonate is weaker, corresponding to larger crystallites, often higher anisotropy, and often higher texture order. This is observed especially in the leaf nucleus, lower helix nucleus, and sheet-like regions of the helix. In a later regime, as local reaction conditions evolve through precipitation, pH change, ion depletion, and silicate polymerization or oligomerization, more oligomerized silica is thought to interact more strongly with carbonate crystallites and inhibit growth, corresponding to smaller crystallites, altered anisotropy, expanded unit cells, and often lower or more heterogeneous order. This is most pronounced in the cores of helix and worm strands. The inference is that co-precipitating silica is not simply a passive matrix but actively shapes carbonate crystallite growth by spatially heterogeneous inhibition as its own chemistry evolves. Because the role of silica monomers versus oligomers is inferred from structural trends and prior chemistry literature rather than measured directly in situ in the same voxelated experiment, this interpretation remains mechanistic but indirect (Frewein et al., 28 Aug 2025).

The significance of biomorph research extends in several directions. For geochemistry and astrobiology, biomorphs strengthen the case that highly life-like mineral forms can emerge from purely inorganic processes while possessing complex internal order. For the problem of distinguishing abiotic biomorphs from biological forms, the literature suggests that spatially resolved texture motifs, local crystal-orientation fields, and crystal-size or core-shell signatures may provide fingerprints of abiotic mineral self-organization. For functional materials, the presence of well-defined but spatially varying crystal alignment implies that tensorial properties such as conductivity, dielectric permittivity, or magnetic susceptibility may ultimately be predictable from the reconstructed three-dimensional orientation field (Frewein et al., 28 Aug 2025).

Several limits and controversies remain explicit. The geometric growth theory is primarily a geometric or kinematic theory with phenomenological closure laws and does not derive speed or curling from first-principles transport-reaction equations, silica condensation chemistry, finite-thickness elasticity, fracture, or anisotropic crystallographic growth (Kaplan et al., 2021). The tomographic work, although bulk-sensitive and three-dimensional, does not directly measure local silica speciation, cannot yet provide fully dynamic correlation between growth and local chemistry, and leaves unresolved the exact nature of lattice occlusions and the precise mechanism by which silica influences carbonate growth (Frewein et al., 28 Aug 2025). More broadly, morphology and size cannot yet be fully controlled because many reaction parameters are strongly correlated, and the nature of silica–crystal interaction remains debated.

A plausible synthesis of the current literature is that silica–carbonate biomorphs are best understood as spatially heterogeneous self-organizing composites in which thin-wall geometry, local crystal texture, diffusion-limited transport, pH-dependent curling tendencies, and evolving silica chemistry are inseparable. In that view, the characteristic biomorph repertoire—vases, corals, shells, worms, helices, leaves, and related curved sheets—does not arise from a single scalar instability but from coupled geometric compatibility and local crystallographic growth regimes acting across scales (Kaplan et al., 2021, Frewein et al., 28 Aug 2025).