mechanical properties of modern calcite- (mergerlia truncata) and phosphate-shelled brachiopods...

Journal of Structural Biology 168 (2009) 396–408

Contents lists available at ScienceDirect

Journal of Structural Biology

journal homepage: www.elsevier .com/ locate/yjsbi

Mechanical properties of modern calcite- (Mergerlia truncata) andphosphate-shelled brachiopods (Discradisca stella and Lingula anatina)determined by nanoindentation

Casjen Merkel a, Julia Deuschle b, Erika Griesshaber a,*, Susan Enders c, Erwin Steinhauser d,Rupert Hochleitner e, Uwe Brand f, Wolfgang W. Schmahl a

a Department für Geo- und Umweltwissenschaften and GeoBioCenter, LMU Munich, Germanyb Max-Planck-Institut für Metallforschung, Stuttgart, Germanyc Department of Engineering Mechanics, University of Nebraska – Lincoln, USAd University of Applied Sciences, Munich, Germanye Mineralogische Staatssammlung München, Munich, Germanyf Department of Earth Sciences, Brock University, St. Catharines, Ontario, Canada

a r t i c l e i n f o a b s t r a c t

Article history:Received 30 March 2009Received in revised form 26 August 2009Accepted 27 August 2009Available online 1 September 2009

Keywords:NanoindentationNanohardnessElastic modulusCalcitic and phosphatic brachiopod shellsLaminated nanocompositeCross-laminated fibrous microstructureNacreBone

1047-8477/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.jsb.2009.08.014

* Corresponding author.E-mail address: [email protected]

We measured distribution patterns of hardness and elastic modulus by nanoindentation on shells of therhynchonelliform brachiopod Mergerlia truncata and the linguliform brachiopods Discradisca stella andLingula anatina. The rhynchonelliformea produce calcitic shells while the linguliformea produce chitino-phosphatic shells. Dorsal and ventral valves, commissure and hinge of the calcitic shell of M. truncatashow different nanohardness values (from 2.3 to 4.6 GPa) and E-modulus (from 52 to 76 GPa). The hard-ness of the biocalcite is always increased compared to inorganic calcite. We attribute the effects to dif-ferent amounts of inter- and intracrystalline organic matrix. Profiles parallel to the radius of curvatureof the valves cutting through the different layers of shell material surprisingly show quite uniform valuesof nanohardness and modulus of elasticity. Nanoindentation tests on the chitinophosphatic brachiopodsD. stella and L. anatina reflect the hierarchical structure composed of laminae with varying degree ofmineralization. As a result of the two-phase composite of biopolymer nanofibrils reinforced with Ca-phosphate nanoparticles, nanohardness, and E-modulus correlate almost linearly from (H = 0.25 GPa,E = 2.5 GPa) to (H = 2.5 GPa, E = 50 GPa). The mineral provides stiffness and hardness, the biopolymerprovides flexibility; and the composite provides fracture toughness. Gradients in the degree of mineral-ization reduce potential stress concentrations at the interface between stiff mineralized and soft non-mineralized laminae. For the epibenthic chitinophosphatic D. stella the lamination is also present but lesspronounced than for the infaunal L. anatina, and the overall distribution of material strength in the cross-sectional profile shows a maximum in the center and a decrease towards the inner and outer shell mar-gins (modulus of elasticity from 30 to 12 GPa, hardness from 1.7 to 0.5 GPa). Accordingly, the two epiben-thic forms, calcitic M. truncata and chitinophosphatic D. stella display fairly bulky (homogeneous)nanomechanical properties of their shell materials, while the burrowing infaunal L. anatina is distinc-tively laminated. The strongly mineralized laminae, which provide the strength to the shell, are also brit-tle, but keeping them as thin as possible, allows some bending flexibility. This flexibility is not requiredfor the epibenthic life style.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

Brachiopods are sessile marine invertebrates that have beenexisting since the early Cambrian (e.g., Rudwick, 1959; Williamset al. 1994). They still live in a wide range of marine habitats andmineralize either low-Mg calcite (the Rhynchonelliformea and

ll rights reserved.

e (E. Griesshaber).

Craniiformea) or Ca-phosphate (the Linguliformea) shells (Williamset al., 1997). Brachiopods are ideal animals for a multitude of stud-ies. Due to their long geologic record in distinct habitats they are ofinterest for evolutionary systematics (e.g., Williams et al., 1994,1997, 1998a) and for studies of paleoclimatic and paleoenviron-ment variations (e.g., Veizer et al., 1999; Bruckschen et al., 1999;Brand et al., 2003; Parkinson et al., 2005). The utilization of twodistinct shell materials (carbonate and phosphate) with distinctshell design strategies renders them highly appropriate for under-

http://dx.doi.org/10.1016/j.jsb.2009.08.014

mailto:[email protected]

http://www.sciencedirect.com/science/journal/10478477

http://www.elsevier.com/locate/yjsbi

C. Merkel et al. / Journal of Structural Biology 168 (2009) 396–408 397

standing biomineralization processes and properties of biomateri-als (e.g., Cusack et al., 1999; Schmahl et al., 2004a, 2006, 2008;Griesshaber et al., 2005b; Merkel et al., 2007; Perez-Huerta et al.,2007).

The shell of the phosphatic brachiopod Lingula anatina is lami-nated. It contains two distinct materials: an entirely organic andsoft primary layer that is followed inward towards the soft tissueof the animal, by a laminated chitinophosphatic secondary layer(Williams et al., 1994; Merkel et al., 2007). The organic primarylayer shields the secondary layer from incipient cracks that can oc-cur while the animal burrows itself into and through the sediment.The secondary layer is built of stacks of alternating mineralizedand non-mineralized b-chitin sheets (Williams et al., 1998a;Williams and Cusack, 1999). These mineralized-organic lamina-tions occur several times within the secondary layer, in such away that a hard and mineralized compact layer lies always ontop of a soft organic layer. The innermost shell portion, the layernext to the soft tissue of the animal, the basal plate, is formed bya hard mineralized layer. The chitin fibrils of all these layers aretwisted around each other and are grouped in bundles (Merkel

Fig. 1. Orientational contrast SEM images of (a and b) highlight microstructuralfeatures of the calcitic brachiopod M. truncata. Well visible is the distinctness inmicrostructure between the primary and the secondary shell layers (a), as well asthe differently oriented stacks of fibers of the secondary layer that results in asection cut through the shell in longitudinally and transversely cut fibers (b andGriesshaber et al. 2007).

et al., 2007). It has been suggested that the mineral that surroundsthe fibers is francolite (Lévêque et al., 2004), a carbonate-substituted apatite-like mineral. These mineral particles haveeither a spherical (Williams et al., 1994) or a cylindrical shape(Merkel et al., 2007). They are aligned onto chitin chains such thattheir crystallographic c-axis is parallel to the chitin fiber axis(Iwata, 1981) and coat these like a sheath. Thus, the flexiblesecondary layer of L. anatina contains several architectural featuresthat aim to divert fractures and simultaneously provides high flex-ibility during the movement of the shell through the sediment.

The shell of Discradisca stella has been described by Williamset al. (1998b), its ultrastructure and some mechanical propertieshave recently been well characterized by Merkel et al. (2007). Ingeneral, the shell of D. stella resembles to some extent that of L.anatina. However, there are two significant distinctions betweenthe shells of these two phosphatic brachiopods: the first is givenby a general absence of the 10–40 lm-sized soft non-mineralizedlayers within the shell of D. stella. Thus, at a first approximationthe shell of D. stella consists only of one major and continuouscompact layer. It does not contain non-mineralized laminae. Thesecond major distinction to the shell ultrastructure of L. anatinais the existence of baculate features (Williams et al., 1998b). Theseare soft non-mineralized layers of around 10 lm thickness that arepresent at distinct positions only within the ventral valve.

The ultrastructure and texture of the shell of Megerlia truncatahas recently been well characterized by Griesshaber et al. (2007)and Schmahl et al. (2008). The outer shell layer of M. truncata isa 50–100 lm thick, nanocrystalline and dense aggregate of calcitecrystallites. It is termed ‘‘primary layer” in the paleontologic

Fig. 2. Nanohardness map across a cross-section of the ventral valve of M. truncata(a) together with the corresponding light microscopy image (b), where the map ofthe nanoindents is well visible. Nanohardness across the cross-section is quiteuniform and scatters between 3 and 4.5 GPa. Fig. A (Appendix) shows the specificanatomic location of the cross-section through the central portion of the ventralvalve that corresponds to the displayed nanoindentation map.

398 C. Merkel et al. / Journal of Structural Biology 168 (2009) 396–408

literature (Williams and Rowell, 1965; Rudwick, 1970; Rowell andGrant, 1987). The thickness of the valves is dominated by thefibrous ‘‘secondary” layer. The fibers are calcite single crystals(Schmahl et al., 2004b; Griesshaber et al., 2007) and they aregrouped into bundles. The morphological fiber axes of these arraysof fibers run in different directions, such that cross-sectionsthrough the valves show longitudinally and transversely cut arrays

Fig. 3. Distribution pattern of hardness (measured as nanohardness) for cross-sections from the dorsal valve (a), the ventral valve (b) and the commissure (ventralvalve) (c). Mean values are given at each data set and highlight clearly thedifferences in nanohardness between the different shell portions. Note the strikingdifference in hardness between the ventral and the dorsal valves.

of fibers (Griesshaber et al., 2007). Generally the fibers are sub-par-allel to the shell vault, but occasionally they can reach angles up to45� with respect to the shell surface. The changing fiber orientationcreates sub-layers of the secondary layer.

Rigid biomaterials, regardless of whether they are calcitic orphosphatic, are hybrid composites and show a complex structuringon several hierarchical levels (e.g., Currey, 1999; Tai et al., 2007;Miserez et al. 2008). Within the hierarchical organization each le-vel contributes to the overall function of the resulting biomaterialand thus influences the mechanical properties of the final product(e.g., Okumura and De Gennes, 2001; Currey, 2005; Rousseau et al.,2005; Miserez et al. 2008). As shown for bivalve (Kamat et al.,2000) and brachiopod shells (Schmahl et al., 2004, 2008; Griess-haber et al., 2006, 2007; Goetz et al., 2009) the microstructure isof profound importance for the shell’s mechanical properties sincethese change as a function of dimension. For example, the smaller acrystallite becomes the less important is the weakening influenceof defects (Ji et al., 2004). In chitinophosphatic brachiopods, thelamination and the degree of mineralization (�10 lm length scale)correlates with the micromechanical properties, while in calciticbrachiopods crystallographic texture plays a major role in the opti-mization of the shell’s materials properties. Thus, since the hierar-chical architecture and small scale mechanical heterogeneity areintrinsic features to all biocomposites (Tai et al., 2007), a completeand thorough understanding of the different parameters that areinvolved in the construction of biomaterials is essential for biomi-metic development of advanced engineering materials (e.g., Mayer,2006). Previous studies regarding the microhardness of calcitic(Griesshaber et al., 2005a, 2007) and chitinophosphatic (Merkelet al., 2007; Schmahl et al., 2008) brachiopods gave a general in-sight into the microstructure–microhardness relationship of theseshells. However, nanohardness testing yields hardness and modu-lus of elasticity variations on a considerably higher spatial resolu-tion. This has so far been only carried out for brachiopods from twodifferent genera: Terebratulina retusa (rhynchonelliform species)and Novocrania anomala (craniid species) (Perez-Huerta et al.,2007). Nanoindentation testing on phosphatic brachiopod shellsas well as nanoindentation analysis on specialized skeletal partsof both calcitic and phosphatic shells, where a high purpose-ori-ented control of the biomaterial is expected, has not been carriedout so far. Thus, the aim of our study is threefold: (1) to obtainan understanding of the hardness and modulus of elasticity distri-bution of the entire shell by conducting nanoindentation maps inboth the specialized parts as well as within the valves of calciticand phosphatic brachiopod shells, (2) to compare nanohardnessand modulus of elasticity results of calcitic and phosphatic bra-chiopod species, and (3) to compare nanohardness results betweenepibenthic and infaunal phosphatic brachiopods. For these pur-poses we have chosen the calcitic brachiopod M. truncata and thetwo phosphatic brachiopods D. stella and L. anatina.

2. Materials and methods

2.1. Specimen preparation

Polished sections of 100 lm thickness have been prepared fromthe calcitic brachiopod shell of M. truncata (habitat: 150 m waterdepth, Mediterranean Sea, France), as well as from the two phos-phatic shells of D. stella (habitat: 10 m water depth, Bali, Indonesia)and L. anatina (habitat: 0.5 m water depth, Japan). The sectionswere cut in two different directions, parallel and perpendicularto the median plane of the shell. A highly smooth surface of theshell wafer was prepared by first polishing the section with dia-mond paste (1 lm particle size) and subsequently attack-polishingthe sample with a suspension of alumina nanoparticles.


2.2. SEM analysis

The microstructure of the samples was investigated with a JEOLHR-SEM equipped with an HKL-EBSD (HKL Technology ‘‘Channel 5”System, Schmidt and Olesen 1989) and an EDX detector. SEMimages were generated using a voltage of 4–16 kV and a beam cur-rent of usually 10 mA.

Fig. 4. Nanohardness versus modulus of elasticity in comparison to inorganic calcite fortooth shell region (b). Corresponding frequency plots of the nanohardness data sets are

2.3. Nanoindentation

After the removal of the carbon coating, the samples wereinvestigated using a SA2 TM (Agilent Technologies, Oak Ridge,TN) nanoindenter. Displacement-controlled nanoindentation wascarried out by loading at a rate of 1 lm/s. Maps with up to 2500indents were obtained across the valves, the commissure and the

the ventral and the dorsal valves (a) as well as for the commissure and the hinge/shown in (c–g).

Table 1Nanohardness and modulus of elasticity results from distinct shell portions of Megerlia truncata, Lingula anatina, and Discradisca stella.

Brachiopodespecies

Hinge Commissure Center (portion ofmaximum curvature)

Dorsalvalve

Ventralvalve

Megerliatruncata

Nanohardness(n) (GPa)

4.14 ± 0.23 3.87 ± 0.34 3.12 ± 0.31 2.85 ± 0.41 4.10 ±.043

Megerliatruncata

E-Modulus(E) (GPa)

70.2 ± 3.3 62.17 ± 5.03 67.57 ± 4.34 63.21 ± 7.15 65.45 ± 4.34

Megerliatruncata

n/E 0.059 0.062 0.046 0.045 0.063

Mineralized layer Organic-rich layer Basal plate

Lingula anatina Nanohardness(n) (GPa)

2.25 ± 0.42 0.40 ± 0.19 1.7 ± 0.74

Lingula anatina E-Modulus(E) (GPa)

43.28 ± 3.28 6.47 ± 2.56 36.95 ± 2.47

Lingula anatina n/E 0.052 0.077 0.046

Mineralized layer withincentral valve portion

Organic-rich layer atthe valve margins

Discradiscastella

Nanohardness(n) (GPa)

1.6 ± 0.34 0.54 ± 0.2

Discradiscastella

E-Modulus(E) (GPa)

31.8 ± 3.34 8.58 ± 3.82

Discradiscastella

n/E 0.005 0.063


hinge (Fig. A in Appendix). Indents with a maximum depth of300 nm were carried out with 30 s between two subsequent inden-tations. The indents were performed with a triangular pyramidalBerkovich indenter with a tip angle of 13� (Barbakadze et al.,2006). Thus a 300 nm deep indent has an edge-length of 2.2 mm.The indenter geometry has the same projected area to depth ratioas the Vickers nanoindenter we have used in our previous studies(Griesshaber et al., 2007; Merkel et al., 2007). The values for thehardness and the modulus of elasticity have been calculatedaccording to Oliver and Pharr (1992), while the continuous stiff-ness measurement (CSM) was performed. An isotropic poisson ra-tio of 0.3 was assumed for the shell material. The modulus ofelasticity E and hardness H were calculated using Eqs. (1) and(2), respectively:

E ¼ 1� v2s

� � 1Er� ð1� v2

i ÞEi

� ��1

ð1Þ

H ¼ Pmax

Að2Þ

Here A is the contact area of the impression, Pmax the maximumload, Ei = 1141 GPa the modulus of elasticity of the Berkovich in-denter, E the modulus of elasticity of the sample, and Er the re-duced modulus of elasticity taken from the measurement. ms andmi (0.35) denote the Poission ratio of the sample and the indenter,respectively. Subsequent to all nanoindentation measurements thesamples were again investigated with SEM and/or polarizing lightmicroscopy in order to determine the exact positions of the inden-tations. To obtain an estimate on the reproducibility of E and H val-ues on brittle materials, clear crystals of inorganic calcite andhydroxyapatite were tested as reference materials with multipleindentations. Histograms indicating the data scatter are shown to-gether with the data.

3. Results

3.1. Calcite-shelled brachiopods

3.1.1. Mergerlia truncataFig. 1 presents a typical microstructural image for M. truncata.

Figs. 2–4 show the overall distribution pattern of nanohardness

from different portions of the shell of M. truncata. The specific shellportions where nanoindentation measurements were carried outare indicated in Fig. A in Appendix. The light microscopy imagein Fig. 2b gives an example of one of the conducted nanoindenta-tion maps. Since the distribution pattern of the moduli of elasticityis graphically similar to the nanohardness distribution it is notshown in a separate profile. Although M. truncata consists of twodistinct valve layers, the primary and the secondary layer, thenanohardness is fairly constant across the cross-sections of thevalves (Figs. 2a and 3a–c). However, there are significant differ-ences in nanohardness for distinct parts of the shell (Table 1). Thisfeature is also well observable in Figs. 4a–g, where we shownanohardness versus modulus of elasticity relationships for dis-tinct shell portions (Fig. 4c–g) together with corresponding fre-quency distribution histograms. The tooth/hinge region is thehardest part of M. truncata and shows 7 % increased mean nanoh-ardness compared to the commissure (Fig. 4b). Most remarkable isthe 36% difference of mean nanohardness between the centralcross-sections of the dorsal and the ventral valves (Fig. 4a). Itshould be noted that for all parts of the shell the nanohardness ishigher than that measured on the (104) cleavage plane of an inor-ganic calcite single crystal (Fig. 4e). Compared to the nanohard-ness, the moduli of elasticity vary slightly less between thedifferent parts of the shell.

3.2. Chitinophosphatic brachiopod shells

3.2.1. Lingula anatinaThe laminated shell structure of L. anatina (Fig. 5) and nanoin-

dentation results for L. anatina are displayed in Figs. 6–9. Fig. A(Appendix) shows the locations in the shell wafer where nanoin-dentation measurements were carried out. The laminated struc-ture of the shell of L. anatina (Fig. 5) with alternating hardmineralized ‘‘compact” layers and soft, non-mineralized layers(the rhythmic units of Williams et al., 1994; Williams and Cusack,1999) is well visible in all images of Figs. 5–7. These laminar unitshave a width in the order of 20–50 lm (Fig. 6a and b) and are sub-divided into lamellea of a width on the submicron-to-micron scale(Fig. 7b and Schmahl et al., 2008). As the lateral size of the indentsis in the order of 2 lm the fine-scale lamination is not alwaysclearly resolvable with nanoindentation. This occasionally pro-duces a variation resembling random data scatter. In the basal

Fig. 6. Nanohardness results across a cross-section of the central valve portion of L. anatinof the figure has been carried out. The laminated nature of the shell consisting of severalwithin the corresponding light microscopy image with the conducted nanoindents (b)hardness variation units in the hardness versus distance diagram.

Fig. 5. The light microscopy image of the figure shows the laminated nature of L.anatina. It is well visible that detachment of a laminar unit (in this case due topreparation of the sample) only occurs within the hard and brittle mineralizedlayers.


plate this fine-scale lamination is most evident (Fig. 6a and b). Themean modulus of elasticity of the hard ‘‘compact” layers is43 ± 6 GPa with a mean nanohardness of 2.25 ± 0.42 GPa, whilethe organic-rich soft layers have a modulus of elasticity of6.5 ± 2.6 GPa and a nanohardness of 0.40 ± 0.2 GPa (Table 1). Thebasal plate has a mean modulus of elasticity of 37 ± 12 GPa andhardness of 1.7 ± 0.74 GPa (Table 1). Note that these mean valuesare calculated by averaging over the fine-scale (1 lm or less) lam-ination. The maximum modulus of elasticity and the maximumhardness of the compact layers increases from the outer borderof the shell (Ha � 1.5 GPa, Ea � 40 GPa) to its central part (Ha � 2 G-Pa, Ea � 55 GPa) and decreases again towards the innermost valvemargin (Figs. 4a and 5b), the valve rim next to the soft tissue of theanimal.

An interesting feature of the shell of L. anatina is the gradual de-crease in nanohardness within each laminar unit (Fig. 6a and 7a, c,and d), a feature that is positively correlated with a gradual in-crease in biopolymer content. The direction of the decrease pointsfrom the valve’s outer to its inner part (Fig. 6a). The lamellar unit is

a. Fig. A (Appendix) shows the specific shell portion where the nanoindentation maplaminar units is obvious in both, the hardness versus distance diagram (a) as well as. Each laminar unit shown in the light microscopy image can easily be related to

Fig. 7. Gradients in nanohardness are present in all laminar units of L. anatina (Fig. 7a and d). These are directly related to gradients in polymer content within such a unit (c).A laminar unit starts with a hard, mineralized layer (see the Vickers microhardness indent value of HV 0.005/10 of 132 in (b). Fig. 7b taken and modified from Merkel et al.2007). This is followed by a succession of organic–inorganic laminae (see the laminae in the light microscopy image of the figure and the internal hardness variations within alaminar unit in a). It ends with a thin soft organic polymer layer (see (a–c) and the Vickers microhardness indent value of HV 0.005/10 of 34 and Merkel et al., 2007).


started by a thin and very hard ‘‘compact” (mineralized) layer(Fig. 7a and b). This layer is followed by a succession of alternatingsoft and hard laminae (Fig. 7a and b). The soft final layer of onelaminar unit (Fig. 7b and c) is followed abruptly by the hard, min-eralized and compact layer of the next lamellar unit (Fig. 6a and7a). Two nanohardness to modulus of elasticity datasets measuredon L. anatina are given in Fig. 8 together with the correspondingfrequency distribution curves (for both nanohardness and modulusof elasticity). For phosphate-shelled brachiopods there is a broaddistribution of modulus of elasticity and nanohardness with astrong linear correlation between the two quantities. The broaddistribution corresponds to the spread between soft organic-rich

and hard mineralized layers and the gradients between thesetwo end-members that compose the shell.

A further feature deduced from nanohardness analysis of L. ana-tina is shown in Fig. 9a–c. A narrow organic-rich band runs alongthe innermost margin of both valves, containing round, hard andprotruding 5–30 lm sized aggregates (Fig. 9a and c). This architec-tural feature can be regarded as a particle-reinforced band. It pro-tects the innermost margin of the valve and extends into theadductor muscles of both valves. Within this band the organic–inorganic composite material is not given by a lamination betweenmineralized and a polymer sheets but by the reinforcement of anorganic band with hard particles.

Fig. 8. Nanohardness versus modulus of elasticity relation together with corresponding frequency distribution diagrams are shown for two distinct parts of the shell of L.anatina: across the entire cross-section (a) and for the innermost valve portion of the shell (b) next to the soft tissue of the animal.


3.3. Discradicsa stella

Nanohardness indentation results of D. stella are shown in Figs.10 and 11. Fig. A (Appendix A) shows the specific shell portionswhere nanoindentation measurements were carried out. Nanoh-ardness and modulus of elasticity values range from H = 0.5 GPaand E = 12 GPa to H = 1.7 GPa and E = 30 GPa (Table 1). The highestvalues are present in the central part of the valve and decrease to-wards the outer and the inner shell margins (Fig. 10a and b). Eventhough D. stella is also a phosphate-shelled brachiopod it shows anentirely different nanohardness distribution pattern in comparisonto L. anatina.

The overall nanohardness distribution pattern starts with a rel-atively soft outer valve margin (denoted with a C in Fig. 8,Ha � 1 GPa, Ea � 25 GPa) that is followed by a hard inner valve por-tion (denoted with a B in Fig. 8, Hi � 1.6 GPa, Ei � 35 GPa) and againturns back to a relatively soft inner shell lamina (denoted with an Ain Fig. 10). At both, the outer and at the inner valve portions thinlayers with relatively high hardness values appear that are sepa-rated from each other by soft organic membranes (Fig. 10a andc). The thickness of the thin layers reaches up to around 4 lm,while the hard layers are expanded up to about 30 lm. The corre-sponding mean hardness values are 0.54 ± 0.2 GPa for the soft and1.6 ± 0.34 GPa for the hard, mineralized layers. The correspondingmoduli of elasticity are 8.58 ± 3.82 and 31.8 ± 3.34 GPa.

Fig. 11 shows nanoindentation versus modulus of elasticity dia-grams with corresponding frequency diagrams for D. stella and L.anatina. Well visible is the trisection (A, B, and C) of the valve ofD. stella, in both the frequency as well as the nanohardness versusmodulus of elasticity diagrams. Furthermore, it is well observablefrom Fig. 9b that the mechanical properties of the shells of D. stellaand L. anatina lie on mixing lines between the two end-members:inorganic hydroxyapatite and the organic biopolymer. It should benoted that the proportion of mixture appears to be different forthese two phosphatic brachiopod species.

Fig. 12 shows fracture surfaces giving an indication of the Ca-phosphate-particle-reinforced biopolymer fiber nanostructure ofthe material composing the shell of D. stella. Fig. 10b shows the‘‘rotated plywood”-type arrangement of the fibrils which form lay-ers of different fiber orientation and the resulting ragged fracturesurfaces with large surface area as described for human lamellarbone by Weiner et al. (1999).

4. Discussion

4.1. Carbonate-shelled brachiopods

Brachiopod shells, like other stiff biomaterials, are hybrid, or-ganic/inorganic, composites with hierarchical microstructures. Avery interesting feature of the phylum brachiopoda is the presenceof two fundamentally different realizations of this concept given bythe utilization of distinct materials for the construction of the shell.In calcite-shelled brachiopods such as M. truncata the fibers areinorganic, consisting of microscale biocalcite single crystals withvery small amounts of nanoscale inter-crystalline and intracrystal-line biopolymer sheets (Schmahl et al., 2004, 2008; Griesshaberet al., 2007). In chitinophosphatic brachiopods such as L. anatinaand D. stella the fibers are essentially organic with a reinforcementof Ca-phosphate nanoparticles in the 5–100 nm size range at-tached to nanoscale fibrils (Fig. 12, see Merkel et al., 2007; Schmahlet al., 2008, for details). The mineral component provides the struc-tural strength and rigidity, which is a prerequisite for fracturetoughness, resistance to abrasion, and protection against mechan-ical failure. Stiff and hard minerals, however, are usually brittle,which is particularly true for calcite but also for apatite. Hence,those minerals produced by organisms need to be suitably nano-and microstructured or modified to avoid brittle fracture and toprovide the functions of a skeleton. This is achieved by the organicmatrix in the composite that provides flexibility and cohesivestrength.

Fig. 9. Distribution pattern of nanohardness along the innermost valve portion. Fig. A (Appendix) shows the specific shell location where the nanoindentation map of thefigure has been carried out. Clearly visible are round and hard grain-like structures (see the light microscopy image of (c) that are embedded into a thin and soft polymer bandthat runs along the innermost margin of the shell and ends in the adductor muscles that move the two valves of the shell (surface plot of (a) and contour map of (b)).


The calcitic terebratulide shell structure is layered, with a nano-crystalline outer, primary layer, and a fiber-composite secondarylayer (some species show a tertiary layer of columnar calcite, Goetzet al., 2009, which is not present here). The secondary layer is com-posed of sub-layers with distinct morphological directions of thefibers. We see no corresponding laminar variation in nanohardnessalong a cross-section through the layers of the shell (the lowerhardness values on left side of Fig. 3c relate to expoxy-filled holesin the surface of the primary layer). In contrast, in our previouswork (Griesshaber et al., 2007; Schmahl et al., 2008) using Vickersmicrohardness indentation (with much higher loads than in thepresent investigation) we observed an overall Vickers hardnessgradient from the hard primary layer to the softer inside of the cal-citic shell. Indents probing transversely cross-sectioned calcite fi-bers showed increased Vickers hardness compared to indents setin sub-layers with longitudinally cut calcite fibers. In our previouspaper (Schmahl et al., 2008), we showed that the hardness valuesdepend on the scale of the indent. For Vickers hardness, the appliedloads of 0.49 N and 0.049 N, respectively, yielded indents withdiagonal sizes on the order of 5 lm or larger, while the load of0.020 N for nanoindentation produces Berkovich edge-lengths inthe order of 2.2 lm on the biocalcite. The single-crystal fibers havedimensions in the order of 5 � 10 � 150 lm3 (Fig. X1, see Griess-

haber et al., 2007, for details). The biopolymer films between thefibers are comparatively thin, usually in the 10–20 nm range (Sch-mahl et al., 2008), reaching 500 nm in rare pathologic circum-stances (Griesshaber et al., in press). Due to their size, usualVickers microindents probe the composite microstructure, andthe measured hardness value depends on the ease of fiber delam-ination or inter-crystalline crack propagation. The nanoindents,however, usually probe hardness and elasticity of the individualcrystallites in the calcitic microstructure as the indents in the reg-ular raster pattern rarely probe such an organic film. Thus, we con-clude that the mechanical parameters of the calcite crystallitesacross a shell cross-section, regardless of the shape and size ofthe crystallites or the microstructure, are very uniform. Perez-Huerta et al. (2007) also studied hardness of craniid brachiopodsusing instrumented indentation, but they applied forces (50 mN)comparable to our Vickers microindentation tests. Their resultsare thus similar to our Vickers results in showing a hardness gradi-ent from the outside (primary layer) to the inner side of the valve.

Consistently with Vickers microhardness results we observe anincreased nanohardness of the biocalcite compared to the hardnessmeasured with indents into the (104) face of inorganic calcite sin-gle crystals. Further, we have evidence that the nanohardness fromthe crystallites of different areas of the shell, i.e. the dorsal and

Fig. 10. Nanohardness distribution pattern for the shell of D. stella. Fig. A(Appendix) shows the specific shell portion where the nanoindentation map ofFig. 8 has been carried out. The shell of D. stella is also laminated but the type oflamination in D. stella is different from that present in L. anatina. We observe threedifferent valve sections with distinct hardness regimes (A, B, and C). The overallhardness distribution pattern of this brachiopod is such that it is hard in the centralvalve portion and soft to the valve’s margins (Fig. 10a and b).


ventral valve, the socket, and tooth region, differ significantly fromeach other. This difference in nanohardness must translate intosome nanostructural differences in the biocalcites composing thosedifferent parts of the shell. In Schmahl et al. (2008), we showedcontrasts in transmission electron micrographs which can beattributed to intracrystalline biopolymers. We expect that theseintracrystalline polymers impede plastic deformation by inhibitingdislocation glide.

The (H/E) ratio (resiliance) describes the relative resistance toplastic and elastic deformation and it is potentially predictive ofquantities such as elastic strain to failure, fracture toughness, andwear resistance (Ashby and Jones, 1998). The H/E ration of thehinge of M. truncata is considerably higher than that of the com-missure (Table 1). The hinge shell portion is composed of socketand tooth where the (H/E) values are 0.059 ± 0.06 and0.047 ± 0.06, respectively. These differences reflect the highermechanical wear stress the socket has to sustain in contrast tothe tooth.

4.2. Phosphate-shelled brachiopods

In contrast to the rather homogeneous nanoscale mechanicalproperties of calcitic brachiopod shells, the nanoscale mechanicalparameters of chitinophosphatic brachiopod shells reflect theirlaminated architecture. The amount of Ca-phosphate particles onthe biopolymer fibers can be varied continuously to adapt materialparameters. The strongly mineralized laminae are hard and stiffand they alternate with less-mineralized or non-mineralized lam-inae, which are soft and compliant. In comparison to a homoge-nously mineralized shell, this laminated architecture provides ahigher flexibility and fracture toughness. Compared to the purebiopolymer component the laminated structure provides enhancedrigidity and structural strength. These properties allow and facili-tate burrowing movements through the sediment which is a prere-quisite for the infaunal life style of L. anatina. The mineralcomponent of the biocomposite is necessary to give the skeletonmaterial sufficient rigidity and elastic stiffness. Strongly mineral-ized laminae, however, are brittle, i.e., susceptible to fracture underload. The progression of fractures along the mineralized layers hasbeen demonstrated by Schmahl et al. (2008). The brittleness of themineral component is a major drawback for the construction of ashell which needs to bend. If a sheet of material bends under load,the bending strain is proportional to the thickness of the sheet (andinversely proportional to the bending radius). By constructing alaminated composite where the brittle laminae are kept thin, thebending strain within those laminae is minimized, and so is thesusceptibility to fracture.

The strategy of nanoparticle-reinforcement further allows tovary the material properties of the biocomposite continuously be-tween two shell end-members, one of which is the pure biopoly-mer and the other the bioapatite. This strategy also allows thecreation of gradient materials by continuously changing the bioap-atite content. Accordingly we find a nearly linear correlation(Fig. 13) of measured hardness and modulus of elasticity values be-tween those two end-members. The gradients potentially avoidstress concentrations arising from differential strain at the inter-face between layers of different moduli of elasticity. Highly differ-ential strains can lead to delaminations between the layers orchipping-off. In contrast to a single-layer laminate the multilayerlaminated structure allows, due to a more homogeneous changein the Young’s modulus from one laminar unit to the next laminarunit a considerable reduction of critical shear forces within theinterface. Thus, the risk of delamination and cracking of laminarunits can be minimized (Perez-Mariano et al., 2006). Most of thehardness (or modulus of elasticity) gradients of the shell materialof L. anatina slope downward towards the inside of the shell (Figs.6 and 7). As the shell is convex, the inner part of a laminate shouldreact to a point load with a larger strain rate than its outer part.Thus, a decrease in modulus of elasticity from outside to inside isrequired and the hardness (as well as the modulus of elasticity)gradients are oriented favorably in order to avoid delamination.

The design principle of organic nanofibrils reinforced with phos-phate nanoparticles is not unlike that in vertebrate bones and teeth.While the biopolymer is mainly the carbohydrate chitin for the bra-chiopod, it is the protein collagen for vertebrates. The hardness andmodulus of elasticity of strongly mineralized laminae in the shell ofLingula anatina reach up to 3 GPa and 55 GPa, respectively. Thesevalues are much higher than the corresponding values for, e.g., hu-man (cortical) bone (0.736 GPa and 25.8 GPa, respectively (Rhoet al., 1997), but distinctively lower than those for human tooth en-amel (3.3–3.9 and 72.7–87.5 GPa (Habelitz et al., 2001).

The shell of the other chitinophosphatic brachiopod, D. stella,with an epibenthic life style, has lower values for the hardnessand the modulus of elasticity than the highly mineralized layersof L. anatina. In D. stella, there is also a strong correlation between

Fig. 11. (a) The nanohardness to modulus of elasticity relationship together with corresponding frequency distribution diagrams for D. stella. In both diagrams wellobservable is the trisection of the shell into the three hardness regimes A, B, and C. (b) A comparison of nanohardness to E-Modulus results for the two phosphate-shelledbrachiopods: L. anatina and D. stella. (b) The frequency distribution diagram for L. anatina.


hardness and modulus of elasticity due to the Ca-phosphate nano-particle-reinforced biopolymer–fiber composite architecture,which allows adaptation of material properties by changing pro-portions of mineral and organic components. The shell shows apattern with higher hardness and stiffness in the central part ofthe cross-section and decreasing values towards the shell margins.A lamination is present, but in more fine-scale and shows a lesspronounced differentiation in material properties than the shellof L. anatina. In the central part (labeled B in Fig. 10) the hardnessand moduli of elasticity are fairly constant, while there are gradi-ents sloping to softer mechanical parameters towards the shellmargins (labeled A and C in Fig. 10). At the borders of the centralpart B and within the gradients A and C there are many thin hardlaminae. Similar to human lamellar bone (Fig. 1 of Weiner et al.,1999), the fibrils in adjacent laminae of D. stella form a ‘‘rotatedplywood” type of arrangement on the 2–10 lm scale (Fig. 12band Schmahl et al., 2008). While it is most likely that the thin hardlaminae observed in the nanoindentation scans are related to ahigher degree of mineralization, the different orientation of fibersbetween those lamellae might also give rise to hardness and stiff-ness variations.

The fiber-composite micro- and nanostructure offers an advan-tageous mechanism for ductile energy dissipation. During crackformation the deformation around the crack-tip is not evenly dis-tributed and has to be partitioned into stretching of fibrils andinterfibrillar shearing. The cross-lamination of fibril orientation(‘‘rotated plywood”, Fig. 12b and Schmahl et al., 2008) and the suc-cession of stiff and compliant laminae also contributes to fracturetoughness and results in a predetermined fracture parallel to theshell at high energy costs. The fracture surfaces are not even butragged and tortuous, they are hierarchically structured, have thesame appearance at every length scale, and are parallel to theshell’s curvature (Merkel et al., 2007; Schmahl et al., 2008). From

Fig. 12b it can be inferred that the fibril direction is rotated by aa few tens of degrees between laminae of a width of about 5 lmwhich are separated by thinner (�1 lm) laminae. Within the mi-cro-laminae there are nano-laminae arranged roughly perpendicu-lar to the micro-lamination. These changes in the fiber orientationlead to a very efficient protection against cracking, because a crackcannot penetrate steadily along the weak direction parallel to a setof fibers but it is diverted when it reaches fibers of different direc-tion in the laminate.

Further, the ends of most of the ruptured fibrils at fracture sur-faces (Fig. 12a) are directed towards the surface instead of beingstatistically oriented. This selective orientation must be due to anattractive force pulling the fibril ends towards the surface. Such aforce can be the result of an adhesive between the biopolymerfibrils. Fantner et al. (2005) described such a so-called self-healingeffect, where the adhesive substance unfolds during loading andrefolds to close the crack after it has been stopped and the loadhas been removed.

5. Concluding summary

1. The shells of both calcitic and chitinophosphatic brachiopodsare examples of distinct shell biomaterials of marine organismswith advanced mechanical properties. They display an adaptedmaterials design with a hierarchical composite architecture thatovercomes the brittleness of the mineral component and theinferior strength of the organic component.

2. The calcitic brachiopod shells are dominantly inorganic likeother carbonate shells such as those of molluscs. Accordingly,the nanomechanical parameters measured here are close to thatof the mineral component. The hardness of the biocalcite isalways increased compared to inorganic calcite while the mod-ulus of elasticity of the biocalcite is similar to that in inorganic

Fig. 12. (a) An SEM image of ruptured fibrils (indicated with arrows) on the fracturesurface of D. stella. The ends of most of the ruptured fibrils are directed towards thefracture surface. (b) The changing fibril orientation between adjacent laminae of theshell resulting in a ‘‘rotated plywood” type cross-lamination.

Fig. 13. Comparison of nanohardness to modulus of elasticity for phosphate- andcalcite-shelled modern brachiopods (L. anatina, D. stella, and M. truncata) in relationto the mineral references: calcite (sampled from Rodeo, Durango, Mexico) andhydroxyapatite (sampled from Snarum, Norway). Phosphate-shelled brachiopodsshow large variations in nanohardness and modulus of elasticity and plot on amixing line between the end-members hydroxyapatite and a pure organic polymer(not specifically indicated mean nanohardness (polymer): 0.7 GPa; mean modulusof elasticity (polymer): 3.5 GPa), while calcite-shelled brachiopods display a largerange in nanohardness with a small scatter in modulus of elasticity. Thus, forcalcite-shelled brachiopod hardness variations are not simply given by two end-members.

Fig. A. Generic sketch of M. truncata, L. anatina and D. stella highlighting those shellportions where nanoindentation maps were measured.


calcite. For the calcitic brachiopod shells there is a differentia-tion of nanohardness between different parts of the shells.The modulus of elasticity is fairly homogeneous throughoutthe entire shell. This is related to variations in the dispositionand/or constitution of organic inter- and intracrystalline matrixmaterial in the calcite.

3. Like in vertebrate bones or teeth the material of chitinophos-phatic brachiopod shells is, on the molecular- to nanoscale, ahybrid composite of biopolymer fibrils reinforced with Ca-phosphate nanoparticles. This structure allows a continuousadaptation of modulus of elasticity and hardness within a largerange by two-phase mixing. The mineral provides stiffness andhardness, the biopolymer provides flexibility; the hierarchicalcomposite structure on all length scales contributes to fracturetoughness.

4. The two epibenthic forms, calcitic M. truncata and chitinophos-phatic D. stella display fairly bulky (unstructured) nanomechan-ical properties of their shell materials, while the burrowinginfaunal L. anatina is distinctively laminated. In addition to arandom nano- or microscale distribution of two mechanicallydistinct phases, the two-phase composite design allows to pro-duce a laminated gradient material where stiff and compliantlaminae units alternate. This allows to tune the properties ofstrength versus flexibility even further. The strongly mineral-ized sheets in the laminate provide mechanical strength while

flexibility of the shell is given by keeping the individual sheetsthin enough to avoid bending strains exceeding their fracturestrength. This flexibility appears to be not required for the lifestyle of the epibenthic brachiopods.

5. Organisms mineralizing calcium carbonate skeletons, includingthe sophisticated mollusc nacre, use this adaptable two-phasemixing concept to a much lesser degree than chitinophosphaticbrachiopods or vertebrates with their collagen nanofibril/Ca-phosphate nanoparticle hybrid composites.


Acknowledgments

We thank Alan Logan for providing samples. We also thankMadlen Fischer and Renate Enders for their kind help with thesample preparation. This work is supported by the GermanResearch Council (Deutsche Forschungs Gemeinschaft, DFG) byGrant No. SCHM 930/8-1.

Appendix A

See Fig. A.

References

Ashby, M.F., Jones, D.R.H., 1998. Engineering Materials 2. Butterworth Heineman,Oxford. 455 pp.

Brand, U., Logan, A., Hiller, N., Richardson, J., 2003. Geochemistry of modernbrachiopods: applications and implications for oceanography andpaleoceanography. Chem. Geol. 198, 305–334.

Barbakadze, N., Enders, S., Gorb, S., Arzt, E., 2006. Local mechanical properties of thehead articulation cuticle in the beetle Pachnoda marginata (Coleoptera,Scarabaeidae). J. Exp. Biol. 209, 722–730.

Bruckschen, P., Oesmann, S., Veizer, J., 1999. Isotope stratigraphy of the EuropeanCarboniferous: proxy signals for ocean chemistry, climate and tectonics. Chem.Geol. 161, 127–163.

Currey, J.D., 1999. The design of mineralised hard tissues for their mechanicalfunctions. J. Exp. Biol. 202, 3285–3294.

Currey, J., 2005. Hierarchies in biomineral structures. Science 309, 253–254.Cusack, M., Williams, A., Buckman, J., 1999. Chemico-structural evolution of

linguloid brachiopod shells. Palaeontology 42, 799–840.Fantner, G.E., Hassenkam, T., Kindt, J.H., Weaver, J.C., Birkedal, H., Pechenik, L.,

Cutroni, J.A., Cidade, G.A.G., Stucky, G.D., Morse, D.E., Hansma, P.K., 2005.Sacrificial bonds and hidden length dissipate energy as mineralized fibrilsseparate during bone fracture. Nat. Mater. 4, 612–616.

Goetz, A.J., Griesshaber, E., Neuser, R.D., Lüter, C., Hühner, M., Harper, E., Schmahl,W.W., 2009. Calcite morphology, texture and hardness in the distinct layers ofrhynchonelliform brachiopod shells. Eur. J. Mineral. 21, 303–315.

Griesshaber, E., Schmahl, W.W., Neuser, R., Job, R., Bluem, M., Brand, U., 2005a.Microstructure of brachiopod shells – an inorganic/organic fiber composite withnanocrystalline protective layer. Mater. Res. Soc. Symp. Proc. 851, 99–104.

Griesshaber, E., Job, R., Pettke, T., Schmahl, W.W., 2005b. Micro-scale physical andchemical heterogeneities in biogenic materials: a combined micro-raman,chemical composition and microhardness investigation. In: Katti, K., Ulm, F.J.,Hellmich, C., Viney, C. (Eds.), Mechanical Properties of Bio-inspired andBiological Materials, MRS Symp. Proc. Series, vol. 844. Materials ResearchSociety, Warrendale, PA, pp. 93–98.

Griesshaber, E., Kelm, K., Sehrbrock, A., Job, R., Schmahl, W.W., Mader, W., 2006. Theultrastructure of brachiopod shells: a mechanically optimized material withhierarchical architecture. In: Bushby, A.J., Ferguson, V.L., Ko, C.C., Oye, M.L.(Eds.), Mechanical Behavior of Biological and Biomimetic Materials, MRS Symp.Proc. Series, vol. 898E. Materials Research Society, Warrendale, PA.

Griesshaber, E., Schmahl, W.W., Neuser, R., Pettke, T., Blüm, M., Brand, U., 2007.Texture and microstructure of terebratulide brachiopod shell calcite – anoptimized materials design with hierarchical architecture. Am. Mineral. 92,722–735.

Griesshaber, E., Kelm, K., Sehrbrock, S., Schmahl, W., Mader, W., Mutterlose, J.,Brand, U., in press. Amorphous components in the shell material of thebrachiopod Megerlia truncata. Eur. J. Mineral.

Habelitz, S., Marshall, S.J., Marshall, G.W., Balooch, M., 2001. Mechanical propertiesof human dental enamel on the nanometre scale. Arch. Oral Biol. 46, 173–183.

Iwata, K., 1981. Ultrastructure and Mineralization of the Shell of Lingula unguis(Linne). J. Fac. Sci. Hokkaido Univ. Ser. IV, vol. 20, No. 1.

Ji, B., Gao, H., Hsia, K.J., 2004. How do slender mineral crystals resist buckling inbiological materials. Phil. Mag. Lett. 84, 631–641.

Kamat, S., Su, X., Ballarini, R., Heuer, A.H., 2000. Structural basis for the fracturetoughness of the shell of the conch Strombus gigas. Nature 405, 1036–1040.

Lévêque, I., Cusack, M., Davis, S.A., Mann, S., 2004. Promotion of fluoroapatitecrystallization by soluble-matrix proteins from Lingula anatina shells. Angew.Chem. Int. Ed. 43, 885.

Mayer, G., 2006. New classes of tough composite materials – lessons from naturalrigid biological systems. Mater. Sci. Eng. C 26, 261–1268.

Merkel, C., Griesshaber, E., Kelm, K., Neuser, R., Jordan, G., Logan, A., Mader, W.,Schmahl, W.W., 2007. Micromechanical properties and structuralcharacterization of modern inarticulated brachiopod shells. J. Geophys. Res.112, G02008. doi:10.1029/2006JG000253.

Miserez, A., James, C., Weaver, P., Aizenberg, J., Dauphin, Y., Fratzl, P., Morse, D., Zok,F., 2008. Effects of laminate architecture on fracture resistance of spongebiosilica: lessons from nature. Adv. Funct. Mater. 18, 1241–1248.

Okumura, K., De Gennes, P.G., 2001. Why is nacre strong? Elastic theory andfracture mechanics for biocomposites with stratified structures. Eur. Phys. J. E4(1), 121–127.

Oliver, W.C., Pharr, G.M., 1992. An improved technique for determining hardnessand elastic modulus using load and displacement sensing indentationexperiments. J. Mater. Res. 7, 1564–1583.

Parkinson, D., Curry, G.B., Cusack, M., Fallik, A.E., 2005. Shell structure, patterns andtrends of oxygen and carbon stable isotopes in modern brachiopod shells.Chem. Geol. 219, 193–235.

Perez-Huerta, A., Cusack, M., Zhu, W.Z., England, J., Hughes, J., 2007. Materialproperties of brachiopod shell ultrastructure by nanoindentation. J. R. Soc.Interface 4 (12), 33V–39V.

Perez-Mariano, J., Lau, K.-H., Sanjurjo, A., Caro, J., Prado, J. M., Colominas, C., 2006.Multilayer coatings by chemical vapour deposition in a fluidized bed reactor atatmospheric pressure (AP/FBR-CVD): TiN/TaN and TiN/W. Surf. Coat. Tech. 201,2174, 2180.

Rho, J.Y., Tsui, T.Y., Pharr, G.M., 1997. Elastic properties of human cortical andtrabecular lamellar bone measured by nanoindentation. Biomaterials 18, 1325–1330.

Rousseau, M., Lopez, E., Stempfle, P., Brendle, M., Franke, L., Guette, A., Naslain, R.,Bourrat, X., 2005. Multiscale structure of sheet nacre. Biomaterials 26, 6254–6262.

Rudwick, M.J.S., 1959. The growth and form of brachiopod shells. Geol. Mag. 96, 1–24.

Rudwick, M.J.S., 1970. Living and Fossil Brachiopods. Hutchinson and Co. LTD,London, 199 pp.

Rowell, A.J., Grant, R.E., 1987. Phylum brachiopoda. In: Bradman, R.S., Cheetham,A.H., Rowell, A.J. (Eds.), Fossil Invertebrates. Blackwell Scientific Publications,Oxford, pp. 445–497.

Schmahl, W.W., Griesshaber, E., Neuser, R., Lenze, A., Brand U., 2004a. Morphologyand texture of the fibrous calcite in terebratulide brachiopode shells. Geochim.Cosmochim. Acta 68 (11), A202–A202, Suppl. S.

Schmahl, W.W., Griesshaber, E., Neuser, R.D., Lenze, A., Job, R., Brand, U., 2004b. Themicrostructure of the fibrous layer of terebratulide brachiopod shell calcite. Eur.J. Miner. 16, 693–697.

Schmahl, W.W., Griesshaber, E., Neuser, R., Pettke, T., Kelm, K., 2006a. Texture andmicrostructure of terebratulid brachiopod shell calcite – a mechanicallyoptimised material with hierarchical architecture. In: Micro- to Nano-Geosciences: Developments and Application, UK Mineralogical Society WinterMeeting, Programme and Abstract Volume, p. 64.

Schmahl, W.W., Griesshaber, E., Merkel, C., Kelm, K., Deuschle, J., Neuser, R.D., Goetz,A.J., Sehrbrock, A., Mader, W., 2008. Hierarchical fibre composite structure andmicromechanical properties of phosphatic and calcitic brachiopod shellmaterials – an overview. Miner. Mag. 72, 541–562.

Schmidt, N.H., Olesen, N.O., 1989. Computer-aided determination of crystal-latticeorientation from electron channeling patterns in the SEM. Canad. Miner. 27 (1),15–22.

Tai, K., Dao, M., Suresh, S., Palazoglu, A., Ortiz C., 2007. Nanoscale heterogeneitypromotes energy dissipation in bone. Nat. Mat. 6. doi:10.1038/nmat1911.

Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Carden, G.A.F., Diener,A., Ebneth, S., Godderis, Y., Jasper, T., Korte, G., Pawellek, F., Podlaha, O.G.,Strauss, H., 1999. Sr-87/Sr-86, delta C-13 and delta O-18 evolution ofPhanerozoic seawater. Chem. Geol. 161, 59–88.

Weiner, S., Traub, W., Wagner, H.D., 1999. Lamellar bone: structure–functionrelations. J. Struct. Biol. 126, 241–255.

Williams, A., Cusack, M., Mackay, S., 1994. Collagenous chitino-phosphatic shell ofthe brachiopod Lingula. Phil. Trans. R. Soc. Lond. B 346, 223–266.

Williams, A., Brunton, C.H., Carlson, S.J., Alvarez, F., Ansell, A.D., Baker, P.G., Bassett,M.G., Blodgett, R.B., Boucot, A.J., Carter, J.L., Cocks, L.R., Cohen, B.L., Copper, P.,Curry, G.B., Cusack, M., Dagys, A.S., Emig, C.C., Gawthrop, A.B., Gourvennec, R.,Grant, R.E., Harper, D.A., Holmer, L.E., Hong-Fei, H., James, M.A., Yu-Gan, J.,Johnson, J.G., Laurie, J.R., Lazarev, S., Lee, D.E., MacKay, S., MacKinnon, D.I.,Mancenido, M.O., Mergl, M., Owen, E.F., Peck, L.S., Popov, L.E., Racheboeuf, P.R.,Rhodes, M.C., Richardson, J.R., Jia-Yu, R., Rubel, M., Savage, N.M., Smirnova, T.N.,Dong-Lee, S., Walton, D., Wardlaw, B., Wright, A.D., 1997. Treatise onInvertebrate Paleontology, Part H, Geological Society of America andUniversity of Kansas Press, Lawrence, KS, pp. 265.

Williams, A., Cusack, M., Buckman, J., Stachel, T., 1998a. Siliceos tablets in the larvalshells of apatitic discinid brachiopods. Science 279, 2094–2096.

Williams, A., Cusack, M., Buckmann, J., 1998b. Chemico-structural phylogeny of thediscinoid brachiopod shell. Phil. Trans. R. Soc. Lond. B 353, 2005–2038.

Williams, A., Cusack, M., 1999. Evolution of a rhythmic lamination in theorganophosphatic shells of brachiopods. J. Struct. Biol. 126, 227–240.

Williams A., Rowell A.J., 1965. Brachiopod anatomy, pp. H6–H57. In: Moore, R.C.(Ed), Treatise on Invertebrate Paleontology, Part H, Geological Society ofAmerica and University of Kansas Press, Lawrence, KS, pp. 257.

http://dx.doi.org/10.1029/2006JG000253

http://dx.doi.org/10.1038/nmat1911

mechanical properties of modern calcite- (mergerlia truncata) and phosphate-shelled brachiopods...

Documents