ArticlePDF Available

Unifying Design Strategies in Demosponge and Hexactinellid Skeletal Systems

Authors:

Abstract and Figures

Biological systems are well known for their ability to construct remarkably complex and mechanically robust skeletal structures from a great diversity of minerals. One such example, silica, is widely used in the synthesis of skeletal elements (spicules) within the phylum Porifera (the sponges). As a result, members of this diverse group have served as useful model systems for analysis of the dynamic processes of biosilicification and for investigating structure function relationships in their often hierarchically ordered skeletal systems. This article describes in detail the skeletal diversity within the two silica-forming sponge classes, the Demospongiae and the Hexactinellida, and through the use of several representative examples, discusses the mechanical consequences of the various modes of construction implemented as well as the potential evolutionary pressures that resulted in their observed structural complexity.
Content may be subject to copyright.
This article was downloaded by: [University of California, Riverside Libraries]
On: 15 August 2012, At: 17:27
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK
The Journal of Adhesion
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/gadh20
Unifying Design Strategies in
Demosponge and Hexactinellid
Skeletal Systems
James C. Weaver a , Garrett W. Milliron a , Peter
Allen b , Ali Miserez c , Aditya Rawal d , Javier Garay
e , Philipp J. Thurner f , Jong Seto g , Boaz Mayzel h
, Larry Jon Friesen i , Bradley F. Chmelka d , Peter
Fratzl g , Joanna Aizenberg j , Yannicke Dauphin k ,
David Kisailus a & Daniel E. Morse l
a Department of Chemical and Environmental
Engineering, University of California, Riverside,
Riverside, CA, USA
b Engineering Department, University of California,
Santa Barbara, Santa Barbara, CA, USA
c Materials Department and Department of
Molecular, Cellular, and Developmental Biology,
University of California, Santa Barbara, Santa
Barbara, CA, USA
d Department of Chemical Engineering, University of
California, Santa Barbara, Santa Barbara, CA, USA
e Department of Mechanical Engineering, Materials
Science and Engineering Program, University of
California, Riverside, Riverside, CA, USA
f Bioengineering Science Research Group, School
of Engineering Science, University of Southampton
Highfield, Southampton, U.K.
g Department of Biomaterials, Max Planck Institute
of Colloids and Interfaces, Potsdam, Germany
h Department of Zoology, Faculty of Life Sciences,
Tel Aviv University, Tel Aviv, Israel
i Department of Biological Sciences, Santa Barbara
City College, Santa Barbara, CA, USA
j School of Engineering and Applied Sciences,
Department of Chemistry and Chemical Biology,
Harvard University, Cambridge, MA, USA
k UMR IDES, University of Paris-Sud, Orsay, France
l Department of Molecular, Cellular and
Developmental Biology, Institute for Collaborative
Biotechnologies, Materials Research Laboratory,
and California Nanosystems Institute, University of
California, Santa Barbara, Santa Barbara, CA, USA
Version of record first published: 05 Feb 2010
To cite this article: James C. Weaver, Garrett W. Milliron, Peter Allen, Ali Miserez,
Aditya Rawal, Javier Garay, Philipp J. Thurner, Jong Seto, Boaz Mayzel, Larry Jon
Friesen, Bradley F. Chmelka, Peter Fratzl, Joanna Aizenberg, Yannicke Dauphin, David
Kisailus & Daniel E. Morse (2010): Unifying Design Strategies in Demosponge and
Hexactinellid Skeletal Systems, The Journal of Adhesion, 86:1, 72-95
To link to this article: http://dx.doi.org/10.1080/00218460903417917
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-
and-conditions
This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden.
The publisher does not give any warranty express or implied or make any
representation that the contents will be complete or accurate or up to
date. The accuracy of any instructions, formulae, and drug doses should be
independently verified with primary sources. The publisher shall not be liable
for any loss, actions, claims, proceedings, demand, or costs or damages
whatsoever or howsoever caused arising directly or indirectly in connection
with or arising out of the use of this material.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Unifying Design Strategies in Demosponge and
Hexactinellid Skeletal Systems
James C. Weaver
1
, Garrett W. Milliron
1
, Peter Allen
2
,
Ali Miserez
3
, Aditya Rawal
4
, Javier Garay
5
,
Philipp J. Thurner
6
, Jong Seto
7
, Boaz Mayzel
8
,
Larry Jon Friesen
9
, Bradley F. Chmelka
4
, Peter Fratzl
7
,
Joanna Aizenberg
10
, Yannicke Dauphin
11
,
David Kisailus
1
, and Daniel E. Morse
12
1
Department of Chemical and Environmental Engineering,
University of California, Riverside, Riverside, CA, USA
2
Engineering Department, University of California, Santa Barbara,
Santa Barbara, CA, USA
3
Materials Department and Department of Molecular, Cellular,
and Developmental Biology, University of California, Santa Barbara,
Santa Barbara, CA, USA
4
Department of Chemical Engineering, University of California,
Santa Barbara, Santa Barbara, CA, USA
5
Department of Mechanical Engineering, Materials Science and
Engineering Program, University of California, Riverside, Riverside,
CA, USA
6
Bioengineering Science Research Group, School of Engineering Science,
University of Southampton Highfield, Southampton, U.K.
7
Department of Biomaterials, Max Planck Institute of Colloids and
Interfaces, Potsdam, Germany
8
Department of Zoology, Faculty of Life Sciences, Tel Aviv University,
Tel Aviv, Israel
9
Department of Biological Sciences, Santa Barbara City College,
Santa Barbara, CA, USA
Received 6 March 2009; in final form 20 August 2009.
This article is dedicated to Professor J. Herbert Waite who continues to be an inspira-
tion for investigating structure-function relationships in biological systems.
One of a Collection of papers honoring J. Herbert Waite, the recipient in February 2009
of The Adhesion Society Award for Excellence in Adhesion Science, Sponsored by 3M.
Address correspondence to David Kisailus, Department of Chemical and Environ-
mental Engineering, University of California, Riverside, CA 92521, USA. E-mail:
david@engr.ucr.edu or Daniel E. Morse, Department of Molecular, Cellular and
Developmental Biology, University of California, Santa Barbara, CA 93106, USA.
E-mail: d_morse@lifesci.ucsb.edu
The Journal of Adhesion, 86:72–95, 2010
Copyright #Taylor & Francis Group, LLC
ISSN: 0021-8464 print=1545-5823 online
DOI: 10.1080/00218460903417917
72
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
10
School of Engineering and Applied Sciences, Department of Chemistry
and Chemical Biology, Harvard University, Cambridge, MA, USA
11
UMR IDES, University of Paris-Sud, Orsay, France
12
Department of Molecular, Cellular and Developmental Biology,
Institute for Collaborative Biotechnologies, Materials Research
Laboratory, and California Nanosystems Institute, University of
California, Santa Barbara, Santa Barbara, CA, USA
Biological systems are well known for their ability to construct remarkably
complex and mechanically robust skeletal structures from a great diversity of
minerals. One such example, silica, is widely used in the synthesis of skeletal
elements (spicules) within the phylum Porifera (the sponges). As a result, members
of this diverse group have served as useful model systems for analysis of the
dynamic processes of biosilicification and for investigating structure function
relationships in their often hierarchically ordered skeletal systems. This article
describes in detail the skeletal diversity within the two silica-forming sponge
classes, the Demospongiae and the Hexactinellida, and through the use of several
representative examples, discusses the mechanical consequences of the various
modes of construction implemented as well as the potential evolutionary pressures
that resulted in their observed structural complexity.
Keywords: Biomineralization; Biosilica; Composite; Fracture mechanics; Porifera
INTRODUCTION
The diversity of silica skeletal systems in the phylum Porifera (the
sponges) is truly remarkable [1–5]. Despite the nearly 200 years of
scientific data available on the subject, however, there is no unified
theory that can explain the observed complexity at the ultrastructural
level of the individual spicules or the mechanisms and design princi-
ples by which they assemble to produce hierarchically organized
skeletal networks.
The two main silicifying sponge classes, the Demospongiae and the
Hexactinellida, differ significantly from one another with respect to
spicule symmetry, structural diversity, and basic histology. While
demosponges consist predominantly of loose aggregations of individual
cells with specialized functions, the hexactinellids are principally
composed of massive multinucleate syncytia
1
[5–7]. These differences
1
A syncytium is a large cell-like mass of cytoplasm containing multiple nuclei and
enclosed in a membrane with no internal cell boundaries. Syncytia frequently result
from the fusion of two or more cells.
Design Strategies in Skeletal Systems 73
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
in histology have a dramatic effect on the dimensional limits of the
spicules synthesized by members of these two sponge classes.
In demosponges, the maximum spicule size encountered is
typically a few millimeters in total length [1,2]. This limitation
is mainly related to the fundamental mechanisms, by which these
spicules are formed. In these sponges, recent evidence suggests that
the central axial filament that provides a substrate for early silica
deposition is synthesized in its entirety prior to silicification [8]. As
a result, the maximum dimensions of a single spicule are fundamen-
tally limited by the extensibility of the individual cells involved in
axial filament synthesis.
High-resolution structural analyses of demosponge spicules have
revealed that the central axial filament is hexagonal, or some deriva-
tive thereof, in cross-section and is surrounded by concentric lamellae
of consolidated silica nanoparticles [9,10]. While the silica surrounding
the axial filament is deposited in layers exhibiting variability in the
degree of silica condensation, there are no detectable quantities of
organic matter present [9,10].
These general design strategies of demosponge spicules differ
significantly from those seen in the hexactinellids. Whereas hexa-
ctinellid spicules also possess a proteinaceous axial filament, it is dis-
tinctly square in cross-section [11], in contrast to the hexagonal form
found in demosponges [12–14]. The cross-sectional morphology of
these axial filaments plays a direct role in establishing the unique
three-axis (or six-rayed) geometry characteristic of the hexactinellid
spicules. Lateral filament outgrowth from the four faces of the main
central axis results in the formation of three distinct axes (two in
addition to the central axis) that intersect one another at right
angles. In addition to the unique axial filament morphology found
in the hexactinellids, the mechanism of axial filament synthesis also
appears to be distinct. Unlike demosponge axial filaments, which are
synthesized in their entirety prior to silica deposition, those in the
hexactinellids appear to grow continuously during spicule biosynth-
esis [15]. This central organic scaffold is reportedly connected to the
surrounding syncytium via an opening in the end of each spicule
ray, and once spicule lateral growth has ceased, the openings are
sealed by the deposition of additional silica [4]. In addition to the
growth potential of the axial filament, because of the syncytial nature
of the sclerocytes, there is practically no limitation to the dimensions
of the synthesized spicules.
The following discussion of sponge spicule architectural diversity
is based on the observations obtained from a wide range of different
demosponge and hexactinellid species. Because of the ultrastructural
74 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
simplicity of demosponge spicules outlined above (consisting of fused
concentric lamellae of consolidated silica nanoparticles), they will be
used as a comparative metric for the following analysis of hexactinellid
skeletal diversity. Based on these observations, we attempt to explain
the basic design features from a mechanics perspective in hopes of
gaining additional insight into the potential evolutionary pressures
that resulted in the observed structural diversity.
MATERIALS AND METHODS
Research Specimens
Aphrocallistes vastus and Rhabdocalyptus dawsoni
Both species were collected by SCUBA at Foley Head, Hotham
Sound=Jervis Inlet, British Columbia at ca. 25 m depth.
Euplectella aspergillum
Specimens of Euplectella aspergillum (of Philippine origin) were
received as dried skeletons.
Monorhaphis chuni
Specimens used in this study were collected from two locations:
1. New Caledonia, near Lifou (Loyalty Island) at a depth of 1905 m
during the CALSUB campaign of the IRD 1989.
2. Norfolk Ridge at a depth of 1200 m during the HALIPRO campaign
in 1996.
Sample Preparation
A. vastus and R. dawsoni
The A. vastus sample, depicted in Figs. 3A and B, was soaked in
freshwater to remove the residual salts, flash frozen in liquid nitrogen,
and lyophilized. All other skeletal material from either A. vastus or
R. dawsoni was soaked in a 5.25%solution of sodium hypochlorite
until all of the organic material had been removed. The samples were
then washed five times with Milli-Q (Millipore-purified) water, and
air-dried at room temperature.
M. chuni
All spicule specimens from M. chuni were surface cleaned with
Milli-Q water and air-dried at room temperature.
Design Strategies in Skeletal Systems 75
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Scanning Electron and Optical Microscopy
Excised portions of the mineralized skeleton of each sponge species
were examined by optical microscopy and then mounted on aluminum
disks using either conductive carbon tabs, silver paint, or conductive
epoxy, depending on the preferred orientation of the sample being
examined. Following mounting, all samples were sputter-coated with
gold and examined with a Tescan Vega TS 5130MM (Brno, Czech
Republic) scanning electron microscope.
Mechanical Testing
Nanoindentation
Spicule segments were embedded in M-Bond AE-15 (M-Line,
Raleigh USA) epoxy resin, sliced into ca. 3-mm thick transverse sec-
tions using a diamond cutting wheel, and polished with diamond lap-
ping films with particles sizes down to 0.1 mm under a constant flow of
fresh water.
For mechanical testing, nanoindentation was performed using a
Triboscan nanoindenter system (Hysitron, Minneapolis, MN, USA)
and cube-corner tips. Indentations were applied with a high-load
transducer permitting applied loads up to 1 N. High loads were
mainly selected to ensure damage in the thin-layer region of the
spicules. For all indents, the peak load was held constant for 10 s
before unloading. Control indentations were made on a fused
quartz control slide under similar experimental conditions. After
indentation, the samples were sputter-coated with gold and
examined with a Tescan Vega TS 5130MM (Brno, Czech Republic)
scanning electron microscope (both before and after sonication).
The images presented in Fig. 9 are representative of the results
obtained.
3-Point Bending Tests
Each of the synthetic glass rods and the giant anchor spicules from
M. chuni measured ca. 5 mm in diameter. The test samples were cut to
length with a diamond cutting wheel, surface cleaned with Milli-Q,
water, and then air dried at room temperature.
Three-point bending tests were conducted with a 5543 Instron
(Norwood, MA, USA) tensile and compression testing machine using
a displacement control rate of 2.5 mm=min. A standard three-point
bend fixture (Instron) was used and the span was 4 cm.
76 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Solid-State
29
Si NMR
Solid-state
29
Si NMR experiments were conducted at UCSB on a
Bruker (Billerica, MA, USA) DMX-300 spectrometer with a wide-bore
7.0 Tesla superconducting magnet operating at a
29
Si Larmor
frequency of 59.6 MHz. The experiments were carried out at room
temperature under magic-angle-spinning (MAS) conditions (8 kHz)
using a 4-mm double-resonance MAS probehead, on powders prepared
from the spicule samples. Single-pulse
29
Si NMR signals were
acquired using 1024 transients with a 60 s recycle delay between the
transients.
1
H heteronuclear decoupling was applied during acquisi-
tion using a SPINAL64 decoupling scheme and a
1
H radio-frequency
field strength of 60 kHz.
29
Si signals are referenced to TMS.
RESULTS AND DISCUSSION
Aphrocallistes vastus
Despite their vast structural diversity (Fig. 1), the smallest of the
hexactinellid spicules exhibit a basic design remarkably similar to
those encountered in demosponges (such as Tethya aurantia;cf. [10]),
FIGURE 1 (A) Structural diversity and (B,C) regiospecificity of hexactinellid
spicules (adapted from [5,34]).
Design Strategies in Skeletal Systems 77
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
consisting of a central axial filament surrounded by concentric
lamellae of consolidated silica nanoparticles. When these spicules
are fractured, like those from demosponges, they fail catastrophically
without any indication of the presence of structural boundaries
(Fig. 2A) via the uninterrupted propagation of a single dominant
crack.
In the skeletal system of Aphrocallistes vastus (Fig. 3A), for exam-
ple, the outer surface is almost completely covered with numerous
such spicules that are held in place by the syncytium and are posi-
tioned perpendicular to the surface (Figs. 3B–D) [5]. While these
spicules do exhibit the unique three-axis symmetry characteristic of
hexactinellids, the rays are unequally developed, resulting in the
formation of the characteristic pinnule (sword shaped) spicules shown
in Fig. 3D. Beneath this outer layer of protective spicules is the main
skeletal lattice [4,5]. This skeletal lattice consists of an intricate net-
work of fused six-rayed spicules (hexactines), each of which measures
ca. 200 mm in length (Fig. 3G).
These spicules are fused together at the end of each ray in such a
manner that results in the formation of a rigid honeycomb-like archi-
tecture (Figs. 3E, F) and can exhibit a remarkable degree of ordering
when viewed in cross-section, with many of the constituent spicules
fused at right angles with respect to one another (Figs. 4A, B).
The organic scaffold (consisting of the six-rayed axial filaments)
is not continuous throughout the lattice, although it may overlap at
the points of fusion between two neighboring spicules (Figs. 4C–E).
FIGURE 2 Failure modes of (A) nonlaminated and (B) laminated spicules. (A)
is from the demosponge Tethya aurantia and (B) is from the hexactinellid
Monorhaphis chuni.
78 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
This observation suggests that the ray length of the constituent
spicules are largely synthesized in their entirety prior to incorpora-
tion into the lattice. While this may be the case, at the points of
spicule fusion there are no indications of structural boundaries,
indicating that the lattice formation is largely a continuous process
and that the process of spicule cementation may occur as a continua-
tion of the lateral growth of the constituent spicules [5]. In A. vastus,
as is the case for many other closely related hexactinellid sponges of
the order Hexactinosida, the incorporation and direct fusion of small
spicules into a rigid skeletal lattice is a common and characteristic
design strategy [4,5]. The maximum length of all spicules in these
species does not exceed a few millimeters in length. Thus, a design
strategy similar to that observed in the demosponge spicules, coupled
with the hierarchical assembly and cementation described above,
apparently is sufficiently robust to survive the loading regimes
experienced by these species.
FIGURE 3 (A–D) External armament of Aphrocallistes vastus (A). (B, C) The
external surface of A. vastus is completely covered with (D) protective pinnule
spicules, each of which measures ca. 650 mm in length. (E-G) Primary skeletal
system of Aphrocallistes vastus. Various progressively magnified views of the
honeycomb-like skeletal network of A. vastus, which is composed entirely of
fused hexactines. In this series of images, the pinnule spicules (shown in
B–D) have been removed.
Design Strategies in Skeletal Systems 79
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Rhabdocalyptus dawsoni
From a structural and mechanical perspective, the most remarkable of
the hexactinellid spicules are the significantly larger ones that form
the basis of the skeletal system of many species such as Rhabdocalyp-
tus dawsoni (Fig. 5A). These spicules, which lie spatially isolated from
one another in R. dawsoni, can frequently measure up to a centimeter
or more in length (Fig. 5B) and are characterized by their distinctive
laminated architecture. These spicules consist of a solid featureless
central cylinder of silica that encases the axial filament and is
surrounded by concentric lamellae of consolidated silica nanoparticles
FIGURE 4 (A) Exterior and (B) interior views of the highly ordered network
of fused hexactines in Aphrocallistes vastus. (C–E) Three-dimensional struc-
tural rendering of the location of the organic scaffold (shown in blue) within
the skeletal lattice of A. vastus. Data used for the 3-D renderings were
obtained from [5] and our optical microscopy studies.
80 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
and organic thin interlayers, forming a microlaminate (cf. Figs. 2B,
12B, C).
The fact that longer cylinders of a given radius are less mechani-
cally robust is intuitive, but to illustrate the point: consider two cylin-
ders of equal radii but having different lengths, both fixed at one of
their ends. If a load is applied orthogonal to the central axis of each
cylinder, then the cylinders will be perturbed from their linear shape
and the bending about the fixed end will either be balanced by a reac-
tionary force or the cylinder will begin to fail. The cylinders will flex
more with increasing distance of the applied force from the fixed
end. Because longer spicules can receive a load further from its fixed
point, they can be subjected to a greater strain per unit force applied
and therefore have a greater chance of failure. Furthermore, surface
abrasions and other defects lead to premature failure because they
act as sites for strain concentration and instigate the formation of
cracks. In nonlaminated spicules, a single crack has a high probability
of resulting in catastrophic failure. Since longer spicules can accumu-
late more abrasions and surface defects, a laminated architecture is an
effective means by which to inhibit crack propagation.
When examined in cross-section, there is a distinct reduction in
silica layer thickness from the spicule core to its periphery. The thin-
ner outer layers significantly limit the depth of crack penetration into
the spicule interior, effectively increasing the damage tolerance of this
FIGURE 5 Skeletal system of Rhabdocalyptus dawsoni. (A) Illustration of
the North Eastern Pacific hexactinellid R. dawsoni, shown approximately
life-size and (B) its mineralized skeletal system consisting largely of laminated
needle-like spicules. ((A) adapted from [5]).
Design Strategies in Skeletal Systems 81
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
composite structure [16–21]. In addition, the number of silica layers
increases with increasing spicule length, with the largest laminated
spicules from R. dawsoni containing 15 or more separate silica layers.
While it is tempting to attribute the reduction in layer thickness that
accompanies spicule diameter to the deposition of a fixed volume of
silica per layer (which would result in a continuous reduction in layer
thickness as a function of spicule radial growth), these trends are not
typically observed.
The benefits of damage tolerance and reduced sensitivity to flaws
imparted by the laminated structure can be predicted from estab-
lished models of fracture mechanics for the two limiting cases shown
in Fig. 6 [23]. In the first, the material is modeled as a series of weakly
bonded brittle plates, each of thickness t, with flaws of length c, and a
fracture toughness K
c
. Provided that the dominant crack (consisting,
say, of several broken plates) is sufficiently blunted by the organic
interlayer to completely mitigate the associated stress intensity,
renucleation can occur only when the stress in the intact layers
reaches its intrinsic strength: ro¼Kc=1:12 ffiffiffiffiffi
pc
p. Since, in this limit,
the stress in the intact layers is uniform, the fracture stress (in
FIGURE 6 Normalized strength vs. crack length for laminated and mono-
lithic materials. The thickness of the organic interlayers (highlighted in gray)
is exaggerated in the illustration.
82 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
normalized form) becomes rfffiffi
t
p=Kc¼ð1a=WÞffiffiffiffiffiffiffiffiffi
t=pc
p=1:12. In the
second case, the material is treated as a homogeneous monolithic
body, also with fracture toughness K
c
. The corresponding fracture
stress is rfffiffi
t
p=Kc¼ffiffiffiffiffiffiffiffiffi
t=W
p=1:12 ffiffiffiffiffiffiffiffiffiffiffiffi
pa=W
p. Comparisons of flaw sensitiv-
ities of the two structures are presented in Fig. 6, for the case wherein
t=W¼0.01 (a laminate with 100 layers) and an intrinsic flaw size
c=t¼0.2. The number of layers selected for this model is arbitrary,
but lies well between two of the cases discussed in this report
(R. dawsoni and M. chuni). Clearly, the effects of the laminated struc-
ture on spicule strength are significant. For instance, for a crack com-
prising five broken layers (a=t¼5 and hence a=W¼0.05), the retained
strength of the weakly bonded laminate is about five times that of
the monolithic body with the same crack dimension. The behavior of
the spicules is expected to lie between these two limits, depending
on the extent to which the organic interlayers mitigate the near-tip
stresses.
Monorhaphis chuni
A modification to the basic laminated spicule design strategy described
in R. dawsoni is observed in Monorhaphis chuni. In this species, and
in stark contrast to all other genera of sediment dwelling hexactinel-
lids (Fig. 7), skeletal support and benthic anchoring is provided by
a single monolithic anchoring spicule measuring up to three meters
long and almost one centimeter thick [22], a portion of one such spicule
is shown in Fig. 8A.
Presumably as a consequence of its large size and in response to the
local environmental conditions (prevailing currents, etc.), the spicule
develops a natural curvature (Figs. 8A, B). This creates spicule zones
of maximum tension and compression and is accompanied by a distinct
asymmetry in silica layer thickness; the thinnest exterior layers under
maximum tension help limit the depth of crack penetration, while the
thickest exterior layers prevent buckling in the zone of maximum com-
pression (Figs. 8C–F) [23]. In many instances, large spicules (greater
than 2 m in length and containing nearly 500 separate silica layers)
develop an elliptical cross-section, with the major axis of the ellipse
coinciding with the direction of curvature (Figs. 8B, C).
The layer asymmetry observed in M. chuni is unique to this species
and is most likely a stress-induced response to the predominantly uni-
directional bending regimes of this monolithic structure, thus, raising
intriguing questions as to the sensory and silica deposition regulatory
mechanisms that result in this remarkable design strategy. Despite its
exceptional rarity and deep dwelling nature (frequently encountered
Design Strategies in Skeletal Systems 83
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
at depths exceeding 2000 m), the giant anchor spicule from M. chuni
has proven to be an exceptionally useful model system for investigat-
ing a wide range of chemical, mechanical, and ultrastructural proper-
ties of laminated spicules, mainly due to its unusually large size
[24–26]. No other described species of sponge synthesizes spicules that
are even remotely comparable in dimensions with that encountered in
this species. This attribute makes it an ideal research subject for
mechanical and compositional studies of biosilica, using techniques
that are not readily applicable to smaller spicules. For instance, using
Raman spectroscopy for chemical mapping of the spicules from this
species [25], it was revealed for the first time that the organic inter-
layers are protein-rich, based on their characteristic C-H stretch and
amide vibrational signatures (Fig. 8G).
Additional information obtained from high load nanoindentation
studies reveal the remarkable energy-dissipating properties of this
laminated architecture [23]. Nanoindentation results demonstrate
FIGURE 7 Holdfast diversity in hexactinellid sponges. In stark contrast to
other sediment-dwelling hexactinellid sponges (from left to right: Hyalonema,
Chaunangium, Semperella), Monorhaphis (far right) is anchored into the sea
floor by a single giant spicule measuring up to 3 m long and nearly 1 cm thick
(adapted from [4,5,35]).
84 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
that when compared with the monolithic material encountered in the
central cylinder region immediately surrounding the axial filament,
the laminated architecture effectively inhibits crack initiation from
FIGURE 8 (A) Design features of the giant anchor spicule of Monorhaphis
chuni (sponge photo adapted from [5]). Unidirectional spicule curvature results
in asymmetrical strain accumulation. The large white arrow in (B) indicates
the direction of bending, resulting in specific zones under tension and compres-
sion. Cross-sections through the spicule shown in (B) [(C) graphical representa-
tion and (D–F) environmental SEM images] reveal the accompanying
asymmetrical silica deposition. (E) The difference in silica layer thickness from
one side of the spicule to the other (in the direction of loading) is nearly 6-fold
[(D) ca. 500 nm per layer (F) ca. 3mm per layer]. (E) The thickest layers, imme-
diately surrounding the central cylinder, measure ca. 10 mm thick. (G) Raman
chemical mapping of M. chuni spicule cross-sections. C-H stretch (2937 cm
1
)
and amide (1655 cm
1
) vibrations through the spicule cortex confirm that the
organic interlayers are protein-rich. C-H stretch vibrations (2937 cm
1
) around
the axial filament confirm the presence of extra-axial filament occluded organ-
ics within the biosilica (Si-O-Si at 800 cm
1
and Si-OH at 975 cm
1
) and an
accompanying decrease in the degree of silica condensation. (B) and (G)
adapted from [25], used with permission from Materials Research Society.
(D) and (F) adapted from [23], used with permission from Wiley.
Design Strategies in Skeletal Systems 85
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
the corners of the indents (Fig. 9). All of the applied energy appears to
be dissipated locally with no net effect on the macroscopic structural
integrity of the spicule. In addition, the extent to which the damage
field extends beyond the site of indentation critically depends on the
relative distance between the organic interlayers [23].
Cross-sections through a fractured spicule (Figs. 10A, B) clearly
reveal the crack-stopping properties of the organic interlayers, which
exhibit a distinctive stair step-like crack pattern resulting from multi-
ple sequential arrests and renucleations during crack propagation. In
addition to the crack deflecting capabilities of the organic interlayers,
a propagating crack can even be confined to a specific radial depth
as it travels parallel to the long axis of the spicule in a helical
FIGURE 9 Regiospecific indentation fracture in the giant anchor spicule of
Monorhaphis chuni. SEM images illustrating the crack morphology and
extent of damage created by (A,D) indentation before (upper) and after (lower)
sonication in the monolithic core, (B,E) the thick layer region immediately sur-
rounding the monolithic core, and (C,F) the region of minimum layer thickness
near the spicule exterior on the side of tensile loading of the spicule shown in
Fig. 8B; core (non-laminated) region, 100 mN (A) before and (D) after sonica-
tion; thick layer region, 200 mN indent, (B) before and (E) after sonication; thin
layer region, 200 mN indent, (C) before and (F) after sonication. The black
triangle in (D–F) represent the locations of the initial impressions in (A–C),
respectively. (A) and (D) adapted from [23], used with permission from Wiley.
86 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
fashion, fracturing only the silica layers immediately adjacent to the
crack trajectory (Fig. 10C). The energy dissipated in this damage mode
is significant if one considers that a 2-m long spicule measuring 5 mm
in diameter can propagate a crack over 1 m long through the spicule
without any significant loss in its structural integrity.
To investigate how the observed micromechanical properties
revealed in the nanoindentation studies affect the global macro-
mechanical properties of the giant anchor spicule of M. chuni, we con-
ducted a series of three-point bending tests (Fig. 11) on these spicules,
similar measurements of which have been reported elsewhere [21,24].
These results were compared with those obtained from duplicate tests
performed on monolithic glass rods of similar dimensions (both ca.
5 mm in diameter) and similar moduli (36 GPa for the synthetic glass
rod and 23 GPa for the spicule). The fracture stress (sometimes
referred to as the ‘‘modulus of rupture’’) was calculated from the
resulting load–displacement curves, using the assumption that the
FIGURE 10 Fracture dynamics of laminated spicules from Monorhaphis
chuni: (A,B) Cracks propagating through a damaged anchor spicule from
M. chuni exhibit a distinct step-mode, clearly illustrating the crack-deflecting
properties of the organic interlayers. (C) In some instances, rather than propa-
gating through the multilayer architecture, cracks can be spatially restricted
to a specific radial depth of the spicule and can travel parallel to its long axis in
a helical fashion (photograph of a fractured spicule illuminated end-on). (A)
adapted from [27], used with permission from AAAS. (B) adapted from [23],
with permission from Wiley.
Design Strategies in Skeletal Systems 87
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
samples behave elastically. Thus, the maximum stress on a cylindrical
beam in bending is: r¼Mr=I, where ris the radius of the cylinder, I
is the moment of inertia, and Mis the bending moment. The largest
bending moment, M, occurs in the middle of the span so that:
M¼PL=4, where Pis the load and Lis the span length. For a cylinder,
the moment of inertia, I, is: I¼pr4=4, where ris the cylinder radius.
The stress at fracture, r
fb
, for a cylinder can be defined as:
rfb ¼PfL=pr3, where P
f
is the load at fracture.
The elastic modulus, E, can also be found from our three-point
bending tests. The maximum deflection, v, occurs at the midpoint
and is: v¼PL3=48EI; thus, the modulus can be found using:
E¼L3=48IðdP=dvÞ¼L3=12pr4ðdP=dvÞ, where the quantity in par-
entheses is the linear slope of the load-displacement curve. Despite
the higher modulus of the synthetic glass rods (Fig. 11C), the yield
strength is 50%higher for the giant anchor spicules, with a corre-
sponding 45%increase in fracture stress, r
fb
(164 MPa for the mono-
lithic glass rod and 237 MPa for the giant anchor spicule), and a
ten-fold increase in toughness (calculated from the areas under the
load vs. displacement curves).
FIGURE 11 Fracture dynamics of the giant anchor spicule of Monorhaphis
chuni and a synthetic glass rod of similar diameter and modulus: (A) Photo-
graph illustrating the 5-mm diameter samples used in this study highlighting
their optical transparency (spicule, left; synthetic glass rod, right). (B) Photo-
graphs of the same structures following failure from a three-point bending test
(spicule, upper; synthetic glass rod, lower) and (C) their corresponding load vs.
displacement curves. (D) A higher magnification view of the fractured spicule
showing clear delamination of the silica layers.
88 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Euplectella aspergillum
In addition to their remarkable mechanical properties, the hexactinel-
lids are also well known for the ability to form extremely complex,
hierarchically ordered, robust skeletal networks from their constitu-
ent spicules [27,28]. One such example is Euplectella aspergillum
(Fig. 12A), a common deep-sea sediment-dwelling hexactinellid from
the Western Pacific [4,5]. As in the case of spicule fusion observed in
A. vastus, the skeletal system of E. aspergillum also consists of a rigid
skeletal lattice, although the design strategies implemented in its
construction are quite different. While the basic structural unit
from which the skeletal framework of A. vastus is constructed, is the
FIGURE 12 Hierarchical organization of the skeletal lattice of Euplectella
aspergillum. (A) Preserved specimens of E. aspergillum clearly illustrating
the cylindrical bodyplan, the external ridge system, the holdfast apparatus,
and the terminal sieve plate. Each of (D) the nonplanar cruciform (stauractine)
spicules consists of a central proteinaceous axial filament surrounded by (B,C)
concentric lamellae of consolidated silica nanoparticles and organic inter-
layers. These are combined into (E) two interpenetrating square lattices
(one shown in blue, one in yellow), which are further reinforced with (F)
bundled spicules organized vertically, horizontally, and diagonally forming
alternating open and closed squares measuring 2.5–3 mm in width. (G) The
resulting checkerboard-like configuration is consolidated with a laminated
cement (color-coded cross-section to highlight the laminated architecture)
and (H) covered by an alternating network of external ridges. The two largest
laminated spicules in (G) are the stauractines shown in (D). A photograph of
the specimen after which (H) was modeled is shown in (I). All images adapted
from [28], used with permission from Elsevier.
Design Strategies in Skeletal Systems 89
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
hexactine (Fig. 3), the framework of the skeletal lattice of E. aspergillum
is composed of stauractine (non-planar cruciform) spicules (Fig. 12D).
Like the spicules from R. dawsoni, these spicules consist of a central
proteinaceous axial filament surrounded by concentric lamellae of
consolidated silica nanoparticles and organic interlayers forming a
laminated composite (Figs. 12B, C) [27,28].
Two intersecting grids of these non-planar cruciform spicules define
a locally quadrate, globally cylindrical, skeletal lattice that provides
the framework, onto which other skeletal constituents are deposited,
as shown in Fig. 12E [27,28]. The grids are supported by bundles
of spicules that form vertical, horizontal, and diagonally ordered
struts covering the resulting square openings in an alternating
checkerboard-like manner, as depicted in Fig. 12F. Each strut consists
of a wide size range of individual spicules, ranging from 5 to 50 mmin
diameter and of variable length. These struts help stabilize the lattice
and provide additional mechanical support [16], forming a series of
nearly uniform quadrate meshes averaging 2.5–3 mm in size. The
incorporation of diagonal bracings as part of this configuration is essen-
tial for supporting the bending, shear, and torsional loads experienced
by the skeletal lattice [29]. The overall cylindrical lattice is capped at its
upper end by a terminal sieve plate and rooted into the sea floor at its
base by a flexible cluster of barbed fibrillar anchor spicules (Fig. 12A)
[30,31]. External diagonally oriented spiral ridges that extend perpen-
dicular to the surface further strengthen the lattice and likely help
prevent ovalization during lateral compression (Figs. 12H, I). A secon-
darily deposited laminated silica matrix (Fig. 12G) that cements the
structure together additionally reinforces the resulting skeletal mass.
At higher magnification, it can be seen clearly that the consolidating
silica cement precisely follows the contours of the underlying spicules,
apparently enhancing the strength of this fiber-reinforced composite,
the structure of which is similar to armored concrete.
It is important to note that the mode of skeletal consolidation
follows the general design principle also exhibited in the dominant
constituent spicules themselves. The significance of this observation
is that, due to its laminated architecture, the cement itself is able to
dissipate a significant amount of energy during loading through crack
deflection at the organic interlayers, while simultaneously distribut-
ing the applied load over the entire network of underlying spicules.
29
Si NMR STUDIES
The mechanical properties of hierarchically structured materials such
as sponge spicules are governed by both their molecular architectures
90 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
as well as by the meso- and micro-structural motifs employed by the
different sponge species, apparently selected as adaptations to their
environments. As the spicules are highly siliceous, at a molecular level
their structures are primarily governed by the relative concentrations
of so-called ‘‘Q
n
’’ silicon species (integer n4), which refer to Si atoms
bonded covalently to four bridging oxygen atoms, nof which are
bonded to other Si atoms [32]. For example, Q
4
silicon species are cova-
lently bonded to four other Si atoms via oxygen atoms; these corre-
spond to fully condensed moieties, the relative concentration of
which corresponds to the extent of cross-linking of the siliceous
matrix. Similarly, in Q
3
silicon species Si atoms are bonded via brid-
ging oxygen atoms to three other Si atoms and, thus, are incompletely
condensed. By determining the relative fractions of the different Q
n
silicon species, the molecular structures of spicules associated with dif-
ferent sponge species can be compared and correlated with their differ-
ent bulk mechanical properties.
Solid-state
29
Si nuclear magnetic resonance (NMR) spectroscopy is
sensitive to the local bonding environments of
29
Si species and can be
FIGURE 13 Room-temperature single-pulse
29
Si MAS NMR spectra of
ground spicules from the sponge species indicated. These data reveal that
regardless of the mode of skeletal construction (and the architecture of the
constituent spicules), there are no significant distinguishable differences in
the degree of silica condensation between the spicules of these five species.
Design Strategies in Skeletal Systems 91
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
used to quantify the relative concentrations of Q
n29
Si species in
spicules from different sponge species [10]. Single-pulse
29
Si NMR
spectra acquired under conditions of magic-angle sample spinning
show two broad signals centered at 101 and 111 ppm, respectively,
which correspond to the Q
3
and Q
429
Si species (Fig. 13). The absence
of downfield signals (<100 ppm) in the spectra reveal no detectable
quantities of Q
0
,Q
1
,orQ
2
species, from which it is established that
all of the silica is present in a highly (but incompletely) cross-linked
structure. Furthermore, the broad
29
Si lineshapes indicate that broad
distributions of Q
3
and Q
429
Si sites are present, consistent with an
amorphous, glass-like structure of the condensed silica networks.
Deconvolution and integration of the quantitative single-pulse
29
Si
NMR spectra in Fig. 13 establish similar extents of cross-linking,
Q
4
=(Q
3
þQ
4
)¼0.65 for all of the spicules measured. This indicates
that molecular structures of the
29
Si matrices are similar for all of
the species investigated, suggesting that observed differences in
sponge spicule mechanical properties may be attributed to differences
in macroscopic (e.g., laminated vs. non-laminated) architectures and
not from significant variability in the degree of silica condensation.
ADDITIONAL OBSERVATIONS
Recent studies investigating early larval development of the hexacti-
nellid Oopsacas minuta reveal that their syncytial architecture is sec-
ondarily derived, arising from the 32-cell stage, after which micromere
fusion results in the formation of a syncytial mass [33]. These results
are significant in that they suggest that the ancestral metazoan group
that gave rise to the Porifera was cellular and not syncytial, thus, rais-
ing intriguing questions as to the selective pressures that may have
favored the evolution of the unique hexactinellid body plan. One pos-
sible answer to this question can be found by exploring the environ-
ment to which the hexactinellids have adapted through the course of
evolution. While virtually all other sponge species occur on solid sub-
strates, the hexactinellids are unique in that they have evolved the
capacity to colonize soft sediments [4,5]. One unifying feature of the
sediment-dwelling hexactinellids is the evolution of a holdfast appara-
tus, which typically consists of basalia (anchor spicules) measuring
tens of centimeters in length (Fig. 7). These long anchor spicules are
essential for soft sediment colonization and hexactinellids lacking
basalia (Fig. 3) are confined to more solid substrates. As discussed
above, only a syncytial body plan (Fig. 14) could facilitate the synthesis
of spicules greater than a few millimeters in length. Consequently, it is
92 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
reasonable to suggest that the syncytial architecture evolved in this
group of early sponges in habitats of reduced solid substrate availabil-
ity, which characterizes the predominantly deep-sea habitats in which
the majority of hexactinellids now survive [4,5].
SUMMARY
As we have seen from these studies, the siliceous skeletal systems of
sponges are well suited to meet the structural needs of each species.
The following summarizes the general design strategies observed in
the siliceous skeletal elements of the Porifera, as discussed in this
report:
1) Spicules greater than a few millimeters in length exhibit
unique laminated architectures, which effectively retard crack
propagation.
2) The number of laminate layers increases with spicule length and
typically decreases in thickness radially outward from the core.
3) The mode of spicule consolidation, when present, follows the gen-
eral design schemes present in the dominant constituent spicules;
laminated spicules use a laminated cement, nonlaminated spicules
use a nonlaminated cement.
4) Large spicules confronting unidirectional bending regimes exhibit
a unique vectorially graded architecture for enhanced fracture
resistance.
FIGURE 14 Illustration of the hexactinellid syncytial architecture and their
associated spicules (adapted from [5]).
Design Strategies in Skeletal Systems 93
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
Additional research into the structure-function relationships of
these materials is expected to provide important new understanding
and inspiration for the fabrication of a new generation of complex,
robust, hierarchically ordered, fracture-resistant three-dimensional
composites.
ACKNOWLEDGMENTS
We thank Frank Zok, J. Herbert Waite, Shane Anderson, Nemil Vora,
Corey Hardin, Amy Butros, and Armand Kuris for helpful suggestions.
DK and NV were supported by a University of California, Riverside
Undergraduate Research Grant (#A01009-07431-40-43). JCW and
DEM were supported by grants from NASA (NAG1-01-003 and
NCC-1-02037); the Institute for Collaborative Biotechnologies, Army
Research Office (DAAD19-03D-0004); the NOAA National Sea Grant
College Program, U.S. Department of Commerce (NA36RG0537,
Project R=MP-92); and the MRSEC Program of the National
Science Foundation (DMR-00-8034). PF acknowledges support by
the Max Planck Research Award from the Alexander von Humboldt
Foundation. BFC acknowledges support from the U.S. National
Science Foundation (grant CBET-0829182).
REFERENCES
[1] Garrone, R., Simpson, T. L., and Pottu-Boumendil, J., Ultrastructure and deposition
of silica in sponges, in Silicon and Siliceous Structures in Biological Systems,T.L.
Simpson and B. E. Volcani (Eds.) (Springer-Verlag, New York, 1981), pp. 495–525.
[2] Hartman, W. D., Form and distribution of silica in sponges, in Silicon and Siliceous
Structures in Biological Systems, T. L. Simpson and B. E. Volcani (Eds.) (Springer-
Verlag, New York, 1981), pp. 453–493.
[3] Schulze, F. E., Trans. R. Soc Edinburgh. 29, 661–673 (1880).
[4] Schulze, F. E., 1887. Report on the Hexactinellida Collected by H. M. S. Challenger
During the Years 1873–1876, Volume XXI (Neill and Company, Edinburgh).
[5] Schulze, F. E., Hexactinellida, in Scientific Results of the German Deep-Sea Expedi-
tion with the Steamboat, Valdivia 1898–1899, C. Chun (Ed.) (Verlag Gustav
Fischer, Jena, Germany, 1904).
[6] Schulze, F. E., Zur Histologie der Hexactinelliden, Sitz. Ber. K. Pr. Akad. Wiss.
Berlin. 14, 198–209 (1899).
[7] Mackie, G. O. and Singla, C. L., Phil. Trans. R. Soc. Lond. 301, 365–400 (1983).
[8] Uriz, M. J., Turon, X., and Becerro, M. A., Cell Tiss. Res. 301, 299–309 (2000).
[9] Schwab, D. W. and Shore, R. E., Nature 232, 501–502 (1971).
[10] Weaver, J. C., Pietrasanta, L. I., Hedin, N., Chmelka, B. F., Hansma, P. K., and
Morse, D. E., J. Struct. Biol. 144, 271–281 (2003).
[11] Reiswig, H. M., Can. Biol. Mar. 12, 505–514 (1971).
[12] Simpson, T. L., (Springer-Verlag, New York, 1984).
[13] Simpson, T. L., Langenbruch, P. F., and Scaleraliaci, L., Zoomorphology 105,
375–382 (1985).
94 J. C. Weaver et al.
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
[14] Weaver, J. C. and Morse, D. E., Microsc. Res. Techniq. 62, 356–367 (2003).
[15] Leys, S. P., Microsc. Res. Techniq. 62, 300–311 (2003).
[16] Clegg, W. J., Kendall, K., Alford, N. M., Button, T. W., and Birchall, J. D., Nature
347, 455 (1990).
[17] Sarikaya, M., Fong, H., Sunderland, N., Flinn, B. D., Mayer, G., Mescher, A., and
Gaino, E., J. Mater. Res. 16, 1420–1428 (2001).
[18] Chai, H. and Lawn, B. R., Acta Mater. 50, 2613–2625 (2002).
[19] Seshadri, M., Bennison, S. J., Jagota, A., and Saigal, S., Acta Mater. 50, 4477–4490
(2002).
[20] Fratzl, P., Gupta, H. S., Fischer, F. D., and Kolednik, O., Advanced Materials 19,
2657–2661 (2007).
[21] Muller, W. E. G., Wang, X. H., Kropf, K., Ushijima, H., Guertsen, W., Eckert, C.,
Tahir, M. N., Tremel, W., Boreiko, A., Schlossmacher, U., Li, J. H., and Schroder,
H. C., Journal of Structural Biology 161 (2), 188–203 (2008).
[22] Miserez, A., Weaver, J. C., Thurner, P. J., Aizenberg, J., Dauphin, Y., Fratzl, P.,
Morse, D. E., and Zok, F. W., Advanced Functional Materials 18, 1241–1248 (2008).
[23] Chun, C., Aus den Tiefen des Weltmeers: Schilderungen von der Deutschen
Tieffee-Expedition, (Gustav Fischer Verlag, Jena, 1900).
[24] Levi, C., Barton, J. L., Guillemet, C., Lebras, E., and Lehuede, P. J., Mater. Sci.
Lett. 8, 337–339 (1989).
[25] Woesz, A., Weaver, J. C., Kazanci, M., Dauphin, Y., Aizenberg, J., Morse, D. E., and
Fratzl, P., Journal of Materials Research 21, 2068–2078 (2006).
[26] Mu¨ ller, W. E. G., Eckert, C., Kropf, K., Wang, X., Schloßmacher, U., Seckert, C.,
Wolf, S., Tremel, W., and Heinz, S., Cell and Tissue Research 329, 363–378 (2007).
[27] Aizenberg, J., Weaver, J. C., Thanawala, M. S., Sundar, V. C., Morse, D. E., and
Fratzl, P., Science 309, 275–278 (2005).
[28] Weaver, J. C., Aizenberg, J., Fantner, G. E., Kisailus, D., Woesz, A., Allen, P.,
Fields, K., Porter, M. J., Zok, F. W., Hansma, P. K., Fratzl, P., and Morse, D. E.,
Journal of Structural Biology 158, 93–106 (2007).
[29] Deshpande, V. S., Ashby, M. F., and Fleck, N. A., Acta Mater. 49, 1035–1040 (2001).
[30] Sundar, V. C., Yablon, A. D., Grazul, J. L., Ilan, M., and Aizenberg, J., Nature 424,
899–900 (2003).
[31] Aizenberg, J., Sundar, V. C., Yablon, A. D., Weaver, J. C., and Chen, G., Proc. Natl.
Acad. Sci. USA 101, 3358–3363 (2004).
[32] Engelhardt, G. and Michel, D., High-Resolution Solid-State NMR of Silicates and
Zeolites, (Wiley, Chichester, 1987).
[33] Leys, S. P., Cheung, E., and Boury-Esnault, N., Integrative and Comparative
Biology 46, 104–117 (2006).
[34] Iijima, I., The Hexactinellida of the Siboga Expedition, (Leiden, Brill, 1926).
[35] Gu¨ nther, A., Report on the Deep-Sea Fishes Collected by H. M. S. Challenger During
the Years 1873–1876, Volume XXII. (Neill and Company, Edinburgh, 1887).
Design Strategies in Skeletal Systems 95
Downloaded by [University of California, Riverside Libraries] at 17:27 15 August 2012
... The basic triaxonic (six-rayed) symmetry of the skeletal formations found in a diverse range of more than 600 species of glass sponges (Hexactinellida) is one of the characteristic structural features [43,44]. It is well recognized that these sponges produce microporous biosilica-based 3D hierarchical constructs with highly sophisticated network-like geometries [3,15,45]; however, the identity of the biopolymer that may be responsible for the patterning of such structures is still under investigation [23]. Figure 11 presents the results of HF-based desilicification, using the "sliding drop technique" [24], of selected fragments isolated from the Euplectella aspergillum glass sponge. ...
... Without a doubt, the modern design strategies of a new generation of engineering materials related to poriferan multiscale hierarchical structures remain a significant trend (Table 1). It is well recognized that they are based on unifying naturally occurring design strategies in sustainable skeletal systems of demosponges, homoscleromorphs, and hexactinellids [45]. described previously as biosilicificators ( Figure 12). ...
... Without a doubt, the modern design strategies of a new generation of engineering materials related to poriferan multiscale hierarchical structures remain a significant trend (Table 1). It is well recognized that they are based on unifying naturally occurring design strategies in sustainable skeletal systems of demosponges, homoscleromorphs, and hexactinellids [45]. ...
Article
Full-text available
Sponges (phylum Porifera) were among the first metazoans on Earth, and represent a unique global source of highly structured and diverse biosilica that has been formed and tested over more than 800 million years of evolution. Poriferans are recognized as a unique archive of siliceous multiscaled skeletal constructs with superficial micro-ornamentation patterned by biopolymers. In the present study, spicules and skeletal frameworks of selected representatives of sponges in such classes as Demospongiae, Homoscleromorpha, and Hexactinellida were desilicified using 10% HF with the aim of isolating axial filaments, which resemble the shape and size of the original structures. These filaments were unambiguously identified in all specimens under study as F-actin, using the highly specific indicators iFluor™ 594-Phalloidin, iFluor™ 488-Phalloidin, and iFluor™ 350-Phalloidin. The identification of this kind of F-actins, termed for the first time as silactins, as specific pattern drivers in skeletal constructs of sponges opens the way to the fundamental understanding of their skeletogenesis. Examples illustrating the biomimetic potential of sophisticated poriferan biosilica patterned by silactins are presented and discussed.
... Simultaneously, a couple of shrimps often inhabit the cylindrical cage-like structure to avoid the attack of predators. Hence it is also called the "Venus flower basket" [13][14][15][16][17][18]. ...
... One fascinating microstructural solution observed in nature is the periodic alternation of stiff (and strong) and soft (and weak) layers, which often characterise biomaterials with excellent fracture properties, as observed in the giant anchoring spicula of the deep-sea sponge Monorhaphis chuni. In these microstructures, a paramount role is played by the soft layers, as they act as crack blunting regions [9], they promote crack deflection [10], and the periodic change in properties caused by their presence is at the base of the inhomogeneity effect, which is believed to contribute to crack arrest [11]. ...
Conference Paper
Full-text available
In this study, we developed a novel microstructure inspired by the deep-sea glass sponge Monoraphis chuni to enhance longitudinal compressive performance of multidirectional carbon fibre reinforced polymer laminates. The microstructure features alternating stiff and soft regions, similar to those observed in the sponge's anchoring spicula. To create this microstructure, we utilized a unique manufacturing process. We then assessed the performance of the microstructure through small-scale notched compression tests, comparing it to an industrially-relevant baseline laminate. Our findings demonstrated a statistically significant increase in failure load and average ligament specific stress at failure compared to the baseline, as well as the ability to delay damage initiation and arrest damage propagation. Our research provides valuable insights for the design of lightweight structures under compression.
... As the applied load increases, the increasing stresses are deflected in the organic phase, and finally, cracking through the inorganic phase is initiated. A typical step-shaped stress-strain curve becomes visible during such failure that can also be observed during the failure of BnM structures [47,50,[53][54][55]. The BnM structure behaves in a similar way. ...
Article
Full-text available
Nature provides various templates for integrating organic and inorganic materials to create high-performance composites. Biological structures such as nacre and the structural elements of the glass sponge are built up in layers, leading to remarkable fracture toughness. In this work, the brick-and-mortar and layer-by-layer structures found in these biological examples have been abstracted and implemented by using an aqueous polymer dispersion in combination with nanoclay particles and sodium water glass. These dispersions were used as impregnation of carbon rovings in order to form bio-inspired contact zones towards the concrete matrix. The bonding behavior was investigated using the Yarn Pull-Out (YPO) test, and a beneficial behavior of the layered polymer–nanoclay dispersions was observed. Thermogravimetric analysis (TGA) was used to determine the organic impregnation content of the roving. Further, light microscopy of the roving cross-sections prior to YPO and visual analyses of the fractured contact zone of split concrete specimens provided information on the quality of the impregnation and the interaction with the concrete matrix.
... These spicules function as anchors to keep the Ea. sponge fixed onto the sea floor (Weaver et al., 2010), and are also called basalia spicules or anchor spicules for their function. For the sake of simplicity, we will refer to them as Ea. ...
Article
Layered architectures are prevalent in tough biological composites, such as nacre and bone. Another example of a biological composite with layered architecture is the skeletal elements—called spicules—from the sponge Euplectella aspergillum. Based on the similarities between the architectures, it has been speculated that the spicules are also tough. Such speculation is in part supported by a sequence of sudden force drops (sawtooth patterns) that are observed in the spicules' force-displacement curves from flexural tests, which are thought to reflect the operation of fracture toughness enhancing mechanisms. In this study, we performed three-point bending tests on the spicules, which also yielded the aforementioned sawtooth patterns. However, based on the analysis of the micrographs obtained during the tests, we found that the sawtooth patterns were in fact a consequence of slip events in the flexural tests. This is put into perspective by our recent study, in which we showed that the spicules' layered architecture contributes minimally to their toughness, and that the toughness enhancement in them is meager in comparison to what is observed in bone and nacre [Monn MA, Vijaykumar K, Kochiyama S, Kesari H (2020): Nat Commun 11:373]. Our past and current results underline the importance of inferring a material's fracture toughness through direct measurements, rather than relying on visual similarities in architectures or force-displacement curve patterns. Our results also suggest that since the spicules do not possess remarkable toughness, re-examining the mechanical function of the spicule's intricate architecture could lead to the discovery of new engineering design principles.
... A hinge design in a lattice can consume energy through buckling [25][26][27][28]. The excellent mechanical properties of the skeleton of E. aspergillum have been extensively studied [29,30]. However, most studies have focused on the performance of sponge spicules [31][32][33][34]. ...
Article
Full-text available
Diatoms have been described as “nanometer‐born lithographers” because of their ability to create sophisticated 3D amorphous silica exoskeletons. The hierarchical architecture of these structures provides diatoms with mechanical protection and the ability to filter, float, and manipulate light. Therefore, they emerge as an extraordinary model of multifunctional materials from which to draw inspiration. In this paper, numerical simulations, analytical models, and experimental tests are used to unveil the structural and fluid dynamic efficiency of the Coscinodiscus species diatom. Then a novel 3D printable multifunctional biomimetic material is proposed for applications such as porous filters, heat exchangers, drug delivery systems, lightweight structures, and robotics. The results demonstrate Nature's role as a materials designer for efficient and tunable systems and highlight the potential of diatoms for engineering materials innovation. Additionally, this paper lays the foundation to extend the structure‐property characterization of diatoms.
Article
Biological structural materials not only exhibit remarkable mechanical properties but also often embody dynamic characteristics such as environmental responsiveness, autonomy, and self-healing, which are difficult to achieve in conventional engineering materials. By merging materials science, synthetic biology, and other disciplines, engineered living materials (ELMs) provide a promising solution to combine living organisms with abiotic components, thus facilitating the construction of functional “living” materials. Like natural materials, ELMs possess vitality and hold immense application potential in areas such as medicine, electronics, and construction, captivating increasing research attention recently. As an emerging branch of ELMs, structural ELMs aim to mimic living biological structural materials by achieving desired mechanical performance while maintaining important “living” characteristics. Here we summarize the recent progress and provide our perspectives for this emerging research area. We first summarize the superiority of structural ELMs by reviewing biological structural materials and biomimetic material design strategies. Subsequently, we provide a systematic discussion on the definition and classifications of structural ELMs, their mechanical performance, and physiological behaviors. Finally, we summarize some critical challenges faced by structural ELMs and highlight directions of future development. We hope this review article can provide a timely summary of the state of the art and relevant perspectives for future development of structural ELMs.
Article
Full-text available
The major problem with the use of ceramics as structural materials is their brittleness. One way of overcoming this problem is to introduce weak interfaces which deflect a growing crack1. Polymer composites of this sort can be easily prepared by surrounding fibres with liquid plastic. To make similar structures with ceramic matrices and fibres is difficult and expensive, however, because traditional ceramic processing techniques of powder compaction and sintering prevent densification and cause cracking2–4. Here we describe a simple, inexpensive way of preparing a ceramic material that contains such weak interfaces. Silicon carbide powder is made into thin sheets which are coated with graphite to give weak interfaces and then pressed together and sintered without pressure. Relative to the monolithic material, the apparent fracture toughness for cracks propagating normal to the weak interfaces is increased more than fourfold, and the work required to break the samples increases by substantially more than a hundredfold. The technique should be readily applicable to other ceramics.
Article
Full-text available
Nanomechanical properties, nanohardness and elastic modulus, of an Antarctic sponge Rosella racovitzea were determined by using a vertical indentation system attached to an atomic force microscope. The Rosella spicules, known to have optical waveguide properties, are 10–20 cm long with a circular cross section of diameter 200–600 μm. The spicules are composed of 2–10-μm-thick layers of siliceous material that has no detectable crystallinity. Measurements through the thickness of the spicules indicated uniform properties regardless of layering. Both the elastic modulus and nanohardness values of the spicules are about half of that of either fused silica or commercial glass optical fibers. The fracture strength and fracture energy of the spicules, determined by 3-point bend tests, are several times those of silica rods of similar diameter. These sponge spicules are highly flexible and tough possibly because of their layered structure and hydrated nature of the silica. The spicules offer bioinspired lessons for potential biomimetic design of optical fibers with long-term durability that could potentially be fabricated at room temperature in aqueous solutions.
Chapter
Sponges, the most primitive multicellular animals, are able to concentrate and precipitate three mineral elements: iron, calcium, and silicon. Iron forms small granules deposited on and within the collagenous structures in horny sponges and freshwater sponges (Garrone, 1978). The iron compound occuring in horny sponges has been identified as a crystalline oxide, lepidocrocite (Towe and Rützler, 1968). Calcium (in the form of carbonate) and silicon (in the form of silica) are generally involved in distinct skeletal structures, the spicules. Both elements can also form secondary deposits fusing spicules together. However, only calcium constitutes massive nonspicular skeletons. Calcareous and siliceous spicules never coexist in the same species, but siliceous spicules can be associated with massive calcareous skeletons. Crystalline calcium (and magnesium) carbonate forms the calcite spicules of calcareous sponges (Jones, 1970) and the calcite or aragonite of the massive basal skeleton of sclerosponges (Hartman and Goreau, 1970, 1975). Amorphous silica is, however, the most abundant inorganic skeletal compound in the phylum Porifera, forming the spicules of the hexactinellids, of the demosponges, and of the sclerosponges. In the last-mentioned class, the siliceous spicules are scattered through the tissues of the animal, and they are also entrapped within the massive, basal, calcareous skeleton which is characteristic of these sponges. Sponges with siliceous spicules may contribute great quantities of silicon to the sediments in their environment, and in some cases sponge skeletal elements form sediment deposits that are more than 1 m thick in certain areas of the Antarctic Ocean (Sara and Vacelet, 1973).
Chapter
Opaline sclerites, spicules composed of hydrated silica, comprise all or part of the skeleton of three of the extant classes of the phylum Porifera, namely, the Demospongiae, Sclerospongiae, and Hexactinellida. The occurrence of mineralized spicules is generally regarded as one of the most distinctive characteristics of the phylum, although three of the orders of the Demospongiae lack them entirely, and in scattered species among other demosponge orders as well, spicules are absent. Furthermore, some of the genera of sclerosponges and Calcarea with non- spicular massive skeletons lack spicular elements. A majority of living sponge species do possess spicules, however, and these exhibit a great variety of forms characteristic of species and hence useful in taxonomic endeavors. Within a single sponge species from one to eight or more kinds of spicules may occur. In a number of lines of demosponges and hexactinellids there is a tendency toward the deposition of secondary deposits of silica on the basic spicule forms with the resulting formation of rigid skeletal frameworks composed of fused or interlocking spicules. Figures 16-13 to 16-30 in this chapter appear sequentially in the section entitled “Evolution of Kinds of Spicules in Sponges.”
Article
This paper presents results from the first comprehensive study of hexactinellid tissue organization by electron microscopy. It is confirmed that the trabecular tissue of Rhabdocalyptus dawsoni, which constitutes the bulk of the cellular material in the animal, is a syncytium. The dermal membrane and other similar membranes are specialized regions of the trabecular syncytium, as are thickened regions provisionally equated with the `cord syncytia' of Reiswig (Coll. int. Cent. natn. Rech. Scient. no. 291, pp. 173-180 (1979)). Trabecular tissue contributes to the walls of the flagellated chambers and provides the processes that form Reiswig's secondary reticulum. It is confirmed that choanocytes are absent. The sponge has conventional `collar bodies' (collar, flagellum and basal cytoplasm) but many collar bodies are syncytially interconnected via narrow `stolons', and there are no nuclei in these complexes in the fully differentiated state. It is suggested that collar bodies are dehiscent, and are periodically replaced. A novel feature is the perforate septum, or junctional `plug'. Plugs are not specialized portions of the cell membranes of adjacent cells. They are complex, disc-shaped structures, probably Golgi secretion products, which are inserted into syncytial bridges and appear to form a filter or partial barrier limiting translocation of materials between differentially specialized portions of the cytoplasm. In this respect they more closely resemble red algal pit connections than junctions found in animals. Gap junctions are absent in Rhabdocalyptus (and probably in all sponges) but a type of septate junction is described. Plugged junctions occur between elements of the trabecular syncytium and collar bodies, and between the latter and cells termed `choanoblasts', which are probably derived from archaeocytes. A developmental sequence is proposed wherein the collar bodies and their interconnecting stolons are produced as outgrowths from choanoblasts, which may function singly or in syncytial groups during this phase. The cytoplasm is originally continuous throughout these systems, but plugging occurs progressively, leading to segregation of collar body complexes from their mother cells. Plugged junctions are seen between a variety of cells and the trabecular tissues in which they lie. These cells, whose characteristics are described, are archaeocytes, thesocytes, choanoblasts, granulated cells, spherulous cells and gametes. Sclerocytes, however, appear to lack specialized connections with surrounding tissues. As noted by Okada (1928), spicules are produced intracellularly in hexactinellids. Spiculation has not been studied in this investigation, and no details have been obtained on embryos or development. Nerves are absent. The system responsible for impulse conduction is almost certainly the trabecular syncytium. Impulses can probably cross plugged junctions, as pores with internal diameters of about 7 nm are seen in them. There is no reason to suppose that the tissues lining the openings in the body wall or the internal water passages are contractile. Tests with the dermal membrane show that its pores are not contractile. Regulation of water flow is therefore held to be a property of the sum total of the collar body flagella. Phagosomes occur both in collar bodies and in the trabecular syncytium. It is assumed that food particles can be taken up throughout the internal surfaces. Mucus nets span the internal lacunae in some places, but information is sketchy. Mucus strands interconnect the collar microvilli and may assist in particle capture. It is suggested that food breakdown products pass directly from collar bodies to choanoblasts and trabecular issues, crossing junctional plugs, essentially a `symplastic' transport mechanism as found in plants. Archaeocytes are probably immobile and do not appear to be involved in digestion. External transport of nutrients via the mesolamella is probably of minor importance, and this densely collagenous material is probably not a pathway for cell migration. However, bacteria, presumably symbionts, do occur widely in the mesolamella. The paper concludes with a review of the phenomenon of syncytialization in plants and animals. Hexactinellids are considered in the same context and the features that set them apart from all other Porifera are listed.
Book
Although the basic principles and methods of high-resolution solid-state NMR are presented in this volume, the emphasis is on applications. References are extensive and include material published in 1986. A seven-page index is included.
Article
The driving force onto cracks, propagating inside a material where the young's modulus varies in a periodic way in a given direction, has been analyzed. A simple design creation to obtain laminate structures without driving force for crack propagation perpendicular to the lamellae has also been derived. A crack in a plane configuration of unit thickness with the crack tip located at a specific point has been considered., and the material has been been assumed to be elastic with a constant Young's modulus. An effective crack stopping occurs, when the ratio of elastic moduli is larger than about 5, and the this dose not depend much on the thickness of the soft layer. Introduction of the largest possible number of thin interlayer with a stiffness of at least five times lower than the base material has been suggested to obtain a tough multilayer based on a brittle base material.
Article
Laminated glass plates are used in several safety, security and transportation applications to enhance their structural integrity. The mechanical behavior of laminates after glass-fracture, for example energy absorbing capacity and residual stiffness, determines their utility. However, the combined influence of glass fragmentation, large deformations and interfacial decohesion has been difficult to assemble in models capable of directing mechanical design. This study provides a framework for such analyses. The cracked plate is modeled as a collection of stiff glass fragments connected by elastomeric bridging ligaments. The behavior of the elastomer layer is represented by an analytical bridging model, validated and calibrated through experiments. For simple regular crack patterns, an analytical model has been developed to study the post-cracking response of laminated glass plates. This model predicts the compliant behavior of a cracked laminated plate as a function of the adhesive strength, thickness and elastic properties of the elastomer layer, and the number of fragments and the size of the plate. Based on the bridging behavior of the elastomeric ligament, an interface/bridging finite element has been formulated for numerical simulations of cracked laminates. Mechanical behavior of cracked laminates has been studied through model experiments and excellent agreement between experiments and theory is observed.