Journal of Surface Engineered Materials and Advanced Technology, 2013, 3, 18-26
http://dx.doi.org/10.4236/jsemat.2013.33A004 Published Online September 2013 (http://www.scirp.org/journal/jsemat)
Surface Characterization of As-Spun and Supercontracted
Nephila clavipes Dragline Silk
Benoit Faugas1, Michael S. Ellison1, Delphine Dean2, Marian S. Kennedy1
1Department of Materials Science & Engineering, Clemson University, Clemson, USA; 2Department of Bioengineering, Clemson
University, Clemson, USA.
Received June 20th, 2013; revised July 19th, 2013; accepted August 5th, 2013
Copyright © 2013 Benoit Faugas et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Dragline spider silks have relatively high mass-based mechanical properties (tensile strength, elongation to break and
rupture energy) and are environmentally responsive (supercontraction). In order to produce new synthetic fibers with
these properties, many research groups have focused on identifying the chemical composition of these fibers and the
structure of the fiber core. Since each fiber also has an outer skin, our study will provide a detailed understanding of the
silk surface morphology, the response of the surface morphology to environmental conditions and processing variables,
and also determine if the silk surface has a definitive patterning of charged amino acids. Specifically, by using force
microscopy and functionalized nanoparticles, the present study examines 1) how the silk surface (topography, average
roughness) is altered due to prior mechanical loading (viz. reeling speed), 2) alterations in morphology due to envi-
ronmental conditions (supercontraction, storage time), and 3) the negatively and positively charged regions along with
the surface using both force and nanoparticle mapping. Roughness data taken on dragline silk collected from Nephila
clavipes spiders revealed that the surface comprised both smooth (5 nm RMS) and rough (65 nm RMS) regions. Super-
contracted silk (from immersion in 0.01 M PBS during AFM testing) showed higher surface roughness values compared
to spider silk tested in the air, indicating that the surface might be reorganized during supercontraction. No correlation
was found between surface roughness and neither collection speed nor aging time for the as-spun or supercontracted
fiber, demonstrating the surface stability of the dragline silk over time in terms of roughness. Both the force microscopy
and the nanoparticle methods suggested that the density of negatively charged amino acids (glutamic acid, aspartic acid)
was higher than that of the positively charged amino acids (lysine, asparagine, and histidine).
Keywords: Spider Dragline Silk; Surface Roughness; Supercontraction; Amino Acids; AFM; Force Microscopy;
Nanoparticles
1. Introduction
Orb-weaver spiders, that is, spiders that produce spiral
wheel-shaped prey-capture webs, are able to produce up
to seven different types of silk. Each silk type has unique
functions in the spider ecosphere including web struc-
ture, prey wrapping, or egg sack construction [1]. The
most well studied silk type is dragline silk, due to its
relatively high tensile strength (around 1.2 GPa) and
large extensibility (around 18%), the combination of
which results in a large work of rupture [2-9]. Interest in
this silk is also due to its ability to respond to water. Un-
restrained fibers have shown to shrink up to 60% in
length relative to their as-spun state when either fully
wetted [10,11] or subjected to a high relative humidity,
i.e., above 70% - 75% [12-14]. This ability is termed
supercontraction and is thought to be the result of break-
ing of the interchains hydrogen bonds in the amorphous
regions of the fiber [12].
Dragline silk is produced in the spider’s abdomen by
the major ampullate gland. This silk is drawn through the
spider’s spinnerets and is a cylindrically shaped fiber
with a diameter of three to five micrometers [9]. Re-
searchers have shown that the fiber has three layers: a
core, a skin and an outer layer. The outermost layer es-
sentially comprises lipids and glycoproteins that fulfill an
array of functions including protection against the envi-
ronment, fortification again mico-organixms, carriers for
pheromones in species recognition, water balance and
lubricant [15,16]. The core includes many thread-like
structures (fibrils), which are about 100 to 150 nm in size
and are lined up along the fiber axis [16]. Inside these

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 19
fibrils, nanocrystallites are found attached together via a
semi-amorphous domain [17].
Many studies have been performed to investigate
chemical and structural composition of the entire fiber
(core, skin and outer layer). The fiber contains two main
proteins: the major ampullate spidroin 1 (MaSp1) and the
major ampullate spidroin 2 (MaSp2). The primary struc-
ture of these proteins, i.e., their amino acid sequence, has
been identified: MaSp1 is made of a polyalanine block
followed by a glycine-rich region made of GGA and
GGX motifs (G = glycine, A = alanine, X = another
amino acid) [18]; MaSp2 also has a polyalanine block
but followed by GPGXX motifs (P = proline) [19]. These
polyalanine blocks are organized in dense nanocrystal-
lites and have an antiparallel β-sheet configuration [11,
20,21], whereas the semi-amorphous matrix has 31-helix
type structures, α-helices and β-turns [21,22]. Currently,
there are no published studies that focus on the structure
of the skin.
This study will provide a detailed understanding of the
silk surface morphology, i.e., the shape and form of
physical features on the surface and the response of the
surface morphology in environmental conditions and
processing variables, and will also determine if the silk
surface has a definitive patterning of charged amino ac-
ids. Specifically, by using force microscopy and func-
tionalized nanoparticles, the present study examines 1)
how the silk surface (morphology, roughness) is altered
due to prior mechanical loading (specifically the reeling
speed), 2) alterations in morphology due to environ-
mental conditions (supercontraction and storage time),
and 3) the negatively and positively charged regions
along the surface using both force mapping and SEM
imaging of functionalized nanoparticles on the silk sur-
face. It is more energetically stable to charge amino acids
to be located on the surface of a protein rather than in the
core, where the environment is neutral and hydrophobic
[23].
2. Materials and Methods
2.1. Spider Dragline Silk Collection
Female N. clavipes spiders were collected near Charles-
ton, South Carolina during the late summer months (July
and August). These spiders were watered by misting their
webs and fed with crickets each day. The dragline silk
was artificially gathered on a metal spool with the use of
a take-up reel [2] using the following steps. Each spider
was sedated by cooling her in a refrigerator for one hour
at approximately 4˚C. She was then rapidly transferred
onto a foam block and attached with curved staples
around her legs and body (Figure 1) [24]. The time to
secure the spider was not enough to warm above 20˚C
and it was assumed that temper during forced silking did
Figure 1. N. clavipes spider restrained onto a Styrofoam
block for silk collection.
not influence the dragline silk [25]. Once secured, the
spider was put under the microscope and forceps were
carefully brought inside the spinnerets. The silk was then
pinched by the forceps and connected to the reel. Since
spiders are thought normally to build their webs by pro-
ducing silk at a rate between 10 - 20 mm/s [9,25], the
collection speeds were selected around these values and
will represent normal and abnormal production speeds.
The dragline silk in this study was collected at rates rep-
resenting normal collection speeds (14.0 mm/s), slower
rates (0.7 and 2.2 mm/s) and faster rates (45.7 and 65.1
mm/s). The silk was collected between six months and
seven years before testing and was then stored inside a
dark cabinet at room temperature and humidity. By
characterizing a range of speeds and aging times, this
work will allow researchers to compare between pub-
lished results that do not identify or contain varied reel-
ing conditions and/or silk storage time.
2.2. Surface Morphology Characterization of
Spider Dragline Silk
To characterize both the surface morphology and rough-
ness of the collected silk, an atomic force microscope
(Veeco Dimension 3100, Digital Instruments) was used.
Scans were completed both in air and aqueous fluid to
compare the silk surface morphology in the as-spun and
supercontracted states respectively. Contact mode scans
were run using silicon nitride cantilevers (spring con-
stant—0.12 N/m, resonant frequency between 14 kHz
and 26 kHz) in both air and fluid (0.01 M of Phosphate
Buffer Saline). To increase the rigidity of the silk fibers
during scanning, fibers were straightened and adhered to
the slide by placing Loctite® glue beads onto the fiber at
5 mm intervals. The average strain of these fibers was
8.7%. Fibers tested in the PBS solution were in the su-
percontracted state. In our case of restrained fibers, su-
percontraction still occurred and internal forces were
induced [12-14]. All roughness measurements were made
using either 1 µm or 2 µm scan boxes. Previous work has

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 20
shown that the scan size influences the reported AFM
roughness value [26]; therefore values were only com-
pared between identical scan sizes. To compare rough-
ness without the influence of individual fiber diameters, a
2nd order fit was applied to every scan in order to remove
the curvature and then average roughness (Ra) values
were calculated using the instrument software (Nano-
Scope v613r1). Before characterizing the surface mor-
phology and surface roughness, initial tests were com-
pleted to determine the sampling influence and possible
wear. These tests were run using a supercontracted fiber.
Five scans over a 1 µm × 1 µm area were run and then
the scan size area was enlarged to 2 µm × 2 µm.
2.3. Determination of Surface Charge Density
Spacial Distribution and Labeling of
Charged Amino Acids with Gold
Nanoparticles
Both force mapping and nanoparticle labeling methods
were used to identify the arrangement of positively and
negatively charged amino acids. For charge density map-
ping, a cantilever tip was functionalized with carboxyl or
amine groups. An array of force-distance curves was
taken and then was analyzed to generate a force map
using the linear Poisson-Boltzmann model equation:
()
tip fiber
tip
4π
D
w
FR e
κ
σσ
εκ
−
= (1)
where Rtip (25 nm) is the radius of the tip, σtip is the
charge density on the tip, σfiber is the charge density on
the fiber, εw is the permittivity of the fluid (6.923 * 10−10
C2/Nm2), κ−1 is the Debye length and D is the tip-sample
separation distance. The surface charge density of the
carboxyl-terminated tips has been previously measured
and was found to be −0.01 C/m2 [27] and the Debye
length in 0.01 M PBS is approximately 3 nm [28]. For
this study, 500 nm × 500 nm square arrays with 20
force-distances curves were taken on each of the 20 rows.
This resulted in a force-distance curve being taken each
25.6 nm. This method was outlined in detail within
previous publications [28,29].
To corroborate the data from force mapping method, a
second technique using nanoparticle markers was used.
Gold nanoparticles (Cytodiagnostics, 20 nm diameter)
were functionalized with carboxyl groups (respectively
amine groups) using 11-mercaptoundecanoic acid (re-
spectively using 11-Amino-1-undecanethiol) to allow the
identification of positively (respectively negative)
charged amino acids. The solution was poured on the
unstretched restrained fiber. EDC and Sulfo-NHS were
added and allowed to react for 2 h at 37˚C, hereby at-
taching the functionalized gold nanoparticles to the tar-
geted amino acids via a stable peptide bond [30]. The
fibers samples were then washed several times in 0.1 M
PBS using sonication for 10 min at 50/60 Hz. To image
the arrangement of the nanoparticles, the Hitachi SU6600
scanning electron microscope in the Clemson University
Electron Microscope Facility in the Advanced Materials
Research Laboratory was used in backscatter mode, util-
izing the variable pressure environment capabilities of
the SU660 to reduce charging effects.
3. Results and Discussion
3.1. Reproducibility of Measurements
The reproducibility of measurements and measurement
artifacts were considered carefully. To calculate the sur-
face roughness for both the as-spun and supercontracted
states, contact AFM was utilized. Potential errors from
this method could have included plowing (scraping) of
the tip during scanning, scanning drift and/or sampling
errors. Plowing could be produced when scanning a soft
sample during contact mode. Dragline silk samples tested
in both air and PBS did not show wear boxes (Figure 2),
which would have resulted if plowing had occurred. The
silk used in these experiments was foced at a speed of
45.7 mm/s and were preformed in PBS (supercontracted
state). Figure 2 showed a fiber scanned initially with a 1
µm scanning box and subsequently using a 2 µm box
centered over the original 1 µm scanning box. In the 2
µm scan, there was no evidence of the prior 1 µm scan
and the morphology did not change (the higher features
were identical between scans). However, scanning the
same area sequentially was affected by drift illustrated
with a shift of the features towards the upper left as seen
in Figure 3. Figure 4 showed the average Ra roughness
as a function of the sequential scan number (both 1 µm
and 2 µm scan boxes). This plot showed that the rough-
ness value changes with each scan number. Since no
wear was present (demonstrated in Figure 2), this change
was attributed to the drift (demonstrated in Figure 3).
350.0 nm
0.0 1.0 μm
350.0 nm
0.0 2.0 μm
Figure 2. A square of 1 µm area (left) was scanned with a
high set point and then reimaged at 2 µm (right). The black
box in the right height image matches the area of the left
height image. Comparison of these AFM scans shows no
evidence of surface wear on the supercontacted spider drag-
line silk resulting from the AFM tip.

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 21
350.0 nm
0.0 2.0 μm
350.0 nm
0.0 2.0 μm
Figure 3. The surface of the spider silk (collected at a rate of
45.7 mm/s) was scanned repetitively using the AFM. Com-
paring the right and left height images show some drift with
the features being shifted to the upper left.
Figure 4. Sequential scans (five) taken of the same area at
two different scan sizes (1 µm × 1 µm and 2 µm × 2 µm).
The drift indicated in Figure 3 results in small changes to
the surface roughness values. This silk had a reeling speed
of 45.7 mm/s and was in the supercontracted state.
3.2. Surface Morphology of As-Spun and
Supercontracted Spider Dragline Silk
The fiber surface scans provided a better understanding
of the dragline fiber morphology in both the as-spun and
supercontracted fibers. There were two distinct surface
morphologies present along both the as-spun (Figure 5)
and supercontracted fibers (Figure 6). These compare-
sons were made using dragline fiber reeled at 65.1 mm/s
and collected six years before testing. One surface type
had large (400 nm diameter) globules present (Figure 6)
and the second had small spherical features of 40 nm
diameter (Figure 5). The large globules having extended
heights on the surface were randomly spread, thus they
can be attributed to lipids and/or glycoproteins attached
to the skin below, in accordance with the vision of sur-
face layers reported by Sponner’s et al. [13]. The small
spherical features were lined up along the longitudinal
fiber axis, which was similar to the findings of Du on N.
50.0 nm
0.0 2.0 μm
Figure 5. Representative AFM scan (height) of the surface
of as-spun (non-supercontracted) N. clavipes dragline silk.
The surface appears to have small beads that line up along
the fiber (the arrow indicates the longitudinal fiber axis)
and the surface roughness (Ra) is 4.4 nm. This silk was ex-
tracted at a reel rate of 65.1 mm/s.
150.0 nm
0.0 2.0 μm
Figure 6. Representative AFM scan images (height) taken
along the surface of supercontracted N. clavipes dragline
silk. Distinct globular shapes are evident, which likely lo-
calized glycoproteins and/or lipids. Deep ridges are also
shown to run parallel to the longitudinal fiber axis (indi-
cated with arrow). This silk was collected at a reeling rate
of 65.1 mm/s and the resulting surface roughness (Ra) is
63.5 nm.
pilipes spiders [31]. Furthermore, it has been noticed on
several of our AFM images taken in aqueous fluid that
there were some cracks on the surface, generally oriented
along the fiber axis (like the black “worm” lines on Fig-
ure 6). Since spider silk was known to have a skin-core
structure with fibrils inside the core, this crack could
have been in fact a fibril. However, if we were able to
see the fibrils, we would have seen more than one fibril
that composed the core, so this was not a likely explana-

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 22
tion. Moreover, a skin cladding surrounded the core and
this would have prevented the AFM from seeing the core.
Thus, a possible explanation was that this outer skin may
have been releasing some stress by forming these cracks.
Since the fiber was stretched, both during the silking
operation and the sample preparation (and by the super-
contraction phenomenon to a lesser extent), this cracking
may have been a result of stretching the fiber above its
surface yield elongation point during the sample prepara-
tion. This could correlate with the results of Zax et al. [7]
who found that a splitting was occurring along and across
the fiber axis after tensile testing.
The response of the surface roughness to the environ-
ment was detected by measuring the silk in air and
aqueous fluid. Both as-spun and supercontracted fibers
showed smooth and rough regions. However, the smooth
regions (Ra ranging between 0 and 20 nm) were more
prevalent in the as-spun samples (31% as shown in Fig-
ure 7) than in the supercontracted samples (7% as shown
in Figure 8). Supercontraction is known to make spider
silks shrink; thus, we may have expected to see an in-
crease in surface roughness when the pre-stressed pro-
teins contract by making the lengthened chains go higher
in height (relative to the surface) in response to their re-
duction in length. A t-test performed for the roughness
values between the as-spun and the supercontracted state
lead to [P(T ≤ t) two-tail] = 2.06 × 10−7, which was less
than 0.05. Thus, supercontraction significantly increased
Figure 7. Distribution of surface roughness values for N.
clavipes silk tested in air. The width of each bar is 5 nm.
Figure 8. Distribution of surface roughness values for N.
clavipes silk tested in 0.01 M PBS (supercontracted state).
Comparison with Figure 7 shows that the distribution
changes as fibers transition from the as-spun to supercon-
tracted states. The width of each bar is 5 nm.
surface roughness. Even though proteins may have been
different on the surface, they still were affected by su-
percontraction and went through conformational changes.
In addition to the influence of moisture in the silk
environment, the collection speed effect on surface
roughness was determined. Measurements taken along
as-spun dragline silk obtained at reeling speeds of 0.7,
2.2, 14.0, 45.7 and 65.1 mm/s (Figure 9) showed that the
average roughness did not change within this range of
collection speed. The same result was obtained for
supercontracted silk (Figure 10). It is important to note
that the roughness values reported in our study may not
have been the “exact” values because of the scan size
area used and also because of the Planefit process used to
remove the arch shaped bow on the images. However,
since every image was taken with the same scan size of 2
µm × 2 µm and was treated with a 2nd order Planefitting,
the trend reported was valid.
Figure 9. Average roughness values obtained using 42 indi-
vidual AFM scans at each collection speed showed that the
surface roughness of the as-spun silk (non-supercontracted)
was independent of collection speed. The diamonds repre-
sent a five-year-old spider silk whereas circle represents
fresh spider silk.
Figure 10. Surface roughness of supercontracted silk (av-
eraged from 42 individual scans at each collection speed) is
independent of collection speed. Roughness values were
obtained using AFM scans. The diamonds represent a five
year-old spider silk whereas circle represents fresh spider
silk.

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 23
Along with collection speed consequences on surface
roughness, the aging time effect (i.e., time since silk col-
lection) was investigated. The silk reeled at 14 mm/s was
collected six months to one year before testing compared
to the other silks that were collected at least five years
before testing. Data showed (Figures 9 and 10) that the
the surface roughness values of this “fresh” silk did not
differ from the “old” silk. This implied that this fiber
aging did not affect spider silk surface roughness, and
demonstrated the surface stability of the dragline silk
over time in terms of roughness. Thus, surface roughness
did not change when increasing collection speed nor
when aging, whether the silk was in the supercontracted
state or not.
3.3. Surface Identification of Charged Amino
Acids and Spatial Distribution of Surface
Charge Density
Sponner et al. [13] suggested within dragline silk fibers
the layer just below the lipid and glycoprotein coat com-
prised proteins having a sequence that was similar to that
of the minor ampullate silk used by spiders for web rein-
forcement. Since the primary structure of the dragline
silk consists of regular alanine and glycine repeat units in
the core, we tried to determine if a pattern for the elec-
trically charged amino acids may be seen on the surface
in terms of spatial frequency. In order to look at the
charged amino acids and their spatial distribution, a map
of surface charge densities was obtained through the use
of AFM force-distance curves taken in a 500 nm box.
The revealed representative map of surface charge densi-
ties in Figure 11(a), showed darker squares as negative
surface charge densities and white squares as positive (or
neutral) surface charge densities. It is crucial to note that
when performing force-distance curves the size of the
AFM tip implied that it interacted with more than one
amino acid on the surface, and thus the surface charge
density values unveiled an average of all the interactions
between the tip and the surrounding amino acids. There-
fore, darker squares represented regions having interac-
tions with more negatively charged amino acids than
positively charged amino acids. Plus, force-distance
curves were taken 25.6 nm apart and the radius of the tip
was 50 nm; thus measurements taken next to each other
partially probed several amino acids. Moreover, the
tip-surface interaction occurred at one point in space (1D)
but the surface charge density value obtained was ex-
tended to a square (2D) in order to get a visual map. The
map given in Figure 11(a) did not show a regular pattern
in the spatial distribution of surface charge densities.
Then charged amino acids may be seen as randomly
spread on the sub-micron scale. However, a much
smaller tip would be required to probe them individually
on the nanoscale. Furthermore, we did not report any
correlation between the surface charge densities shown in
the map and special features on the associated AFM
height image (see Figure 11(b)). Though, it is important
to note that a drift has been identified when scanning the
silk surface; thus, the force map might have been taken
with a small shift relative to the AFM image and change
our vision of a possible match between surface charge
densities and features.
The quantitative distribution of surface charge density
values given in Figure 12 revealed a slight predomi-
nance of negative surface charge densities around −0.2
C/m2. Values around −0.3 C/m2 corresponded to values
close to the control, where AFM force-distance curves
were taken on a silicon wafer coated with gold and func-
tionalized with carboxyl groups; thus force-distance
curves leading to surface charge densities values around
−0.2 C/m² probe more than one negatively charged
(a)
15.0 nm
0.0 500 nm
(b)
Figure 11. (a) This representative image is a surface charge
density map of supercontracted spider silk. The negative
surface charge density areas (darker squares) are more
prevalent than are the positive ones (white squares). The
scale bar is in C/m2, and positive values (in white) are omit-
ted from the scale bar for visual clarity; x and y axes are in
nm. (b) The associated AFM height image.

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 24
(a)
(b)
Figure 12. (a) Distribution of surface charge density values
for N. clavipes silk tested in 0.01 M PBS (supercontracted
state). This silk was collected at a rate of 14 mm/s. The
width of each bar is 0.02 C/m2. 13% of all the surface
charge density values (maximum peak) are comprised be-
tween −0.02 C/m2 and 0 C/m2. (b) Comparative distribution
of positive (dotted line) and negative (continuous line) sur-
face charge densities taken from the data in (a) as a func-
tion of absolute charge density.
amino acid, and this also meant that they can be spread a
few nanometers apart. Hence, there were more negatively
charged amino acids found on the surface compared to
positively charged amino acids.
In order to check the previous findings obtained with
force mapping, scanning electron microscopy images of
single spider dragline silk (supercontracted) coated with
gold nanoparticles functionalized with COOH and NH2
were taken (Figure 13). In Figure 13(a), the image dis-
played a fiber treated with COOH functionalized gold
nanoparticles that attached specifically to the positively
charged amino acids lysine, arginine and histidine. In
Figure 13(b), the image depicted a fiber treated with
NH2 functionalized gold nanoparticles that attached spe-
cifically to the negatively charged amino acids glutamic
acid and aspartic acid. These SEM images for amino acid
surface characterization revealed a higher amount of
negatively charged amino acids spread over the surface
(Figure 13(b)) compared to positively charged amino
acids (Figure 13(a)), confirming the findings with force
mapping. This result also correlates with the findings
(a)
(b)
Figure 13. (a) SEM images of silk fibers exposed to solutions
of PBS and surface modified gold nanoparticles. (a) This
representative image of fibers with negatively charged
nanoparticles indicate that the positively charged amino
acids are randomly spread along the fiber surface. (b) This
image shows that the negatively charged amino acids
(marked with the positively charged nanoparticles) are
more densely packed. The silk used in these tests was col-
lected at 0.7 mm/s and was collected six years before testing.
from bulk amino acid analysis that showed that nega-
tively charged amino acids were found in a higher
amount than the positively charged amino acids [2,20,32].
A possible pattern can be seen on the image of Figure
13(b) that located the negatively charged amino acids. It
was seen that gold nanoparticles were arranged as if
forming a helix around the fiber. Other images coming
from this same fiber also showed this type of arrange-
ment at different locations, but in another report, the fiber
analyzed did not show this pattern at all [23]. Images that
located positively charged amino acids did not demon-
strate any pattern, and these amino acids were spread
randomly on the surface. It could be seen that there were
some gold nanoparticles trapped between the fiber and
the microscope slide. These gold nanoparticles might not
have been attached to the targeted amino acids since they
might have been only gold nanoparticles that could have
not been removed by washing. It is also important to no-
tice that we only saw about half of the fiber because the

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 25
other half was “stuck” to the microscope slide. Hence,
we did not have access to information regarding the fiber
structure between the fiber and microscope slide, and it is
clear that we were missing the targeted amino acids.
Furthermore, it is interesting to consider that in our ex-
periments fibers were subjected to supercontraction be-
fore the attachment of gold nanoparticles. Indeed, re-
strained fibers were in contact with the aqueous gold
nanoparticles solution before the peptide bond reaction
could occur. This means that protein chains on the sur-
face were allowed to reorient and change conformation
due to supercontraction, as suggested by our roughness
experiments. Thus, the side-chains of charged amino
acids may have been available on the surface (or un-
available if they turned inside) for gold nanoparticles to
attach. Hence, SEM images may have not shown all the
charged amino acids present on the surface. It is also
worth noting that some clusters of gold nanoparticles can
be identified on the dragline silk surface. It is unlikely
that the entire surface covered by the cluster contain the
type of amino acids we were looking at, meaning that we
would have had a polypeptide chain with a long sequence
of glutamic/aspartic acid. However, these clusters were
micrometers wide and hid a certain amount of these
amino acids since negatively charged amino acids can be
found few nanometers apart, as demonstrated by the
force microscopy experiments. Thus, these clusters af-
fected the vision we had of their distribution.
4. Conclusion
This study determined the surface morphology of artifi-
cially forced N. clavipes spider dragline silk through the
use of atomic force microscopy. The silk surface, in both
the as-spun and supercontracted states, showed large and
small globule surface features and exhibited different
level of roughness ranging from 5 nm to 65 nm. A rela-
tively large surface roughness was measured when the
spider silk was tested in a water-based environment,
indicating that proteins on the surface were affected by
the supercontraction process. Although changes in envi-
ronment resulted in roughness alteration, there was no
correlation between the collection speed and or the col-
lection time with the roughness for in either the as-spun
or supercontracted states. Characterization of the chemi-
cal composition of the N. clavipes dragline silk surface
showed that both positively and negatively charged
amino acids are present on the surface. The areal density
of negatively charged amino acids (glutamic acid and
aspartic acid) was higher than the positively charged
ones (lysine, asparagine, and histidine). Initial results
indicate that the negatively charged amino acids were
spaced regularly at some locations along the fiber, while
the positively charged amino acids did not show a pat-
terned arrangement appearing randomly arranged on the
surface.
5. Acknowledgements
The authors would like to thank Mr. Donald Mulwee of
the Clemson University Electron Microscopy facility for
discussions on sample preparation and imaging condi-
tions. In addition, they would like to acknowledge Mr.
Matthew Marchewka and Mr. Joseph Durst for helpful
discussions as well as Mr. Godfrey Kimball for useful
editorial comments on this manuscript. This project was
partially funded by the Atlantis MILMI program, which
supports the cooperation between EU and US higher
education institutions. Funding support was also pro-
vided by the US Department of Education under contract
#P116J080033 (University of Central Florida) and the
European Commission under contract #2008-1750/001-
001 CPT-USTRAN (Université Bordeaux 1) and by the
Clemson University Department of Materials Science
and Engineering.
REFERENCES
[1] F. Vollrath, “Strength and Structure of Spiders’ Silks,”
Reviews in Molecular Biotechnology, Vol. 74, No. 2,
2000, pp. 67-83. doi:10.1016/S1389-0352(00)00006-4
[2] J. C. Zemlin, “A Study of the Mechanical Behavior of
Spider’s Silks,” Technical Report 69-29-CM, AD 684333,
US Army Natick Laboratories, Natick, 1968.
[3] R. W. Work, “The Force-Elongation Behavior of Web
Fibers and Silks Forcibly Obtained from Orb-Web-Spin-
ning Spider,” Textile Research Journal, Vol. 46, 1976, pp.
485-492.
[4] J. M. Gosline, P. A. Guerette, C. S. Ortlepp and K. N.
Savage, “The Mechanical Design of Spider Silks: From
Fibroin Sequence to Mechanical Function,” The Journal
of Experimental Biology, Vol. 202, 1999, pp. 3295-3303.
[5] S. L. Stauffer, S. L. Coguill and R. V. Lewis, “Compa-
rison of Physical Properties of three Silks from Nephila
clavipes and Araneus Gemmoides,” The Journal of
Arachnology, Vol. 22, No. 1, 1994, pp. 5-11.
[6] M. Hudspeth, X. Nie, W. N. Chen and R. V. Lewis, “Ef-
fect of Loading Rate on Mechanical Properties and Frac-
ture Morphology of Spider Silk,” Biomacromolecules,
Vol. 13, No. 8, 2012, pp. 2240-2246.
doi:10.1021/bm3003732
[7] D. B. Zax, D. Armanios, S. Horak, C. Malowniak and Z.
Yang, “Variation of Mechanical Properties with Amino
Acid Content in the Silk of Nephila clavipes,” Biomac-
romolecules, Vol. 5, No. 3, 2004, pp. 732-738.
doi:10.1021/bm034309x
[8] B. O. Swanson, T. A. Blackledge, J. Beltran and C. Y.
Hayashi, “Variation in the Material Properties of Spider
Dragline Silk Across Species,” Applied Physics A: Mate-
rials Science & Processing, Vol. 82, No. 2, 2006, pp.
213-218. doi:10.1007/s00339-005-3427-6
[9] P. M. Cunniff, S. A. Fossey, M. A. Auerbach, J. W. Song,

Surface Characterization of As-Spun and Supercontracted Nephila clavipes Dragline Silk 26
D. Kaplan, W. W. Adams, R. K. Eby, D. Mahoney and D.
L. Vezie, “Mechanical and Thermal Properties of Drag-
line Silk from the Spider Nephila clavipes,” Polymers for
Advanced Technologies, Vol. 5, No. 8, 1994, pp. 401-
410. doi:10.1002/pat.1994.220050801
[10] R. W. Work, “A Comparative Study of the Supercontrac-
tion of Major Ampullate Silk Fibers of Orb-Web-Build-
ing Spiders (Araneae),” Journal of Arachnology, Vol. 9,
No. 3, 1981, pp. 299-308.
[11] L. W. Jelinski, A. Blye, O. Liivak, C. Michal, G. LaVerde,
A. Seidel, N. Shah and Z. Yang, “Orientation, Structure,
Wet-Spinning, and Molecular Basis for Supercontraction
of Spider Dragline Silk,” International Journal of Bio-
logical Macromolecules, Vol. 24, No. 2-3, 1999, pp. 197-
201. doi:10.1016/S0141-8130(98)00085-3
[12] T. A. Blackledge, C. Boutry, S. C. Wong, A. Baji, A.
Dhinojwala, V. Sahni and I. Agnarsson, “How Super Is
Supercontraction? Persistent versus Cyclic Responses to
Humidity in Spider Dragline Silk,” Journal of Experi-
mental Biology, Vol. 212, No. 13, 2009, pp. 1980-1988.
doi:10.1242/jeb.028944
[13] K. N. Savage, P. A. Guerette and J. M. Gosline, “Super-
contraction Stress in Spider Webs,” Biomacromolecules,
Vol. 5, 2004, pp. 675-679. doi:10.1021/bm034270w
[14] I. Agnarsson, P. A. Guerette and J. M. Gosline, “Super-
contraction Forces in Spider Dragline Silk Depend on
Hydration Rate,” Zoology, Vol. 112, No. 5, 2009, pp.
325-331. doi:10.1016/j.zool.2008.11.003
[15] A. Sponner, W. Vater, S. Monajembashi, E. Unger, F.
Grosse and K. Weisshart, “Composition and Hierarchical
Organization of a Spider Silk,” PloS One, Vol. 2, No. 10,
2007, pp. 1-8. doi:10.1371/journal.pone.0000998
[16] K. Augsten, P. Mühlig and C. Herrmann, “Glycoproteins
and Skin-Core Structure in Nephila clavipes Spider Silk
Observed by Light and Electron Microscopy,” Scanning,
Vol. 22, 2000, pp. 12-15. doi:10.1002/sca.4950220103
[17] L. Eisoldt, A. Smith and T. Schiebel, “Decoding the Se-
crets of Spider Silk,” Materials Today (Kidlington, Eng-
land), Vol. 14, No. 3, 2011, pp. 80-86.
[18] M. Xu and R. V. Lewis, “Structure of a Protein Superfi-
ber: Spider Dragline Silk,” Proceedings of the National
Academy of Sciences, Vol. 87, No. 18, 1990, pp. 7120-
7124. doi:10.1073/pnas.87.18.7120
[19] M. Hinman, J. Jones and R. V. Lewis, “Synthetic Spider
Silk: A Modular Fiber,” Trends in Biotechnology, Vol. 18,
No. 9, 2000, pp. 374-379.
doi:10.1016/S0167-7799(00)01481-5
[20] J. O. Warwicker, “Comparative Studies of Fibroins: II.
The Crystal Structures of Various Fibroins,” Journal of
Molecular Biology, Vol. 2, 1960, pp. 350-362.
doi:10.1016/S0022-2836(60)80046-0
[21] M. Creager, J. E. Jenkins, L. A. Thagard-Yeaman, A. E.
Brooks, J. A. Jones, R. V. Lewis, G. P. Holland and J. L.
Yarger, “Solid-State NMR Comparison of Various Spi-
ders’ Dragline Silk Fiber,” Biomacromolecules, Vol. 11,
No. 8, 2010, pp. 2039-2043. doi:10.1021/bm100399x
[22] J. D. Van Beek, S. Hess, F. Vollrath and B. H. Meier,
“The Molecular Structure of Spider Dragline Silk: Fold-
ing and Orientation of the Protein Backbone,” PNAS:
Proceedings of the National Academy of Sciences, Vol.
99, No. 16, 2002, pp. 10266-10271.
[23] R. H. Garrett and C. M. Grishman, “Biochemistry,”
Brooks/Cole, 1999.
[24] B. Faugas, “Surface Characterization of Nephila clavipes
Dragline Silk,” Master’s Thesis, Clemson University,
Clemson, 2012.
[25] F. Vollrath, B. Madsen and Z. Shao, “The Effect of Spin-
ning Conditions on the Mechanics of a Spider’s Dragline
Silk,” Proceedings of the Royal Society B: Biological Sci-
ences, Vol. 268, No. 1483, 2001, pp. 2339-2346.
doi:10.1098/rspb.2001.1590
[26] P. J. Ramón-Torregrosa, M. A. Rodríguez-Valverdea, A.
Amirfazlia and M. A. Cabrerizo-Vílcheza, “Factors Af-
fecting the Measurement of Roughness Factor of Surfaces
and its Implications for Wetting Studies,” Colloids and
Surfaces A: Physicochemical and Engineering Aspects,
Vol. 323, No. 1-3, 2008, pp. 83-93.
doi:10.1016/j.colsurfa.2007.10.032
[27] J. Vandiver, D. Dean, N. Patel, W. Bonfield and C. Ortiz,
“Nanoscale Variation in Surface Charge by Synthetic
Hydroxyapatite Detected by Chemically and Spatially
Specific High Resolution Force Spectroscopy,” Biomate-
rials, Vol. 25, No. 3, 2005, pp. 271-83.
doi:10.1016/j.biomaterials.2004.02.053
[28] B. Zimmerman, J. Chow, A. G. Abbott, M. S. Ellison, M.
S. Kennedy and D. Dean, “Variation of Surface Charge
Along the Surface of Wool Fibers Assessed by High-
Resolution Force Spectroscopy,” Journal of Engineered
Fibers and Fabrics, Vol. 6, No. 2, 2011, pp. 61-66.
[29] B. Zimmerman, “Mechanical and Chemical Characteriza-
tion of Biological Composite Structures,” Master’s Thesis,
Clemson University, Clemson, 2009.
[30] G. T. Hermanson, “Bioconjugate Techniques,” Academic
Press, Boston, 2008.
[31] N. Du, Z. Yang, X. Y. Liu, Y. Li and H. Y. Xu “Struc-
tural Origin of the Strain-Hardening of Spider Silk,” Ad-
vanced Functional Materials, Vol. 21, No. 4, 2011, pp.
772-778. doi:10.1002/adfm.201001397
[32] S. Lombardi and D. Kaplan, “The Amino Acid Composi-
tion of Major Ampullate Gland Silk (Dragline) of Nephila
clavipes (Araneae, Tetragnathidae),” Journal of Arach-
nology, Vol. 18, No. 3, 1990, pp. 297-306.