<- Physical Foundations Next: Web Architecture ->

Module 1: Silk Mechanics and Chemistry

Spider silk is the most remarkable biopolymer on Earth. Pound for pound tougher than steel, more extensible than nylon, and produced at room temperature from water -- silk represents 400 million years of evolutionary optimization. This module dissects the molecular architecture, spinning process, and mechanical properties of all seven silk types, deriving their behavior from a two-phase crystalline-amorphous composite model.

Featured Lectures

Silk Biology: From DNA to Fiber

Biophysics & Computational Tools for Silk Formation — G. Holland

Silk Biology: From DNA to Fiber

The journey from a spidroin gene to a finished silk fiber spans six levels of biological organization: gene → mRNA → protein → liquid crystal dope → spinning duct → solid fiber. Each step is precisely controlled, and the spider can modulate the final fiber properties by adjusting conditions at each stage.

Step 1: Spidroin Genes

Spidroin genes are among the largest known: MaSp1 encodes a ~350 kDa protein (~3,500 amino acids) from a single exon. The gene structure is unusual: a short N-terminal domain (NTD, ~130 aa), a massive repetitive region (2,000-3,000 aa), and a short C-terminal domain (CTD, ~110 aa).

\[ \text{Gene: } \underbrace{\text{NTD}}_{130\,\text{aa}} - \underbrace{(\text{motif})_n}_{2000\text{--}3000\,\text{aa}} - \underbrace{\text{CTD}}_{110\,\text{aa}} \]

The repetitive region contains tandem repeats of short motifs (GGX, GPGXX, An) that encode the structural elements of the fiber. Gene size: ~12-15 kb for MaSp1.

Step 2: Transcription & Translation

Spidroin mRNA is produced in the tail of each silk gland, where epithelial cells are specialized for protein secretion. The mRNA has a poly(A) tail and 5' cap but minimal UTR structure. Translation on rough ER produces the full-length spidroin, which is secreted into the gland lumen.

Production rate: a single gland cell can produce ~10\(^6\) spidroin molecules per day. The major ampullate gland contains ~600 secretory cells, giving a total production of ~0.5 mg silk protein per day.

Step 3: Liquid Crystal Dope

In the gland lumen, spidroins self-assemble into a concentrated liquid crystalline solution at 30-50% w/v protein. At this concentration, the proteins should aggregate and precipitate — but they don't, thanks to the NTD and CTD domains:

NTD: pH-sensitive dimerization domain. At pH 6.3 (storage), NTD is monomeric and keeps proteins soluble
CTD: constitutive dimer that links two spidroin chains — creates the building block for fiber
The repetitive regions remain disordered in solution (intrinsically disordered proteins)

Step 4: The Spinning Duct

The S-shaped duct (20 mm long, tapering from 500 \(\mu\)m to 10 \(\mu\)m) applies three simultaneous transformations:

Ion exchange: Na\(^+\)/Cl\(^-\) pumped out, K\(^+\)/PO\(_4^{3-}\) pumped in (kosmotropic → chaotropic shift)
pH drop: 6.3 → 5.7 (triggers NTD dimerization → chain linking)
Shear alignment: elongational flow orients spidroins along fiber axis

Step 5: Phase Transition

At the critical point in the duct, the combination of shear + pH + ion change triggers an irreversible liquid-to-solid phase transition:

Alanine-rich blocks fold into \(\beta\)-sheet crystallites (antiparallel, H-bonded)
Glycine-rich blocks remain amorphous (random coil / 3\(_1\)-helix)
Water is extracted through the duct wall (80% removed)
The fiber solidifies within ~1 ms of the phase transition

Step 6: Finished Fiber

The final fiber has a core-shell structure:

Core: aligned \(\beta\)-sheet crystallites (20-30% by volume) embedded in amorphous matrix
Skin: thin lipid/glycoprotein coating (controls humidity response)
Diameter: 1-10 \(\mu\)m (species-dependent)
The spider can tune crystallinity (and thus stiffness) by varying pulling speed

\[ \text{Crystallinity} \propto \dot{\gamma}^{0.3}, \quad E \propto V_c^{1.5}, \quad \varepsilon_{break} \propto V_c^{-0.8} \]

The Central Paradox: How Does 30% Protein Stay Soluble?

At 30-50% concentration, most proteins would aggregate instantly. Spider silk dope stays liquid because of a remarkable pH-controlled molecular switch:

\[ \text{pH 6.3 (storage):} \quad \text{NTD monomer} \xrightarrow{\text{keeps soluble}} \text{liquid dope} \]\[ \text{pH 5.7 (duct):} \quad \text{NTD dimer} \xrightarrow{\text{chain linking}} \text{fiber nucleation} \]

The NTD has a single histidine residue (pKa ~6.0) that protonates at pH 5.7, triggering a conformational change from monomer to dimer. This links spidroin chains end-to-end, creating the high-molecular-weight polymers needed for fiber formation. The CTD dimer provides the other end-to-end link, creating an effectively infinite polymer chain. Derive the critical pH:

\[ K_{dimer}(\text{pH}) = K_0 \cdot \frac{[\text{H}^+]}{[\text{H}^+] + K_a^{His}} = K_0 \cdot \frac{1}{1 + 10^{\text{pH} - \text{p}K_a}} \]

1. The Seven Types of Spider Silk

Orb-weaving spiders (Araneidae, Tetragnathidae, Nephilidae) produce up to seven distinct silk types, each from a specialized gland connected to specific spigots on the spinnerets. Each silk is optimized for a different mechanical function:

1. Major Ampullate (MA)

Dragline and web frame silk. Strongest silk type: tensile strength ~1.1 GPa, toughness ~160 MJ/m^3. Produced by the major ampullate glands in the abdomen. The spider's safety line.

2. Minor Ampullate (MiA)

Temporary scaffolding spiral during web construction. Stiffer than MA but less extensible (~5% strain). Used as auxiliary thread and reinforcement.

3. Flagelliform (Flag)

Capture spiral of the orb web. Extraordinarily extensible: up to 270% strain. Low stiffness but very high toughness. Coated with aggregate glue.

4. Tubuliform (Tu)

Egg sac construction. High stiffness and tensile strength. Dense, tough sheet silk that protects eggs from mechanical damage, desiccation, and parasites.

5. Aciniform (Ac)

Prey wrapping and sperm web. Highest toughness of all silks (~250 MJ/m^3). Good balance of strength and extensibility. Swathing bands.

6. Aggregate (Ag)

Aqueous glycoprotein glue coating on capture spiral. Not a true fiber -- forms viscous droplets. Contains hygroscopic compounds (GABamide, KNO3) that absorb water.

7. Pyriform (Py)

Attachment discs that anchor silk to surfaces. Forms a cement-like bond. Highest stiffness of all silks. Spun as multiple fine fibers that fuse into a disc.

2. Spidroin Protein Architecture

All spider silks are composed of spidroins -- large (250-350 kDa) repetitive proteins with conserved N- and C-terminal domains flanking a long repetitive core. The two main dragline spidroins are MaSp1 and MaSp2:

MaSp1 (Major Ampullate Spidroin 1): Contains GGX repeats (where X = Ala, Leu, Tyr, Gln) that form 3₁-helical structures in the amorphous matrix, plus poly-alanine blocks (A)₄₋₈ that form crystalline\(\beta\)-sheets. MaSp1 is more hydrophobic than MaSp2.
MaSp2 (Major Ampullate Spidroin 2): Contains GPGXX repeats (where X = Gly, Gln, Tyr) that form \(\beta\)-spiral structures (similar to elastin). The proline content makes MaSp2 more elastic. Also contains poly-Ala crystalline blocks. MaSp2 is responsible for supercontraction behavior.

The \(\beta\)-sheet crystallites serve as physical crosslinks in an amorphous polymer matrix. This gives silk its extraordinary combination of properties -- the crystallites provide strength while the amorphous regions provide extensibility. The silk fiber is thus modeled as a two-phase composite:

\(\sigma = E_c \cdot V_c \cdot \varepsilon + E_a \cdot V_a \cdot \varepsilon\)

Rule of mixtures for the elastic regime: \(E_c\) = crystallite modulus (~160 GPa),\(V_c\) = crystallite volume fraction (~0.15-0.25),\(E_a\) = amorphous modulus (~2 GPa),\(V_a = 1 - V_c\). Beyond the yield strain (~2%), the amorphous phase undergoes unfolding and chain alignment, producing strain hardening.

3. Hierarchical Structure of Silk

Spider silk exhibits a hierarchical structure spanning from angstrom-scale amino acids to micrometer-scale fibers. Each level of hierarchy contributes distinct mechanical properties.

4. The Spinning Process

The transformation from liquid protein dope to solid silk fiber is a marvel of biological engineering. The spinning dope is a concentrated aqueous solution (30-50% w/v protein) stored in the gland lumen. As it passes through the spinning duct, a series of physical and chemical changes convert it to a solid fiber:

Storage in gland lumen: Spidroins are stored as a liquid crystalline solution. The N- and C-terminal domains prevent premature aggregation by maintaining pH-dependent solubility.
Acidification: pH drops from ~7.2 (gland) to ~6.3 (duct exit). This triggers conformational changes in the terminal domains, promoting protein-protein interactions.
Ion exchange: Na⁺ and Cl⁻ are absorbed from the dope; K⁺ and PO₄³⁻ are secreted in. Phosphate promotes \(\beta\)-sheet formation (kosmotropic effect).
Shear alignment: The narrowing duct (funnel shape) creates shear flow. Above a critical shear rate \(\dot{\gamma}_c\), the elongated spidroins align along the fiber axis.
Water extraction: ~80% of water is removed through the duct epithelium. Protein concentration increases from ~30% to ~80%.
Crystallization: Aligned, dehydrated poly-Ala blocks stack into \(\beta\)-sheet crystallites via hydrogen bonding. This is irreversible.

The spider controls fiber properties by varying the spinning speed. The critical shear rate for crystallization is:

\(\dot{\gamma}_c = \frac{v}{r_{duct}} \approx \frac{k_B T}{6\pi \eta R_g^3}\)

where v is the spinning speed (1-70 cm/s), \(r_{duct}\) is the duct radius (~10-50 um), \(\eta\) is the dope viscosity, and \(R_g\) is the spidroin radius of gyration. Faster spinning = higher crystallinity = stiffer fiber.

This is why forcibly silked fibers (reeled at high speed) have different properties than naturally spun silk. The spider dynamically tunes its silk by adjusting spinning speed, valve geometry, and the relative contribution of different spidroins.

5. Mechanical Properties and the Two-Phase Model

Major ampullate (dragline) silk has mechanical properties that surpass most engineering materials:

Tensile strength: 1.1 GPa (steel: 0.5 GPa, Kevlar: 3.6 GPa)
Toughness: 160 MJ/m³ (steel: 6 MJ/m³, Kevlar: 50 MJ/m³)
Extensibility: 30% (steel: 0.8%, nylon: 18%)
Density: 1.3 g/cm³ (steel: 7.8 g/cm³)

The key insight is that silk's extraordinary toughness arises from its unique stress-strain behavior: initial elastic stiffness (from both phases), followed by a yield point and extensive strain hardening (from amorphous chain unfolding and alignment). The total energy absorbed is the integral:

\(U = \int_0^{\varepsilon_f} \sigma(\varepsilon) \, d\varepsilon\)

This is why silk outperforms steel in toughness despite lower ultimate strength: silk's 30% extensibility gives it ~50x more area under the stress-strain curve than steel's 0.8%.

The full two-phase stress-strain model divides the response into three regimes:

\(\sigma(\varepsilon) = \begin{cases} (E_c V_c + E_a V_a) \varepsilon & \varepsilon < \varepsilon_y \\ \sigma_y + E_a V_a (\varepsilon - \varepsilon_y) \cdot h + \alpha E_c V_c (\varepsilon - \varepsilon_y)^2 & \varepsilon_y < \varepsilon < \varepsilon_h \\ \sigma_h + E_c V_c (\varepsilon - \varepsilon_h) & \varepsilon > \varepsilon_h \end{cases}\)

Regime 1: linear elastic. Regime 2: amorphous unfolding with strain hardening (h = hardening coefficient). Regime 3: crystallite loading to failure. The \(\alpha\) term captures nonlinear stiffening from chain alignment.

Simulation: Stress-Strain Curves for All 7 Silk Types

Left: Stress-strain curves generated from the two-phase model for all seven silk types. Note the enormous extensibility of flagelliform silk (~270%) versus the high strength of MA silk (~1.1 GPa). Center: Toughness comparison with engineering materials. Right: Model fit against experimental data for MA dragline silk.

Python

script.py93 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Stress-strain curves for all 7 silk types
# Data from: Blackledge & Hayashi (2006), Gosline et al. (1999), Vollrath & Knight (2001)

def two_phase_model(strain, E_c, V_c, E_a, V_a, eps_yield, strain_hardening):
    """Two-phase composite model: crystalline beta-sheets + amorphous matrix"""
    stress = np.zeros_like(strain)
    for i, eps in enumerate(strain):
        if eps < eps_yield:
            # Linear elastic: both phases contribute
            stress[i] = (E_c * V_c + E_a * V_a) * eps
        else:
            # Post-yield: crystallites intact, amorphous phase stretches
            yield_stress = (E_c * V_c + E_a * V_a) * eps_yield
            stress[i] = yield_stress + E_a * V_a * (eps - eps_yield) * strain_hardening
            # Strain hardening from unfolding and alignment
            stress[i] += 0.3 * E_c * V_c * (eps - eps_yield)**2
    return stress / 1e9  # Convert to GPa

# Silk type parameters: [name, E_c(GPa), V_c, E_a(GPa), V_a, eps_yield, max_strain, strain_hard, color]
silk_types = [
    ("Major Ampullate\n(dragline)", 160e9, 0.20, 2.0e9, 0.80, 0.02, 0.30, 1.5, '#c084fc'),
    ("Minor Ampullate\n(scaffolding)", 120e9, 0.18, 1.5e9, 0.82, 0.02, 0.05, 1.0, '#a78bfa'),
    ("Flagelliform\n(capture spiral)", 3e9, 0.05, 0.5e9, 0.95, 0.03, 2.70, 0.3, '#22d3ee'),
    ("Tubuliform\n(egg sac)", 140e9, 0.22, 3.0e9, 0.78, 0.015, 0.20, 2.0, '#f472b6'),
    ("Aciniform\n(prey wrapping)", 100e9, 0.15, 2.5e9, 0.85, 0.02, 0.80, 1.2, '#34d399'),
    ("Aggregate\n(glue droplets)", 0.5e9, 0.02, 0.01e9, 0.98, 0.05, 10.0, 0.1, '#fbbf24'),
    ("Pyriform\n(attachment)", 130e9, 0.25, 2.0e9, 0.75, 0.01, 0.10, 3.0, '#f87171'),
]

fig, axes = plt.subplots(1, 3, figsize=(16, 6))
fig.patch.set_facecolor('#030712')
fig.suptitle("Spider Silk Mechanical Properties: 7 Silk Types", fontsize=15, color='white', fontweight='bold', y=1.01)

# Panel 1: Stress-strain curves
ax1 = axes[0]
ax1.set_facecolor('#1a0a2e')
for name, E_c, V_c, E_a, V_a, eps_y, max_eps, sh, color in silk_types:
    eps = np.linspace(0, max_eps, 500)
    sigma = two_phase_model(eps, E_c, V_c, E_a, V_a, eps_y, sh)
    label = name.split("\n")[0]
    ax1.plot(eps * 100, sigma, color=color, linewidth=2, label=label)
ax1.set_xlabel('Strain (%)', color='white', fontsize=11)
ax1.set_ylabel('Stress (GPa)', color='white', fontsize=11)
ax1.set_title('Stress-Strain Curves\n(Two-Phase Model)', color='#d8b4fe', fontsize=11)
ax1.set_xlim(0, 100)
ax1.set_ylim(0, 1.5)
ax1.tick_params(colors='white'); ax1.grid(True, alpha=0.2, color='#a78bfa')
for sp in ax1.spines.values(): sp.set_color('#7c3aed')
ax1.legend(fontsize=7, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white', loc='upper left')

# Panel 2: Toughness comparison
ax2 = axes[1]
ax2.set_facecolor('#1a0a2e')
# Toughness = area under stress-strain curve (MJ/m^3)
materials = ['MA Silk', 'Flagelliform', 'Aciniform', 'Kevlar', 'Steel', 'Nylon', 'Rubber']
toughness = [160, 150, 250, 50, 6, 80, 100]
colors_t = ['#c084fc', '#22d3ee', '#34d399', '#fbbf24', '#94a3b8', '#f472b6', '#f87171']
bars = ax2.barh(materials, toughness, color=colors_t, edgecolor='white', linewidth=0.5, height=0.6)
ax2.set_xlabel('Toughness (MJ/m^3)', color='white', fontsize=11)
ax2.set_title('Toughness Comparison\nSilk vs Engineering Materials', color='#d8b4fe', fontsize=11)
ax2.tick_params(colors='white', labelsize=10); ax2.grid(True, alpha=0.2, color='#a78bfa', axis='x')
for sp in ax2.spines.values(): sp.set_color('#7c3aed')
for bar, val in zip(bars, toughness):
    ax2.text(val + 3, bar.get_y() + bar.get_height()/2, f'{val}', va='center', color='white', fontsize=9)

# Panel 3: Two-phase model fit for MA silk
ax3 = axes[2]
ax3.set_facecolor('#1a0a2e')
# Experimental data points (Gosline et al., 1999)
exp_strain = np.array([0, 0.5, 1, 2, 3, 5, 8, 12, 18, 22, 27, 30]) / 100
exp_stress = np.array([0, 0.15, 0.3, 0.5, 0.52, 0.54, 0.58, 0.65, 0.80, 0.95, 1.10, 1.15])
ax3.scatter(exp_strain * 100, exp_stress, c='#f472b6', s=50, zorder=5, edgecolors='white',
            linewidth=0.5, label='Experimental (Gosline)')
# Model fit
eps_model = np.linspace(0, 0.30, 500)
sigma_model = two_phase_model(eps_model, 160e9, 0.20, 2.0e9, 0.80, 0.02, 1.5)
ax3.plot(eps_model * 100, sigma_model, color='#c084fc', linewidth=2.5, label='Two-phase model')
ax3.set_xlabel('Strain (%)', color='white', fontsize=11)
ax3.set_ylabel('Stress (GPa)', color='white', fontsize=11)
ax3.set_title('MA Silk: Model vs Data\nCrystalline + Amorphous Phases', color='#d8b4fe', fontsize=11)
ax3.tick_params(colors='white'); ax3.grid(True, alpha=0.2, color='#a78bfa')
for sp in ax3.spines.values(): sp.set_color('#7c3aed')
ax3.legend(fontsize=10, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#030712')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

6. Supercontraction

One of silk's most remarkable properties is supercontraction: when wetted (above ~70% relative humidity), dragline silk contracts by up to 50% of its original length. This is driven by water molecules disrupting the hydrogen bonds in the amorphous phase, allowing the previously extended chains to relax to a higher-entropy (coiled) configuration.

The thermodynamic driving force is entropy. The extended amorphous chains are in a low-entropy state (ordered, aligned). When hydrogen bonds are disrupted by water, the chains relax to their maximum-entropy (random coil) configuration, contracting the fiber. This is the same physics as rubber elasticity:

\(\sigma = nk_BT\left(\lambda - \frac{1}{\lambda^2}\right)\)

where n is the crosslink density (crystallites per unit volume), \(k_B\) is Boltzmann's constant, T is temperature, and \(\lambda = L/L_0\) is the stretch ratio. For \(\lambda < 1\) (contraction), the stress is negative (compressive), driving the fiber to shorten.

Supercontraction has biological significance: morning dew causes the web to contract, tensioning sagging threads and restoring the web's prey-catching geometry. It effectively gives the spider a self-repairing web that retensions itself after perturbation.

The MaSp2 spidroin is primarily responsible for supercontraction due to its high proline content (GPGXX motifs). The proline ring constrains the backbone, and the hydrogen bonds stabilizing the extended conformation are more water-accessible than those in MaSp1. Species with higher MaSp2/MaSp1 ratios show greater supercontraction.

Simulation: Supercontraction Model

Left: Rubber elasticity model comparing dry and wet silk stress-strain behavior. The wet silk has fewer effective crosslinks (H-bonds disrupted), resulting in lower stress and a contracted equilibrium. Center: Supercontraction percentage vs. relative humidity showing the sigmoid transition at ~70% RH. Right: Free energy landscape showing how wetting shifts the equilibrium from extended to contracted conformation.

Python

script.py88 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Supercontraction model: entropy-driven rubber elasticity
# sigma = nkT(lambda - 1/lambda^2) where lambda = L/L0

# Panel 1: Supercontraction stress-strain (wet vs dry)
# Panel 2: Contraction ratio vs humidity
# Panel 3: Energy landscape during wetting

fig, axes = plt.subplots(1, 3, figsize=(16, 6))
fig.patch.set_facecolor('#030712')
fig.suptitle("Spider Silk Supercontraction", fontsize=15, color='white', fontweight='bold', y=1.01)

# Panel 1: Rubber elasticity model
ax1 = axes[0]
ax1.set_facecolor('#1a0a2e')

lam = np.linspace(0.5, 3.0, 500)  # stretch ratio

# Dry silk: high crystallinity, stiff
n_dry = 2e26     # crosslink density (m^-3) in dry silk
kT = 4.11e-21    # J at 25C
sigma_dry = n_dry * kT * (lam - 1.0 / lam**2)

# Wet silk: amorphous regions relax, rubber-like
n_wet = 5e25     # fewer effective crosslinks after H-bond disruption
sigma_wet = n_wet * kT * (lam - 1.0 / lam**2)

ax1.plot(lam, sigma_dry / 1e6, color='#c084fc', linewidth=2.5, label='Dry silk')
ax1.plot(lam, sigma_wet / 1e6, color='#22d3ee', linewidth=2.5, label='Wet silk (supercontracted)')
ax1.axhline(0, color='white', linewidth=0.5, alpha=0.3)
ax1.axvline(1.0, color='white', linewidth=0.5, alpha=0.3, linestyle='--')
ax1.fill_between(lam, sigma_dry / 1e6, sigma_wet / 1e6, alpha=0.1, color='#a78bfa')
ax1.set_xlabel('Stretch Ratio (lambda = L/L0)', color='white', fontsize=11)
ax1.set_ylabel('Stress (MPa)', color='white', fontsize=11)
ax1.set_title('Rubber Elasticity Model\nsigma = nkT(lam - 1/lam^2)', color='#d8b4fe', fontsize=11)
ax1.set_ylim(-200, 1200)
ax1.tick_params(colors='white'); ax1.grid(True, alpha=0.2, color='#a78bfa')
for sp in ax1.spines.values(): sp.set_color('#7c3aed')
ax1.legend(fontsize=10, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

# Panel 2: Contraction vs relative humidity
ax2 = axes[1]
ax2.set_facecolor('#1a0a2e')
RH = np.linspace(0, 100, 200)
# Sigmoid transition: contraction onset at ~70% RH
contraction = 50 / (1 + np.exp(-0.15 * (RH - 72)))  # max ~50% contraction
ax2.plot(RH, contraction, color='#e879f9', linewidth=2.5)
ax2.fill_between(RH, 0, contraction, alpha=0.15, color='#e879f9')
ax2.axvline(70, color='#fbbf24', linewidth=1.5, linestyle='--', label='Critical RH (~70%)')
ax2.axhline(50, color='#f87171', linewidth=1, linestyle=':', alpha=0.5, label='Max contraction (~50%)')
# Data points (Guinea et al. 2005)
rh_data = [30, 40, 50, 60, 70, 75, 80, 85, 90, 95, 100]
cont_data = [0.5, 1.0, 1.5, 3, 10, 20, 32, 40, 45, 48, 50]
ax2.scatter(rh_data, cont_data, c='#34d399', s=50, zorder=5, edgecolors='white', linewidth=0.5, label='Data (Guinea et al.)')
ax2.set_xlabel('Relative Humidity (%)', color='white', fontsize=11)
ax2.set_ylabel('Contraction (%)', color='white', fontsize=11)
ax2.set_title('Supercontraction vs Humidity\nSigmoid transition at ~70% RH', color='#d8b4fe', fontsize=11)
ax2.tick_params(colors='white'); ax2.grid(True, alpha=0.2, color='#a78bfa')
for sp in ax2.spines.values(): sp.set_color('#7c3aed')
ax2.legend(fontsize=9, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

# Panel 3: Free energy landscape
ax3 = axes[2]
ax3.set_facecolor('#1a0a2e')
x = np.linspace(0, 1, 200)  # order parameter (0=disordered, 1=ordered H-bonds)
# Dry: H-bonds stabilize extended conformation
G_dry = -5 * x**2 + 8 * x**4 - 2 * x  # double well, biased to ordered state
# Wet: water disrupts H-bonds, disordered state favored
G_wet = -1 * x**2 + 8 * x**4 + 3 * x  # biased to disordered state
ax3.plot(x, G_dry - G_dry.min(), color='#c084fc', linewidth=2.5, label='Dry (H-bonds intact)')
ax3.plot(x, G_wet - G_wet.min(), color='#22d3ee', linewidth=2.5, label='Wet (H-bonds disrupted)')
ax3.annotate('Extended\nconformation', xy=(0.85, 0.5), fontsize=9, color='#c084fc', ha='center')
ax3.annotate('Contracted\n(entropy max)', xy=(0.15, 0.5), fontsize=9, color='#22d3ee', ha='center')
ax3.set_xlabel('Order Parameter (H-bond fraction)', color='white', fontsize=11)
ax3.set_ylabel('Free Energy (arb. units)', color='white', fontsize=11)
ax3.set_title('Free Energy Landscape\nWetting shifts equilibrium', color='#d8b4fe', fontsize=11)
ax3.tick_params(colors='white'); ax3.grid(True, alpha=0.2, color='#a78bfa')
for sp in ax3.spines.values(): sp.set_color('#7c3aed')
ax3.legend(fontsize=10, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#030712')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

7. Energy Absorption: Why Silk Beats Steel

The toughness (energy absorbed per unit volume before failure) is the integral of the stress-strain curve. For a material with linear elastic behavior to failure:

\(U_{elastic} = \frac{1}{2}\sigma_f \varepsilon_f = \frac{\sigma_f^2}{2E}\)

For steel: \(U_{steel} = \frac{(0.5 \times 10^9)^2}{2 \times 200 \times 10^9} \approx 0.6 \;\text{MJ/m}^3\). But steel can yield and plastically deform, adding ~5 MJ/m³ for a total of ~6 MJ/m³.

For MA silk, the two-phase model gives three contributions:

\(U_{silk} = \underbrace{\frac{1}{2}\sigma_y \varepsilon_y}_{\text{elastic}} + \underbrace{\int_{\varepsilon_y}^{\varepsilon_h} \sigma(\varepsilon)\,d\varepsilon}_{\text{unfolding}} + \underbrace{\int_{\varepsilon_h}^{\varepsilon_f} \sigma(\varepsilon)\,d\varepsilon}_{\text{crystallite loading}} \approx 160 \;\text{MJ/m}^3\)

The dominant contribution (~75%) comes from the amorphous chain unfolding regime. This is the key design principle: silk achieves extreme toughness not through brute strength but through controlled, energy-dissipating deformation of its amorphous phase. This principle is now being applied to biomimetic materials design.

Module Summary

7 Silk Types

Major/minor ampullate, flagelliform, tubuliform, aciniform, aggregate, pyriform -- each from a specialized gland with distinct mechanical properties

Spidroin Proteins

MaSp1 (GGX, hydrophobic) and MaSp2 (GPGXX, proline-rich) form beta-sheet crystallites in amorphous matrix

Two-Phase Model

sigma = E_c*V_c*eps + E_a*V_a*eps; crystallites (~20% vol) provide strength, amorphous matrix provides extensibility

Spinning Process

Liquid dope -> acidification -> ion exchange -> shear alignment -> water extraction -> crystallization; spider controls properties via speed

Supercontraction

~50% shortening when wetted; entropy-driven rubber elasticity: sigma = nkT(lam - 1/lam^2); self-tensioning web

Toughness Champion

MA silk: 160 MJ/m^3 vs steel 6 MJ/m^3; from amorphous chain unfolding, not brute strength

8. Detailed Biochemistry of Each Silk Type

Each of the seven silk types is encoded by distinct spidroin genes, produced in specialized glands with unique morphologies, and exhibits a characteristic amino acid composition that dictates its mechanical properties. Below is a comprehensive breakdown of every silk type.

1. Major Ampullate (MA) Silk — Dragline & Frame

Spidroin Genes

MaSp1 (~250 kDa) and MaSp2 (~300 kDa). MaSp1/MaSp2 ratio varies by species: Nephila clavipes ~80:20, Argiope aurantia ~50:50.

Amino Acid Composition

Gly: 42%, Ala: 25%, Pro: 2-10% (MaSp2-dependent), Ser: 4%, Gln: 10%, Tyr: 4%, Leu: 3%. The high Gly+Ala content (>65%) is characteristic of silks with \(\beta\)-sheet crystallites.

Gland Morphology

Large pear-shaped glands (major ampullate glands) in the anterior abdomen. Each gland has a tail (protein synthesis), sac (storage lumen), and S-shaped duct (20 mm). Paired glands connect to anterior spinnerets.

Mechanical Properties

E = 10 GPa, \(\sigma_f\) = 1.1 GPa, \(\varepsilon_f\) = 30%, U = 160 MJ/m³. Fiber diameter: 1-8 μm (species-dependent).

Analogy: Like a climber's dynamic rope but 5x tougher per unit weight. If you could make a rope of MA silk the diameter of a pencil, it could stop a falling Boeing 747.

2. Minor Ampullate (MiA) Silk — Scaffolding

Spidroin Gene

MiSp1 (~250 kDa). Contains GGX repeats similar to MaSp1 but lacks the GPGXX motifs of MaSp2. Also has a distinctive GGYGGY spacer motif.

Amino Acid Composition

Gly: 45%, Ala: 28%, Pro: <1%, Ser: 6%, Tyr: 3%. Very low proline content means minimal supercontraction — the scaffolding stays dimensionally stable when wet.

Gland Morphology

Smaller pear-shaped glands posterior to the major ampullate glands. Connected to median spinnerets. Shorter duct (~10 mm) than MA glands.

Mechanical Properties

E = 12 GPa, \(\sigma_f\) = 0.6 GPa, \(\varepsilon_f\) = 5%, U = 30 MJ/m³. Fiber diameter: 1-4 μm. Stiff but brittle compared to MA silk.

Analogy: Like a stiff fishing line — holds its shape under load but snaps rather than stretching. Perfect for the temporary spiral that guides web construction.

3. Flagelliform (Flag) Silk — Capture Spiral

Spidroin Gene

Flag (~360 kDa). Dominated by GPGGX pentapeptide repeats forming \(\beta\)-spiral spring-like structures. Contains very few poly-Ala blocks — hence minimal crystallinity.

Amino Acid Composition

Gly: 44%, Pro: 18%, Ala: 8%, Val: 9%, Ser: 3%. The exceptionally high proline content gives the backbone flexibility by disrupting regular secondary structure.

Gland Morphology

Small, elongated flagelliform glands connected to posterior spinnerets. The duct is shorter than MA (~5 mm) and produces a single fiber per gland, which is then coated with aggregate glue.

Mechanical Properties

E = 0.003 GPa, \(\sigma_f\) = 0.5 GPa, \(\varepsilon_f\) = 270%, U = 150 MJ/m³. Fiber diameter: 1-5 μm. The most extensible silk type.

Analogy: Like a bungee cord — stretches to nearly 4x its length before breaking. Absorbs the kinetic energy of a flying insect without catapulting it back out.

4. Tubuliform (Tu) Silk — Egg Sac

Spidroin Gene

TuSp1 (~320 kDa). Unique among spidroins: contains long Ala-rich and Ser-rich blocks that form extensive \(\beta\)-sheet crystallites. High crystallinity (~30%).

Amino Acid Composition

Gly: 15%, Ala: 28%, Ser: 25%, Pro: <1%, Gln: 8%. The unusually high serine content (vs other silks) creates additional H-bonding capacity in the crystalline phase.

Gland Morphology

Large cylindrical (tubuliform) glands, also called cylindriform glands. Present only in female spiders. Multiple glands per spinneret produce many fibers simultaneously for sheet silk.

Mechanical Properties

E = 14 GPa, \(\sigma_f\) = 0.4 GPa, \(\varepsilon_f\) = 20%, U = 40 MJ/m³. Fiber diameter: 5-15 μm. Thicker fibers than other silk types.

Analogy: Like a tough canvas tarp — stiff, dense, protective. Shields hundreds of eggs from mechanical damage, parasitoid wasps, fungal infection, and desiccation.

5. Aciniform (Ac) Silk — Prey Wrapping

Spidroin Gene

AcSp1 (~300 kDa). Contains complex, non-repetitive sequences unlike other spidroins. The repeat unit (~200 amino acids) is much longer than MA silk repeats (~30 aa).

Amino Acid Composition

Gly: 20%, Ala: 15%, Ser: 18%, Pro: 3%, Val: 8%, Leu: 6%. More diverse amino acid profile than other silks; lower Gly content correlates with different secondary structure.

Gland Morphology

Numerous small, globular aciniform glands (~100-500 per spider). The most abundant gland type. Connected to all three pairs of spinnerets. Each produces a very fine fiber.

Mechanical Properties

E = 8 GPa, \(\sigma_f\) = 0.7 GPa, \(\varepsilon_f\) = 80%, U = 250 MJ/m³. Fiber diameter: 0.5-2 μm. The toughest silk of all seven types.

Analogy: Like ultra-tough shrink wrap — conforms to prey, absorbs struggling energy, and is nearly impossible to tear. The spider's strait jacket.

6. Aggregate (Ag) Silk — Glue Droplets

Spidroin Gene

AgSF (Aggregate Silk Factor, ~350 kDa). Not a typical spidroin — a glycoprotein with extensive O-linked glycosylation. Contains mucin-like domains that are inherently hygroscopic.

Composition

Not a typical amino acid fiber. Contains glycoproteins + low-molecular-weight organic salts: GABamide (isethionic acid derivative), KNO₃, K₂HPO₄. These hygroscopic compounds absorb atmospheric water to maintain glue viscosity.

Gland Morphology

Multilobed aggregate glands connected to posterior spinnerets. Each lobe produces a slightly different glue component. The glue is applied as a coating on flagelliform core fibers, then beads up into droplets via Rayleigh instability.

Mechanical Properties

Not a structural fiber. Adhesive energy: ~0.05 J/m². Individual droplets: 20-50 μm diameter, spaced 1-2 mm apart on capture spiral. Viscoelastic: viscosity ~10 Pa·s, elasticity ~100 Pa.

Analogy: Like microscopic drops of honey on a clothesline — sticky enough to trap a fly, and self-hydrating to stay effective in changing weather.

7. Pyriform (Py) Silk — Attachment Discs

Spidroin Gene

PySp1 (~330 kDa). Contains highly repetitive sequences rich in proline and serine. Unique among spidroins in forming a two-component system: fibers embedded in a cement matrix.

Amino Acid Composition

Gly: 18%, Ala: 22%, Ser: 15%, Pro: 12%, Gln: 8%. The high proline content contributes to the cement-like adhesive matrix component rather than structural flexibility.

Gland Morphology

Pear-shaped pyriform glands, numerous (~50-100 per spider). Connected to anterior spinnerets. Multiple spigots produce many fine fibers simultaneously that fuse into an attachment disc on contact with a surface.

Mechanical Properties

E = 15 GPa (disc composite), \(\sigma_f\) = 0.8 GPa, \(\varepsilon_f\) = 10%, U = 20 MJ/m³. Disc diameter: 100-500 μm. Highest stiffness and adhesion of all silk types.

Analogy: Like a biological superglue — the spider's anchor bolts. Each disc can support the entire spider's weight many times over, adhering to wood, glass, metal, or leaf surfaces.

Comprehensive Silk Type Comparison

Property	MA	MiA	Flag	Tu	Ac	Ag	Py
Gene	MaSp1/2	MiSp1	Flag	TuSp1	AcSp1	AgSF	PySp1
MW (kDa)	250-300	250	360	320	300	350	330
E (GPa)	10	12	0.003	14	8	~0	15
σ (GPa)	1.1	0.6	0.5	0.4	0.7	N/A	0.8
ε (%)	30	5	270	20	80	~1000	10
U (MJ/m³)	160	30	150	40	250	N/A	20
Diameter (μm)	1-8	1-4	1-5	5-15	0.5-2	20-50*	disc
Gly (%)	42	45	44	15	20	var	18
Ala (%)	25	28	8	28	15	var	22
Pro (%)	2-10	<1	18	<1	3	var	12
Ser (%)	4	6	3	25	18	var	15

* Aggregate "diameter" refers to glue droplet size, not fiber diameter.

9. Silk Production Rate & Energetics

Silk production is metabolically expensive. A spider must balance silk investment against energy return from prey capture — an optimization problem that evolution has been tuning for hundreds of millions of years.

Production Rates

Dragline (walking/rappelling): 1-2 cm/s, fiber diameter ~3 μm
Forced reeling (lab): up to 10 cm/s, produces stiffer silk due to higher shear alignment
Ballooning (dispersal): up to 100 cm/s, very fine fibers (~0.5 μm) for maximum drag-to-weight ratio
Web construction: An orb weaver completes a web in ~30 minutes, using ~0.5 mg of silk protein distributed across ~20 m of total thread length

Metabolic Cost of Silk Production

The metabolic cost of silk production can be derived from the thermodynamics of peptide bond formation. Each peptide bond requires hydrolysis of ATP and GTP during ribosomal translation:

\(\Delta G_{\text{peptide}} \approx +8\text{--}16 \;\text{kJ/mol (thermodynamic cost per bond)}\)

Including amino acid biosynthesis + tRNA charging + translation: ~4 ATP equivalents per peptide bond

For a typical spidroin of ~3000 amino acids (MW ≈ 300 kDa), we derive the total metabolic cost:

\(E_{\text{spidroin}} = N_{\text{bonds}} \times \Delta G_{\text{peptide}} \times \frac{1}{\eta_{\text{metabolic}}}\)

\(= 2999 \times 4 \;\text{ATP} \times 30.5 \;\text{kJ/mol ATP} \times \frac{1}{0.40}\)

\(\approx 915 \;\text{kJ/mol spidroin} \approx 3.05 \;\text{kJ per mg silk}\)

At ~50% efficiency of converting metabolic energy to usable silk: effective cost ≈ 1.5 J/mg. This matches calorimetric measurements (Prestwich, 1977).

Daily Silk Budget

An orb weaver typically uses ~0.5 mg per web. At 1.5 J/mg, each web costs ~0.75 J in silk production energy. For a spider with a resting metabolic rate of ~10 mW, this represents ~75 seconds of metabolic output — seemingly cheap, but prey capture is uncertain.

Silk Recycling: 90% Amino Acid Recovery

Most orb weavers eat their old web before building a new one, recovering ~90% of the amino acids through proteolytic digestion. This dramatically reduces the net cost:

\(E_{\text{net}} = \frac{E_{\text{silk}} - R \cdot E_{\text{silk}}}{\bar{n}_{\text{prey}}} = \frac{E_{\text{silk}}(1 - R)}{\bar{n}_{\text{prey}}}\)

where R = 0.90 (recycling fraction), \(\bar{n}_{\text{prey}}\) = average number of prey per web. Net cost per web: 0.75 J × (1 - 0.90) = 0.075 J per web — only 10% of the non-recycled cost. For a spider catching ~2 prey/day at ~50 J each, the silk investment yields a ≈1300:1 energy return on investment.

10. Water Extraction & Fiber Formation: The Spinning Duct

The spinning duct is where the magic happens — liquid protein solution transforms into solid silk fiber through a precisely orchestrated sequence of physical and chemical changes. This process has no industrial equivalent; no synthetic process achieves the same fiber properties from aqueous solution at ambient conditions.

Liquid Crystal Spinning Process

Stage 1: Storage (Gland Lumen)

Spinning dope: 30-50% w/v protein in aqueous solution. pH 6.3. High ionic strength (150 mM NaCl). Spidroins exist as micellar structures — the hydrophobic poly-Ala blocks are sequestered in the interior, preventing premature \(\beta\)-sheet aggregation. The N-terminal domain dimerizes in a pH-dependent manner, keeping chains soluble.

Stage 2: The S-Shaped Duct

The duct is S-shaped, ~20 mm long, tapering from ~500 μm diameter at the gland to ~10 μm at the spigot exit. This geometry creates both elongational flow (from tapering) and shear flow (from the S-curves). The duct epithelium is lined with ion-pumping cells.

Stage 3: Ion Exchange

Na⁺/Cl⁻ pumped OUT of the dope; K⁺/PO₄³⁻ pumped IN (Vollrath & Knight, 2001). Phosphate is a kosmotropic (water-structure-making) ion that promotes hydrophobic collapse of the poly-Ala blocks, nucleating \(\beta\)-sheet crystallites. K⁺ stabilizes the emerging fiber structure.

Stage 4: pH Drop

pH drops from 6.3 → 5.7 along the duct. This triggers a conformational switch in the N-terminal domain: the antiparallel dimer interface changes, releasing the chains for intermolecular \(\beta\)-sheet assembly. The C-terminal domain unfolds, exposing its poly-Ala blocks. This pH drop is the key trigger for fiber solidification.

Stage 5: Shear Alignment

Elongational flow from duct tapering orients the rod-like spidroins along the fiber axis. Above a critical Weissenberg number (Wi ≥ 1), the chains cannot relax fast enough and become permanently aligned. This creates the anisotropic structure essential for high tensile strength.

Stage 6: Water Extraction

~80% of water is removed through the duct epithelium via aquaporin channels. Protein concentration rises from ~30% to ~80% w/v. The remaining water is trapped in the amorphous phase and contributes to the plasticizing effect that gives silk its extensibility.

Derivation: Shear Rate in Tapering Duct

For Poiseuille flow through a tapered cylinder, the wall shear rate at position z along the duct is:

\(\dot{\gamma}(z) = \frac{4Q}{\pi r(z)^3}\)

where Q is the volumetric flow rate (~1 nL/s for dragline) and r(z) is the local duct radius. As the duct tapers from 250 μm to 5 μm, the shear rate increases by a factor of (250/5)³ = 125,000. At the exit: \(\dot{\gamma} \sim 10^5 \;\text{s}^{-1}\).

Derivation: Weissenberg Number

The Weissenberg number compares the polymer relaxation time to the flow timescale. When Wi ≥ 1, chains cannot relax and become aligned:

\(\text{Wi} = \lambda \cdot \dot{\gamma} = \frac{\eta_0 M_w}{c R T} \cdot \frac{4Q}{\pi r^3}\)

where \(\lambda\) is the spidroin relaxation time (~0.01-0.1 s for a 300 kDa protein in concentrated solution). At the proximal duct (r = 250 μm), Wi ≪ 1 (liquid-like). At the distal duct (r = 5 μm), Wi ≫ 1 (solid-like, aligned). The transition occurs at the critical radius where Wi = 1.

Spider Control of Fiber Properties

The spider actively controls fiber properties by varying three parameters:

Pulling speed: Faster pulling → higher shear rate → more \(\beta\)-sheet crystallization → stiffer, stronger but less extensible silk. Forcibly silked fibers (10 cm/s) are ~2x stiffer than naturally spun fibers (1 cm/s).
Spinneret valve opening: The muscular valve at the spinneret tip controls fiber diameter. Narrower opening → thinner fiber → higher surface-to-volume ratio → faster water loss → different crystallinity.
Humidity during spinning: Higher ambient humidity slows water extraction → more amorphous phase → more extensible silk. Spiders adjust silk properties seasonally and even web-to-web based on conditions.

11. Silk vs Engineering Materials: A Quantitative Comparison

To fairly compare silk with engineering materials, we use specific properties — normalized by density — because a lighter material that matches a heavier one's absolute strength is actually superior for most applications.

Specific Strength and Specific Stiffness

\(\text{Specific Strength} = \frac{\sigma_f}{\rho}, \qquad \text{Specific Stiffness} = \frac{E}{\rho}\)

MA silk: \(\sigma_f/\rho = 1.1/1.3 = 0.85\) GPa/(g/cm³). Steel: \(\sigma_f/\rho = 0.5/7.8 = 0.064\) GPa/(g/cm³). Silk is 13x stronger than steel per unit weight.

Why Silk Occupies a Unique Material Space

On an Ashby plot of specific strength vs specific toughness, silk occupies a region that no synthetic material can match. The key insight is silk's unique combination of:

High toughness (from amorphous chain unfolding — rubber-like energy absorption)
Moderate stiffness (from \(\beta\)-sheet crystallites — ceramic-like resistance to deformation)
Low density (organic polymer — 6x lighter than steel)

Kevlar is stronger but brittle. Rubber is extensible but weak. Steel is stiff but heavy and low-toughness. Only silk combines all three properties because its two-phase nanocomposite architecture has no synthetic equivalent.

Biomimetic Silk: The State of the Art

Several companies are attempting to produce recombinant spider silk proteins:

Bolt Threads: Yeast-produced MaSp-based proteins (Microsilk™). Achieved ~30% of natural silk toughness. Used in consumer products (Stella McCartney dress, 2017).
Spiber Inc.: Brewed Protein™ from engineered microorganisms. Partnered with The North Face for Moon Parka. Focus on tunable material properties.
AMSilk: Transgenic E. coli-produced recombinant spidroins. First company to receive medical device approval for spider silk-based products (surgical mesh).

Why artificial silk still falls short: The gap between natural and recombinant silk arises from three factors: (1) incorrect protein folding — recombinant spidroins lack the chaperone-assisted folding pathway of the silk gland; (2) no liquid crystal alignment — industrial spinning cannot replicate the S-shaped duct's ion exchange + pH gradient + shear alignment; (3) wrong crystallite size distribution — natural silk has a specific distribution of \(\beta\)-sheet crystallite sizes (2-6 nm) that is critical for toughness; recombinant silk tends to form either too-large or too-small crystallites.

12. Silk as a Biomaterial: Medical & Industrial Applications

Silk's biocompatibility, tunable degradation rate, and exceptional mechanical properties make it one of the most versatile biomaterials. Note: most current medical silk products use Bombyx mori (silkworm) silk, not spider silk, due to scalability.

Sutures

FDA-approved silk sutures since the 1900s (Ethicon PERMA-HAND). These use degummed B. mori silk (fibroin), which is braided and coated with wax/silicone. Silk sutures degrade in ~1-2 years in vivo. Spider silk sutures would offer superior fatigue resistance but are not commercially viable due to supply constraints.

Nerve Conduits

Tubular silk scaffolds (electrospun or cast) guide axon regrowth across nerve gaps of 10-30 mm. The silk tube provides mechanical support, prevents fibrous ingrowth, and degrades at a rate matching nerve regeneration (~1 mm/day). Studies show 80% functional recovery in rat sciatic nerve models (Allmeling et al., 2008).

Drug Delivery

Silk microspheres (1-100 μm diameter) for controlled drug release. The \(\beta\)-sheet crystallinity controls degradation rate: higher crystallinity = slower release. Can deliver antibiotics, growth factors, or chemotherapy agents. Release kinetics tunable from hours to months by varying processing conditions.

Bulletproof Vests

Theoretical: a spider silk vest would be ~3x lighter than Kevlar for the same ballistic protection. This follows from silk's higher specific toughness (123 vs 35 MJ/m³ per g/cm³). However, producing enough spider silk for a vest (~1 kg) would require milking ~1 million spiders.

Air Filters

Electrospun silk nanofibers (~100-500 nm diameter) capture PM2.5 particles with 98.8% efficiency at a pressure drop ~50% lower than commercial HEPA filters. The silk fibers' electrostatic charge and high surface area provide both mechanical and electrostatic filtration. Biodegradable after use.

Optical Fibers & Photonics

Silk waveguides transmit light with losses of ~1-2 dB/cm at visible wavelengths. Biocompatible photonic sensors can be implanted to monitor tissue oxygenation, pH, or glucose in vivo. Silk's optical transparency and tunable refractive index (n ≈ 1.54) make it ideal for bio-integrated photonics.

Derivation: Ballistic Limit Velocity V₅₀

The ballistic limit velocity V₅₀ is the projectile speed at which 50% of impacts are defeated by the armor. For a fiber-based armor of areal density \(\rho_A\):

\(V_{50} = \sqrt{\frac{2 \cdot U \cdot t}{\rho_p}} = \sqrt{\frac{2 \cdot U \cdot \rho_A}{\rho_p \cdot \rho_f}}\)

where U = volumetric toughness of the fiber, t = armor thickness, \(\rho_p\) = projectile areal density (mass/cross-section), \(\rho_f\) = fiber density. For equal areal density armor: \(V_{50,silk}/V_{50,Kevlar} = \sqrt{U_{silk}/U_{Kevlar}} = \sqrt{160/50} \approx 1.79\). Spider silk armor would stop 79% faster projectiles than Kevlar of the same weight.

13. Cribellate vs Ecribellate Silk: Two Solutions to Prey Capture

Spiders have evolved two fundamentally different strategies for making capture threads sticky. This represents one of the major evolutionary transitions in spider biology, with profound consequences for web ecology and species diversification.

Cribellate Silk (Ancient Strategy)

Produced by the cribellum, a sieve-like spinning plate bearing thousands of tiny spigots. The spider combs the cribellate silk with a specialized leg structure (the calamistrum) to produce woolly, dry capture threads.

Thousands of nanometer-scale fibrils (~20-50 nm diameter)
Adhesion via van der Waals forces — no glue required
Works in any humidity (dry adhesion)
More energy-expensive to produce (combing is metabolically costly)
Cannot be recycled as efficiently as glue silk
Found in ancient spider families (Deinopidae, Uloboridae, Filistatidae)

Ecribellate Silk (Modern Strategy)

Uses aggregate glue droplets coating a flagelliform core fiber. The glue is an aqueous glycoprotein solution containing hygroscopic salts that absorb atmospheric water to maintain stickiness.

Glue droplets 20-50 μm diameter, spaced 1-2 mm apart
Adhesion via viscous + capillary forces (JKR contact mechanics)
Performance depends on humidity (too dry = glue hardens; too wet = glue dissolves)
Cheaper to produce per unit length (~50% less energy)
Fully recyclable — spider eats old web, recovers 90% of amino acids + salts
Found in derived families (Araneidae, Tetragnathidae, Theridiidae)

Derivation: Adhesion Force Comparison

The two adhesion mechanisms can be compared quantitatively:

\(F_{\text{crib}} = n \cdot F_{\text{vdW}} = n \cdot \frac{A \cdot R}{6D^2}\)

Cribellate: n = number of contacting nanofibrils (~5000), A = Hamaker constant (~6.5×10⁻²⁰ J), R = fibril radius (~20 nm), D = contact distance (~0.3 nm). Total: F ≈ 40 μN per capture thread.

\(F_{\text{ecrib}} = \frac{3}{2}\pi \gamma R_d \cdot \left(1 + \frac{\eta \dot{\varepsilon}}{\gamma/R_d}\right)\)

Ecribellate (JKR + viscous): \(\gamma\) = surface energy of glue (~0.05 J/m²), R₄ = droplet radius (~25 μm), \(\eta\) = glue viscosity (~10 Pa·s). The viscous term increases with pull-off speed, providing rate-dependent adhesion — faster prey escape attempts encounter stronger adhesion.

Why Ecribellate Spiders Dominate

The transition from cribellate to ecribellate silk occurred ~200 Mya and enabled an explosive radiation of spider diversity. Ecribellate spiders now represent >90% of araneomorph species. Two factors drive this dominance: (1) lower production cost — glue silk requires ~50% less metabolic energy per unit length because no combing is needed and the flagelliform core fiber is simpler than cribellate meshwork; (2) full recyclability — the aqueous glue dissolves completely during web ingestion, recovering both the protein and the hygroscopic salts, whereas cribellate nanofibril meshwork is harder to digest completely. The net energy savings compound daily: an ecribellate spider investing the same metabolic budget in web construction can build a ~40% larger web, capturing proportionally more prey.

Simulation: Advanced Silk Biophysics

Four-panel simulation: (1) Ashby plot showing all silk types and engineering materials in specific strength vs specific toughness space — note how silk occupies a unique region. (2) Spinning duct physics: shear rate, Weissenberg number, and protein concentration profiles along the S-shaped duct. (3) Silk production energetics: metabolic cost per web vs expected prey energy return, showing the optimal silk investment and the effect of 90% recycling. (4) Cribellate vs ecribellate adhesion force dynamics — cribellate (van der Waals) builds slowly as nanofibrils conform; ecribellate (glue) establishes contact rapidly but depends on viscous dissipation.

Python

script.py225 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Comprehensive 4-panel simulation
# 1. Ashby plot: silk types + engineering materials (specific strength vs specific toughness)
# 2. Spinning duct physics: shear rate, Weissenberg number, concentration along duct
# 3. Silk production energetics: cost per web vs prey capture probability
# 4. Cribellate vs ecribellate adhesion force vs contact time

fig, axes = plt.subplots(2, 2, figsize=(16, 14))
fig.patch.set_facecolor('#030712')
fig.suptitle("Spider Silk: Advanced Biophysics Simulations", fontsize=16, color='white', fontweight='bold', y=0.98)

# ========== Panel 1: Ashby Plot ==========
ax1 = axes[0, 0]
ax1.set_facecolor('#1a0a2e')

# Material data: [name, strength_GPa, toughness_MJ/m3, density_g/cm3, color, marker]
materials = [
    ("MA Silk", 1.1, 160, 1.3, '#c084fc', 'o'),
    ("MiA Silk", 0.6, 30, 1.3, '#a78bfa', 's'),
    ("Flagelliform", 0.5, 150, 1.3, '#22d3ee', '^'),
    ("Tubuliform", 0.4, 40, 1.3, '#f472b6', 'D'),
    ("Aciniform", 0.7, 250, 1.3, '#34d399', 'v'),
    ("Pyriform", 0.8, 20, 1.3, '#f87171', 'P'),
    ("Steel", 0.5, 6, 7.8, '#94a3b8', 'x'),
    ("Kevlar 49", 3.6, 50, 1.44, '#fbbf24', '+'),
    ("Carbon Fiber", 3.5, 25, 1.75, '#fb923c', '1'),
    ("Nylon 6,6", 0.95, 80, 1.14, '#e879f9', '2'),
    ("Rubber", 0.05, 100, 0.92, '#f87171', '3'),
    ("Bone", 0.13, 4, 2.0, '#d1d5db', '4'),
    ("Tendon", 0.08, 7.5, 1.12, '#86efac', 'h'),
    ("Muscle", 0.3, 0.07, 1.06, '#fca5a5', '*'),
]

for name, sigma, U, rho, color, marker in materials:
    sp_strength = sigma / rho  # GPa/(g/cm3) = specific strength
    sp_toughness = U / rho     # MJ/m3 / (g/cm3) = specific toughness
    ax1.scatter(sp_toughness, sp_strength, c=color, s=120, marker=marker, zorder=5, edgecolors='white', linewidth=0.5)
    offset_x = 3
    offset_y = 0.02
    if name == "Kevlar 49":
        offset_y = 0.1
    elif name == "Steel":
        offset_x = -15
    ax1.annotate(name, (sp_toughness, sp_strength), fontsize=7, color=color,
                 xytext=(offset_x, offset_y), textcoords='offset points')

# Draw silk region
from matplotlib.patches import FancyBboxPatch
silk_patch = FancyBboxPatch((20, 0.25), 180, 0.75, boxstyle="round,pad=5",
                             facecolor='#c084fc', alpha=0.08, edgecolor='#c084fc', linewidth=1.5, linestyle='--')
ax1.add_patch(silk_patch)
ax1.text(60, 0.95, "SILK\nDOMAIN", fontsize=11, color='#c084fc', alpha=0.5, fontstyle='italic')

ax1.set_xscale('log')
ax1.set_yscale('log')
ax1.set_xlabel('Specific Toughness (MJ/m\u00b3 per g/cm\u00b3)', color='white', fontsize=10)
ax1.set_ylabel('Specific Strength (GPa per g/cm\u00b3)', color='white', fontsize=10)
ax1.set_title('Ashby Plot: Silk in Material Space', color='#d8b4fe', fontsize=12)
ax1.set_xlim(0.01, 500)
ax1.set_ylim(0.01, 5)
ax1.tick_params(colors='white')
ax1.grid(True, alpha=0.15, color='#a78bfa', which='both')
for sp in ax1.spines.values(): sp.set_color('#7c3aed')

# ========== Panel 2: Spinning Duct Physics ==========
ax2 = axes[0, 1]
ax2.set_facecolor('#1a0a2e')

# S-shaped duct: 20 mm long, tapering from 500 um to 10 um
z = np.linspace(0, 20e-3, 500)  # position along duct (m)
# Radius: S-shaped taper
r = 250e-6 - 240e-6 * (1 / (1 + np.exp(-500 * (z - 10e-3))))  # sigmoid taper
r = np.clip(r, 10e-6, 500e-6)

# Flow rate Q ~ 1 nL/s (typical)
Q = 1e-12  # m^3/s

# Shear rate at wall: gamma_dot = 4Q / (pi * r^3)
gamma_dot = 4 * Q / (np.pi * r**3)

# Relaxation time of spidroin (lambda ~ 0.01-0.1 s)
lam_relax = 0.05  # s
Wi = lam_relax * gamma_dot  # Weissenberg number

# Protein concentration (increases as water extracted)
c_protein = 30 + 50 * (z / z[-1])**1.5  # % w/v

ax2_twin = ax2.twinx()
ax2_twin2 = ax2.twinx()
ax2_twin2.spines['right'].set_position(('outward', 55))

p1, = ax2.plot(z * 1e3, gamma_dot, color='#c084fc', linewidth=2.5, label='Shear rate')
ax2.set_ylabel('Shear Rate (s\u207b\u00b9)', color='#c084fc', fontsize=10)
ax2.set_xlabel('Position Along Duct (mm)', color='white', fontsize=10)
ax2.tick_params(axis='y', colors='#c084fc')

p2, = ax2_twin.plot(z * 1e3, Wi, color='#22d3ee', linewidth=2.5, label='Wi')
ax2_twin.set_ylabel('Weissenberg Number', color='#22d3ee', fontsize=10)
ax2_twin.tick_params(axis='y', colors='#22d3ee')
ax2_twin.axhline(1.0, color='#fbbf24', linewidth=1.5, linestyle='--', alpha=0.7)
ax2_twin.text(12, 1.2, 'Wi = 1 (alignment onset)', fontsize=8, color='#fbbf24')

p3, = ax2_twin2.plot(z * 1e3, c_protein, color='#34d399', linewidth=2.5, label='[Protein]')
ax2_twin2.set_ylabel('[Protein] (% w/v)', color='#34d399', fontsize=10)
ax2_twin2.tick_params(axis='y', colors='#34d399')

ax2.set_title('Spinning Duct: Shear Rate,\nWi Number, Protein Conc.', color='#d8b4fe', fontsize=12)
ax2.tick_params(axis='x', colors='white')
ax2.grid(True, alpha=0.15, color='#a78bfa')
for sp in ax2.spines.values(): sp.set_color('#7c3aed')
for sp in ax2_twin.spines.values(): sp.set_color('#7c3aed')
for sp in ax2_twin2.spines.values(): sp.set_color('#7c3aed')
ax2.legend(handles=[p1, p2, p3], loc='upper left', fontsize=8, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

# ========== Panel 3: Silk Production Energetics ==========
ax3 = axes[1, 0]
ax3.set_facecolor('#1a0a2e')

# Silk mass per web (mg)
silk_mass = np.linspace(0.1, 2.0, 200)

# Energy cost of silk production: 1.5 J/mg
E_silk = 1.5  # J/mg
silk_cost = E_silk * silk_mass  # J per web

# Recycling recovery: 90% amino acid recovery
recycling_fraction = 0.90
net_cost = silk_cost * (1 - recycling_fraction)  # J per web after recycling

# Prey capture probability increases with web size (more silk = larger web)
# Saturating function
p_capture = 1 - np.exp(-2.5 * silk_mass)

# Expected energy return: average prey ~ 50 J, capture rate per web
E_prey = 50  # J per prey item
expected_return = E_prey * p_capture

# Net energy balance
net_energy = expected_return - silk_cost

ax3.plot(silk_mass, silk_cost, color='#f87171', linewidth=2.5, label='Silk cost (no recycling)')
ax3.plot(silk_mass, net_cost, color='#fbbf24', linewidth=2.5, label='Net cost (90% recycling)')
ax3.plot(silk_mass, expected_return, color='#34d399', linewidth=2.5, label='Expected prey energy')
ax3.fill_between(silk_mass, net_cost, expected_return, where=expected_return > net_cost,
                  alpha=0.15, color='#34d399', label='Profit zone')
ax3.fill_between(silk_mass, net_cost, expected_return, where=expected_return <= net_cost,
                  alpha=0.15, color='#f87171', label='Loss zone')

# Optimal silk investment
optimal_idx = np.argmax(expected_return - net_cost)
ax3.axvline(silk_mass[optimal_idx], color='#e879f9', linewidth=1.5, linestyle='--',
            label=f'Optimal: {silk_mass[optimal_idx]:.1f} mg')
ax3.scatter([silk_mass[optimal_idx]], [expected_return[optimal_idx]], c='#e879f9', s=100, zorder=5, edgecolors='white')

ax3.set_xlabel('Silk Investment per Web (mg)', color='white', fontsize=10)
ax3.set_ylabel('Energy (J)', color='white', fontsize=10)
ax3.set_title('Silk Production Energetics:\nCost vs Prey Capture', color='#d8b4fe', fontsize=12)
ax3.tick_params(colors='white')
ax3.grid(True, alpha=0.15, color='#a78bfa')
for sp in ax3.spines.values(): sp.set_color('#7c3aed')
ax3.legend(fontsize=8, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white', loc='center right')

# ========== Panel 4: Cribellate vs Ecribellate Adhesion ==========
ax4 = axes[1, 1]
ax4.set_facecolor('#1a0a2e')

# Contact time (seconds)
t_contact = np.linspace(0.001, 2.0, 500)

# Cribellate: van der Waals adhesion from nanofibril mesh
# F_crib = n * F_vdW per fibril; builds up as fibrils conform
n_fibrils = 5000  # number of cribellate nanofibrils in contact
A_Hamaker = 6.5e-20  # J (Hamaker constant for protein-chitin)
R_fibril = 20e-9  # m (fibril radius)
d_contact = 0.3e-9  # m (contact distance)
F_single_vdW = A_Hamaker * R_fibril / (6 * d_contact**2)  # N (Hamaker sphere-flat)

# Time-dependent conformation: fibrils gradually wrap around prey
conformity = 1 - np.exp(-3 * t_contact)
F_cribellate = n_fibrils * F_single_vdW * conformity * 1e6  # uN

# Ecribellate: viscous glue adhesion (JKR model + viscous dissipation)
gamma_glue = 0.05  # J/m^2 (surface energy of aggregate glue)
R_droplet = 25e-6  # m (glue droplet radius)
# JKR pull-off: F = 3/2 * pi * gamma * R
F_JKR = 1.5 * np.pi * gamma_glue * R_droplet * 1e6  # uN (constant once contact made)

# Viscous contribution increases with contact time (glue flows)
eta_glue = 10  # Pa.s (viscosity of aggregate glue)
viscous_factor = 1 + 2 * (1 - np.exp(-t_contact / 0.3))
F_ecribellate = F_JKR * viscous_factor  # uN

ax4.plot(t_contact, F_cribellate, color='#c084fc', linewidth=2.5, label='Cribellate (van der Waals)')
ax4.plot(t_contact, F_ecribellate, color='#22d3ee', linewidth=2.5, label='Ecribellate (glue)')
ax4.fill_between(t_contact, F_cribellate, F_ecribellate,
                  where=F_cribellate > F_ecribellate, alpha=0.1, color='#c084fc')
ax4.fill_between(t_contact, F_cribellate, F_ecribellate,
                  where=F_ecribellate > F_cribellate, alpha=0.1, color='#22d3ee')

# Crossover point
crossover_idx = np.argmin(np.abs(F_cribellate - F_ecribellate))
if crossover_idx > 0 and crossover_idx < len(t_contact) - 1:
    ax4.scatter([t_contact[crossover_idx]], [F_cribellate[crossover_idx]],
                c='#fbbf24', s=100, zorder=5, edgecolors='white')
    ax4.annotate(f'Crossover: {t_contact[crossover_idx]:.2f} s',
                 (t_contact[crossover_idx], F_cribellate[crossover_idx]),
                 fontsize=9, color='#fbbf24', xytext=(15, 15), textcoords='offset points',
                 arrowprops=dict(arrowstyle='->', color='#fbbf24'))

ax4.set_xlabel('Contact Time (s)', color='white', fontsize=10)
ax4.set_ylabel('Adhesion Force (\u03bcN)', color='white', fontsize=10)
ax4.set_title('Cribellate vs Ecribellate:\nAdhesion Force Dynamics', color='#d8b4fe', fontsize=12)
ax4.tick_params(colors='white')
ax4.grid(True, alpha=0.15, color='#a78bfa')
for sp in ax4.spines.values(): sp.set_color('#7c3aed')
ax4.legend(fontsize=10, facecolor='#1a0a2e', edgecolor='#7c3aed', labelcolor='white')

plt.tight_layout()
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#030712')

Click Run to execute the Python code

Code will be executed with Python 3 on the server

References

Gosline, J. M., Guerette, P. A., Ortlepp, C. S., & Savage, K. N. (1999). The mechanical design of spider silks. Journal of Experimental Biology, 202, 3295-3303.
Vollrath, F. & Knight, D. P. (2001). Liquid crystalline spinning of spider silk. Nature, 410, 541-548.
Blackledge, T. A. & Hayashi, C. Y. (2006). Silken toolkits: biomechanics of silk fibers spun by the orb web spider Argiope argentata. Journal of Experimental Biology, 209, 2452-2461.
Guinea, G. V., Elices, M., Perez-Rigueiro, J., & Plaza, G. R. (2005). Stretching of supercontracted fibers: a link between spinning and the variability of spider silk. Journal of Experimental Biology, 208, 25-30.
Termonia, Y. (1994). Molecular modeling of spider silk elasticity. Macromolecules, 27, 7378-7381.
Hayashi, C. Y., Shipley, N. H., & Lewis, R. V. (1999). Hypotheses that correlate the sequence, structure, and mechanical properties of spider silk proteins. International Journal of Biological Macromolecules, 24, 271-275.
Plaza, G. R., Guinea, G. V., Perez-Rigueiro, J., & Elices, M. (2006). Thermo-hygro-mechanical behavior of spider dragline silk. Polymer, 47, 2435-2446.
Rising, A. & Johansson, J. (2015). Toward spinning artificial spider silk. Nature Chemical Biology, 11, 309-315.
Prestwich, K. N. (1977). The energetics of web-building in spiders. Comparative Biochemistry and Physiology Part A, 57, 321-326.
Allmeling, C., Jokuszies, A., Reimers, K., Kall, S., & Vogt, P. M. (2008). Spider silk fibres in artificial nerve constructs promote peripheral nerve regeneration. Cell Proliferation, 41, 408-420.
Opell, B. D. & Bond, J. E. (2001). Changes in the mechanical properties of capture threads and the evolution of modern orb-weaving spiders. Evolutionary Ecology Research, 3, 567-581.
Blamires, S. J., Blackledge, T. A., & Tso, I. M. (2017). Physicochemical property variation in spider silk: ecology, evolution, and synthetic production. Annual Review of Entomology, 62, 443-460.
Heim, M., Keerl, D., & Scheibel, T. (2009). Spider silk: from soluble protein to extraordinary fiber. Angewandte Chemie International Edition, 48, 3584-3596.
Agnarsson, I., Kuntner, M., & Blackledge, T. A. (2010). Bioprospecting finds the toughest biological material: extraordinary silk from a giant riverine orb spider. PLoS ONE, 5, e11234.
Sahni, V., Blackledge, T. A., & Dhinojwala, A. (2010). Viscoelastic solids explain spider web stickiness. Nature Communications, 1, 19.

<- Physical Foundations Next: Module 2 -- Web Architecture ->