Disordered Sticker–Spacer Model

• The disordered sticker–spacer model is defined by polymer chains with random spacer lengths and specific binding stickers, crucially influencing network connectivity and thermodynamics.
• Mean-field theories and simulation methods like graph Monte Carlo and Brownian dynamics quantify bond formation, cooperativity effects, and percolation thresholds in these systems.
• Rheological behavior is governed by sticker bond lifetimes and chain relaxation times, driving transitions from evanescent gels to classical gel phases with universal scaling laws.

A disordered sticker–spacer model describes associative polymers composed of linear chains where specific binding sites (“stickers”) are interspersed with flexible, inert “spacer” segments of random length. In these systems, the spatial and statistical distribution of stickers along the backbone is not periodic but described by a quenched random process—spacer segment lengths $\ell_1, \ldots, \ell_{f+1}$ are drawn from a prescribed probability distribution $P(\ell)$. This structural disorder plays a pivotal role in determining the thermodynamic, percolative, and rheological behavior of the resulting physical network. The model underlies both mean-field statistical frameworks and simulation-based explorations of sol–gel transitions, network connectivity, and nontrivial dynamical regimes such as “evanescent” gels, with comprehensive treatments found in (Choi et al., 2020) and (Robe et al., 2023).

1. Model Architecture: Statistical Placement and Bond Variables

Each of $N$ polymer chains in a system volume $V$ carries $f$ stickers separated by $M$ spacers. Stickers serve as specific associative binding sites capable of reversible bonding, typically limited by a valence $\nu$ (the maximum number of bonds per sticker, generally $\nu=1$ or $2$). Disorder is introduced through the sequence of spacer lengths $\{\ell_i\}$ drawn independently from $P(\ell)$, e.g., Poisson or uniform distributions (Robe et al., 2023). This architecture yields random sticker positions along each chain—formally a quenched disorder—leading to local heterogeneity in potential crosslinking sites.

Relevant concentrations are $c = N/V$ (polymer chains per volume) and $c_s = f N / V$ (stickers per volume). Association between stickers $i$ and $j$ is characterized by an energy gain $\epsilon_{ij}$ and a “bond volume” $v_{ij}$, which encodes the entropic cost of crosslink spatial localization.

2. Mean-Field Theories of Bond Percolation and Free Energy

Within mean-field theory, spatial correlations are omitted and each sticker interacts uniformly with others. The probability of a bond between stickers of type $i$ and $j$ is

$p_{ij} = v_{ij} \exp(-\beta \epsilon_{ij}),$

where $\beta = 1/(k_B T)$. The mean number of bonds per $i$-sticker is

$$\langle n_b^{(i)} \rangle = \frac{1}{n_i} \sum_j n_j \lambda_{ij}, \quad \lambda_{ij} = v_{ij} e^{-\beta \epsilon_{ij}},$$

with $n_i$ the per-chain valence for sticker type $i$.

Percolation—the onset of a system-spanning network—is determined by the average number of bonds per sticker. The critical mean-field condition is

$$\langle n_b \rangle \equiv \sum_i \frac{n_i}{f} \langle n_b^{(i)} \rangle = 1.$$

This yields a closed-form critical polymer concentration for percolation:

$c_{perc} = \frac{1}{f \lambda}$

in the symmetric, single-sticker-type case.

The mean-field contribution to the free energy from sticker bonds is given by

$$F_{int} = -k_B T \sum_{i < j} p_{ij} \epsilon_{ij},$$

summed over all sticker types.

3. Cooperativity Effects in Bonding

Bond cooperativity modifies the energy gain for new bonds depending on the number of pre-existing bonds a sticker participates in. Introducing a cooperativity parameter $\Delta \epsilon$, the effective bonding energy when a sticker already has $n_{prior}$ bonds becomes

$$\epsilon_{eff}^{(i,j)} = \epsilon_{ij} + n_{prior} \Delta \epsilon.$$

The corresponding bond probability is

$$p_{ij}(n_{prior}) = v_{ij} \exp[-\beta(\epsilon_{ij} + n_{prior}\Delta\epsilon)].$$

In mean-field, substituting the average $\langle n_{prior} \rangle = \langle n_b \rangle$ yields an effective bond propensity $\lambda_{ij}^{eff}$, allowing recalculation of the percolation threshold. Positive cooperativity ($\Delta\epsilon < 0$) lowers the critical concentration for percolation, while negative cooperativity can suppress network formation entirely (Choi et al., 2020).

4. Simulation Approaches: Graph Monte Carlo and Brownian Dynamics

Two classes of simulation approaches have validated and extended mean-field models:

• Graph-based Monte Carlo simulations (Choi et al., 2020): Stickers are nodes and reversible bonds are edges. Random sticker–spacer chains are generated by drawing spacer lengths from $P(\ell)$, assigning stickers, and tracking dynamic bond formation and breakage with Metropolis criteria (including entropy differences between intra- and inter-cluster bonds). The largest-cluster fraction $S_{max}$ and bond fraction $\phi_b$ quantitate percolation behavior. Simulations confirm mean-field predictions in the noncooperative regime, with deviations arising for strong cooperativity and small clusters.
• Non-equilibrium Brownian dynamics (Robe et al., 2023): Polymer chains are modeled as bead–spring systems with a subset of beads as stickers and random sticker positions according to $P(\ell)$. Reversible binding (formation and breaking) is handled via explicit MC kinetics with acceptance probabilities set by interaction energies, and network topology is analyzed using graph algorithms (e.g., SPQR trees) to quantify elastically active chain density and network lifetime.

Both methods capture important features of disorder and dynamic association, with graph MC excelling at static percolation and Brownian dynamics integrating rheological observables.

5. Rheological Regimes and Dynamical Transitions

Rheological behavior is governed by the relationship between sticker bond lifetime and polymer relaxation time. The sticker dissociation time is

$\tau_b = \tau_0 \exp(\Delta E / k_B T),$

with $\Delta E$ the binding energy, while single-chain Rouse time is

$\tau_R \simeq (\zeta b^2 / k_B T) (N^2 / 3\pi^2).$

For $\tau_b \lesssim \tau_R$, a network may percolate instantaneously but possesses no steady-state elasticity—termed an “evanescent” gel. Only when $\tau_b \gg \tau_R$ does a classical gel phase emerge with a measurable elastic plateau $G_\epsilon$.

Critical rheological findings include:

• Zero-shear viscosity diverges near the sol–gel transition as $\eta_0 \sim \Delta^{-s}$ with $s \approx 0.9$.
• The elastic plateau modulus scales as $G_\epsilon \sim \Theta(c - c_g)[(c - c_g)/c_g]^z$ with $z \approx 2.6$.
• At criticality, the stress relaxation modulus $G(t)$ exhibits universal power-law decay: $G(t) \sim t^{-n}$, $n \approx 0.73$, with hyperscaling $n = z/(z + s)$ (Robe et al., 2023).

6. Impact of Architecturally Disordered Spacers

Quenched disorder in spacer length (compared to periodic assignment) imparts several statistical and physical consequences:

• The percolation threshold $c_{perc}$ is modestly increased in the presence of disorder.
• Disordered placement results in broader distributions of loop sizes and network mesh spacings, leading to a modest broadening of the power-law regime in $G(t)$ and in the storage/loss moduli $G', G''$ versus frequency.
• The onset of the elastic plateau in $G_\epsilon(c)$ is diminished for concentrations just above the threshold.
• Universal scaling exponents ($s$, $z$, $n$) and hyperscaling relations are preserved, consistent with percolation universality (Robe et al., 2023).

7. Summary of Key Parameters and Observables

The following table summarizes principal parameters:

Symbol	Definition	Notes
$N$	number of polymer chains	—
$V$	total system volume	$c=N/V$
$f$	stickers per chain (valence)	—
$\ell_i$	spacer length (random)	Drawn from $P(\ell)$
$P(\ell)$	spacer length distribution	Poisson, uniform, etc.
$\nu$	maximum bonds per sticker	$\nu=1$ or $2$
$\epsilon_{ij}$	sticker–sticker binding energy	May depend on types $i, j$
$\Delta\epsilon$	bond cooperativity parameter	Shift per existing bond
$v_{ij}, v_0$	bonding/bond reference volume	—
$\lambda_{ij}$	mean-field bond propensity ($v_{ij}e^{-\beta\epsilon_{ij}}$)	—
$c_{perc}$	percolation threshold (mean-field, simulation)	Closed form in symmetric case
$S_{max}$	largest-component fraction in MC simulations	—
$G_\epsilon$	elastic plateau modulus	$G_\epsilon \sim (c-c_g)^{z}$
$\eta_0$	zero-shear viscosity	Diverges at $c_g$ as $\eta_0 \sim \Delta^{-s}$
$\tau_b, \tau_R$	sticker dissociation, Rouse timescales	Control sol–gel crossover, evanescence

System behavior is thus tunable via sticker density (mean and distribution of $\ell$), binding energies $\epsilon_{ij}$, and cooperativity $\Delta\epsilon$. Spacer disorder weakly shifts percolation and rheology, but preserves universal scaling and phase boundaries.

References:

(Choi et al., 2020): Generalized models for bond percolation transitions of associative polymers (Robe et al., 2023): Evanescent Gels: Competition Between Sticker Dynamics and Single Chain Relaxation