Cahn–Hoffman Field in Interfacial Energy

• Cahn–Hoffman field is a vector field that encodes anisotropic surface energy based on variational principles and a homogeneous extension of surface energy density.
• It is constructed from the gradient of an extended energy function, linking interfacial normals with anisotropic stress and curvature modifications in two-dimensional systems.
• The field's applications include modeling interface evolution, surface diffusion, and contact-line dynamics, crucial for predicting equilibrium and dynamic behaviors in anisotropic materials.

The Cahn–Hoffman field, also known as the Cahn–Hoffman $\boldsymbol{\xi}$-vector, is a geometrically and variationally motivated vector field that encodes the anisotropy of interfacial energy for material interfaces. The field emerges naturally from the first variation of total surface energy in systems with anisotropic surface tension, and is fundamental to the mathematical modeling of interface evolution, surface diffusion, and contact line migration in contexts such as solid-state dewetting with both weak and strong anisotropy (Jiang et al., 2018).

1. Mathematical Definition and Construction

Let $\gamma(n)$ denote the surface-energy density, assumed to be a $\mathcal{C}^2$ function on the unit circle $S^1$, where $n$ is the unit normal vector to the interface in $\mathbb{R}^2$. To extend the theory to general vectors, $\gamma$ is homogeneously extended to nonzero $p \in \mathbb{R}^2$ as

$$\hat{\gamma}(p) = |p|\, \gamma\!\left( \frac{p}{|p|} \right),$$

where the homogeneity relation $\hat{\gamma}(\lambda p) = \lambda \hat{\gamma}(p)$ holds for all $\lambda > 0$. The Cahn–Hoffman vector is defined as

$\xi(n) = \nabla_p \hat{\gamma}(p)\Big|_{p = n},$

satisfying the constraint

$\xi(n) \cdot n = \hat{\gamma}(n) = \gamma(n).$

In two dimensions, parameterizing $n = (-\sin\theta, \cos\theta)$ and $\tau = (\cos\theta, \sin\theta)$, the vector admits the explicit form

$\xi = \gamma(\theta)\, n - \gamma'(\theta)\, \tau,$

where $\gamma'(\theta)$ is the derivative of $\gamma$ with respect to angle $\theta$.

2. Variational Principle and Physical Emergence

For a smooth open curve $\Gamma$, consider a one-parameter perturbation family $\Gamma^\varepsilon := X(\cdot, \varepsilon)$ with variation vector field $$V(\rho, \varepsilon) = \partial_\varepsilon X(\rho, \varepsilon)$$. The total interfacial energy

$E(\Gamma) = \int_\Gamma \gamma(n(s))\, ds$

changes under variation as

$$\delta E(\Gamma; V) = \lim_{\varepsilon \to 0} \frac{E(\Gamma^\varepsilon) - E(\Gamma)}{\varepsilon} = -\int_\Gamma (\partial_s \xi)^{\perp} \cdot n\, (V_0 \cdot n)\, ds + [\xi^\perp \cdot V_0]_{s=0}^{s=L},$$

where “$^\perp$” denotes a $90^\circ$ clockwise rotation. The appearance of $\xi$ reflects its interpretation as the derivative of the surface energy with respect to interface normal, i.e., $\partial \gamma/\partial n$ in a vectorial sense. The boundary term involving $\xi$ gives rise to the dynamic contact-angle (Herring) condition.

3. Anisotropic Curvature and Interface Geometry

The anisotropic curvature $\kappa_\gamma$ generalizes the classical notion of curvature by incorporating the anisotropy encoded in $\gamma$. For a vector field $w(s)$ on $\Gamma$,

$\nabla_s \cdot w = \partial_s w \cdot \tau$

defines the surface divergence. Then,

$$\kappa_\gamma = -\nabla_s \cdot \xi = -\partial_s \xi \cdot \tau,$$

so that the spatial variation of $\xi$ along the interface (projected onto the tangent) furnishes the anisotropic modification of geometric curvature. In two dimensions, the identity $$(\partial_s\xi)^\perp \cdot n = \partial_s\xi \cdot \tau$$ connects the standard and rotated projections.

4. Chemical Potential and Thermodynamic Driving Force

In the context of interfacial thermodynamics, the Gibbs–Thomson relation defines the (dimensionless) chemical potential $\mu$ along the interface:

$$\mu = \frac{\delta E}{\delta \Gamma} = -(\partial_s\xi)^{\perp} \cdot n = -\partial_s \xi \cdot \tau = \kappa_\gamma.$$

Here, $s$ is the arclength parameter, $n$ the unit normal, $\tau$ the unit tangent, and $\partial_s\xi$ is the derivative of $\xi$ along the curve $\Gamma$. The chemical potential thereby reflects the local anisotropic curvature and provides the driving force for morphological evolution via surface diffusion.

5. Interface Evolution, Surface Diffusion, and Contact-Line Dynamics

Surface diffusion–driven motion is governed by the normal velocity

$v_n = \partial_t X \cdot n = \partial_{ss} \mu,$

where $\partial_{ss}\mu$ is the second arclength derivative of chemical potential. The interface evolution equation is

$$\partial_t X(s, t) = (\partial_{ss} \mu)\, n, \quad 0 < s < L(t).$$

At the contact points ($s=0, L$), mass conservation imposes a no-flux condition $\partial_s \mu = 0$. The dynamic Herring contact-angle condition at the contact line requires

$$\xi(n_c) \cdot n_c = \sigma = \gamma_{VS} - \gamma_{FS},$$

where $n_c$ is the substrate normal, and $\gamma_{VS}$, $\gamma_{FS}$ are vapor–solid and film–substrate energies, respectively. The contact-point velocity is controlled by

$$\frac{dx_c}{dt} = \pm \eta\left[\,\xi_2 - (\gamma_{VS} - \gamma_{FS})\right],$$

which relaxes the contact line until the equilibrium Herring condition $\xi_2 = \sigma$ is satisfied.

6. Explicit Forms for Common Anisotropy Functions

Several explicit $\gamma$ and corresponding $\xi$ forms are used to model physical interfaces:

Anisotropy Model	Surface Energy $\gamma$	$\xi(n)$ Expression
$k$-fold smooth crystal	$1 + \beta\cos(k\theta)$	$(1+\beta\cos k\theta)n + k\beta\sin k\theta\,\tau$
Riemannian-metric form	$\sum_{i=1}^K \sqrt{n^T G_i n}$	$\sum_{i=1}^K \frac{G_in}{\sqrt{n^T G_i n}}$
Cusped (e.g., truncated polygon)	$1 + \beta |\cos \tfrac{k\theta}{2}|$	$$[1+\beta f(\theta)]n - \beta\frac{\cos(\tfrac{k\theta}{2})(-\frac{k}{2}\sin(\tfrac{k\theta}{2}))}{\sqrt{\delta^2+\cos^2(\tfrac{k\theta}{2})}}\,\tau$$

Weak anisotropy prevails when $0 < \beta < 1/(k^2-1)$, while strong anisotropy arises for $\beta > 1/(k^2-1)$.

7. Physical Interpretation and Significance

The Cahn–Hoffman vector $\xi(n)$ serves as a “vectorial surface-tension,” whose tip-to-tail locus describes the Wulff shape—the equilibrium shape of a crystal under surface energy anisotropy. Variation in $\xi$ along the interface generalizes curvature and determines the local thermodynamic driving force for interface motion. The field thus forms a compact, geometrically-intrinsic encoding of interfacial anisotropy, and is essential in analytic and numerical treatments of interfacial evolution, especially where classical isotropic curvature fails to capture the underlying physics. The Herring boundary condition, expressible directly in terms of $\xi$, generalizes Young’s law to systems with anisotropic energies, governing equilibrium contact angles and dynamic contact-line evolution (Jiang et al., 2018).