---
title: Kirchhoff–Love Material Model
url: https://www.emergentmind.com/topics/kirchhoff-love-material-model
type: topic
---

# Kirchhoff–Love Material Model

The Kirchhoff–Love material model is the thin-structure model obtained by combining Kirchhoff–Love kinematics with a constitutive relation between curvature and bending moments or shell energies. In its classical plate form, the unknown is the transverse deflection \(w=w(x,y,t)\) of a mid-surface \(\Omega\subset\mathbb{R}^2\), and the homogeneous isotropic dynamic equation reads \(D\Delta^2 w+\rho h\,\ddot w=f\), with flexural rigidity \(D=\frac{E h^3}{12(1-\nu^2)}\). In generalized forms, the scalar rigidity \(D\) is replaced by a fourth-order stiffness tensor, and the framework extends to anisotropic, orthotropic, heterogeneous, piezoelectric, thermoelastic, inelastic, beam, and shell settings while retaining the characteristic Kirchhoff–Love assumption of vanishing transverse shear deformation [2008.01693][2102.00747][2603.18164].

## 1. Kinematic foundation

Kirchhoff–Love theory is a reduction of three-dimensional elasticity to a lower-dimensional model under the assumptions of a thin plate or shell, small strains, and negligible transverse shear deformation. Its defining kinematic statement is that normals to the mid-surface remain straight and normal after deformation, so transverse shear strains are suppressed. In the plate setting, this leads to the standard displacement field
\[
u(x,y,z,t)=u_0(x,y,t)-z\,w_x(x,y,t),\qquad
v(x,y,z,t)=v_0(x,y,t)-z\,w_y(x,y,t),\qquad
w(x,y,z,t)=w(x,y,t),
\]
with \(z\in[-h/2,h/2]\) the thickness coordinate; in pure bending, the model effectively uses only the scalar field \(w\) [2008.01693]. The same hypothesis appears in orthotropic plate modeling, where the transverse displacement is the only mid-surface displacement and
\[
u_1(x,y,z)=-z\,u_x(x,y),\qquad
u_2(x,y,z)=-z\,u_y(x,y),\qquad
u_3(x,y,z)=u(x,y),
\]
implying \(e_{13}=e_{23}=0\) and \(\sigma_{13}=\sigma_{23}=0\) [2110.00421].

For shells, the geometry is written directly in terms of the midsurface and its normal. A generic point has position
\[
\mathbf r(\xi,\eta,\zeta)=\mathbf x(\xi,\eta)+\zeta\,\mathbf n(\xi,\eta),\qquad \zeta\in[-h/2,h/2],
\]
and the Green–Lagrange strain decomposes as
\[
\mathbf S=\mathbf A+\zeta\,\mathbf B,
\]
with \(\mathbf A\) the membrane strain tensor and \(\mathbf B\) the bending strain tensor; Kirchhoff–Love assumptions enforce the vanishing of shear and thickness-stretch components [2105.09288]. In geometrically exact beam language, the same idea appears as the shear-free constraint
\[
\mathbf g_2(s)\cdot \mathbf t(s)\equiv 0,\qquad \mathbf g_3(s)\cdot \mathbf t(s)\equiv 0,
\]
equivalently \(\mathbf g_1(s)\equiv \mathbf t(s)/\|\mathbf t(s)\|\), so the cross-section remains orthogonal to the centerline [1609.00119].

The curvature measures are generated by second derivatives of the deflection. For plates,
\[
\kappa_{xx}=-w_{xx},\qquad
\kappa_{yy}=-w_{yy},\qquad
\kappa_{xy}=-2w_{xy},
\]
although the normalization of \(\kappa_{xy}\) varies across conventions; the common point is that curvature is linear in second derivatives of \(w\) [2008.01693]. For shells, curvature is encoded through changes in the second fundamental form, and in fully nonlinear shell reductions it appears together with the first and third fundamental forms of the deformed midsurface [2603.18164].

## 2. Constitutive relations and energy forms

At the core of the Kirchhoff–Love material model is a linear relation between bending moments and curvatures. In the generalized anisotropic plate setting,
\[
M_{\alpha\beta}(x,y,t)=\sum_{\gamma,\delta}D_{\alpha\beta\gamma\delta}(x,y)\,\kappa_{\gamma\delta}(x,y,t),
\]
where \(D_{\alpha\beta\gamma\delta}(x,y)\) is a fourth-order bending stiffness tensor, possibly anisotropic and spatially varying [2008.01693]. A common PDE formulation writes the same law as
\[
M=\mathbb C(x):\kappa,\qquad \kappa_{ij}=-\partial_{ij}w,
\]
with tensor symmetries \(C_{ijkl}=C_{jikl}=C_{ijlk}=C_{klij}\) and strong ellipticity imposed in inverse and analysis settings [2102.00747].

For homogeneous isotropic plates, the constitutive law simplifies to
\[
M_{xx}=-D(\kappa_{xx}+\nu\kappa_{yy}),\qquad
M_{yy}=-D(\kappa_{yy}+\nu\kappa_{xx}),\qquad
M_{xy}=-D(1-\nu)\kappa_{xy},
\]
with \(D=\frac{E h^3}{12(1-\nu^2)}\) [2008.01693]. In matrix form, the isotropic bending stiffness reads
\[
\mathbf D
=
D
\begin{pmatrix}
1 & \nu & 0\\
\nu & 1 & 0\\
0 & 0 & (1-\nu)/2
\end{pmatrix},
\]
and yields the standard bending energy density based on the Hessian \(D^2u\) [2102.00747][1905.02030].

Orthotropic plates retain the same kinematics but relax isotropy. For a rectangular orthotropic plate with one-dimensional reinforcement, the bending energy becomes
\[
\mathbb E_B(u)=\frac{d^3\mathcal K}{24}\int_\Omega
\left[
\frac{E_1}{E_2}u_{xx}^2+u_{yy}^2+2\nu_{12}u_{xx}u_{yy}+2(1-\nu_{12})u_{xy}^2
\right]dx\,dy,
\]
or, equivalently,
\[
\mathbb E_B(u)=\frac{d^3\mathcal K}{24}\int_\Omega
\left[
\nu_{12}|\Delta u|^2+(1-\nu_{12})|D^2u|^2+\kappa\,u_{xx}^2
\right]dx\,dy,
\qquad
\kappa=\frac{E_1-E_2}{E_2},
\]
which isolates the anisotropy term in the longitudinal direction [2110.00421].

For shells derived from three-dimensional hyperelasticity, the constitutive structure can be genuinely nonlinear. Starting from the Ciarlet–Geymonat energy, the reduced shell energies depend on the first, second, and third fundamental forms of the deformed midsurface, and their coefficients depend on the Lamé coefficients, the thickness, and the mean and Gaussian curvatures of the reference configuration. The purely volumetric contribution leads to two-dimensional energies depending on the mean and Gaussian curvatures of both the undeformed and deformed midsurfaces [2603.18164]. This suggests that, in shell form, the Kirchhoff–Love material model is not restricted to a moment–curvature tensor law but can be an energy density on geometric invariants of the midsurface.

| Variant | Constitutive statement | Source |
|---|---|---|
| Isotropic plate | \(M_{xx},M_{yy},M_{xy}\) determined by \(D\) and \(\nu\) | [2008.01693] |
| General anisotropic plate | \(M_{\alpha\beta}=D_{\alpha\beta\gamma\delta}\kappa_{\gamma\delta}\) | [2102.00747] |
| Orthotropic plate | bending energy with \(E_1,E_2,\nu_{12},\mu_{12}\) | [2110.00421] |
| Nonlinear shell | energy depends on first, second, third fundamental forms and curvatures | [2603.18164] |

## 3. Governing equations, dynamics, and boundary data

In generalized plate form, the equilibrium equation is
\[
\partial_\alpha\partial_\beta M_{\alpha\beta}(x,y,t)+\rho(x,y)h(x,y)\,\ddot w(x,y,t)=f(x,y,t),
\]
with
\[
M_{\alpha\beta}=D_{\alpha\beta\gamma\delta}(x,y)\,\kappa_{\gamma\delta},\qquad
\kappa_{\gamma\delta}=-\partial_{\gamma\delta}w.
\]
For homogeneous isotropic plates this reduces to
\[
D\Delta^2 w(x,y,t)+\rho h\,\ddot w(x,y,t)=f(x,y,t)
\]
[2008.01693]. In the static inverse-problem literature the same model is written as
\[
\operatorname{div}\operatorname{div}\big(\mathbb C(x):\nabla^2w(x)\big)=q(x)\quad\text{in }\Omega,
\]
or, in components,
\[
\sum_{i,j,k,l=1}^2\partial_i\partial_j\big(C_{ijkl}(x)\,\partial_k\partial_lw(x)\big)=q(x)
\]
[2102.00747]. Mixed formulations replace the fourth-order scalar equation by a first-order system in the deflection and the moment tensor, for example
\[
\boldsymbol{\sigma}+\mathbb A\,\HESS u=0,\qquad
-\DIV\VDIV \boldsymbol{\sigma}=f,\qquad
u=\partial_n u=0,
\]
or, in DPG form, by equations for the deflection \(u\), its gradient, the bending moment \(M\), and suitable skeleton traces [2112.14497][1805.08864].

Boundary data are expressed in terms of the displacement, its normal derivative, the normal bending moment, and the effective Kirchhoff shear force. With \(n\) the unit outward normal and \(s\) a tangential direction,
\[
M_{nn}=n_\alpha n_\beta M_{\alpha\beta},\qquad
M_{ns}=n_\alpha s_\beta M_{\alpha\beta},\qquad
Q_n=-\frac{\partial M_{nn}}{\partial n}-\frac{\partial M_{ns}}{\partial s}.
\]
The standard boundary conditions are:
\[
w=0,\qquad \frac{\partial w}{\partial n}=0 \quad\text{(clamped)},
\]
\[
w=0,\qquad M_{nn}=0 \quad\text{(simply supported)},
\]
\[
M_{nn}=0,\qquad Q_n=0 \quad\text{(free)}
\]
[2008.01693]. In inverse boundary value problems, the natural Cauchy data are
\[
w|_{\partial\Omega},\quad \partial_n w|_{\partial\Omega}
\]
and the corresponding moment and shear data, which define a Dirichlet-to-Neumann map for the stiffness tensor [2102.00747].

Dynamic analysis proceeds by modal decomposition. For free vibrations, one sets \(f=0\) and seeks
\[
w(x,y,t)=\phi(x,y)e^{i\omega t},
\]
which yields the eigenvalue problem
\[
\partial_\alpha\partial_\beta\big(D_{\alpha\beta\gamma\delta}(x,y)\,\partial_{\gamma\delta}\phi(x,y)\big)
-\omega^2\rho(x,y)h(x,y)\phi(x,y)=0.
\]
This formulation underlies the computation of natural frequencies, resonance, and beats in generalized Kirchhoff–Love plates [2008.01693].

## 4. Generalizations of the material model

The classical isotropic plate is only one member of a broader Kirchhoff–Love family. Orthotropic deck models replace a single rigidity by direction-dependent stiffnesses and yield equations such as
\[
\frac{d^3\mathcal K}{12}\big(\Delta^2u+\kappa\,u_{xxxx}\big)=f
\]
for rectangular plates with different longitudinal and transverse rigidities [2110.00421]. In shell theory, geometrically linear Kirchhoff–Love models on curved midsurfaces use tangential differential calculus, with membrane strain
\[
\boldsymbol{\varepsilon}_{\Gamma,\mathrm{Memb}}(\boldsymbol{u})
=
\tfrac12\big[\nabla_\Gamma^{\mathrm{cov}}\boldsymbol{u}
+
(\nabla_\Gamma^{\mathrm{cov}}\boldsymbol{u})^T\big]
\]
and bending strain
\[
\boldsymbol{\varepsilon}_{\Gamma,\mathrm{Bend}}(\boldsymbol{u})
=
-
\sum_{i=1}^d
\nabla_\Gamma^{\mathrm{cov}}\big(\nabla_\Gamma u_i\big)\,n_i,
\]
with constitutive relations
\[
\tilde{\boldsymbol n}_\Gamma=t\,\boldsymbol C_\Gamma:\boldsymbol{\varepsilon}_{\Gamma,\mathrm{Memb}},\qquad
\boldsymbol m_\Gamma=\frac{t^3}{12}\,\boldsymbol C_\Gamma:\boldsymbol{\varepsilon}_{\Gamma,\mathrm{Bend}}
\]
for shells [2602.14566].

Piezoelectric Kirchhoff–Love shells preserve the same shear-free kinematics but augment the constitutive structure with piezoelectric and dielectric tensors. The three-dimensional electroelastic relations are reduced through the thickness by expanding the electric potential as
\[
\phi(\mathbf r(\xi,\eta,\zeta))
\approx
\phi^{(0)}(\mathbf x(\xi,\eta))
+\zeta\,\phi^{(1)}(\mathbf x(\xi,\eta))
+
\left[\zeta^2-\left(\frac h2\right)^2\right]\phi^{(2)}(\mathbf x(\xi,\eta)),
\]
leading to coupling of \(\phi^{(1)}\) with membrane strains and \(\phi^{(2)}\) with bending strains [2105.09288]. In peridynamic formulations, the same Kirchhoff–Love shell kinematics are combined with three-dimensional rate-form constitutive laws, including \(J_2\) plasticity, plane-stress enforcement by iterative adjustment of \(\hat D_{33}\), and bond-level damage models for brittle and ductile fracture [2107.13062].

Thermoelastic extensions couple the plate equation to heat conduction. A nonlinear thermoelastic Kirchhoff–Love plate is written, in reduced form, as
\[
w_{tt}-\gamma\Delta w_{tt}+a(-\Delta w)\Delta^2 w+\alpha\Delta\theta
=
f(-\Delta w,-\nabla\Delta w),
\]
\[
\beta\theta_t-\eta\Delta\theta+\sigma\theta-\alpha\Delta w_t=0,
\]
with simply supported mechanical boundary conditions \(w=\Delta w=0\) and \(\theta=0\) at the boundary [1612.00798]. Defect-based shell models impose the Kirchhoff–Love constraint \(\boldsymbol d=\boldsymbol n\), which suppresses director stretch and transverse shear; in that setting only in-surface dislocation densities remain as active incompatibility sources, and the nontrivial incompatibility equations reduce to Gauss- and Codazzi-type relations [1506.07641].

A plausible implication is that “Kirchhoff–Love material model” names a kinematic class rather than a single constitutive law: the same no-shear hypothesis supports isotropic, orthotropic, piezoelectric, thermoelastic, inelastic, and nonlinear shell models with substantially different constitutive content.

## 5. Analysis, identification, and numerical formulations

The material parameters entering Kirchhoff–Love models are accessible to analysis and inverse identification. For the two-dimensional Kirchhoff–Love plate equation, global recovery of bending stiffness, Poisson coefficient, and Lamé parameters from boundary Cauchy data is established in the inverse-problem setting [2102.00747]. In the presence of a rigid inclusion inside a thin isotropic plate, the inclusion can be identified from a boundary couple field together with measurements of the induced transversal displacement and its normal derivative; under suitable regularity assumptions the resulting stability estimate is of log type [1806.08700].

Transformation-based interpretations of the model are more restrictive than is sometimes assumed. For an anisotropic heterogeneous plate, the Kirchhoff–Love equation is form invariant for a class of transformations with a vanishing Hessian; in that case the transformed plate remains a pure Kirchhoff–Love medium, without the in-plane body forces and pre-stresses that appear in other approaches [1901.00067]. This addresses a recurrent misconception: exact form invariance is not stated for arbitrary transformations but for that specific class.

A large numerical literature reflects the fourth-order nature of the model. Finite-difference discretizations for generalized Kirchhoff–Love plates use second-order spatial differences together with explicit predictor-corrector or implicit Newmark–Beta time stepping; stability analysis yields stable time steps, and the reported schemes are stable and second-order accurate [2008.01693]. Ultraweak DPG formulations recast the model into first-order systems involving the deflection, its gradient, the bending moment, and skeleton traces, and then prove well-posedness and quasi-optimal convergence for lowest-order discrete schemes with approximated test functions [1805.08864]. For polygonal meshes, a \(C^1\) conforming virtual element method of arbitrary order \(k\ge2\) gives correct approximation of the buckling spectrum, optimal-order convergence for buckling modes, and double order for the buckling coefficients [1905.02030]. The fully discrete plates complex on polygonal meshes yields an exact Hilbert-complex-based discretization for mixed Kirchhoff–Love problems, together with full stability and convergence analysis [2112.14497].

Higher-order continuity requirements can also be handled geometrically or by mixed reformulation. Catmull–Clark subdivision bases provide the \(C^1\) continuity required by Kirchhoff–Love shell theory for isogeometric Galerkin vibration analysis of piezoelectric shells [2105.09288]. Mixed-hybrid Bulk Trace FEM alleviates the higher-order continuity requirements of displacement-based Kirchhoff–Love beam and shell formulations by allowing standard \(C^0\)-continuous Lagrange elements in the bulk domain while retaining higher-order accurate trace discretizations of the embedded structures [2602.14566]. Rotation-free peridynamics offers a meshfree alternative that naturally discretizes shell theories requiring higher-order smoothness on completely unstructured surface meshes [2107.13062].

## 6. Applicability, related theories, and limitations

Kirchhoff–Love models are intended for thin plates, shells, and slender beams in regimes where transverse shear deformation is negligible. Across the cited plate, shell, and beam formulations, the theory is consistently tied to small strains and either small deflections or geometrically exact but shear-free kinematics, depending on the setting [2008.01693][1609.00119][2602.14566]. Their principal advantage is dimensional reduction: the mechanics are represented on the midsurface or centerline, while bending is retained through second-derivative or curvature terms.

The principal limitation is the same assumption that gives the model its simplicity. For moderately thick plates or shells, transverse shear can be significant, and the Mindlin–Reissner or Simo–Reissner theories are more appropriate [2008.01693][2102.00747][1609.00119]. Linear Kirchhoff–Love plate models also exclude plasticity, damage, viscoelasticity, and other nonclassical effects unless these are explicitly added, as in peridynamic, thermoelastic, or hyperelastic shell generalizations [2107.13062][1612.00798][2603.18164]. In transformation problems, exact form invariance is confined to transformations with vanishing Hessian rather than generic mappings [1901.00067].

Another common misconception is that the model is intrinsically isotropic. The record here is the opposite: generalized Kirchhoff–Love formulations explicitly admit anisotropic, orthotropic, and spatially varying stiffness tensors, variable thickness and density, piezoelectric couplings, and shell energies derived from three-dimensional hyperelasticity [2008.01693][2110.00421][2105.09288][2603.18164]. What remains invariant across these variants is not a single material law but the structural hypothesis that normals remain straight and normal and that shear deformation is excluded. Under that reading, the Kirchhoff–Love material model is best understood as the canonical shear-free constitutive framework for bending-dominated thin structures.

Source: https://www.emergentmind.com/topics/kirchhoff-love-material-model