Skip to content

Effective Mass

For derivations, response-specific masses, experimental extraction, and failure tests, see the canonical Effective Mass article.

For a band En(k)E_n(\mathbf k) expanded near k0\mathbf k_0,

En(k)≈En(k0)+ℏ22∑ij(m−1)ijΔkiΔkj,E_n(\mathbf k) \approx E_n(\mathbf k_0) + \frac{\hbar^2}{2} \sum_{ij} (m^{-1})_{ij} \Delta k_i\Delta k_j,

where Δk=k−k0\Delta\mathbf k=\mathbf k-\mathbf k_0 and

(m−1)ij=1ℏ2∂2En∂ki ∂kj∣k0.(m^{-1})_{ij} = \frac{1}{\hbar^2} \frac{\partial^2 E_n} {\partial k_i\,\partial k_j} \biggr\rvert_{\mathbf k_0}.

In one dimension,

m∗=ℏ2d2E/dk2.m^\ast = \frac{\hbar^2} {d^2E/dk^2}.

The band velocity used in semiclassical wave-packet dynamics is

vn(k)=1ℏ∇kEn(k).\mathbf v_n(\mathbf k) = \frac{1}{\hbar} \nabla_{\mathbf k}E_n(\mathbf k).
  • A smooth isolated band is being approximated locally in k\mathbf k.
  • The expansion point and coordinate axes are specified.
  • The effective-mass tensor describes curvature of a band dispersion, not the bare particle mass.
  • The simple parabolic approximation is local and may fail far from the expansion point.

Effective mass is most useful near band extrema, where the linear term in En(k)E_n(\mathbf k) vanishes. At a band minimum the effective mass tensor is positive in stable directions. At a band maximum, the curvature can be negative; this is one reason hole language is useful in semiconductors.

In anisotropic bands, the tensor form matters. Reducing it to one scalar is an additional approximation tied to a direction, symmetry, or averaged response.

  • Treating m∗m^\ast as a universal material constant independent of band, direction, density, and energy.
  • Forgetting that negative curvature gives negative electron effective mass in the electron-band description.
  • Using the parabolic approximation through a van Hove singularity or band crossing.
  • Confusing transport, cyclotron, optical, and density-of-states effective masses.

For a one-dimensional parabolic band E(k)=E0+ℏ2k2/(2m0)E(k)=E_0+\hbar^2k^2/(2m_0), what is m∗m^\ast?

Solution

The second derivative is d2E/dk2=ℏ2/m0d^2E/dk^2=\hbar^2/m_0. Therefore

m∗=ℏ2ℏ2/m0=m0.m^\ast = \frac{\hbar^2}{\hbar^2/m_0} = m_0.
  • N. W. Ashcroft and N. D. Mermin, Solid State Physics, Holt, Rinehart and Winston, 1976.
  • C. Kittel, Introduction to Solid State Physics, 8th ed., Wiley, 2004.
  • M. P. Marder, Condensed Matter Physics, 2nd ed., Wiley, 2010.