Differential Forms and the Exterior Derivative

1. Differential Forms and the Wedge Product

Let's begin by formally defining what a differential form is. As you'll see, it's a specific kind of tensor field you've already been working with, but singling it out provides access to a powerful new set of tools.

A differential p-form is simply a totally antisymmetric $(0,p)$ tensor field.

• A 0-form ($p=0$) is a scalar field $f$.
• A 1-form ($p=1$) is a covector field (a dual vector field) $\omega$.
• A 2-form ($p=2$), like the electromagnetic Faraday tensor $F_{\mu \nu }$, is a $(0,2)$ tensor field satisfying $F_{\mu \nu }=-F_{\nu \mu }$.

The following video from Frederic Schuller's lecture series provides a rigorous introduction to differential forms and the essential operation that combines them: the wedge product.

The collection of all p-forms on a manifold $M$ is denoted ${\Omega }^p(M)$. The wedge product endows the direct sum of these spaces, $\Omega (M)={⨁}_p{\Omega }^p(M)$, with the structure of a graded algebra, known as the Grassmann algebra or exterior algebra.

2. The Exterior Derivative Operator d

Now that we have our objects (differential forms) and a way to multiply them (the wedge product), we need a way to differentiate them. This is the role of the exterior derivative, d. It's an operator that takes a p-form and produces a (p+1)-form, written $d:{\Omega }^p(M)\to {\Omega }^{p+1}(M)$.

There are two common ways to define d: a coordinate-free definition and a coordinate-based one. Both are equivalent and offer different insights.

The Coordinate-Free Definition

The Schuller video you just watched continues with a beautiful, abstract definition of d that doesn't rely on any coordinate system. This is the most fundamental definition.

The Coordinate-Based Definition

While the coordinate-free definition is elegant, a coordinate-based formula is often more practical for calculations. Given a p-form $\omega$ with components ${\omega }_{{\mu }_1\dots {\mu }_p}$ in some coordinate basis, its exterior derivative $d\omega$ is a (p+1)-form with components:
$(d\omega {)}_{{\mu }_1\dots {\mu }_{p+1}}=(p+1){\partial }_{[{\mu }_1}{\omega }_{{\mu }_2\dots {\mu }_{p+1}]}$
where the square brackets denote total antisymmetrization over all enclosed indices.

Let's unpack this for the first few cases:

• 0-form (function $f$): $\omega =f$. $d\omega$ is a 1-form.
  $(df{)}_{\mu }={\partial }_{\mu }f$
  This is just the gradient. In basis form, $df=({\partial }_{\mu }f)d{x}^{\mu }$.

• 1-form ($\omega ={\omega }_{\nu }d{x}^{\nu }$): $d\omega$ is a 2-form.
  $(d\omega {)}_{\mu \nu }=2{\partial }_{[\mu }{\omega }_{\nu ]}={\partial }_{\mu }{\omega }_{\nu }-{\partial }_{\nu }{\omega }_{\mu }$
  You might recognize this form from the definition of the Faraday tensor from the vector potential, $F_{\mu \nu }={\partial }_{\mu }{A}_{\nu }-{\partial }_{\nu }{A}_{\mu }$.

This definition is presented clearly in the following reading.

3. The Fundamental Property: $d^2=0$

The single most important property of the exterior derivative is that applying it twice always yields zero.
$d(d\omega )=0\text{for any form }\omega$
This is often written compactly as $d^2=0$.

The proof in a coordinate basis is surprisingly simple and relies on the symmetry of second partial derivatives (Schwarz's theorem).

This simple identity has profound consequences, leading to a rich mathematical structure. It allows us to classify forms into two important categories.

Definition: Closed and Exact Forms

• A p-form $\omega$ is called closed if $d\omega =0$.
• A p-form $\omega$ is called exact if it is the exterior derivative of some (p-1)-form $\alpha$, i.e., $\omega =d\alpha$.

From $d^2=0$, it immediately follows that every exact form is closed.
If $\omega =d\alpha$, then $d\omega =d(d\alpha )=d^2\alpha =0$.

The much deeper question is the converse: is every closed form exact? The answer, remarkably, depends on the global topology of the manifold.

• On "simple" manifolds without holes (like ${\mathbb{R}}^n$), the answer is yes. This is Poincaré's Lemma.
• On manifolds with non-trivial topology (like a torus or a sphere with a point removed), there can be closed forms that are not exact.

The study of the "failure" of closed forms to be exact is called de Rham cohomology, a powerful tool that connects differential geometry to topology. Schuller's video gives an excellent introduction to this idea.

4. Application: Electromagnetism Revisited

The language of differential forms provides a breathtakingly elegant formulation of Maxwell's equations. Your background in the tensor formulation of electromagnetism will make this especially clear.

Let the vector potential be a 1-form $A=A_{\mu }d{x}^{\mu }$.
The Faraday tensor is a 2-form $F$ defined as the exterior derivative of $A$:
$F=dA$
In components, this is $F_{\mu \nu }={\partial }_{\mu }{A}_{\nu }-{\partial }_{\nu }{A}_{\mu }$, which is the definition you already know.

Now consider the exterior derivative of $F$:
$dF=d(dA)=d^{2}A=0$
This is an identity! The equation $dF=0$ automatically contains the two homogeneous Maxwell's equations (Gauss's law for magnetism, $\nabla \cdot \mathbf{B}=0$, and Faraday's law of induction, $\nabla \times \mathbf{E}+\frac{\partial \mathbf{B}}{\partial t}=0$). In this language, these laws are not independent equations of motion but are geometric identities that follow from the existence of the potential 1-form $A$.

The invariance of $F$ under the gauge transformation $A\to A^{\prime }=A+d\lambda$ (where $\lambda$ is a 0-form) is also immediately obvious:
$F^{\prime }=d(A^{\prime })=d(A+d\lambda )=dA+d^2\lambda =dA=F$
The remaining two (inhomogeneous) Maxwell's equations can be written as $d\star F=\star J$, where $\star$ is the Hodge star operator (which we will not cover in detail now, but it is a map from p-forms to (n-p)-forms that requires a metric) and $J$ is the current 1-form. The full set of Maxwell's equations in vacuum is just $dF=0$ and $d\star F=0$.

Conclusion

In this lesson, you have been introduced to the powerful language of differential forms. It provides not just a notational simplification, but a deeper understanding of the structures of calculus on manifolds.

Key Takeaways:

• Differential forms are totally antisymmetric covariant tensor fields.
• The wedge product ($\wedge$) combines forms while preserving antisymmetry, creating the exterior algebra.
• The exterior derivative ($d$) is a metric-independent operator that maps p-forms to (p+1)-forms.
• The most crucial property of d is $d^2=0$.
• This property implies that all exact forms ($\omega =d\alpha$) are automatically closed ($d\omega =0$). Whether the converse is true depends on the manifold's topology.
• This framework provides an exceptionally elegant formulation of physical laws, such as Maxwell's equations of electromagnetism.

We now have all the essential machinery in place to discuss the final piece of calculus on manifolds. In our next lesson, we will see how differential forms and the exterior derivative come together in the generalized Stokes' theorem, which provides a unified and powerful way to define integration on manifolds.