Differential Geometry – How to Check if a 2-Form is Exact

differential-formsdifferential-geometry

Consider the 2-form
$$\sigma=\frac{x_1 dx_2 \wedge dx_3 + x_2dx_3\wedge dx_1+ x_3 dx_1 \wedge dx_2}{(x_1^2+x_2^2+x_3^2)^{3/2}}.$$
I need to show if it is exact or not. Suppose it is exact, then there exists a 1-form

$$\alpha=f_1dx_1+f_2dx_2+f_3dx_3,$$
such that
$$d\alpha=\displaystyle\sum_i df_i\wedge dx_i=\sigma.$$
Now I use $$df_i=\frac{\partial f_i}{\partial x_1}dx_1+\frac{\partial f_i}{\partial x_2}dx_2+\frac{\partial f_i}{\partial x_3}dx_3,$$
then $d\alpha$ becomes
$$\left(\frac{\partial f_2}{\partial x_1}-\frac{\partial f_1}{\partial x_2}\right)dx_1\wedge dx_2+\left(\frac{\partial f_3}{\partial x_2}-\frac{\partial f_2}{\partial x_3}\right)dx_2\wedge dx_3+\left(\frac{\partial f_1}{\partial x_3}-\frac{\partial f_3}{\partial x_1}\right)dx_3\wedge dx_1.$$
I do not know how to proceed, is this even the right method?

Best Answer

Generally it's easier to prove that a form is NOT exact.

For instance, an exact form is necessarily closed. So, if your form was exact it would be closed. And to verify this later property is just a matter of differentiation -which is easier than looking for an antiderivative, as you seem to be trying. Of course, if $d\sigma = 0$, this says nothing about being exact or not.

Another idea: if your form was exact, $\sigma = d\alpha$, for some $\alpha$, then, by Stokes' theorem, its integral over a closed surface $S$ (that is, with no boundary, or empty boundary, $\partial S = \emptyset$) would be zero:

$$ \int_S d\alpha = \int_{\partial S} \alpha = 0 \ . $$

So you could try to find a closed surface $S$ such that $\int_S \sigma \neq 0$.

Where to look for such a surface? Well, there is a "meta criterion" for this: since this is, presumably, an exercise in a beginner's differential geometry book, it can NOT be too far away from your knowledge. :-) So, I would try with the first closed surface that came to my mind without hesitation.

Or also, looking at that denominator, you could think of a surface where it becomes really, really, and I mean really, simple. (For instance, an sphere of radius 1.)