MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

How to write a 4-form of topological charge which would correspond to non-zero element of the homotopy group $\pi_4(S^3)$ or $\pi_4(S^2)$ (both are equal to $Z_2$) ? ?

An example of such a mapping (non-trivial), or even a homotopy of maps ($0\leq \tau\le 1$), is (or `localized' maps $R^4=H \to S^3$)

$f(h) = \frac{q-1}{q+1}$, where $q=h\ i\ \bar{h}+\tau\ i= -\bar{q}$ ; i.e. $\bar{f}=f^{-1}$ ;

here $h$ - quaternion, $q$ - imaginary quaternion, $f\in S^3$ - quaternion of unit length.

One can check that the framed 1-manifold $f^{-1}(-1)$ is non-trivial (see $\tau=1$).

Motivation: to describe topological charges (and quasi-charges) in the frame field theory (or Absolute Parallelism), see arXiv: gr-qc/0610076 .

This map is of great symmetry, in a sense like: $f(gh\bar{g})=gf\bar{g}=f(ghg)$, where $g=e^{i\alpha/2}$. Moreover, there is a more huge symmetry, a kind of gauge (gradient) symmetry, I quess:

$q(h)=q(h\ e^{i\alpha(h)})$ .

So, if one write quaternion $h$ as a pair of complex number (spinor), $h=a\ e^{i\phi} + j\ b\ e^{i\psi}$, so $q=i+(a^2-b^2)i -2abe^{i(\phi-\psi)}k$, and it is evident that the differential, combination $d\phi+d\psi$ in no way can arise in any differential form constructed from only $f$-field.

The cases $\pi_3(S^2)$ and $\pi_7(S^4)$ have the known solution relating to the Hopf invariant (if I am not mistaken) and having the form (pullback form)

$\alpha\wedge d\alpha$, where $d\alpha$ is the volume form on the sphere $S^2$ or $S^4$;

that is, for the first case, $d\alpha\propto{\rm Re}(\omega d\omega\wedge d\omega) $, where $\omega$ is a unit imaginary quaternion, the coordinates on the $S^2$; $\omega^2=-1$. (${\rm Tr}()$ suits (instead of ${\rm Re}()$ if the units are represented with the Pauli matrices, $i_p=i\sigma_p$.)

share|cite|improve this question
I'm not sure I understand this question. Where does the differential form live? Or are you asking for a map (à la Skyrme, say) $\mathbb{R}^5 \to S^3$, such that its restriction to the sphere at infinity gives the generator of $\pi_4(S^3)$? – José Figueroa-O'Farrill Nov 13 '10 at 14:43
I think Ivan Zhogin is after some differential form similar to the 3-form one can integrate on $S^3$ to get the Hopf invariant of a mapping. So he'd want a construction of differential forms on $S^4$ from maps from $S^4$ to $S^3$ and maps from $S^4$ to $S^2$, which when integrated yielded topological invariants (mod 2, say). Ivan, have you looked at Bott and Tu's book on Differential forms in Algebraic Topology? – j.c. Nov 13 '10 at 20:35
How is this ever going to work? Don't differential forms only detect real homotopy? – Daniel Pomerleano Nov 13 '10 at 22:15
@Daniel: actually rational homotopy (by Sullivan,...), but your objection stands. I don't know how to probe a torsion class via differential forms. – José Figueroa-O'Farrill Nov 13 '10 at 23:19
Try to look for Bott and Tu via Google. Homotopy groups of spheres are torsion outside of $\pi_n S^n$ and $\pi_{4n-1}S^{2n}$ so $\pi_4 S^3$ isn't detectable by standard differential form technology -- things like pull-backs, wedge products, integration, etc. – Ryan Budney Nov 15 '10 at 6:14

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.