1

Let $X$ be a topological space, $R$ is a commutative, unital ring. In a proof from lecture there is claimed that $H^*(S^1\times X,\{\text{pt}\}\times X;R)\cong H^*(D^1\times X,\partial D^1\times X;R)$ in singular cohomology and I'm not sure how to justify it.

It is $\partial(D^1)=S^0=\{-1\}\coprod \{+1\}$ and maybe it's something like exicion and homotopy invariance, but I don't know how to apply excision here in detail. I would be happy if you explain me how to prove that $H^*(S^1\times X,\{\text{pt}\}\times X;R)\cong H^*(D^1\times X,\partial D^1\times X;R)$.

Najib Idrissi
  • 54,185
alg
  • 371
  • 1
  • 11

1 Answers1

0

The inclusion $* \times X \subset S^1 \times X$ is a cofibration (the singleton $*$ is a sub-CW-complex of $S^1$, hence $* \subset S^1$ is a cofibration, and by this question the product of a cofibration with an identity map is a cofibration), thus: $$H^*(S^1 \times X, * \times X) \cong \tilde{H}^*\bigl( (S^1 \times X) / (* \times X) \bigr)$$ (this is a standard fact, see e.g. Hatcher, Proposition 2.22 which deals with the case of homology; the cohomological case is identical).

Similarly you have: $$H^*(D^1 \times X, \partial D^1 \times X) \cong \tilde{H}^*\bigl( (D^1 \times X) / (\partial D^1 \times X) \bigr).$$

But now the two spaces $(S^1 \times X) / (* \times X)$ and $(D^1 \times X) / (\partial D^1 \times X)$ are homeomorphic: in the first one the boundary of $D^1$ is already collapsed, while in the second one you collapse at the same time the boundary of $D^1$ and the subspace corresponding to $* \times X$ (drawing a picture might be helpful here). Thus they have isomorphic cohomology, and you get the result.

Najib Idrissi
  • 54,185
  • thank you. This result from hatcher is new for me, but it seems to be very useful. Meanwhile I was able to prove $H^(S^1,{\text{pt}};R)\cong H^(D^1,\partial D^1;R)$ with the long exact sequence in cohomology for $(S^1,{\text{pt}})$ and with excision, but for products I only know the universal coefficient theorem in singular homology, such that I don't know how to prove it for products. Your proof is great, thanks! – alg Feb 06 '16 at 00:41