Proof and counter-example that a chain $c_{R, n} \ne \partial c$. Where is the error?

89 Views Asked by At

If $R > 0$ and $n \in \mathbb{Z}$ we can define the singular 1-cube $c_{R, n}\colon [0, 1] \rightarrow \mathbb{R}^2$ by

$$c_{R, n}(t) = (R\cos(2\pi n t), R\sin(2\pi n t))$$

We know that $c_{R, n} \ne \partial c$ for any 2-chain $c$ in $\mathbb{R}^2 - 0$. For assume $c_{R, n} = \partial c$ for some 2-chain $c$ in $\mathbb{R}^2$. Then

$$2\pi n = \int_{\partial c_{R, n}} d\theta = \int_{c_{R, n}}d(d\theta)) = 0$$

where $\theta$ is the 2nd coordinate function in the polar coordinate function (i.e. the one that gives us the angle of a point in $\mathbb{R}^2$) since $d\theta$ is closed.

I am confused, however, because if we define the 2-cube $c$ (in $\mathbb{R}^2 - 0$) by

$$c(s, t) = s\cdot c_{R, n}(t)$$

we get

$$\partial c(x) = -c(0,x) + c(1, x) + c(x, 0) - c(x, 1) = c_{R, n}(x)$$

since

$$c(x, 0) = x\cdot c_{R, n}(0) = x\cdot c_{R, n}(1) = c(x, 1)$$

I'm clearly doing something(s) wrong, and I'd like to understand what. Any help will be appreciated.

1

There are 1 best solutions below

0
On BEST ANSWER

That really isn't much of an answer but ... as you asked for it ... well, the mistake is that the image of the function $$c:[0,1]\times[0,1]\to\mathbb{R}^2$$ given by $c(s,t)=s\cdot c_{R,n}(t)$ is not contained in $\mathbb{R}^2\setminus\{0\}$, because $c(0,t)=(0,0)$ for any $t\in[0,1]$. So, your construction of $c$ simply shows that $c_{R,n}$ is exact (i.e. it is the boundary of some $2$-chain) in $\mathbb{R}^2$, but doesn't allow you to conclude anything about what happens in $\mathbb{R}^2\setminus\{0\}$.

By the way, your definition of $c$ is essentially the construction of the cone on $c_{R,n}$ by means of the linear structure of $\mathbb{R}^2$, with vertex exactly in $(0,0)$.