Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I already know that the function $$ f(x) = \begin{cases} \exp(- \frac{1}{x^2}), \quad x > 0 \\ 0 , \quad x \leq 0 \end{cases} $$

is infinitely differentiable throughout $\mathbb R$. The only real problem, of course, lies in showing that $f^{(k)} (0) = 0$ for any positive integer $k$. What I have not been able to deduce is that

$$ \phi(x) = \begin{cases} \exp(- \frac{1}{1 - x^2}), \quad |x| < 1 \\ 0 , \quad |x| \geq 1 \end{cases} $$

is also infinitely differentiable throughout $\mathbb R$, using the previous function. The problem now is finding out what happens at $x = 1,-1$. Does the substitution $\zeta ^2 = 1 - x^2$ work, or is there another way to prove this?

Thank you all!

share|improve this question
add comment

2 Answers 2

up vote 5 down vote accepted

This is much easier if you first show that

$g(x) = \begin{cases} e^{-\frac{1}{x}} & \text{for } x \gt 0 \\\\ 0 & \text{for } x \leq 0 \end{cases}$

is smooth. Then it's easy to see that $f(x) = g(x^2)$ and $\phi(x) = g(1-x^2)$, i.e. $f$ and $\phi$ are compositions of smooth functions.

share|improve this answer
That's much better indeed, thank you! –  Jose L. Lykón Mar 13 '11 at 1:59
add comment

Hint 1: For every positive $c$ and every real $a$, the function $f_{a,c}$ defined by $f_{a,c}(x)=\exp(-c/(x-a))$ if $x>a$ and $f_{a,c}(x)=0$ if $x\le a$ is infinitely differentiable on the real line.

Hint 2: use $f_{a,c}$ to build a function infinitely differentiable on the real line, zero for $x\ge a$ and positive for $x<a$.

Hint 3 : use the decomposition of $1/(1-x^2)$ as a linear combination of $1/(1\pm x)$ to write $\phi$ as a product of two infinitely differentiable functions.

share|improve this answer
This is a good detour as well, thanks a bunch. –  Jose L. Lykón Mar 13 '11 at 2:02
@Jose Not sure this qualifies as a detour if one keeps in mind the general case of an interval $[a,b]$... but nevermind. –  Did Mar 13 '11 at 8:22
add comment

Your Answer


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

Not the answer you're looking for? Browse other questions tagged or ask your own question.