Calculate the value of $k$ such that the following limit has a finite solution, $L$ such that $L \ne 0$:
$$\lim_{x\rightarrow0} \frac{(e^{x^2}-x^2-1)(\cos(x)-1)}{x^k}$$
I use the Taylor Series expansions of $e^x$ and $\cos(x)$ and simplify the above expression to the following:
$$\lim_{x\rightarrow0} \frac{-\frac{1}{4}x^6+(\frac{1}{48}-\frac{1}{12})x^8+(\frac{1}{144})x^{10}}{x^k}$$
Now I have a mental roadblock. For $k<6$ the above limit goes to zero and for $k>6$ this expression should diverge but in my head it goes to zero again.... I am trying to think of a simple example to convince myself but can't. Can someone please help me understand this?
Thanks.
You are almost there
$$\lim_{x\rightarrow0} \frac{-\frac{1}{4}x^6+(\frac{1}{48}-\frac{1}{12})x^8+(\frac{1}{144})x^{10}}{x^k}=\\ =\lim_{x\rightarrow0} -\frac{1}{4}x^{6-k}+(\frac{1}{48}-\frac{1}{12})x^{8-k}+(\frac{1}{144})x^{10-k} $$
If any of the exponents is negative, then the limit goes to infinity, so you must have $k\le 6$. Also if all exponents are (stricly) positive (this happens iff $k<6$) then the limit goes to zero.
Then you must have $k=6$.
Why? If $k>6$, the first (at least) summand goes to infinity.
The standard recipe for limits with polynomials is: "When $x$ goes to infinity, the highest degree term rules; when $x$ goes to zero, the lowest degree term rules". Here, this suggests to factor the ruling term from the fraction and write it as
$$ x^{6-k} ( a + b x^2 +cx^4)$$ for some non-zero $a,b,c$. Then, we quickly see that as $x\to 0$ the factor in parentheses tends to the constant $a$, and we have the three cases: if $k<6$ the left factor tends to zero, if $k>6$ it tends to infinity, if $k=6$ it tends to one (and hence the limit to $a$).