It Telescopes? : Extreme Edition

Calculus Level 5

Let f 0 ( x ) = e x f_{0}(x) = e^{x} , and for all n = 1 , 2 , 3 , 4 , . . . n = 1,2,3,4,... , let f n ( x ) = e f n 1 ( x ) f_{n}(x)=e^{f_{n-1}(x)} . If

S = n = 0 2018 [ 0 1 [ k = 0 n f k ( x ) ] d x ] S = \sum_{n=0}^{2018} \left[ \int_{0}^{1} \left [ \prod_{k=0}^{n} f_{k}(x) \right] dx \right]

Then what natural number j j satisfies 0 > ln ( ln ( ln ( . . . ( ln ( S ) ) . . . ) ) ) j times 0>\underbrace{\ln(\ln(\ln(...(\ln(S))...)))}_{j \text{ times}} ?

Details and Assumptions:

  • Undefined is not less than 0.
  • ' j j times' means that there are j j nested natural logs.


The answer is 2020.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

Brandon Monsen
Feb 27, 2018

Let's begin by noting the first few f n ( x ) f_{n}(x) .

f 0 ( x ) = e x , f 1 ( x ) = e e x , f 2 ( x ) = e e e x , f 3 ( x ) = e e e e x , . . . f_{0}(x) = e^{x}, \ f_{1}(x) = e^{e^{x}}, \ f_{2}(x) = e^{e^{e^{x}}}, \ f_{3}(x) = e^{e^{e^{e^{x}}}}, \ ...

We can see that f n ( x ) f_{n}(x) is a power tower of n + 1 n+1 e e 's with an x x topping them off. Let P n ( x ) P_{n}(x) be the product inside the integral. Then

P n ( x ) = k = 0 n f k ( x ) = e x e e x e e e x . . . e e e . . . x n + 1 e’s P_{n}(x) = \prod_{k=0}^{n} f_{k}(x) = e^{x}e^{e^{x}}e^{e^{e^{x}}} ... \underbrace{e^{e^{e^{...^{x}}}}}_{n+1 \text{ e's}}

Notice that d d x e e e . . . x n + 1 e’s = P n ( x ) \frac{d}{dx} \underbrace{e^{e^{e^{...^{x}}}}}_{n+1 \text{ e's}} = P_{n}(x) Then we can rewrite the integral as follows:

0 1 P n ( x ) d x = 0 1 d d x e e e . . . x n + 1 e’s d x = 0 1 d ( e e e . . . x n + 1 e’s ) = e e e . . . 1 n + 1 e’s e e e . . . 0 n + 1 e’s = e e e . . . 1 n + 1 e’s e e e . . . 1 n e’s \int_{0}^{1} P_{n}(x) dx = \int_{0}^{1} \frac{d}{dx} \underbrace{e^{e^{e^{...^{x}}}}}_{n+1 \text{ e's}} dx = \int_{0}^{1} d \left( \underbrace{e^{e^{e^{...^{x}}}}}_{n+1 \text{ e's}} \right) = \underbrace{e^{e^{e^{...^{1}}}}}_{n+1 \text{ e's}} - \underbrace{e^{e^{e^{...^{0}}}}}_{n+1 \text{ e's}} = \underbrace{e^{e^{e^{...^{1}}}}}_{n+1 \text{ e's}} - \underbrace{e^{e^{e^{...^{1}}}}}_{n \text{ e's}}

Now, looking at S S we get

S = n = 0 2018 0 1 P n ( x ) d x = n = 0 2018 e e e . . . 1 n + 1 e’s e e e . . . 1 n e’s S = \sum_{n=0}^{2018} \int_{0}^{1} P_{n}(x) dx = \sum_{n=0}^{2018} \underbrace{e^{e^{e^{...^{1}}}}}_{n+1 \text{ e's}} - \underbrace{e^{e^{e^{...^{1}}}}}_{n \text{ e's}}

Which is a nice telescoping sum! The sum can then be easily be evaluated as S = e e e . . . 1 2019 e’s 1 = e e e . . . 2019 e’s 1 S = \underbrace{e^{e^{e^{...^{1}}}}}_{2019 \text{ e's}} - 1 = \underbrace{e^{e^{e^{...}}}}_{2019 \text{ e's}} - 1 . The 1 has been dropped since the e 1 e^{1} at the end of the power tower is simply equal to e e . Now note that

e e e . . . 2019 e’s > e e e . . . 2019 e’s 1 > e e e . . . 2018 e’s \underbrace{e^{e^{e^{...}}}}_{2019 \text{ e's}} > \underbrace{e^{e^{e^{...}}}}_{2019 \text{ e's}}-1 >\underbrace{e^{e^{e^{...}}}}_{2018 \text{ e's}}

And also note that since the natural log is strictly increasing over its domain, applying ln \ln to all three preserves the inequality. Lastly, note that applying ln \ln to any power tower of e e e e . . . e e^{e^{e^{e^{...^{e}}}}} consisting of finitely many e e 's removes one from the power tower. Thus, applying ln \ln 2018 times to both sides gives

e > ln ( ln ( ln ( ln ( . . . ln ( S ) . . . ) ) ) ) 2018 times > 1 1 > ln ( ln ( ln ( ln ( . . . ln ( S ) . . . ) ) ) ) 2019 times > 0 0 > ln ( ln ( ln ( ln ( . . . ln ( S ) . . . ) ) ) ) 2020 times > e > \underbrace{\ln(\ln(\ln(\ln(...\ln(S)...))))}_{2018 \text{ times}} > 1 \\ 1 > \underbrace{\ln(\ln(\ln(\ln(...\ln(S)...))))}_{2019 \text{ times}} > 0 \\ 0 > \underbrace{\ln(\ln(\ln(\ln(...\ln(S)...))))}_{2020 \text{ times}} > -\infty

And so our answer is 2020 \boxed{2020}

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...