How can we prove that f'(0) = 0 if the series converges

  • Thread starter Thread starter ptolema
  • Start date Start date
  • Tags Tags
    Convergence Proof
Click For Summary

Homework Help Overview

The problem involves a continuous function f defined around 0, with a focus on the derivative f'(0) and its relationship to the convergence of a series involving f(1/n). The goal is to prove that if f'(0) exists and the series converges, then f'(0) must equal 0.

Discussion Character

  • Exploratory, Assumption checking

Approaches and Questions Raised

  • Participants discuss the implications of using l'Hôpital's rule and the continuity of f' at 0. There is exploration of the limit definition of the derivative and its relation to the behavior of f(1/n) as n approaches infinity. Questions arise about the convergence of the series c/n and its implications for the value of c.

Discussion Status

The discussion is ongoing, with participants exploring different interpretations of the relationship between f'(0) and the convergence of the series. Some guidance has been offered regarding the implications of c being non-zero, but no consensus has been reached.

Contextual Notes

There is an assumption that the series converges, and previous work has established that f(0) equals 0. The participants are navigating the implications of these conditions without resolving the core question.

ptolema
Messages
82
Reaction score
0

Homework Statement



let f be a continuous function on an interval around 0, and let an=f(1/n) (for large enough n)
prove that if f'(0) exists and
conv-1.jpg
converges, then f'(0)=0

Homework Equations



proved earlier in the problem that if the series converges, then f(0)=0.

The Attempt at a Solution



for this part of the problem, i found that
lhop.jpg
by l'Hopital, so f' is continuous at 0. how can i then draw the conclusion that f'(0)=0? i know that the series should tie in somehow, but I'm getting stuck on making the connection.
 
Physics news on Phys.org
Call c=f'(0). Then you know c=lim n->inf f(1/n)/(1/n). So for large n, f(1/n) is 'like' c/n. Does the series c/n converge?
 
ok, so i just did the work to get to that part. so for large n, f'(0)/n is approx f(1/n), and f(1/n) is approx 0. since f'(0)/n is "like" 0, f'(0) approaches 0 for large n. is that something like what you were saying?
 
ptolema said:
ok, so i just did the work to get to that part. so for large n, f'(0)/n is approx f(1/n), and f(1/n) is approx 0. since f'(0)/n is "like" 0, f'(0) approaches 0 for large n. is that something like what you were saying?

Not at all! f'(0)/n may approach zero for large n, but that's because n->infinity! It doesn't say anything at all about f'(0). The hint I'm trying to give you is that the series c/n does not converge if c is not zero.
 

Similar threads

Replies
26
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 15 ·
Replies
15
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
7
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 48 ·
2
Replies
48
Views
6K