Associated Topics || Dr. Math Home || Search Dr. Math

### Infinite Series of 1/n

```
Date: 01/13/98 at 08:56:17
From: Michael Theemling
Subject: Sum of an infinite series of 1/n when n=1,2,3...

This is a question I encountered in calculus a long time ago while I
was still a young math student. I understand the answer is divergence,
or the sum is infinity, but do not conceptually understand why that is
so, especially since the terms eventually go to 0.

1 +1/2+1/3+1/4+...

For n>1 and n<1 it is relatively easy to comprehend, but not this one.
I saw a proof before but never really got it.

I just want to add your site is outstanding. As a student (now
knowledge in-depth.
```

```
Date: 01/26/98 at 10:59:21
From: Doctor Joe
Subject: Re: Sum of an infinite series of 1/n when n=1,2,3...

Dear Michael,

The divergence of 1 + 1/2 + 1/3 + 1/4 + ... is famous (or rather
infamous!). The proof requires some geometrical interpretation and
a minimal knowledge of integral calculus.

Let f(x) = 1/x.

The next thing you need to do is to sketch this reciprocal function
over the interval (1,infinity).

Next, construct "boxes" that cover the region bounded by the curve
y=f(x), the y=1 line, the x-axis, and the y=n+1 line in the following
way:

On each integer interval a rectangle is constructed by taking the
height to be the maximum value over that interval. For example, on the
interval [1,2], the maximum value of the function y = 1/x is 1.

y

1 |     ____
|    |    |
|    |    |
|    |    |
1/2 |    |    |____
|    |    |    |
1/3 |    |    |    |
|    |    |    |
|    |    |    |
-----------------------  x
0    1    2    3 ...

If we form the sum of the areas of the first n rectangles, then the
value is:

1 + 1/2 + 1/3 + 1/4 + ... + 1/n

Since the curve is concave downwards, the sum of areas of the n+1
rectangles is clearly greater than the area under the curve f(x) = 1/x
on the interval [1,n+1].

So, we have the following inequality:

n+1
Int | 1/x < 1 + 1/2 + 1/3 + 1/4 + ... + 1/n
1

for each n, positive integer.

Evaluating the integral on the left, we have ln(n+1).

Suppose that the sum 1 + 1/2 + 1/3 + 1/4 + 1/5 + ... converges to,
say, a real number S.  Then,

lim      ln (n+1) < or equal to S.
n->infinity

Note that the sequence {ln(n+1)} diverges to infinity. Thus, we have
arrived at a contradiction, and therefore our supposition is false.
Hence, 1 + 1/2 + 1/3 + ... diverges.

(Classical analysts call this series the harmonic series.)

-Doctor Joe, The Math Forum
Check out our web site!  http://mathforum.org/dr.math/
```

```
Date: 04/03/00 at 08:36:42
From: Dr. Schwa
Subject: harmonic series

To prove that the sum diverges, you can also use a comparison test
to a series that obviously diverges.

Group the terms together:
(1) + (1/2) + (1/3 + 1/4) + (1/5 + 1/6 + 1/7 + 1/8) + (1/9 + ... + 1/16)
+ ...
taking twice as many terms each time.

This is clearly greater than
(1) + (1/2) + (1/4 + 1/4) + (1/8 + 1/8+ 1/8 + 1/8) + (1/16 + ... + 1/16)
+ ...
since we replaced each term with something smaller or equal.

Then, each of the parenthesized groups is 1/2, so the original harmonic
series is greater than
1/2 + 1/2 + 1/2 + 1/2 + 1/2 + ...

Since there are infinitely many 1/2's, the series diverges.

-Dr. Schwa, The Math Forum
Check out our web site!  http://mathforum.org/dr.math/
```
Associated Topics:
High School Sequences, Series

Search the Dr. Math Library:

 Find items containing (put spaces between keywords):   Click only once for faster results: [ Choose "whole words" when searching for a word like age.] all keywords, in any order at least one, that exact phrase parts of words whole words

Submit your own question to Dr. Math
Math Forum Home || Math Library || Quick Reference || Math Forum Search