Infinite Products

From Sums to Products

An infinite product has the form

\prod_{n=1}^{\infty} a_n = a_1a_2a_3\cdots,

where each $a_n$ is a real or complex number. As with infinite series, the meaning of the expression is defined by finite approximations. The $N$ -th partial product is

P_N=\prod_{n=1}^{N}a_n.

The infinite product converges to $L$ if

\lim_{N\to\infty} P_N=L

exists and $L\neq 0$ . The nonzero condition is important. If the product tends to $0$ , the factors may still contain useful information, but the product is usually called divergent in the standard theory of infinite products.

Products Near One

Most infinite products in number theory have factors close to $1$ . They are often written as

\prod_{n=1}^{\infty}(1+b_n).

Here the behavior of the product is controlled by the size of the terms $b_n$ . If the numbers $b_n$ are small, then multiplication by $1+b_n$ changes the product only slightly.

The basic comparison is with the logarithm:

\log\prod_{n=1}^{N}(1+b_n) = \sum_{n=1}^{N}\log(1+b_n).

For small $b_n$ ,

\log(1+b_n)\approx b_n.

Thus infinite products are closely related to infinite series. In many cases, convergence of

\sum_{n=1}^{\infty} b_n

controls convergence of

\prod_{n=1}^{\infty}(1+b_n).

A Simple Example

Consider

\prod_{n=2}^{\infty}\left(1-\frac1{n^2}\right).

Each factor can be factored as

1-\frac1{n^2} = \frac{(n-1)(n+1)}{n^2} = \frac{n-1}{n}\cdot\frac{n+1}{n}.

The finite product becomes

\prod_{n=2}^{N}\left(1-\frac1{n^2}\right) = \left(\frac12\cdot\frac32\right) \left(\frac23\cdot\frac43\right) \cdots \left(\frac{N-1}{N}\cdot\frac{N+1}{N}\right).

Most factors cancel. The result is

\prod_{n=2}^{N}\left(1-\frac1{n^2}\right) = \frac{N+1}{2N}.

Taking $N\to\infty$ , we obtain

\prod_{n=2}^{\infty}\left(1-\frac1{n^2}\right) = \frac12.

This example shows that infinite products may have exact values, just as infinite series may have exact sums.

Vanishing Products

Not every infinite product with positive factors has a positive limit. Consider

\prod_{n=2}^{N}\left(1-\frac1n\right).

Since

1-\frac1n=\frac{n-1}{n},

we get

\prod_{n=2}^{N}\left(1-\frac1n\right) = \frac12\cdot\frac23\cdot\frac34\cdots\frac{N-1}{N} = \frac1N.

Hence the partial products tend to $0$ . In this case the infinite product does not converge to a nonzero value.

The difference between the two examples is the size of the correction terms. The product

\prod_{n=2}^{\infty}\left(1-\frac1{n^2}\right)

converges because the series

\sum_{n=2}^{\infty}\frac1{n^2}

converges. The product

\prod_{n=2}^{\infty}\left(1-\frac1n\right)

vanishes because the harmonic series

\sum_{n=2}^{\infty}\frac1n

diverges.

Infinite Products in Number Theory

Infinite products enter number theory through prime factorization. The unique factorization of integers suggests that sums over all positive integers may sometimes be rewritten as products over primes.

The most important example is Euler’s product for the zeta function:

\sum_{n=1}^{\infty}\frac1{n^s} = \prod_{p}\left(1-\frac1{p^s}\right)^{-1},

valid for real $s>1$ , and later for complex $s$ with real part greater than $1$ . The product ranges over all primes.

This formula is one of the central bridges between arithmetic and analysis. The left side is a series over all positive integers. The right side is a product over primes. Unique factorization is the reason the two expressions agree.

Why Products Matter

Infinite products make prime numbers visible inside analytic expressions. A series such as

\sum_{n=1}^{\infty}\frac1{n^s}

does not explicitly mention primes. Its Euler product does. This allows analytic properties of functions to encode arithmetic information about primes.

For this reason, infinite products are a basic tool in analytic number theory. They prepare the way for Euler products, Dirichlet series, $L$ -functions, and the analytic study of prime distribution.