Stein’s Section 6.
2: The Zeta Function
Angelicia Romero and Rebeca Hernandez
May 2025
Overview
The zeta function ζ(s) is initially defined for real numbers s > 1 by the series ∞ 1
P
n=1 ns , which
converges absolutely in this region. Stein’s goal in Section 6.2 is to extend this function
beyond Re(s) > 1 to as much of the complex plane as possible. To do this, he introduces
a new function ξ(s), which is constructed from ζ(s) but has better symmetry and analytic
properties.
Motivation
One of the most important reasons we study the zeta function is to see its connection to
prime numbers. Euler discovered the identity
Y −1
1
ζ(s) = 1− s , for Re(s) > 1.
p prime
p
This equation tells us information about the distribution of prime numbers. The product
over primes reveals a link between multiplicative structures and anaylsis. By extending
ζ(s) beyond Re(s) > 1 we will learn more about how this structure behaves. For example,
studying the zeros of ζ(s) tells us how we can accurately estimate the number of primes up
to a given value. The Prime Number Theorem states that π(x) ∼ logx x , but the error in this
approximation depends of the location of the nontrivial zeros of the ζ(s). If all these zeros lie
on the critical line Re(s) = 1/2, as the Riemann Hypotheis suggest, then we can get better
bounds on the error. So by understanding how ζ(s) behaves in the complex plane near and
on the critical line, we can shift our understanding of how primes are distributed.
Key Terms
P∞
• Zeta Function: Defined by the series ζ(s) = 1
n=1 ns , which converges for Re(s) > 1.
• Analytic (Holomorphic): A function that can be expressed as a power series and
is differentiable in the complex numbers.
1
• Meromorphic: A function that is analytic everywhere except at isolated points where
it has poles (i.e., where it goes to infinity in a controlled way).
• Pole: A type of singularity where a function diverges. A simple pole means it behaves
1
like s−a near s = a.
R∞
• Gamma Function: Defined by the integral Γ(s) = 0 ts−1 e−t dt, which generalizes
the factorial function: Γ(n) = (n − 1)!.
• Theta Function: θ(u) = ∞ −πn2 u
P
n=−∞ e , a special function with the functional equa-
1
tion θ(u) = u θ(1/u).
√
• Symmetry of ξ(s): ξ(s) = π −s/2 Γ(s/2)ζ(s) satisfies ξ(s) = ξ(1 − s).
• Integral: Integrals are used here to express functions like ζ(s) in a continuous way
that can be extended to more values of s. For example, the Gamma function itself is
defined using an integral, and an important step in extending ζ(s) is writing it in terms
of integrals involving functions like ψ(u) = (θ(u) − 1)/2, which is used to isolate the
non-constant part of the theta function. This makes it easier to analyze and extend
the domain of the function.
• Entire Function: A function that is analytic over the entire complex plane.
• Uniform Convergence on Compact Sets: A technical condition that guarantees
the limit of a sequence of analytic functions is still analytic. Used in the proof that
the integral expression for ξ(s) is entire.
Theorems and Intuition
Proposition 2.1: The series ζ(s) = ∞ 1
P
n=1 ns converges for Re(s) > 1 and defines a holo-
morphic function on that half-plane.
Proof Idea: For s = σ + it, we have |n−s | = n−σ . When σ > 1, this decay ensures
absolute convergence of the series. The series also converges uniformly on compact subsets
of Re(s) > 1, making ζ(s) holomorphic there. R∞
Theorem 2.2: For Re(s) > 1, the identity π −s/2 Γ(s/2)ζ(s) = 0 us/2−1 ψ(u) du holds,
2
where ψ(u) = (θ(u) − 1)/2, and θ(u) = n∈Z e−πn u .
P
Key Idea: This integral representation connects ζ(s) with the Gamma function and the
theta function. It provides a smooth (analytic) form that allows extension beyond Re(s) > 1.
Theorem 2.3: The function ξ(s) = π −s/2 Γ(s/2)ζ(s) has an analytic continuation to all
of C, is holomorphic except for simple poles atPs = 0 and s = 1, and satisfies ξ(s) = ξ(1 − s).
Proof: Given the theta function: θ(t) = ∞
2 2
n=−∞ e−πn t , we can let f (x) = e−πx t . The
R∞
Fourier transform is fˆ(ξ) = −∞ e−πx t e−2πixξ dx = √1t e−πξ /t . Using the Poisson summation
2 2
formula, ∞
P∞ P∞
ˆ −πn2 t
= √1t ∞ −πm2 /t
P P
n=−∞ f (n) = m=−∞ f (m), we get θ(t) = n=−∞ e m=−∞ e =
1 1
R ∞ s/2−1
√
t
θ t . Next, we consider the Mellin Transform, I(s) := 0 (θ(t)−1)t dt for ℜ(s) > 1.
2
R1 R∞
Splitting the integral gives us I(s) = 0 (θ(t) − 1)ts/2−1 dt + 1 (θ(t) − 1)ts/2−1 dt. Apply-
R1 R∞
ing the modular transformation θ(t) = √1t θ 1t gives us 0 (θ(t) − 1)ts/2−1 dt = 1 (θ(u) −
R∞
1)u−s/2−1/2 du. Putting everything together gives us I(s) = 1 (θ(t)−1)(ts/2−1 +t−(s−1)/2−1 )dt.
This representation is symmetric in s and 1 − s and can be extended
R∞ to all s ∈ C. From
a change of variable, we get the identity π −s/2 Γ 2s ζ(s) = 12 0 (θ(t) − 1)ts/2−1 dt. From
this, we obtain the zeta function ξ(s) = 21 s(s − 1)π −s/2 Γ 2s ζ(s), which gives us the func-
tional equation ξ(s) = ξ(1 − s) which key symmetry of the zeta function and underlies
the Rcritical line ℜ(s) = 1/2. From the Mellin transform representation, we get ζ(s) =
1 ∞ ts−1
Γ(s) 0 et −1
dt (ℜ(s) > 1) and can analytically extend this integral using a change of con-
tour, and the use of modular symmetry of θ(t).
Why it matters: This reflection identity gives the zeta function a symmetric structure
about the line Re(s) = 1/2, which is fundamental to the Riemann Hypothesis.
Theorem 2.4: ζ(s) can be continued to a meromorphic function on all of C with a single
simple pole at s = 1.
Sketch of Argument: Using the identity ζ(s) = ξ(s)/(π −s/2 Γ(s/2)), the poles of Γ(s/2)
at s = 0, −2, −4, . . . are canceled by corresponding zeroes of ξ(s), so the only singularity of
ζ(s) is at s = 1.
Proposition 2.5 and Corollary 2.6: For Re(s) > 0, the zeta function satisfies ζ(s) =
1
s−1
+ H(s), where H(s) is holomorphic in that half-plane.
Key Use: This decomposition is helpful when estimating the behavior of ζ(s) near
Re(s) = 1, especially in analytic number theory.
Proposition 2.7: For s = σ + it, and any 0 ≤ σ0 ≤ 1, there exists a constant cε such
that
|ζ(s)| ≤ cε |t|1−σ0 +ε , and |ζ ′ (s)| ≤ cε |t|ε .
Why it matters: This shows the zeta function and its derivative grow slowly on vertical
lines. These bounds are important when studying the zeros and growth of ζ(s) near the
critical line.
These theorems show that the zeta function extends to a meromorphic function on the
entire complex plane, with a single singularity at s = 1. The construction of ξ(s), theta
and Gamma functions, and the integral representations all allow this continuation. The
symmetry ξ(s) = ξ(1 − s) is essential in understanding ζ(s) properties, and the growth
bounds solidify its behavior near Re(s) = 1.
