Mobius Function

Introduction

In the study of multiplicative number theory, certain arithmetic functions arise naturally when analyzing the structure of integers. Among these, the Mobius function stands out as one of the most fundamental and elegant. It encodes essential information about the prime factorization of integers in a remarkably compact form.

The Mobius function μ(n) was introduced by August Ferdinand Mobius in 1832. Despite its simple definition, this function serves as the cornerstone of multiplicative number theory, appearing in the celebrated Mobius inversion formula, which provides a powerful tool for inverting summatory relationships between arithmetic functions.

The function takes only three possible values: −1 0 and 1 This simplicity belies its deep importance: the Mobius function connects to the Riemann zeta function, the distribution of prime numbers, and fundamental questions about the arithmetic structure of integers.

Formal Definition

The Mobius function μ:ℤ+→{−1,0,1} is defined as follows. For any positive integer n

μ(n)={[1,if *n=1],[(−1)k,if *n=(p_1)*(p_2)⋯(p_k)* (distinct primes)],[0,if *p2*(∣_^)(n)* for some prime *p])

In words, μ(n) equals 1 if n is a product of an even number of distinct primes (including the empty product for n=1, equals −1 if n is a product of an odd number of distinct primes, and equals 0 if n has any repeated prime factor.

An integer n with no repeated prime factors is called squarefree. The Möbius function is nonzero precisely on squarefree integers. The set of squarefree integers has natural density 6/(π2)≈0.608, meaning roughly 61 of positive integers are squarefree.

Computing Initial Values

To develop intuition, we compute μ(n) for small values of n

μ(1)=1 (the empty product of zero primes)

μ(2)=−1 (one prime factor)

μ(3)=−1 (one prime factor)

μ(4)=0 (since 4=2 has a repeated prime)

μ(5)=−1 (one prime factor)

μ(6)=1 (since 6=2⋅3 has two distinct primes)

μ(30)=−1 (since 30=2⋅3⋅5 has three distinct primes)

μ(12)=0 (since 12=2⋅3 has a repeated prime)

Fundamental Properties

Multiplicativity

The Möbius function is multiplicative, meaning that for any two coprime positive integers m and n (that is, gcd(m,n)=1):

μ*(m*n)=μ(m)⋅μ(n)

This property follows directly from the definition. If m and n share no common prime factors, then m*n is squarefree if and only if both m and n are squarefree. Moreover, the number of prime factors of m*n equals the sum of the number of prime factors of m and n, so (−1)((k_m)+(k_n))=(−1)(k_m)⋅(−1)(k_n).

Multiplicativity is crucial because it means that to compute μ(n) for any n, we only need to know μ at prime powers. For a prime p:

μ(pk)={[−1,if *k=1],[0,if *k≥2])

Sum Over Divisors

Perhaps the most important property of the Möbius function is the following identity. For any positive integer n:

(∑_d*(∣_^)(n)^)(μ(d))={[1,if *n=1],[0,if *n>1])

This identity states that the sum of μ(d) over all divisors d of n equals 1 when n=1 and 0 otherwise. Using the notation of the identity function and Dirichlet convolution, this can be written as μ∗1=ε, where ε(n) is 1 if n=1 and 0 otherwise.

To prove this, let n=(p_1)(a_1)*(p_2)(a_2)⋯(p_k)(a_k) be the prime factorization of n>1. Only squarefree divisors contribute to the sum, so:

(∑_d*(∣_^)(n)^)(μ(d))=(∑_S⊆{(p_1),…,(p_k)}^)(−1)=(∑_j=0^k)((k/j))*(−1)j=(1−1)k=0)

The last equality uses the binomial theorem. This elegant cancellation is the key to why the Möbius function enables inversion formulas.

The Möbius Inversion Formula

The Möbius inversion formula is a fundamental tool that allows us to recover a function from its summatory function. If ƒ and g are arithmetic functions related by:

g(n)=(∑_d*(∣_^)(n)^)(ƒ(d))

then the original function ƒ can be recovered using:

ƒ(n)=(∑_d*(∣_^)(n)^)(μ(d))⋅g*(n/d)=(∑_d*(∣_^)(n)^)(μ)*(n/d)⋅g(d)

To verify this, substitute the definition of g into the inversion formula:

(∑_d*(∣_^)(n)^)(μ(d))⋅g*(n/d)=(∑_d*(∣_^)(n)^)(μ(d))*(∑_e*(∣_^)(n/d)^)(ƒ(e))=(∑_e*(∣_^)(n)^)(ƒ(e))*(∑_d*(∣_^)(n/e)^)(μ(d))

The inner sum equals 1 when n/e=1 (i.e., e=n) and 0 otherwise, yielding ƒ(n) as required.

Classical Application: Euler Totient Function

The Euler totient function ϕ(n) counts integers from 1 to n that are coprime to n. A classical identity states:

(∑_d*(∣_^)(n)^)(ϕ(d))=n

Applying Möbius inversion with g(n)=n and ƒ=ϕ, we obtain:

ϕ(n)=(∑_d*(∣_^)(n)^)(μ(d))⋅n/d=n*(∑_d*(∣_^)(n)^)(μ(d)/d)

Since both μ and the function n↦1/n are multiplicative, so is their pointwise product. Using multiplicativity, for n=(p_1)(a_1)⋯(p_k)(a_k):

ϕ(n)=n*(∏_p*(∣_^)(n)^)(1−1/p)

This is the well-known product formula for the Euler totient function.

Connection to the Riemann Zeta Function

The Möbius function has a deep connection to the Riemann zeta function. Recall that for Re(s)>1, the Riemann zeta function is defined as:

ζ(s)=(∑_n=1^∞)(1/(ns))=(∏_p* prime^)(1/(1−p(−s)))

The Dirichlet series for the Möbius function is the reciprocal of the zeta function:

(∑_n=1^∞)(μ(n)/(ns))=1/ζ(s)

To prove this, we use the Euler product. For each prime p:

(∑_k=0^∞)(μ(pk)/(p(k*s)))=1+μ(p)/(ps)+0+0+⋯=1−1/(ps)

Taking the product over all primes:

(∑_n=1^∞)(μ(n)/(ns))=(∏_p^)(1−1/(ps))=1/ζ(s)

This relationship means that the Möbius function encodes information about the zeros of the zeta function, connecting it to the distribution of prime numbers through the Prime Number Theorem and its refinements.

Dirichlet Convolution

The Möbius function plays a central role in the theory of Dirichlet convolution. For two arithmetic functions ƒ and g, their Dirichlet convolution is defined as:

(ƒ∗g)*(n)=(∑_d*(∣_^)(n)^)(ƒ(d))⋅g*(n/d)

Under this operation, the set of arithmetic functions forms a commutative ring with the identity element ε (where ε(1)=1 and ε(n)=0 for n>1).

The constant function 1*(n)=1 for all n has the Möbius function as its Dirichlet inverse:

μ∗1=ε

This is precisely the sum-over-divisors property proven earlier. The Möbius inversion formula is simply the statement that if g=ƒ∗1, then ƒ=g∗μ.

The Mertens Function

The summatory function of the Möbius function is called the Mertens function:

M(x)=(∑_n≤x^)(μ(n))

The behavior of M(x) is intimately connected to the Riemann Hypothesis. Specifically, the Riemann Hypothesis is equivalent to the bound:

M(x)=O(x(1/2+ε))

for any ε>0. This shows how the seemingly simple Möbius function connects to one of the deepest unsolved problems in mathematics.

What is known unconditionally is that M(x)=o(x), which is equivalent to the Prime Number Theorem. The Mertens function oscillates between positive and negative values, and understanding its growth is central to analytic number theory.

Generalizations

The Mobius function generalizes to other algebraic structures. On any locally finite partially ordered set (poset), one can define a Mobius function μ(x,y) for pairs x≤y that satisfies an analogous inversion formula. This broader framework, developed by Gian-Carlo Rota, unifies many combinatorial identities.

For the divisibility poset on positive integers (where d≤n means d*(∣_^)(n), the poset Mobius function μ(d,n) equals μ*(n/d) recovering the classical Mobius function.

Another important generalization is the Liouville function λ(n)=(−1)Ω(n) where Ω(n) counts prime factors with multiplicity. While μ(n) vanishes on non-squarefree integers, λ(n) is always ±1 Both functions share the property that their Dirichlet series relate to ζ(s)

Summary

The Mobius function μ(n) is a fundamental arithmetic function that takes values in {−1,0,1} It equals (−1)k when n is a product of k distinct primes, and 0 when n has a repeated prime factor.

The key property is that (∑_d*(∣_^)(n)^)(μ(d))=0 for all n>1 This enables the Mobius inversion formula: if g(n)=(∑_d*(∣_^)(n)^)(ƒ(d)) then ƒ(n)=(∑_d*(∣_^)(n)^)(μ(d))*g*(n/d)

In the language of Dirichlet convolution, the Mobius function is the inverse of the constant function 1 we have μ∗1=ε Its Dirichlet series is 1/ζ(s) connecting it to the Riemann zeta function and the distribution of primes.

The Mobius function is multiplicative, meaning μ*(m*n)=μ(m)*μ(n) when gcd(m,n)=1 This allows efficient computation and leads to elegant formulas such as ϕ(n)=n*(∏_p*(∣_^)(n)^)(1−1/p) for the Euler totient function.

The summatory Mertens function M(x)=(∑_n≤x^)(μ(n)) connects to the Riemann Hypothesis, and squarefree integers (where μ≠0 have density 6/π2 The Mobius function thus serves as a bridge between elementary arithmetic and deep analytic number theory.