TheoremComplete

Primitive Roots - Main Theorem

The existence and structure of primitive roots modulo prime powers and special moduli form one of the cornerstones of elementary number theory. These results reveal precisely when the multiplicative group modulo nn has maximal cyclic structure.

TheoremPrimitive Roots Modulo Primes

Let pp be a prime number. Then there exist Ο•(pβˆ’1)\phi(p-1) primitive roots modulo pp.

Moreover, for any prime pp, the multiplicative group (Z/pZ)βˆ—(\mathbb{Z}/p\mathbb{Z})^* is cyclic of order pβˆ’1p-1.

This fundamental result guarantees that every prime has primitive roots. The cyclicity of (Z/pZ)βˆ—(\mathbb{Z}/p\mathbb{Z})^* makes modular arithmetic modulo primes particularly well-behaved.

ExamplePrimitive Roots Modulo 17

For p=17p = 17, we have Ο•(pβˆ’1)=Ο•(16)=8\phi(p-1) = \phi(16) = 8 primitive roots.

Since 33 is a primitive root modulo 1717, all primitive roots are 3k3^k where gcd⁑(k,16)=1\gcd(k, 16) = 1: k∈{1,3,5,7,9,11,13,15}k \in \{1, 3, 5, 7, 9, 11, 13, 15\}

Computing: 31=3,33=27≑10,35≑5,37≑11,39≑14,311≑7,313≑12,315≑6(mod17)3^1 = 3, 3^3 = 27 \equiv 10, 3^5 \equiv 5, 3^7 \equiv 11, 3^9 \equiv 14, 3^{11} \equiv 7, 3^{13} \equiv 12, 3^{15} \equiv 6 \pmod{17}.

So the primitive roots modulo 1717 are: {3,5,6,7,10,11,12,14}\{3, 5, 6, 7, 10, 11, 12, 14\}.

TheoremStructure of $(\mathbb{Z}/p^k\mathbb{Z})^*$

Let pp be an odd prime and kβ‰₯1k \geq 1. Then (Z/pkZ)βˆ—(\mathbb{Z}/p^k\mathbb{Z})^* is cyclic of order Ο•(pk)=pkβˆ’1(pβˆ’1)\phi(p^k) = p^{k-1}(p-1).

In particular, primitive roots exist modulo pkp^k for all odd primes pp and all kβ‰₯1k \geq 1.

The cyclicity of (Z/pkZ)βˆ—(\mathbb{Z}/p^k\mathbb{Z})^* for odd prime powers is remarkable and contrasts sharply with the behavior for powers of 22.

ExamplePrimitive Roots Modulo 25

We have Ο•(25)=Ο•(52)=20\phi(25) = \phi(5^2) = 20. Since 22 is a primitive root modulo 55, we check if it works modulo 2525.

Compute 2Ο•(5)=24=16(mod25)2^{\phi(5)} = 2^4 = 16 \pmod{25}. Since 16≑̸1(mod25)16 \not\equiv 1 \pmod{25}, the element 22 is a primitive root modulo 2525 (and modulo all 5k5^k).

The order is ord25(2)=20\text{ord}_{25}(2) = 20, confirming it generates all Ο•(25)=20\phi(25) = 20 units.

TheoremStructure of $(\mathbb{Z}/2^k\mathbb{Z})^*$ for $k \geq 3$

For kβ‰₯3k \geq 3, the group (Z/2kZ)βˆ—(\mathbb{Z}/2^k\mathbb{Z})^* is not cyclic. Instead: (Z/2kZ)βˆ—β‰…Z/2ZΓ—Z/2kβˆ’2Z(\mathbb{Z}/2^k\mathbb{Z})^* \cong \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2^{k-2}\mathbb{Z}

The group has exponent 2kβˆ’22^{k-2}, not Ο•(2k)=2kβˆ’1\phi(2^k) = 2^{k-1}.

This explains why primitive roots don't exist modulo powers of 22 greater than 44. The multiplicative group factors as a product of two cyclic groups rather than being cyclic itself.

ExampleNo Primitive Root Modulo 16

For n=16n = 16, we have Ο•(16)=8\phi(16) = 8. The group (Z/16Z)βˆ—={1,3,5,7,9,11,13,15}(\mathbb{Z}/16\mathbb{Z})^* = \{1, 3, 5, 7, 9, 11, 13, 15\}.

Every element has order dividing 44:

  • 32=9,34=81≑1(mod16)3^2 = 9, 3^4 = 81 \equiv 1 \pmod{16}, so ord16(3)=4\text{ord}_{16}(3) = 4
  • 52=25≑9,54≑1(mod16)5^2 = 25 \equiv 9, 5^4 \equiv 1 \pmod{16}, so ord16(5)=4\text{ord}_{16}(5) = 4
  • 72=49≑1(mod16)7^2 = 49 \equiv 1 \pmod{16}, so ord16(7)=2\text{ord}_{16}(7) = 2

No element has order 88, confirming the group is not cyclic.

The structure is Z/2ZΓ—Z/4Z\mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/4\mathbb{Z} with generators {βˆ’1,3}\{-1, 3\}.

TheoremComplete Classification

Primitive roots modulo nn exist if and only if: n∈{1,2,4,pk,2pk:pΒ oddΒ prime,kβ‰₯1}n \in \{1, 2, 4, p^k, 2p^k : p \text{ odd prime}, k \geq 1\}

For all other moduli, (Z/nZ)βˆ—(\mathbb{Z}/n\mathbb{Z})^* is not cyclic.

This complete classification tells us exactly when we can find generators for the multiplicative group modulo nn.

Remark

For composite nn not of the specified forms, the group (Z/nZ)βˆ—(\mathbb{Z}/n\mathbb{Z})^* decomposes as a product of cyclic groups by the Chinese Remainder Theorem. For instance: (Z/12Z)βˆ—β‰…(Z/4Z)βˆ—Γ—(Z/3Z)βˆ—β‰…Z/2ZΓ—Z/2Z(\mathbb{Z}/12\mathbb{Z})^* \cong (\mathbb{Z}/4\mathbb{Z})^* \times (\mathbb{Z}/3\mathbb{Z})^* \cong \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}

This product structure explains why primitive roots don't exist modulo such nn.

These structural theorems completely characterize when primitive roots exist and provide the foundation for discrete logarithm-based cryptography.