tinytheorems - about groups - rings - modules - models atom/rss

Exploration 7: Sylow theory

.

Questions:


Recall Cauchy’s theorem, which tells you that each prime divisor of G|G| corresponds to a subgroup of that order. But those are far from the only subgroups of GG. How do we get at the other subgroups of GG? That’s where Sylow theory comes in.

Recall that a pp-subgroup is a subgroup of prime power order pnp^n. Sylow’s theorems are all about identifying the pp-subgroups of a given group. Here is the first theorem, which can be seen as a generalization of Cauchy’s theorem:

First Sylow Theorem: A finite group GG contains a pp-subgroup of order pkp^k for every prime power divisor pkp^k of G|G|.

We proceed by strong induction on G|G|. The base case is simple: if G=p|G|=p, then GG itself is our required subgroup of order p1p^1.

Otherwise, given that pkp^k divides G|G|, we must show that GG contains a subgroup of order pkp^k. The inductive hypothesis will let us claim that any group smaller than GG whose order is divisible by some prime power pjp^j, has a subgroup of order pjp^j.

First, recall that all groups satisfy the class equation: G=Z(G)+iG/CG(gi)|G|=|Z(G)|+\sum_i|G|/|C_G(g_i)| Given that pkp^k divides the LHS, G|G|, it must divide the RHS as well. Then pp divides the RHS, and by properties of divisibility, pp either divides both Z(G)|Z(G)| and the sum iG/CG(gi)\sum_i|G|/|C_G(g_i)|, or it doesn’t divide either of them.

If pp divides Z(G)|Z(G)|, then by Cauchy’s theorem, Z(G)Z(G) contains an element of order pp. This element aZ(G)a\in Z(G), being central, must generate a normal subgroup of GG. That means we can consider the quotient G/aG/\<a\>.

  • Since pkp^k divides G|G| and a=p|\<a\>|=p, it follows that pk1p^{k-1} divides G/a|G/\<a\>|. By inductive hypothesis, G/aG/\<a\> must contain a subgroup HH of order pk1p^{k-1}.
  • But since every coset in G/aG/\<a\> contains the same number of elements (namely a=p|\<a\>|=p elements) this means the union of the pk1p^{k-1} cosets that make up HH within GG form a subset of GG of size pkp^k.
  • Since HH is a subgroup, this subset is the required subgroup of order pkp^k.

Otherwise, if pp fails to divide iG/CG(gi)\sum_i|G|/|C_G(g_i)|, then at least one of the G/CG(gi)|G|/|C_G(g_i)| in the sum must be indivisible by pp.

  • If G|G| is divisible by pkp^k but G/CG(gi)|G|/|C_G(g_i)| is indivisible by pp, it follows that CG(gi)|C_G(g_i)| contains all the factors pp of G|G| and is therefore divisible by pkp^k.
  • Since the order of CG(gi)C_G(g_i) is divisible by pkp^k, we’d like use the inductive hypothesis on CG(gi)C_G(g_i) (because centralizers are always subgroups of GG) to immediately show that it must contain a pp-subgroup of order pkp^k, as required. But to use the inductive hypothesis, we need CG(gi)<G|C_G(g_i)|<|G|.
  • We can show CG(gi)<G|C_G(g_i)|<|G| by showing that G/CG(gi)>1|G|/|C_G(g_i)|>1. We show this by showing G/CG(gi)=1|G|/|C_G(g_i)|=1 is false — it would imply a singleton conjugacy class, but in the class equation, singleton conjugacy classes are counted in Z(G)|Z(G)|, not the sum iG/CG(gi)\sum_i|G|/|C_G(g_i)|. Thus G/CG(gi)char "338=1|G|/|C_G(g_i)|\ne 1, so G/CG(gi)>1|G|/|C_G(g_i)|>1.

In summary, if there isn’t a central element of order pp that lets us inductively find a subgroup of order pkp^k, there must be a centralizer that contains a subgroup of order pkp^k.

If there is a pp-subgroup for every prime power pkp^k that divides G|G|, it follows that there is a maximal pp-subgroup, in the sense that no other pp-subgroups properly contain it. The order of this maximal subgroup, known as a Sylow pp-subgroup, must be the largest pkp^k that divides G|G|. Sylow pp-subgroups need not be proper (it can be the whole group). The key property of Sylow pp-subgroups that we like to use is the following:

Theorem: Any pp-subgroup containing a Sylow pp-subgroup PP must be equal to PP.

By definition, no other pp-subgroup properly contains a Sylow pp-subgroup. Thus if a pp-subgroup contains a Sylow pp-subgroup, it must be improperly contained, i.e. the two subgroups are equal.

Sylow pp-subgroups need not be unique: there can be many different Sylow pp-subgroups in a given group. The first Sylow theorem implies that Sylow pp-subgroups always exist. So how many are there? This leads us to the second Sylow theorem:

Second Sylow Theorem: The number of Sylow pp-subgroups contained in a finite group GG is equivalent to 1modp1\bmod p.
Lemma: Let PP be a Sylow pp-subgroup, and let gGg\in G be an element of order pkp^k. Then gPg1=PgPg^{-1}=P implies gPg\in P.

Assuming gPg1=PgPg^{-1}=P (i.e. gg is in the normalizer of PP), that makes PP invariant under conjugation under gg. Since PP is already invariant under conjugation by its own elements, PP is invariant under conjugation by both gg and PP, making it a normal subgroup of R=g,PR=\<g,P\>. Then the quotient R/PR/P, being generated by gPgP, must be cyclic, and it must be a pp-group since gg has order pkp^k. Then by R=R/PP|R|=|R/P|\cdot |P|, the order of RR is the product of the order of two pp-groups, making RR a pp-group itself. But by containing the Sylow pp-subgroup PP, RR must equal PP. This implies R=g,P=PR=\<g,P\>=P, and therefore gPg\in P.

Now let Sylp(G)\Syl_p(G) denote the set of all Sylow pp-subgroups of GG, and consider the conjugation action of GG on Sylp(G)\Syl_p(G). That is to say, for every gGg\in G and PSylp(G)P\in\Syl_p(G), gPg1gPg^{-1} is another Sylow pp-subgroup in Sylp(G)\Syl_p(G). Let P,QP,Q be two Sylow pp-subgroups of Sylp(G)\Syl_p(G). Consider the PP-orbit OPO_P of QQ under this action (all the subgroups obtainable by conjugating QQ by elements of PP). By the orbit-stabilizer theorem, we have: OP=PStabP(Q)|O_P|=\frac{|P|}{|Stab_P(Q)|} Here, OP|O_P| can only be a power of pp since its factors come from the order of the pp-subgroup P=pk|P|=p^k. Note that if p0=1p^0=1, that would imply the orbit is {Q}\{Q\} and that QQ is invariant under conjugation by all elements gPg\in P. But by the lemma, if QQ is invariant under conjugation by an element gg of order pkp^k, then gQg\in Q. Since this is true for all gPg\in P, we have PQP\subseteq Q, and since subgroups containing Sylow pp-subgroups are equal, P=QP=Q. Therefore there is only one orbit of order p0=1p^0=1, and all the others have order a power of pp.

Since conjugation by gGg\in G is always a transitive group action, every Sylow pp-subgroup in Sylp(G)\Syl_p(G) can be obtained by conjugating QQ by elements in GG. Thus Sylp(G)\Syl_p(G) can be considered the union of all PP-orbits of QQ. Again, there is one orbit of order 11 among all other orbits of order a power of pp, so we have Sylp(G)1modp|\Syl_p(G)|\equiv 1\bmod p.

Recall the Fundamental Theorem of Finite Abelian Groups, which states that every finite abelian group GG can be decomposed into a direct product of pp-groups: cyclic groups of prime power order.

Now consider the non-abelian groups.

In this section, we describe the third Sylow theorem.

Sylow theory deals with the structure of finite groups, and hinges on the Sylow theorems that help immensely with proving groups solvable.

Define a Sylow pp-subgroup as a maximal pp-subgroup of GG (it has prime power order and not contained in any other pp-subgroup.)

Third Sylow Theorem: The number of Sylow pp-subgroups npn_p is congruent to 1 mod p1\text{ mod }p. If the group is order pnqp^n q, then npn_p divides qq.
  • First part:
    • Let SS be the set of all Sylow pp-subgroups of GG, which necessarily have the same order pnp^n since they are maximal.
    • The conjugate of a subgroup is a subgroup of the same order. Then conjugating the Sylow pp-subgroups in SS permutes that set.
    • Conjugating the subgroups in SS by gg will fix the Sylow pp-subgroups that contain gg, because subgroups are closed under group product.
    • Conjugating the subgroups in SS by PP will fix only PP itself (since all Sylow pp-subgroups are maximal, only PP contains elements only in PP.) This means all other Sylow pp-subgroups will be permuted to another Sylow pp-subgroup.
    • But then conjugating by PP permutes S{P}S\setminus\{P\} with no fixed points.
    • Each element of PP generates a cyclic subgroup HH, which by Lagrange’s Theorem, divides P=pn|P|=p^n .
    • Furthermore, the cycle length kk of this permutation must divide the order of that cyclic subgroup, because conjugating by HH kk times must lead you back to the same subgroup. Therefore, every cycle in this permutation must be a power of pp.
    • The fact that there are no fixed points implies that there are no cycles of length 11, and therefore every cycle in the permutation is divisible by pp.
    • If a permutation on a set has only cycles divisible by pp, then the size of the set is divisible by pp. This means S{P}=S1|S\setminus\{P\}|=|S|-1 is divisible by pp, which implies S1 mod p|S|\equiv 1\text{ mod }p.
  • Second part:
    • Conjugation by an arbitrary element gGg\in G is a permutation of SS.
    • Define the homomorphism f:GSSf:G\to S_{|S|}, which maps elements of GG to the corresponding permutation in the symmetric group of SS, SSS_{|S|}.
    • The kernel KK of this homomorphism is a normal subgroup of GG containing all the elements of gg that fix SS via conjugation. (i.e. they lead to the identity permutation in SSS_{|S|}.) Then elements of G/KG/K represent all the distinct permutations of SS by conjugation of some element in GG.
    • But since there are exactly S|S| distinct Sylow pp-subgroups reachable by such permutations, the number of distinct permutations divide S|S|, so G/K|G/K| divides S|S|.
    • Since G=G/KK|G|=|G/K||K|, we know that G=(kS)K|G|=(k|S|)|K| for some kk. We can rearrange this into K=GkS|K|=\frac{|G|}{k|S|} or K=pnqkS|K|=\frac{p^nq}{k|S|}.
    • Since S1 mod p|S|\equiv 1\text{ mod }p, we know S|S| doesn’t contain any factor pp, therefore S|S| divides qq.

An immediate application arises from the Third Sylow Theorem. First, npn_p divides G/pn|G|/p^n where pnp^n is the highest power of pp dividing G|G|. If you look at the divisors of G/pn|G|/p^n and find that 11 is the only divisor congruent to 1 mod p1~\text{mod }p, then that means np=1n_p=1, so for the given pp there is only one Sylow pp-subgroup PP. But subgroups of unique order are normal, meaning you can take G/PG/P to reduce the order of the group, and if the resulting group is solvable then GG is solvable.

Here’s some specific applications of that idea:

Theorem: Every group with order pqpq (p,qp,q primes) is solvable.
  • Assume that p>qp>q. (If p=qp=q then it’s a prime power order, therefore solvable.)
  • Third Sylow Theorem: The number of Sylow pp-subgroups npn_p is congruent to 1 mod p1\text{ mod }p. If the group is order pnqp^n q, then npn_p divides qq.
  • In this case, since npn_p divides the prime qq it must be either 11 or qq. It cannot be qq because np1( mod p)n_p\equiv 1(\text{ mod } p) but p>qp>q. Therefore np=1n_p=1.
  • Let G1G_1 be that single Sylow pp-subgroup. Lagrange’s Theorem says the order of every subgroup divides the order of the group pqpq. We know that pp-subgroups have prime power order pkp^k, so the order of G1G_1 must be pp.
  • G1G_1 is the only pp-subgroup, so it is the only order pp subgroup. Recall that subgroups with unique order are normal. Therefore G1GG_1\lhd G.
  • Since pp-groups like G1G_1 are solvable, G1GG_1\lhd G indicates that GG is solvable.

Theorem: Every group GG with order pnqmp^nq^m (p,qp,q primes) where none of q,q2,,qnq,q^2,\ldots,q^n are congruent to 1 mod p1\text{ mod }p is solvable.
  • Third Sylow Theorem: The number of Sylow pp-subgroups npn_p is congruent to 1 mod p1\text{ mod }p. If the group is order pnqp^n q, then npn_p divides qq.
  • Since npn_p divides qmq^m it must be a power of qq between 11 and qmq^m.
  • But npn_p is congruent to 1 mod p1\text{ mod }p.
  • Since none of q,q2,,qnq,q^2,\ldots,q^n are congruent to 1 mod p1\text{ mod }p, npn_p must be 11 – there is only one Sylow pp-subgroup, PP.
  • Recall that subgroups with unique order are normal. Therefore PP is normal. Since PP has prime power pnp^n it is solvable. So we have PGP\lhd G and therefore GG is solvable.

And here’s the general application:

Theorem: A group GG has a normal Sylow pp-subgroup when G/pn|G|/p^n (pnp^n being the highest power of pp dividing G|G|) has no nontrivial divisors congruent to 1 mod p1\text{ mod }p.
  • TODO

divisors of 12 => 1 mod 5 == 1 iff

Theorem: Every group with order < 60 is solvable.
  • Every integer between 11 and 5959 is either prime, a prime power, or the product of two primes, except for the following:

    ┌──┬─────────┐
    │# │primes   │
    ├──┼─────────┤
    │12│2 2 3    │
    │18│2 3 3    │
    │20│2 2 5    │
    │24│2 2 2 3  │
    │28│2 2 7    │
    │30│2 3 5    │
    │36│2 2 3 3  │
    │40│2 2 2 5  │
    │42│2 3 7    │
    │44│2 2 11   │
    │45│3 3 5    │
    │48│2 2 2 2 3│
    │50│2 5 5    │
    │52│2 2 13   │
    │54│2 3 3 3  │
    │56│2 2 2 7  │
    └──┴─────────┘
  • Most of these are of order pnqmp^nq^m for prime p,qp,q, and so we need only prove either p,p2,pnchar "3381 mod qp,p^2,\ldots p^n\not\equiv 1\text{ mod }q or p,p2,pnchar "3381 mod qp,p^2,\ldots p^n\not\equiv 1\text{ mod }q to show that they are solvable:

    ┌──┬─────────┬────────────────────────────┐
    │# │primes   │                            │
    ├──┼─────────┼────────────────────────────┤
    │12│2 2 3    │ 3 ≢ 1 (mod 4)              │
    │18│2 3 3    │ 2 ≢ 1 (mod 3)              │
    │20│2 2 5    │ 2,4 ≢ 1 (mod 5)            │
    │24│2 2 2 3  │ ?                          │
    │28│2 2 7    │ 2,4 ≢ 1 (mod 7)            │
    │30│2 3 5    │ --                         │
    │36│2 2 3 3  │ ?                          │
    │40│2 2 2 5  │ 2,4,8 ≢ 1 (mod 5)          │
    │42│2 3 7    │ --                         │
    │44│2 2 11   │ 2,4 ≢ 1 (mod 11)           │
    │45│3 3 5    │ 3,9 ≢ 1 (mod 5)            │
    │48│2 2 2 2 3│ ?                          │
    │50│2 5 5    │ 2 ≢ 1 (mod 5)              │
    │52│2 2 13   │ 2,4 ≢ 1 (mod 13)           │
    │54│2 3 3 3  │ 2 ≢ 1 (mod 3)              │
    │56│2 2 2 7  │ ?                          │
    └──┴─────────┴────────────────────────────┘
  • TODO

< Back to category Exploration 7: Sylow theory (permalink)
Exploration 6: Groups as matrices An unconventional intro to group theory