Section 7.

3 : The Minimal Polynomial

Definition : Let T be a linear operator on a FDVS. A polynomial p(t) is called a minimal

Theorem 7.12 : Let p(t) be a minimal polynomial of a linear operator T on a FDVS V.

(a): For any polynomial g(t) if g(T) = O, then p(t) divides g(t). In particular p(t) divides
the characteristic polynomial of T.

(b): The minimal polynomial of T is unique.

Proof (a) : Let g(t) be a polynomial for which g(T) = O.

By division algorithm, there exist polynomials q(t) and r(t) such that

g(t) = q(t) p(t) + r(t) where r(t) = 0 or deg r(t) < deg p(t).

Then g(T) = q(T) p(T) + r(T)

i.e. O = q(T). O + r(T)

which implies r(T) = O.

As deg r(t) < deg p(t) and p(t) is the minimal polynomial of T, so r(t) must be zero.

Hence , p(t) divides g(t).

(b): Suppose that p1(t) and p2(t) are each minimal polynomials of T. Then p 1(t) divides

p2(t) by part (a). Since p1(t) and p2(t) have the same degree, we have p2(t) = c p1(t) for

some non zero scalar c . As p1(t) and p2(t) are monic, so c = 1. Hence p 1(t) = p2(t).

Definition : Let A ∈ Mn×n(F). The minimal polynomial p(t) of A is the monic polynomial
of least positive degree for which p(A) = O, i.e. the zero matrix.

Theorem 7.13 : Let T be a linear operator on a FDVS V and let β be an ordered basis
for V. Then the minimal polynomial of T is the same as the minimal polynomial of [T] β .
Corollary : For any A ∈ Mn×n(F) , the minimal polynomial of A is the same as the
minimal polynomial of LA .

Theorem 7.14 : Let T be a linear operator on a FDVS V and let p(t) be the minimal
polynomial of T. A scalar λ is an eigenvalue of T if and only if p( λ) = 0. Hence the
characteristic polynomial and the minimal polynomial of T have the same zeros.

Proof : Let f(t) be the characteristic polynomial of T. Since p(t) divides f(t), there exist

polynomials q(t) and r(t) such that f(t) = q(t) p(t) .

If λ is a zero of p(t) then f(λ) = q(λ) p(λ) = q(λ).0 = 0. So λ is a zero of f(t) i.e. λ is an

eigenvalue of T.

Conversely, suppose λ is an eigenvalue of T and let x ∈ V be an eigenvector

corresponding to λ. Then T(x) = λ x and p(T)(x) = p(λ)x. As p(T) = O

so 0 = O(x) = p(T)(x) = p(λ)x . Since x being an eigenvector is nonzero, it follows that

p(λ ) = 0 and so λ is a zero of p(t) also.

Corollary : Let T be a linear operator on a FDVS V with minimal polynomial p(t) and
characteristic polynomial f(t). Suppose that f(t) factors as

f(t) = (λ1 – t)^ n1 (λ2 – t)^ n2 ……….(λ k – t)^ nk

where λ1 ,λ2, ………,.,λ k are the distinct eigenvalues of T. Then there exist integers
m1 , m2, …… mk such that 1 ≤ mi ≤ ni for all i = 1, 2,…..k, and

p(t) = (λ1 – t)^ m1 (λ2 – t)^ m2 ………. (λk – t)^ mk

Example 1 : Compute the minimal Polynomial of the matrix

3 −1 0
A=(0 2 0)
1 −1 2
Sol. : The characteristic polynomial of A is det (A – tI)
 3−𝑡 −1 0
i.e. det ( 0 2−𝑡 0 ) i.e. f(t) = - (t – 2)2 (t – 3)
1 −1 2−𝑡
As the minimal polynomial p(t) divides f(t) and they have the same zeros, so possibilities
for p(t) are

(t – 2)(t – 3) or (t – 2) 2 (t – 3)

If we let p(t) = (t – 2)(t – 3) then p(A) = (A – 2I)(A – 3I) = A2 – 5A + 6I = O.

Thus p(t) is a polynomial of least degree satisfying p(A) = O, so the minimal polynomial

of A is p(t) = (t – 2)(t – 3).

Example 2 : Let T be the linear operator on R 2 defined by T(a, b) = ( 2a+5b, 6a+b ).

2 5
Sol. : Let β be standard ordered basis for R 2. Then [T]β = ( ). The characteristic
6 1
2−𝑡 5
polynomial of T is f(t) = | | = (t - 7) (t + 4) .
6 1−𝑡
As characteristic polynomial and minimal polynomial have the same zeros,

therefore the minimal polynomial is also (t - 7) (t + 4) .

Example 3. : Let D be the linear operator on P 2 (R) defined by D(g(x)) = g‫(׳‬x) .

Compute the minimal polynomial of D.

0 1 0
Sol. : Let β = { 1, x, x2 } be standard ordered basis for P2 (R). Then [D]β = (0 0 2) .
0 0 0
The characteristic polynomial of D is - t3.

The possibilities for minimal polynomial are t , t 2 , t3 .

As D ≠ O and D2 ≠ O but D 3 = O . Therefore the minimal polynomial of D is t3.

Theorem 7.15 : Let T be a linear operator on an n – dimensional vector space V such

Proof : Since V is a T – cyclic space , Ǝ a nonzero x ϵ V such that

β = { x, T(x), ….,T n-1(x) } is a basis for V . ( By Theorem 5.22 ).

Let g(t) = a0 + a1 t +……..+ ak tk be a polynomial of degree k < n . Then ak ≠ 0

and g(T)(x) = a0 x + a1 T(x) +……..+ ak Tk(x) .

Observe that g(T)(x) is a linear combination of elements of β such that ak ≠ 0 .

Then g(T)(x) ≠ 0. ( Since β is linearly independent ) . Hence g(T) ≠ O .

Therefore, the minimal polynomial of T has degree n which is also the degree of

the characteristic polynomial. Hence, f(t) = (- 1) n p(t).

Theorem 7.16 : Let T be a linear operator on a FDVS V. Then T is diagonalizable

if and only if the minimal polynomial of T is of the form

p(t) = ( t – λ1 ) ( t – λ2 )…………. ( t – λ k )

where λ1 , λ2 ,…… λk are the distinct eigenvalues of T .

Proof : Not to be done.

Example 4 : Determine all matrices A ϵ M2×2(R) for which A2 – 3A + 2I = O.

Sol. : Let g(t) = t2 - 3t + 2 = ( t – 1) ( t – 2 ) .

As g(A) = A2 – 3A + 2I = O , the minimal polynomial p(t) of A divides g(t).

Hence the possibilities for p(t) are t – 1 or t – 2 or ( t – 1) ( t – 2 ).

If p(t) = t – 1 then A = I satisfies the given condition.

If p(t) = t – 2 then A = 2I satisfies the given condition.

If p(t) = ( t – 1) ( t – 2 ) then A is diagonalizable with eigenvalues 1 and 2

1 0
and hence A is similar to ( ).
0 2
Example 5 : Let A ϵ Mn×n(R) satisfy A3 = A . Show that A is diagonalizable .

Sol. : Let g(t) = t3 - t = t ( t – 1) ( t +1 ) . Then g(A) = O and hence the minimal

polynomial p(t) of A divides g(t). Hence the possibilities for p(t) are t, t – 1 or t + 1

or t ( t – 1) or t ( t + 1 ) or ( t – 1) ( t + 1) or t ( t – 1) ( t +1 ) . In each of these

possibilities p(t) has no repeated factors . Thus A is diagonalizable ( by Theorem 7.16 ).

Example 6 : In example 3 we have seen that the minimal polynomial of differential

operator D is t3. Thus it has repeated zeros. So by Theorem 7.16, D is not

diagonalizable.

Assignment : Do questions 5, 8 and 10 from exercise 7.3 .