← Back to all lessons
Class 12Algebra9:15Published 9 Jun 2026

Matrix Identities: Finding a Value and Proving a Cubic Relation

Two worked matrix problems: find the unknown in the relation A squared equals kA minus 2I for a 2 by 2 matrix, then prove a 3 by 3 matrix satisfies A cubed minus 23A minus 40I equals the zero matrix.

This lesson works through two exercise questions on matrix powers. First it computes the square of a 2 by 2 matrix and compares it term by term against kA minus 2I to solve for k. Then it builds up the square and cube of a 3 by 3 matrix using the row into column rule, and substitutes them to show that A cubed minus 23A minus 40I gives the zero matrix. Each step is shown so you can follow the multiplication and the final subtraction.

What you'll learn

  • How to multiply two matrices using the row into column rule
  • How to find an unknown constant by comparing matrices entry by entry
  • How to build a matrix cube from its square, then check a cubic matrix identity reduces to the zero matrix

Lesson chapters

0:08Question 17: solving for k in A squared equals kA minus 2I
0:21Computing A squared for the 2 by 2 matrix
1:09Comparing entries to find k
3:04Second example: the 3 by 3 matrix identity
3:35Finding the square of the 3 by 3 matrix
4:55Finding the cube of the matrix
7:00Substituting to prove the result is zero

Lesson notes

This lesson solves two exercise questions on matrix powers. The first finds an unknown constant by comparing two matrices entry by entry, and the second proves a 3 by 3 matrix satisfies a cubic identity.

Question 17: solving for kk

We are given

A=[3242],I=[1001]A = \begin{bmatrix} 3 & -2 \\ 4 & -2 \end{bmatrix}, \qquad I = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}

and told that A2=kA2IA^2 = kA - 2I. We need the value of kk.

Step 1: compute A2A^2

Multiply AA by itself using the row into column rule:

A2=[3242][3242]=[986+41288+4]=[1244]A^2 = \begin{bmatrix} 3 & -2 \\ 4 & -2 \end{bmatrix}\begin{bmatrix} 3 & -2 \\ 4 & -2 \end{bmatrix} = \begin{bmatrix} 9 - 8 & -6 + 4 \\ 12 - 8 & -8 + 4 \end{bmatrix} = \begin{bmatrix} 1 & -2 \\ 4 & -4 \end{bmatrix}

Step 2: build kA2IkA - 2I

kA2I=k[3242]2[1001]=[3k22k4k2k2]kA - 2I = k\begin{bmatrix} 3 & -2 \\ 4 & -2 \end{bmatrix} - 2\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} = \begin{bmatrix} 3k - 2 & -2k \\ 4k & -2k - 2 \end{bmatrix}

Step 3: compare entries

Setting A2=kA2IA^2 = kA - 2I and matching corresponding entries:

[1244]=[3k22k4k2k2]\begin{bmatrix} 1 & -2 \\ 4 & -4 \end{bmatrix} = \begin{bmatrix} 3k - 2 & -2k \\ 4k & -2k - 2 \end{bmatrix}

From the top left, 3k2=13k - 2 = 1, so 3k=33k = 3 and k=1k = 1. The other entries confirm it: 2k=2-2k = -2, 4k=44k = 4, and 2k2=4-2k - 2 = -4 all give k=1k = 1.

k=1k = 1

Second example: a 3 by 3 identity

Given

A=[123321421],A = \begin{bmatrix} 1 & 2 & 3 \\ 3 & -2 & 1 \\ 4 & 2 & 1 \end{bmatrix},

show that A323A40I=0A^3 - 23A - 40I = 0.

Step 1: find A2A^2

A2=AA=[1948112814615]A^2 = A \cdot A = \begin{bmatrix} 19 & 4 & 8 \\ 1 & 12 & 8 \\ 14 & 6 & 15 \end{bmatrix}

Be careful to copy each entry exactly when you carry it forward.

Step 2: find A3A^3

Multiply AA by A2A^2:

A3=AA2=[123321421][1948112814615]=[63466969623924663]A^3 = A \cdot A^2 = \begin{bmatrix} 1 & 2 & 3 \\ 3 & -2 & 1 \\ 4 & 2 & 1 \end{bmatrix}\begin{bmatrix} 19 & 4 & 8 \\ 1 & 12 & 8 \\ 14 & 6 & 15 \end{bmatrix} = \begin{bmatrix} 63 & 46 & 69 \\ 69 & -6 & 23 \\ 92 & 46 & 63 \end{bmatrix}

Step 3: substitute and subtract

Now form A323A40IA^3 - 23A - 40I:

[63466969623924663]23[123321421]40[100010001]\begin{bmatrix} 63 & 46 & 69 \\ 69 & -6 & 23 \\ 92 & 46 & 63 \end{bmatrix} - 23\begin{bmatrix} 1 & 2 & 3 \\ 3 & -2 & 1 \\ 4 & 2 & 1 \end{bmatrix} - 40\begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix}

The scaled matrices are

23A=[234669694623924623],40I=[400004000040]23A = \begin{bmatrix} 23 & 46 & 69 \\ 69 & -46 & 23 \\ 92 & 46 & 23 \end{bmatrix}, \qquad 40I = \begin{bmatrix} 40 & 0 & 0 \\ 0 & 40 & 0 \\ 0 & 0 & 40 \end{bmatrix}

Subtracting entry by entry, for example the top left is 632340=063 - 23 - 40 = 0, and every other entry also cancels:

A323A40I=[000000000]A^3 - 23A - 40I = \begin{bmatrix} 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end{bmatrix}

Hence proved.

Key takeaways

  • To find an unknown in a matrix equation, compute each side and match corresponding entries; any one matching entry gives the value, and the rest confirm it.
  • Multiply matrices with the row into column rule, and copy entries carefully when reusing a result.
  • Build a cube as AA2A \cdot A^2, then substitute into the identity; if the matrix satisfies the relation, all entries subtract to the zero matrix.