Concept:
Matrix multiplication is fundamentally different from real number algebra because matrix multiplication is generally non-commutative, meaning $AB \neq BA$ in most cases. We must preserve order when expanding algebraic terms.
Step 1: Analyze statement (i).
Let us expand the left-hand side expansion keeping multiplication order strict:
\[
(A + B)(A - B) = A(A - B) + B(A - B) = A^2 - AB + BA - B^2
\]
For this to equal $A^2 - B^2$, the terms $-AB + BA$ must cancel out, which requires $AB = BA$. Since matrix multiplication is non-commutative, this statement is not always true.
Step 2: Analyze statement (ii).
As established, matrix multiplication is non-commutative in general ($AB \neq BA$). There are specific matrices that commute, but it is not always true for any arbitrary square matrices.
Step 3: Analyze statement (iii).
Let us expand $(A + B)^2$:
\[
(A + B)^2 = (A + B)(A + B) = A(A + B) + B(A + B) = A^2 + AB + BA + B^2
\]
This algebraic expansion perfectly preserves the multiplication order without assuming commutativity. Therefore, this statement is always true.
Step 4: Analyze statement (iv).
In matrix algebra, it is possible to have two non-zero matrices whose product is a zero matrix (these are called divisors of zero). For example:
Let $A = \begin{bmatrix} 0 & 1
0 & 0 \end{bmatrix} \neq 0$ and $B = \begin{bmatrix} 1 & 0
0 & 0 \end{bmatrix} \neq 0$.
Their product is $AB = \begin{bmatrix} 0 & 0
0 & 0 \end{bmatrix} = 0$.
Hence, $AB = 0$ does not necessarily imply $A=0$ or $B=0$. This statement is not always true.
Thus, only statement (iii) is true, matching option (C).