What Is a Complex Number?
Solve x^2 + 1 = 0 and you hit a wall. Rearranging gives
x^2 = -1, and no real number squares to a negative — squaring
always lands on zero or above. The
quadratic formula
runs into the same wall whenever the discriminant goes negative: it asks for the square
root of a negative number, which the real line simply does not contain.
So we do the boldest thing in algebra — we invent the missing number.
Define a brand-new symbol i, the imaginary unit,
by the single defining property
i^2 = -1, \qquad \text{equivalently} \qquad i = \sqrt{-1}.
That one rule is the whole foundation. Everything else follows from ordinary algebra with
i^2 rewritten as -1 wherever it appears.
Solving the impossible equation
Watch x^2 + 1 = 0 dissolve, one line at a time.
Step 1 — isolate the square. Subtract 1 from both
sides:
x^2 = -1.
Step 2 — take the square root. A square root comes with both a positive and
a negative branch, so we keep the \pm:
x = \pm\sqrt{-1}.
Step 3 — name the new number. By definition \sqrt{-1} = i,
so the two solutions are simply
x = i \qquad \text{and} \qquad x = -i.
Step 4 — check. Substituting back, i^2 + 1 = -1 + 1 = 0.
The equation that had no real answer has exactly two answers — once you allow i.
A complex number is a real part plus an imaginary part
Mix a real number with a multiple of i and you get a
complex number:
z = a + bi.
Here a is the real part, written
\operatorname{Re}(z), and b is the
imaginary part, written \operatorname{Im}(z)
(note: the imaginary part is the real number b, not
bi). Ordinary real numbers are just the special case
b = 0, so the reals sit comfortably inside the complex numbers.
Arithmetic: it's just algebra with i² = −1
Adding — combine like with like
Step 1 — group real with real, imaginary with imaginary. Addition is
componentwise; the i terms gather together just like a common
factor:
(a + bi) + (c + di) = (a + c) + (b + d)i.
For example (3 + 2i) + (1 + 4i) = 4 + 6i — add the reals, add the
imaginaries, done.
Multiplying — expand, then use i² = −1
Step 1 — expand the brackets exactly as you would for two real binomials
(every term times every term):
(a + bi)(c + di) = ac + adi + bci + bd\,i^2.
Step 2 — replace i^2 with -1.
This is the only new move; the last term flips sign:
= ac + adi + bci - bd.
Step 3 — collect real and imaginary parts. Group the two real terms and the
two imaginary terms:
(a + bi)(c + di) = (ac - bd) + (ad + bc)i.
For example (3 + 2i)(1 + 4i) = (3 - 8) + (12 + 2)i = -5 + 14i.
The conjugate cleans things up
Flip the sign of the imaginary part and you get the complex conjugate,
written \bar{z}:
\overline{a + bi} = a - bi.
Step 1 — multiply a number by its conjugate. Using the multiplication rule
with c = a and d = -b:
z \bar{z} = (a + bi)(a - bi) = a^2 - (bi)^2 = a^2 - b^2 i^2.
Step 2 — apply i^2 = -1. The cross terms
+abi and -abi cancelled, and the last
term becomes +b^2:
z\bar{z} = a^2 + b^2.
The imaginary part vanished — z\bar{z} is always a non-negative
real number. That single fact is what makes division by a complex number work
(multiply top and bottom by the conjugate to clear the i from the
denominator), and it is the seed of the
modulus
we meet next.
Adjoin a symbol i with i^2 = -1 to the
real numbers. Then:
-
the defining rule is i^2 = -1 (so i = \sqrt{-1});
-
every complex number has the form z = a + bi with real part
a and imaginary part b;
-
addition is componentwise: (a + bi) + (c + di) = (a + c) + (b + d)i;
-
multiplication expands and uses i^2 = -1:
(a + bi)(c + di) = (ac - bd) + (ad + bc)i;
-
the conjugate \bar{z} = a - bi satisfies
z\bar{z} = a^2 + b^2 \ge 0, always real.
The quadratic
formula solves ax^2 + bx + c = 0 with
x = \frac{-b \pm \sqrt{b^2 - 4ac}}{2a}.
When the discriminant b^2 - 4ac is negative there are no real
roots — but there are two complex ones. Take x^2 - 2x + 5 = 0,
which you could also reach by
completing the
square to (x - 1)^2 = -4. Here
b^2 - 4ac = 4 - 20 = -16, so
x = \frac{2 \pm \sqrt{-16}}{2} = \frac{2 \pm 4i}{2} = 1 \pm 2i.
The two roots are 1 + 2i and 1 - 2i —
a conjugate pair, as complex roots of a real polynomial always are. With
i in hand, every quadratic has exactly two roots.
See a product computed
Slide the four coefficients of z_1 = a + bi and
z_2 = c + di. The diagram works the multiplication rule live:
the real part ac - bd and the imaginary part
ad + bc update as you move the knobs, so you can watch
i^2 = -1 doing its job in the -bd term.