Complex numbers as points in a plane, and the polar description of those points. Defines the modulus as the distance from the origin and the argument as the set of angles that reach the number, singling out the principal argument in the interval from negative pi to pi. Establishes the properties of both, then assembles the polar form: the modulus times cosine theta plus i sine theta, abbreviated cis, and converts in both directions.
Subject: Trigonometry · 66 slides · symbolic lesson
Open the interactive version of this deck
Title
Trigonometry · Section 11.7
Giving every complex number a distance and a direction
Objectives
Complex numbers arrived in Section 3.4 as an algebraic device for solving equations. This lesson gives them a home in the plane, and then gives that plane a polar address system.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 991-997 — learning goals for this half of the section
You have met complex numbers twice before: in Intermediate Algebra as a way of writing down a square root of a negative number, and in Section 3.4 as the numbers that make every polynomial factor. In both places they were symbols you manipulated, with no picture attached.
That changes here. A complex number carries two real numbers, a real part and an imaginary part, and two real numbers locate a point in a plane. Once the number is a point, everything you learned about polar coordinates in Section 11.4 applies to it, and the distance and the angle get new names: the modulus and the argument.
The pay-off comes at the end of the lesson, when the two names are packed back into a single expression called a polar form. Almost nothing here is new mathematics. It is the polar coordinates you already know, applied to a new set of objects, with a careful convention added for choosing one angle out of infinitely many.
Warm-up
Discussion prompt
Section 11.4 gave every point in the plane a polar pair: a distance out and an angle round. Write down, without looking back, the two equations that take a polar pair to rectangular coordinates, and the two facts that take rectangular coordinates back to a polar pair.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 991 — the section opens by recalling Section 11.4
Write your four equations down before revealing anything. If you cannot produce them from memory, go back to Section 11.4 for five minutes before continuing, because this whole lesson leans on them.
Two of the equations are safe and two are treacherous. Going from polar to rectangular is completely safe: the cosine and the sine already know which quadrant the angle points into, so their signs come out right without any thought from you.
Going the other way is where care is needed. The equation for the distance has two solutions, one positive and one negative, and the equation for the tangent has infinitely many, half of them pointing in exactly the wrong direction. Each of those two difficulties turns into a definition in this lesson: one is settled by insisting the distance be non-negative, the other by choosing a standard interval for the angle.
Section
Part 1
Concept
A complex number carries two real numbers inside it, and two real numbers are exactly what it takes to locate a point in a plane.
\[ z = a + bi \quad \longleftrightarrow \quad (a, b), \qquad a = \operatorname{Re}(z), \; b = \operatorname{Im}(z) \]
Figure (svg): A complex plane with the horizontal axis labelled Re and the vertical axis labelled Im, marked in multiples of i. Three points are plotted: minus 4 plus 2i up and to the left, minus 3i straight down on the vertical axis, and 3 out along the horizontal axis.
The horizontal axis is relabelled the real axis and the vertical axis the imaginary axis, marked off in multiples of the imaginary unit. The plane these two axes determine is called the complex plane.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 991 — the complex plane
The correspondence in the middle line is the whole idea, and it is worth reading slowly in both directions. Given a number, take its real part across and its imaginary part up. Given a point, read its horizontal coordinate as the real part and its vertical coordinate as the coefficient of the imaginary unit.
Look at the three plotted numbers in the figure. The number with a nonzero part in each direction sits off both axes. The purely imaginary one sits on the vertical axis, three units down. The real number 3 sits on the horizontal axis, and this is the important special case: the real numbers, which you have always pictured on a line, are exactly the horizontal axis of this plane.
Notice the labelling of the vertical axis. It is marked in multiples of the imaginary unit rather than in plain numbers, because a point one unit up is the number i, not the number 1. That habit will save you from sign and factor errors later.
Concept
Because the ordered pair of real and imaginary parts is exactly a pair of rectangular coordinates, the expression a plus bi is called the rectangular form of the number.
Two complex numbers are equal precisely when their real parts agree and their imaginary parts agree. That is what makes each part a function of the number rather than an artefact of how it happens to be written.
\[ a + bi = c + di \;\Longleftrightarrow\; a = c \text{ and } b = d \]
well-defined — No matter how a complex number is written down, its real part is always the same number and its imaginary part is always the same number.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 991 — Re and Im are well-defined
The word rectangular is not decoration. The pair of parts is literally a pair of rectangular coordinates, which is why the expression a plus bi earns the same adjective. Later in the lesson the same number will get a polar form, and the two names will sit side by side exactly as they did for points.
The equality statement deserves more attention than it usually gets. It says that a complex number can be split into its two parts in only one way. If that failed, the real part would depend on how you happened to write the number down, and the whole picture would collapse, because a point would have two different addresses.
In practice this matters when you are asked for the parts of something like 3i or negative 117. You have to write the missing part in explicitly, as zero, before reading anything off. Do that on paper every time until it becomes automatic; it is the single most common place to lose a mark in this section.
Worked example
Four numbers, and two of them need rewriting before anything can be read off.
\[ z_1 = \sqrt{3} - i, \quad z_2 = -2 + 4i, \quad z_3 = 3i, \quad z_4 = -117 \]
Put the first in standard form
Why: Make the coefficient of i explicit, including its sign.
\[ \sqrt{3} - i = \sqrt{3} + (-1)i, \quad \operatorname{Re} = \sqrt{3}, \; \operatorname{Im} = -1 \]
Read the second directly
Why: It is already written as a real part plus a real multiple of i.
\[ -2 + 4i: \quad \operatorname{Re} = -2, \; \operatorname{Im} = 4 \]
Supply the missing real part of the third
Why: A purely imaginary number has real part zero.
\[ 3i = 0 + 3i, \quad \operatorname{Re} = 0, \; \operatorname{Im} = 3 \]
Supply the missing imaginary part of the fourth
Why: A real number has imaginary part zero.
\[ -117 = -117 + 0i, \quad \operatorname{Re} = -117, \; \operatorname{Im} = 0 \]
Check that every imaginary part is a real number
Why: The imaginary part is the coefficient of i, not the term containing i. For the third number that makes the imaginary part 3, not 3i, and 3 is a perfectly ordinary real number.
\[ \operatorname{Im}(3i) = 3 \ne 3i \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 992-993 — Example 11.7.1, the Re and Im of each part
This is the easy half of the example, and the whole difficulty is bookkeeping. Two of the four numbers arrive with a part missing and one arrives with a hidden minus sign, so all three have to be rewritten before anything can be read.
Watch what happens with the first number. Written as it is, it looks as though the imaginary part might be i, or negative i. Rewriting it so that the coefficient of the imaginary unit is explicit settles it: the imaginary part is negative one.
The check at the end is the point of the slide. The imaginary part is a real number, always. It is the coefficient sitting in front of the imaginary unit, not the term that contains it. If you find yourself writing an answer with an i in it for an imaginary part, you have made this error, and the plotting picture is the cure: the imaginary part is a height, and a height is an ordinary real number.
Sorting
Sort into buckets
Place each complex number by looking only at the signs of its two parts.
Do this one by the signs alone, with no arithmetic. The sign of the real part decides left or right and the sign of the imaginary part decides up or down, so the quadrant falls out of two glances.
Two of the six have a part equal to zero, and those are the ones worth slowing down on. A number with no real part cannot be left or right of the vertical axis, so it lies on it; a number with no imaginary part lies on the horizontal axis. Those two cases will come back very shortly as the numbers whose arguments cannot be found from a tangent.
The last two items look complicated because of the radicals, but the radicals change nothing: the root of three is a positive number and the root of two is a positive number, so only the signs written in front of them matter.
Picture it
Figure (svg): Two panels. The left panel is a complex plane holding three of the four numbers: root 3 minus i in the fourth quadrant, minus 2 plus 4i in the second, and 3i on the imaginary axis. The right panel is the real axis alone with a break drawn in it, so that minus 117 can be shown far out to the left of the small window that holds the other three.
Three of them fit in one window. The fourth is 117 units out along the negative real axis, so the second panel breaks the axis rather than pretending the scale is uniform, which is exactly what the book does.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 993 — the plot at the end of Example 11.7.1
Spend a moment on the honesty of this picture. Three of the four numbers fit comfortably in a window a few units wide. The fourth is 117 units away, which is about thirty times further than anything else in the example.
You cannot draw both at one scale on one page, so the book breaks the axis, and so does the right-hand panel here. The two small slashes are the standard signal that a stretch of the axis has been cut out. Everything to the left of the break is drawn at a completely different scale from everything to the right.
Why bother? Because the plot is not decoration in this section; it is how you decide the quadrant, and the quadrant is what the algebra cannot give you. A rough sketch with the axes labelled and the point in roughly the right place is enough for every problem in the exercise set, and it is worth drawing every single time.
Trap
A very common line, written while reading off the parts:
\[ \operatorname{Im}(3i) = 3i \]
\[ \operatorname{Im}(-2 + 4i) = 4i \]
Wrong in both places.
The imaginary part is the real coefficient sitting in front of the imaginary unit. The name is historical and misleading: the imaginary part of a complex number is a real number.
\[ \operatorname{Im}(3i) = 3, \qquad \operatorname{Im}(-2+4i) = 4 \]
The test is the plotting: the imaginary part is how far up the point sits, and a height is a real number.
The name is the whole problem. Imaginary part sounds as though it should be the imaginary piece of the number, which is the term containing the imaginary unit. It is not: it is the real coefficient multiplying that unit.
The plotting picture is the reliable defence. The imaginary part is how far up the point sits, and a height on a graph is an ordinary real number. Nobody would say a point is 4i units above the axis.
There is a practical reason this matters beyond marks. Every formula in this lesson takes the imaginary part and squares it or divides by it, and those operations only make sense for real numbers. Feed 4i into the modulus formula in place of 4 and you will get the wrong answer, because the square of 4i is negative sixteen rather than sixteen.
Section
Part 2
Concept
Plot the number, then describe its point in polar coordinates instead of rectangular ones, insisting that the distance be non-negative. The two entries of that polar pair get their own names.
Figure (svg): A complex plane with one point marked in the first quadrant. An arrow runs from the origin to the point; dashed lines drop from the point to each axis, marking the real part along the horizontal and the imaginary part up the vertical. A curved arrow near the origin sweeps from the positive real axis up to the arrow, labelled theta.
\[ z = a + bi \;\longleftrightarrow\; (a, b) \;\longleftrightarrow\; (r, \theta), \quad r \ge 0 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 991-992 — Definition 11.2
Read the double correspondence carefully. The number gives a pair of rectangular coordinates, and that point has polar coordinates too. All Definition 11.2 does is give the two polar entries names that belong to the number rather than to the point.
The figure is worth keeping in your head for the rest of the section. Four quantities are marked on it: the real part along the bottom, the imaginary part up the side, the length of the arrow, and the angle the arrow makes with the positive real axis. The first two you read off directly; the last two are what you have to compute.
The one condition attached is that the distance be non-negative, and it is doing real work. In Section 11.4 you were allowed a negative first coordinate, which meant walk backwards along the ray. That freedom is deliberately given up here, and the next slide explains what is bought with it.
Notation
Annotate
On: \( |z| = r, \quad \theta \in \arg(z), \quad \operatorname{Arg}(z) \in (-\pi, \pi] \)
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 991 — Definition 11.2
Take these four annotations one at a time, because three different notations are being introduced at once and they are easy to blur together.
The bars mean distance and nothing else. Lower-case arg is a set, so an answer to a question about it should look like a set, written with braces. Capital-A Arg is a single angle, so an answer to a question about it should be one number with no letter k anywhere in it. The difference between the third and fourth bullets is just a capital letter, which is exactly why this section trips people up in writing rather than in thinking.
The last annotation is the fussiest and the most useful. The interval is open at the left and closed at the right, so a half turn is recorded as positive pi and never as negative pi. That decision looks arbitrary until you meet a negative real number, where it is the only thing standing between you and two different correct answers.
Concept
Why is the modulus a single number? Because for a point away from the origin there are exactly two possible distances in polar coordinates, one positive and one negative, and the definition throws the negative one away.
\[ r^2 = a^2 + b^2 \;\Longrightarrow\; r = \pm\sqrt{a^2+b^2}, \quad r \ge 0 \;\Longrightarrow\; r = \sqrt{a^2+b^2} \]
Figure (svg): A complex plane with the single point root 3 minus i marked in the fourth quadrant and an arrow drawn to it. Three curved arrows sweep from the positive real axis to that arrow: a short clockwise one, a long counter-clockwise one almost all the way round, and a spiral that goes clockwise more than a full turn.
Nothing similar pins down the angle. Every angle that reaches the point is coterminal with every other, so they differ by whole turns, and there are infinitely many of them.
\[ \arg(z) = \left\{ \operatorname{Arg}(z) + 2\pi k \;\middle|\; k \text{ an integer} \right\} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — the remarks after Definition 11.2
The first line of algebra is the whole reason the modulus is well behaved. Squaring and adding the two parts gives an equation for the distance, and an equation like that always has two solutions, one positive and one negative. The definition simply throws the negative one away, and what is left is one number.
Now look at the picture and try to do anything similar for the angle. Three arcs are drawn, all ending at the same point: a short clockwise one, a long counter-clockwise one, and a spiral that goes more than a full turn. Every one of them is a legitimate argument. There is no natural rule that prefers one of them, because they all genuinely reach the number.
So the modulus is a number and the argument is a set, and that asymmetry is permanent. The set has a very simple shape, though: pick any one argument and every other is that one plus a whole number of turns, which is what the last line records.
Concept
Coterminal angles are exactly two pi apart, so an interval two pi long contains exactly one of them. The convention picks the interval from negative pi to pi, excluding the left end and including the right.
Figure (svg): A number line of angles from about minus 7.6 to 7.6, with a shaded band from minus pi to pi. Three dots mark the arguments of one complex number: minus 6.81, minus 0.52 and 5.76. Only the middle dot lies inside the shaded band. The left end of the band carries an open circle and the right end a filled circle.
Excluding one end is what stops a half turn from having two names. Without it, both negative pi and pi would qualify, and the principal argument would not be a single value after all.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — why only one angle lies in the interval
The set of arguments is drawn here as what it really is: an evenly spaced row of numbers on an angle line, two pi apart. The shaded band is the window that the convention chooses, and you can see immediately that exactly one dot falls inside it.
Why must it be exactly one? Because the dots are spaced two pi apart and the window is two pi long. A window that long always catches at least one dot, and if both ends counted it could occasionally catch two. Excluding one end removes that last possibility.
Which end gets excluded is a pure convention, and this book excludes the left one. That means an angle of exactly a half turn is reported as positive pi. Different books and different calculators sometimes make the other choice, so when you meet a principal argument outside this course, check the interval before trusting the sign.
Worked example
Find the modulus, the set of arguments and the principal argument.
\[ z = \sqrt{3} - i, \quad P(\sqrt{3}, -1) \]
Square the two parts and add
Why: This is the polar relation from Section 11.4.
\[ r^2 = (\sqrt{3})^2 + (-1)^2 = 3 + 1 = 4 \]
Take the non-negative root
Why: Both signs solve the equation; the definition keeps the non-negative one.
\[ r = \pm 2, \quad r \ge 0 \;\Longrightarrow\; |z| = 2 \]
Find the tangent of the angle
Why: Imaginary part over real part, then rationalise.
\[ \tan(\theta) = \frac{-1}{\sqrt{3}} = -\frac{\sqrt{3}}{3} \]
Place the angle in the right quadrant
Why: The real part is positive and the imaginary part negative, so the point is in Quadrant IV.
\[ \theta = -\frac{\pi}{6} + 2\pi k \]
Figure (svg): The number root 3 minus i in the fourth quadrant of the complex plane, with an arrow from the origin, the horizontal leg of length root 3 and the vertical leg of length 1 drawn dashed, a right-angle marker, and a curved arrow sweeping clockwise from the positive real axis through an angle of minus pi over 6.
List the set, then pick the principal one
Why: Add every whole turn to get the set; keep the member inside the window.
\[ \arg(z) = \left\{-\tfrac{\pi}{6} + 2\pi k \;\middle|\; k \in \mathbb{Z}\right\}, \quad \operatorname{Arg}(z) = -\tfrac{\pi}{6} \]
Check the principal value is the only one in range
Why: The neighbours in the set are eleven pi over six, about 5.76, and negative thirteen pi over six, about negative 6.81. Both sit far outside the interval, and negative pi over six, about negative 0.52, is comfortably inside it.
\[ -\pi < -\tfrac{\pi}{6} \le \pi, \qquad \tfrac{11\pi}{6} \approx 5.76 > \pi \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — Example 11.7.1, part 1
This is the model solution for the whole exercise set, so learn its shape rather than its numbers. Modulus first, because it never depends on the quadrant. Then the tangent, then the quadrant, then the set, then the one principal value.
The step people skip is the fourth. The tangent equation is satisfied both by an angle in the fourth quadrant and by one in the second, and nothing in the equation itself chooses between them. The signs of the two parts do: a positive real part with a negative imaginary part can only be down and to the right.
Look at the figure while you read the steps. The two legs are the real and imaginary parts, the hypotenuse is the modulus, and the reference triangle has sides root three, one and two, which you should recognise instantly as the thirty-sixty-ninety triangle. That recognition is what makes the angle exact rather than a decimal.
Fill the middle
Exercise 13 of the section asks for the polar data of this number. Start with its distance from the origin.
Fill in the blanks
|3 + 4i| = \sqrt4}^2} = \sqrt25} = 5
Why: Nine plus sixteen is twenty-five, whose square root is five. Both parts are squared, so the signs of the parts never matter to the modulus: the numbers 3 + 4i, 3 − 4i, −3 + 4i and −3 − 4i all have modulus 5 and sit on one circle.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercise 13
Fill the three blanks before checking. Everything here is squaring and adding, and the only thing to be careful about is that both parts get squared, including any minus sign.
That last point has a nice consequence, mentioned in the explanation. Changing the sign of either part does not change the modulus at all. The four numbers you can build from parts of size three and four all sit at distance five from the origin, one in each quadrant, on a single circle of radius five.
This is also why the modulus can never tell you the quadrant. It has thrown that information away deliberately, which is precisely why the argument has to be computed separately.
Worked example
The same procedure, but this time the angle has no exact value in the table.
\[ z = -2 + 4i, \quad P(-2, 4) \]
Compute the modulus
Why: Four plus sixteen is twenty, and twenty is four times five.
\[ r = \sqrt{(-2)^2 + 4^2} = \sqrt{20} = 2\sqrt{5} \approx 4.47 \]
Compute the tangent
Why: Imaginary part over real part.
\[ \tan(\theta) = \frac{4}{-2} = -2 \]
Read what the arctangent gives
Why: The arctangent always returns an angle between negative pi over two and pi over two, so it lands in Quadrant IV.
\[ \arctan(-2) \approx -1.107 \]
Move it to Quadrant II
Why: The point has a negative real part and a positive imaginary part, so add a half turn.
\[ \theta = \pi + \arctan(-2) = \pi - \arctan(2) \approx 2.034 \]
Figure (svg): The number minus 2 plus 4i in the second quadrant, with an arrow from the origin, dashed legs of length 2 to the left and 4 up, and two curved arrows: a large one from the positive real axis counter-clockwise to the arrow, and a smaller one from the negative real axis down to the arrow, marking the reference angle arctan 2.
Record the set and the principal value
Why: Whole turns again; only the one value lies in the window.
\[ \arg(z) = \{\pi - \arctan(2) + 2\pi k\}, \quad \operatorname{Arg}(z) = \pi - \arctan(2) \]
Check the answer lands where the point is
Why: Put the angle back through cosine and sine at radius two root five: the result is the original number, so the angle and the modulus agree with the point.
\[ 2\sqrt5\cos(2.034) \approx -2, \qquad 2\sqrt5\sin(2.034) \approx 4 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — Example 11.7.1, part 2
Most numbers in the world give an angle that is not in any table, and this example shows what an exact answer looks like then. You leave the arctangent in the answer. Writing about 2.03 is a decimal approximation, not an exact value, and both are worth being able to produce.
The middle steps are the ones to study. The arctangent of negative two is a Quadrant IV angle, because the arctangent function is built to return only angles between negative a quarter turn and a quarter turn. Our point is in Quadrant II. Adding a half turn moves the answer from one to the other without changing its tangent, and that is the standard repair.
The final two forms of the answer are the same angle written two ways: a half turn plus the arctangent of negative two, or a half turn minus the arctangent of two. The second is tidier because the arctangent of a positive number is the reference angle, which the picture shows directly.
Error analysis
Annotate
On: \( \tan(\theta) = -2 \;\Longrightarrow\; \operatorname{Arg}(-2+4i) = \arctan(-2) \approx -1.107 \)
\[ -1.107 \text{ points into Quadrant IV}, \quad -2+4i \text{ is in Quadrant II} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — Theorem 11.15 and the quadrant discussion
Read the line and try to locate the error yourself before opening the annotations. The tempting thought is that the mistake is arithmetic, and it is not: the value of the arctangent is correct to three places.
The error is one of logic. The tangent equation is true of every argument of this number, but it is also true of every angle pointing in the exactly opposite direction, because tangent repeats every half turn rather than every full turn. So the equation alone can never finish the job.
The repair is mechanical once you see it. Plot the point, name the quadrant, and if the real part is negative add a half turn to whatever the arctangent gave you. If the result then falls outside the window from negative pi to pi, subtract a full turn to bring it back in. Do those three checks in that order and this error cannot survive.
Worked example
Some numbers need no calculation at all.
\[ z = 3i = 0 + 3i, \quad P(0, 3) \]
Locate the point
Why: Zero real part, so it is on the imaginary axis, three units up.
\[ P(0,3) \text{ lies on the positive imaginary axis} \]
Read the distance
Why: Distance from the origin along an axis is just the size of the non-zero part.
\[ |z| = 3 \]
Read the angle
Why: Straight up is a quarter turn counter-clockwise.
\[ \theta = \tfrac{\pi}{2} + 2\pi k \]
Figure (svg): Two panels. The left panel shows 3i on the imaginary axis with an arrow straight up and a quarter-turn arc labelled pi over 2. The right panel is a broken real axis with an arrow running left to minus 117 and a half-turn arc above the origin labelled pi, with a note that the scale is broken.
Write the set and the principal value
Why: Only the quarter turn itself lies in the window.
\[ \arg(z) = \left\{\tfrac{\pi}{2} + 2\pi k\right\}, \quad \operatorname{Arg}(z) = \tfrac{\pi}{2} \]
Check why the tangent rule was not used
Why: The tangent rule divides by the real part, which is zero here, so it does not apply. Theorem 11.15 handles this case separately, and it gives exactly the quarter turn.
\[ \operatorname{Re}(z) = 0, \; \operatorname{Im}(z) > 0 \;\Longrightarrow\; \arg(z) = \left\{\tfrac{\pi}{2} + 2\pi k\right\} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 993 — Example 11.7.1, part 3
This one is here to show you when not to compute. The point is three units straight up. You do not need a distance formula to know it is three units from the origin, and you do not need an arctangent to know that straight up is a quarter turn.
It is also the first case where the tangent rule genuinely cannot be used. The real part is zero, so the fraction in Theorem 11.15 would be a division by zero. That is not a technicality to be worked around; it is the reason the theorem lists this case separately.
If you ever find your calculator returning an error on the arctangent step, this is almost certainly why. Stop and look at where the point is: on an axis, the answer is one of four angles, and you can write it down immediately.
Worked example
The last one is a real number, and a large one, but the work is no harder.
\[ z = -117 = -117 + 0i, \quad P(-117, 0) \]
Read the distance
Why: The point is 117 units from the origin along the negative real axis.
\[ |z| = 117 \]
Read the angle
Why: Pointing in the negative real direction is a half turn from the positive one.
\[ \theta = \pi + 2\pi k = (2k+1)\pi \]
Correct the printed line
Why: The book prints this step as theta equals pi plus two pi, then as two k plus one, all times pi, times k. The middle term has lost its k and the last has gained one; the set it then states, the odd multiples of pi, is right.
\[ \text{printed: } \theta = \pi + 2\pi = (2k+1)\pi k \;\to\; \text{correct: } \theta = \pi + 2\pi k = (2k+1)\pi \]
Pick the principal value
Why: The odd multiples of pi are pi, negative pi, three pi and so on; the window admits only the first.
\[ \arg(z) = \{(2k+1)\pi\}, \quad \operatorname{Arg}(z) = \pi \]
Check the boundary case carefully
Why: Negative pi is also an odd multiple of pi and it reaches the same point, but the interval is open at the left end, so it is excluded and pi is kept. This is the one number in the example where the choice of window actually bites.
\[ -\pi \notin (-\pi, \pi], \qquad \pi \in (-\pi, \pi] \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 993 — Example 11.7.1, part 4
Two things are going on here, so separate them. The mathematics is easy: a negative real number points in the negative real direction, which is a half turn from the positive one, and its distance from the origin is its size.
The book's printed line for the set of angles is garbled. It reads as a half turn plus two pi, and then as two k plus one all times pi, times k. The first expression has lost its k and the second has gained a spare one. The correct line is a half turn plus a whole number of turns, which is the same as the odd multiples of pi, and that is what the book's next sentence uses, so only the one line is wrong.
The check step is where this example earns its place. Negative pi is an odd multiple of pi, and it reaches the same point, so why is the answer positive pi? Because the window excludes its left end. This is the one number in the example where that convention decides the answer, which is why the book says the value just barely lies in the interval.
Prediction
\[ z = -5, \quad \arg(z) = \{\ldots, -3\pi, -\pi, \pi, 3\pi, \ldots\} \]
Predict first
Every one of those angles reaches the point. Which one is the principal argument?
Correct: Pi, because the interval includes its right end
Why: The interval is open on the left and closed on the right, so negative pi is excluded and pi is admitted. The two angles do reach the same point, which is exactly why one of them has to be ruled out: otherwise the principal argument would not be a single value.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — the interval from negative pi to pi
Commit to one option before revealing. If you chose the negative one because it looks smaller, you have made a reasonable guess about a convention, and conventions have to be learned rather than guessed.
The two angles really do reach the same point, and that is the whole reason a choice must be made. If both counted as principal, a number would have two principal arguments, and the word principal would mean nothing.
Remember the shape of the interval rather than a rule about half turns: open at the left, closed at the right. Then every boundary case answers itself, including the one lurking in the next part of the lesson, where multiplying two numbers can push an argument past the edge of the window.
Concept
At the origin the two quantities behave completely differently, and the reason is worth stating precisely.
| at z = 0 | what happens | verdict |
|---|---|---|
| modulus | the origin has exactly one polar representation with a non-negative distance, namely distance zero | |0| = 0, perfectly well defined |
| arguments | the pair (0, θ) is the origin for every angle θ whatsoever | arg(0) is the whole real line |
| principal argument | no single angle is distinguished among all of them | Arg(0) is left undefined |
\[ |0| = 0, \qquad \arg(0) = (-\infty, \infty), \qquad \operatorname{Arg}(0) \text{ is undefined} \]
This is not a gap in the theory. It is the honest consequence of the origin having no direction, and every later rule that mentions the argument will quietly assume the number is not zero.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 992 — arg(0) and Arg(0)
Zero is the exception in this section, and the table is arranged so that you can see exactly where the two quantities part company.
For the distance, nothing goes wrong. The origin is zero units from itself, and no other non-negative number describes it, so the modulus is a perfectly ordinary value. For the angle, everything goes wrong at once: the polar pair with distance zero lands on the origin whatever angle you write, so every angle whatsoever is an argument of zero.
The set of arguments is therefore the entire real line rather than an evenly spaced row, and a window two pi long no longer catches exactly one of them; it catches all of them in that range. Nothing distinguishes one, so the principal argument is left undefined. This is not a hole in the theory. It is the honest statement that the origin has no direction, and from here on any rule mentioning the argument quietly assumes the number is not zero.
Matching
Match the pairs
Why: Equal parts of the same sign give a quarter of a half turn: positive and positive is π/4, negative and negative is −3π/4. For the third, the tangent is 6 over −6 root 3, so the reference angle is π/6 and Quadrant II makes it π − π/6 = 5π/6. For the fourth, equal parts of opposite signs put it in Quadrant IV at −π/4.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercises 1, 5, 8 and 10
Work each one by the two-step routine rather than by matching shapes. Find the reference angle from the sizes of the two parts, then place it using the signs.
Three of the four have parts equal in size, so the reference angle is an eighth of a turn every time, and only the quadrant differs. Two positives give a Quadrant I answer, two negatives a Quadrant III answer measured clockwise, and a positive with a negative a Quadrant IV answer.
The remaining number is the interesting one. Its parts are not equal, but the ratio of their sizes is one over the root of three, which you should recognise as the tangent of a thirtieth of a turn, and the signs put it in Quadrant II. If you had to reach for a calculator there, add that ratio to your list of recognisable tangents.
Worked example
Not every modulus is irrational. This one is a Pythagorean triple in disguise.
\[ z = -7 + 24i \]
Compute the modulus
Why: Forty-nine plus five hundred and seventy-six is six hundred and twenty-five.
\[ |z| = \sqrt{(-7)^2 + 24^2} = \sqrt{625} = 25 \]
Compute the tangent
Why: Imaginary part over real part.
\[ \tan(\theta) = \frac{24}{-7} \]
Find the reference angle
Why: Use the sizes of the two parts, which is what the arctangent of a positive number gives.
\[ \arctan\left(\tfrac{24}{7}\right) \approx 1.287 \]
Put it in Quadrant II
Why: Negative real part, positive imaginary part, so subtract the reference angle from a half turn.
\[ \operatorname{Arg}(z) = \pi - \arctan\left(\tfrac{24}{7}\right) \approx 1.855 \]
Check by converting back
Why: Cosine of 1.855 is about negative 0.28 and sine is about 0.96; multiplied by 25 these give negative 7 and 24, the parts we started with.
\[ 25\cos(1.855) \approx -7, \qquad 25\sin(1.855) \approx 24 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercise 15
This is a good exercise to do in full because the numbers are large enough that guessing is impossible and clean enough that the answer is obviously right when you get there.
Seven, twenty-four and twenty-five are a Pythagorean triple, so the modulus comes out as a whole number. It is worth knowing two or three of these triples by sight, since textbook problems use them constantly to keep the arithmetic tidy.
The angle needs the full quadrant repair. The arctangent of twenty-four over seven is about 1.287, an angle in Quadrant I, and the point is in Quadrant II, so the principal argument is a half turn minus that. The check converts the answer back into a pair of parts, which is the most useful check in the whole lesson: if the parts you recover are not the ones you started with, the quadrant step went wrong.
Trap
Asked for the principal argument of root three minus i, a common answer:
\[ \operatorname{Arg}(z) = -\tfrac{\pi}{6} + 2\pi k \]
And asked for the set of arguments:
\[ \arg(z) = -\tfrac{\pi}{6} \]
Both wrong, and in opposite directions.
The capital-A version is one angle. The lower-case version is a set, and a set is written with braces and a description of what it contains.
\[ \operatorname{Arg}(z) = -\tfrac{\pi}{6} \]
\[ \arg(z) = \left\{-\tfrac{\pi}{6} + 2\pi k \;\middle|\; k \in \mathbb{Z}\right\} \]
A quick self-check: if your answer to a capital-A question still has a k in it, you have answered a different question.
This trap costs more marks in this section than any genuine misunderstanding, and it is entirely a matter of reading the question.
Use the shape of your answer as an alarm. If a question asked for the capital-A version and your answer still contains the letter k, you have answered the other question. If a question asked for the lower-case version and your answer is a single angle with no braces, you have answered the other question.
The two are related in one line, and that line is worth memorising: the set consists of the principal value plus every whole number of turns. Once you have found the principal value, writing the set costs you nothing but the braces.
Section
Part 3
Concept
Every property in Theorem 11.14 follows from one fact: the modulus of a number is how far its point is from the origin.
Figure (svg): A right triangle in the complex plane with its right angle at 4 on the real axis: the horizontal leg of length 4, the vertical leg of length 3, and the hypotenuse running from the origin to the point 4 plus 3i, labelled with the square root of 4 squared plus 3 squared, which is 5.
\[ |z| = \sqrt{a^2+b^2} = \sqrt{\operatorname{Re}(z)^2 + \operatorname{Im}(z)^2} \]
Squaring both parts destroys their signs, which is why the four numbers with parts plus or minus four and plus or minus three all have modulus five and sit on one circle.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 993-994 — Theorem 11.14 and the proof of its first three parts
Everything in Theorem 11.14 is a consequence of the picture on this slide, so it is worth being able to redraw the picture from memory. The number sits at the corner of a right triangle whose legs are its two parts, and the modulus is the hypotenuse.
That is why the formula is the Pythagorean theorem and the distance formula at the same time. The distance from the origin to the point, computed with the distance formula from algebra, is the same expression as the hypotenuse computed with Pythagoras.
Notice what the squaring does. It destroys both signs, so four numbers with the same pair of part sizes and different sign patterns all share one modulus and sit on a single circle. Keep that circle in mind: the modulus tells you which circle the number is on, and the argument tells you where on it.
Notation
Annotate
On: \( |zw| = |z||w|, \quad |z^n| = |z|^n, \quad \left|\frac{z}{w}\right| = \frac{|z|}{|w|} \)
\[ |z| \ge 0, \qquad |z| = 0 \iff z = 0 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 993 — Theorem 11.14
Three of these rules are ordinary algebra and one absence is the interesting part, so read the fourth annotation as carefully as the other three.
The product rule is the engine. Multiply two complex numbers and the distances multiply, which is a surprisingly strong statement: multiplication in the plane, which looks complicated when you expand it, does something very simple to the lengths. The power rule is the product rule applied repeatedly and the quotient rule is the product rule applied to a reciprocal, so there is really only one theorem here.
What is missing is any statement about a sum. That absence is not an oversight by the authors. There is no true equation of that kind, and the trap slide two ahead shows the counterexample. Whenever you cannot remember whether a rule exists, ask yourself whether it is about multiplying or adding: multiplication behaves, addition does not.
Intuition
Reusing absolute-value bars for a brand-new object would be reckless if the two ever disagreed. They never do.
Take a real number and read it as a complex number with imaginary part zero. Its modulus is the square root of its square, which is exactly the definition of its absolute value.
\[ z = a + 0i \;\Longrightarrow\; |z| = \sqrt{a^2 + 0^2} = \sqrt{a^2} = |a| \]
So on the real axis the new notation reduces to the old one, and in both settings it means the same thing: the distance from the number to zero. That is why the bars are unambiguous.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 994 — footnote 5, on the absolute-value notation
Reusing notation is dangerous, so it is worth checking that nothing breaks. Take a real number, regard it as a complex number with imaginary part zero, and run it through the modulus formula.
What comes out is the square root of its square, which is exactly the definition of absolute value that you have been using since Intermediate Algebra. So on the real axis the new bars mean exactly what the old bars meant, and the notation is unambiguous.
The deeper point is that both notions mean the same thing anyway: distance from zero. On the number line that distance is the absolute value, and in the plane it is the modulus. The bars have always meant how far from zero, and complex numbers simply extend the idea from a line to a plane.
Worked example
Multiply two general complex numbers, take the modulus, and watch the cross terms cancel.
\[ z = a + bi, \quad w = c + di \]
Multiply the two numbers
Why: Expand and use that i squared is negative one.
\[ zw = (ac - bd) + (ad + bc)i \]
Write the modulus of the product
Why: Square each part and add under the root.
\[ |zw| = \sqrt{(ac-bd)^2 + (ad+bc)^2} \]
Expand both squares
Why: The two middle terms are the same size with opposite signs.
\[ = \sqrt{a^2c^2 - 2abcd + b^2d^2 + a^2d^2 + 2abcd + b^2c^2} \]
Cancel the cross terms
Why: Only four terms survive.
\[ = \sqrt{a^2c^2 + a^2d^2 + b^2c^2 + b^2d^2} \]
Factor in pairs
Why: Take a squared out of the first two and b squared out of the last two.
\[ = \sqrt{a^2(c^2+d^2) + b^2(c^2+d^2)} \]
Factor again
Why: The common factor is the second bracket.
\[ = \sqrt{(a^2+b^2)(c^2+d^2)} \]
Split the root
Why: The product rule for radicals, legitimate because both factors are non-negative.
\[ = \sqrt{a^2+b^2}\,\sqrt{c^2+d^2} = |z||w| \]
Check the cancellation was real
Why: The cross terms were negative two abcd and positive two abcd. With a, b, c, d equal to 3, 4, 1, 1 they are negative 24 and 24, and the surviving four terms give 9 plus 9 plus 16 plus 16, which is 50, whose root is about 7.07.
\[ \sqrt{50} \approx 7.07 = 5\cdot\sqrt{2} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 994 — the proof of the Product Rule
This proof has no clever idea in it at all, which is worth saying out loud. You multiply the numbers out, take the modulus, and grind. The whole thing works because two cross terms cancel.
Watch those cross terms. Expanding the first square produces a term with a minus sign in front of two abcd, and expanding the second produces the same quantity with a plus sign. They vanish together, and what remains has a common factor that can be pulled out in two stages.
The last step splits a root of a product into a product of roots, which is legal here because both factors are sums of squares and so never negative. The check at the end runs the same cancellation with actual numbers, so you can see that the disappearing terms are real quantities that really do cancel rather than symbols that happen to look alike.
Worked example
A proof deserves one concrete test, and this one can be done entirely by hand.
\[ z = 3 + 4i, \quad w = 1 + i \]
Find the two moduli
Why: Nine plus sixteen, then one plus one.
\[ |z| = \sqrt{25} = 5, \qquad |w| = \sqrt{2} \approx 1.414 \]
Multiply the numbers
Why: Expand, then collect using i squared equal to negative one.
\[ zw = (3+4i)(1+i) = 3 + 3i + 4i - 4 = -1 + 7i \]
Take the modulus of the product
Why: One plus forty-nine is fifty.
\[ |zw| = \sqrt{(-1)^2 + 7^2} = \sqrt{50} = 5\sqrt{2} \]
Figure (svg): Three arrows from the origin in one complex plane: a short one to 1 plus i of length about 1.41, a medium one to 3 plus 4i of length 5, and a long one to minus 1 plus 7i of length about 7.07, which is 5 times 1.41.
Compare with the product of the moduli
Why: Five times root two is the same number.
\[ |z||w| = 5\sqrt{2} \approx 7.071 = |zw| \]
Check the power rule on the same numbers
Why: Raise the second number to the fourth power: it comes out as negative four, whose modulus is four, and root two to the fourth is also four.
\[ (1+i)^4 = -4, \quad |-4| = 4 = (\sqrt{2})^4 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 993 — Product Rule and Power Rule, Theorem 11.14
A proof tells you something is true everywhere; a numerical check tells you that you have understood what it says. Do both, especially with a rule you are meeting for the first time.
The arithmetic here is small enough to follow in your head. Multiplying the two numbers gives a point that is much further out than either factor, and the picture shows that clearly: the product's arrow is noticeably longer than either of the arrows it came from, and its length is five root two, about 7.07.
Notice what the check does not say. Nothing in this lesson tells you the direction of the product yet, only its length. The direction turns out to follow an equally simple rule, that the arguments add, but that is Theorem 11.16 and belongs to the next lesson. For now the picture should leave you curious about why the product ended up pointing where it did.
Worked example
The power rule is the product rule applied over and over, and induction is how that sentence is made rigorous.
\[ P(n): \quad |z^n| = |z|^n \]
Check the base case
Why: One factor, nothing to prove.
\[ |z^1| = |z| = |z|^1 \]
Assume the statement at k
Why: This is the induction hypothesis, assumed for one unspecified k at least one.
\[ |z^k| = |z|^k \]
Split off one factor
Why: Properties of exponents.
\[ |z^{k+1}| = |z^k z| \]
Apply the product rule
Why: The two factors are complex numbers, so the rule just proved applies.
\[ |z^k z| = |z^k|\,|z| \]
Use the hypothesis
Why: Replace the first factor.
\[ |z^k|\,|z| = |z|^k |z| = |z|^{k+1} \]
Check the machinery did what it claimed
Why: The last line is the statement at k plus one, so truth at k forces truth at k plus one; with the base case, the principle of mathematical induction gives it for every natural n. A numerical spot check: the modulus of one plus i cubed is root two cubed, about 2.83, and one plus i cubed is negative two plus two i, whose modulus is indeed about 2.83.
\[ |(1+i)^3| = |-2+2i| = \sqrt{8} \approx 2.83 = (\sqrt{2})^3 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 994 — the proof of the Power Rule
If induction is rusty, notice how little of it there is here. You check the statement for one factor, assume it for k factors, and get it for k plus one by splitting off a single factor and using the rule you already proved.
The only genuinely mathematical line is the third: a power with exponent k plus one is a power with exponent k times one more copy of the number. Everything after that is the product rule and the assumption, applied in that order.
The numerical check is worth doing yourself with a different exponent. Cube one plus i by hand, find the modulus of the answer, and compare it with the root of two cubed. Seeing the two agree once makes the induction feel like bookkeeping rather than magic, which is what it is.
Worked example
Division is multiplication by a reciprocal, so the quotient rule needs almost no new work.
\[ w \ne 0 \;\Longrightarrow\; |w| \ne 0 \]
Write the quotient as a product
Why: Dividing by w is multiplying by one over w.
\[ \left|\frac{z}{w}\right| = \left|z\cdot\frac{1}{w}\right| \]
Apply the product rule
Why: Both factors are complex numbers.
\[ \left|z\cdot\frac{1}{w}\right| = |z|\left|\frac{1}{w}\right| \]
Use the reciprocal fact
Why: The book leaves this one line as an exercise; it says the modulus of a reciprocal is the reciprocal of the modulus.
\[ \left|\frac{1}{w}\right| = \frac{1}{|w|} \]
Put the two together
Why: This is the quotient rule.
\[ \left|\frac{z}{w}\right| = \frac{|z|}{|w|} \]
Check it on the earlier pair
Why: With three plus four i over one plus i the quotient is three and a half plus a half i, whose modulus is about 3.536, and five divided by root two is the same number.
\[ \left|\frac{3+4i}{1+i}\right| = \left|\tfrac{7}{2} + \tfrac{1}{2}i\right| = \frac{\sqrt{50}}{2} \approx 3.536 = \frac{5}{\sqrt{2}} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 994-995 — the proof of the Quotient Rule
Notice the strategy, because it recurs throughout mathematics: rather than proving a new statement from scratch, rewrite it as a case of one you already have. Dividing by a number is multiplying by its reciprocal, so a quotient is secretly a product.
The one gap is the reciprocal fact in the middle, which the book leaves as an exercise. It is worth doing: write the reciprocal of a number with the conjugate trick, take the modulus, and the result falls out in two lines.
The hypothesis that the denominator is not zero is doing real work. The modulus of zero is zero, and the last line divides by the modulus of the denominator, so allowing a zero denominator would divide by zero twice over, once in the complex numbers and once in the real ones.
Trap
Products behave, so sums surely behave too:
\[ |z + w| = |z| + |w| \]
Wrong. Test it on three plus four i and negative one plus i:
\[ |2 + 5i| = \sqrt{29} \approx 5.39 \]
\[ |z| + |w| = 5 + \sqrt2 \approx 6.41 \]
Figure (svg): Two arrows drawn head to tail in the complex plane: one from the origin to 3 plus 4i, then one from there to 2 plus 5i. A third arrow runs straight from the origin to 2 plus 5i. The direct route, about 5.39 long, is shorter than the two legs added, which come to about 6.41.
Adding complex numbers puts their arrows head to tail, and going round two sides of a triangle is never shorter than going along the third. So the sum rule fails, and what replaces it is an inequality, not an equation.
\[ |z+w| \le |z| + |w| \]
Equality happens only when the two arrows point the same way, which is the degenerate triangle with no area.
Products behaved so nicely that sums seem bound to follow, and they do not. The counterexample is small enough to check in your head: the two moduli are 5 and about 1.41, but the modulus of the sum is only about 5.39, which is well short of their total.
The picture explains why in one glance. Adding complex numbers places their arrows head to tail, so the sum is the third side of a triangle, and a side of a triangle is never longer than the other two together. The statement that replaces the false equation is therefore an inequality, and it is called the triangle inequality for exactly this reason.
There is one case of equality, and it is the case where the triangle collapses: when the two arrows point in the same direction, the detour costs nothing. So the equation is not always false. It is just not a rule, and a rule is what you would be using it as.
Discrimination
Sort into buckets
Sort each claim, assuming z and w are any complex numbers and w is not zero.
Sort these before checking. Two of the six fail, and both failures involve adding or subtracting, which is the pattern to take away.
The negation item is worth a moment. Multiplying a number by negative one reflects its point through the origin, which changes the direction completely and the distance not at all. You can also get it from the product rule, since the modulus of negative one is one.
The subtraction item fails for a reason that is quicker than a counterexample: a modulus can never be negative, but the difference of two moduli easily can be. Take the smaller number first and the right-hand side is negative while the left-hand side is not, so the claim cannot possibly hold in general.
Tweak it
Parameter explorer
The curve plots the modulus of a complex number whose imaginary part is fixed, as its real part runs along the horizontal axis. Slide the imaginary part. Where is the modulus smallest, and what is that smallest value?
\[ \sqrt{x^2 + ({b})^2} \]
Move the slider slowly and watch two things: the height of the lowest point of the curve, and how quickly the curve straightens out as you move away from the middle.
The lowest point sits where the real part is zero, and its height is exactly the size of the imaginary part. That makes sense from the picture of the plane: among all numbers on a fixed horizontal line, the one closest to the origin is the one directly above it.
Far out to either side the curve becomes almost straight, with slopes one and negative one. That is the statement that when the real part is enormous, the imaginary part hardly matters and the modulus is nearly the size of the real part. Also notice what never happens: the curve never dips below the horizontal axis, because a distance cannot be negative.
Section
Part 4
Concept
There is no clean formula for the argument, but there is a clean equation that every argument satisfies, provided the real part is not zero.
\[ \theta \in \arg(z), \; \operatorname{Re}(z) \ne 0 \;\Longrightarrow\; \tan(\theta) = \frac{\operatorname{Im}(z)}{\operatorname{Re}(z)} \]
Figure (svg): A complex plane with the dashed line through the origin at forty-five degrees. Two arrows lie along it in opposite directions, one to 2 plus 2i and one to minus 2 minus 2i. A small arc marks a quarter of a quarter turn for the first and a large clockwise arc marks three quarters of a half turn for the second.
Notice how much the equation does not say. It is satisfied by every angle in the set, which is what you want, and also by every angle pointing in the exactly opposite direction, which is not.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — Theorem 11.15, first bullet
Theorem 11.15 is the only general statement available about the argument, and it is deliberately weak. It gives you an equation the angle satisfies, not a formula for the angle.
The figure shows the weakness precisely. One straight line through the origin carries two opposite directions, and every angle pointing along either direction has the same tangent. So the equation cannot distinguish a number from its negative, and the two have principal arguments a half turn apart.
Read the theorem, then, as a tool for finding the reference angle and nothing more. The quadrant has to come from somewhere else, and the only reliable source is the pair of signs of the two parts, which is why every worked solution in this lesson plots the point first.
Concept
When the real part is zero the fraction is undefined, so Theorem 11.15 lists those numbers separately. Each case is read straight off the picture instead.
| the number | where it sits | arg(z) |
|---|---|---|
| Re(z) ≠ 0 | off the imaginary axis | the angles with tan θ = Im(z)/Re(z), in the right quadrant |
| Re(z) = 0, Im(z) > 0 | up the positive imaginary axis | π/2 plus whole turns |
| Re(z) = 0, Im(z) < 0 | down the negative imaginary axis | −π/2 plus whole turns |
| Re(z) = Im(z) = 0 | the origin | every real number at all |
\[ \operatorname{Re}(z) = 0, \; \operatorname{Im}(z) > 0 \;\Longrightarrow\; \arg(z) = \left\{\tfrac{\pi}{2} + 2\pi k\right\} \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — Theorem 11.15
The table is the rest of Theorem 11.15, and each row exists because the first row's fraction breaks down.
If the real part is zero the fraction has a zero denominator, so the theorem stops and reads the answer off the picture instead. Straight up is a quarter turn, straight down is a quarter turn the other way, and those two cases cover every purely imaginary number.
The last row is zero again, with its full set of arguments and no principal one. Notice how the four rows between them cover every complex number without overlap: either the real part is nonzero, or it is zero and the imaginary part is positive, negative, or zero as well. That kind of exhaustive case split is what makes a theorem usable rather than merely true.
Worked example
The clearest demonstration that the tangent cannot finish the job on its own.
\[ z = 2 + 2i, \qquad w = -2 - 2i \]
Compute both tangents
Why: Each is its imaginary part over its real part, and the two minus signs cancel in the second.
\[ \tan(\theta_z) = \frac{2}{2} = 1, \qquad \tan(\theta_w) = \frac{-2}{-2} = 1 \]
Compute both moduli
Why: Four plus four under the root, twice.
\[ |z| = |w| = \sqrt{8} = 2\sqrt{2} \approx 2.83 \]
Place the first
Why: Both parts positive, so Quadrant I, at the reference angle itself.
\[ \operatorname{Arg}(z) = \tfrac{\pi}{4} \]
Place the second
Why: Both parts negative, so Quadrant III: go a half turn the other way from the same reference angle.
\[ \operatorname{Arg}(w) = \tfrac{\pi}{4} - \pi = -\tfrac{3\pi}{4} \]
Check the two really are different numbers
Why: Feed each principal argument back through cosine and sine at the common radius. The first returns 2 plus 2i and the second returns negative 2 minus 2i, so the same tangent has indeed produced two different points.
\[ 2\sqrt2\,\operatorname{cis}\left(\tfrac{\pi}{4}\right) = 2 + 2i, \qquad 2\sqrt2\,\operatorname{cis}\left(-\tfrac{3\pi}{4}\right) = -2 - 2i \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — the discussion after Theorem 11.15
This pair is the cleanest possible demonstration of the previous slide. Both tangents are one, both moduli are the same, and yet the numbers are as far apart as two numbers on a circle can be.
Work through how each principal argument was placed. The reference angle is an eighth of a turn in both cases, because the parts are equal in size. For the first, both parts are positive, so nothing needs doing. For the second, both are negative, so the direction is a half turn away, and reporting it inside the window means going clockwise rather than counter-clockwise.
The check converts both answers back. That is always available to you, costs two trigonometric values, and catches every quadrant error immediately. Make it a habit on every problem in this exercise set, not just the ones you are unsure of.
Step zero
\[ z = -12 - 5i \]
Discussion prompt
Exercise 17 asks for the principal argument of this number. Before computing anything, what two questions should you answer, and what do their answers tell you about the size and sign of the final angle?
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercise 17
Write your two questions down before revealing. The instinct is to press a calculator button first and think afterwards, and this slide exists to delay that by thirty seconds.
The first question, whether the real part is zero, decides which part of Theorem 11.15 you are in. The second, which quadrant, decides what has to be done to the arctangent's output. Answering both before computing means the final answer arrives with its own error check attached.
Here the answer must lie between a half turn clockwise and a quarter turn clockwise, which is a narrow window. A bare arctangent returns a small positive number, nowhere near that window, so if you had computed first you would have had to notice afterwards that the answer pointed into the wrong quadrant. Knowing the target range in advance means you cannot miss it.
Ranking
Put in order
Order the steps of a complete rectangular-to-polar calculation.
Why: Plotting comes first because the quadrant is what the arctangent cannot supply, and naming the quadrant before computing the angle is what stops the classic error. The modulus can be computed at any point, since it never depends on the quadrant.
Drag the steps into order before checking. Only one ordering question really matters: does the quadrant get named before or after the angle is computed?
Before, always. Naming the quadrant first turns the arctangent into a source of a reference angle, which is all it can reliably be. Naming it afterwards means comparing an answer you already have with a picture, which is exactly the situation in which people talk themselves into keeping a wrong answer.
The modulus is the free step. It never depends on the quadrant, so it can be computed at any point without risk, and putting it early means you have something correct written down before the delicate part begins.
Comparison
Comparison matrix
| z | |z| | Arg(z) | quadrant or axis |
|---|---|---|---|
| 9 + 9i | 9√2 | π/4 | Quadrant I |
| 6i | 6 | π/2 | positive imaginary axis |
| −2 | 2 | π | negative real axis |
| −3√2 + 3i√2 | 6 | 3π/4 | Quadrant II |
| −5i | 5 | −π/2 | negative imaginary axis |
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercises 1, 3, 4, 6 and 9
Fill every blank before checking. The rows are arranged to move from a generic number through the four points on the axes, which are the cases most often got wrong under time pressure.
For the two axis rows, resist any urge to compute. A point straight up has a quarter-turn argument and a modulus equal to the size of its imaginary part; a point straight down has the same modulus and the negative of that angle. The negative real axis is the half turn, reported as positive pi because of the window.
The two remaining moduli are the only arithmetic in the table. In both cases the parts are equal in size, so the modulus is that size times the root of two, and in the fourth row that product simplifies to a whole number, which is worth noticing before reaching for a calculator.
Counterexample
Discussion prompt
Someone claims: if two non-zero complex numbers have the same value of the imaginary part over the real part, they have the same principal argument. Give a counterexample, and then say what extra piece of information does decide the principal argument.
Answer in writing before revealing. Finding a counterexample is a skill, and the useful move here is to look for the cheapest way to change a number without changing the ratio of its parts.
Negating both parts does exactly that: the two minus signs cancel in the fraction, so the ratio is untouched, while the point moves to the diametrically opposite position. That gives you a counterexample for free from any number you like.
The second half of the question is the more valuable one. A counterexample does not just refute a claim; it tells you what the claim was missing. Here what is missing is the pair of signs, which is to say the quadrant. The tangent is a ratio, and forming a ratio of two numbers destroys the information about their individual signs, keeping only whether they agree.
Section
Part 5
Concept
Everything needed to marry the geometry to the algebra is already on the table. Take the rectangular form and replace each part by its polar expression.
Figure (svg): A point in the first quadrant of the complex plane sitting on a dashed quarter circle of radius r. An arrow of length r runs from the origin to it, a dashed vertical line marks the real part as r cos theta and a dashed horizontal line marks the imaginary part as r sin theta, with the angle theta arced at the origin.
\[ a = r\cos(\theta), \qquad b = r\sin(\theta) \]
\[ z = a + bi = r\cos(\theta) + r\sin(\theta)i = r\left[\cos(\theta) + i\sin(\theta)\right] \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — the derivation before Definition 11.3
Everything on this slide was available to you an hour ago. The two substitutions are Theorem 11.7 from Section 11.4, and the expression being substituted into is the rectangular form.
Do the algebra slowly once. Replace the real part by the modulus times the cosine, replace the imaginary part by the modulus times the sine, and then factor the modulus out of both terms. What is left inside the bracket depends only on the angle, and that is the crucial structural fact: the size of the number and the direction of the number have been separated into two factors.
Look at the figure while you do it. The point sits on a circle of radius r, and the two dashed lines are the two substitutions drawn. Every point on that circle has the same modulus and a different angle, which is exactly what the two factors are recording.
Concept
The bracket that appeared is so common it gets its own abbreviation, and then the whole statement becomes short enough to write in one breath.
\[ \operatorname{cis}(\theta) = \cos(\theta) + i\sin(\theta) \]
Definition 11.3, a polar form — If z is a complex number and theta is any one of its arguments, the expression given by the modulus of z times cis of theta is called a polar form of z.
\[ z = |z|\,\operatorname{cis}(\theta), \qquad \theta \in \arg(z) \]
The article matters. Since there are infinitely many arguments, there are infinitely many polar forms of the same number, and the book says a polar form rather than the polar form for exactly that reason.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — Definition 11.3
The abbreviation is only that: three letters standing for cosine plus i sine, taken from the first letters of cosine, i and sine. Nothing is hidden inside it, and whenever you are stuck, the first move is always to write it out in full.
Read the definition's article carefully. It says a polar form, not the polar form, and the reason is the previous part of the lesson: there are infinitely many arguments, so there are infinitely many equally correct polar forms of the same number. Using the principal argument gives the one most people would write, but nothing forces it.
Two things are being carried by the two factors, and it is worth naming them. The first factor carries size and the second carries direction. That separation is why the polar form is worth having, and the next lesson will show that it is exactly the separation multiplication respects.
Notation
Annotate
On: \( z = |z|\,\operatorname{cis}(\theta) = |z|\left[\cos(\theta) + i\sin(\theta)\right] \)
Figure (svg): A unit circle in the complex plane with five points marked on it: cis 0 at 1, cis of pi over 2 at i, cis pi at minus 1, cis of two pi over three in the second quadrant and cis of minus three pi over four in the third quadrant. Every point sits on the circle.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 995 — Definition 11.3 and the cis abbreviation
Four small observations, and the second is the one to keep. The cis factor always lands on the unit circle, because cosine squared plus sine squared is one, so its modulus is one no matter what the angle is. It carries direction and no size at all.
That makes the whole expression easy to read: a number one unit long pointing the right way, scaled up by however long the number actually is. The figure marks five such directions on the unit circle, including the two you should know instantly, that cis of zero is 1 and cis of a quarter turn is the imaginary unit.
The last annotation is the recipe for converting back, and it is worth stating as a sentence: the real part is the modulus times the cosine and the imaginary part is the modulus times the sine. That single sentence is every polar-to-rectangular problem in the exercise set.
Picture it
Figure (svg): A complex plane showing two moves. First a solid arrow runs from the origin two units along the positive real axis. Then a curved arrow sweeps clockwise three quarters of a half turn, carrying the tip round to the third quadrant, where a dashed arrow points at the number minus root 2 minus i root 2.
Read the two factors as two moves. The modulus says how far to walk along the positive real axis, and the argument says how far to rotate what you have walked. The number you land on is the value of the polar form.
Reading the two factors as two physical moves makes polar form much harder to misremember. Walk out along the positive real axis as far as the modulus tells you, then rotate what you have walked by the argument.
The picture performs exactly that for the number in the next worked example. Two units out, then three quarters of a half turn clockwise, and the tip lands in the third quadrant at a point whose parts are both the negative of the root of two.
This reading also explains why the modulus is not allowed to be negative. Walking a negative distance and then rotating would give the same destinations twice over, and the whole point of insisting on a non-negative distance was to make the description unique.
Worked example
Converting to rectangular form is always the same move: write out what cis means.
\[ z = 4\,\operatorname{cis}\left(\tfrac{2\pi}{3}\right) \]
Expand the abbreviation
Why: Cosine plus i times sine, all multiplied by the modulus.
\[ z = 4\left[\cos\left(\tfrac{2\pi}{3}\right) + i\sin\left(\tfrac{2\pi}{3}\right)\right] \]
Evaluate the two values
Why: Two pi over three is in Quadrant II with reference angle pi over three.
\[ \cos\left(\tfrac{2\pi}{3}\right) = -\tfrac{1}{2}, \qquad \sin\left(\tfrac{2\pi}{3}\right) = \tfrac{\sqrt{3}}{2} \]
Multiply through
Why: Four halves is two.
\[ z = 4\left(-\tfrac12 + i\tfrac{\sqrt3}{2}\right) = -2 + 2i\sqrt{3} \]
Read off the parts
Why: The real part is the first term and the imaginary part is the coefficient of i.
\[ \operatorname{Re}(z) = -2, \qquad \operatorname{Im}(z) = 2\sqrt{3} \approx 3.46 \]
Figure (svg): The number 4 cis of two pi over three drawn on a dashed circle of radius 4: an arrow from the origin turned two thirds of a half turn, ending in the second quadrant, with dashed lines showing the real part minus 2 and the imaginary part 2 root 3, about 3.46.
Check the modulus survived the conversion
Why: Squaring and adding gives four plus twelve, sixteen, whose root is four: the same modulus the polar form started with, as it must be.
\[ |-2 + 2i\sqrt3| = \sqrt{4 + 12} = \sqrt{16} = 4 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 996 — Example 11.7.2, part 1a
Every polar-to-rectangular conversion is the same three moves, and this example is the one to copy. Write out what cis means, evaluate the cosine and the sine, and multiply through by the modulus.
The only place a mistake can hide is in the middle step. The angle is two thirds of a half turn, which is in Quadrant II, so the cosine is negative and the sine is positive, and the reference angle is a third of a half turn. If your two values do not have the signs the quadrant demands, stop there rather than multiplying.
The check is the one you should run on every conversion in either direction. The modulus of the answer must equal the modulus you started with, because converting a number between two descriptions cannot change how far it is from the origin. Here squaring and adding gives sixteen, whose root is four, exactly the factor in front of the cis.
Worked example
Same move, negative angle, and the picture from two slides ago is the answer drawn.
\[ z = 2\,\operatorname{cis}\left(-\tfrac{3\pi}{4}\right) \]
Expand the abbreviation
Why: Nothing changes for a negative angle.
\[ z = 2\left[\cos\left(-\tfrac{3\pi}{4}\right) + i\sin\left(-\tfrac{3\pi}{4}\right)\right] \]
Evaluate the two values
Why: Negative three pi over four lands in Quadrant III, where both are negative, with reference angle pi over four.
\[ \cos\left(-\tfrac{3\pi}{4}\right) = -\tfrac{\sqrt2}{2}, \qquad \sin\left(-\tfrac{3\pi}{4}\right) = -\tfrac{\sqrt2}{2} \]
Multiply through
Why: Two times root two over two is root two.
\[ z = -\sqrt{2} - i\sqrt{2} \approx -1.414 - 1.414i \]
Read off the parts
Why: They are equal, and both negative.
\[ \operatorname{Re}(z) = -\sqrt2 = \operatorname{Im}(z) \]
Check against the quadrant
Why: Two negative parts means Quadrant III, which is where an argument of negative three pi over four points; and the modulus is the root of two plus two, which is two, as the polar form said.
\[ |-\sqrt2 - i\sqrt2| = \sqrt{2+2} = 2 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 996 — Example 11.7.2, part 1b
The same three moves with a negative angle, which changes nothing about the procedure. A negative angle just means the rotation is clockwise.
The quadrant check is again where the work is. Three quarters of a half turn clockwise from the positive real axis lands in Quadrant III, where both the cosine and the sine are negative, and the reference angle is an eighth of a turn. So both values are the negative of root two over two, and both parts of the answer come out equal and negative.
Compare your answer with the picture two slides back, which drew this very number as a walk followed by a turn. Being able to match a computed answer to a drawn one is a better check than repeating the arithmetic, because it tests a different part of your understanding.
Worked example
Two conversions whose answers should feel obvious afterwards, which is the point of doing them.
\[ z = 3\,\operatorname{cis}(0), \qquad z = \operatorname{cis}\left(\tfrac{\pi}{2}\right) \]
Expand the first
Why: Cosine of zero is one, sine of zero is zero.
\[ 3\left[\cos(0) + i\sin(0)\right] = 3[1 + 0i] = 3 \]
Read its parts
Why: It is a real number, as an argument of zero should give.
\[ \operatorname{Re} = 3, \qquad \operatorname{Im} = 0 \]
Expand the second
Why: Here the modulus is 1, left unwritten.
\[ \cos\left(\tfrac{\pi}{2}\right) + i\sin\left(\tfrac{\pi}{2}\right) = 0 + i(1) = i \]
Read its parts
Why: The imaginary unit itself.
\[ \operatorname{Re} = 0, \qquad \operatorname{Im} = 1 \]
Check both against the unit circle
Why: Any cis has modulus one, since cosine squared plus sine squared is one, so the second number must sit on the unit circle, and i does. The first is that circle scaled by three and rotated by nothing, landing at 3 on the real axis.
\[ |\operatorname{cis}(\theta)| = \sqrt{\cos^2\theta + \sin^2\theta} = 1 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 996 — Example 11.7.2, parts 1c and 1d
These two conversions look too easy to be worth doing, and that is exactly why they are worth doing: they tell you what the notation means in the cases you can verify independently.
An argument of zero means no rotation at all, so the number stays on the positive real axis and the polar form is just the modulus written twice over. That is the general fact that a positive real number has principal argument zero.
The second is the more memorable. A modulus of one and a quarter turn gives the imaginary unit exactly, which says that the imaginary unit is the point one unit from the origin, a quarter of the way round. If you remember one value of cis, remember that one: it turns the mysterious definition of the imaginary unit into a position on a circle.
Fill the middle
Exercise 32 of the section. The angle is a Quadrant IV special angle, so both values are exact.
Fill in the blanks
12\,\operatorname\tfrac{1}{2}\left(-\tfrac6___\right) = 12\left(___ - i\tfrac___}___\right) = ___ - 6i\sqrt___
Why: Cosine of negative pi over three is one half and sine is negative root three over two, because negative pi over three is in Quadrant IV with reference angle pi over three. Twelve times one half is 6, and twelve times root three over two is 6 root 3, about 10.39, so the number is 6 minus 6i root 3.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercise 32
Fill both blanks before checking. The angle is a third of a half turn clockwise, which puts it in Quadrant IV, where the cosine is positive and the sine is negative.
That sign pattern is the only thing that can go wrong here. If you wrote a negative cosine, you have placed the angle in Quadrant II or III, and the quickest fix is to sketch the rotation: a third of a half turn is a small clockwise move that stays well to the right of the vertical axis.
Notice how the twelve makes the arithmetic pleasant. Twelve times a half is six, and twelve times the root of three over two is six times the root of three, about 10.39. Textbook moduli are chosen to cancel the twos in the denominators of the special-angle values, so if your answer has ugly fractions in it, suspect an error before suspecting the problem.
Worked example
Going the other way needs two pieces of information, and both were computed in Example 11.7.1. Nothing new has to be found.
\[ \sqrt3 - i, \quad -2 + 4i, \quad 3i, \quad -117 \]
Assemble the first
Why: Modulus 2, principal argument negative pi over six.
\[ \sqrt{3} - i = 2\,\operatorname{cis}\left(-\tfrac{\pi}{6}\right) \]
Assemble the second
Why: Modulus two root five, argument a half turn less the arctangent of two.
\[ -2 + 4i = 2\sqrt{5}\,\operatorname{cis}\left(\pi - \arctan(2)\right) \]
Assemble the third
Why: Modulus 3, a quarter turn.
\[ 3i = 3\,\operatorname{cis}\left(\tfrac{\pi}{2}\right) \]
Assemble the fourth
Why: Modulus 117, a half turn.
\[ -117 = 117\,\operatorname{cis}(\pi) \]
Check the first by converting it back
Why: Cosine of negative pi over six is root three over two and sine is negative one half; doubling those gives root three and negative one, so the polar form really does reduce to the number it came from.
\[ 2\left[\tfrac{\sqrt3}{2} + i\left(-\tfrac12\right)\right] = \sqrt{3} - i \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 996 — Example 11.7.2, part 2
The book's phrase for this is shamelessly mining the earlier example, and that is the lesson: once the modulus and an argument are known, a polar form costs no work at all. You assemble it.
Two of the four are worth a second look. The one with the arctangent shows that an exact polar form is allowed to contain an arctangent; you are not obliged to produce a decimal. The one with the modulus 117 shows that nothing about the method depends on the number being small.
The check runs the first one backwards, and that is the honest way to verify a polar form. Expand the cis, multiply by the modulus, and confirm that the rectangular number you started with comes back. If it does not, the argument was placed in the wrong quadrant, which is by far the most likely error.
Worked example
When the angle is written as a half turn minus an arctangent, the conversion needs a right triangle rather than a table of special angles.
\[ z = 50\,\operatorname{cis}\left(\pi - \arctan\left(\tfrac{7}{24}\right)\right) \]
Name the inner angle
Why: Call the arctangent alpha; it is the acute angle of a right triangle with legs 7 and 24.
\[ \alpha = \arctan\left(\tfrac{7}{24}\right), \quad \tan\alpha = \tfrac{7}{24} \]
Find its cosine and sine
Why: The hypotenuse is the root of 49 plus 576, which is 625, so 25.
\[ \cos\alpha = \tfrac{24}{25}, \qquad \sin\alpha = \tfrac{7}{25} \]
Apply the difference identities for a half turn
Why: Cosine of pi minus alpha flips sign; sine of pi minus alpha does not.
\[ \cos(\pi - \alpha) = -\tfrac{24}{25}, \qquad \sin(\pi-\alpha) = \tfrac{7}{25} \]
Multiply by the modulus
Why: Fifty times each of the two fractions.
\[ z = 50\left(-\tfrac{24}{25} + i\tfrac{7}{25}\right) = -48 + 14i \]
Check the modulus comes back to fifty
Why: Squaring and adding gives 2304 plus 196, which is 2500, whose root is 50; and the point is in Quadrant II, which is where an angle a little under a half turn points.
\[ |-48 + 14i| = \sqrt{2304 + 196} = \sqrt{2500} = 50 \]
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 1004 — Exercise 39
This is the hardest conversion in the exercise set, and it rewards a picture of a right triangle rather than a calculator.
The inner angle is the acute angle of a triangle with legs seven and twenty-four, and those legs belong to a Pythagorean triple with hypotenuse twenty-five. So the cosine and sine of that inner angle are exact fractions, and nothing decimal is needed anywhere in the problem.
The other ingredient is the pair of identities for a half turn minus an angle: the cosine changes sign and the sine does not. If those are rusty, read them off the unit circle rather than memorising them again. After that the modulus of fifty cancels the twenty-fives and the answer comes out in whole numbers, which the final check confirms by rebuilding the modulus from them.
Edge cases
\[ -2\,\operatorname{cis}(0) \quad \text{versus} \quad 2\,\operatorname{cis}(\pi) \]
Discussion prompt
Both expressions evaluate to the number negative 2. Only one of them is a polar form in the sense of Definition 11.3. Which one, and what exactly rules the other out?
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, p. 991 — the requirement that r be non-negative
Answer in writing before revealing. Both expressions do evaluate to the same number, so the question is genuinely about the definition rather than about arithmetic.
Definition 11.3 builds the expression out of the modulus of the number, and the modulus was defined to be non-negative. So a leading negative factor cannot be a modulus, and the expression, while it evaluates correctly, is not a polar form in this book's sense.
It is worth seeing this as a deliberate trade. Section 11.4 allowed a negative first coordinate, which gave flexibility but meant a point had even more names than coterminal angles alone would produce. Here that flexibility is spent to buy uniqueness of the modulus, and what is left over, the freedom in the angle, is then tamed separately by the principal-argument convention.
Real world
Discussion prompt
An alternating voltage is described by two numbers: how big the swing is, and how far it lags or leads a reference wave. Engineers write such a signal as a single complex number. Which part of the polar form carries the size, which carries the lag, and why would rectangular form be the wrong choice for this job?
Write an answer before revealing. The question is really about which description of a number matches the description of the thing being modelled.
An alternating signal is naturally described by a size and a lag, so the polar form matches it factor for factor: the modulus is the amplitude and the argument is the phase. In rectangular form both physical quantities are smeared across both parts, and neither can be read without computing.
There is a second reason, which the next lesson supplies. Passing a signal through a circuit element scales it and shifts its phase, and in polar form that is multiplication: the moduli multiply and the arguments add. A calculation that would be a mess of expansions in rectangular form becomes multiplying two numbers and adding two angles. That is what the polar form is for.
Section
Part 6
Pattern
Figure (svg): A two-way diagram. On the left a box reads rectangular form, a plus bi; on the right a box reads polar form, the modulus times cis theta. An upper arrow runs left to right carrying the rules for the modulus and the tangent of theta, with a warning to check the quadrant. A lower arrow runs right to left carrying a equals r cosine theta and b equals r sine theta.
This is the procedure for every problem in the exercise set, in both directions, and the diagram is arranged so that the asymmetry between the two arrows is visible.
Going from rectangular to polar is the direction that needs care, and all of the care is in one place: the tangent loses the quadrant, so you must supply it yourself from the signs of the parts. Plot first, name the quadrant, then compute.
Going the other way needs no care at all, because the cosine and the sine already carry the sign information. That asymmetry is also why converting back is such a good check: the safe direction can be used to audit the risky one, and it costs two trigonometric values.
Check
Check your understanding
Which angle is the principal argument of the number with real part negative 3 and imaginary part negative 3?
Answer: A
Why: Both parts are negative, so the point is in Quadrant III, and the reference angle is π/4 because the parts are equal in size. Measuring clockwise from the positive real axis gives −3π/4, which lies inside the window from negative pi to pi.
Three of the four options are angles with the correct reference value, so the question is entirely about placement. Work out the quadrant from the signs before looking at the options.
Both parts are negative, which is Quadrant III, and the parts are equal in size, which makes the reference angle an eighth of a turn. Measured clockwise, that is three quarters of a half turn in the negative direction, and that value lies inside the window. The counter-clockwise version names the same direction but is too large for the window, so it is an argument and not the principal one.
Check
Check your understanding
Which statement can fail for some complex numbers z and w?
Answer: B
Why: There is no sum rule. Adding complex numbers adds their arrows head to tail, and the direct route is at most as long as the two legs, with equality only when the arrows point the same way. The other three are the product, power and quotient rules of Theorem 11.14.
This asks which rule can fail, so scan the four statements for the one about addition. Products, powers and quotients all have rules; sums do not.
If you want to see the failure rather than recall it, take the pair from the trap slide: the moduli are 5 and about 1.41, and the modulus of the sum is about 5.39 rather than about 6.41. What is true instead is an inequality, with equality only when the two numbers point the same way.
Check
Check your understanding
Which complex number does the polar form with modulus 6 and argument three pi over four equal?
Answer: A
Why: Three pi over four is in Quadrant II, where the cosine is negative root two over two and the sine is positive root two over two. Multiplying both by 6 gives −3√2 for the real part and 3√2 for the imaginary part, each about 4.243.
Two things have to be right: the quadrant and the size. Three quarters of a half turn counter-clockwise points into Quadrant II, so the real part must be negative and the imaginary part positive, which eliminates two options immediately.
The remaining discrimination is by size. Both surviving candidates point the right way, but only one has modulus 6; the other is about 4.24 from the origin. Checking the modulus of your answer against the modulus in the polar form is the fastest way to catch that kind of slip.
Explain it to yourself
Discussion prompt
In two or three sentences, explain why the modulus of a complex number is a single number while its argument is an infinite set, and why that difference forces the extra definition of a principal argument.
Write your two or three sentences before revealing. Being able to say why the two quantities behave differently is a better test of understanding than any computation in this lesson.
The key contrast is between an equation with finitely many solutions and one with infinitely many. Squaring and adding gives an equation for the distance with two solutions, and one requirement removes one of them. Rotating by a whole turn gives infinitely many angles reaching the same point, and no requirement of that kind can cut the list down to one.
So the convention has to do the work instead, by choosing an interval. The half-openness of that interval is not fussiness: without it a number pointing along the negative real axis would have two principal arguments, and the definition would fail at exactly the place it is most often used.
Exit ticket
\[ z = -1 + i\sqrt{3} \]
Discussion prompt
Find the modulus and the principal argument of this number, write a polar form, and then convert that polar form back to check it.
This puts the whole lesson into one problem, and the third part, converting back, is the part to take seriously.
The modulus is the root of one plus three, which is two. The tangent is negative the root of three, whose reference angle is a third of a half turn, and the signs put the point in Quadrant II, so the principal argument is a half turn less that reference angle.
Now convert back. The cosine of two thirds of a half turn is negative a half and the sine is the root of three over two, and doubling those returns the number you started with. If your conversion back had produced the number with both signs swapped, that would have been the signal that you had taken the arctangent's Quadrant IV answer and forgotten to move it.
Recap
| object | what it is | how many values |
|---|---|---|
| Re(z), Im(z) | the two real numbers in a + bi, and the coordinates of its point | one each, always |
| |z|, the modulus | the distance from the point to the origin | one, because the distance is required to be non-negative |
| arg(z) | every angle whose terminal side reaches the point | infinitely many, spaced 2π apart |
| Arg(z) | the one argument in the window from −π to π, excluding −π | one, unless z is 0, where it is undefined |
| a polar form | the modulus times cis of any argument | infinitely many, one per argument |
\[ z = a + bi = |z|\,\operatorname{cis}(\theta), \quad |z| = \sqrt{a^2+b^2}, \quad \tan\theta = \frac{b}{a} \]
Next, Section 11.7 continues with what the polar form was built for: multiplying two of them multiplies the moduli and adds the arguments, which gives DeMoivre's Theorem for powers and, running it backwards, every one of the nth roots of a complex number.
Stitz & Zeager, College Trigonometry, Ch. 11 Applications of Trigonometry — §11.7 Polar Form of Complex Numbers §11.7, pp. 991-997 — Definitions 11.2 and 11.3, Theorems 11.14 and 11.15
OpenStax Algebra and Trigonometry 2e, §10.5 Polar Form of Complex Numbers §10.5 — the same material in OpenStax Algebra and Trigonometry
Five objects and one question about each: how many values does it have? The two parts and the modulus have one each, the set of arguments has infinitely many, the principal argument has one unless the number is zero, and a polar form has one per argument. Nearly every error in this section is really a confusion about which of those five is being asked for.
The procedural core is two formulas and one warning. The modulus is the square root of the sum of the squares of the parts, the tangent of any argument is the imaginary part over the real part, and that second formula cannot tell you the quadrant, so you must get it from the signs and a sketch.
The next lesson is where the polar form pays for itself. Multiplying two numbers in polar form multiplies the moduli and adds the arguments, which gives a power rule known as DeMoivre's Theorem and, run backwards, produces all of the nth roots of a complex number, evenly spaced around a circle.
Want this taught 1-on-1? Alexander tutors Trigonometry — $55/session, free consultation.