Section 5.6 Complex Numbers
Section 5.6 Complex Numbers
Section 5.6
5.6 Complex Numbers
An Introductory
Introductory Tale
Try to remember back when you were in the second grade and Miss
Thomas asked you to find the number x that made 2 x = 4 , ok maybe not in
those exact words. You, of course, told her right away x = 2 , but then the
next year in third grade Miss Anderson asked for the number that made
x + 3 = 0 and you were stumped. Of course Miss Anderson used this problem
to introduce you to negative numbers. But then, just when you were getting
comfortable with negative numbers, Miss Hanson in fourth grade asked for the
solution of 5 x = 13 , which naturally you said didn’t exist, but once again you
were forced to enlarge your thinking about numbers and include rational
numbers into your numeric repertoire. By this time you were a junior in high
school and asked to solve x 2 − 2 = 0 , which after another embarrassing
interlude upon announcing “there wasn’t any,” you were introduced to
irrational numbers and the real number system. It was about then that you
were starting to feel your teachers were playing some kind of “switch and
bait” scam, making up new numbers on the fly simply to find solutions to
problems they couldn’t solve. And the final cruncher came when you were in
high school and asked to solve x 2 + 1 = 0 . By this time you were starting to get
comfortable with the real numbers so there was no question that this equation
had no solution; everyone knew it was impossible for the square of a number
plus one to be zero. It just can’t happen. But to your amazement, your
teacher pulled another rabbit out of the hat and served up two answers called
imaginary numbers x = ± i , where i = −1 , which seemed at the time
ridiculous since all your past teachers had taught that it was impossible to
take the square root of a negative number. So you protested and told your
teacher something like “that’s a lot of …,” But over time you started to
accept those “imaginary” numbers and even start to call them by a less
controversial name “complex numbers.” Now, by the time you got to college
you were waiting for some professor introduce hyper-complex numbers in
order to solve some hyper-complex equation like
x 5 + 13 x 4 − 2 x3 + x 2 − 5 = 0
However, your professor told you not to worry, because a young man about
your age had proven that any polynomial equation
Section 5.6 541 Complex Numbers
a0 x n + a1 x n −1 + + an −1 x + an = 0
If you think this story is rather far-fetched, it isn’t. It is, in fact the
history of numbers and how humankind has adapted its interpretation of their
meaning to fit its needs. And if you think humankind didn’t kick and scream
with the introduction of each new number system, think again. It was not too
many hundred years ago that even the greatest mathematicians didn’t
“believe” in complex numbers.
Oh yes, and that young lad who showed that every polynomial equation
with integer coefficients always has complex roots, that was Karl Frederick
Gauss (1777-1855), arguably the greatest mathematician who ever lived, who
proved the fundamental theorem of algebra in 1798 at the ripe old age of 21.
Margin Note:
Note In 1545 the Italian mathematician Girolamo Cardano in the
celebrated Ars Magna solved the simultaneous equations
x + y = 10, xy = 40
getting the solution
x = 5 + −15, y = 5 − −15 .
Cardano did not give any interpretation of the square root of a negative
number, although he did say that if they obey the usual rules of algebra, then
the solution could be verified.
Complex Numbers
Although historically complex number were introduced to solve
polynomial equations, today they are crucial in the foundations of many areas
of mathematics, including harmonic analysis, ordinary and partial differential
equations, analytic number theory, analytic function theory, as well as many
areas of engineering and science, including theoretical physics where analytic
function theory constitutes much of the foundation of quantum mechanics.
Someone once argued that real numbers are more “natural” than
complex numbers since real numbers measure things we can all see and feel,
like a person’s height or weight, whereas no one can physically “experience”
complex numbers. The person who makes such a claim just doesn’t know
1
The real numbers of course are just special types of complex numbers.
Section 5.6 542 Complex Numbers
where with these operations, the complex numbers satisfies the axioms of an
algebraic field, complete with additive identity (the real number 0) and
multiplicative identity (the real number 1). The complex numbers do not form
an ordered field, but they do have a very important property the real numbers
do not, such as containing solutions of equations not contained in the real
number system. For example the equation z 2 + 1 = 0 . Algebraists would say
that the complex numbers are algebraically closed field whereas the real
numbers are not.
2
You may not think of them as numbers yet but after we learn to add, subtract, multiply and divide
them, you will.
3
The word “imaginary” is unfortunate since there is nothing really imaginary about complex
numbers. The number was coined in the 1500s when many people didn’t think of complex
numbers as numbers. Unfortunately, the name as stuck.
Section 5.6 543 Complex Numbers
( 0,1) . Keep in mind the complex plane is not the same as the Cartesian
plane. In the Cartesian plane points are simply that, points. In the complex
plane points are representations of complex numbers that we can multiply and
divide.
Margin Note: The “+” in x + iy is unlike any “+” you have seen before. It is
not addition in the traditional sense, but a “placeholder” which separates the
real and pure imaginary parts of the complex number; you don’t actually add
x + iy together to get something else.
Polar Coordinates
Recall that a point ( x, y ) in the Cartesian plane can be written in polar
coordinates ( r ,θ ) where the relationship between the variables is
and so any complex number z = x + iy can also be
x = r cos θ , y = r sin θ
expressed in polar coordinates as
Since the argument θ can wrap around the origin several times, either
clockwise or counterclockwise, the principle argument of a complex number is
the unique argument that lies in the interval ( −π , π ] . Thus, the complex
number i has argument π / 2 , −1 has argument π , and −i has argument −π / 2
(or sometimes we say 3π / 2 ).
4
There are two ways to introduced the complex exponential; one is to define the exponential in terms of
real sines and cosines, then show this “new animal” i.e. the complex exponential, has all the properties we
a b
a b
expect of an “exponential” function, such as e e = e
a +b
( )
, e = e ab and so on. Another way to
introduce the complex exponential is by means of Euler’s theorem, which shows how the domain of
functions of a real variable can be extended to the complex plane. The problem with this approach,
however, is that it requires more knowledge of complex infinite series that we have available in our brief
introduction to complex variables. Nevertheless, we will proceed “formally” by presenting a proof of
Euler’s theorem, where several (true) facts about infinite series are taken from granted. .
Section 5.6 546 Complex Numbers
where eiθ is the result of replacing θ with iθ in the Taylor series expansion
of eθ . Replacing θ by −θ gives the reciprocal
Proof:
Taylor series expansions of eθ , sin θ , cos θ about θ = 0 , which converge for
all real values θ , are
θ3 θ5 θ7
sin θ = θ − + −
3! 5! 7!
θ2 θ4 θ6
cos θ = 1 − + − +
2! 4! 6!
θ2 θ3 θ4
eθ = 1 + θ + + + +
2! 3! 4!
2 3 4
e iθ
= 1 + iθ +
( iθ ) +
+
( iθ ) +
( iθ )
2! 4! 3!
θ2 θ4 θ6 θ3 θ5 θ7
= 1 − + − + + i θ − + − ▌
2! 4! 6! 3! 5! 7!
= cos θ + i sin θ
Note: The value of Euler’s theorem lies in the fact it allows us to work with
exponentials and all their wonderful manipulative properties rather than with
trigonometric functions and their less than desirable manipulative properties.
Using Euler’s theorem, we can prove eiθ has all the properties of one expects
iθ1 n
of an exponential function, such as e eiθ 2 = ei (θ1 +θ2 ) , ( eθ ) = e nθ using the
trigonometric identities of the sine and cosine functions. Also from Euler’s
equation we have the identity
5
The sine and cosine function are measured in radians in this formula.
6
The rearrangement of terms is allowed since the Taylor series are absolutely convergent.
Section 5.6 547 Complex Numbers
a) e − x +3 y = e − x ( cos 3 y + i sin 3 y )
b) e( −3+ 2i )t = e−3t ( cos 2t + i sin 2t )
c) e −ix = cos x − i sin x
which makes the visual aspect of complex numbers much easier. Note that
See Figure 4.
7
We get so used to representing complex numbers as points in the plane that we use them interchangeably.
8
We really should say θ goes from −π to + π since we have adopted the negative angle convention,
but people are sloppy about this so we simply say 0 to 2π .
9
Note the convention of sometimes placing the " i " in front of the constants in the exponent and
sometimes at the end. Also the argument of −i is often represented interchangeably by either 3π / 2 or
−π / 2 .
Section 5.6 548 Complex Numbers
a) 1 + i = 2 cos ( π / 4 ) + sin ( π / 4 ) = 2 ei (π / 4)
b) i = cos (π / 2 ) + i sin ( π / 2 ) = ei (π / 2)
c) 1 + 2i = 5 cos ( π / 3) + i sin (π / 3) = 5 ei (π / 3)
d) −1 = cos ( π ) + i sin (π ) = eiπ
Margin Note: The equation eiπ = −1 in Example 2d), which can be rewritten
eiπ + 1 = 0 , is an amazing equation, called Euler’s identity and contains
(arguably) the five most important numbers in all of mathematics;
0,1, i, π , and e .
Section 5.6 549 Complex Numbers
Multiplication:
Multiplication: (
z1 z 2 = r1eiθ1 )( r e ) = r r e (
2
iθ 2
1 2
i θ1 +θ 2 )
or
( x1 + iy1 )( x2 + iy2 ) = ( x1 x2 − y1 y2 ) + i ( x1 y2 + x2 y1 )
The product of two complex numbers is best interpreted using polar
coordinates. In the complex plane, the product of two complex numbers is a
complex number whose magnitude is the product of the magnitudes of the two
numbers, and whose argument is the sum of the arguments of the two
numbers.
Section 5.6 550 Complex Numbers
a) ( 2 + 3i )( 3 + i ) = 3 + 11i
2 + i 1 + 3i 2 + i 2 − 3 1 + 6 1 7i
c) = = 2 2 + 2 2 i = − +
1 − 3i 1 + 3i 1 − 3i 1 + 3 1 + 3 10 10
Section 5.6 551 Complex Numbers
6eπ i
d) πi/2
= 3eπ i
2e
1 1 −iθ 1
e) = e = ( cos θ − i sin θ )
z r r
n
z n = ( reiθ ) = r n einθ = a eiφ
where φ is the argument of a n . Taking the absolute value of each side of this
1/ n
equation yields r n = a from which we conclude r = a . The equation now
reduces to einθ = eiφ from which we conclude nθ = φ or θ = φ / n . Hence, an
iφ
nth root of the complex number a = a e is
1/ n 1/ n
z= a eiφ / n = a cos (φ / n ) + i sin (φ / n ) .
1/ n
However, this is not the only root since zk ≡ a ei(φ + 2π k ) / n) , k = 0,1, 2,..., n − 1 also
satisfies z n = a , which can be seen by direct computation
n
(a 1/ n
ei (φ + 2π ) / n ) ) = a ei (φ + 2π ) = a eiφ e 2π i = a eiφ = a .
i φ + 2π k ) / n
e (
1/ n
a1/ n = a
1/ n φ + 2π k φ + 2π k
=a cos + i sin
n
, k = 0,1, 2,..., n − 1
n
Hence, to find the nth root of a complex number a , first find the principal root,
root
1/ n
whose absolute value is a and whose argument is arg ( a ) / n . The other
1/ n
n − 1 roots are equally spaced points on the same circle (radius a ) as the
principle value, the angle between them being 2π / n .
Example 4 (Roots)
0 + 2π k 0 + 2π k
a) 4 = 2 cos + i sin = 2 cos ( kπ ) + i sin ( kπ ) , k = 0,1
2 2
which gives 2 and − 2 respectively.
10
When we say complex number, we of course mean the real numbers as well.
Section 5.6 553 Complex Numbers
b)
π / 2 + 2π k π / 2 + 2π k π π
i = 1 cos + i sin = cos + kπ + i sin + kπ , k = 0,1
2 2 4 4
1 i 1 i
which gives the two roots on the unit circle + ,− − .
2 2 2 2
1/ 3 π / 2 + 2π k π / 2 + 2π k
c) i1/ 3 = 1 cos + i sin , k = 0,1, 2
3 3
π 2π k π 2π k
ak = cos + + i sin + , k = 0,1, 2
6 3 6 3
shown in Figure 5.
The fractional powers z p / q are found in the same way the roots of a
complex number were found and are
p/q pφ + 2π k pφ + 2π k
a p/ q = a cos + i sin , k = 0,1,..., q − 1
q q
A useful technique to finding the values of z p / q is to first find the q roots z1/ q ,
then raise each root to the pth power.
Section 5.6 554 Complex Numbers
Complex Functions
In calculus you performed analytical calculations, such as differentiation
and integration, on a host of real-valued functions f : → of one variable,
such as f ( x) = sin x, g ( x) = e x and so on. We now introduce complex-
complex-valued
2
functions of a complex variablevariable f : → , but unlike functions like f ( x) = x
where one plots x on the horizontal axis and y on the vertical axis of the
Cartesian plane, we have a problem when it comes to plotting complex functions
w = f ( z ) . The problem lies in the fact that both domain and codomain are
complex planes, so in effect we need four dimensions to draw a graph of a
function, a rather difficult problem in a 3-dimensional world. To overcome this
problem we draw two copies of the complex plane11, the domain plane called the
z-plane and the codomain plane called the w-plane.
plane See Figure 1.
11
It is really the same complex plane, it is simply more illustrative to draw two separate copies.
Section 5.6 555 Complex Numbers
a) w = i, z = 1 + i
Ans:
Ans z = 2, θ = π / 4, w + z = 1 + 2i, wz = −1 + i, w / z = (1 + i ) / 2
b) w = 2, z = −i
Ans: z = 1, θ = 3π / 2, w + z = 2 − i, wz = −2i, w / z = 2i
c) w = i, z = −i
Ans z = 1, θ = 3π / 2, w + z = 0, wz = 1, w / z = −1
d) w = −1 − i, z = 1 + i
Ans:
Ans z = 2, θ = π / 4, w + z = 0, wz = −2i, w / z = −1
e) w = 1 + i, z = −1 + i
Ans:
Ans z = 2, θ = 3π / 4, w + z = 2i, wz = −2, w / z = −i
reiθ = 2e ( )
i π /2 2
a) 2i Ans:
Ans: reiθ = 2e ( )
i 3π / 4
b) −1 + i
reiθ = e ( )
i 3π / 2
c) −i Ans:
Ans: reiθ = 8e ( )
i 7 π /8
d) 2 − 2i
Ans: reiθ = 3e ( )
i 0
e) 3
a) e3π i
b) 2eiπ /4
Section 5.6 557 Complex Numbers
2 2
Ans: 2eiπ / 4 = 2 cos (π / 4 ) + i sin (π / 4 ) = 2 + = 2 2
2 2
c) e 2π i
d) e5π i
e) 5e3π i /4
2 2
Ans: 5e3π i / 4 = 5 cos ( 3π / 4 ) + i sin ( 3π / 4 ) = 5 −i
2 2
100
4. Evaluate (1 + i ) .
Ans:
100
(1 + i )
100
= ( 2 eiπ / 4 )
= 250 e 25π i
= 250 cos ( 25π ) + i sin ( 25π )
= 250 ( −1)
= −250
5. Show that the complex conjugate of the sum of two complex numbers is
the sum of the conjugates; that is ( w + z ) = w + z .
Ans:
2
6. Verify the identity z = zz for z = 2 + 3i .
Section 5.6 558 Complex Numbers
Ans:
a) z 3 ( ) ( )
Ans: Re z 3 = x 3 − 3 xy 2 ,Im z 3 = 3x 2 y − y 3
x y
b) 1/ z Ans: Re (1 / z ) = 2 2
,Im (1 / z ) = − 2
x +y x + y2
9. Compute
a) i Ans:
b) −i Ans:
c) 1+ i Ans:
3
d) −1 Ans:
4
e) −1 Ans:
n
( cos θ + i sin θ ) = cos nθ + i sin nθ
Ans: The formula is clearly true when n = 1 . Assuming the formula is true
for n , we can write
n +1 n
( cos θ + i sin θ ) = ( cos θ + i sin θ )( cos θ + i sin θ )
= ( cos θ + i sin θ )( cos nθ + i sin nθ )
= ( cos θ cos nθ − sin θ sin nθ ) + i ( cos θ sin nθ + sin θ cos nθ )
= cos ( n + 1) θ + i sin ( n + 1) θ
Section 5.6 559 Complex Numbers
9.. (Primitive Roots of Unity) The n roots of unity are the n roots of the
equation z n = 1 . Find and plot the roots when n = 1, 2,3, 4, and 8 .
Ans: Writing both sides of the equation z n = 1 in polar form and using de
Moivre’s Formula, we have
But two complex numbers are equal if and only if their magnitudes are equal
and their angles are equal, hence r = 1 and the angles nθ and 0 are equal up to
the aiddition of a multiple of 2π . Hence, this gives us θ = 0 + 2kπ where k
can be any of the numbers 0,1,2,…. . In principle this gives us an infinite
number of roots, one for each k , but not all the answers are different since
changing the angle by 2π does not change z Hence, we are left with the
roots
2kπ 2kπ
zk = cos + i sin n , k = 0,1, 2,..., n − 1 .
n
These n complex numbers are scattered equally around the unit circle in the
complex plane. Figure 5 shows the 8 roots of unity.
10. (Fractional
(Fractional Powers) Find the following.
a) i 3/ 2
Ans:
Section 5.6 560 Complex Numbers
i 3/ 2 = e (3/ 2)ln i
= e( 3/2 )( iπ / 2)
= e3π i / 4
= cos ( 3π / 4 ) + i sin ( 3π / 4 )
2 2
=− +i
2 2
3/4
b) ( −1)
Ans:
3/ 4
( −1) = e( 3/ 4 )( iπ )
= e( 3/ 4)( iπ )
= e3π i / 4
3π 3π
= cos + i sin
4 4
2 2
=− +i
2 2
c) 1+ i
Ans:
1/ 2
1 + i = (1 + i )
= e( ) ( )
1/ 2 ln 1+ i
(1/ 2 )( ln 2 + iπ / 4 )
=e
= e( )
1/ 2 ln 2
eiπ /8
= 21/ 4 ( cos (π / 8 ) + i sin ( π / 8 ) )
1.098 + 0.455 i
d) ii
Ans:
Section 5.6 561 Complex Numbers
i ln ( i )
ii = e
=e(
i iπ / 2 )
= e −π / 2
0.208
1
= −i .
i
Ans:
Ans Multiply numerator and denominator by i .