Polar Coordinates

Definition:
A two-dimensional coordinate system that specifies the location of a point P via 2 entities:
1) r be the distance measured from the origin O to the point P .
2) θ being the angle measured from the +ve x-axis, to the segment OP

Visual:

Polar coordinates offer a unique way to describe points in a plane, particularly in contexts where
the relationship between points is more naturally expressed in terms of distance and direction
from a central point.

Important equations for conversion:

Equation Description
r 2 = x2 + y 2 Relationship between Cartesian coordinates (x, y) and polar coordinates (r, θ)
x = r cos θ Conversion from polar to Cartesian coordinate x
y = r sin θ Conversion from polar to Cartesian coordinate y
y
x = tan θ Relationship between Cartesian coordinates and the angle θ in polar coordinates
r = 0 ∀θ The origin in polar coordinates

example:
Find the equation of the parabola (P ) : y = x2 − 2x in polar coordinates:
solution:
r sin(θ) = r2 cos2 (θ) − 2r cos(θ)
r(r cos2 (θ) − 2 cos(θ) − sin(θ)) = 0
(r = 0) ∪ (r = tan(θ) sec(θ) + 2 sec(θ))

1
Describing Polar Curves + Sketching:
(1) Explicitly: r = f (θ)

For most polar curves, we can isolate r into a function of θ, namely r = f (θ)
Cardioids

1. Horizontal Cardioid: (cusp at the origin + symmetric about the x-axis):

r = a(1 ± cos θ)

2. Vertical Cardioid: (cusp at the origin + symmetric about the y-axis):

r = a(1 ± sin θ)

r = 2(1 + sin(θ)) r = 1 − cos(θ)

(2) Implicitly: F (r, θ) = 0

Some polar curves are described by a relation between r and θ

Lemniscates:

1. Lemniscate of Bernoulli:

r2 = a2 cos(2θ)
2. Lemniscate of Gerono:

r2 = a2 sin(2θ)

r2 = 4 cos(2θ) r2 = sin(2θ)

2
Exercises:

1) Describe the line ax + by = c in polar coordinates. (a, b, c ̸= 0)

solution:
Substituting x = r cos(θ) and y = r sin(θ), we get:
ar cos(θ) + br sin(θ) = c
r(a cos(θ) + b sin(θ) = c
c
r=
a cos(θ) + b sin(θ)
−a
However, notice the restriction from the denominator that tan(θ) ̸=
b
−a −a
In the case where tan(θ) = , we get y = x
b b
Substituting this in our original line, we get c = 0, which is false, hence no restrictions.

2) Sketch the region defined by sec(θ) < r < 2 for θ ∈ [0, π/4]
solution:
First off, we have to find what the bounds for r correspond to.
r1 = sec(θ) =⇒ r1 cos(θ) = 1 =⇒ x = 1, a vertical line.
r2 = 2 =⇒ (r2 )2 = 4 =⇒ x2 + y 2 = 4, the circle (C) : (O, 2)
For the angle, we start off on the x-axis and reach the line y = x (tan θ = 1)
Our region is the one to the left of x = 1 enclosed by (C) for the corresponding θ.

3
Derivatives and Tangent Lines:
Given a polar curve described explicitly as r = f (θ), then:
d
dy dy/dθ dθ (r sin(θ)) f ′ (θ) sin(θ) + f (θ) cos(θ)
= = =
dx dx/dθ d
dθ (r cos(θ))
f ′ (θ) cos(θ) − f (θ) sin(θ)

Worked out example:

We will find an equation of the tangent line to the cardioid (C) : r = 1 − cos(θ) when θ = π/4
 
dy
The tangent line has an equation (T ) : y = ax + b where a =
dx θ=π/4
dy sin2 (θ) + cos(θ) − cos2 (θ)
f (θ) = 1 − cos(θ) =⇒ f ′ (θ) = sin(θ) =⇒ =
dx sin(θ) cos(θ) − sin(θ) + sin(θ) cos(θ)
√
2 √
dy cos(2θ) − cos(θ) 0− 2
After some simplification: = =⇒ a = √ √ =1+ 2
dx sin(θ)(1 − 2 cos(θ)) 2
−
2 (1 2)
√
Our equation for the line simplifies to (T ) : y = (1 + 2)x + b
√
2
To find b, we need the point of tangency, namely P (1 − 2 , π/4) (plug θ = π/4 in (C))
√ √
Since we are working in cartesian, this corresponds to x = (1 − 22 ) cos(π/4) = 2−1
2
√ √ √ √
2 2−1 2−1 2−1
and y = (1 − 2 ) sin(π/4) = 2 =⇒ P ( 2 , 2 )
√
2−1
√ √2−1 √
2
Substituting into (T ), we get b = 2 − (1 + 2)( 2 ) = 2 −1
√
√ 2
(T ) : y = (1 + 2)x + −1
2

4
Extra Sheet 6:
Exercise 1:
Consider the line (d) : y = 4 − x and the hyperbola (H) : xy = 3 where x, y > 0
1) Find the intersection point of (d) and (H)
p
2) Prove that the equation of (H) is given by r = 2 csc(2θ)
3) Describe then sketch the region bounded by (d) and (H)

Exercise 2: Sketch the following regions

1) 0 < r < 4 for θ ∈ [0, π/3]

2) sec(θ) tan(θ) ≤ r ≤ 2 csc θ for any θ

3) |r| < cos(θ) + sin(θ) for any θ

Exercise 3:
Consider the lemniscate (L) : r2 = cos(2θ)
p
1) Explain briefly why we can write r = cos(2θ)
2) Find the points of (L) were the tangent drawn is horizontal.
2 tan(θ)
You may need the identity tan(2θ) =
1 − tan2 (θ)

Exercise 4: Describe in polar coordinates the below region

(d1 ) : y = x (d2 ) : y = −2x

5
6
Multivariable Calculus
This branch of calculus extends the concepts of single-variable calculus to functions of several
variables.

(1) Functions of Several Variables:

Functions of several variables are prevalent in various fields of science like engineering, physics
and economics.

1) Electric Potential in Electrostatics:

The electric potential V at a point P (x, y, z) in space due to the distribution of charge q at the
origin is:
1 q
V (x, y, z) = ·p
4πϵ0 x + y2 + z2
2

2) Stress Distribution in a Rectangular Beam:

Consider a rectangular beam of inertia I subjected to a bending moment M . The normal stress
σ at any point within the cross-section of the beam depends on the distance y from the neutral
axis and the distance z along the length of the beam.
M
σ(y, z) = · yz
I

3) Cobb-Douglas Production Function:

The quantity of production Q depends on the labor L and capital K.

Q(L, K) = ALα K β

Definition:
Ley y = f (x1 , x2 , · · · xn ) where f : Rn → R
x1 , x2 · · · xn are called the independent variables, whereas y is the dependent variable.
The function takes n real numbers and outputs one real number in return.
For this course, we will mostly deal with functions of 2 variables.

Important:

Let z = f (x, y) be a general function of 2 variables, where Df ⊆ R2

The image of any point (x, y) ∈ Df , is a point in 3d space of coordinates P (x, y, z)
The general shape traced by this function is called a surface.

1) Explicitly: z = f (x, y)

x = f (u, v)

2) Implicitly: F (x, y, z) = 0 3) Parametrically : y = g(u, v)

z = h(u, v)


7
(2) Domain of Functions:
In this context, we look at domain of functions of 2 variables, namely z = f (x, y)

For functions of one variable f (x), we represented the domain Df on a number line. We say
that the domain of f lives in R, or at least a subset.

For functions of two variables f (x, y), the domain will live in R × R or R2 , or at least a subset.
For such functions, not only will we specify the domain, but also sketch it in the xy-plane.

Examples:
x+y
1) Consider the function f (x, y) =
y − x2
For our restrictions, we have that y − x2 ̸= 0 ⇐⇒ y ̸= x2
These are all the points in the plane, that are not on the parabola (P ) : y = x2
Formally we write Df = (x, y) ∈ R2 : y ̸= x2


ln(xy)
2) For this example, we will find the domain of f (x, y) = p
4 − x2 − y 2
For our restrictions, we have xy > 0 and 4 − x2 − y 2 > 0
Df = (x, y) ∈ R2 : xy > 0 & x2 + y 2 < 4


Give yourself some time to understand the restrictions.

When we say the product of two numbers x and y is positive, we really mean that these 2
numbers have the same sign, meaning the corresponding region of x and y are the first and
third quadrants.
For the second restriction, we have that x2 + y 2 < 4, which is the area limited inside the circle
centered at O and of radius 2.

8
(3) Range of Functions:
As discussed previously, z = f (x, y) is an equation of a surface, an object that lives in 3d space.
However, we ask the question: Does this surface cover the entirety of R.
In other words, if I solve f (x, y) = c, can I always find at least 1 z?
The answer is of course not, similar to functions of 1 variable, y = f (x) = x2 does not cover
the negative numbers, as it is not in the range of f .
A similar notion appears for functions of several variables.
Examples:

1) Consider the surface z = f (x, y) = x2 + y 2

This equation describes a paraboloid, notice that x2 + y 2 is non-negative, implying z > 0

The ground in orange is that of z = 0, namely the xy-plane.

Notice how the surface is always above that plane, simply because z > 0.
Then we say that Rf = [0, ∞)

2) We do not always need to visualize surfaces to find their range.

x2
Take z = g(x, y) =
x2 + y 2
Notice that since everything is squared, we have that g(x, y) > 0 or z > 0
However, nothing prevents it from being 0, notice that g(0, 1) = 0 then {0} ∈ Rf .
x2
We also know that x2 + y @ ≥ x2 =⇒ ≤1
x2 + y 2
Then Rg = [0, 1]

9
(4) Level Curves:
z = f (x, y) is the equation of a surface.
z = c is an equation of a horizontal plane parallel to the xy-plane.
We can cut the surface with this plane at level of z, equivalently we are solving f (x, y) = c.
But f (x, y) = c for z = c is of the form F (x, y) = 0 for z = c, meaning we are really looking at
a curve that is elevated c units upwards.
This curve on the surface, for this particular value of z = c is called a level curve.

Examples:
1) Take the function z = f (x, y) = y − x2

The surface (blue) is cut by two planes, one at the level z = 0 and one at z = −20
It’s hard to notice but the curve traced out by this upper cut is a parabola, and so is the lower.
We do not need to visualize the surface to know that the general level curves are parabolas.
Solving f (x, y) = c ⇐⇒ y − x2 = c ⇐⇒ y = x2 + c
That is an equation of a parabola being shifted vertically.
I can project these level curves on the xy-plane to get a better visual.

2) We will describe the level curves for z = g(x, y) = x2 + y 2

Solving g(x, y) = c ⇐⇒ x2 + y 2 = c
This is an equation of a circle as long as c > 0
Important Note:
Since really z = c tells us when cutting the surface what we get, if g(x, y) = c does have a
solution, then really we are saying that c ∈
/ Rg
Namely, the restrictions on c, can tell the range of the function.

10
Contour Plots:

Definition:
The plot where the level curves are projected and represented on the xy-plane.
Take or instance z = f (x, y) = x2 + y 2 + 1
Solving f (x, y) = c ⇐⇒ x2 + y 2 = c − 1
This is an equation of a circle for c > 1 and the origin for c = 1
c = 1 =⇒ x2 + y 2 = 0 =⇒ x = y = 0 O(0, 0)
c = 2 =⇒ x2 + y 2 = 1 circle (C2 ) : (O, 1)
c = 5 =⇒ x2 + y 2 = 4 circle (C5 ) : (O, 2)

The contour plot for c ∈ [0, 1]:

The surface and some of its level curves:

11
(5) Differentiation in Several Variables:

Recall the definition of a derivative for single variable functions:

f (x + h) − f (x)
f ′ (x) = lim
h→0 h
We look at how much the function changes as we have infinitesimal change in x.

This notion still holds for functions of several variables, however, we cannot be changing several
variables to observe the change in the function.
In other words, we fixate all variables except one, and look at the change in the function as we
change that one variable.
Taking the derivative ”with respect to” one variable x1 is referred to as the partial derivative
of f with respect to x1
Denoted by ”∂”, pronounced as ”del” and is derived from the Greek letter ”delta”, we write:
∂f
∂x1
For function of two variables f (x, y)

∂f f (x + h, y) − f (x, y)
(x, y) = lim
∂x h→0 h
∂f f (x, y + h) − f (x, y)
(x, y) = lim
∂y h→0 h

For single variable functions, f ′ (a) measures the slope of the tangent drawn to f at x = a.
∂f
For a function of two variables, (a, b) measures the slope of the tangent line of the level curve
∂x
generated by cutting z = f (x, y) with the plane y = b at x = a
Visual:

12
(5.1) Notations:

1) The Leibnitz Notation:

Consider a general function of 2 variables f (x, y)

∂f
refers to first order partial derivative of f (x, y) wrt x
∂x
∂f
refers to first order partial derivative of f (x, y) wrt y
∂y
∂2f
 
∂ ∂f
= refers to second order partial derivative of f (x, y) wrt x
∂x ∂x ∂x2
∂2f
 
∂ ∂f
= refers to second order partial derivative of f (x, y) wrt y
∂y ∂y ∂y 2
∂2f
 
∂ ∂f ∂f
= refers to first order partial derivative of (x, y) wrt y
∂y ∂x ∂y∂x ∂x

2) The Subscript Notation:

Consider a general function of 2 variables f (x, y)

fx refers to first order partial derivative of f (x, y) wrt x

fy refers to first order partial derivative of f (x, y) wrt y
fxx refers to second order partial derivative of f (x, y) wrt x
fyy refers to second order partial derivative of f (x, y) wrt y
fxy refers to first order partial derivative of fx (x, y) wrt y

Example:

Let f (x, y) = x2 y + 2 sin(y)

We are going to compute all partial derivatives up to second order.
∂f
= fx = 2xy Notice how y is treated as a constant when diff. wrt x
∂x
∂f
= fy = x2 + 2 cos(y)
∂y
∂2f ∂
= fxx = (2xy) = 2y
∂x2 ∂x
∂2f ∂ 2
= fyy = (x + 2 cos(y)) = −2 sin(y)
∂y 2 ∂y
∂2f ∂ 2
= fyx = (x + 2 cos(y)) = 2x
∂x∂y ∂x
∂2f ∂
= fxy = (2xy) = 2x
∂y∂x ∂y

13
(5.2) The Mixed Derivative Theorem:

Let f (x, y) be a function having continuous second order mixed derivatives fxy and fyx in a
region then:
fxy (x, y) = fyx (x, y)

In general, order does not matter as long as we take n derivatives wrt x and m wrt y.
example: fxyxx = fyxxx = fxxyx = fxxxy

Worked out example:

1) We will verify the mixed derivative theorem for the function f (x, y) = x2 y ln(xy)
 
∂ 2 2 1
fx = y · (x ln(xy)) = y · 2x · ln(xy) + x · = 2xy ln(xy) + xy
∂x x
 
∂ 1
fxy = x · (2y ln(xy) + y) = x 2 ln(xy) + 2y · + 1 = 2x ln(xy) + 3x
∂y y
 
∂ 1
fy = x2 · (y ln(xy)) = x2 ln(xy) + y · = x2 ln(xy) + x2
∂y y
 
∂ 2 2 2 1
fyx = (x ln(xy) + x ) = 2x ln(xy) + x · + 2x = 2x ln(xy) + 3x
∂x x

2) Starting from the mixed derivative theorem, namely fxy = fyx , we will show that fxxy = fyxx
We know that fxxy = (fx )xy = (fx )yx , since the theorem also applies to the function fx .
Hence, fxxy = fxyx
But fxyx = (fxy )x = (fyx )x , since we know that fxy = fyx
Hence fxyx = fyxx , and the proof is complete.

3) The equation fx + cfy = 0 is called a transport equation.

 x
The general solution is f (x, y) = g y −
c
Consider the case where c = −1, we get fx = fy , whose solution is a function g(x + y)
We will show that if fx = fy , then all the derivatives of f of same order n ≥ 2 are equal.
Starting from the fact that fx = fy and differentiating wrt x =⇒ fxx = fyx
Similar argument leads to the fact that fyy = fxy
But from the mixed derivative theorem, we have that fxy = fyx =⇒ fxx = fyy = fxy = fyx
The statement is then verified for n = 2.
Assume that for k ≥ 2 we have that fxx · · · x yy · · · y = fxx · · · x yy · · · y for some 0 ≤ i, j ≤ k
| {z } | {z } | {z } | {z }
i times k−i times j times k−j times

Differentiating both sides wrt y =⇒ fxx · · · x yy · · · y = fxx · · · x yy · · · y (true for k + 1)

| {z } | {z } | {z } | {z }
i times k+1−i times j times k+1−j times

From the mixed derivative theorem, we know that order does not matter, implying any partial
derivative of order n for any 0 ≤ i, j ≤ n, we have these partial derivatives being equal.

14
(6) Limits:
Recall from single variable calculus the notion of limits:
We write lim f (x) = L ⇐⇒ lim f (x) = lim f (x) = L
x→c x→c+ x→c−
We say that as x becomes arbitrarily small close to c, the values that the function f (x) gets
arbitrarily close to L. This has to hold whether x approaches c from ”the left side” or from
”the right side”.

For functions of two variables f (x, y), we are interested in the behaviour of f as we approach
any point (a, b) ∈ Df in a certain ”direction”.

Definition:
A path for functions of 2 variables is a curve in the xy-plane that passes through a point (a, b)
This path of the form F (x, y) = 0 satisfies then the relation F (a, b) = 0
If we can express y explicitly as y = g(x), then we can say that:

lim f (x, y) = lim f (x, g(x)) = lim h(x)

(x,y)→(a,b) x→a x→a

Note 1:
We say that lim f (x, y) = L ⇐⇒ it’s true along any path.
(x,y)→(a,b)

Note 2:
We say that lim f (x, y) = DN E ⇐⇒ ∃ at 2 distinct paths y = g1 (x) and y = g2 (x) :
(x,y)→(a,b)

lim f (x, g1 (x)) ̸= lim f (x, g2 (x))

x→a x→a

Important note:
Proving that a limit exists cannot be done using paths, however, if we can find two paths that
do not lead to the same answer, then the limit simply does not exist.
Methods to actually find limits are discussed later in the pdf.
examples:
xy − 2
1) Find lim √ √
(x,y)→(0,0) x+ y+3
xy − 2 2
Through direct substitution, we notice that lim √ √ =−
(x,y)→(0,0) x+ y+3 3

xy
2) Let f (x, y) = . Find lim or show that it does not exist.
x2 + y2 (x,y)→(0,0)

solution:
0
Along x = 0; lim f (0, y) = lim = lim 0 = 0
y→0 y→0 y 2 y→0

x2 1 1
Along y = x; lim f (x, x) = lim 2 2
= lim = ̸= 0
x→0 x→0 x + x x→0 2 2
Thus lim f (x, y) = DN E
(x,y)→(0,0)

15
Generalized Paths:

Definition:
A general path is a family of curves that pass through a particular point (a, b)

General Form:
Fm (x, y) = 0 : Fm (a, b) = 0 ∀ m
If we can isolate x or y, we get x = fm (y) or y = gm (x)
example:
For the point (0, 0), which mostly what we deal with, some general paths would be:

y = mx y = mx2 x = my 2 y = xm

Importance:

1) If lim f (x, y) = lim f (x, gm (x)) = h(m), then we say that lim f (x, y) = DN E
(x,y)→(0,0) x→0 (x,y)→(0,0)

Since the result is a function of m, we can plug two values of m, getting h(m1 ) ̸= h(m2 ) which
is really just checking two paths but in a manner where I do not have to look for different paths,
but rather take paths that differ by this m.

Worked out example:

x2
Consider the function f (x, y) = , we are interested in lim f (x, y)
x2 + y 2 (x,y)→(0,0)

m2 y 2 m2
Along the general path x = my; lim f (my, y) = lim = = h(m)
y→0 y→0 m2 y 2 + y 2 m2 + 1
Then we say that lim f (x, y) = DN E
(x,y)→(0,0)
1
This is because for instance, h(1) = ̸= h(0) = 0
2
1
It’s as if we’re saying try along x = y, you’ll get , then try along x = 0, you’ll get 0, which are
2
two different answers, meaning the limit does not exist.
You might think for this example you could have guessed the paths, which is fine, but for harder
examples, this saves time trying to think of paths that lead to different answers.

2) If lim f (x, y) = lim f (x, gm (x)) = L, then we cannot know anything.

(x,y)→(0,0) x→0

Since we get a fixed number L for any m, this does not mean that the limit is equal to L. You
just found infinite paths that lead to the same answer, but this does not imply that you’ve
checked all paths.
Worked out example:
4x2 y
Consider lim
(x,y)→(0,0) x4 + y 2

4mx3 4mx
Along y = mx; lim 4 2 2
= lim 2 = 0 for all m ̸= 0
x→0 x + m x x→0 x + m2

For m = 0; lim 0 = 0, still does not tell anything.

x→0
4x4 1
Along y = x2 ; lim = 4 4
= ̸= 0 =⇒ Limit DN E
x→0 x +x 2

16
Methods to find limits:
(1) Simplification of indeterminate forms:

This can be via rationalizing, taking a common factor· · ·

examples:
xy − y − x + 1 0
1) lim 2
= I.F.
(x,y)→(1,0) x −1 0
xy − y − x + 1 (x − 1)(y − 1) y−1 1
lim = lim = lim =−
(x,y)→(1,0) x2 − 1 (x,y)→(1,0) (x − 1)(x + 1) (x,y)→(1,0) x + 1 2

p
x2 + y 2 − 1 0
2) lim 2 2
= I.F.
(x,y)→(0,0) x + y − 1 0
p p p
x2 + y 2 − 1 x2 + y 2 − 1 x2 + y 2 − 1
lim = lim = lim
(x,y)→(0,0) x2 + y 2 − 1
p p p
(x,y)→(0,0) ( x2 + y 2 )2 − 12 (x,y)→(0,0) ( x2 + y 2 − 1)( x2 + y 2 + 1)

1 1
lim p = =1
(x,y)→(0,0) 2 2
x +y +1 0+1

(2) The Squeeze Theorem:

Theorem:
If we have the following conditions satisfied simultaneously:
1) h(x, y) ≤ f (x, y) ≤ g(x, y) for (x, y) ∈ Df
2) lim h(x, y) = lim g(x, y) = L
(x,y)→(a,b) (x,y)→(a,b)

Then by the squeeze theorem, we have that lim f (x, y) = L

(x,y)→(a,b)

example:
x3 − 3xy 2
Show using the squeeze theorem that lim =0
(x,y)→(0,0) x2 + y 2

solution:
x3 − 3xy 2
 2
x − 3y 2 x2
  
Notice that =x ≤x ≤ x(1) = x
x2 + y 2 x2 + y 2 x2 + y 2
x3 − 3xy 2
 2
x − 3y 2 −3x2 − 3y 2
  
Also ≥x ≥x = −3x
x2 + y 2 x2 + y 2 x2 + y 2
x3 − 3xy 2
We come to the conclusion that −3x ≤ ≤x
x2 + y 2
x3 − 3xy 2
But lim −3x = lim x = 0 =⇒ lim = 0 by the squeeze theorem.
(x,y)→(0,0) (x,y)→(0,0) (x,y)→(0,0) x2 + y 2

17
(3) Polar Coordinates:
Consider a general function of 2 variables f (x, y)
The lim f (x, y) can be changed via the transformation of coordinates to polar.
(x,y)→(0,0)

lim f (x, y) = lim f (r cos θ, r sin θ) = lim g(r, θ)

(x,y)→(0,0) r→0 r→0
∀θ ∀θ

1) We say that lim f (x, y) = DN E if lim f (r cos θ, r sin θ) = h(θ)

(x,y)→(0,0) r→0
∀θ
example:
x2 + y 2
Use polar coordinates to show that lim does not exist.
(x,y)→(0,0) x2 + y 3

solution:
x2 + y 2 r2 1 1
lim 2 3
= lim 3 = lim 3 = = h(θ)
(x,y)→(0,0) x + y 2 2 3
r→0 r cos (θ) + r sin (θ) 2
r→0 cos (θ) + r sin (θ) cos2 (θ)
∀θ ∀θ

x2 + y2
Then lim = DN E
(x,y)→(0,0) x2 + y3

2) We say that lim f (x, y) = L ⇐⇒ lim f (r cos θ, r sin θ) = L

(x,y)→(0,0) r→0
∀θ
example:
Use polar coordinates to find lim y ln(x2 + y 2 )
(x,y)→(0,0)

solution:
lim y ln(x2 + y 2 ) = lim r sin θ ln(r2 ) = (2 sin θ) lim {r ln |r|} = (2 sin θ)(0) = 0
(x,y)→(0,0) r→0 r→0
∀θ ∀θ

Thus, lim y ln(x2 + y 2 ) = 0

(x,y)→(0,0)

Recall:
If lim |f (x)| = 0 =⇒ lim f (x) = 0
x→c x→c
proof:

−|f (x)| ≤ f (x) ≤ |f (x)|

lim −|f (x)| ≤ lim f (x) ≤ lim |f (x)|
x→c x→c x→c

But lim |f (x)| = 0 =⇒ 0 ≤ lim f (x) ≤ 0

x→c x→c

By the squeeze theorem, lim f (x) = 0

x→c

For application:
0 ≤ |(2 sin θ)r ln(r)| ≤ 2r ln(r)

18
(4) Maclaurin Series

The Maclaurin series for a function of two variables f (x, y) is an extension of the single-variable
Maclaurin series to functions of two variables.
The Maclaurin series expansion of f (x, y) is given by:
∞ X
∞
!
X 1 ∂ (n+m) f
f (x, y) = (0, 0) xn y m
n!m! ∂xn ∂y m
n=0 m=0

Generally, if f (x, y) is analytic in some neighborhood of (0, 0), then the Maclaurin series will
converge to f (x, y) in some disk centered at (0, 0).
We can use this to our advantage to compute limits near the point (0, 0)
In the case where f (x, y) = g(x)h(y), the Maclaurin series expansion for f is given by:
∞
! ∞ !
X g (n) (0) n X h(n) (0)
n
f (x, y) = x y
n! n!
n=0 n=0
examples:
sin(x) + sin(y)
1) Use Maclaruin series to prove that lim =1
(x,y)→(0,0) x+y
solution:
∞
X x2n+1
Recall that sin(x) = (−1)k
(2n + 1)!
n=0
2n
X
(x2n+1 + y 2n+1 ) = (x + y) (−1)n x2n−k y k
k=0
sin(x) + sin(y)
If f (x, y) = , then:
x+y
x3 x5 y3 y5 1 3 1
x− 3! + 5! − ··· + y − 3! + 5! − ··· (x + y) − 3! (x + y3) + 5! (x
5 + y5) − · · ·
f (x, y) = =
x+y x+y
1
(x + y) − + y)(x2 − xy + y 2 ) + 5!1 (x4 − x3 y + x2 y 2 − xy 3 + y 4 ) − · · ·
3! (x
f (x, y) =
x+y
1 1
f (x, y) = 1 − (x2 − xy + y 2 ) + (x3 − x2 y + x2 y 2 − xy 3 + y 4 ) + · · ·
3! 5!
1 1
lim f (x, y) = 1 − (0) + (0) − · · · = 1
(x,y)→(0,0) 3! 5!

2 −y
2) Find the first few terms of the expansion for f (x, y) = ex around (0, 0)
solution:
x4 y2 y3
  
2 1
f (x, y) = ex ·e−y 2
= 1+x + + ··· 1−y+ − + · · · = 1−y+x2 + y 2 −x2 y+· · ·
2! 2! 3! 2

19
LaTeXed by Jawad Ibrahim
Summer 2024

20