Partial Derivatives

Partial Derivatives

Introduction

So far we have dealt with the calculus of functions of a single variable.

But, in the real world, physical quantities often depend on two or more
variables, so we turn our attention to functions of several variables and
extend the basic ideas of differential calculus to such functions.

Functions of Two Variables

The temperature 𝑻 at a point on the surface of the earth at any given time
depends on the longitude 𝒙 and latitude 𝒚 of the point. We can think of 𝑻 as
being a function of the two variables 𝒙 and 𝒚. We indicate this functional
dependence by writing

𝑇 = 𝑓(𝑥, 𝑦).

The volume 𝑽 of a circular cylinder depends on its radius 𝒓 and its height 𝒉. In
fact, we know that

𝑉 = 𝜋 𝑟 2 ℎ.

We say that 𝑽 is a function of 𝒓 and 𝒉, and we write

𝑉(𝑟, ℎ) = 𝜋 𝑟 2 ℎ.

1
 Partial Derivatives

Definition: A function 𝒇 of two variables is a rule that assigns to each

ordered pair of real numbers (𝑥, 𝑦) in a set 𝑫 a unique real number denoted
by 𝑓(𝑥, 𝑦). The set 𝑫 is the domain of 𝒇 and its range is the set of values
that 𝒇 takes on.

We often write

𝑧 = 𝑓(𝑥, 𝑦)

where the variables 𝒙 and 𝒚 are independent variables and 𝑧 is the

dependent variable. Compare this with the notation 𝑦 = 𝑓(𝑥) for functions
of a single variable. A function of two variables is just a function whose domain
is a subset of ℝ2 and whose range is a subset of ℝ.

ℝ𝟐 ℝ

2
 Partial Derivatives

Another way of visualizing the behavior of a function of two variables is to

consider its graph. The graph of a function of one variable is a curve 𝑪 with
equation 𝑦 = 𝑓(𝑥), so the graph of a function of two variables is a surface 𝑺
with equation 𝑧 = 𝑓(𝑥, 𝑦). We can visualize the graph 𝑺 of 𝒇 as lying directly
above or below its domain 𝑫 in the 𝑥𝑦 − plane.

ℝ𝟑

Find the domain and the range of the function

𝑧 = sin(𝑥 2 + 𝑦 2 )

Solution

Domain: All the 𝑥𝑦 − plane i.e. (𝑥, 𝑦) ∈ ℝ2

Range: 𝑧 ∈ [−1,1]

3
 Partial Derivatives

Find the domain and the range of the function 𝑧 = ln(𝑥 + 𝑦).
Sketch the domain in the 𝑥𝑦 − plane.

Solution y

Domain: All points in the 𝑥𝑦 −

plane such that The domain

x + y =0
𝑥+𝑦 >0
x

Range: 𝑧 ∈ ℝ

Find the domain and the range of the function 𝑧 = sin−1 (𝑥 + 𝑦).
Sketch the domain in the 𝑥𝑦 − plane.

Solution y

Domain: All points in the 𝑥𝑦 −

1 x + y =1
plane such that
1 x
−1 ≤ 𝑥 + 𝑦 ≤ 1 -1

x + y = −1 Domain
-1
Range: 𝑧 ∈ [− 𝜋⁄2 , 𝜋⁄2]

4
 Partial Derivatives

Find the domain and the range of the function 𝑧 = cos −1 (𝑥 2 + 𝑦 2 ).

Sketch the domain in the 𝑥𝑦 − plane.

Solution
y
Domain: All points in the 𝑥𝑦 −
plane such that
Domain 1
2 2 x
0≤𝑥 +𝑦 ≤1

The points on and inside the unit circle

-1
x 2 + y 2 =1
Range: 𝑧 ∈ [0, 𝜋⁄2]

Find the domain and the range of the function 𝑧 = ln( 𝑦 − cosh 𝑥).
Sketch the domain in the 𝑥𝑦 − plane.

Solution

Domain: All points in the 𝑥𝑦 −

plane such that

𝑦 − cosh 𝑥 > 0
𝑦 = cosh 𝑥
∴ 𝑦 > cosh 𝑥

Range: 𝑧 ∈ ℝ

5
 Partial Derivatives

Partial Derivatives

Partial derivatives measure the rate of change of a function with respect to

one of the variables, keeping all other variables fixed.

In general, if 𝒇 is a function of two variables 𝒙 and 𝒚, suppose we let only

𝒙 vary while keeping 𝒚 fixed (constant). Then we are really considering a
function of a single variable 𝒙. If this function has a derivative, then we call it
the partial derivative of 𝑓 with respect to 𝑥 and denote it by

𝜕𝑓
= 𝑓𝑥
𝜕𝑥

Similarly, the partial derivative of 𝑓 with respect to 𝑦, denoted by

𝜕𝑓
= 𝑓𝑦
𝜕𝑦

which is obtained by keeping 𝒙 fixed (constant). To distinguish partial

derivatives from ordinary derivatives we used the symbol 𝝏 rather than 𝒅
previously used.

Definition: If 𝒇 is a function of two variables 𝒙 and 𝒚, its partial

derivatives are the functions 𝒇𝒙 and 𝒇𝒚 defined by

𝜕𝑓 𝑓(𝑥 + ∆𝑥, 𝑦) − 𝑓(𝑥, 𝑦)

= 𝑓𝑥 (𝑥, 𝑦) = lim
𝜕𝑥 ∆𝑥 → 0 ∆𝑥

𝜕𝑓 𝑓(𝑥, 𝑦 + ∆𝑦) − 𝑓(𝑥, 𝑦)

= 𝑓𝑦 (𝑥, 𝑦) = lim
𝜕𝑦 ∆𝑦 → 0 ∆𝑦

6
 Partial Derivatives

Find the first partial derivatives for the function

𝑓(𝑥, 𝑦) = 𝑥 4 + 𝑦 3 + sin 𝑥 tanh 𝑦 + 6

Solution

To find 𝒇𝒙 , keep 𝒚 as a constant and differentiate 𝒇(𝒙, 𝒚) with respect to 𝒙

𝑓𝑥 = 4𝑥 3 + 0 + cos 𝑥 tanh 𝑦 + 0

To find 𝒇𝒚 , keep 𝒙 as a constant and differentiate 𝒇(𝒙, 𝒚) with respect to 𝒚

𝑓𝑦 = 0 + 3𝑦 2 + sin 𝑥 sech2 𝑦 + 0

Find the first partial derivatives for the function

𝑓(𝑥, 𝑦) = (𝑥)𝑦 + 𝑒 𝑥 sec 𝑥 tan−1 𝑦

Solution

𝑓𝑥 = 𝑦 (𝑥)𝑦−1 + [ 𝑒 𝑥 sec 𝑥 tan 𝑥 + 𝑒 𝑥 sec 𝑥 ] tan−1 𝑦

1
𝑓𝑦 = (𝑥)𝑦 ln 𝑥 + 𝑒 𝑥 sec 𝑥 [ ]
1 + 𝑦2

7
 Partial Derivatives

Find the first partial derivatives for the function

𝑓(𝑥, 𝑦) = (ln 𝑦)(sinh 𝑥)

Solution

𝑓𝑥 = (ln 𝑦)(sinh 𝑥) . ln(ln 𝑦) . cosh 𝑥

1
𝑓𝑦 = sinh 𝑥 . (ln 𝑦)(sinh 𝑥 −1) .
𝑦

Find the first partial derivatives for the function

𝑓(𝑥, 𝑦) = cosh(𝑥 + 𝑦 2 ) + sin−1 (2𝑥 ln 𝑦)

Solution

1
𝑓𝑥 = sinh(𝑥 + 𝑦 2 ) + [ 2𝑥 ln 2 . ln 𝑦 ]
√1 − (2𝑥 ln 𝑦)2

1 1
𝑓𝑦 = 2𝑦 sinh(𝑥 + 𝑦 2 ) + [ 2𝑥 . ]
√1 − (2𝑥 ln 𝑦)2 𝑦

8
 Partial Derivatives

Find 𝑧𝑥 and 𝑧𝑦 for the function

𝑥5 𝑦8
𝑧= (𝑥)ln 𝑥 + 3𝑦
𝑒 + tan 𝑥
Solution

ln 𝑥 2
∵ (𝑥)ln 𝑥 = 𝑒 ln[ (𝑥) ]
= 𝑒 [ ln 𝑥 . ln 𝑥 ]
= 𝑒 (ln 𝑥)

(ln 𝑥)2
𝑥5 𝑦8
∴𝑧=𝑒 + 3𝑦
𝑒 + tan 𝑥

(ln 𝑥)2
1 (𝑒 3𝑦 + tan 𝑥)(5𝑥 4 𝑦 8 ) − (𝑥 5 𝑦 8 )(sec 2 𝑥)
𝑧𝑥 = 𝑒 . 2(ln 𝑥) . +
𝑥 (𝑒 3𝑦 + tan 𝑥)2

(𝑒 3𝑦 + tan 𝑥)(8𝑥 5 𝑦 7 ) − (𝑥 5 𝑦 8 )(3𝑒 3𝑦 )

𝑧𝑦 =
(𝑒 3𝑦 + tan 𝑥)2

9
 Partial Derivatives

Geometric interpretation of partial derivatives

We recall that the equation 𝑧 = 𝑓(𝑥, 𝑦) represents a surface 𝑺. If 𝑓(𝑎, 𝑏) =

𝑐, then the point 𝑃(𝑎, 𝑏, 𝑐) lies on 𝑺. By fixing 𝑦 = 𝑏, we are restricting our
attention to the curve 𝑪𝟏 in which the vertical plane 𝑦 = 𝑏 intersects 𝑺.
Likewise, the vertical plane 𝑥 = 𝑎 intersects 𝑺 in a curve 𝑪𝟐 . Both of the curves
𝑪𝟏 and 𝑪𝟐 pass through the point 𝑃.

Thus the partial derivatives 𝑓𝑥 (𝑎, 𝑏) and 𝑓𝑦 (𝑎, 𝑏) can be interpreted

geometrically as the slopes of the tangent lines 𝑻𝟏 and 𝑻𝟐 at 𝑃(𝑎, 𝑏, 𝑐) to

the curves 𝑪𝟏 and 𝑪𝟐 of 𝑺 in the planes 𝑦 = 𝑏 and 𝑥 = 𝑎, respectively.

10
 Partial Derivatives

Functions of More Than Two Variables

Partial derivatives can also be defined for functions of three or more

variables. For example, if 𝒇 is a function of three variables 𝒙, 𝒚 and 𝒛, then its
partial derivative with respect to 𝒙 is defined as

𝜕𝑓 𝑓(𝑥 + ∆𝑥, 𝑦, 𝑧) − 𝑓(𝑥, 𝑦, 𝑧)

= 𝑓𝑥 (𝑥, 𝑦, 𝑧) = lim
𝜕𝑥 ∆𝑥 → 0 ∆𝑥

and it is found by keeping 𝒚 and 𝒛 as constants and differentiating 𝒇(𝒙, 𝒚, 𝒛)

with respect to 𝒙.

In general, if 𝒖 is a function of 𝒏 variables, 𝑢 = 𝑓(𝑥1 , 𝑥2 , ⋯ , 𝑥𝑘 , ⋯ , 𝑥𝑛 ),

its partial derivative with respect to the 𝑘 th variable 𝒙𝒌 is

𝜕𝑓
= 𝑓𝑥𝑘 (𝑥1 , 𝑥2 , ⋯ , 𝑥𝑘 , ⋯ , 𝑥𝑛 )
𝜕𝑥𝑘

𝑓(𝑥1 , 𝑥2 , ⋯ , 𝑥𝑘 + ∆𝑥𝑘 , ⋯ , 𝑥𝑛 ) − 𝑓(𝑥1 , 𝑥2 , ⋯ , 𝑥𝑘 , ⋯ , 𝑥𝑛 )

= lim
∆𝑥𝑘 → 0 ∆𝑥𝑘

Find the first partial derivatives for the function

𝑓(𝑥, 𝑦, 𝑧) = 𝑦𝑒 (𝑥 𝑦) ln 𝑧 + cot(𝑧 sinh−1 𝑥)

Solution

To find 𝒇𝒙 , keep 𝒚 and 𝒛 as constants and differentiate 𝒇(𝒙, 𝒚, 𝒛) with respect

to 𝒙

11
 Partial Derivatives

𝑧
𝑓𝑥 = 𝑦 2 𝑒 (𝑥 𝑦) ln 𝑧 − csc 2 (𝑧 sinh−1 𝑥) .
√1 + 𝑥 2

To find 𝒇𝒚 , keep 𝒙 and 𝒛 as constants and differentiate 𝒇(𝒙, 𝒚, 𝒛) with respect

to 𝒚

𝑓𝑦 = [𝑥 𝑦 𝑒 (𝑥 𝑦) + 𝑒 (𝑥 𝑦) ] ln 𝑧 + 0

To find 𝒇𝒛 , keep 𝒙 and 𝒚 as constants and differentiate 𝒇(𝒙, 𝒚, 𝒛) with respect

to 𝒛

𝑦 𝑒 (𝑥 𝑦)
𝑓𝑧 = − csc 2 (𝑧 sinh−1 𝑥) . sinh−1 𝑥
𝑧

Find 𝑓𝑤 for the function

csc 𝑦
𝑓(𝑤, 𝑥, 𝑦, 𝑧) = 𝑧 2 sec −1 𝑤 + + cosh(√𝑤 tan 𝑥)
𝑥 2 − ln 𝑧
Solution

To find 𝒇𝒘 , keep 𝒙, 𝒚 and 𝒛 as constants and differentiate 𝒇(𝒘, 𝒙, 𝒚, 𝒛) with

respect to 𝒘

1 1
𝑓𝑤 = 𝑧 2 . + 0 + sinh(√𝑤 tan 𝑥) . . tan 𝑥
𝑤 √𝑤 2 − 1 2 √𝑤

𝑧2 tan 𝑥 . sinh(√𝑤 tan 𝑥)

∴ 𝑓𝑤 = +
𝑤 √𝑤 2 − 1 2 √𝑤

12
 Partial Derivatives

Higher Derivatives

If 𝒇 is a function of two variables, then its partial derivatives 𝑓𝑥 and 𝑓𝑦 are also
functions of two variables, so we can consider their partial derivatives

(𝑓𝑥 )𝑥 , (𝑓𝑥 )𝑦 , (𝑓𝑦 ) and (𝑓𝑦 )

𝑥 𝑦

which are called the second partial derivatives of 𝒇. We can use the
following notation

𝜕 𝜕𝑓 𝜕2𝑓
(𝑓𝑥 )𝑥 = 𝑓𝑥𝑥 = ( )= 2
𝜕𝑥 𝜕𝑥 𝜕𝑥

𝜕 𝜕𝑓 𝜕2𝑓
(𝑓𝑥 )𝑦 = 𝑓𝑥𝑦 = ( )=
𝜕𝑦 𝜕𝑥 𝜕𝑦 𝜕𝑥

𝜕 𝜕𝑓 𝜕2𝑓
(𝑓𝑦 )𝑥 = 𝑓𝑦𝑥 = ( )=
𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦

𝜕 𝜕𝑓 𝜕2𝑓
(𝑓𝑦 )𝑦 = 𝑓𝑦𝑦 = ( )= 2
𝜕𝑦 𝜕𝑦 𝜕𝑦

Thus the notation

𝜕2𝑓
𝑓𝑥𝑦 =
𝜕𝑦 𝜕𝑥

means that we first differentiate with respect to 𝒙 and then with respect to 𝒚,
whereas in computing 𝑓𝑦𝑥 the order is reversed.

13
 Partial Derivatives

Find the second partial derivatives for the function

𝑓(𝑥, 𝑦) = 𝑥 3 𝑦 + sin 𝑦 + cosh 𝑥

Solution

𝑓𝑥 = 3𝑥 2 𝑦 + sinh 𝑥 𝑓𝑦 = 𝑥 3 + cos 𝑦

𝑓𝑥𝑥 = 6𝑥𝑦 + cosh 𝑥 𝑓𝑦𝑥 = 3𝑥 2

𝑓𝑥𝑦 = 3𝑥 2 𝑓𝑦𝑦 = − sin 𝑦

Find the fourth partial derivative 𝑓𝑥𝑥𝑦𝑧 for the function

𝑓(𝑥, 𝑦, 𝑧) = 𝑥 3 𝑒 2𝑦 + 𝑧 sinh 𝑥

Solution

𝑓𝑥 = 3 𝑥 2 𝑒 2𝑦 + 𝑧 cosh 𝑥

𝑓𝑥𝑥 = 6 𝑥 𝑒 2𝑦 + 𝑧 sinh 𝑥

𝑓𝑥𝑥𝑦 = 12 𝑥 𝑒 2𝑦

𝑓𝑥𝑥𝑦𝑧 = 0

Hint:

𝜕4𝑓
𝑓𝑥𝑥𝑦𝑧 =
𝜕𝑧 𝜕𝑦 𝜕𝑥 2

14
 Partial Derivatives

Note that:

In Example 13, 𝑓𝑥𝑦 = 𝑓𝑦𝑥 . This is not just a coincidence. The mixed
partial derivatives 𝑓𝑥𝑦 and 𝑓𝑦𝑥 are equal for most functions that one meets in

practice. The following theorem gives conditions under which we can assert
that 𝑓𝑥𝑦 = 𝑓𝑦𝑥 .

Clairaut’s Theorem: Suppose 𝒇 is defined on a disk 𝑫 that contains the

point (𝑎, 𝑏). If the functions 𝑓𝑥𝑦 and 𝑓𝑦𝑥 are both continuous on 𝑫, then

𝑓𝑥𝑦 (𝑎, 𝑏) = 𝑓𝑦𝑥 (𝑎, 𝑏)

Partial derivatives of order 3 or higher can also be defined. For instance,

𝜕 𝜕2𝑓 𝜕3𝑓
𝑓𝑥𝑦𝑦 = (𝑓𝑥𝑦 )𝑦 = ( )= 2
𝜕𝑦 𝜕𝑦 𝜕𝑥 𝜕𝑦 𝜕𝑥

and using Clairaut’s Theorem it can be shown that

𝑓𝑥𝑦𝑦 = 𝑓𝑦𝑥𝑦 = 𝑓𝑦𝑦𝑥

if these functions are continuous. So, in Example 14

𝑓𝑥𝑥𝑦𝑧 = 𝑓𝑥𝑦𝑥𝑧 = 𝑓𝑦𝑧𝑥𝑥 = 𝑓𝑦𝑥𝑧𝑥 = 𝑓𝑧𝑥𝑥𝑦

15
 Partial Derivatives

Chain Rule

For functions of more than one variable, the Chain Rule has several
versions, each of them giving a rule for differentiating a composite function.

▪ Case (1)

If 𝑧 = 𝑓(𝑥, 𝑦) and each of the variables 𝒙 and 𝒚 is, in turn, a function of a

variable 𝒕 such that

𝑥 = 𝑔(𝑡) and 𝑦 = ℎ(𝑡)

This means that 𝒛 is a composite function of the argument 𝒕,

𝑧 = 𝑓(𝑔(𝑡), ℎ(𝑡))

and the Chain Rule gives a formula for differentiating 𝒛 as a function of 𝒕.

𝑑𝑧 𝜕𝑧 𝑑𝑥 𝜕𝑧 𝑑𝑦
= . + .
𝑑𝑡 𝜕𝑥 𝑑𝑡 𝜕𝑦 𝑑𝑡

16
 Partial Derivatives

Find 𝑑𝑧⁄𝑑𝑡 at 𝑡 = 0 for

𝑧 = 𝑥 2 𝑦 + 3 𝑥 𝑦 4 where 𝑥 = sin 2𝑡 and 𝑦 = cos 𝑡.

Solution

The Chain Rule gives

𝑑𝑧 𝜕𝑧 𝑑𝑥 𝜕𝑧 𝑑𝑦
= . + .
𝑑𝑡 𝜕𝑥 𝑑𝑡 𝜕𝑦 𝑑𝑡

= (2 𝑥 𝑦 + 3 𝑦 4 )(2 cos 2𝑡) + (𝑥 2 + 12 𝑥 𝑦 3 )(− sin 𝑡)

It’s not necessary to substitute the expressions for 𝒙 and 𝒚 in terms of 𝒕. We

simply observe that when 𝑡 = 0, we have 𝑥 = 0 and 𝑦 = 1. Therefore

𝑑𝑧
| = (3)(2) + (0)(0) = 6
𝑑𝑡 𝑡=0

▪ Case (2)

Suppose that

𝑧 = 𝑓(𝑥, 𝑦)

is a differentiable function of 𝒙 and 𝒚, where

𝑥 = 𝜑(𝑠, 𝑡) and 𝑦 = 𝜓(𝑠, 𝑡)

are differentiable functions of 𝒔 and 𝒕. Then

17
 Partial Derivatives

𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑧 𝜕𝑦 𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑧 𝜕𝑦
= . + . = . + .
𝜕𝑠 𝜕𝑥 𝜕𝑠 𝜕𝑦 𝜕𝑠 𝜕𝑡 𝜕𝑥 𝜕𝑡 𝜕𝑦 𝜕𝑡

Find 𝜕𝑧⁄𝜕𝑠 and 𝜕𝑧⁄𝜕𝑡 for

2)
𝑧 = ln(𝑥 2 + 𝑦) where 𝑥 = 𝑒 (𝑠+𝑡 and 𝑦 = 𝑠 2 + 𝑡

Solution

Applying Case 2 of the Chain Rule, we get

𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑧 𝜕𝑦 2𝑥 2 1
= . + . =[ 2 ] [ 𝑒 (𝑠+𝑡 ) ] + [ 2 ] [ 2𝑠 ]
𝜕𝑠 𝜕𝑥 𝜕𝑠 𝜕𝑦 𝜕𝑠 𝑥 +𝑦 𝑥 +𝑦

𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑧 𝜕𝑦 2𝑥 2 1
= . + . =[ 2 ] [ 2𝑡 𝑒 (𝑠+𝑡 ) ] + [ 2 ][ 1 ]
𝜕𝑡 𝜕𝑥 𝜕𝑡 𝜕𝑦 𝜕𝑡 𝑥 +𝑦 𝑥 +𝑦

18
 Partial Derivatives

Hint:

Case 2 of the Chain Rule contains three types of variables: 𝒔 and 𝒕 are
independent variables, 𝒙 and 𝒚 are called intermediate variables and 𝒛 is

the dependent variable. To remember the Chain Rule, it’s helpful to draw the

tree diagram.

▪ General Version

Now we consider the general situation in which a dependent variable is a

function of 𝒏 intermediate variables, each of which is, in turn, a function of 𝒎
independent variables. Notice that there are 𝒏 terms, one for each intermediate
variable.

Write out the Chain Rule for the case where

𝑤 = 𝑓(𝑥, 𝑦, 𝑧, 𝑡) and

𝑥 = 𝑔(𝑢, 𝑣), 𝑦 = ℎ(𝑢, 𝑣), 𝑧 = 𝜓(𝑢, 𝑣) and 𝑡 = 𝜑(𝑢, 𝑣)

Solution

19
 Partial Derivatives

𝜕𝑤 𝜕𝑤 𝜕𝑥 𝜕𝑤 𝜕𝑦 𝜕𝑤 𝜕𝑧 𝜕𝑤 𝜕𝑡
= . + . + . + .
𝜕𝑢 𝜕𝑥 𝜕𝑢 𝜕𝑦 𝜕𝑢 𝜕𝑧 𝜕𝑢 𝜕𝑡 𝜕𝑢

𝜕𝑤 𝜕𝑤 𝜕𝑥 𝜕𝑤 𝜕𝑦 𝜕𝑤 𝜕𝑧 𝜕𝑤 𝜕𝑡
= . + . + . + .
𝜕𝑣 𝜕𝑥 𝜕𝑣 𝜕𝑦 𝜕𝑣 𝜕𝑧 𝜕𝑣 𝜕𝑡 𝜕𝑣

If 𝑢 = 𝑥 4 𝑦 + 𝑦 2 𝑧 3 , where

𝑥 = 𝑟 𝑠 𝑒 𝑡 , 𝑦 = 𝑟 𝑠 2 𝑒 −𝑡 and 𝑧 = 𝑟 2 𝑠 sin 𝑡

Find the value of 𝜕𝑢⁄𝜕𝑠 when 𝑟 = 2, 𝑠 = 1, 𝑡 = 0.

Solution

With the help of the tree diagram, we have

𝜕𝑢 𝜕𝑢 𝜕𝑥 𝜕𝑢 𝜕𝑦 𝜕𝑢 𝜕𝑧
= . + . + .
𝜕𝑠 𝜕𝑥 𝜕𝑠 𝜕𝑦 𝜕𝑠 𝜕𝑧 𝜕𝑠

= (4 𝑥 3 𝑦)(𝑟 𝑒 𝑡 ) + (𝑥 4 + 2 𝑦 𝑧 3 )(2 𝑟 𝑠 𝑒 −𝑡 ) + (3 𝑦 2 𝑧 2 )(𝑟 2 sin 𝑡)

When 𝑟 = 2, 𝑠 = 1 and 𝑡 = 0, we have 𝑥 = 2, 𝑦 = 2 and 𝑧 = 0, so

𝜕𝑢
= (64)(2) + (16)(4) + (0)(0) = 192
𝜕𝑠

20
 Partial Derivatives

Implicit Differentiation

The Chain Rule can be used to give a more complete description of the
process of implicit differentiation that was introduced in Sections 2.5. We
suppose that an equation of the form

𝐹(𝑥, 𝑦) = 0

defines 𝒚 implicitly as a differentiable function of 𝒙, that is, 𝑦 = 𝑓(𝑥), where

𝐹(𝑥, 𝑓(𝑥)) = 0

If 𝑭 is differentiable, we can apply Case 1 of the Chain Rule to differentiate

both sides of the equation 𝐹(𝑥, 𝑦) = 0 with respect to 𝒙. Since both 𝒙 and 𝒚
are functions of 𝒙, we obtain

𝜕𝐹 𝑑𝑥 𝜕𝐹 𝑑𝑦
. + . =0
𝜕𝑥 𝑑𝑥 𝜕𝑦 𝑑𝑥

But 𝑑𝑥⁄𝑑𝑥 = 1, so if 𝜕𝐹 ⁄𝜕𝑦 ≠ 0 we solve for 𝑑𝑦⁄𝑑𝑥 and obtain

𝑑𝑦 𝐹𝑥
=−
𝑑𝑥 𝐹𝑦

Find 𝑑𝑦⁄𝑑𝑥 If tan(𝑥 2 𝑦) = 𝑥 + 𝑥 𝑦 2 .

Solution

The given equation can be written as

𝐹(𝑥, 𝑦) = tan(𝑥 2 𝑦) − 𝑥 − 𝑥 𝑦 2 = 0

21
 Partial Derivatives

𝑑𝑦 𝐹𝑥 2 𝑥 𝑦 sec 2 (𝑥 2 𝑦) − 1 − 𝑦 2
=− =−
𝑑𝑥 𝐹𝑦 𝑥 2 sec 2 (𝑥 2 𝑦) − 2 𝑥 𝑦

𝑑𝑦 1 + 𝑦 2 − 2 𝑥 𝑦 sec 2 (𝑥 2 𝑦)
∴ =
𝑑𝑥 𝑥 2 sec 2 (𝑥 2 𝑦) − 2 𝑥 𝑦

Note that:

Now we suppose that 𝒛 is given implicitly as a function 𝑧 = 𝑓(𝑥, 𝑦) by an

equation of the form

𝐹(𝑥, 𝑦, 𝑧) = 0

This means that

𝐹(𝑥, 𝑦, 𝑓(𝑥, 𝑦)) = 0

If 𝑭 and 𝒇 are differentiable, then

𝜕𝐹 𝜕𝑥 𝜕𝐹 𝜕𝑦 𝜕𝐹 𝜕𝑧
. + . + . =0
𝜕𝑥 𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑧 𝜕𝑥

𝜕 𝜕
But (𝑥) = 1 and ( 𝑦) = 0
𝜕𝑥 𝜕𝑥

so this equation becomes

𝜕𝐹 𝜕𝐹 𝜕𝑧
+ . =0
𝜕𝑥 𝜕𝑧 𝜕𝑥

If 𝜕𝐹 ⁄𝜕𝑧 ≠ 0, we solve for 𝜕𝑧⁄𝜕𝑥 and obtain

𝜕𝑧 𝐹𝑥
=−
𝜕𝑥 𝐹𝑧

22
 Partial Derivatives

The formula for 𝜕𝑧⁄𝜕𝑦 is obtained in a similar manner

𝜕𝑧 𝐹𝑦
=−
𝜕𝑦 𝐹𝑧

Find 𝑧𝑥 and 𝑧𝑦 if 𝑦 𝑧 + 𝑥 ln 𝑦 = 𝑧 2

Solution

𝐹(𝑥, 𝑦, 𝑧) = 𝑦 𝑧 + 𝑥 ln 𝑦 − 𝑧 2 = 0

𝜕𝑧 𝐹𝑥 ln 𝑦 ln 𝑦
𝑧𝑥 = =− =− =
𝜕𝑥 𝐹𝑧 𝑦−2𝑧 2𝑧−𝑦

𝑥 𝑥
) 𝑧+(
𝑧 + (
𝜕𝑧 𝐹𝑦 𝑦 𝑦)
𝑧𝑦 = =− =− =
𝜕𝑦 𝐹𝑧 𝑦−2𝑧 2𝑧−𝑦

23