Topics:

• Second Order Partial Derivatives

• Higher Order Partial Derivatives
------------------------------------------------------
Second & Higher Order Partial Derivatives
Second-order partial derivatives are derivatives of a function of multiple variables,
where the differentiation is performed twice with respect to different variables.
These derivatives provide insight into the function's curvature and how the variables
interact.
 How to Compute 2nd Order Derivatives
We will compute the first order partial derivatives initially.
Consider the function 𝒇(𝒙, 𝒚), then there will be two first order partial
derivatives:
𝝏𝒇
1) 𝒇𝒙 =
𝝏𝒙
𝝏𝒇
2) 𝒇𝒚 =
𝝏𝒚

Now if we need to compute the second order partial derivatives:

We will use the solutions of the first order dervatives.
Since the function 𝒇(𝒙, 𝒚) was initially a two variable function thus we
will further compute the second derivative partially with respect to 𝒙 and
𝒚 using the solutions of the first order dervatives.
Thus, we will have four second order partial derivatives for 𝒇(𝒙, 𝒚):
𝝏 𝝏𝒇
➢ 𝒇𝒙𝒙 = ( )
𝝏𝒙 𝝏𝒙

The second derivative with respect to 𝒙 while holding 𝒚 constant. This

indicates the curvature of the function along the 𝒙-direction.

𝝏 𝝏𝒇
➢ 𝒇𝒚𝒚 = ( )
𝝏𝒚 𝝏𝒚

The second derivative with respect to 𝒚, holding 𝒙 constant.

 𝝏 𝝏𝒇 𝝏 𝝏𝒇
➢ 𝒇𝒙𝒚 = ( ) 𝒂𝒏𝒅 𝒇𝒚𝒙 = ( )
𝝏𝒚 𝝏𝒙 𝝏𝒙 𝝏𝒚

The mixed second derivative, where differentiation is first done with

respect to 𝒙 and then 𝒚 (or vice versa). This measures how the function
changes as both variables change.
Important Note:

• 𝒇𝒙𝒚 and 𝒇𝒚𝒙 are called the mixed partial derivatives.

• 𝒇𝒚𝒙 = 𝒇𝒙𝒚 , when function is continuous and twice differentiable.
• The variable which appears first is generally the one you would want
to differentiate with respect to first.
For example,
𝒇𝒚𝒙 means that you will first differentiate with respect to 𝒚, and then
will differentiate with respect to 𝒙.

Example 1:
Example 2:
Compute the 2nd order partial derivatives of
𝒇(𝒙, 𝒚) = 𝒙𝒚𝟐 + 𝟑𝒙𝟐 𝒆𝒚
Solution:
1st Order Partial Derivatives: 𝒇𝒙 and 𝒇𝒚
Finding partial derivative of 𝒇 Finding partial derivative of 𝒇
w.r.t 𝒙 w.r.t 𝒚
𝝏𝒇 𝝏 𝝏𝒇 𝝏
= (𝒙𝒚𝟐 + 𝟑𝒙𝟐 𝒆𝒚 ) = (𝒙𝒚𝟐 + 𝟑𝒙𝟐 𝒆𝒚 )
𝝏𝒙 𝝏𝒙 𝝏𝒚 𝝏𝒚
𝝏𝒇 𝝏 𝝏 𝟐 𝝏𝒇 𝝏 𝟐 𝝏 𝒚
= 𝒚𝟐 (𝒙) + 𝟑𝒆𝒚 (𝒙 ) =𝒙 (𝒚 ) + 𝟑𝒙𝟐 (𝒆 )
𝝏𝒙 𝝏𝒙 𝝏𝒙 𝝏𝒚 𝝏𝒙 𝝏𝒙
𝝏𝒇 𝝏𝒇
= 𝒚𝟐 (𝟏) + 𝟑𝒆𝒚 (𝟐𝒙) = 𝒙(𝟐𝒚) + 𝟑𝒆𝒚 𝒙𝟐
𝝏𝒙 𝝏𝒚

This implies that This implies that

𝒇𝒙 = 𝒚𝟐 + 𝟔𝒆𝒚 𝒙 𝒇𝒚 = 𝟐𝒙𝒚 + 𝟑𝒆𝒚 𝒙𝟐
2nd Order Partial Derivatives:

Finding 𝒇𝒙𝒙 and 𝒇𝒚𝒙

Finding partial derivative of 𝒇𝒙 Finding partial derivative of 𝒇𝒚
w.r.t 𝒙 w.r.t 𝒙
𝝏 𝝏𝒇 𝝏 𝟐 𝝏 𝝏𝒇 𝝏
( )= (𝒚 + 𝟔𝒆𝒚 𝒙) ( )= (𝟐𝒙𝒚 + 𝟑𝒆𝒚 𝒙𝟐 )
𝝏𝒙 𝝏𝒙 𝝏𝒙 𝝏𝒙 𝝏𝒚 𝝏𝒙
𝝏 𝝏 𝝏 𝟐
𝑓𝑥𝑥 = 0 + 6𝒆𝒚 (𝒙) 𝑓𝑦𝑥 = 2𝑦 (𝑥) + 3𝒆𝒚 (𝒙 )
𝝏𝒙 𝝏𝒙 𝝏𝒙
This implies that This implies that
𝒇𝒙𝒙 = 𝟔𝒆𝒚 𝒇𝒚𝒙 = 𝟐𝒚 + 𝟔𝒙𝒆𝒚

Similarly, finding 𝒇𝒙𝒚 and 𝒇𝒚𝒚 :

Finding partial derivative of 𝒇𝒙 Finding partial derivative of 𝒇𝒚
w.r.t 𝒚 w.r.t 𝒚
𝝏 𝝏𝒇 𝝏 𝟐 𝝏 𝝏𝒇 𝝏
( )= (𝒚 + 𝟔𝒆𝒚 𝒙) ( )= (𝟐𝒙𝒚 + 𝟑𝒆𝒚 𝒙𝟐 )
𝝏𝒚 𝝏𝒙 𝝏𝒚 𝝏𝒚 𝝏𝒚 𝝏𝒚
𝝏 𝟐 𝝏 𝝏 𝝏 𝒚
𝒇𝒙𝒚 = (𝒚 ) + (𝟔𝒙𝒆𝒚 ) 𝒇𝒚𝒚 = 𝟐𝒙 (𝒚) + 𝟑𝒙𝟐 (𝒆 )
𝝏𝒚 𝝏𝒚 𝝏𝒚 𝝏𝒚

This implies that This implies that

𝒇𝒙𝒚 = 𝟐𝒚 + 𝟔𝒙𝒆𝒚 𝒇𝒚𝒚 = 𝟐𝒙 + 𝟑𝒙𝟐 𝒆𝒚
 Practice Exercises
1) Find the second order partial derivatives of the following:
𝒙𝟐 𝒚 − 𝟒 𝒙 𝒛 + 𝒚𝟐
a) 𝒈(𝒙, 𝒚, 𝒛) =
𝒙𝒚𝒛
b) 𝒉(𝒙, 𝒚, 𝒛) = 𝐬𝐢𝐧(𝒙𝟐 𝒚 − 𝒛) + 𝐜𝐨𝐬 (𝒙𝟐 − 𝒚𝒛)
2) Verify that
𝒖(𝒙, 𝒚, 𝒕) = 𝟓 𝒔𝒊𝒏(𝟑𝝅𝒙) 𝒔𝒊𝒏(𝟒𝝅𝒚) 𝒄𝒐𝒔(𝟏𝟎𝝅𝒕)
is a solution to the wave equation
𝒖𝒕𝒕 = 𝟒(𝒖𝒙𝒙 + 𝒖𝒚𝒚 ).
3) Verify that
𝒙 𝒚 − 𝟐𝟓 𝒕
𝒖(𝒙, 𝒚, 𝒕) = 𝟐𝒔𝒊𝒏 ( ) 𝒔𝒊𝒏 ( ) 𝒆 𝟏𝟔
𝟑 𝟒
is a solution to the heat equation
𝒖𝒕 = 𝟗(𝒖𝒙𝒙 + 𝒖𝒚𝒚 )
Example:
Consider the function

𝒇(𝒙, 𝒚, 𝒛) = 𝒛 𝒍𝒏[𝒙𝟐 𝒚 𝒄𝒐𝒔(𝒛)]

Solution:
 Higher Order Partial Derivatives
In multivariate calculus, higher-order partial derivatives extend the
concept of first-order partial derivatives to describe the rate of change of
a function with respect to its variables multiple times. They provide
deeper insights into the behavior of functions, especially in applications
like optimization, physics, and machine learning. The concept of partial
derivatives was first formalized by mathematicians such as Leonhard
Euler and Joseph-Louis Lagrange in the 18th century.
What Are Higher-Order Partial Derivatives?
Higher-order partial derivatives are derivatives of a function taken
more than once, either with respect to the same variable or with respect to
different variables.
For example:

Second-order partial derivatives: These are obtained by

differentiating a first-order partial derivative again.
𝜕2 𝑓 𝜕2 𝑓
o Examples:
𝜕𝑥 2, 𝜕𝑥𝜕𝑦

Third-order and higher derivatives: These are obtained by

continuing the differentiation process.
𝜕3 𝑓 𝜕3 𝑓
o Examples:
𝜕𝑥 3, 𝜕𝑥 2 𝜕𝑦
 How to Continue Taking Derivatives?
1. Start with the First Derivative: Compute the first partial
derivatives of the function with respect to each variable.
2. Move to Higher Orders: Treat each first derivative as a new
function and differentiate again. Repeat this process according to the
desired order of differentiation.
3. Respect the Variable Order: When working with mixed
derivatives, ensure you follow the order specified.
For example:
𝝏𝟑 𝒇
means differentiating twice with respect to 𝒙, then once with
𝝏𝒙𝟐 𝝏𝒚
respect to 𝒚.
4. Check Continuity for Clairaut’s Theorem: If all the mixed partial
derivatives up to a certain order are continuous, the order of
differentiation does not matter:

𝝏𝟐 𝒇 𝝏𝟐 𝒇
=
𝝏𝒙𝝏𝒚 𝝏𝒚𝝏𝒙
How Does It Work for Continuous Functions?
For functions that are sufficiently smooth (i.e., their derivatives exist and
are continuous up to the required order):
• The process of differentiation can be repeated as many times as
needed, following the same principles as for first-order derivatives.
• The results of higher-order derivatives describe more variation in
behavior of the function, such as curvature or changes in slope in
multiple directions.
 • For continuous functions we can make interchanges within the
differentiation order.
For instance, if 𝒇(𝒙, 𝒚) = 𝟒𝒙𝟑 𝒚𝟐 we can verify that 𝒇𝒙𝒚𝒚 = 𝒇𝒚𝒙𝒚 .

For example:
𝝏𝟐 𝒇
• gives the concavity (maxima/minima) in the 𝒙-direction.
𝝏𝒙𝟐

𝝏𝟐 𝒇
• describes how the rate of change in the 𝒙-direction is affected
𝝏𝒙𝝏𝒚
by changes in 𝒚.

Step-by-Step Examples
Example 1: Algebraic Function
Find the third-order partial derivatives of 𝒇(𝒙, 𝒚) = 𝒙³𝒚² + 𝟐𝒙𝒚.
Solution:
Step 1: First Order Partial Derivatives
𝜕𝑓
= 3𝑥²𝑦² + 2𝑦
𝜕𝑥
𝜕𝑓
= 2𝑥³𝑦 + 2𝑥
𝜕𝑦
Step 2: Second Order Partial Derivatives
𝜕2𝑓
= 6𝑥𝑦²
𝜕𝑥 2
𝜕2𝑓
= 2𝑥³
𝜕𝑦 2
𝜕2𝑓 𝜕2𝑓
= = 6𝑥²𝑦 + 2
𝜕𝑥𝜕𝑦 𝜕𝑦𝜕𝑥
Step 3: Third Order Partial Derivatives
𝜕3𝑓
= 6𝑦²
𝜕𝑥 3
𝜕3𝑓
= 6𝑦 + 6𝑥𝑦
𝜕𝑥 2 𝜕𝑦
𝜕3𝑓 𝜕3𝑓
= = 6𝑥
𝜕𝑥𝜕𝑦 2 𝜕𝑦𝜕𝑥 2

Example 2: Exponential and Trigonometric Function

Find the second and third-order partial derivatives of
𝒈(𝒙, 𝒚) = 𝒆ˣ 𝐬𝐢𝐧(𝒚).
Solution:
Step 1: First Order Partial Derivatives
𝝏𝒈
= 𝒆ˣ 𝐬𝐢𝐧(𝒚)
𝝏𝒙
𝝏𝒈
= 𝒆ˣ𝒄𝒐𝒔(𝒚)
𝝏𝒚

Step 2: Second Order Partial Derivatives

𝝏𝟐 𝒈
= 𝒆ˣ𝒔𝒊𝒏(𝒚)
𝝏𝒙𝟐
𝝏𝟐 𝒈
= −𝒆ˣ𝒔𝒊𝒏(𝒚)
𝝏𝒚𝟐
𝝏𝟐 𝒈 𝝏𝟐 𝒈
= = 𝒆ˣ𝒄𝒐𝒔(𝒚)
𝝏𝒙𝝏𝒚 𝝏𝒚𝝏𝒙
Step 3: Third Order Partial Derivatives
𝝏𝟑 𝒈
= 𝒆ˣ𝒔𝒊𝒏(𝒚)
𝝏𝒙𝟑
𝝏𝟑 𝒈 𝝏𝟑 𝒈
= = −𝒆ˣ𝒄𝒐𝒔(𝒚)
𝝏𝒙𝟐 𝝏𝒚 𝝏𝒙𝝏𝒚𝟐
𝝏𝟑 𝒈
= −𝒆ˣ𝒄𝒐𝒔(𝒚)
𝝏𝒚𝟑

Summary:
To compute higher-order partial derivatives:
1. Start by computing first-order partial derivatives.
2. Use these results iteratively to find second, third, or higher-order
derivatives.
3. Ensure the function is continuous for applying Clairaut’s theorem,
which guarantees the equality of mixed derivatives when applicable.
Higher-order derivatives help capture complex behaviors in multivariable
functions, making them essential tools in advanced mathematical and
computational applications.
 Practice Questions
Find the indicated derivative for each of the following functions.
1. Find 𝒇𝒙𝒙𝒚𝒛𝒛 for 𝒇(𝒙, 𝒚, 𝒛) = 𝒛𝟑 𝒚𝟐 𝒍𝒏(𝒙)
𝟏𝟐𝒛𝒚
Answer: 𝒇𝒙𝒙𝒚𝒛𝒛 = − 𝒙𝟐

𝝏𝟑 𝒇
2. Find for 𝒇(𝒙, 𝒚) = 𝒆𝒙𝒚
𝝏𝒚𝝏𝒙𝟐
𝝏𝟑 𝒇
Answer: 𝝏𝒚𝝏𝒙𝟐 = 𝟐𝒚𝒆𝒙𝒚 + 𝒙𝒚𝟐 𝒆𝒙𝒚

3. Compute all second-order partial derivatives of

𝒉(𝒙, 𝒚) = 𝒙²𝒚 + 𝒍𝒏(𝒚).
Answers:
𝝏𝒉 𝝏𝒉 𝟏
= 𝟐𝒙𝒚, = 𝒙𝟐 +
𝝏𝒙 𝝏𝒚 𝒚
𝟐
𝝏 𝒉
= 𝟐𝒚
𝝏𝒙𝟐
𝝏𝟐 𝒉 𝟏
= −
𝝏𝒚𝟐 𝒚𝟐
𝝏𝟐 𝒉 𝝏𝟐 𝒉
= = 𝟐𝒙
𝝏𝒙𝝏𝒚 𝝏𝒚𝝏𝒙

𝝏𝟑 𝒇
4. Find for 𝒇(𝒙, 𝒚) = 𝒙𝒄𝒐𝒔(𝒚).
𝝏𝒙𝝏𝒚𝟐
𝝏𝟑 𝒇
Answer: = −𝒙𝒄𝒐𝒔(𝒚)
𝝏𝒙𝝏𝒚𝟐

5. Verify that the second order mixed derivatives are equal

for 𝒌(𝒙, 𝒚) = 𝒙𝟐 𝒚𝟑 + 𝒆ʸ 𝐬𝐢𝐧(𝐱).
 Answer: Mixed derivatives are equal.

𝝏𝟐 𝒌 𝝏𝟐 𝒌
= = 𝟑𝒙²𝒚² + 𝒆ʸ𝒄𝒐𝒔(𝒙)
𝝏𝒙𝝏𝒚 𝝏𝒚𝝏𝒙

𝝏𝟑 𝒎
6. Determine the third-order partial derivative for
𝝏𝒚𝟑
𝒎(𝒙, 𝒚) = 𝒕𝒂𝒏(𝒙𝒚)
Answer:
𝝏𝟑 𝒎
= 𝟑𝒙³𝒕𝒂𝒏²(𝒙𝒚)𝒔𝒆𝒄²(𝒙𝒚)
𝝏𝒚𝟑
Practice Questions:
Calculate the 2nd order partial derivatives of 𝒇.
i. 𝑓 (𝑥, 𝑦) = 𝑥 2 𝑦
ii. 𝑓 (𝑥, 𝑦) = 𝑥 2 + 2𝑥𝑦 + 𝑦 2
iii. 𝑓 (𝑥, 𝑦) = 𝑥𝑒 𝑦
2𝑥
iv. 𝑓 (𝑥, 𝑦) = 𝑦
v. 𝑓 (𝑥, 𝑦) = 5 + 𝑥 2 𝑦 2
vi. 𝑓 (𝑥, 𝑦) = 𝑒 𝑦𝑥
vii. 𝑓 (𝑥, 𝑦) = 100𝑒 𝑥𝑦
viii. 𝑓 (𝑥, 𝑦) = 5𝑥𝑒 −2𝑦
ix. 𝑓 (𝑥, 𝑦) = sin (𝑥𝑦)
x. 𝑓 (𝑥, 𝑦) = 𝑥 2 𝑦 + cos 𝑦 + 𝑦 sin 𝑥
xi. 𝑓 (𝑥, 𝑦) = ln(𝑥 + 𝑦)
xii. 𝑓 (𝑥, 𝑦) = 𝑥𝑒 𝑦 + 𝑦 + 1