Math301 Summary (A.

Fathy)

𝜕 𝜕 𝜕 𝜕 𝜕 𝜕
⃗⃗ = ( , , ).
 Nabla (del) operator: 𝛻 = 𝑖⃗ 𝜕𝑥 + 𝑗⃗ 𝜕𝑦 + 𝑘 𝜕𝑧 𝜕𝑥 𝜕𝑦 𝜕𝑧

 Gradient of a scalar field

grad(𝑓(𝑥, 𝑦, 𝑧)) = ∇𝑓(𝑥, 𝑦, 𝑧) = (𝑓𝑥 (𝑥, 𝑦, 𝑧), 𝑓𝑦 (𝑥, 𝑦, 𝑧), 𝑓𝑧 (𝑥, 𝑦, 𝑧)).

Directional derivative: is the rate of change (with respect to distance) of 𝑓(𝑥, 𝑦, 𝑧) at

𝑃0 in the direction of the unit vector 𝑢

⃗⃗

𝐷𝑢⃗⃗ 𝑓(𝑃0 ) = ∇𝑓(𝑃0 ). 𝑢

⃗⃗ .

At 𝑃0
𝛻𝑓
 𝑓 increases most rapidly in the direction of |𝛻𝑓| (rate of change= |∇𝑓|).

−∇𝑓
 𝑓 decreases most rapidly in the direction of (rate of change= −|∇𝑓|).
|∇𝑓|

 Any direction orthogonal to 𝛻𝑓 (𝛻𝑓 ≠ 0) is a direction of zero change of 𝑓.

 𝛻𝑓 is orthogonal to the level surface 𝑓(𝑥, 𝑦, 𝑧) = 𝑐 everywhere.

 Tangent Planes and normal lines for a general surface

 The equation of the tangent plane to the surface of 𝑓(𝑥, 𝑦, 𝑧) = 𝑐 at 𝑃0 (𝑥0 , 𝑦0 , 𝑧0 ) is

∇𝑓(𝑃0 ) ⋅ (𝑟⃗ − ⃗⃗⃗⃗⃗⃗⃗⃗

𝑂𝑃0 ) = 0

where: 𝑟⃗ = (𝑥, 𝑦, 𝑧) is the position vector.

 The equation of the normal line to the surface of 𝑓(𝑥, 𝑦, 𝑧) = 𝑐 at 𝑃0 (𝑥0 , 𝑦0 , 𝑧0 ) is in

1. Vector form:

𝑟⃗ = ⃗⃗⃗⃗⃗⃗⃗⃗
𝑂𝑃0 + 𝑡𝑛⃗⃗; 𝑡 ∈ ℝ.

2. Parametric form:

𝑥 = 𝑥0 + 𝑎𝑡, 𝑦 = 𝑦0 + 𝑏𝑡, 𝑧 = 𝑧0 + 𝑐𝑡; 𝑡 ∈ ℝ.

where: 𝑛⃗⃗ = (𝑎, 𝑏, 𝑐) = 𝛻𝑓(𝑃0 ) is the direction vector of the line.

1
 To get the local extrema and saddle points of a function of two variables 𝑓(𝑥, 𝑦):
1. Find: 𝑓𝑥 (𝑥, 𝑦), 𝑓𝑦 (𝑥, 𝑦), 𝑓𝑥𝑥 (𝑥, 𝑦), 𝑓𝑦𝑦 (𝑥, 𝑦), 𝑓𝑥𝑦 (𝑥, 𝑦).
2. Solve [𝑓𝑥 (𝑥, 𝑦) = 0 and 𝑓𝑦 (𝑥, 𝑦) = 0] for 𝑥, 𝑦 to get the critical points.
𝑓𝑥𝑥 (𝑥, 𝑦) 𝑓𝑥𝑦 (𝑥, 𝑦) 2
3. Find the discriminant: 𝐷(𝑥, 𝑦) = | | = 𝑓𝑥𝑥 (𝑥, 𝑦)𝑓𝑦𝑦 (𝑥, 𝑦) − (𝑓𝑥𝑦 (𝑥, 𝑦)) .
𝑓𝑥𝑦 (𝑥, 𝑦) 𝑓𝑦𝑦 (𝑥, 𝑦)
2
4. At each critical point (𝑎, 𝑏), evaluate 𝐷(𝑎, 𝑏) = 𝑓𝑥𝑥 (𝑎, 𝑏)𝑓𝑦𝑦 (𝑎, 𝑏) − (𝑓𝑥𝑦 (𝑎, 𝑏))
If 𝐷(𝑎, 𝑏) > 0
If 𝑓𝑥𝑥 (𝑎, 𝑏) > 0, or equivalently 𝑓𝑦𝑦 (𝑎, 𝑏) > 0, then the function has a local minimum at (𝑎, 𝑏, 𝑓(𝑎, 𝑏)).
If 𝑓𝑥𝑥 (𝑎, 𝑏) < 0, or equivalently 𝑓𝑦𝑦 (𝑎, 𝑏) < 0, then the function has a local maximum at (𝑎, 𝑏, 𝑓(𝑎, 𝑏)).
If 𝐷(𝑎, 𝑏) < 0, then the function has a saddle point at (𝑎, 𝑏, 𝑓(𝑎, 𝑏)).
If 𝐷(𝑎, 𝑏) = 0, then no conclusion can be drawn.

 To use Lagrange’s multiplier to find the absolute extrema of a function of three variables
𝒇(𝒙, 𝒚, 𝒛) subject to the constraint 𝒈(𝒙, 𝒚, 𝒛) = 𝒄
1. Solve: 𝛁𝒇(𝒙, 𝒚, 𝒛) = 𝝀𝛁𝒈(𝒙, 𝒚, 𝒛) and 𝒈(𝒙, 𝒚, 𝒛) = 𝒄
⟹ [𝒇𝒙 (𝒙, 𝒚, 𝒛) = 𝝀𝒈𝒙 (𝒙, 𝒚, 𝒛), 𝒇𝒚 (𝒙, 𝒚, 𝒛) = 𝝀𝒈𝒚 (𝒙, 𝒚, 𝒛), 𝒇𝒛 (𝒙, 𝒚, 𝒛) = 𝝀𝒈𝒛 (𝒙, 𝒚, 𝒛)
and 𝒈(𝒙, 𝒚, 𝒛) = 𝒄] for 𝑥, 𝑦, 𝑧, 𝜆.
2. Find the value of the function 𝒇 for each ordered triple (𝑥0 , 𝑦0 , 𝑧0 ) obtained in step (1).
3. The maximum value of the function in step (2) is the absolute maximum.
The minimum value of the function in step (2) is the absolute minimum.

 Double integrals of a continuous function 𝑓(𝑥, 𝑦) over the region 𝐷 are evaluated in
Cartesian coordinates as iterated integrals of type:

2 𝑏 𝑢 (𝑥)
I) ∬𝐷 𝑓(𝑥, 𝑦) 𝑑𝐴 = ∫𝑎 ∫𝑢 (𝑥) 𝑓(𝑥, 𝑦)𝑑𝑦 𝑑𝑥.
1

2 𝑑 𝑔 (𝑦)
II) ∬𝐷 𝑓(𝑥, 𝑦) 𝑑𝐴 = ∫𝑐 ∫𝑔 (𝑦) 𝑓(𝑥, 𝑦)𝑑𝑥 𝑑𝑦.
1

 When dealing with regions with circular boundaries, it is easier to deal with polar
coordinates to evaluate double integrals.

 Double integrals of a continuous function 𝑓(𝑥, 𝑦) over the region 𝐷 , with circular
boundary, are evaluated in polar coordinates as the iterated integrals

2 𝛽 𝑟
∬𝐷 𝑓(𝑥, 𝑦) 𝑑𝐴 = ∫𝛼 ∫𝑟 𝑓(𝑟𝑐𝑜𝑠𝜃, 𝑟𝑠𝑖𝑛𝜃) 𝑟𝑑𝑟 𝑑𝜃.
1

2
 Triple integrals of a continuous function 𝑓(𝑥, 𝑦, 𝑧) over the region 𝐸 are evaluated in
Cartesian coordinates as iterated integrals of type:
2 𝑔 (𝑥,𝑦)
I) ∭𝐸 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑉 = ∬𝐷 [∫𝑔 (𝑥,𝑦) 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑧] 𝑑𝐴 ;
1 1

𝐷1 : is the projection of 𝐸 onto the 𝑥𝑦 −plane.

2 ℎ (𝑦,𝑧)
II) ∭𝐸 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑉 = ∬𝐷 [∫ℎ (𝑦,𝑧) 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑥] 𝑑𝐴 ;
2 1

𝐷2 : is the projection of 𝐸 onto the 𝑦𝑧 −plane.

2 𝐼 (𝑥,𝑧)
III) ∭𝐸 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑉 = ∬𝐷 [∫𝐼 (𝑥,𝑧) 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑦] 𝑑𝐴 ;
3 1

𝐷3 : is the projection of 𝐸 onto the 𝑥𝑧 −plane.

 Double integral applications

 𝐴 = ∬𝐷 𝑑𝐴: gives the area of the region 𝐷.

 𝑉 = ∬𝐷 𝑓(𝑥, 𝑦) 𝑑𝐴: gives the volume of the solid bounded above by the surface of the
function 𝑧 = 𝑓(𝑥, 𝑦) and below by the 𝑥𝑦 −plane provided 𝑓(𝑥, 𝑦) ≥ 0 ∀ (𝑥, 𝑦) ∈ 𝐷,

𝐷: is the projection of the surface of the function 𝑧 = 𝑓(𝑥, 𝑦) onto the 𝑥𝑦 −plane.

 𝑚 = ∬𝐷 𝜌(𝑥, 𝑦) 𝑑𝐴: gives the mass of the flat body bounded by the region 𝐷,

𝜌(𝑥, 𝑦): is the density of the body (mass per unit area).

 𝑄 = ∬𝐷 𝜎(𝑥, 𝑦) 𝑑𝐴: gives the total charge of the flat body bounded by the region 𝐷,

𝜎(𝑥, 𝑦): is the charge density of the body (charge per unit area).

 Triple integral applications

 𝑉 = ∭𝐸 𝑑𝑉: gives the volume of the solid 𝐸.

 𝑚 = ∭𝐸 𝜌(𝑥, 𝑦, 𝑧)𝑑𝑉: gives the mass of the body bounded by 𝐸,

𝜌(𝑥, 𝑦, 𝑧): is the density of the body (mass per unit volume).

 𝑄 = ∭𝐸 𝜎(𝑥, 𝑦, 𝑧)𝑑𝑉 : gives the total charge of the body bounded by 𝐸,

𝜎(𝑥, 𝑦, 𝑧): is the charge density of the body (charge per unit volume).

3
 Relation between Cartesian (𝒙, 𝒚, 𝒛) and cylindrical coordinates (𝒓, 𝜽, 𝒛)

𝑥 = 𝑟𝑐𝑜𝑠𝜃, 𝑦 = 𝑟𝑠𝑖𝑛𝜃, 𝑧 = 𝑧
𝑦
𝑟 2 = 𝑥 2 + 𝑦 2 , 𝑡𝑎𝑛𝜃 = ,𝑧 = 𝑧
𝑥

 Relation between Cartesian (𝒙, 𝒚, 𝒛) and spherical coordinates (𝝆, 𝜽, 𝝋)

𝑥 = 𝜌𝑠𝑖𝑛𝜑𝑐𝑜𝑠𝜃, 𝑦 = 𝜌𝑠𝑖𝑛𝜑𝑠𝑖𝑛𝜃, 𝑧 = 𝜌𝑐𝑜𝑠𝜑

𝑧 𝑦
𝑝2 = 𝑥 2 + 𝑦 2 + 𝑧 2 , 𝑐𝑜𝑠𝜑 = , 𝜃 = tan−1 | |
𝜌 𝑥

 𝜃 = 𝜃; 𝑥 > 0, 𝑦 > 0

 𝜃 = 𝜋 − 𝜃; 𝑥 < 0, 𝑦 > 0

 𝜃 = 𝜋 + 𝜃; 𝑥 < 0, 𝑦 < 0

 𝜃 = 2𝜋 − 𝜃; 𝑥 > 0, 𝑦 < 0

 Triple integrals of a continuous function 𝑓(𝑥, 𝑦, 𝑧) over the solid 𝐸 are evaluated as
iterated integrals in:

 Cylindrical coordinates as:

𝜃2 𝑟2 𝑧2 (𝑟,𝜃)

∭ 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑉 = ∫ ∫ ∫ 𝑓(𝑟𝑐𝑜𝑠𝜃, 𝑟𝑠𝑖𝑛𝜃, 𝑧)𝑟𝑑𝑧𝑑𝑟𝑑𝜃

𝐸 𝜃1 𝑟1 𝑧1 (𝑟,𝜃)

and is typically used when there is a symmetry about a line (𝑧 −axis).

 Spherical coordinates as:

𝜑2 𝜃2 𝜌2 (𝜃, 𝜑)

∭ 𝑓(𝑥, 𝑦, 𝑧)𝑑𝑉 = ∫ ∫ ∫ 𝑓(𝜌𝑠𝑖𝑛𝜑𝑐𝑜𝑠𝜃, 𝜌𝑠𝑖𝑛𝜑𝑠𝑖𝑛𝜃, 𝜌𝑐𝑜𝑠𝜑) 𝜌2 𝑠𝑖𝑛𝜑𝑑𝜌𝑑𝜃𝑑𝜑

𝐸 𝜑1 𝜃1 𝜌1 (𝜃, 𝜑)

and is typically used when there is a symmetry about a point (the origin).

4
 A vector valued function 𝑟⃗ of a real variable is a rule that assigns a vector (in ℝ𝑛 ) to each 𝑡
in some domain 𝐷 of real numbers.
 If the codomain of 𝒓 ⃗⃗ is ℝ𝟐 or ℝ𝟑 , then as 𝑡 takes all the values in the domain 𝐷, the range
of 𝑟⃗ is geometrically a curve.
⃗⃗⃗⃗⃗⃗
𝑑𝑟(𝑡0 )
⃗⃗ =
 The unit tangent vector at the point 𝑡 = 𝑡0 is 𝑇 𝑑𝑡
⃗⃗⃗⃗⃗⃗ .
𝑑𝑟(𝑡0 )
| |
𝑑𝑡

 A vector field is a function 𝐹⃗ that assigns a vector to each point in its domain.
 A vector field 𝐹⃗ of space vectors defined on a domain in ℝ3 can be written in terms of its
components
𝐹⃗ = (𝑀(𝑥, 𝑦, 𝑧), 𝑁(𝑥, 𝑦, 𝑧), 𝑃(𝑥, 𝑦, 𝑧)).
 Conservative Vector Fields
A vector field 𝐹⃗ is said to be conservative, over some domain, if it can be expressed as the
gradient of some scalar field 𝑓
𝐹⃗ = ∇𝑓(𝑥, 𝑦, 𝑧).
 The curl of a vector field
If 𝐹⃗ = (𝑀(𝑥, 𝑦, 𝑧), 𝑁(𝑥, 𝑦, 𝑧), 𝑃(𝑥, 𝑦, 𝑧)) is a vector field and the partial derivatives for its
components exist, then its curl is given by
𝑖⃗ 𝑗⃗ ⃗⃗
𝑘
𝜕 𝜕 𝜕
curl(𝐹⃗ ) = 𝛻 × 𝐹⃗ = | |.
𝜕𝑥 𝜕𝑦 𝜕𝑧
𝑀(𝑥, 𝑦, 𝑧) 𝑁(𝑥, 𝑦, 𝑧) 𝑃(𝑥, 𝑦, 𝑧)
 Theorem
If 𝐹⃗ is a vector field defined over all of ℝ3 whose components have continuous partial
derivatives and curl(𝐹⃗ ) = ⃗⃗0, then 𝐹⃗ is a conservative vector field.
 The divergence of a vector field
If 𝐹⃗ = (𝑀(𝑥, 𝑦, 𝑧), 𝑁(𝑥, 𝑦, 𝑧), 𝑃(𝑥, 𝑦, 𝑧)) is a vector field and the partial derivatives for its
components exist, then its divergence is given by
div(𝐹⃗ ) = 𝛻 ⋅ 𝐹⃗ = 𝑀𝑥 (𝑥, 𝑦, 𝑧) + 𝑁𝑦 (𝑥, 𝑦, 𝑧) + 𝑃𝑧 (𝑥, 𝑦, 𝑧).
 Remark
If 𝐹⃗ = (𝑀(𝑥, 𝑦, 𝑧), 𝑁(𝑥, 𝑦, 𝑧), 𝑃(𝑥, 𝑦, 𝑧)) is a vector field and its components have
continuous partial derivatives, then
div (curl(𝐹⃗ )) = 0.
 The parametric representation of
 The segment from 𝑃1 (𝑥1 , 𝑦1 , 𝑧1 ) to 𝑃2 (𝑥2 , 𝑦2 , 𝑧2 ) is
𝑟⃗(𝑡) = ⃗⃗⃗⃗⃗⃗⃗⃗
𝑂𝑃1 + 𝑡𝑃 ⃗⃗⃗⃗⃗⃗⃗⃗⃗
1 𝑃2 ; 0 ≤ 𝑡 ≤ 1.
𝑥2 𝑦2
 The ellipse (𝑎2 + 𝑏2 = 1, 𝑧 = 𝑐) oriented CCW as viewed from above is
𝑟⃗(𝑡) = (𝑎𝑐𝑜𝑠𝑡, 𝑏𝑠𝑖𝑛𝑡, 𝑐); 0 ≤ 𝑡 ≤ 2𝜋.