MATH F342 (Differential Geometry) Dr. Sangita Yadav Department of Mathematics BITS Pilani, Pilani Campus, Rajasthan

MATH F342 (Differential Geometry)
Dr. Sangita Yadav

Department of Mathematics
BITS Pilani
Pilani Campus BITS Pilani, Pilani Campus, Rajasthan
Lecture 6
January 23, 2024
Dr. Sangita Yadav (BITS Pilani) MATH F342 (Differential Geometry) January 27, 2024 2 / 35
Point of Inflexion
Definition (Point of Inflexion)

The point P = ⃗r(u0 ) on the curve for which ⃗r ′′ (u0 ) = 0
is called a point of inflexion and the tangent line at P is
called inflexional.
Point of Inflexion
Definition (Point of Inflexion)

The point P = ⃗r(u0 ) on the curve for which ⃗r ′′ (u0 ) = 0
is called a point of inflexion and the tangent line at P is
called inflexional.
Corollary
Let k be the smallest integer⩾ 2 such that ⃗r (k) (0) ̸= 0,
then the equation of the osculating plane is
⃗ − ⃗r(0), ⃗r ′ (0), ⃗r (k) (0)] = 0.
[R
When curve is not parameterised by arc length parameter
When curve is not parameterised by arc length parameter
For ⃗r(t) where t is not necessarily an arc length
parameter, 2
′ d⃗r ds ′′ d2⃗r ds d⃗r d2 s
⃗r = and ⃗r = 2 + .
ds dt ds dt ds dt2
By linearlity, 3 2

⃗ − ⃗r(0), ⃗r (0), ⃗r (0)] =
′ ′′ ds d⃗
r
⃗ − ⃗r(0), , d ⃗
r
[R R .
dt ds ds2
ds
Since ̸= 0, again the equation for the osculating
dt
plane is
[R⃗ − ⃗r(0), ⃗r ′ (0), ⃗r ′′ (0)] = 0.
Also the equation at point of inflection remains
[R⃗ − ⃗r(0), ⃗r ′ (0), ⃗r (k) (0)] = 0.
Example
Find the osculating plane of ⃗r(t) = (t, t3 , 0); t ∈ R at
P = (1, 1, 0).
Example
P = (1, 1, 0).
Sol. Here, P = ⃗r(1) and

⃗r ′ (t) = (1, 3t2 , 0) → ⃗r ′ (1) = (1, 3, 0),
⃗r ′′ (t) = (0, 6t, 0) → ⃗r ′ (1) = (0, 6, 0).
Hence the osculating plane has the equation
x−1 y−1 z
⃗ r(1), ⃗r ′ (1), ⃗r ′′ (1)] =
[R−⃗ 1 3 0 = 0 ⇒ z = 0.
0 6 0
Example
P = (0, 0, 0).
Example
P = (0, 0, 0).
Sol. Here, P = ⃗r(0) and
⃗r ′ (t) = (1, 3t2 , 0) → ⃗r ′ (0) = (1, 3, 0),
⃗r ′′ (t) = (0, 6t, 0) → ⃗r ′′ (0) = (0, 0, 0).
Hence (0, 0, 0) is a point of inflexion. So
⃗r ′′′ (t) = (0, 6, 0) ̸= 0. Hence the osculating plane has
the equation
x−1 y−1 z
⃗ ′ ′′′
[R−⃗r(0), ⃗r (0), ⃗r (0)] = 1 3 0 = 0 ⇒ z = 0.
0 6 0
The curve which doesn’t posses the osculating plane
Consider the curve defined by

 2
 (u, 0, e−1/u ) when u > 0
2
⃗r(u) = (u, e−1/u , 0) when u < 0
(0, 0, 0) when u = 0.

For the above vector valued function, u = 0 is a point of

inflexion. Infact, the vector valued function has all its
derivatives with ⃗r (k) (0) = 0, k ⩾ 0. Hence the curve
does not possess the osculating plane.
Lecture 7
January 27, 2024
Section 1.6
Prinicipal Normal and Binormal
Normal Plane
Let ⃗r be a regular path of class m ⩾ 1. The normal
plane to ⃗r at ⃗r(0) is a plane through ⃗r(0) perpendicular
to ⃗t(0).
Its equation is given by
⃗ − ⃗r(0)) = 0.
⃗t(0) · (R
Normal Plane
Let ⃗r be a regular path of class m ⩾ 1. The normal
plane to ⃗r at ⃗r(0) is a plane through ⃗r(0) perpendicular
to ⃗t(0).
⃗ − ⃗r(0)) = 0.
⃗t(0) · (R
Example
The normal plane of ⃗r(u) = (u, u3 , 0) at ⃗r(0) = (0, 0, 0)
is
(1, 0, 0) · (x, y, z) = 0 ⇒ x = 0.
Principal Normal Line
The principal normal line of a regular path ⃗r of class

m ⩾ 1 at ⃗r(0) is the intersection of the osculating plane
and the normal plane to ⃗r at ⃗r(0).
Principal Normal Line
The principal normal line of a regular path ⃗r of class

m ⩾ 1 at ⃗r(0) is the intersection of the osculating plane
and the normal plane to ⃗r at ⃗r(0).
Example
The principal normal line of ⃗r(u) = (u, u3 , 0) at
⃗r(0) = (0, 0, 0) is
x = z = 0.
Principal Normal Vector
A unit vector parallel to the principal normal line of a
regular path ⃗r of class m ⩾ 1 at ⃗r(0) is called a principal
normal vector ⃗n(0) to ⃗r at ⃗r(0).
Please Note that principal normal vector lies on the
osculating plane and perpendicular to the tangent vector.
Please Note that principal normal vector lies on the
osculating plane and perpendicular to the tangent vector.
Example
The principal normal vector of ⃗r(u) = (u, u3 , 0) at
⃗r(0) = (0, 0, 0) is
⃗n(0) = ĵ = (0, 1, 0)‘.
Section 1.7
Curvature and Torsion
Binormal Line
Definition
The binormal line to a regular path ⃗r of class m ⩾ 1 at
⃗r(0) is a line through ⃗r(0) and perpendicular to the
osculating plane to ⃗r at ⃗r(0).
Example
The binormal line of ⃗r(u) = (u, u3 , 0) at ⃗r(0) = (0, 0, 0)
is
x = y = 0.
Unit Binormal Vector
The unit binormal vector ⃗b(0) to ⃗r at ⃗r(0) is a unit
vector parallel to the binormal line to ⃗r at ⃗r(0) such that
⃗t(0), ⃗n(0), ⃗b(0) form a right handed frame.
If ⃗r(s) is regular of class m ⩾ 2 with arc length

′′
parameter and ⃗r (0) ̸= 0, then generally we take
′′
⃗r (0)
⃗n(0) = ′′ and ⃗b(0) = ⃗t(0) × ⃗n(0).
|⃗r (0)|
If ⃗r(s) is regular of class m ⩾ 2 with arc length

′′
parameter and ⃗r (0) ̸= 0, then generally we take
′′
⃗r (0)
⃗n(0) = ′′ and ⃗b(0) = ⃗t(0) × ⃗n(0).
|⃗r (0)|
Example
The unit binormal vector of ⃗r(u) = (u, u3 , 0) at
⃗r(0) = (0, 0, 0) is ⃗b(0) = ⃗t(0) × ⃗n(0) = î × ĵ = k.
Rectifying Plane
The plane containing the tangent vector and the

binormal vector is called the rectifying plane.
⃗ − ⃗r(0)) = 0.
⃗n(0) · (R
Rectifying Plane
The plane containing the tangent vector and the

binormal vector is called the rectifying plane.
⃗ − ⃗r(0)) = 0.
⃗n(0) · (R
Example
The rectifying plane of ⃗r(u) = (u, u3 , 0) at
⃗r(0) = (0, 0, 0) is y = 0.
Geomerical View
Planes of moving frame
Alternatively we can use these moving vectors to find
osculating, normal and rectifying planes.
Osculating plane to ⃗r(s) at its point ⃗r(s0 ) is a plane
through ⃗r(s0 )and perpendicular to ⃗b(s0 ) (spanned by
⃗t(s0 ) and ⃗n(s0 )).
Normal plane to ⃗r(s) at its point ⃗r(s0 ) is a plane
through ⃗r(s) and perpendicular to ⃗t(s0 ) . (spanned by
⃗n(s0 ) and ⃗b(s0 )).
Rectifying plane to ⃗r(s) at its point ⃗r(s0 ) is a plane
through ⃗r(s0 ) and perpendicular to ⃗n(s0 ). (spanned by
⃗t(s0 ) and ⃗b(s0 )).
Example
Find the vectors ⃗t(0), ⃗n(0), ⃗b(0) and hence the
osculating, normal and rectifying planes to the path
⃗r(u) = (cos u, sin u, u).
Example
Find the vectors ⃗t(0), ⃗n(0), ⃗b(0) and hence the
osculating, normal and rectifying planes to the path
⃗r(u) = (cos u, sin u, u).
Sol. Note : Here parameter is not arc-length. So process

needs to be amended.
d⃗r ds d⃗r √
= (− sin u, cos u, 1) ⇒ = = 2.
du du du
d⃗r
1 1
∴ ⃗t(0) = du d⃗r
= 0, √ , √ .
| du | 2 2
d d⃗r

d2⃗r

d d⃗r du ds − cos u − sin u
= = d⃗r
= , ,0 .
ds2 ds ds | du | 2 2
∴ ⃗n(0) = (−1, 0, 0).

1 1
Hence ⃗b(0) = ⃗t(0) × ⃗n(0) = 0, − √ , √ . The
2 2
osculating plane at ⃗r(0) = (1, 0, 0) is
⃗b(0) · (x − 1, y, z) = 0 ⇒ y − z = 0.
The normal plane at ⃗r(0) = (1, 0, 0) is

⃗t(0) · (x − 1, y, z) = 0 ⇒ y + z = 0.
The rectifying plane at ⃗r(0) = (1, 0, 0) is

⃗n(0) · (x − 1, y, z) = 0 ⇒ x = 1.
Curvature
Definition
The Curvature vector of a regular curve ⃗r(s) of class
m ⩾ 2 with arc-length parameter s is
⃗ d⃗t d2⃗r
K= = .
ds ds2
Curvature
Definition
⃗ d⃗t d2⃗r
K= = .
ds ds2
Definition
⃗
The Curvature of ⃗r(s) is defined by κ = |K|.
Curvature
Definition
⃗ d⃗t d2⃗r
K= = .
ds ds2
Definition
⃗
The Curvature of ⃗r(s) is defined by κ = |K|.
Definition
1
if κ ̸= 0, then ρ = is called the radius of curvature of
κ
⃗r(s).
Curvature vector gives the rate of change (with respect
to the arc length) at which the tangent vector is turning.
d⃗t
In fact, we can write = κ⃗n (by choosing unit principal
ds
d2⃗r
normal in the direction of 2 ).
ds
Torsion
The torsion of a regular curve ⃗r(s) of class m ⩾ 2 with
arc length parameter s is given by
d⃗b
τ = − · ⃗n.
ds
d⃗b
Since |⃗b(s)| = 1 ∀s, so ds
⊥ ⃗b(s) ⇒ ⃗b(s) · ⃗n(s) = 0, ∀s.
d⃗b d⃗n
· ⃗n(s) + ⃗b(s) · = 0.
ds ds
On the other hand, since ⃗b(s) = ⃗t(s) × ⃗n(s),

Differentiating with respect to s
d⃗b d⃗t d⃗n
= × ⃗n(s) + ⃗t(s) × .
ds ds ds
d⃗b ⃗ d⃗b
Hence ds · t = −κ⃗n · ⃗b = 0. Consequently ds ⊥ ⃗t also.
d⃗b d⃗b
Hence ds is a multiple of ⃗n, in fact ds = −τ⃗n. Thus we
see that torsion, in magnitude, represents the arc-rate of
change of ⃗b. This also measures the arc-rate of turning
of the osculating plane.
d⃗n
What is ?
ds
⃗ ⃗ d⃗n d⃗b ⃗ ⃗ d⃗t

As ⃗n = b × t ⇒ = ×t+b× .
ds ds ds
d⃗n
∴ = τ⃗b − κ⃗t.
ds
Serret-Frenet Equations
The information of how the Serret-Frenet frame rotates
is stored in the following Serret-Frenet equations (or
formulas) :
d⃗t
= κ⃗n,
ds
d⃗n
= τ⃗b − κ⃗n,
ds
d⃗b
= −τ⃗n.
ds
Matrix Form
Serret-Frenet equations can be written in matrix form as
 ′  
⃗t (s) ⃗t(s)
 
0 κ(s) 0
⃗n ′ (s) = −κ(s) 0 τ (s) ⃗n(s) .
⃗b ′ (s) 0 −τ (s) 0 ⃗b(s)
Matrix Form
Serret-Frenet equations can be written in matrix form as
 ′  
⃗t (s) ⃗t(s)
 
0 κ(s) 0
⃗n ′ (s) = −κ(s) 0 τ (s) ⃗n(s) .
⃗b ′ (s) 0 −τ (s) 0 ⃗b(s)
It can be shown that ⃗t, ⃗n, ⃗b, κ, τ as well as osculating,

normal and rectifying planes at corresponding points of
equivalent paths are same, i.e., they give property of
regular curves of class m ⩾ 2.
Curvature in Arbitrary Parameter
Theorem
If ⃗r is a regular path of class m ⩾ 2, then
|⃗r ′ × ⃗r ′′ |
κ(t) = .
|⃗r ′ |3
Proof.
′ ′ ′′ d|⃗r ′ (t)|⃗
⃗
As ⃗r (t) = |⃗r (t)|t ⇒ ⃗r (t) = t + |⃗r ′ (t)|2 κ⃗n.
dt
Therefore,
|⃗r ′ × ⃗r ′′ | = |⃗r ′ |3 κ.
Remarks
For a plane parameterized curve
⃗r(t) = (x(t), y(t), 0)
|x′ (t)y ′′ (t) − x′′ (t)y ′ (t)|
κ(t) = .
(x′ (t)2 + y ′ (t)2 )3/2
For the graph y = f (x) of a function f (x) of class
m ⩾ 2,
|f ′′ (x)|
κ(x) = .
(1 + f ′ (x)2 )3/2
For a polar curve r = f (θ) in xy-plane
|2f ′ (θ)2 + f (θ)2 − f (θ)f ′′ (θ)|
κ(θ) = .
(f (θ)2 + f ′ (θ)2 )3/2
For the point P0 (x0 , y0 ) of a level curve f (x, y) = 0
of a function f of class m ⩾ 2 with fy (P0 ) ̸= 0,
|fxx fy2 − 2fxy fx fy + fyy fx2 |

κ(x0 , y0 ) = .
(fx2 + fy2 )3/2 )
P0
Homework Proofs left as an exercise.
Examples
⃗r(t) = (a cos(t), a sin t, 0); 0 ⩽ t ⩽ 2π with a > 0 is
a circle of radius a. Its curvature is κ(t) = a1 . The
radius of curvature of the circle coincides with its
radius.
Examples
radius.
⃗r(t) = (a cos t, a sin t, bt); 0 ⩽ t ⩽ 2π, with a > 0
and b ̸= 0 is a circular helix. Its curvature is
a
κ(t) = a2 +b 2 , a constant.
Examples
radius.
⃗r(t) = (a cos t, a sin t, bt); 0 ⩽ t ⩽ 2π, with a > 0
and b ̸= 0 is a circular helix. Its curvature is
a
κ(t) = a2 +b 2 , a constant.
For a straight line

⃗r(t) = (a1 t + b1 , a2 t + b2 , a3 t + b3 ) : t ∈ R with
(a1 , a2 , a3 ) ̸= (0, 0, 0), κ(t) = 0, constant.
Conversely, if κ(t) = 0, constant, then the curve
must be a straight line.
Torsion
Theorem
Torsion of a curve ⃗r(t) of class m ⩾ 3 at t = t0 with
κ(t0 ) ̸= 0, is given by
[⃗r ′ (t0 ), ⃗r ′′ (t0 ), ⃗r ′′′ (t0 )]

τ (t0 ) = .
|⃗r ′ (t0 ) × ⃗r ′′ (t0 )|2
Torsion
Theorem
Torsion of a curve ⃗r(t) of class m ⩾ 3 at t = t0 with
κ(t0 ) ̸= 0, is given by
[⃗r ′ (t0 ), ⃗r ′′ (t0 ), ⃗r ′′′ (t0 )]

τ (t0 ) = .
|⃗r ′ (t0 ) × ⃗r ′′ (t0 )|2
Proof.
Please note that ⃗r ′ (t0 ) × ⃗r ′′ (t0 ) = |⃗r ′ (t0 )|3 κ⃗b. Moreover
r|⃗
⃗r ′′ (t) = d|⃗ r ′ |2 κ⃗n. Hence
dt t + |⃗
⃗r ′′′ (t) = α⃗t + β⃗n + |⃗r ′ |3 κ(τ⃗b − κ⃗t). Using the formula
for κ and solving for τ we get the result.
Example
Torsion of a circular helix:

⃗r(t) = (a cos t, a sin t, bt); 0 ⩽ t ⩽ 2π, a > 0, b ̸= 0,
b
τ= .
a2 + b 2
Example
Torsion of a circular helix:

⃗r(t) = (a cos t, a sin t, bt); 0 ⩽ t ⩽ 2π, a > 0, b ̸= 0,
b
τ= .
a2 + b 2
Note: For the circular helix, both curvature and torsion
is constant. Conversely any curve with constant
curvature and constant torsion can be obtained from
circular helix above by translation and rotation in R3 .
Example
τ
Show that for ⃗r(t) = (3t2 , 3t − t3 , 3t + t3 ), κ is a
constant.
Example
τ
constant.
(A curve for which κτ is constant is called a general helix.
Circular helix is its special case. )
Example
τ
constant.
(A curve for which κτ is constant is called a general helix.
Circular helix is its special case. )
Sol. Here
⃗r ′ (t) = (6t, 3(1 − t2 ), 3(1 + t2 ),

⃗r ′′ (t) = (6, −6t, 6t),
and ⃗r ′′′ (t) = (0, −6, 6).
So ⃗r ′ × ⃗r ′′ = (−36t, 18(1 − t2 ), 18(1 + t2 ).
√
⇒ |⃗r ′ × ⃗r ′′ | = 18 2(1 + t2 )2
√
and |⃗r ′ (t)| = 3 2(1 + t2 )2 ,
|⃗r ′ × ⃗r ′′ | 1
∴κ= = .
|⃗r ′ (t)|3 3(1 + t2 )2
Now [⃗r ′ , ⃗r ′′ , ⃗r ′′′ ] = 216
[⃗r ′ (t), ⃗r ′′ (t), ⃗r ′′ (t)]
⇒τ =
|⃗r ′ × ⃗r ′′ |2
1
= .
3(1 + t2 )2
Thanks for your
attention!

MATH F342 (Differential Geometry) Dr. Sangita Yadav Department of Mathematics BITS Pilani, Pilani Campus, Rajasthan

Uploaded by

Copyright:

Available Formats

MATH F342 (Differential Geometry) Dr. Sangita Yadav Department of Mathematics BITS Pilani, Pilani Campus, Rajasthan

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MATH F342 (Differential Geometry) Dr. Sangita Yadav Department of Mathematics BITS Pilani, Pilani Campus, Rajasthan

Uploaded by

Copyright:

Available Formats

MATH F342 (Differential Geometry)

Dr. Sangita Yadav

Definition (Point of Inflexion)

Definition (Point of Inflexion)

Sol. Here, P = ⃗r(1) and

Consider the curve defined by

For the above vector valued function, u = 0 is a point of

The principal normal line of a regular path ⃗r of class

The principal normal line of a regular path ⃗r of class

⃗n(0) = ĵ = (0, 1, 0)‘.

If ⃗r(s) is regular of class m ⩾ 2 with arc length

If ⃗r(s) is regular of class m ⩾ 2 with arc length

The plane containing the tangent vector and the

The plane containing the tangent vector and the

Sol. Note : Here parameter is not arc-length. So process

The normal plane at ⃗r(0) = (1, 0, 0) is

The rectifying plane at ⃗r(0) = (1, 0, 0) is

On the other hand, since ⃗b(s) = ⃗t(s) × ⃗n(s),

⃗ ⃗ d⃗n d⃗b ⃗ ⃗ d⃗t

It can be shown that ⃗t, ⃗n, ⃗b, κ, τ as well as osculating,

|fxx fy2 − 2fxy fx fy + fyy fx2 |

Homework Proofs left as an exercise.

For a straight line

[⃗r ′ (t0 ), ⃗r ′′ (t0 ), ⃗r ′′′ (t0 )]

[⃗r ′ (t0 ), ⃗r ′′ (t0 ), ⃗r ′′′ (t0 )]

Torsion of a circular helix:

Torsion of a circular helix:

⃗r ′ (t) = (6t, 3(1 − t2 ), 3(1 + t2 ),

You might also like

Homework Proofs left as an exercise.