17/12/2023, 21:42 Velocity Gradients

AD

Purpose Built process by MASS Open

Read the article to find out how they transformed the site into a reforested landscape. Autodesk

Search Continuum Mechanics Website

Velocity Gradients Search

home > deformation & strain > velocity gradients

Introduction
Velocity gradients are absolutely essential to analyses involving path dependent materials, such as the plastic deformation of metals.
Granted, rubber can usually be treated as hyperelastic, which implies no path dependence. Although viscous damping of rubber can
be addressed with velocity gradients. Regardless, velocity gradients are useful to better understanding deformations and especially
useful to determining energetically conjugate stresses and strains, which is important to all materials.

Velocity Gradient
The velocity gradient is to velocities what the deformation gradient is to displacements. The velocity gradient is represented by L and
defined by

∂vx ∂vx ∂vx

⎡ ⎤
∂x ∂y ∂z

⎢ ⎥
⎢ ⎥
⎢ ⎥
∂v ⎢ ∂vy ∂vy ∂vy ⎥
L = = ⎢ ⎥
⎢ ∂x ∂y ∂z ⎥
∂x ⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
∂vz ∂vz ∂vz
⎣ ⎦
∂x ∂y ∂z

Note that the derivative is with respect to the current coordinates, x, not the reference coordinates, X. This means it is an Eulerian
quantity rather than a Lagrangian quantity. The velocity gradient can also be written as

∂vi
Lij =
∂xj

or as

Lij = vi,j

Although this notation relies on the reader knowing that the "j" represents xj instead of Xj .

Velocity Gradient Example

The velocity gradient is ideally suited to applications involving fluids. With fluids, it is impossible to specify X. This leaves x
as the only practical coordinates to describe the velocity field in.

Suppose the velocity field is

1 −y
v = ( , , 0)
5 − x 10

This gives the velocity of any particle in the fluid as a function of the particle's position at that instant.

The velocity gradient is

https://www.continuummechanics.org/velocitygradient.html 1/12
17/12/2023, 21:42 Velocity Gradients
1
0 0
⎡ (5−x)
2 ⎤

⎢ −1 ⎥
⎢ 0 0⎥
L = ⎢ ⎥
y
10
⎢ ⎥
⎢ ⎥

⎣ ⎦
0 0 0

x
It is still desirable to find a way to express L in terms of F in order to make
it possible to convert from Lagrangian to Eulerian modes. This process
begins with F ˙.

d ∂x ∂ dx ∂v
˙
F = ( ) = ( ) =
dt ∂X ∂X dt ∂X

The next step is to apply the chain rule to the above result

∂v ∂v ∂x
˙
F = = ( )( )
∂X ∂x ∂X

The first partial derivative is L and the second is simply F. So the above equation can be written as

Ḟ = L ⋅ F

−1
Post multiplying both sides by F gives the equation for L in terms of F.

−1
˙
L = F ⋅ F

The two equations above are incredibly useful, although that is probably not apparent at this point. With the one immediately above, it
is possible to calculate the Eulerian quantity L by using only the Lagrangian quantity, F.

Tension Velocity Gradient Example

Recall the Almansi strain example in which the displacements were
expressed in terms of X and then x. This time, in order to
calculate velocity gradients, we must first have velocities.
y
So this time, let

X
Lf
x = ( ) LF
Lo

If the object is being stretched such that

LF = Lo + A t
x
Lo L
where A is any positive constant, then

X
x = ( ) (Lo + A t)
Lo

dx
and vx is , so this gives
dt

X
vx = ( )A
Lo

https://www.continuummechanics.org/velocitygradient.html 2/12
17/12/2023, 21:42 Velocity Gradients

This is nice, but no help to a velocity gradient because the velocity is not expressed as a function of x. But this can be
overcome by combining the two above equations to get

Ax
vx =
Lo + A t

And now the velocity gradient can be evaluated.

∂vx A A
L11 = = =
∂x Lo + A t LF

and

A/LF 0 0
⎡ ⎤

L = ⎢ 0 0 0⎥
⎣ ⎦
0 0 0

A A
This result emphasizes how a velocity gradient is an Eulerian quantity because L11 =
LF
rather than Lo
. As the object
stretches, LF increases with time and L11 decreases.

Velocity Gradient Using F

y
−1
This time, repeat the above example, but use L = Ḟ ⋅ F to
demonstrate that it gives the same result.

Recall that

x = (
X
) (Lo + A t) = (
X
) LF Lf
Lo Lo

The deformation gradient is

⎡
Lo +A t
0 0⎤
x
L
Lo

⎢
⎢
⎥
⎥
Lo
⎢ ⎥
F = ⎢ ⎥
⎢ 0 1 0⎥
⎢ ⎥
⎢ ⎥

⎣ ⎦
0 0 1

and its inverse is

Lo Lo
⎡ 0 0⎤ ⎡ 0 0⎤
Lo +A t LF

⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥
−1 ⎢ ⎥ ⎢ ⎥
F = ⎢ ⎥ = ⎢ ⎥
⎢ 0 1 0⎥ ⎢ 0 1 0⎥
⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥

⎣ ⎦ ⎣ ⎦
0 0 1 0 0 1

We also need Ḟ. It is

https://www.continuummechanics.org/velocitygradient.html 3/12
17/12/2023, 21:42 Velocity Gradients
A
⎡ 0 0⎤
Lo

⎢ ⎥
⎢ ⎥
˙ ⎢ ⎥
F = ⎢ ⎥
0 0 0
⎢ ⎥
⎢ ⎥
⎢ ⎥

⎣ ⎦
0 0 0

−1
˙ and
So the product of F gives
F

A Lo A
0 0 0 0⎤ 0 0
⎡ Lo ⎤⎡ LF
⎡ LF ⎤

⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
−1 ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
L = Ḟ ⋅ F = ⎢ ⎥ ⎢ ⎥ = ⎢ ⎥
⎢ 0 0 0⎥ ⎢ 0 1 0⎥ ⎢ 0 0 0⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥

⎣ ⎦⎣ ⎦ ⎣ ⎦
0 0 0 0 0 1 0 0 0

Which is identical to the previous example.

Velocity Gradients and Abaqus UMATS

The Abaqus finite element package has a feature called a "UMAT," which is short for USER MATERIAL subroutine. This
subroutine gives the user an opportunity to write a subroutine that calculates the stress state in a material given the
deformation gradients at the beginning, Ft , and end Ft+Δt , of a time step. (Abaqus documentation here.)

One could treat the rubber as hyperelastic and calculate the stress based solely on Ft+Δt . Or one could use both
deformation gradients to compute the velocity gradient and use this to include viscoelastic or hysteretic damping in the
stress calculation.

For example, suppose that Abaqus executes a time step during a transient simulation that is Δt = 0.1sec . At the beginning
of the time step (which was just the end of the previous step), the deformation gradient is

1.30 0 0
⎡ ⎤
Ft = ⎢ 0 0.90 0 ⎥
⎣ ⎦
0 0 0.90

And at the end of the step, it is

1.40 0 0
⎡ ⎤
Ft+Δt = ⎢ 0 0.85 0 ⎥
⎣ ⎦
0 0 0.85

So Ḟ is

1
Ḟ = (Ft+Δt − Ft )
Δt

1.40 0 0 1.30 0 0
⎛⎡ ⎤ ⎡ ⎤⎞
1
= ⎜⎢ 0 0.85 0 ⎥ − ⎢ 0 0.90 0 ⎥⎟
0.1
⎝⎣ ⎦ ⎣ ⎦⎠
0 0 0.85 0 0 0.90

1.0 0 0
⎡ ⎤
= ⎢ 0 −0.5 0 ⎥
⎣ ⎦
0 0 −0.5

https://www.continuummechanics.org/velocitygradient.html 4/12
17/12/2023, 21:42 Velocity Gradients

We need the inverse of a deformation gradient as well. Take the inverse of the later one.

−1
1.40 0 0 0.7143 0 0
⎡ ⎤ ⎡ ⎤
−1
F = ⎢ 0 0.85 0 ⎥ = ⎢ 0 1.1765 0 ⎥
⎣ ⎦ ⎣ ⎦
0 0 0.85 0 0 1.1765

So L is

1.0 0 0 0.714 0 0 0.714 0 0

⎡ ⎤⎡ ⎤ ⎡ ⎤
−1
˙
L = F ⋅ F = ⎢ 0 −0.5 0 ⎥⎢ 0 1.177 0 ⎥ = ⎢ 0 −0.588 0 ⎥
⎣ ⎦⎣ ⎦ ⎣ ⎦
0 0 −0.5 0 0 1.177 0 0 −0.588

Rate of Deformation and Spin Tensors

The velocity gradient can be decomposed into symmetric and antisymmetric parts as
follows.

1 T
1 T
L = (L + L ) + (L − L )
2 2

The first term, (L + LT )/2, is symmetric and looks and behaves very much like the small strain tensor discussed earlier, except of
course that it involves velocities rather than displacements. It is in fact a strain rate tensor in an Eulerian context. But it is formally
called the Rate of Deformation Tensor, and assigned the symbol, D. (I have never, ever seen it called a "deformation rate tensor".) So
D is defined as

1 T
D = (L + L )
2

T
The second term, (L − L )/2, is a new animal. It is an antisymmetric tensor that looks and behaves somewhat different than
anything we've discussed to date. The antisymmetric part is called the spin tensor and given the letter, W. We will see that it is directly
related to the rate of rotation of an object, but not its deformation at all. It is defined as

1 T
W = (L − L )
2

So the velocity gradient equation can be written as

L = D + W

The components of the rate of deformation tensor, D, are

L12 +L21 L13 +L31

⎡ L11 ⎤
2 2

⎢ ⎥
⎢ ⎥
⎢ ⎥
L21 +L12 L23 +L32
⎢ ⎥
D = L22
⎢ 2 2 ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
L31 +L13 L32 +L23
⎣ ⎦
L33
2 2

The components of the spin tensor, W, are

https://www.continuummechanics.org/velocitygradient.html 5/12
17/12/2023, 21:42 Velocity Gradients
L12 −L21 L13 −L31
⎡ 0 ⎤
2 2

⎢ ⎥
⎢ ⎥
⎢ ⎥
L21 −L12 L23 −L32
⎢ ⎥
W = 0
⎢ 2 2 ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
L31 −L13 L32 −L23
⎣ ⎦
0
2 2

In terms of partials, the rate of deformation tensor is

∂vx ∂vx ∂vy ∂vx ∂vz

1 1
⎡ ( + ) ( + )⎤
∂x 2 ∂y ∂x 2 ∂z ∂x

⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ∂vy ∂vx ∂vy ∂vy ∂vz ⎥
1 1
D = ⎢ ( + ) ( + )⎥
⎢ 2 ∂x ∂y ∂y 2 ∂z ∂y ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
∂vz ∂vx ∂vz ∂vy ∂vz
1 1
⎣ ( + ) ( + ) ⎦
2 ∂x ∂z 2 ∂y ∂z ∂z

and the spin tensor is

∂vx ∂vy ∂vx ∂vz

1 1
⎡ 0 ( − ) ( − )⎤
2 ∂y ∂x 2 ∂z ∂x

⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
∂vy ∂vy
⎢ 1 ∂vx 1 ∂vz ⎥
W = ⎢ ( − ) 0 ( − )⎥
⎢ 2 ∂x ∂y 2 ∂z ∂y ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
⎢ ⎥
∂vz ∂vx ∂vz ∂vy
1 1
⎣ ( − ) ( − ) 0 ⎦
2 ∂x ∂z 2 ∂y ∂z

Spin Tensor and Angular Velocity Vector

y
Consider the object in the figure rotating counter-clockwise about the z
axis at an angular velocity, ωz . Calculate the W21 component of the spin
tensor.

The following equations describe the rotation.

x = X cos ωz t − Y sin ωz t

y = X sin ωz t + Y cos ωz t

Take the time derivative of the equations to obtain vx and vy .

vx = −ωz X sin ωz t − ωz Y cos ωz t

x
vy = ωz X cos ωz t − ωz Y sin ωz t

This gives us vx and vy in terms of X and Y . But we need vx and vy in terms of x and y instead, in order to compute the
spin tensor components. This can be accomplished by first inverting the first pair of equations to obtain X and Y in terms of
x and y, and substituting these into the second pair of equations.

Inverting the first equations gives

X = x cos ωz t + y sin ωz t

Y = −x sin ωz t + y cos ωz t

−1
(Note that this amounts to starting with x = R ⋅ X and inverting it to obtain X = R ⋅ x )

https://www.continuummechanics.org/velocitygradient.html 6/12
17/12/2023, 21:42 Velocity Gradients
Substituting these inverted equations into the second pair containing the velocity components gives a remarkably simple
result.

vx = −ωz y

vy = ωz x

It is now possible to easily calculate the W21 component of the spin tensor.

1 ∂vy ∂vx 1 ∂(ωz x) ∂(−ωz y)

W21 = ( − ) = ( − ) = ωz
2 ∂x ∂y 2 ∂x ∂y

This simple 2-D example demonstrates how a spin tensor can contain components of the angular velocity vector.

The 3-D spin tensor can be summarized as

0 −ωz ωy
⎡ ⎤

W = ⎢ ωz 0 −ωx ⎥

⎣ ⎦
−ωy ωx 0

where ω is the angular velocity vector. It is directly related to rotation about an axis discussed on the rotation matrix page. Recall that it
contained a rotation angle, α, about an axis, p = (px , py , pz ). The relationship is

1 ∂vz ∂vy 1 ∂vx ∂vz

ωx = α̇ px = ( − ) ωy = α̇ py = ( − )
2 ∂y ∂z 2 ∂z ∂x

1 ∂vy ∂vx
ωz = α̇ pz = ( − )
2 ∂x ∂y

Rate of Deformation and Spin Example

If the velocity gradient is

0.5 0.4 0.3

⎡ ⎤
L = ⎢ 0.2 −0.1 −0.2 ⎥
⎣ ⎦
−0.3 0.4 −0.1

Then the rate of deformation tensor is

AD
0.5 0.3 0.0

Purpose Built
⎡ ⎤
D = ⎢ 0.3 −0.1 0.1 ⎥

process by MASS
⎣ ⎦
0.0 0.1 −0.1

and the spin tensor is

Read the article to find out how they
transformed the site into a
reforested landscape.
0.0 0.1 0.3
⎡ ⎤
W = ⎢ −0.1 0.0 −0.3 ⎥
⎣ ⎦
−0.3 0.3 0.0 Autodesk

Rate of Deformation and True Strain

https://www.continuummechanics.org/velocitygradient.html 7/12
17/12/2023, 21:42 Velocity Gradients
The next webpage on true strains will go into this topic in more depth, but for completeness, it will be quickly summarized here.

The normal components of the rate of deformation tensor also give the time rates of change of true strain. The relationships are as
follows.

True
d LF
D11 = ϵ̇ = ln( )
1
dt Lo

Also,

True
D11 + D22 + D33 = ϵ̇
Vol

And this applies regardless of the strain level or orientation.

Velocity Gradients and the Polar Decomposition

Merging the polar decomposition together with the velocity gradient gives some very interesting results. Start with

F = R ⋅ U

The time derivative is

Ḟ = R ⋅ U̇ + Ṙ ⋅ U

The inverse of F is

−1 −1 −1
F = U ⋅ R

−1
So Ḟ ⋅ F is

−1 −1 −1
˙ ˙ ˙
L = F ⋅ F = (R ⋅ U + R ⋅ U)(U ⋅ R )

−1 −1 −1 −1
= R ⋅ U̇ ⋅ U ⋅ R + Ṙ ⋅ U ⋅ U ⋅ R

−1
But U ⋅ U cancels out, leaving

−1 −1 −1 −1
˙ ˙ ˙
L = F ⋅ F = R ⋅ (U ⋅ U ) ⋅ R + R ⋅ R

−1 T
And R = R , so finally

−1 T T
L = R ⋅ (U̇ ⋅ U ) ⋅ R + Ṙ ⋅ R

−1 T
The result is interesting because of how L has become partitioned into two contributing terms. The first term, R ⋅ (U̇ ⋅ U ) ⋅ R ,
˙ is zero, and the entire first
gives the contribution due solely to deformations. This is evident because if there is no deformation, then U
term is zero as well.

If deformations are taking place, then the first term can have both symmetric and antisymmetric contributions to L. The antisymmetric
(spin) contribution could come from simple shear, for example, D/T style. Note that since U is always related to the reference
T
coordinate system, like Green strain for example, it is multiplied by R and R to rotate it to the current orientation just like

T
V = R ⋅ U ⋅ R

Note the similarity!

https://www.continuummechanics.org/velocitygradient.html 8/12
17/12/2023, 21:42 Velocity Gradients

T T
The second term, R ˙ ˙ ˙
⋅ R , is just as interesting. Suppose there is only rigid body rotation taking place. Then U = 0 and only R ⋅ R
T
is left to contribute to L. But this must be entirely "spin" because there is no deformation taking place. So Ṙ ⋅ R produces an
antisymmetric result, which can easily be re-interpreted into ω = (ωx , ωy , ωz ).

2-D Rigid Body Rotation Example

T
˙
Start with a 2-D rotation matrix, R, and calculate R ⋅ R .

cos θ − sin θ
R = [ ]
sin θ cos θ

so Ṙ is

− sin θ − cos θ
˙
R = ω[ ]
cos θ − sin θ

T
And R is

T
cos θ sin θ
R = [ ]
− sin θ cos θ

So multiplying the two together gives

T
− sin θ − cos θ cos θ sin θ
˙
R ⋅ R = ω[ ][ ]
cos θ − sin θ − sin θ cos θ

0 −1
= ω[ ]
1 0

So the result is indeed an antisymmetric spin tensor.

Velocity Gradient and Green Strain

The last relationship to develop is between the velocity gradient and the time derivative of the Green strain tensor. Start with the
definition of Green strain.

1
T
E = [F ⋅ F − I]
2

Take the derivative of E to get

1 T
T
˙ ˙ ˙
E = [F ⋅ F + F ⋅ F]
2

T
˙
Now insert L ⋅ F for F and FT ⋅ L
T ˙
for F .

1 T T T
˙
E = [F ⋅ L ⋅ F + F ⋅ L ⋅ F]
2

T
Next, notice that each term begins with F and ends with F. So things can be regrouped as follows.

https://www.continuummechanics.org/velocitygradient.html 9/12
17/12/2023, 21:42 Velocity Gradients

T
L + L
T
Ė = F ⋅ ( ) ⋅ F
2

1 T
But ˙ and
is just D. This gives the very useful relationship between E
(L + L ) D. AD
2

T
Ė = F ⋅ D ⋅ F

And it can be inverted to give

−T ˙ −1
D = F ⋅ E ⋅ F

These equations relating Ė and D will be used to identify pairs of

energetically conjugate stress and strain tensors.

Limitations
In closing, it is worth noting that the usual limitations of small strains and/or small rotations that complicate most strain calculations do
not apply to velocity gradients. Velocity gradient calculations are not limited by any of these issues. This is because, as an operator on
velocity rather than displacement, a velocity gradient is a calculation over an infinitesimal time step. During this time step, an object's
orientation and level of strain only changes by an infinitesimal amount. And since this amount of change is not finite, it does not
complicate matters in any way. So velocity gradients are always applicable regardless of the level of strain or rotation.

https://www.continuummechanics.org/velocitygradient.html 10/12
17/12/2023, 21:42 Velocity Gradients

10 Comments 1 Login


Join the discussion…

LOG IN WITH OR SIGN UP WITH DISQUS ?

Name

 1 Share Best Newest Oldest

Roshan − ⚑
3 years ago

Hi, great job on putting together this website. I use it as a go-to reference for my work as I find it
easy to access. But I would like to point out that your calculation of velocity gradient using
UMAT is not correct. Please refer to https://link.springer.com/c...

0 0 Reply • Share ›

Xiang − ⚑
3 years ago

Back to "Velocity Gradient" topic, you mentioned "it is an Eulerian quantity rather than a
Lagrangian quantity", but actually it is opposite. Velocities defined in current frame should be
Lagrangian quantities.

0 0 Reply • Share ›

BobMcG Moderator > Xiang − ⚑

3 years ago

No. The velocity gradient presented here is absolutely Eulerian, not Lagrangian.

0 0 Reply • Share ›

Shriram Krishna > BobMcG − ⚑

3 years ago

Hi. Could you please explain how? With regards to time derivatives it makes
sense that Lagrangian = Eulerian + Convective terms. How does this work
with respect to gradients?

0 0 Reply • Share ›

BobMcG Moderator > Shriram Krishna − ⚑

3 years ago

Sorry, I don't know how Lagrangian is Eulerian plus convective

terms.

0 0 Reply • Share ›

Shriram Krishna > BobMcG − ⚑

3 years ago

I'm sorry If I was not clear. I was talking about the substantial
derivative of a quantity. I'm speaking more from a CFD background.
When we speak of an Eulerian V/s Lagrangian framework. How
does taking derivative w.r.t to a current configuration make it a
Eulerian quantity? An intuitive explanation might help. Or if you
could point me to where you state the difference between
Lagrangian and Eulerian quantities within this website it would be
quite helpful. My uderstanding is that quantities which live in the
reference configuration(Green Lagrange strains, First Piola and
Second Piola Kirchoff Stress etc) are Lagrangian quantities.
Because they are descibed w.r.t basis in the reference config.
Similarly with Eulerian quantities. Could you please give me an
analogy for when we take gradients ? Also, thank you for this
website. It's a treasure trove for those learning continuum
mechanics
https://www.continuummechanics.org/velocitygradient.html 11/12
17/12/2023, 21:42 Velocity Gradients
mechanics.

0 0 Reply • Share ›

BobMcG Moderator > Shriram Krishna

− ⚑
3 years ago

I still don't have a good answer for you. In my mind, taking a

derivative w.r.t. current coords is simply called an Eulerian quantity,
while doing that w.r.t the orig coords is simply called a Lagrangian
quantity.

0 0 Reply • Share ›

Hiroki Endo > BobMcG − ⚑

2 years ago

In continuum mechanics, it is related to Objectivity whether the

velocity gradient is Eulerian or Lagrangian. According to MIT's class
paper at the page 91, the velocity gradient is not objective. Thus,
velocity gradient depends on the observer.

l Sh

Thank You Strain Chapter - $4.99 Entire Website - $12.95

Thank you for visiting this webpage. For $4.99, you receive two formatted For $12.95, you receive two formatted
Feel free to email me if you have PDFs (the first for 8.5" x 11" pages, PDFs (the first for 8.5" x 11" pages,
questions. the second for tablets) of the complete the second for tablets) of the entire
strain chapter. website.
Also, please consider visiting an
advertiser above. Doing so helps to Click here to see a sample page in Click here to see a sample page in
each of the two formats. each of the two formats.
cover website hosting fees.

Email address to receive PDFs Email address to receive PDFs

Bob McGinty
bmcginty@gmail.com

Copyright © 2012 by Bob McGinty

Table of Contents
Hydrostatic & Deviatoric Strains True Strain

https://www.continuummechanics.org/velocitygradient.html 12/12