Continuum Mechanics
Table of Contents
1. Stress
1.1. Cauchy Stress Tensor
The force per area on a surface is called the traction \( \mathbf{t} \). The Cauchy stress tensor \( \boldsymbol{\sigma} \) is defined to satisfy \[ \mathbf{t} = \boldsymbol{\sigma}\hat{\mathbf{n}} \] where \( \hat{\textbf{n}} \) is the unit normal vector of the surface.
The stress is being applied to a control volume which is a infinitesimal inertial volume. The control volume is at a quasi-equalibrium, that is, the changes in tensile and shear stresses across the volume are infinitesimal and the forces and torques should sum to zero. This justifies the Cauchy stress tensor being symmetric.
In a coordinate system, \( \sigma_{ij} \) is the force in \( i \)th direction, when the surface is facing \( j \)th direction.
1.2. Volumetric Stress Tensor
- Mean Normal Stress Tensor
The isotropic stress can change the volumn of the stressed body. \[ \pi \mathbf{I} := \frac{1}{3}\operatorname{tr} (\boldsymbol{\sigma}) \mathbf{I}. \] The three accounts for the three direction that the stress is applied.
1.3. Pressure
In its full generality, \[ p := \zeta \nabla\cdot \mathbf{u} - \pi \] where \( \zeta \) is teh volume viscosity, \( \mathbf{u} \) is the flow velocity.
1.4. Deviatoric Stress Tensor
- \( \boldsymbol{\tau} \), \( s_{ij} \)
- Stress Deviator Tensor
Deviation of the Cauchy stress tensor from the isotropy. \[ \boldsymbol{\tau} := \boldsymbol{\sigma} - \pi \mathbf{I}. \]
1.5. Linear Stress
Assuming
- The Cauchy stress tensor is Galilean invariant
- \( p \) is independent of the strain \( \boldsymbol{\varepsilon} \)
- The fluid is isotropic,
The relation (constitutive equation) between the strain and stress is given as: \[ \boldsymbol{\sigma} = -p\mathbf{I} + \lambda \operatorname{tr}(\boldsymbol{\varepsilon}) \mathbf{I} + 2\mu \boldsymbol{\varepsilon} \] where \( \lambda \) is the second viscosity.
The equation can also be written as: \[ \boldsymbol{\sigma} = -(p - \zeta \nabla\cdot \mathbf{u})\mathbf{I} + \mu \left( \nabla \mathbf{u} + (\nabla \mathbf{u})^{\mathsf{T}} - \frac{2}{3}(\nabla \cdot \mathbf{u})\mathbf{I} \right) \] using the bulk viscosity.
2. Viscosity
2.1. Newton's Law of Viscosity
\[ \boldsymbol{\tau} = \mu (\nabla \mathbf{u} + (\nabla \mathbf{u})^{\mathsf{T}} \] where \(\boldsymbol{\tau}\) is the deviatoric stress, \( \mu \) is the dynamic viscosity, and \(\mathbf{u}\) is the flow velocity.
Coordinate representation would be \[ \tau_{ij} = \mu \left(\frac{\partial u_i}{\partial x_j} + \frac{\partial u_j}{\partial x_i} \right). \]
The fluid that satisfies the Newton's law of viscosity is called the Newtonian fluid.
2.2. Dynamic Viscosity
- \(\mu\), \( \eta \)
- Shear Viscosity
The relational constant between stress and momentum change. It has the unit of \(\rm Pa\cdot s\) in SI, poise in cgs.
2.3. Kinematic Viscosity
- Momentum Diffusivity
\[ \nu = \frac{\mu}{\rho}. \]
2.4. Second Viscosity
The stress occurs due to irreversible resistance, over the reversible resistance by ientropic bulk modulus.
2.5. Bulk Viscosity
- Volume Viscosity
\[ \zeta := \lambda + \frac{2}{3}\mu \] where \( \lambda \) is the second viscosity, and \( \mu \) is the dynamic viscosity.
2.6. Effective Viscosity
For an object moving in a fluid with viscosity \( \mu \), if the size of the object is smaller than the mean free path of the fluid particles, the effective viscosity \( \mu_{\mathrm{eff}} \) deviates from the macroscopic viscosity \( \mu \). \[ \eta_{\mathrm{eff}} = \eta \frac{1}{1+\frac{b}{pr}} \] where \( p \) is the pressure, \( b = 8.22\times 10^{-3}\ \mathrm{Pa\cdot m}\) is a constant, and \( r \) is the radius of the particle.
This quantity is used when establishing the mechanical equalibrium using the Stokes' law: \[ 6\pi r \eta_{\mathrm{eff}} v = F. \] From the equalibrium, the radius \( r \) can be obtained. \[ r = \sqrt{\frac{9\eta_{\mathrm{eff}} v_t}{2g\rho}}. \] where \( v_t \) is the terminal velocity of free falling particle, \( g \) is the gravitational constant, \( \rho \) is the density of the fluid.
3. Strain
3.1. Infinitesimal Strain
\[ \boldsymbol{\varepsilon} = \frac{1}{2}(\nabla \mathbf{u} + (\nabla \mathbf{u})^{\rm T}) \] This is the symmetric part of the gradient of the displacement field \( \mathbf{u} \).
4. Fick's Laws of Diffusion
4.1. First Law
\[ \mathbf{J} = -D\nabla \varphi \] where \(\mathbf{J}\) is the diffusion flux, amount of substance per area per time, and \(D\) is the diffusion coefficient of diffusivity, and \(\nabla\varphi\) is the concentration gradient.
- \(D\) has the unit of area per time.
- \(\varphi\) has the unit of concentration, the amount of substance per volume.
4.2. Second Law
- \[ \frac{\partial \varphi}{\partial t} = D\nabla^{\cdot 2}\varphi. \]
- This is the same form as the heat equation.
5. Fourier's Law
- The Law of Heat Conduction
\[ \mathbf{q} = -k\nabla T \] where \(\mathbf{q}\) is the heat flux density, \(k\) is the thermal conductivity.
5.1. Heat Equation
- \[
\frac{\partial u}{\partial t} = \alpha \nabla^{\cdot 2} u
\]
- where \(\alpha\) is the thermal diffusivity.
- \[
\alpha = \frac{k}{\rho c}
\]
- where \(k\) is the thermal conductivity, \(\rho\) is the density of the material, \(c\) is the specific heat capacity of the material.
5.2. Heat Kernel
- Fundamental solution to the heat equation.
5.2.1. Example
- \[ K(t, x, y) = \exp(t \nabla^{\cdot 2})(x, y) = \frac{1}{(4\pi t)^{d/2}}e^{-\|x-y\|^2/4t}. \]
- For every smooth function \(\phi\) of compact support: \[ \lim_{t\to 0}\int_{\mathbb{R}^d}K(t, x, y)\phi(y)\,dy = \phi(x). \]
6. Coordinate Specification
6.1. Lagrange Specification
- Material Coordinates
Lagrangian specification of the flow field represents individual fluid parcels. \[ \mathbf{X}(\mathbf{x}_0, t) \]
6.2. Eulerian Specification
- Specification at fixed location.
- \[ \mathbf{u}(\mathbf{x}, t) \]
7. Knudsen Number
\[ \mathrm{Kn} := \frac{\lambda}{L} \] where \( \lambda \) is the mean free path of particles, and \( L \) is the representative physical length scale of a system.
If Knudsen number is near or greater than one, statistical methods should be used instead of treating the system as a continuum.
7.1. Properties
It is related to Mach number and Reynolds number by \[ \mathrm{Kn} = \frac{\mathrm{Ma}}{\mathrm{Re}} \sqrt{\frac{\gamma \pi}{2}}. \]
8. Material Derivative
Advective Derivative, Convective Derivative, Derivative Following the Motion, Hydrodynamic Derivative, Lagrangian Derivative, Particle Derivative, Substantial Derivative, Substantive Derivative, Stokes Derivative, Total Derivative
8.1. Definition
\[ \frac{D}{Dt} := \frac{\partial }{\partial t} + \mathbf{u}\cdot\nabla. \]
The first term describes the change due to time at a fixed position, and the second term describes the change due to movement in space at a fixed time, which combine to describe the rate of change while moving along with the flow.
9. Boltzmann Equation
- Boltzmann Transport Equation (BTE)
For a conservative system, \[ \frac{\partial f}{\partial t} + \frac{\mathbf{p}}{m}\cdot \frac{\partial f}{\partial \mathbf{r}} + \mathbf{F}\cdot \frac{\partial f}{\partial \mathbf{p}} = \left( \frac{\partial f}{\partial t} \right)_{\mathrm{coll}} \] where \( f \) is the probability density function, and the derivatives are written in the denominator layout convention.
9.1. Probability Density Function
For a system in which the velocity of the particles are not equilibrated, not even locally, the probability density function \( f(\mathbf{r}, \mathbf{p}, t) \) is a good description of the system.
It is the number density in the phase space, satisfying \[ dN = f(\mathbf{r}, \mathbf{p}, t)\, \mathrm{d}^3 \mathbf{r}\, \mathrm{d}^3 \mathbf{p} \] where \( \mathrm{d}N \) is the number of particles in a small chunk of the phase space \( \mathrm{d}^3\mathbf{r}\,\mathrm{d}^3\mathbf{p} \).
9.2. Derivation
According to the Liouville's theorem, the density of state in the phase space does not change in a conservative system. When one is interested in the microscopic behavior, the force once called friction disappears and just molecular interactions remains, constituting a conservative system.
There's still collisions that can disturb the probability density function. So, \[ \frac{\mathrm{d}f}{\mathrm{d}t} = \left( \frac{\partial f}{\partial t} \right)_{\mathrm{coll}}. \] The change of \( f \) due to collision is swept under the simple looking collision term, and it is not an easy task to find the exact form of this term. The total derivative is taken along the trajectory of physical parcel in the phase space, therefore when expanded
\begin{align*} \frac{\partial f}{\partial t} &+ \frac{\mathrm{d}\mathbf{r}}{\mathrm{d} t} \cdot \frac{\partial f}{\partial \mathbf{r}} + \frac{\mathrm{d} \mathbf{p}}{\mathrm{d} t}\cdot \frac{\partial f}{\partial \mathbf{p}} = \left( \frac{\partial f}{\partial t} \right)_{\mathrm{coll}} \\ \implies \frac{\partial f}{\partial t} &+ \frac{\mathbf{p}}{m}\cdot \frac{\partial f}{\partial \mathbf{r}} + \mathbf{F}\cdot \frac{\partial f}{\partial \mathbf{p}} = \left( \frac{\partial f}{\partial t} \right)_{\mathrm{coll}}. \end{align*}9.3. Derived Definitions
- Velocity \( \mathbf{v} := \mathbf{p}/m \)
- Flow Velocity \[ \mathbf{u} := \langle \mathbf{v} \rangle = \int \mathbf{v} f(\mathbf{r}, \mathbf{v}, t)\, \mathrm{d}^3 \mathbf{v}. \]
- Peculiar Velocity \[ \mathbf{w} := \mathbf{v} - \mathbf{u}. \]
- Pressure \[ p := \frac{1}{3} \rho \langle w^2 \rangle. \]
- Pressure Tensor \[ \mathbf{P} := -\boldsymbol{\sigma} = \rho \langle \mathbf{w} \mathbf{w}^{\mathsf{T}}\rangle \] where \( \boldsymbol{\sigma} \) is the stress tensor.
- Heat Flux \[ \mathbf{h} := \frac{1}{2} \rho \langle w^2 \mathbf{w} \rangle. \]
10. Continuity Equations
- Transport Equation
They are often derived form the Boltzmann equation.
10.1. Mass Continuity
\[ \frac{\partial \rho}{\partial t} = -\nabla\cdot(\rho\mathbf{u}). \] It arises from the global mass conservation.
10.2. Momentum Continuity
\[ \rho \frac{\mathrm{D}\mathbf{u}}{\mathrm{D}t} = \vec{f}_m(\rho, \mathbf{u}, p, T) \] where \(D/Dt\) is the material derivative, the rate of change while moving along with the flow.
- The Eulerian specification is used, since the Lagrangian specification is unstable.
- One form of this is Navier-Stokes equation.
10.3. Energy Continuity
\[ \rho\frac{\mathrm{D}}{\mathrm{D}t}f(\mathbf{u}, T) = f_e(\rho, \mathbf{u}, p, T) \] where the \(f\) is the local energy function.
10.4. Equations of States
- \[ f_{s,1}(p, \rho, T) = 0 \]
- \[ f_{s,2}(\mu, T) = 0 \]
11. Navier-Stokes Equation
11.1. General
Assuming the fluid is isothermal, \[ \rho\frac{\mathrm{D}\mathbf{u}}{\mathrm{D}t}=\nabla\cdot\boldsymbol{\sigma}+\rho\mathbf{f} \] where \( \rho \) is the mass density, \( \mathbf{u} \) is the flow velocity, \( \mathrm{D}/\mathrm{D} t \) is the material derivative, \(\boldsymbol{\sigma}\) is Cauchy stress tensor, and \( \mathbf{f} \) is the body force.
11.2. Compressible
We assume the linear stress. Substituting, we get: \[ \rho \frac{\mathrm{D} \mathbf{u}}{\mathrm{D} t} = - \nabla p + \nabla \cdot \left[ \mu \left( \nabla \mathbf{u} + (\nabla \mathbf{u})^{\mathsf{T}} - \frac{2}{3}(\nabla \cdot \mathbf{u})\mathbf{I} \right) \right] + \nabla (\zeta \nabla \cdot \mathbf{u}) + \rho \mathbf{f}. \]
11.3. Incompressible
For a incompressible fluid \( \nabla\cdot \mathbf{u} = 0 \), simplifying the equation: \[ \rho\frac{\mathrm{D}\mathbf{u}}{\mathrm{D}t}=-\nabla p+\mu\nabla^2\mathbf{u}+\rho\mathbf{f}. \]
Note that \(\nabla\cdot \boldsymbol{\tau} = \mu\nabla^{\cdot 2}\mathbf{u}\) here, and this fluid can be called Newtonian.
12. Reynolds Number
Determining factor for the generation of turbulent flow.
\[ \mathrm{Re} := \frac{u L}{\nu} \] where \( u \) is the mean velocity of the fluid, \( L \) is the characteristic length of the system, and \( \nu \) is the kinematic viscosity.
Turbulence occurs when the momentum is not transferred fast enough, with high velocity, long distance, and low momentum diffusion. Turbulence starts to form around \( \mathrm{Re} \sim 10^{3} \).
13. Lamb Vector
- Named after physicist Horace Lamb
13.1. Vorticity
\[ \boldsymbol{\omega} = \nabla\times \mathbf{u}. \]
13.2. Definition
- The cross product of vorticity vector \(\boldsymbol{\omega}\) and velocity vector \(\mathbf{u}\) of the flow field: \[ \mathbf{l} = \mathbf{u}\times \boldsymbol{\omega}. \]
- This appears in the convective acceleration term of the material derivative in the Navier-Stokes equation.
13.3. Beltrami Flow
- Flow in which the vorticity vector and the velocity vector are parallel.
- Flow in which the Lamb vector is zero.
14. Reference
- Lagrangian and Eulerian specification of the flow field - Wikipedia
- Material derivative - Wikipedia
- The Stress Tensor and Traction Vector - YouTube
- Viscosity - Wikipedia
- Fick's laws of diffusion - Wikipedia
- Heat equation - Wikipedia
- Boltzmann equation - Wikipedia
- Building the simplest fluid simulation that still makes sense - YouTube
- How do you simulate what isn‘t there – and still make sense of it? - YouTube
- Navier–Stokes equations - Wikipedia
- Lamb vector - Wikipedia
- Infinitesimal strain theory - Wikipedia