Stokes flow explained

Stokes flow (named after George Gabriel Stokes), also named creeping flow or creeping motion,[1] is a type of fluid flow where advective inertial forces are small compared with viscous forces.[2] The Reynolds number is low, i.e.

Re\ll1

. This is a typical situation in flows where the fluid velocities are very slow, the viscosities are very large, or the length-scales of the flow are very small. Creeping flow was first studied to understand lubrication. In nature, this type of flow occurs in the swimming of microorganisms and sperm.[3] In technology, it occurs in paint, MEMS devices, and in the flow of viscous polymers generally.

The equations of motion for Stokes flow, called the Stokes equations, are a linearization of the Navier–Stokes equations, and thus can be solved by a number of well-known methods for linear differential equations.[4] The primary Green's function of Stokes flow is the Stokeslet, which is associated with a singular point force embedded in a Stokes flow. From its derivatives, other fundamental solutions can be obtained.[5] The Stokeslet was first derived by Oseen in 1927, although it was not named as such until 1953 by Hancock.[6] The closed-form fundamental solutions for the generalized unsteady Stokes and Oseen flows associated with arbitrary time-dependent translational and rotational motions have been derived for the Newtonian[7] and micropolar[8] fluids.

Stokes equations

The equation of motion for Stokes flow can be obtained by linearizing the steady state Navier–Stokes equations. The inertial forces are assumed to be negligible in comparison to the viscous forces, and eliminating the inertial terms of the momentum balance in the Navier–Stokes equations reduces it to the momentum balance in the Stokes equations:

\boldsymbol{\nabla}\sigma+f=\boldsymbol{0}

where

\sigma

is the stress (sum of viscous and pressure stresses),[9] [10] and

f

an applied body force. The full Stokes equations also include an equation for the conservation of mass, commonly written in the form:
\partial\rho
\partialt

+\nabla(\rhou)=0

where

\rho

is the fluid density and

u

the fluid velocity. To obtain the equations of motion for incompressible flow, it is assumed that the density,

\rho

, is a constant.

Furthermore, occasionally one might consider the unsteady Stokes equations, in which the term

\rho

\partialu
\partialt
is added to the left hand side of the momentum balance equation.[1]

Properties

Re\to0.

Instantaneity
  • A Stokes flow has no dependence on time other than through time-dependent boundary conditions. This means that, given the boundary conditions of a Stokes flow, the flow can be found without knowledge of the flow at any other time.
    Time-reversibility
  • An immediate consequence of instantaneity, time-reversibility means that a time-reversed Stokes flow solves the same equations as the original Stokes flow. This property can sometimes be used (in conjunction with linearity and symmetry in the boundary conditions) to derive results about a flow without solving it fully. Time reversibility means that it is difficult to mix two fluids using creeping flow.

    While these properties are true for incompressible Newtonian Stokes flows, the non-linear and sometimes time-dependent nature of non-Newtonian fluids means that they do not hold in the more general case.

    Stokes paradox

    An interesting property of Stokes flow is known as the Stokes' paradox: that there can be no Stokes flow of a fluid around a disk in two dimensions; or, equivalently, the fact there is no non-trivial solution for the Stokes equations around an infinitely long cylinder.[11]

    Demonstration of time-reversibility

    A Taylor–Couette system can create laminar flows in which concentric cylinders of fluid move past each other in an apparent spiral.[12] A fluid such as corn syrup with high viscosity fills the gap between two cylinders, with colored regions of the fluid visible through the transparent outer cylinder.The cylinders are rotated relative to one another at a low speed, which together with the high viscosity of the fluid and thinness of the gap gives a low Reynolds number, so that the apparent mixing of colors is actually laminar and can then be reversed to approximately the initial state. This creates a dramatic demonstration of seemingly mixing a fluid and then unmixing it by reversing the direction of the mixer.[13] [14] [15]

    Incompressible flow of Newtonian fluids

    In the common case of an incompressible Newtonian fluid, the Stokes equations take the (vectorized) form:

    \begin{align}\mu\nabla2u-\boldsymbol{\nabla}p+f&=\boldsymbol{0}\\ \boldsymbol{\nabla}u&=0\end{align}

    where

    u

    is the velocity of the fluid,

    \boldsymbol{\nabla}p

    is the gradient of the pressure,

    \mu

    is the dynamic viscosity, and

    f

    an applied body force. The resulting equations are linear in velocity and pressure, and therefore can take advantage of a variety of linear differential equation solvers.

    Cartesian coordinates

    With the velocity vector expanded as

    u=(u,v,w)

    and similarly the body force vector

    f=(fx,fy,fz)

    , we may write the vector equation explicitly,

    \begin{align} \mu\left(

    \partial2u
    \partialx2

    +

    \partial2u
    \partialy2

    +

    \partial2u
    \partialz2

    \right)-

    \partialp
    \partialx

    +fx&=0\\ \mu\left(

    \partial2v
    \partialx2

    +

    \partial2v
    \partialy2

    +

    \partial2v
    \partialz2

    \right)-

    \partialp
    \partialy

    +fy&=0\\ \mu\left(

    \partial2w
    \partialx2

    +

    \partial2w
    \partialy2

    +

    \partial2w
    \partialz2

    \right)-

    \partialp
    \partialz

    +fz&=0\\ {\partialu\over\partialx}+{\partialv\over\partialy}+{\partialw\over\partialz}&=0 \end{align}

    We arrive at these equations by making the assumptions that

    P=\mu\left(\boldsymbol{\nabla}u+(\boldsymbol{\nabla}u)T\right)-pI

    and the density

    \rho

    is a constant.

    Methods of solution

    By stream function

    The equation for an incompressible Newtonian Stokes flow can be solved by the stream function method in planar or in 3-D axisymmetric cases

    Type of functionGeometryEquationComments
    Stream function,

    \psi

    2-D planar

    \nabla4\psi=0

    or

    \Delta2\psi=0

    (biharmonic equation)

    \Delta

    is the Laplacian operator in two dimensions
    Stokes stream function,

    \Psi

    3-D spherical

    E2\Psi=0,

    where

    E={\partial2\over\partialr2}+{\sin{\theta}\overr2}{\partial\over\partial\theta}\left({1\over\sin{\theta}}{\partial\over\partial\theta}\right)

    For derivation of the

    E

    operator see Stokes stream function#Vorticity
    3-D cylindrical
    2
    L
    -1

    \Psi=0,

    where

    L-1=

    \partial2
    \partialz2

    +

    \partial2
    \partial\rho2

    -

    1
    \rho
    \partial
    \partial\rho
    For

    L-1

    see [16]

    By Green's function: the Stokeslet

    The linearity of the Stokes equations in the case of an incompressible Newtonian fluid means that a Green's function,

    J(r)

    , exists. The Green's function is found by solving the Stokes equations with the forcing term replaced by a point force acting at the origin, and boundary conditions vanishing at infinity:

    \begin{align} \mu\nabla2u-\boldsymbol{\nabla}p&=-F\delta(r)\\ \boldsymbol{\nabla}u&=0\ |u|,p&\to0asr\toinfty \end{align}

    where

    \delta(r)

    is the Dirac delta function, and

    F\delta(r)

    represents a point force acting at the origin. The solution for the pressure p and velocity u with |u| and p vanishing at infinity is given by[1]

    u(r)=FJ(r),    p(r)=

    Fr
    4\pi|r|3

    where

    J(r)={1\over8\pi\mu}\left(

    I
    |r|

    +

    rr
    |r|3

    \right)

    is a second-rank tensor (or more accurately tensor field) known as the Oseen tensor (after Carl Wilhelm Oseen). Here, r r is a quantity such that

    F(rr)=(Fr)r

    .

    The terms Stokeslet and point-force solution are used to describe

    FJ(r)

    . Analogous to the point charge in electrostatics, the Stokeslet is force-free everywhere except at the origin, where it contains a force of strength

    F

    .

    For a continuous-force distribution (density)

    f(r)

    the solution (again vanishing at infinity) can then be constructed by superposition:

    u(r)=\intf\left(r'\right)J\left(r-r'\right)dr',    p(r)=\int

    f\left(r'\right)\left(r-r'\right)
    4\pi\left|r-r'\right|3

    dr'

    This integral representation of the velocity can be viewed as a reduction in dimensionality: from the three-dimensional partial differential equation to a two-dimensional integral equation for unknown densities.

    By Papkovich–Neuber solution

    The Papkovich–Neuber solution represents the velocity and pressure fields of an incompressible Newtonian Stokes flow in terms of two harmonic potentials.

    By boundary element method

    Certain problems, such as the evolution of the shape of a bubble in a Stokes flow, are conducive to numerical solution by the boundary element method. This technique can be applied to both 2- and 3-dimensional flows.

    Some geometries

    Hele-Shaw flow

    Hele-Shaw flow is an example of a geometry for which inertia forces are negligible. It is defined by two parallel plates arranged very close together with the space between the plates occupied partly by fluid and partly by obstacles in the form of cylinders with generators normal to the plates.

    Slender-body theory

    Slender-body theory in Stokes flow is a simple approximate method of determining the irrotational flow field around bodies whose length is large compared with their width. The basis of the method is to choose a distribution of flow singularities along a line (since the body is slender) so that their irrotational flow in combination with a uniform stream approximately satisfies the zero normal velocity condition.[9]

    Spherical coordinates

    Lamb's general solution arises from the fact that the pressure

    p

    satisfies the Laplace equation, and can be expanded in a series of solid spherical harmonics in spherical coordinates. As a result, the solution to the Stokes equations can be written:

    \begin{align}u&=

    n=infty
    \sum
    n=-infty,n ≠ -1

    \left[

    (n+3)r2\nablapn
    2\mu(n+1)(2n+3)

    -

    nxpn
    \mu(n+1)(2n+3)

    \right]

    n=infty
    +...\\sum
    n=-infty

    [\nabla\Phin+\nabla x (x\chin)]\\ p&=

    n=infty
    \sum
    n=-infty

    pn\end{align}

    where

    pn,\Phin,

    and

    \chin

    are solid spherical harmonics of order

    n

    :

    \begin{align}pn&=rn

    m=n
    \sum
    m=0
    m(\cos\theta)(a
    P
    mn

    \cosm\phi+\tilde{a}mn\sinm\phi)\\ \Phin&=rn

    m=n
    \sum
    m=0
    m(\cos\theta)(b
    P
    mn

    \cosm\phi+\tilde{b}mn\sinm\phi)\\ \chin&=rn

    m=n
    \sum
    m=0
    m(\cos\theta)(c
    P
    mn

    \cosm\phi+\tilde{c}mn\sinm\phi)\end{align}

    and the

    m
    P
    n
    are the associated Legendre polynomials. The Lamb's solution can be used to describe the motion of fluid either inside or outside a sphere. For example, it can be used to describe the motion of fluid around a spherical particle with prescribed surface flow, a so-called squirmer, or to describe the flow inside a spherical drop of fluid. For interior flows, the terms with

    n<0

    are dropped, while for exterior flows the terms with

    n>0

    are dropped (often the convention

    n\to-n-1

    is assumed for exterior flows to avoid indexing by negative numbers).

    Theorems

    Stokes solution and related Helmholtz theorem

    See also: Stokes' law. The drag resistance to a moving sphere, also known as Stokes' solution is here summarised. Given a sphere of radius

    a

    , travelling at velocity

    U

    , in a Stokes fluid with dynamic viscosity

    \mu

    , the drag force

    FD

    is given by:

    FD=6\pi\muaU

    The Stokes solution dissipates less energy than any other solenoidal vector field with the same boundary velocities: this is known as the Helmholtz minimum dissipation theorem.[1]

    Lorentz reciprocal theorem

    The Lorentz reciprocal theorem states a relationship between two Stokes flows in the same region. Consider fluid filled region

    V

    bounded by surface

    S

    . Let the velocity fields

    u

    and

    u'

    solve the Stokes equations in the domain

    V

    , each with corresponding stress fields

    \sigma

    and

    \sigma'

    . Then the following equality holds:

    \intSu(\boldsymbol{\sigma}'n)dS=\intSu'(\boldsymbol{\sigma}n)dS

    Where

    n

    is the unit normal on the surface

    S

    . The Lorentz reciprocal theorem can be used to show that Stokes flow "transmits" unchanged the total force and torque from an inner closed surface to an outer enclosing surface. The Lorentz reciprocal theorem can also be used to relate the swimming speed of a microorganism, such as cyanobacterium, to the surface velocity which is prescribed by deformations of the body shape via cilia or flagella.[17] The Lorentz reciprocal theorem has also been used in the context of elastohydrodynamic theory to derive the lift force exerted on a solid object moving tangent to the surface of an elastic interface at low Reynolds numbers.[18] [19]

    Faxén's laws

    Faxén's laws are direct relations that express the multipole moments in terms of the ambient flow and its derivatives. First developed by Hilding Faxén to calculate the force,

    F

    , and torque,

    T

    on a sphere, they take the following form:

    \begin{align} F&=6\pi\mua\left(1+

    a2
    6

    \nabla2\right)

    infty(x)|
    v
    x=0

    -6\pi\muaU\ T&=8\pi\mua3(\Omegainfty(x)-\omega)|x=0\end{align}

    where

    \mu

    is the dynamic viscosity,

    a

    is the particle radius,

    vinfty

    is the ambient flow,

    U

    is the speed of the particle,

    \Omegainfty

    is the angular velocity of the background flow, and

    \omega

    is the angular velocity of the particle.

    Faxén's laws can be generalized to describe the moments of other shapes, such as ellipsoids, spheroids, and spherical drops.

    References

    External links

    Notes and References

    1. Kim, S. & Karrila, S. J. (2005) Microhydrodynamics: Principles and Selected Applications, Dover. .
    2. Book: Kirby, B.J.. Micro- and Nanoscale Fluid Mechanics: Transport in Microfluidic Devices.. 2010. Cambridge University Press. 978-0-521-11903-0. 2010-01-15. 2019-04-28. https://web.archive.org/web/20190428234717/http://www.kirbyresearch.com/textbook/. dead.
    3. Dusenbery, David B. (2009). Living at Micro Scale. Harvard University Press, Cambridge, Massachusetts
    4. Book: L. Gary Leal

      . Leal, L. G. . L. Gary Leal . Advanced Transport Phenomena: Fluid Mechanics and Convective Transport Processes . 2007.

    5. Chwang, A. and Wu, T. (1974). "Hydromechanics of low-Reynolds-number flow. Part 2. Singularity method for Stokes flows" . J. Fluid Mech. 62(6), part 4, 787–815.
    6. Web site: Brennen . Christopher E . Singularities in Stokes' Flow . caltech.edu . 18 July 2021 . 1 . en . 10 September 2021 . https://web.archive.org/web/20210910023610/http://brennen.caltech.edu/fluidbook/basicfluiddynamics/Stokesflow/singularities.pdf . dead .
    7. Generalized fundamental solutions for unsteady viscous flows . Physical Review E . 2001 . Shu, Jian-Jun . Chwang, Allen T. . 63 . 5 . 051201 . 10.1103/PhysRevE.63.051201 . 11414893 . 1403.3247 . 2001PhRvE..63e1201S . 22258027 .
    8. Fundamental solutions for micropolar fluids . Journal of Engineering Mathematics . 2008 . Jian-Jun . Shu . Lee, J.S. . 61 . 1 . 69–79 . 10.1007/s10665-007-9160-8 . 1402.5023 . 2008JEnMa..61...69S . 3450011 .
    9. Book: George Batchelor

      . Batchelor, G. K.. George Batchelor . Introduction to Fluid Mechanics . 2000 . Cambridge University Press . 978-0-521-66396-0.

    10. Happel, J. & Brenner, H. (1981) Low Reynolds Number Hydrodynamics, Springer. .
    11. Book: Lamb, Horace. Hydrodynamics. 1945. Dover Publications. New York. 602–604. Sixth.
    12. [C. David Andereck]
    13. Dusenbery, David B. (2009). Living at Micro Scale, pp.46. Harvard University Press, Cambridge, Massachusetts .
    14. Archived at Ghostarchive and the Wayback Machine: Web site: Laminar Flow . YouTube.
    15. Web site: Untitled Document.
    16. 7 . 4 . 529–549. Payne . LE. WH Pell. The Stokes flow problem for a class of axially symmetric bodies. Journal of Fluid Mechanics. 1960. 1960JFM.....7..529P. 10.1017/S002211206000027X. 122685039 .
    17. Stone. Howard A.. Samuel, Aravinthan D. T. . Propulsion of Microorganisms by Surface Distorsions. Physical Review Letters. November 1996. 77. 19. 19. 4102–4104. 1996PhRvL..77.4102S . 10.1103/PhysRevLett.77.4102 . 10062388.
    18. Daddi-Moussa-Ider. A.. Rallabandi, B.. Gekle, S.. Stone, H. A.. Reciprocal theorem for the prediction of the normal force induced on a particle translating parallel to an elastic membrane. Physical Review Fluids. August 2018. 3. 8 . 084101. 10.1103/PhysRevFluids.3.084101. 1804.08429 . 2018PhRvF...3h4101D . 55619671 .
    19. Rallabandi. B.. Saintyves, B.. Jules, T.. Salez, T. Schönecker, C.. Mahadevan, L.. Stone, H. A.. Rotation of an immersed cylinder sliding near a thin elastic coating. Physical Review Fluids. July 2017. 2. 7 . 074102. 10.1103/PhysRevFluids.2.074102. 1611.03552 . 2017PhRvF...2g4102R . 9790910 .