Equation of Motion for Viscous Flow
Ain A. Sonin Department of Mechanical Engineering Massachusetts Institute of Technology Cambridge, Massachusetts 02139 2003 (9th edition)

1. 2. 3. 4. 5. Surface stress …………………………………………………………. The stress tensor ……………………………………………………… Symmetry of the stress tensor ………………………………………… Equation of motion in terms of the stress tensor ……………………… Stress tensor for Newtonian fluids ……………………………………. The shear stresses and ordinary viscosity …………………………. The normal stresses ……………………………………………….. General form of the stress tensor; the second viscosity …………… 6. The Navier-Stokes equation …………………………………………… 7. Boundary conditions ………………………………………………….. 2 3 7 9 12 12 13 18 21 23

Appendix A: The Navier-Stokes and mass conservation equations in cylindrical coordinates, for incompressible flow ………………….24 Appendix B: Properties of selected fluids …………………………..……… 26

„ Ain A. Sonin 2002



Surface Stress

Quantities like density, velocity, and pressure are defined by a value at every point in the v v fluid at every time t. The density r(r ,t) and pressure p( r,t) are scalar fields. They have a v v numerical value at every point in space at any instant in time. The velocity v (r,t) is a vector field; it is defined by a direction as well as a magnitude at every point.

Fig. 1: A surface element at a point in a continuum.

The surface stress is a more complicated type of quantity. One cannot talk of the stress at a point without first defining the particular surface through that point on which the stress v acts. A small fluid surface element centered at the point r is defined by its area dA (the v prefix d indicates a very small but finite quantity) and by its outward unit normal vector n . v The stress exerted by the fluid on the side toward which n points on the surface element is defined as v dF v s = lim dAÆ0 dA (1)

v where dF is the force exerted on the surface by the fluid on that side (only one side is involved). In the limit dA Æ 0 the stress is independent of the magnitude of the area, but will v in general depend on the orientation of the surface element, which is specified by n . In other words, v v v v s = s ( x,t, n) . (2)

v v The fact that s depends on n as well as x, y, z and t appears at first sight to complicate matters considerably. One apparently has to deal with a quantity that depends on six v independent variables (x, y, z, t, and the two that specify the orientation n ) rather than four.

We shall see that the stress on any surface anywhere in the fluid can in general be specified in terms of six scalar functions of x.n .t. disc-shaped fluid particle at r . therefore. When shear forces are present. 2 The Stress Tensor The first and simplest thing that Newton's law implies about the surface stress is that. v v v v v v s ( r. . can be expressed in terms of a single scalar field v p( r.t. with very small area dA and thickness dh.pn (no shear forces) (3) v where p. Newton's law imposes a somewhat more complicated v v constraint on the relationship between s and n . to that on a surface element with an opposite v orientation . y.n . 2.3 v Fortunately. nature comes to our rescue. In the absence of shear stresses. but opposite in direction. at v a given point. and its magnitude is independent of the surface orientation. Newton's law requires that the surface stress have the particularly simple form v v s = . it must depend v on n in a relatively simple way. so that the two faces of the disc are brought toward coincidence in space. that is.n) = -s (r . in the absence of shear forces. This is Pascal's principle. at any point in the fluid. These six are the independent components of a quantity called the stress tensor. the inertial term on the left and the body force term on the right become arbitrarily small compared with the two surface force terms. We find that because s is a stress. is a function of r and t only. as shown in Fig. anywhere in the fluid. the stress is always normal to the surface on which it acts.t) provided there are no shear forces. which states that in the absence of shear forces. When we let dh approach zero. and t. The equation of motion for this fluid particle reads v v Dv v v v v rdhdA = s ( n)dA + s (.n)dA + rdhdAG Dt (5) v where G is the body force per unit mass. This gives rise to the relatively simple form of the equation of motion for inviscid flow.. n) (4) v This result can be obtained by considering a thin. and (4) follows immediately. the stress on any surface. We have seen that. the stress on a surface element with an orientation n must be equal in magnitude. the magnitude of the normal compressive stress. One side of the disc has an orientation v v n and the other . as they always are in practice except when the fluid is totally static in some reference frame. z.

Let us take as "reference stresses. and a surface oriented in the positive z-direction (Fig. in terms of their components: v v v v v s (i ) = t xx i + t yx j + t zxk v v v v v s ( j ) = t xy i + t yy j + t zy k v v v v v s ( k ) = t xz i + t yz j + t zzk (6) . Newton's law also implies that the stress has a more profound attribute. 3). We can write these three reference stresses. 2: Illustration for equation (4) Figure 3: Reference stresses at a point in the continuum. which of course are vectors. the values of the stresses that are exerted on a surface oriented in the positive xdirection." at a given point r and instant t. a surface oriented in the positive y-direction. The stress at a given point depends on the orientation of the v surface element on which it acts. which leads to the concept of the stress tensor.4 Fig.

The areas of the three orthogonal faces are related to dA by dAx = cosq nxdA = nxdA dAy = cos qnydA = nydA dAz = cosq n zdA = nzdA (7) Fig. y. as shown in Fig.. etc. and s ( k ) or. and the reference stresses themselves also depend on x. q nx is the angle between n and the x-axis and nx is the x-component of n . and the second indicates the outward normal of the surface on which it acts. Consider a fluid particle which at time t has the shape of a small tetrahedron centered at v x. respectively. y. in terms of their nine components t xx . In (6) the t ij ' s are of course functions of position x.. t yx and t zx represent the x. we have simply not indicated this dependence. 4. y. z. y. The first subscript on t ij identifies the direction of the stress. and so on. y. t yx . Consider what Newton's law tells us about the forces acting on the tetrahedron as we let v it shrink in size toward the point r around which it is centered. 4: Tetrahedron-shaped fluid particle at (x. z. (Fig. One of its four faces has an area dA and an arbitrary outward normal n .. y and z directions. and time t.5 Thus. We shall now show. t xx .. t zz . z). that the stress on a surface having v v v v anyvorientation n at the point r can be expressed in terms of the reference stresses s ( i ). and t. and z components of the stress acting on the surface whose outward normal is oriented in the positive x-direction. z. again by using Newton's law. v v v s ( j ) . and the other three faces have outward normals in the negative x. 3). more specifically. Since the ratio of the mass . where dAx represents the area of the surface whose outward normal is in the negative xv v direction.

v v v v v v s ( n)dA + s ( j )dAx + s (k )dAz = 0 . v v v v Thus the stress s ( n) acting at x. n) = t ij (r . both the mass times acceleration and the body force become arbitrarily small compared with the surface force as the tetrahedron is shrunk to a point (c. Using this result and (7) for the areas.t. These nine quantities.6 of the tetrahedron to the area of any one of its faces is proportional to the length of any one of the sides.t on a surface with any arbitrary orientation n can be expressed in terms of the nine reference stress components (10) txx ty x txz txy ty y tzy txz tyz tzz . etc. Hence. and using (10). or z and where it is understood that any term which contains the same index twice actually represents the sum of all such terms with all possible values of the repeated index (for example. each of which depends on position and time. (8) becomes v v v v v v v v s ( n) = s (i )nx + s ( j )ny + s ( k )nz . (9) Alternatively. if we use (6)vtovwrite the reference stresses in terms of their components.i direction is the v negative of the stress on a surface in the + i direction. where i and j can represent x. one can compute the surface stress acting on any surface drawn through that point by determining v the components of the outward unit normal n of the surface involved.f. in the limit as the tetrahedron is shrunk to a point. that is. Once the stress tensor components are known at a given point. (8) v Now we know from (4) that the stress on a surface pointing in the .t)n j . (11) . are the stress tensor components. we obtain the components of s ( n) as v s x (n) = t xx nx + t xy ny + t xz nz v s y (n) = t yx n x + t yy ny + t yz nz v s z ( n) = t zxnx + t zyny + t zznz . In this notation (10) reads v v v s i ( r. Equation (10) can be written more succinctly in conventional tensor notation. y. (5) and the paragraph that follows it). the surface forces on the four faces must balance. sii ≡ sx x + sy y + szz).

q and z directions. t ij being the i-component of the stress on the surface whose normal points in the j-direction. the reference stresses are the stresses on the surfaces pointing in the positive x. stress tensor components." and not just an arbitrary bunch of nine scalar quantities? The answer lies in the special way the values of these nine quantities transform when one changes one's reference frame from one coordinate system to another. v has a doubly infinite set of values corresponding to different surface orientations n . z. it does so almost as a solid body (i. the stress tensor represents the nine components of the three reference v stresses at the point r and time t in question. that is. Admittedly. and dz (Fig. at any given position and time. If the cube has an angular velocity q z in the zdirection.e. at essentially zero angular distortion). the stress tensor would be comprised of the components of the stresses acting on the three surfaces having outward normals in the positive r. A set of nine quantities t ij that transform in this manner is by definition a tensor of second rank. 5). where the fluid particle is reduced to a point. Since we shall be taking the limit where dx. y. d z Æ 0. etc. whose three components transform so that the magnitude and direction of the vector remain invariant. we can safely assume that the values of the density. a tensor of zeroth rank is a scalar. velocity. Equation (10) tells us that when a coordinate change is made. In a cylindrical coordinate system. Why are the quantities t ij "tensor components. but not on the orientation of the surface on which the stress acts. For convenience. Physically. if the cube rotates by an infinitesimal amount. (A tensor of first rank is a vector. The reference stresses are by custom chosen as the stresses on the three surface elements that have outward normals in the direction of the positive axes of the coordinate system being used. a finite angular distortion would ˙ require infinite shear in a viscous fluid.7 The importance of the stress tensor concept in continuum theory is this: It allows us to describe the state of stress in a continuum in terms of quantities that depend on position and time. we can derive from Newton's law . dy. nine such quantities are needed (actually only six are independent. Still. and the stress tensor is made up of the nine components of these three stresses. dy. since in the limit dx. y. it is far easier to deal with them than with a single quantity which. dy. dz Æ 0. say. the three sums t ij nj must transform as components of a vector. What is more. Thus in our Cartesian coordinates. t ij = t ji for i # j. as we shall see shortly). The proof follows from considering the angular acceleration of a little fluid particle at x. and rotates like a solid body. a single quantity whose magnitude remains invariant with coordinate changes. and z directions. are almost uniform throughout the cube.) 3 Symmetry of the Stress Tensor One further piece of information emerges from applying Newton's law to an infinitesimal fluid particle: The stress tensor is in most cases symmetric. we let it be shaped like a little cube with infinitesimal sides dx.

x and y being the Cartesian coordinates fixed in the rotating cube. relative to an axis running through the center of the cube parallel to the z-axis. The net torque about an axis through the cube's center. Equation (13) is obtained ˙ by writing the cube’s angular velocity as vq = q (t)r .8 Fig. 5: Illustration of the reason for the stress tensor's symmetry. dt (12) Iz = Ú Ú Ú r(x -dx 2 -dy 2 -dz 2 2 + y 2 )dxdydx = r[(dx) 2 + (dy) 2 ] 12 dxdydz (13) is the moment of inertia of the cube and Tz is the net torque acting on the cube. On the face with n = -i .axis. Let us assume for the sake . that at any given instant its angular velocity increases according to Iz where +dx 2 +dy 2 +dz 2 ˙ dq z = Tz . where r 2 = x 2 + y 2 . On the face with n = i . v v 5). is caused by the shear forces (the pressure forces act through the cube’s center) and by any volumetric torque exerted by the external body force field. The torque in (12) is obtained by considering the stresses acting on the cube (Fig. written in angular momentum form for a material volume. A body force field like gravity acts through the cube's center of mass and exerts no torque about that point. the corresponding stresses have the same magnitudes but opposite directions [see (10) or (4)]. parallel to the z. for example. there is by definition a stress t xx in the positive xv v direction and a stress t yx in the positive y-direction.

2 t xydy dz + t zdx dydz 2 2 (14) = (t yx . x. with sides dx. z. Ferrohydrodynamics. or y. the stress tensor has only six independent components. 6. Although the sides are small. this reduces to t yx = t xy . in which case (16) shows that the off-diagonal or shear terms in the stress tensor are symmetric. for example (e. Volumetric body torque can exist in magnetic fluids. or z. z) as in Fig. E. in Cartesian coordinates they are in the order x. they are not zero and the components of the stress tensor will have slightly different values on the faces of the cube than at the center . y. y. dy Æ 0. The net torque in the z-direction around the particle's center would then be Tz = 2 dx dx t yx dydz . j. (17) This means that three of the nine components of the stress tensor can be derived from the remaining ones. Rosensweig. 1985. the result that the off-diagonal stress tensor components must satisfy t ji = t ij + t k . that is. In what follows we shall assume that volumetric body torque is absent. y). or. y. and z axes at time t. 4 Equation of Motion in Terms of the Stress Tensor A general equation of motion in differential form may be derived by applying Newton's law to a small but finite fluid particle.g.9 v of generality. dy. z. (16) where i. more generally. Consider again a particle which at time t has the shape of a cube centered about (x.g.t xy + tz )dxdydz From (12) .tz . see R.t xy + t z = ˙ r dq z [(dx)2 + (dy)2 ] 12 dt (15) As we approach a point in the fluid by letting dx. t ji = t ij (i#j) . that the external body force may exert a torque t per unit volume at the particle's location. x. Chapter 8). k form a right-hand triad (e. however. and dz parallel to the x.(14) we see that t yx .

6. For example. y. the center of the cube. and t ij - ∂t ij dx ∂x 2 at the opposite face. Fig. respectively. except that the first subscript x is replaced by y and z. The equation of motion can now be written down directly for the cubical fluid particle in Fig. The x-component of the equation states that the mass times the acceleration equals the net surface force plus the body force acting on the particle: . expressed in terms of the stress tensor. we identify the quantity within the brackets as the net x-component of surface force per unit volume at a point in a fluid. Figure 6 shows all those stresses which act on the cube in the x-direction. The net x-component of surface force on the cube is obtained by multiplying the stresses by the areas on which they act and summing: Ê ∂t xx ∂t xy ∂t xz ˆ Á + + dxdydz . The arrows indicate the directions of the stresses for positive values of tij [see (10)]. then their values will be t ij + ∂t ij dx ∂x 2 at the face whose outward normal is in the positive x-direction. The expressions for the y and z components are similar.10 of the cube. if the stress tensor components t ij are specified at (x. Ë ∂x ∂y ∂z ¯ (18) Since dxdydz is the particle's volume. z). 6: x-direction surface stresses acting on a fluid particle.

This yields r Dv x Ê ∂t xx ∂t xy ∂t xz ˆ =Á + + + rGx . incomplete as it stands. Dt Ë ∂x ∂y ∂z ¯ (20a) For the y and z components we obtain similarly r Dv y Ê ∂t yx ∂t yy ∂t yz ˆ =Á + + + rGy Dt Ë ∂x ∂y ∂z ¯ Dvz Ê ∂t zx ∂t zy ∂t zz ˆ =Á + + + rGz Dt Ë ∂x ∂y ∂z ¯ (20b) r (20c) or. and Gx is the xcomponent of the external body force per unit mass. however. one must specify the stress tensor components and the body force components. and z is implied. and is not based on any assumption other than that the continuum hypothesis applies. the acceleration term is absent and the gravitational loads induced by the weight of the structure itself are often negligible compared with externally applied forces.1 Eq. y. r Dvi ∂t ij = + rGi Dt ∂x j (20) where a summation over j=x. for example. the force G per unit mass is well known and is of the same form for all substances. Equation (20) states that at a given point and time. the mass per unit volume times the acceleration in the i-direction (the left-hand term) equals the the net surface force per unit volume in the i-direction (the first term on the right) plus the body force per unit volume in the i-direction (the second term on the right). (20) is. 1In static solid deformations. just as one must define the forces acting on a solid particle before one can derive its motion. whether fluid or solid. In a gravitational field. D/Dt represents the substantial derivative. In such cases the equation of motion reduces to the simple statement that the net surface stress per unit volume is zero at every point in the medium. . The specification of the body v force is straightforward. To complete it. which is defined elsewhere.11 rdxdy dz Dvx Ê ∂t xx ∂t xy ∂t xz ˆ =Á + + dx dy dz + rdxdy dzGx Dt Ë ∂x ∂y ∂z ¯ Here. The form of the stress tensor is different for different classes of materials. The equation applies quite generally to any continuous distribution of matter. more succinctly.

is that it keeps deforming. The defining attribute of a simple fluid. however. e. just like an elastic solid (where stresses are proportional to strains) is the simplest type of deformable solid. Obviously. the faster the rate of shear deformation. 7 depicts the deformation of a fluid particle as it moves between time t and time t+dt. The Newtonian model of fluid response is based on three assumptions: (a) (b) (c) shear stress is proportional to the rate of shear strain in a fluid particle. dy dx dg xy . shear stress is zero when the rate of shear strain is zero. is applied to it. It is observed. where stresses and strains are linearly related. that a fluid tends to resist the rate of deformation: the higher the applied shear stress.g t xy . The shear stresses and the ordinary viscosity To implement the Newtonian assumptions we consider first a typical shear term in the tensor. or straining. no unique relation can exist between the shear stresses and the shear strains if strain can increase indefinitely at constant shear. The simplest model of a solid continuum is the well-known elastic one.12 5 Stress Tensor for Newtonian Fluids There remains the task of specifying the relationship between the stress tensor components and the flow or deformation field. the stress to rate-of-strain relation is isotropic—that is. however. no matter how small. In many fluids the relation between stress and rate of strain in a fluid particle is linear under normal conditions. A Newtonian fluid is the simplest type of viscous fluid. Fig. there is no preferred orientation in the fluid. In this interval the shear stress t xy produces in the fluid particle an incremental angular strain dg xy Ê ∂v x ˆ Ê ∂vy ˆ Á Á dydt dxdt Ë ∂y ¯ Ë ∂x ¯ = + . as long as any shear stress.

viscosity coefficient. one of the stress tensor's diagonal terms. (22) The coefficient of proportionality is the same in all three shear stresses because a Newtonian fluid is isotropic. (22c) Ê ∂v ∂v ˆ t ij = m Á i + j ˜ Ë ∂x j ∂xi ¯ (i ≠ j) . its corner A is at (x. The derivation of such a term's form is not as simple as that of the shear terms. it moves and deforms as in Fig. Between t and t+dt. 7: Shear deformations in a fluid particle. that is. that is. We will again be considering the limit of a particle "at a point". The normal stresses Next consider a typical normal stress. y and z axes. or "ordinary". Similarly. The rate of angular (or shear) strain in the fluid particle as seen by an observer sitting on it is therefore Dg xy ∂vx ∂v y = + . but can nevertheless be done in fairly physical terms by noting that linear and shear deformations generally occur hand in hand. Dt ∂y ∂x The Newtonian assumptions (a) and (b) thus require that (21) t xy = m Dg xy Ê ∂v ∂v ˆ = mÁ x + y . The sides AB and AD will in general rotate by unequal . The trick is to find how the linear stresses and deformations are related to the shear stresses and deformations. 8. At time t . t xz = m Dg xz Ê ∂v x ∂v z ˆ =m + Ë ∂z ∂x ¯ Dt Dg yz Ê ∂v ∂v ˆ = mÁ y + z Ë ∂z ∂y ¯ Dt (22b) t yz = m or in general. y. z). and is a property of the fluid. Ë ∂y ∂x ¯ Dt (22a) where the coefficient of proportionality m is called the shear.13 Fig. say t xx . Consider a small fluid particle which at time t is a small cube with sides of length h parallel to the x. the limit h Æ 0.

t yx . strain rates on the (x'. we know the relationship between the shear stress and the rate of angular strain of the particle in the (x. A force balance in the x'-direction requires that ¢ t xx = t xx + t yy + t yx . y) frame. This will result in a shear deformation of the particle. expressed in terms of the stress tensor components in the original and the rotated reference frames. it will give rise to linear deformations in the x' and y' directions which are rotated 45o relative to the x and y axes. Since the reference frames are arbitrary. Now.t ¢ = 2t yx .14 amounts. If we can connect the shear stresses in this frame and the stresses in the rotated (x'. 2 (23) Similarly. The shear deformation will cause one of the diagonals AC and BD to expand and the other to contract. We start by considering the forces acting on one half of the fluid particle in Fig. 8: Why shear and linear deformations are related. Since we are considering the limit dh Æ 0. the relationship between stresses and rates of strain for the (x'. y') frame. and the shear strain rates in the (x. where the ratio of volume to area vanishes. y') frame. we will arrive at a relation between the stresses and the strain rates in the (x'. Figure 9 shows the surface forces on particle ABD. that is. ¢ yy (25) . a force balance in the y'-direction on the triangular particle ACD requires that t yy = ¢ Adding (23) and (24) we obtain t xx + t yy . y') frame must be general in form. y) frame and the Fig. y') frame. 8: the triangular fluid particle ABD as shown in Fig. 2 (24) t xx . 9. the equation of motion for the particle will reduce to the statement that the surface forces must be in balance.

t ¢ = 2 m ¢ yy Dg xy Dt (26) which relates the diagonal stress tensor terms in the (x'. 8. To close the loop we must relate the angular strain rate in the (x. y) frame to the strain rates in the (x'. 10: Deformations of the two triangular particles in Fig. Fig. y) frame. and d in the figure are related to the incremental linear strains de and angular strains dg in the (x. 9: Stresses on two halves of the particle in Fig. y') frame. this becomes t xx . c. y) frame by c h a de y = h de x = (27) .15 Using the relation (22a) between the shear stress and the rate of strain. Figure 10 shows the deformations of the triangular particles ABD and ACD between t and t+dt. 9. b. y') frame to the angular strain rate in the (x. The deformations a. Fig.

h Here. y) plane: de x ¢ . de y is its linear strain in the y-direction. Recalling that ACD is an isoscoles triangle at time t. The linear strain in the x' direction can be computed in terms of these quantities from the fractional stretching of the diagonal AC. which is oriented in the x' direction. Between t and t+dt.t y¢y ¢ = 2m Á x ¢ . 11. (30) The differentials refer to changes following the fluid particle. y) frame by using (26). Dt Dt Dt If we now eliminate the reference to the (x.16 dg xy = b+ d . the linear strain suffered by the fluid particle's side parallel to the x' axis is .de y¢ = dg xy . dex ¢ = = ˜ Á x y xy (AC) 2Ëh h h ¯ 2 h 2 (28) The linear strain in the y' direction is obtained similarly from the fractional stretching of the diagonal BD of the triangular particle ABD as de y¢ = d(BD) 1 = (de x + de y . and dg xy is the angular strain in the x-y plane. (BD) 2 (29) The sum of the last two equations shows that the difference of the linear strains in the x' and y' directions is equal to the angular strain in the (x.dg xy ) . Ë Dt Dt ¯ (32) (31) The linear strain rates can be evaluated in terms of the velocity gradients by referring to Fig. we obtain a+ d b+c + d(AC) 2 2 = 1 Ê c + a + b + d ˆ = 1 (de + de + dg ) . The rates of strain following the fluid motion are therefore related by De x ¢ De y ¢ Dg xy = . de x is the linear strain (increase in length divided by length) of the particle in the xdirection. and that the deformations between t and t+dt are infinitesimally small. we obtain De y ¢ ˆ Ê De t x ¢x ¢ .

A similar equation is obtained for the linear strain rate in the y' direction.17 ∂vx ¢ dxdt ∂v ∂x ¢ de x ¢ = = x ¢ dt dx ∂x ¢ so that De x ¢ ∂vx ¢ = . z') plane. Ë ∂x ¢ ∂y ¢ ¯ Similarly we obtain. Ë ∂x ¢ ∂z ¢ ¯ (35) t x ¢x ¢ = t x ¢x ¢ + t y ¢y¢ + t z¢z ¢ 3 + 2m ∂v x ¢ 2 Ê ∂vx ¢ ∂v y ¢ ∂vz ¢ ˆ . (34) t x ¢ x ¢ . we now obtain Ê ∂v ∂v ˆ t x ¢x ¢ .t z ¢ z ¢ = 2m Adding equations (34) and (35) we get Ê ∂v x ¢ ∂vz ¢ ˆ . 11: Linear deformations of a fluid particle. Dt ∂x ¢ (33) Fig.mÁ + + ∂x ¢ 3 Ë ∂x ¢ ∂y ¢ ∂z¢ ¯ (36) Since the coordinate system (x'. y') is arbitrary.y ¢ . by viewing the particle in the (x'.t y¢y ¢ = 2m Á x ¢ . this relationship must apply in any coordinate system. We thus have our final result: . Using these relations in (32).

which is superposed on the mean drift motion associated with flow. and no sum is implied in this equation when one writes down the general form of the diagonal terms by setting j=i . bouncing against other molecules. y and z directions. An individual molecule in a fluid executes a random thermal motion.m— ⋅ v ∂x 3 (37) (t xx + t yy + t zz ) 3 =- t ii 3 (38) is the "mechanical" pressure. 1967. it is evident that all the terms of the Newtonian stress tensor can be represented by the equation Ê ∂v ∂v ˆ 2 Ê v t ij = . From these expressions and (22) for the off-diagonal terms. to be distinguished from the "thermodynamic" pressure which is discussed below.Ë pm + m— ⋅ v ˆ d ij + mÁ i + j ˜ ¯ 3 Ë ∂x j ∂x i ¯ where (39) dij = 1 if i=j = 0 if i#j is the Kronecker delta. and serves as a measure of local normal compressive stress in viscous flows where that stress is not the same in all directions. The mechanism whereby stress is exerted by one fluid region against another is actually a molecular one. respectively. K. Batchelor.pm + 2m where the quantity pm = - ∂vx 2 v . it can be shown that pm as defined in (38) is in fact equal to the average normal compressive stress on the surface of a sphere centered on the point in question. The mechanical pressure is the negative of the average value of the three diagonal terms of the stress tensor. Note that although the definition is phrased in terms of the normal stresses on surfaces pointing in the x. Cambridge University Press. Note that (39) represents any single component of the tensor. The mechanical pressure is a well defined physical quantity.18 t xx = . General form of the stress tensor and the second viscosity Expressions similar to (37) are obtained for t yy and t zz . An Introduction to Fluid Mechanics. p. and is a true scalar since the trace of a tensor remains invariant under coordinate transformations. in the limit as the sphere's radius approaches zero (see G. except that ∂vx ∂x is replaced by ∂v y ∂y and ∂v z ∂z . Normal stress on a surface arises from average momentum transfer by the fluid .141 ff).

The shear viscosity is mainly a function of temperature for both gases and liquids. (40) Here. each molecule imparting an impulse as it collides with the surface and rebounds. on the other hand. on the other hand. Molecules which move by random thermal motion transverse to the flow from a higher mean velocity region toward a lower mean velocity region carry more streamwise momentum than those moving in the opposite direction. one big difference between gases and liquids: the viscosity of gases increases with temperature. The viscosity of water. Equation (39) contains only a single empirical coefficient. and the net transfer of the streamwise molecular momentum manifests itself as a shear stress on the macroscopic level at which we view the fluid. A second coefficient is. The viscosity of air. alternative names are "second coefficient of viscosity" and "bulk viscosity"). Normal stress is exerted even in a static. . Shear stress arises when there is a mean velocity gradient in the direction transverse to the flow. which represents the rate of change of fluid volume per unit volume as seen by an observer moving with the fluid. In gases the molecules are sparsely distributed and spend most of their time in free flight rather than in collisions with each other. and is called the expansion viscosity (Batchelor. the dependence on pressure being relatively weak. C. for example. however. The molecular theory of the shear viscosity coefficient is quite different for gases and liquids. It is customary to assume a simple linear relationship which may be thought of as being in the spirit of the original Newtonian postulates. O. introduced in our quest for a complete set of flow equations when we invoke the fluid's equation of state and are forced to ask how the "thermodynamic" pressure which appears in that equation is related to the mechanical pressure pm . Arguments derived from statistical thermodynamics suggest that this equilibrium pressure may differ from the mechanical pressure when the fluid is composed of complex molecules with internal degrees of freedom. Hirschfelder.19 molecules executing their random thermal motion. nondeforming fluid. usually at a rate much faster than the increase in gases. decreases by almost a factor of four over the same temperature range. Molecular Theory of Liquids Gases and Liquids). The equation of state expresses the fluid's density as a function of temperature and pressure under equilibrium conditions. The quantity that provides the simplest measure v of rate of density change is the divergence of the velocity vector. but in fact rests on much more tenuous experimental grounds: v pm = p . Bird. The "thermodynamic" pressure which appears in that equation is therefore the hypothetical pressure that would exist if the fluid were in static equilibrium at the local density and temperature. while the viscosity of liquids decreases.l— ⋅ v . In liquids. the molecules spend most of their time in the short-range force fields of their neighbors (see for example J. F. however. There is. B. Curtiss and R. — ⋅ v . the shear or ordinary coefficient of viscosity m. l is an empirical coefficient which has the same dimension as the shear viscosity m. and that the difference should depend on the rate at which the fluid density or pressure is changing with time. increases by 20% when temperature increases from 18o C to 100o C. An Introduction to Fluid Dynamics.

The physical v interpretation of — ⋅ v is that it represents the rate of change of a fluid particle's volume recorded by an observer sitting on the particle. however. This implies that the thermodynamic pressure tends to be higher than the mechanical pressure when the v mechanical pressure is decreasing (volume increasing. the Newtonian stress tensor reads Ê ∂v ∂v ˆ È Ê2 v˘ t ij = . It can be shown rigorously that l=0 for dilute monatomic gases. and lower than the v mechanical pressure when the pressure is increasing (volume decreasing. Only when density changes are induced either over extremely small distances (e. For water l is about three times larger than m. except in very special and difficult-to-achieve circumstances. Indeed. we can approximate the stress tensor by Ê ∂v ∂v ˆ t ij = .Í p + Ë m . in the interior of shock waves. including most compressible flows where the fluid's density is changing. divided by the particle's instantaneous volume. where they occur over a molecular scale) or over very short time scales (e. the effect on the flow of the term which involves — ⋅ v and the expansion viscosity is usually very small even in compressible flows.20 Thermodynamic second-law arguments show that l must be positive. and for complex liquids like benzene it can be over 100 times v larger.pd ij + m Á i + j ˜ Ë ∂x j ∂xi ¯ or (42) t xx = .l ˆ — ⋅ v ˙d ij + m Á i + j ˜ . — ⋅ v > 0). — ⋅ v < 0).g.p + 2m ∂vx ∂x ∂v y ∂y ∂vz ∂z (43) t yy = . Written in terms of the thermodynamic pressure p. therefore. in high-intensity ultrasound) will the v term involving — ⋅ v be large enough to have a noticeable effect on the equation of motion.p + 2 m t zz = . Nevertheless. In other words. For most flows. the thermodynamic pressure always tends to "lag behind” the mechanical pressure when a change is occurring. attempts to study the expansion viscosity are hampered by the difficulty of devising experiments where its effect is significant enough to be accurately measured. The difference depends.g. on both the rate of v expansion (— ⋅ v ) and the molecular composition of the fluid (via l: see below).p + 2m . ¯ ˚ Î 3 Ë ∂x j ∂xi ¯ (41) v The term — ⋅ v is associated with the dilation of the fluid particles.

pp 469-470. An Introduction to Fluid Mechanics. Vol. Batchelor. II. Ë ∂z ∂y ¯ v Equations (42) and (43) are rigorously valid in the limit of incompressible flow (— ⋅ v ª 0 ). P. some experimental values can be found for example in the paper by L. pp. (40). Fluid Mechanics. 1949). 153-156. is suspect. Vol. for experiments have shown that the assumed linear relation between the mechanical and thermodynamic pressures. the Newtonian assumption of linearity between the shear stresses and rates of shear strain is very accurately obeyed in a large class of fluids under wide ranges of flow conditions. 73-74. (41). on — ⋅ v and thus on the particular flow field. Lifshitz. when it is large enough to be measured accurately. this equation can be written in vector notation as . For the expansion viscosity in gases. Landau and E. Razier. By contrast. B. or L.e.21 Ê ∂v ∂v ˆ t xy = t yx = m Á x + y Ë ∂y ∂x ¯ t xz = t zx = m Ê ∂v x ∂vz ˆ + Ë ∂z ∂x ¯ (43) cont’d Ê ∂v ∂v ˆ t yz = t zy = m Á y + z . O. That the term which involves l is usually negligible is fortunate. B. Lieberman. K. pp 1415-1422. Physical Review. pp. The theory of the expansion viscosity is discussed in J. F. (20): r Dvi ∂ =Dt ∂xi È Ê2 ˆ v˘ ∂ Í p + Ë 3 m . v i. is inserted into the general equation of motion. Zeldovich and Y. B. as are most liquids with relatively simple molecular structure. Zel'dovich and Y. often turns out to be not a fluid property but dependent on the rate of expansion. I.l ¯ — ⋅ v ˙ + ∂x ˚ Î j È Ê ∂v j ∂vi ˆ ˘ Á Í m Ë ∂x + ∂x ˜ ˙ + rGi i j¯˚ Î (44) For constant m and l. Bird's Molecular Theory of Gases and Liquids. Y. The value of l . Physics of Shock Waves and High-Temperature Hydrodynamic Phenomena. P. M. Hirschelder. All gases at normal conditions are Newtonian. 304-309. Raizer's Physics of Shock Waves and High-Temperature Hydrodynamic Phenomena. Curtiss and R. For further discussion of the expansion viscosity. Vol. C. 75. D. 6 The Navier-Stokes Equation The Navier-Stokes equation is the equation which results when the Newtonian stress tensor. pp. see also the editorial footnote by Hayes and Probstein in Y. N. see for example G.

∂x j and one obtains from (44) or (45) (47) r ∂ 2 vi Dvi ∂p =+m + rGi . and independently by Siméon-Denis Poisson in 1831. The Navier-Stokes equation of motion was derived by Claude-Louis-Marie Navier in 1827.+ mÁ 2 + 2 + 2 ˜ + rGy Ë ∂t Ë ∂x ∂x ∂y ∂z ¯ ∂y ∂y ∂z ¯ Ê ∂v Ê ∂ 2v ∂ 2v ∂ 2v ˆ ∂v ∂v ∂v ˆ ∂p r Á z + v x z + v y z + vz z = .+ mÁ 2z + 2z + 2z + rGz Ë ∂t Ë ∂x ∂x ∂y ∂z ¯ ∂z ∂y ∂z ¯ (50a) (50b) (50c) Appendix A gives the equations in cylindrical coordinates. (48) reads Ê ∂v Ê ∂ 2 v ∂ 2 vx ∂ 2 vx ˆ ∂v ∂v ∂v ˆ ∂p r Á x + vx x + v y x + vz x = . Dt (49) As mentioned above. ∂v j v = —⋅ v = 0 . operating in (45) on the vector v .+ m Á 2x + + 2 + rGx Ë ∂t Ë ∂x ∂x ∂y ∂z ¯ ∂x ∂y 2 ∂z ¯ Ê ∂ 2 v y ∂ 2 vy ∂ 2 vy ˆ Ê ∂v y ∂vy ∂vy ∂vy ˆ ∂p Á r + vx + vy + vz = . just like D/Dt on the left side is the v well-known scalar operator that operates on v . For incompressible flows with constant viscosity. Dt ∂xi ∂x j ∂x j (48) or. (48) or (49) are in many cases a very good approximation even when the flow is compressible. v v Dv v r = -—p + m—2 v + r G .22 v v Dv Ê1 v v r = -—p + Ë m + l ˆ —(— ⋅ v) + m—2 v + rG ¯ 3 Dt where —2 = (45) ∂ ∂ ∂ 2 + 2 + ∂x ∂y ∂z2 (46) v is a scalar operator. Their motivations of the stress tensor were based on what amounts to a molecular view of how stresses are exerted by one . Written out fully in Cartesian coordinates. in vector form.

together with the mass conservation equation plus whatever other equations are required to form a complete set. A solution yields the velocity components and pressure at the boundaries. rate-of-strain argument. . The proof for the continuity of t ij is essentially the same as the one for equation (4). including across phase interfaces like the boundaries between the fluid and a solid and between two immiscible fluids. the boundary conditions consistent with the continuum hypothesis are that (a) the velocity components and (b) the stress tensor components must be everywhere continuous. 1. Barré de Saint Venant (in 1843) and George Gabriel Stokes (in 1845) derived the equation starting with the linear stress vs. from which one obtains the stress tensor components via equation (42) [or (43)] and the stress vector from (11). In the absence of surface tension. with the requirement that the v equation of motion must be satisfied at every point for any orientation n of the surface.23 fluid particle against another. Surface tension gives rise to a discontinuity in the normal stress at the interface between two immiscible fluids. similar to the disc depicted in Fig. That this must be so can be proved by applying mass conservation and the equation of motion to a small disc-shaped control volume at a point in space. with the boundary conditions appropriate to the particular problem at hand. Boundary conditions A particular flow problem may in principle be solved by integrating the Navier-Stokes equation. and considering the limit where the thickness of the disc go to zero. Later.

for incompressible flow Fig. its stress tensor.2+ 2 r ∂q 2 r 2 ∂q ∂z ˚ r ∂q Î r ∂r Ë ∂ r ¯ r Ê ∂vz ∂v v ∂v ∂v + vr z + q z + v z z ˆ = Ë ∂t ∂r r ∂q ∂z ¯ ∂p È 1 ∂ Ê ∂vz ˆ 1 ∂ 2 vz ∂ 2 v z ˘ + rGz + mÍ r + 2 2 + ∂z ∂z 2 ˙ Î r ∂r Ë ∂r ¯ r ∂q ˚ (A. A. z). q .3) .1: Cylindrical coordinate system Navier-Stokes equation of motion Ê ∂v v ∂v v 2 ∂v ∂v ˆ rÁ r + v r r + q r .24 Appendix A The Navier-Stokes equation.2) r (A.q + v z r ˜ = Ë ∂t ∂r r ∂q r ∂z ¯ ∂p È 1 ∂ Ê ∂vr ˆ vr 1 ∂ 2 vr 2 ∂vq ∂ 2 vr ˘ + mÍ r + 2 ˙ + rGr .1) r v ∂v v v Ê ∂vq ∂v ∂v + vr q + q q + r q + vz q ˆ = Ë ∂t ∂r r ∂q r ∂z ¯ 1 ∂ 2 vq 2 ∂vr ∂ 2 vq ˘ 1 ∂p È 1 ∂ Ê ∂vq ˆ vq + mÍ r + + 2 ˙ + rGq .2+ 2 ∂r r ∂q 2 r2 ∂q ∂z ˚ Î r ∂r Ë ∂ r ¯ r (A. and the mass conservation equation in cylindrical coordinates (r.

p + 2 m t zz = .25 Stress tensor components (cylindrical coordinates) t rr = .p + 2m ∂vz ∂z (A.p + 2m ∂v r ∂r Ê 1 ∂vq vr ˆ + Ë r ∂q r ¯ t qq = .4) 1 ∂vr ˘ È ∂ Êv t rq = t qr = m Í r Ë q ˆ + ˚ Î ∂r r ¯ r ∂q ˙ t qz = t zq = m Ê ∂vq 1 ∂vz ˆ + Ë ∂z r ∂q ¯ t rz = t rz = m Ê ∂vr ∂vz ˆ + Ë ∂z ∂r ¯ Mass conservation equation ∂r 1 ∂ ( rrvr ) + 1 ∂ (rvq ) + ∂ ( rvz ) = 0 + r ∂q ∂z ∂t r ∂r (A.5) .

18x103 Glycerin 1.49 0.39x103 Mercury 1.00x10-5 * Specific heat at constant pressure cp (J/kgK) 5.40x10-10 1.92x10-5 Air 1.164* 1.286 5.597 1.00x103 1.36x104 1.0262 3.19* 1.150 Coefficient of thermal expansion b (K-1) 3.8x10-4 4.6x10-10 4.41x10-3 * Isothermal Compressibility kT (m2/N) 1.00x10-5 * 1.98x10-5 0.39x102 * Calculated from ideal gas relationships.21x103 * Helium 0.69 1.7x10-10 2.00x103 * Water 1.0x10-4 3.55x10-3 8.26 Appendix B Properties of selected fluids at 20o C=293K and 1bar=105 N/m-2 Fluid Density r (kg/m3) Viscosity m (kg/m s) Thermal conductivity k (W/m K) 0.41x10-3 * 1.00x10-3 0.26x103 (C3H80 3) 1.82x10-4 0. .