Differential Operator
Differential Operator
The expression in the second parenthesis of Eq. (2.22) is the spatial differential operator ∇
(nabla, del), which has a vector character. In the cartesian coordinate system, the operator
nabla ∇ is defined as:
(2.23)
Using the above differential operator, the change of the quantity Q is written as:
(2.24)
The increment dQ of Eq. (2.24) is obtained either by applying the product dx · ∇, or by taking
the dot product of the vector dx and ∇Q. If Q is a scalar quantity, then ∇Q is a vector or a
first-order tensor with definite components. In this case, ∇Q is called the gradient of the
scalar field. Equation (2.24) indicates that the spatial change of the quantity Q assumes a
maximum if the vector ∇Q (gradient of Q) is parallel to the vector dx. If the vector ∇Q is
perpendicular to the vector dx, their product will be zero. This is only possible if the spatial
change dx occurs on a surface with Q = const. Consequently, the quantity Q does not
experience any changes. The physical interpretation of this statement is found in Fig. 2.3. The
scalar field is represented by the point function temperature that changes from the surface T
to the surface T + dT. In Fig. 2.3, the gradient of the temperature field is shown as ∇T, which
is perpendicular to the surface T = const. at point P. The temperature probe located at P
moves on the surface T = const. to the point M, thus measuring no changes in temperature (α
= π/2, cosα = 0). However, the same probe experiences a certain change in temperature by
moving to the point Q, which is characterized by a higher temperature T + dT (0 < α < π/2).
The change dT can immediately be measured if the probe is moved parallel to the vector ∇ T.
In this case, the displacement dx (see Fig. 2.3) is the shortest (α = 0, cosα = 1). Performing
the similar operation for a vector quantity as seen in Eq. (2.21) yields:
FIGURE 2.3 Physical explanation of the gradient of scalar field.
(2.25)
The right-hand side of Eq. (2.25) is identical with:
(2.26)
In Eq. (2.26) the product dx · ∇ can be considered as an operator that is applied to the vector
V, resulting in an increment of the velocity vector. Performing the scalar multiplication
between dx and ∇ gives:
(2.27)
with ∇V as the gradient of the vector field, which is a second-order tensor. To perform the
differential operation, first the ∇ operator is applied to the vector V, resulting in a second-
order tensor. This tensor is then multiplied with the vector dx in a scalar manner that results
in a first-order tensor or a vector. From this operation, it follows that spatial change of the
velocity component can be expressed as the scalar product of the vector dx and the second-
order tensor ∇V, which represents the spatial gradient of the velocity vector. Using the spatial
derivative from Eq. (2.27), the substantial change of the velocity is obtained by:
(2.28)
where the spatial change of the velocity is expressed as:
(2.29)
Dividing Eq. (2.29) by dt yields the convective part of the acceleration vector:
(2.30)
The substantial acceleration is then:
(2.31)
The differential dt may symbolically be replaced by Dt indicating the material character of
the derivatives. Applying the index notation to velocity vector and nabla operator, performing
the vector operation, and using the Kronecker delta, the index notation of the material
acceleration A is:
(2.32)
Equation (2.32) is valid only for the cartesian coordinate system, where the unit vectors do
not depend upon the coordinates and are constant. Thus, their derivatives with respect to the
coordinates disappear identically. To arrive at Eq. (2.32) with a unified index i, we renamed
the indices. To decompose the above acceleration vector into three components, we cancel
the unit vector from both sides in Eq. (2.32) and get:
(2.33)
To find the components in xi direction, the index i assumes subsequently the values from 1 to
3, while the summation convention is applied to the free index j. As a result we obtain the
three components: