Vector Differentiation in Physics: Vector-Valued Functions, Moving Bases, and Kinematics

Vector differentiation is the study of how a vector changes with respect to a scalar parameter such as time, arc length, or an angle. It is one of the basic mathematical operations of mechanics because position, velocity, acceleration, momentum, angular momentum, force, and many other physical quantities are vectors whose magnitudes and directions may both change. A mathematically careful treatment also explains why differentiating vectors in polar, cylindrical, spherical, or rotating coordinates requires more than differentiating their scalar components [1, 2, 3].

This article focuses on derivatives of vector-valued functionshttps://physicslibrary.org/encyclopedia/Bijective.html and on their use in mechanics. Gradient, divergence, curl, and related differential operators acting on scalar and vector fields belong more naturally to vector calculus, although the final sections show how the two subjects connect.

1 A vector-valued function

Let a vector depend on a scalar parameter t:

(1)

For motion in three-dimensional Euclidean space one may write

(2)

where e_x,e_y,e_z are fixed Cartesian basis vectors.

The derivative is defined by the same limit used for scalar functions, except that the difference quotient is now a vector:

(3)

provided this vector limit exists.

Equivalently, A is differentiable at t if there exists a vector B such that

(4)

with

(5)

Then

(6)

This form makes the local-linear meaning of the derivative explicit: over a sufficiently short interval, the change in the vector is the derivative multiplied by the parameter increment, plus a remainder that becomes negligible faster than Δt.

2 Componentwise differentiation in a fixed basis

If the basis vectors are constant in time,

(7)

Therefore

(8)

Thus a vector-valued function in a fixed Cartesian basis is differentiable exactly when its component functions are differentiable, and its derivative is obtained component by component.

This simple rule is often the first rule learned for vector differentiation. It is correct only because the Cartesian basis is fixed. Later we will see that a moving basis contributes its own derivatives.

3 Geometric meaning: the derivative is tangent to a curve

A vector-valued function

(9)

may represent a curve in physical space. For two nearby parameter values, the difference

(10)

is a chord joining two nearby points of the curve. The difference quotient

(11)

points along that chord. As Δt → 0, the chord direction approaches the tangent direction to the curve.

For a particle position r(t),

(12)

is therefore tangent to the path. The acceleration is the derivative of the velocity,

(13)

Acceleration need not point along the path because it measures the change of the entire velocity vector, including both its magnitude and its direction.

4 Differentiation with respect to another parameter

A curve can be parameterized by a quantity other than time. Suppose

(14)

and s = s(t). The chain rule gives

(15)

If s is arc length, then

(16)

so the vector

(17)

is the unit tangent to the path. Since

(18)

where v = |v| is speed,

(19)

This decomposition will be useful when differentiating the velocity.

5 Product rules for vectors

The familiar scalar product rule extends naturally to vector products. Let f(t) be a scalar and let A(t) and B(t) be differentiable vectors.

For scalar multiplication,

(20)

For the dot product,

(21)

For the cross product,

(22)

The order in the cross-product rule matters because

(23)

These identities can be proved directly from the limit definition using bilinearity of the dot and cross products.

6 Derivative of a vector magnitude

Let

(24)

with A≠0. Since

(25)

differentiation gives

(26)

Therefore

(27)

If

(28)

then

(29)

Only the part of dA∕dt parallel to A changes the magnitude. A perpendicular derivative changes the direction without changing the magnitude to first order.

A useful mechanics example is

(30)

provided v≠0.

7 Derivative of a unit vector

Let

(31)

be a unit vector. Differentiating

(32)

gives

(33)

Hence

(34)

The derivative of a unit vector is therefore perpendicular to the unit vector itself.

A more explicit formula follows from e = A∕A:

(35)

The bracket removes the component of dA∕dt parallel to A. What remains is the perpendicular part responsible for turning the vector.

This result is fundamental in orbital mechanics, rigid-body kinematics, moving coordinate systems, and Differential Geometry.

8 Velocity and acceleration along a curved path

Write the velocity as

(36)

where e_t is the unit tangent. Differentiation gives

(37)

The derivative of e_t must be perpendicular to e_t. For a smooth plane curve, define the radius of curvature ρ and principal normal e_n by

(38)

Using ds∕dt = v,

(39)

Therefore

(40)

Thus

(41)

changes the speed, while

(42)

changes the direction of motion. Uniform circular motion has dv∕dt = 0 but still has nonzero acceleration because e_t continuously changes direction.

9 A physics example: differentiating angular momentum

For a particle with position r and momentum p,

(43)

Using the cross-product differentiation rule,

(44)

For constant mass,

(45)

so

(46)

Newton’s second law gives

(47)

Therefore

(48)

Since torque is defined by

(49)

we obtain

(50)

This derivation shows how a basic vector product rule becomes one of the central dynamical relations of mechanics.

10 The crucial complication: a moving basis

Suppose a vector is written in a basis whose directions themselves depend on time:

(51)

Now the product rule gives

(52)

The last three terms vanish only when the basis vectors are fixed.

This is the source of many additional terms in polar, cylindrical, spherical, and rotating-coordinate equations of motion. They are not artificial corrections; they arise because the coordinate directions themselves change.

11 Polar-coordinate basis derivatives

In the plane, define

(53)

and

(54)

Differentiating with respect to 𝜃 gives

(55)

(56)

The chain rule then gives

(57)

(58)

The position vector is

(59)

Differentiating,

(60)

Differentiating again,

(61)

The term

(62)

is the inward centripetal contribution, while

(63)

is produced by simultaneous radial motion and rotation of the basis.

These terms follow entirely from differentiating the basis vectors correctly.

12 Rotating frames and the transport theorem

The polar-coordinate result is a special case of a more general rule. Let a rotating basis e′₁,e′₂,e′₃ have angular velocity ω relative to an inertial frame. Each basis vector satisfies

(64)

For

(65)

differentiation gives

(66)

This is the vector transport theorem.

Applying the transport theorem to a particle position relative to a moving origin gives the inertial velocity

(67)

Differentiating once more gives the inertial acceleration

(68)

The additional terms are, respectively, the Coriolis, Euler, and centrifugal contributions. Their origin is vector differentiation in a rotating basis.

13 Vector fields and the derivative following a particle

The previous sections considered a vector that depends directly on one scalar parameter. Physics also uses vector fields such as velocity, electric field, or magnetic field:

(69)

A particle following a trajectory r(t) samples different points of the field. The ordinary chain rule applied componentwise gives

(70)

In vector notation this is

(71)

In continuum mechanics and fluid mechanics this is called the material or substantial derivative. The first term measures explicit change of the field at a fixed point, while the second measures change caused by motion through a spatially nonuniform field.

This formula is also a useful bridge to vector calculus. The derivative of a vector-valued trajectory and the spatial differentiation of a vector field are related but distinct ideas.

14 Differentials and infinitesimal changes

If A(t) is differentiable, its differential is

(72)

This equation is best interpreted as the linear part of the change in A associated with a small change dt. More precisely,

(73)

where

(74)

This interpretation prevents a common confusion: dA is not an arbitrary tiny vector inserted by notation; it is the linear map generated by the derivative acting on the small scalar increment dt.

15 Common errors

Several mistakes recur in mechanics problems involving vector differentiation.

• Differentiating only components in a moving basis. In polar, spherical, or rotating coordinates the basis vectors also have derivatives.
• Assuming zero speed derivative means zero acceleration. Constant speed does not imply constant velocity. Uniform circular motion is the standard counterexample.
• Treating the derivative of a unit vector as parallel to the vector. Since the magnitude is fixed, the derivative of a unit vector is perpendicular to it.
• Reversing factors in a differentiated cross product. The correct order is preserved:

  (75)

• Confusing vector differentiation with vector differential operators. dA∕dt differentiates a vector-valued function with respect to a scalar. Quantities such as ∇⋅ A and ∇× A describe spatial derivatives of a vector field.

16 Worked example: a planar spiral

Consider

(76)

(77)

where u and Ω are constants. Then

(78)

(79)

(80)

(81)

The polar-coordinate velocity is

(82)

The acceleration is

(83)

Even though both scalar rates u and Ω are constant, the acceleration is not zero because the polar basis rotates and because the tangential speed rΩ changes as the radius changes.

This is a compact demonstration of why vector differentiation is more than differentiating scalar coefficients.

17 Relation to classical mechanics

Vector differentiation underlies many of the standard equations of classical mechanics:

(84)

(85)

(86)

(87)

It also supplies the kinematic machinery behind curvilinear coordinates, orbital frames, rotating rigid bodies, Coriolis effects, and non-inertial reference frames. For this reason physics texts often introduce vector differentiation early in mechanics, before moving on to full equations of motion [1, 2].

Summary

The derivative of a vector-valued function is defined by the vector limit

(88)

In a fixed Cartesian basis it is obtained by differentiating components. In a moving basis, the basis vectors must also be differentiated. Product rules apply to scalar multiplication, dot products, and cross products. The derivative of a unit vector is perpendicular to that unit vector, which leads naturally to tangential-normal acceleration and to the extra terms appearing in polar and rotating coordinates.

The central lesson for mechanics is simple:

A vector can change because its components change, because its direction changes, or because the basis used to describe it changes.

A correct derivative must account for all three possibilities.

References

References