Solving The Wave Equation

Back to Contents

    In physics, following the example of James Clerk Maxwell, we can combine the third and fourth of Maxwells Equations under the conditions that occur in vacuum and get an equation that has as its solution a description of an electric or a magnetic field. In the case of an electric field that equation has it that

(Eqn 1)

We now want to tease a proper algebraic description of E out of that equation.

    Begin with Euclids first common notion: things that stand equal to the same thing also stand equal to each other. We can say that in reverse: things that stand equal to each other also stand equal, both of them, to some other thing. We assume that neither of the terms in Equation 1 equals a constant, unvarying zero (we dont want to consider uniform, stationary electric fields), so that some other thing also does not equal a constant, unvarying zero. The terms in Equation 1 involve differentiating E with respect to mutually exclusive coordinates, so that some other thing does not involve either of those coordinates. And both terms involve the first power of E, so that some other thing must also contain the first power of E. Thus we infer that each of the terms in Equation 1 stands equal to E multiplied by some constant;

(Eqn 2)

and

(Eqn 3)

In those equations k represents a constant.

    In those equations I have tacitly assumed that the second term opposes the first term; that is, when the second derivative grows larger the second term grows larger in the negative sense and when the second derivative diminishes the second term diminishes in the negative sense. The k2 term thus acts as a negative feedback on the derivative. That mathematical fact ensures that E remains finite and bounded, which satisfies the requirement that the energy manifested in the combined electric and magnetic fields remain conserved.

    To simplify our analysis lets assume that E varies only with the elapse of time and in the x-direction. Under that assumption Equation 2 becomes

(Eqn 4)

We can now "factor" the operator that acts on E, so Equation 4 becomes

(Eqn 5)

If we apply the operator ∂x-ik to that equation, we get

(Eqn 6)

which tells us, when we compare it with Equation 4, that E=(∂x-ik)E represents a solution of the wave equation. And that fact means that we can rewrite Equation 5 as

(Eqn 7)

which we can solve readily for E to get

(Eqn 8)

in which E0, the amplitude of the solution, represents the antilogarithm of the constant of integration.

    If we use Equation 8 to substitute into the definition of E, we can determine the algebraic form of E, which we get as

(Eqn 9)

In the realm of mathematics that stands as a legitimate solution of Equation 2, but in the realm of physics it represents a forcefield that increases or decreases without bound, which represents a violation of the conservation of energy theorem. Thus, in physics we dismiss Equation 9 as a legitimate description of the electric field and use only Equation 8 with both positive and negative arguments.

    We can use the same technique to solve Equation 3, getting

(Eqn 10)

In this case we can delete the negative-signed argument from the solution on the basis of the statement that time does not elapse backward. Again we get an expression analogous to what we see in Equation 9 and again we dismiss it from the realm of physics.

    Equation 1 tells us that applying the second derivative with respect to distance to the proper description of the electric field yields the same result as does applying the second derivative with respect to time and dividing by the speed of light. In order for that statement to stand true to mathematical reality in all circumstances we must have

(Eqn 11)

that is, our description of the electric field must consist of two factors, each of which encodes only one of the variables in Equation 1. In addition, Equation 1 tells us, because the differential operators have the same form, that S(x) and T(t) must both have the same algebraic form.

    If we hold the value of x constant and let the value of t increase, Equation 11 will become a constant fractional multiple of Equation 10 and if we hold the value of t constant and let the value of x vary, Equation 11 will become a constant fractional multiple of Equation 8. We thus infer that

(Eqn 12)

For convenience I have removed the prime from E: since we have dismissed the expression in Equation 9 as a solution in physically real situations, we can do this without causing confusion. Having the product of two exponential functions, we can see that if we hold the value of x constant, its exponential provides the fraction to multiply Equation 10 and that if we hold the value of t constant, its exponential provides the fraction to multiply Equation 8. Indeed, strictly speaking, I should have shown the constant fractions explicitly in Equations 8 and 10 as unknowns.

    That solution describes a sinusoidal pattern, both in space and in time. If we hold the value of t constant, then as the value of x changes the value of E rises and falls in a wave-like pattern. On the other hand, if we hold the value of x constant and let the value of t increase, then the value of E vibrates like the motion of an harmonic oscillator. If the values of both variables change, then the wave pattern moves: the equation

(Eqn 13)

describes a given value of E moving in either the positive or negative x-direction at the speed c. Because its solution describes a wave-like phenomenon, we call Equation 1 the wave equation.

    But Equation 1 has applicability beyond describing electric fields propagating through vacuum. It also describes magnetic fields propagating through vacuum: indeed it must do so because Maxwells Equations necessitate that magnetic fields accompany electric fields that do not emanate directly from charged bodies. The wave equation also describes vibrations propagating through uniform elastic media; that is, it describes sound. Any phenomenon that we can describe in analogy with sound we can describe with the appropriate form of the wave equation.

habg

Back to Contents