llcotp.com/LL1/12-determination-of-the-potential-energy-from-the-period-of-oscillation.md
2022-12-17 23:41:00 +01:00

57 lines
2.4 KiB
Markdown

---
title: Determination of the potential energy from the period of oscillation
---
Let us consider to what extent the form of the potential energy U(x) of a
field in which a particle is oscillating can be deduced from a knowledge of the
period of oscillation T as a function of the energy E. Mathematically, this
involves the solution of the integral equation (11.5), in which U(x) is regarded
as unknown and T(E) as known.
We shall assume that the required function U(x) has only one minimum
in the region of space considered, leaving aside the question whether there
exist solutions of the integral equation which do not meet this condition.
For convenience, we take the origin at the position of minimum potential
energy, and take this minimum energy to be zero (Fig. 7).
U
U=E
121U
X
X1
X2
FIG. 7
28
Integration of the Equations of Motion
§12
In the integral (11.5) we regard the co-ordinate x as a function of U. The
function x(U) is two-valued: each value of the potential energy corresponds
to two different values of X. Accordingly, the integral (11.5) must be divided
into two parts before replacing dx by (dx/dU) dU: one from x = X1 to x = 0
and the other from x = 0 to X = X2. We shall write the function x(U) in
these two ranges as x = x1(U) and x = x2(U) respectively.
The limits of integration with respect to U are evidently E and 0, so that
we have
T(E) =
=
If both sides of this equation are divided by V(a - E), where a is a parameter,
and integrated with respect to E from 0 to a, the result is
dUdE
or, changing the order of integration,
dE
The integral over E is elementary; its value is TT. The integral over U is
thus trivial, and we have
since x2(0) = x1(0) = 0. Writing U in place of a, we obtain the final result:
(12.1)
Thus the known function T(E) can be used to determine the difference
x2(U)-x1(U). - The functions x2(U) and x1(U) themselves remain indeter-
minate. This means that there is not one but an infinity of curves U = U(x)
§13
The reduced mass
29
which give the prescribed dependence of period on energy, and differ in such
a way that the difference between the two values of x corresponding to each
value of U is the same for every curve.
The indeterminacy of the solution is removed if we impose the condition
that the curve U = U(x) must be symmetrical about the U-axis, i.e. that
x2(U) = 1(U) III x(U). In this case, formula (12.1) gives for x(U) the
unique expression
(12.2)