Differential Equations 4: Phy310 - Mathematical Methods For Physicists I
Differential Equations 4: Phy310 - Mathematical Methods For Physicists I
LECTURE 27
DIFFERENTIAL EQUATIONS 4
This equation is homogeneous since each term contains y(x) or a derivative. It is linear because
each y, y 0 , or y 00 appears as the first power – and no products.
Equation (1) can have two linearly independent solutions. Let us find (at least) one solution
of Eq. (1) using a generalized power series. Using the first solution we can develop the second
independent solution. We will also later prove that a third independent solution does not exist.
Let us write the most general solution of Eq. (1) as
In some cases we can have a source term as well in the ODE, leading to a nonhomogeneous,
linear, second-order ODE,
y 00 + p(x)y 0 + q(x)y = r(x). (3)
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
The function r(x), represents a source (such as electrostatic charge) or a driving force (as in a
driven oscillator).
Calling this solution yp , we may add to it any solution of the corresponding homogeneous
equation, Eq. (1). Hence the most general solution of Eq. (3) is
We have to fix the constants c1 and c2 and that will be done by applying boundary conditions.
At the moment let us assume that r(x) = 0 and that our differential equation is homogeneous.
We will attempt to develop a solution of our linear, second-order, homogeneous differential equation,
Eq. (1), by substituting in a power series with undetermined coefficients.
This generalized power series has a parameter, which is the power of the lowest nonvanishing
term of the series.
As a testbed let us apply this method to an important differential equation, the linear (classical)
oscillator equation
y 00 + w2 y = 0. (5)
y = c1 y1 (x) + c2 y2 (x)
= c1 sin ωx + c2 cos ωx. (6)
y(x) = xk a0 + a1 x + a2 x2 + a3 x3 + · · ·
X∞
= aλ xk+λ , a0 6= 0, (7)
λ=0
2/7
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
Let us substitute the series forms of y(x) and y 00 (x) into Eq. (6). We get
∞
X ∞
X
aλ (k + λ)(k + λ − 1)xk+λ−2 + ω 2 aλ xk+λ = 0. (8)
λ=0 λ=0
The uniqueness of power series tells us that, the coefficients of each power of x on the left-hand
side of Eq. (8) must vanish individually. We have
a0 k(k − 1)xk−2 + a1 k(k + 1)xk−1 + a2 (k + 1)(k + 2)xk + a3 (k + 2)(k + 3)xk+1 + · · ·
+aλ (k + λ)(k + λ − 1)xk+λ−2 + · · · + ω 2 a0 xk + ω 2 a1 xk+1
+ω 2 a3 xk+3 + · · · + ω 2 aλ xk+λ + · · ·
= 0. (9)
The lowest power of x appearing in Eq. (10) is xk−2 , for λ = 0. The requirement that the
coefficient vanish yields
a0 k(k − 1) = 0. (11)
We had chosen a0 as the coefficient of the lowest nonvanishing term of the series, Eq. (7), hence,
by definition, a0 6= 0. Therefore we have the constraint
k(k − 1) = 0. (12)
This equation, coming from the coefficient of the lowest power of x, is called the indicial equation.
The indicial equation and its roots (or indices of the regular singular point of the ODE) play a
crucial role in our attempt to find the solutions.
We have two choices for k, k = 0 or k = 1. We see that a1 is arbitrary if k = 0 and necessarily
zero if k = 1. Thus we will set a1 equal to zero.
Case k = 0
We have the general term in the equation
3/7
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
Since a0 6= 0 we have
a2 · 2 · 1 + ω 2 a0 = 0,
a3 · 3 · 2 + ω 2 a1 = 0,
a4 · 4 · 3 + ω 2 a2 = 0,
a5 · 5 · 4 + ω 2 a3 = 0,
..
.
aj+2 (j + 2)(j + 1) + ω 2 aj = 0.
a2 · 2 · 1 + ω 2 a0 = 0,
a4 · 4 · 3 + ω 2 a2 = 0,
..
.
aj+2 (j + 2)(j + 1) + ω 2 aj = 0.
ω2
aj+2 = − aj . (14)
(j + 1)(j + 2)
Case k = 1
We have the general term in the equation
Since a0 6= 0 we have
a2 · 2 · 3 + ω 2 a0 = 0,
a3 · 3 · 4 + ω 2 a1 = 0,
a4 · 4 · 5 + ω 2 a2 = 0,
a5 · 5 · 6 + ω 2 a3 = 0,
..
.
aj+2 (j + 2)(j + 3) + ω 2 aj = 0. (16)
4/7
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
a2 · 2 · 3 + ω 2 a0 = 0,
a4 · 4 · 5 + ω 2 a2 = 0,
..
.
aj+2 (j + 2)(j + 3) + ω 2 aj = 0. (17)
ω2
aj+2 = − aj . (18)
(j + 2)(j + 3)
For k = 0 we have
ω 2n
a2n = (−1)n a0 , (19)
(2n!)
and our solution is
For k = 1 we have
ω 2n
a2n = (−1)n a0 . (21)
(2n + 1)!
We obtain
Thus we have arrived at two (independent) series solutions of the linear oscillator equation using
the method of generalized series substitution (Frobenius’ method).
Equation (7) is an expansion about the origin, x0 = 0. It is also possible to replace Eq. (7)
with
∞
X
y(x) = aλ (x − x0 )k+λ , a0 6= 0. (23)
λ=0
The choice x0 = 1 has some advantages when we attempt to find the series solution for Legendre,
Chebyshev, and hypergeometric equations.
Note that the point x0 should not be chosen at an essential singularity. Frobenius’ method will
probably fail there. The series solution about x0 an ordinary point or regular singular point will
5/7
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
be valid where it converges. A divergence of some sort can be expected when |x − x0 | = |zs − x0 |,
where zs is the closest singularity to x0 (in the complex plane).
Symmetry of solutions
Suppose we got one solution with even symmetry
This is not just an accident but a direct consequence of the form of the ODE. Writing a general
ODE as
L(x)y(x) = 0, (26)
in which L(x) is a differential operator, we see that for the linear oscillator equation, L(x) is even
under parity. That is,
L(x) = L(−x). (27)
This is telling us that if y(x) is a solution of the differential equation, then y(−x) is also a
solution. Then, we can resolve any solution into even and odd parts
1 1
y(x) = [y(x) + y(−x)] + [y(x) − y(−x)] , (28)
2 2
where the first bracket on the right gives an even solution, the second an odd solution.
We can see that Legendre, Bessel, and Hermite equations (or differential operators) all exhibit
this even parity. Solutions of all of them may be expressed as series of even powers of x and separate
series of odd powers of x.
The Laguerre differential operator has neither even nor odd symmetry. Hence its solutions
cannot be expected to exhibit even or odd parity.
6/7
PHY310 - MATHEMATICAL METHODS FOR PHYSICISTS I Odd Term 2019
Fuchs’ Theorem
The answer to the question on when Frobenius’ method can be expected to work is given by Fuchs’
theorem. It states that we can always obtain at least one power-series solution, provided we are
expanding about a point that is an ordinary point or at worst a regular singular point.
Note that if xp(x) and x2 q(x) can be expanded as convergent power series, ∞ n
P
n=0 an x , we say
that the differential equation is regular (or has a non-essential singularity) at the origin. If we
attempt an expansion about an irregular or essential singularity, Frobenius’ method may fail.
Fortunately, the more important equations of mathematical physics (Legendre, Bessel, La-
guerre), have no irregular singularities in the finite plane.
In summary, if we are expanding about an ordinary point or at worst about a regular singularity,
the series substitution approach will yield at least one solution (Fuchs’ theorem). Whether we get
one or two distinct solutions depends on the roots of the indicial equation.
Let us note the following points.
i. If the two roots of the indicial equation are equal, we can obtain only one solution by this
series substitution method.
ii. If the two roots differ by a nonintegral number, two independent solutions may be obtained.
iii. If the two roots differ by an integer, the larger of the two will yield a solution. The smaller
may or may not give a solution, depending on the behavior of the coefficients.
7/7