01 / 21

Frobenius Method
Regular Singular Points

Learn how to solve differential equations at points where standard power series fail: the Frobenius method extends your toolkit to handle singular points.

Differential Equations

What is a Singular Point?

Consider a second-order linear ODE in standard form:

A point \( x = x_0 \) is called an ordinary point if both \( P(x) \) and \( Q(x) \) are analytic (have convergent Taylor series) at \( x_0 \). Otherwise, \( x_0 \) is a singular point.

Regular vs Irregular Singular Points

Not all singular points are created equal. A singular point \( x_0 \) is regular if both:

Condition 1
\( (x - x_0)\,P(x) \) is analytic at \( x_0 \)
Condition 2
\( (x - x_0)^2\,Q(x) \) is analytic at \( x_0 \)

If either condition fails, the singular point is irregular. The Frobenius method applies only at regular singular points.

Classifying Points: A Visual Guide

Ordinary Point
\( P(x), Q(x) \) analytic
Use power series Singular Point
\( P(x) \) or \( Q(x) \)
not analytic Regular Singular
\( (x-x_0)P, (x-x_0)^2 Q \) analytic
Use Frobenius method Irregular Singular Conditions fail Advanced methods needed

The Indicial Equation

At a regular singular point \( x_0 \), we assume a Frobenius series:

Here \( r \) is an unknown exponent (possibly non-integer or negative). Substituting into the ODE and collecting the lowest-power terms yields the indicial equation:

The two roots \( r_1, r_2 \) of this quadratic (with \( r_1 \geq r_2 \)) determine the form of the two linearly independent solutions.

Three Cases for the Indicial Roots

Let \( r_1 \geq r_2 \). The behavior of the two solutions depends on \( r_1 - r_2 \):

Case 1
\( r_1 - r_2 \notin \mathbb{Z} \): Two independent Frobenius series, one for each root.
Case 2
\( r_1 = r_2 \): One Frobenius series; second solution contains a \( \ln(x - x_0) \) term.
Case 3
\( r_1 - r_2 \in \mathbb{Z}^+ \): First root always gives a Frobenius series; second may need a logarithmic term.

Power Series vs Frobenius Method

Power Series Frobenius
Point type Ordinary Regular singular
Assumed form \( \sum a_n(x-x_0)^n \) \( \sum a_n(x-x_0)^{n+r} \)
Exponent \( r \) Always 0 (integer) Found via indicial eq.
Logarithmic terms Never Sometimes (Cases 2, 3)
First step Substitute directly Find \( p_0, q_0 \), solve indicial eq.

How to Apply the Frobenius Method

1
Standard form: Write the ODE as \( y'' + P(x)y' + Q(x)y = 0 \).
2
Classify \( x_0 \): Check that \( (x-x_0)P(x) \) and \( (x-x_0)^2 Q(x) \) are analytic (regular singular).
3
Find \( p_0 \) and \( q_0 \): Evaluate the limits as \( x \to x_0 \).
4
Solve the indicial equation: \( r(r-1) + p_0 r + q_0 = 0 \) to find \( r_1, r_2 \).
5
Build recurrence: Substitute \( y = \sum a_n x^{n+r} \) and match coefficients to find \( a_n \).
6
Determine case: Use \( r_1 - r_2 \) to decide if the second solution needs a logarithmic term.

Example 1: Classify the Point

Step 1: Divide by \( 2x^2 \) to get standard form:

Step 2: Check regularity at \( x_0 = 0 \):

\( x \cdot P(x) \)
\( x \cdot \frac{1}{2x} = \frac{1}{2} \) : analytic ✓
\( x^2 \cdot Q(x) \)
\( x^2 \cdot \bigl(-\frac{x+1}{2x^2}\bigr) = -\frac{x+1}{2} \) : analytic ✓

Example 1: Indicial Equation

Step 3: Compute \( p_0 \) and \( q_0 \):

\( p_0 \)
\( \displaystyle\lim_{x\to 0} x \cdot P(x) = \frac{1}{2} \)
\( q_0 \)
\( \displaystyle\lim_{x\to 0} x^2 \cdot Q(x) = -\frac{1}{2} \)

Step 4: The indicial equation is:

Example 2: Bessel-type Equation

Standard form: \( y'' + \frac{1}{x}y' + \frac{x^2 - 1/4}{x^2}y = 0 \)

Check: \( xP(x) = 1 \) (analytic ✓), \( x^2 Q(x) = x^2 - \frac{1}{4} \) (analytic ✓).

So \( p_0 = 1 \) and \( q_0 = -\frac{1}{4} \). The indicial equation:

Example 3: Equal Roots (Case 2)

Standard form: \( y'' + \frac{3}{x}y' + \frac{1+x}{x^2}y = 0 \)

\( p_0 = \lim_{x\to 0} x \cdot \frac{3}{x} = 3 \), and \( q_0 = \lim_{x\to 0} x^2 \cdot \frac{1+x}{x^2} = 1 \).

Why Does Frobenius Work?

1
Fuchs' theorem guarantees at least one Frobenius-type solution exists at every regular singular point.
2
The factor \( (x - x_0)^r \) absorbs the singular behavior. Multiplying by this factor "regularizes" the problem so the remaining power series converges.
3
The indicial equation is the solvability condition for the leading coefficient \( a_0 \neq 0 \). It determines which values of \( r \) are consistent.
4
The recurrence relation for \( a_n \) follows from matching coefficients of each power of \( (x - x_0) \), just like the ordinary power series method.

Famous Equations Using Frobenius

Many of the most important equations in physics and engineering have regular singular points that are crucial to their analysis:

Bessel
\( x^2y'' + xy' + (x^2 - \nu^2)y = 0 \)
(Regular singular point at \(x=0\))
Legendre
\( (1-x^2)y'' - 2xy' + \ell(\ell+1)y = 0 \)
(Regular singular points at \(x=\pm 1\))
Laguerre
\( xy'' + (1-x)y' + ny = 0 \)
(Regular singular point at \(x=0\))

Common Mistakes to Avoid

✕ Forgetting Standard Form
The leading coefficient of \( y'' \) must be 1 before identifying \( P(x) \) and \( Q(x) \). Always divide through first.
✕ Skipping the Classification
Applying Frobenius at an irregular singular point will fail. Always verify both regularity conditions.
✕ Wrong Indicial Equation
Mixing up \( p_0 \) and \( q_0 \), or using \( P(x_0) \) instead of \( \lim_{x \to x_0} (x-x_0)P(x) \). These are different things.

Summary: The Three Cases

Condition Solution \( y_1 \) Solution \( y_2 \)
\( r_1 - r_2 \notin \mathbb{Z} \) \( x^{r_1}\sum a_n x^n \) \( x^{r_2}\sum b_n x^n \)
\( r_1 = r_2 \) \( x^{r_1}\sum a_n x^n \) \( y_1\ln x + x^{r_1}\sum b_n x^n \)
\( r_1 - r_2 \in \mathbb{Z}^+ \) \( x^{r_1}\sum a_n x^n \) \( Cy_1\ln x + x^{r_2}\sum b_n x^n \)

Question 1 of 5

For the ODE \( x^2y'' + xy' + x^2y = 0 \), what is \( x_0 = 0 \)?

Question 2 of 5

What is the indicial equation for \( 2x^2y'' + 3xy' - y = 0 \) at \( x_0 = 0 \)?

Question 3 of 5

If the indicial roots are \( r_1 = 3 \) and \( r_2 = 3 \), which case applies?

Question 4 of 5

For the ODE \( x^3y'' + xy' + y = 0 \), what type of point is \( x_0 = 0 \)?

Question 5 of 5

In the Frobenius series \( y = \sum_{n=0}^{\infty} a_n x^{n+r} \), what determines the value of \( r \)?