Characteristic Roots Classify Linear ODE Solutions
How an algebraic polynomial determines growth, decay, and oscillation in constant-coefficient equations.
By Theory Commons Editors4 min readPublished Aug 22, 2026
For ay′′+by′+cy=0, try y=ert. Substitution produces
ar2+br+c=0,
the characteristic equation.
Three root patterns
Distinct real roots give y=C1er1t+C2er2t. A repeated root gives y=(C1+C2t)ert. Complex roots α±iβ give
y=eαt(C1cosβt+C2sinβt).
The real part controls exponential growth or decay; the imaginary part controls oscillation frequency.
Why the trial solution is legitimate
The operator L[y]=ay′′+by′+cy sends ert to (ar2+br+c)ert. Because ert is never zero, L[ert]=0 exactly when r is a root of the characteristic polynomial. The trial is therefore not a guess pulled from nowhere: exponentials are eigenfunctions of constant-coefficient differentiation.
When roots are distinct, the two exponential solutions are linearly independent. A second-order equation needs two independent solutions because two initial conditions, usually y(0) and y′(0), must determine two constants.
Check your understanding
Without solving for constants, classify the behavior of y′′+6y′+9y=0. Will generic solutions oscillate?
Show the reasoning
The polynomial is (r+3)2, so the repeated root is −3. Solutions have the form (C1+C2t)e−3t. They decay without sinusoidal oscillation; the factor t comes from the repeated root.