Fact-checked by Grok 2 weeks ago

Annihilator method

The annihilator method is a systematic for determining particular solutions to nonhomogeneous linear ordinary differential equations with constant coefficients, particularly when the nonhomogeneous term consists of polynomials, exponentials, , or their products. It leverages differential operators—such as powers of the D = \frac{d}{dx}—that reduce the specified forcing function to zero, thereby transforming the original nonhomogeneous into a higher-order homogeneous one whose general solution encompasses the desired particular solution. This approach builds on the method of undetermined coefficients by providing a structured way to guess the form of the particular solution without for the coefficients alone. An is defined as a linear A(D) such that A(D)[g(x)] = 0, where g(x) is the nonhomogeneous term; for instance, D^{n+1} annihilates any of n, (D - \alpha) annihilates e^{\alpha x}, and (D^2 + \omega^2) annihilates \sin(\omega x) or \cos(\omega x). For composite functions, the annihilator is the product of individual operators, ensuring the method applies efficiently to common forcing functions encountered in applications like physics and . The procedure typically begins by solving the associated homogeneous equation to obtain the complementary solution y_c. An appropriate annihilator is then identified and applied to both sides of the original equation L = g(x), yielding A(L) = 0, a new homogeneous equation. The general solution to this elevated-order equation is found, and the components overlapping with y_c are discarded to isolate the particular solution y_p. Finally, the general solution is y = y_c + y_p, with constants determined via initial or boundary conditions if provided. This method's advantages include its algorithmic nature and avoidance of integration, making it preferable over for suitable g(x).

Background

Linear Differential Equations with Constant Coefficients

Linear ordinary differential equations (ODEs) with constant coefficients form a fundamental class in the study of differential equations, characterized by their linear structure and unchanging coefficients. The general form of an nth-order linear with constant coefficients is given by a_n y^{(n)}(x) + a_{n-1} y^{(n-1)}(x) + \dots + a_1 y'(x) + a_0 y(x) = g(x), where a_n, a_{n-1}, \dots, a_0 are constants with a_n \neq 0, y^{(k)}(x) denotes the kth of y(x), and g(x) represents the nonhomogeneous forcing term. This equation models numerous physical phenomena, such as vibrations and electrical circuits, due to the tractability provided by the constant coefficients. For the homogeneous case, where g(x) = 0, the equation simplifies to a_n y^{(n)}(x) + a_{n-1} y^{(n-1)}(x) + \dots + a_0 y(x) = 0. The solution method relies on the , obtained by assuming a solution of the form y(x) = e^{rx} and substituting it into the homogeneous equation, yielding the a_n r^n + a_{n-1} r^{n-1} + \dots + a_1 r + a_0 = 0. This equation in r has n roots (counting multiplicities), which determine the form of the general . The roots of the dictate the basis functions for the general solution. For a distinct real root r_k, the corresponding solution component is c_k e^{r_k x}, where c_k is an arbitrary constant. If a real root r has multiplicity m, the solutions include e^{rx}, x e^{rx}, \dots, x^{m-1} e^{rx}. For roots \alpha \pm i\beta (with \beta \neq 0) of multiplicity one, the real-valued solutions are e^{\alpha x} \cos(\beta x) and e^{\alpha x} \sin(\beta x); for multiplicity m, the set extends to include factors of x^j for j = 0 to m-1 multiplied by these trigonometric functions. The general solution is a of these n linearly independent basis functions, ensuring a complete n-dimensional solution space. To facilitate analysis, these equations can be expressed using the D = \frac{d}{dx}, where higher powers represent higher derivatives. The homogeneous equation then becomes p(D) y = 0, with p(D) = a_n D^n + a_{n-1} D^{n-1} + \dots + a_0 being the evaluated at the D. This notation highlights the and commutativity of derivatives, aiding in and solution construction.

Homogeneous and Nonhomogeneous Equations

A homogeneous linear differential equation with constant coefficients is expressed in operator notation as p(D) y = 0, where D = \frac{d}{dx} denotes the differentiation operator and p(D) = a_n D^n + a_{n-1} D^{n-1} + \cdots + a_1 D + a_0 is a polynomial of degree n with constant coefficients a_i (and a_n \neq 0). The general solution to this equation consists of a linear combination of n linearly independent fundamental solutions: y_h(x) = c_1 y_1(x) + c_2 y_2(x) + \cdots + c_n y_n(x), where the c_i are arbitrary constants determined by initial or boundary conditions. This structure arises from the fact that the solution space forms an n-dimensional vector space under addition and scalar multiplication, as guaranteed by the linearity of the operator. In contrast, a nonhomogeneous with constant coefficients has the form p(D) y = g(x), where g(x) is a given non-zero forcing (also called the nonhomogeneous term). The general to this is the sum of the homogeneous and a particular : y(x) = y_h(x) + y_p(x), where y_p(x) is any specific satisfying p(D) y_p = g(x). The arbitrary constants in y_h(x) allow adjustment to meet initial conditions, while y_p(x) accounts for the influence of g(x). Solutions to the associated homogeneous equation fail to satisfy the nonhomogeneous equation because p(D) y_h = 0 \neq g(x) for g(x) \not\equiv 0. However, the linearity of p(D) ensures the applies: if y_h solves the homogeneous equation and y_p solves the nonhomogeneous one, then p(D)(y_h + y_p) = p(D) y_h + p(D) y_p = 0 + g(x) = g(x), so their sum solves the full equation. This principle underpins the decomposition of the solution and extends to linear combinations of multiple particular solutions when g(x) is a sum of terms. The foundational work on homogeneous linear differential equations with constant coefficients dates to the , with key developments by Leonhard Euler and . Euler established a general solution method using exponential functions, as detailed in his correspondence around 1743.

The Annihilator Method

Definition and Principle

The annihilator method provides a systematic approach to finding particular solutions for nonhomogeneous linear ordinary differential equations with constant coefficients by leveraging . Central to this method is the concept of an , defined as a linear A(D), where D = \frac{d}{dx} denotes the differentiation operator, such that A(D) g(x) = 0 for the given nonhomogeneous term g(x). This operator effectively "annihilates" the forcing function, transforming it into the zero function. The underlying principle exploits the properties of operator composition. Consider the original nonhomogeneous equation p(D) y = g(x), where p(D) is the characteristic operator of the homogeneous part. Since A(D) g(x) = 0, applying A(D) to both sides yields A(D) p(D) y = 0, converting the problem into a higher-order homogeneous equation whose general solution encompasses both the solution to the original homogeneous equation p(D) y = 0 and the desired particular solution y_p. This transformation allows the particular solution to be extracted from the broader solution space. Within the general solution to A(D) p(D) y = 0, the particular y_p consists of those components that do not belong to the solution space of the original homogeneous p(D) y = 0. Any terms overlapping with the homogeneous solution are discarded to isolate y_p, ensuring it satisfies the nonhomogeneous without redundancy. A key aspect of the method is the selection of a minimal , which is the lowest-degree that satisfies A(D) g(x) = 0. Higher-degree annihilators may also work but introduce unnecessary complexity by expanding the order of the transformed beyond what is required.

Annihilators for Common Forcing Functions

The annihilator method identifies linear differential operators with constant coefficients that, when applied to a forcing function g(x), yield zero. These operators are essential for transforming nonhomogeneous equations into homogeneous ones. The following table summarizes annihilators for standard classes of forcing functions encountered in linear differential equations with constant coefficients.
Forcing Function g(x)Annihilator
Polynomial: x^mD^{m+1}
Exponential: e^{ax}D - a
Trigonometric: \sin(bx) or \cos(bx)D^2 + b^2
Product: x^m e^{ax}(D - a)^{m+1}
Product: x^m e^{ax} \sin(bx) or x^m e^{ax} \cos(bx)((D - a)^2 + b^2)^{m+1}
For a sum of forcing functions, such as g(x) = e^{ax} + \sin(bx), the annihilator is the product of the individual annihilators, (D - a)(D^2 + b^2), assuming the factors are relatively prime; if not, the is used instead. Constant multiples of these functions, like c \cdot x^m where c is a nonzero constant, share the same annihilator form, as the operators are linear and homogeneous. The effectiveness of D^{m+1} for polynomials of degree m stems from the fact that the (m+1)-th derivative of x^m is zero, mirroring how the Taylor expansion of such a polynomial terminates after m+1 terms, leaving higher derivatives null.

Step-by-Step Procedure

The annihilator method provides a systematic approach to finding a particular solution y_p for a nonhomogeneous linear ordinary differential equation (ODE) with constant coefficients, of the form p(D) y = g(x), where p(D) is a polynomial differential operator and g(x) is the forcing function. This technique leverages the fact that applying an annihilator operator A(D) to both sides of the equation yields a higher-order homogeneous equation whose general solution encompasses both the original homogeneous solution y_h and the particular solution y_p, along with extraneous terms. The method is particularly effective for forcing functions that are polynomials, exponentials, sines, cosines, or their products, as these admit simple annihilators. To apply the method, follow these steps:
  1. Express the given nonhomogeneous ODE in operator form as p(D) y = g(x), where p(D) is the characteristic polynomial operator corresponding to the left-hand side, and identify the minimal annihilator A(D) such that A(D) g(x) = 0. The minimal annihilator is the lowest-order differential operator that reduces g(x) and all its derivatives to zero; for common forms like polynomials or , these are tabulated in standard references.
  2. Apply the to both sides of the original to obtain the higher-order homogeneous equation A(D) p(D) y = 0. Solve this auxiliary equation by finding its characteristic roots and writing the general solution as y = y_c, which decomposes into the original homogeneous solution y_h (from p(D) y = 0), the particular solution y_p, and extraneous terms y_{extra} arising from the additional factors in A(D). The form of y_c is a of basis functions like e^{rx}, x^k e^{rx}, e^{\alpha x} \cos(\beta x), and e^{\alpha x} \sin(\beta x), determined by the roots of the for A(D) p(D).
  3. From the general solution y_c of the auxiliary equation, isolate y_p by discarding all terms that belong to the original homogeneous solution y_h. This ensures y_p is free of arbitrary constants already accounted for in y_h; the remaining terms, including any multiplicity adjustments from step 4, form y_p. If no overlap occurs, y_p consists directly of the non-homogeneous components introduced by A(D).
  4. Verify that the isolated y_p satisfies the original nonhomogeneous by substituting it back into p(D) y = g(x) and confirming equality. The full is then y = y_h + y_p, where constants in y_h are determined by initial or boundary conditions if provided.
When roots of the auxiliary overlap with those of the original p(D), indicating multiplicity in the forcing function relative to the homogeneous part, adjust the form of y_p by multiplying the standard basis terms by appropriate powers of x to ensure . Specifically, if a r of multiplicity m in A(D) coincides with a root of multiplicity s in p(D), include terms up to x^{m+s-1} e^{rx} (or analogous for complex roots) in the decomposition, but discard the first s powers when isolating y_p. This rule prevents redundancy and guarantees a valid particular solution.

Applications and Examples

Basic Example

Consider the second-order linear nonhomogeneous y'' + y' - 2y = x. To solve using the annihilator method, first find the solution to the associated homogeneous equation y'' + y' - 2y = 0. The is r^2 + r - 2 = 0, with roots r = 1 and r = -2. Thus, the homogeneous solution is y_h = c_1 e^{x} + c_2 e^{-2x}. The nonhomogeneous term x is a of degree 1, annihilated by the D^2, where D = \frac{d}{dx}, since D(x) = 1 and D^2(x) = 0. Applying D^2 to both sides of the original equation yields the higher-order homogeneous equation D^2 (D^2 + D - 2)y = 0. The is r^2 (r^2 + r - 2) = 0, with r = 0 (multiplicity 2), r = 1, and r = -2. The general solution to this equation is y = a + b x + c_1 e^{x} + c_2 e^{-2x}. The particular solution y_p consists of the terms arising from the annihilator's roots (the polynomial part), so assume y_p = a + b x. Substitute y_p into the original ODE: y_p' = b and y_p'' = 0, giving $0 + b - 2(a + b x) = x \implies b - 2a - 2 b x = x. Equating coefficients of like terms yields the system -2b = 1, \quad b - 2a = 0. Solving, b = -\frac{1}{2} and a = -\frac{1}{4}. Thus, y_p = -\frac{1}{2} x - \frac{1}{4}. To verify, compute y_p' = -\frac{1}{2}, \quad y_p'' = 0, so y_p'' + y_p' - 2 y_p = 0 - \frac{1}{2} - 2\left( -\frac{1}{2} x - \frac{1}{4} \right) = -\frac{1}{2} + x + \frac{1}{2} = x, which matches the nonhomogeneous term. The general solution is y = y_h + y_p = c_1 e^{x} + c_2 e^{-2x} - \frac{1}{2} x - \frac{1}{4}.

Advanced Example with Initial Conditions

To illustrate the application of the annihilator method to a more complex nonhomogeneous linear differential equation with initial conditions, consider the initial value problem y'' - 4y' + 5y = e^{x} \sin 2x, \quad y(0) = 0, \quad y'(0) = 1. The associated homogeneous equation y'' - 4y' + 5y = 0 has the characteristic equation r^2 - 4r + 5 = 0, with roots r = 2 \pm i. Thus, the homogeneous solution is y_h(x) = e^{2x} (A \cos x + B \sin x). The forcing function e^{x} \sin 2x is annihilated by the operator (D - 1)^2 + 4 = D^2 - 2D + 5, as this operator corresponds to the characteristic roots $1 \pm 2i and eliminates functions of the form e^{x} (c_1 \cos 2x + c_2 \sin 2x). Applying this annihilator to the original equation transforms it into the fourth-order homogeneous equation (D^2 - 2D + 5)(D^2 - 4D + 5)y = 0, whose general solution is y(x) = y_h(x) + e^{x} (C \cos 2x + D \sin 2x). The roots $1 \pm 2i from the annihilator do not overlap with the homogeneous roots $2 \pm i, so there is no resonance; the particular solution y_p(x) = e^{x} (C \cos 2x + D \sin 2x) requires no multiplicative polynomial factor. To find the coefficients C and D, substitute y_p into the original equation. Differentiating gives y_p'(x) = e^{x} \bigl[ (C + 2D) \cos 2x + (D - 2C) \sin 2x \bigr], y_p''(x) = e^{x} \bigl[ (-3C + 4D) \cos 2x + (-4C - 3D) \sin 2x \bigr]. Then, y_p'' - 4 y_p' + 5 y_p = e^{x} \bigl[ (-2C - 4D) \cos 2x + (4C - 2D) \sin 2x \bigr] = e^{x} \sin 2x. Equating coefficients yields the system -2C - 4D = 0, \quad 4C - 2D = 1. Solving, C = -2D from the first equation, so $4(-2D) - 2D = 1 implies -8D - 2D = 1, -10D = 1, D = -\frac{1}{10}, and C = \frac{1}{5}. Thus, y_p(x) = e^{x} \left( \frac{1}{5} \cos 2x - \frac{1}{10} \sin 2x \right). The general solution is y(x) = e^{2x} (A \cos x + B \sin x) + e^{x} \left( \frac{1}{5} \cos 2x - \frac{1}{10} \sin 2x \right). Applying the initial conditions, y(0) = A + \frac{1}{5} = 0 gives A = -\frac{1}{5}. Differentiating the general solution yields y'(x) = e^{2x} \bigl[ (2A + B) \cos x + (2B - A) \sin x \bigr] + e^{x} \bigl[ (C + 2D) \cos 2x + (D - 2C) \sin 2x \bigr], and since C + 2D = 0 and D - 2C = -\frac{1}{2}, this simplifies to y'(x) = e^{2x} \bigl[ (2A + B) \cos x + (2B - A) \sin x \bigr] - \frac{1}{2} e^{x} \sin 2x, so y'(0) = 2A + B = 1. Substituting A = -\frac{1}{5} gives B = \frac{7}{5}. The unique solution is therefore y(x) = e^{2x} \left( -\frac{1}{5} \cos x + \frac{7}{5} \sin x \right) + e^{x} \left( \frac{1}{5} \cos 2x - \frac{1}{10} \sin 2x \right). This example demonstrates how the annihilator method efficiently identifies the form of y_p for product forcing functions involving exponentials and sines, with initial conditions then determining the full solution constants.

Limitations and Comparisons

Limitations of the Method

The annihilator method is inherently limited to nonhomogeneous linear differential equations with constant coefficients where the forcing function g(x) can be annihilated by a linear differential operator with constant coefficients of finite order. Not all forcing functions possess such simple annihilators; for instance, functions like e^{x^2} or \ln |x| do not satisfy any finite-order linear homogeneous differential equation with constant coefficients, rendering the method inapplicable without resorting to alternative approaches. For non-standard or composite forcing functions outside the typical forms (such as polynomials, exponentials, sines, or cosines), the construction of an appropriate annihilator becomes ad hoc and may require extensive trial and error. A significant challenge arises when the roots of the annihilator overlap with those of the of the homogeneous equation, necessitating careful adjustment of multiplicities to extract the particular solution y_p from the enlarged solution space. This process can complicate the identification of y_p, as terms from the homogeneous solution must be excluded, often leading to modified trial forms like multiplying by powers of x. For higher-order equations, the method grows increasingly cumbersome, as applying the elevates the order of the auxiliary homogeneous , resulting in a substantially larger solution space that must be solved and partitioned to isolate y_p. Additionally, there is no systematic guarantee of finding the minimal-order for complex composite forcing functions, potentially requiring iterative refinements to ensure efficiency.

Comparison to Variation of Parameters

The annihilator method provides a systematic framework that builds upon and formalizes the method of undetermined coefficients for finding solutions to nonhomogeneous linear equations with constant coefficients. In the method of undetermined coefficients, the form of the is postulated based on the type of forcing g(x), such as assuming a of the same for g(x) or an exponential form for exponential g(x), with adjustments if it overlaps with the homogeneous . The annihilator method, by contrast, employs operators to "annihilate" g(x), deriving the form from the roots of the combined , which eliminates much of the guesswork and extends naturally to higher-order equations. This makes it particularly advantageous for complex cases where undetermined coefficients might become cumbersome, though both methods are limited to specific classes of g(x) like , exponentials, sines, cosines, and their products. Compared to the method, the annihilator approach offers a non-integral-based alternative that simplifies computations for applicable cases but sacrifices generality. works for any continuous g(x) and even variable-coefficient equations by assuming the particular solution as a of the fundamental homogeneous solutions with parameters determined via the and integration, often leading to more involved calculations especially for higher-order systems. The annihilator method circumvents these integrals entirely, relying instead on algebraic manipulation of , yet it fails when g(x) cannot be annihilated by a constant-coefficient or when coefficients are , rendering it inapplicable to broader scenarios where remains the go-to technique. The method is preferable over both alternatives when g(x) consists of polynomials, exponentials, sinusoids, or finite combinations thereof, as predefined annihilator tables facilitate rapid form determination without extensive or . It proves less ideal for arbitrary g(x) or non-constant coefficients, where the guess-based of undetermined coefficients falters for non-standard forms and becomes essential despite its heavier computational load. Historically, the annihilator method was popularized in 20th-century textbooks, such as those by Nagle, Saff, and Snider, as a pedagogical tool to unify and teach solutions for constant-coefficient nonhomogeneous equations.

References

  1. [1]
    4.6: Undetermined Coefficients - Annihilator Approach
    May 3, 2025 · Solve nonhomogeneous linear differential equations with constant coefficients using the annihilator method. Solve initial value problems with ...
  2. [2]
    5.5: Annihilation - Mathematics LibreTexts
    Oct 9, 2023 · As noted above, annihilators are derivatives and as you'll recall from calculus, derivatives lower the power on each term of the polynomial, ...
  3. [3]
    [PDF] Constant coefficient linear ODEs: review from 18.03 Class 1, 18.031
    Of course, for other physical systems we will use other symbols. The general nth order constant coefficient differential equation has the form anx(n) + an-1x(n- ...
  4. [4]
    Differential Equations - Basic Concepts - Pauls Online Math Notes
    Nov 16, 2022 · We will take the material from the Second Order chapter and expand it out to nth n th order linear differential equations. As we'll see almost ...<|control11|><|separator|>
  5. [5]
    [PDF] Linear nth Order ODEs with constant coefficients. - OU Math
    This can be written as. Ly = f(x) where L is the linear differential operator. L = Dn + a1Dn-1 + ··· + an-1D + an. 2. We will focus first of all on solving the ...
  6. [6]
    [PDF] 4.2 The Homogeneous Linear Equation with Constant Coefficients
    This equation is called the auxiliary equation or the characteristic equation of the given differential equation (4.11). If y = emx is a solution of (4.11) then ...
  7. [7]
  8. [8]
    [PDF] Chapter 3 - Constant Coefficients
    The use of the complex exponential function greatly simplifies the treat- ment of differential equations with constant coefficients.
  9. [9]
    [PDF] Higher Order Linear Differential Equations - Penn Math
    Jul 28, 2015 · Expressed as a linear differential operator, the equation is P(D)y = 0, where. P(D) = Dn + a1Dn-1 + ··· + an-1D + an. Definition. A linear ...
  10. [10]
    Linear Differential Equations - Department of Mathematics at UTSA
    Nov 5, 2021 · If the constant term is the zero function, then the differential equation is said to be homogeneous, as it is a homogeneous polynomial in the ...
  11. [11]
    [PDF] Chapter 2 Second Order Differential Equations - UNCW
    So far, we only know how to solve constant coefficient, homogeneous equations. So, by adding a nonhomogeneous term to such equations we will need to find ...
  12. [12]
    [PDF] Differential equations and linear algebra - Keith Conrad
    The equation (1.1) can be written in the condensed form p(D)(y)=0. That is, solutions y to (1.1) form the kernel of the differential operator p(D) on C∞(R).<|control11|><|separator|>
  13. [13]
    [PDF] Learning Mathematics from the Master: A Collection of Euler
    In short, Euler declared to his mentor that he had found a general method for solving homogeneous linear differential equations with constant coefficients.
  14. [14]
    [PDF] chapter 1 - UCLA | Bionics Lab
    Historical Background, Part II: Euler, Lagrange, and Laplace. The greatest ... homogeneous linear differential equations with constant coefficients (Sections 3.1, ...
  15. [15]
    How to Solve Nonhomogeneous Linear ODEs using Annihilators
    Dec 9, 2015 · How to Solve Nonhomogeneous Linear ODEs using Annihilators · The left side can be rewritten as ( D 2 – 4 D + 3 ) y . · The derivative of 5 = 0; ...
  16. [16]
    [PDF] Math 385 Supplement: the method of undetermined coefficients
    We know how to find all solutions of the annihilator equation (since it is a constant coefficient homogeneous DE), and this will tell us the correct form for ...
  17. [17]
  18. [18]
    [PDF] Nagle R.K., Saff E.B., Snider A.D. Fundamentals of differential ...
    The annihilator method applies to equations of the form. (6) where L is a linear differential operator with constant coefficients and the forcing term is a.
  19. [19]
    [PDF] Contents - University of Utah Math Dept.
    Example 6.34 (Annihilator Method). Find the shortest trial solution for the ... to the same human ear limitations. Two piano keys are more than 7 Hz ...
  20. [20]
    Differential Equations - Variation of Parameters
    Nov 16, 2022 · The method of Variation of Parameters is a much more general method that can be used in many more cases. However, there are two disadvantages to the method.<|control11|><|separator|>