Integration by parts - Ciallo Wiki

1 Fundamental idea

Integration by parts is a technique for integrating a product of functions by transferring differentiation from one factor to another. It is especially effective when one factor becomes simpler after differentiation and the other factor can be integrated with little difficulty. The method is a direct consequence of the product rule in differentiation and is one of the standard tools of elementary and advanced calculus.

1.1 Relationship to the product rule

The product rule states that if \(u(x)\) and \(v(x)\) are differentiable, then \[ \frac{d}{dx}[u(x)v(x)] = u'(x)v(x) + u(x)v'(x). \] Rearranging this identity makes it possible to express one product term in a form suitable for integration. Integration by parts is therefore not an independent principle, but an algebraic restatement of a familiar differentiation law.

1.2 Derivation of the formula

1.2.1 Indefinite form

Starting from the product rule, \[ d(uv)=u\,dv+v\,du, \] and integrating both sides gives \[ uv=\int u\,dv+\int v\,du. \] Solving for one of the integrals yields the standard integration by parts formula: \[ \int u\,dv = uv-\int v\,du. \] This identity is the core of the method for indefinite integrals.

1.2.2 Definite form

For a definite integral over \([a,b]\), the corresponding formula is \[ \int_a^b u\,dv = \bigl[uv\bigr]_a^b - \int_a^b v\,du. \] The boundary term \(\bigl[uv\bigr]_a^b\) often plays an important role, since it may simplify the evaluation or even make the entire integral vanish.

1.3 Geometric and algebraic intuition

Algebraically, the method replaces a difficult product with a new integral that is often easier to handle. Geometrically, the transformation shifts the burden of complexity from one factor to another, rather than directly computing the area under a product. This change is particularly useful when one function has a simple derivative but a complicated antiderivative, while the other has the opposite behavior.

2 Basic formulas

2.1 Standard integration by parts identity

The standard identity is \[ \int u\,dv = uv-\int v\,du. \] In practice, one chooses a factor of the original integrand as \(u\), identifies the remaining part as \(dv\), computes \(du\) and \(v\), and then substitutes into the formula.

2.2 Rewriting products for easier integration

Many integrals become accessible only after a product is rewritten in a more suitable form. For example, a logarithm may be isolated as \(u\), while a polynomial or exponential factor is treated as \(dv\). The success of the method depends on choosing a decomposition that reduces complexity in the remaining integral.

2.3 Repeated application

Integration by parts can be used more than once when the resulting integral still contains a product. Repetition often leads to a simpler expression, a reducible recurrence, or a closed-form answer after several steps.

2.3.1 Tabular integration

Tabular integration is a shorthand method for repeated parts integration, especially effective for a polynomial times an exponential or trigonometric function. One factor is differentiated repeatedly until it becomes zero, while the other is integrated repeatedly. The resulting terms are then combined with alternating signs.

2.3.2 Cyclic integration

Some integrals reappear after one or more applications of the method, creating a cycle. In such cases, the original integral can be solved algebraically by moving the repeated term to one side and isolating it. This is common for certain products of trigonometric functions and exponentials.

3 Choosing \(u\) and \(dv\)

3.1 Common selection heuristics

A good choice of \(u\) usually becomes simpler when differentiated, while \(dv\) should be easy to integrate. This balancing act is central to the method. If the choice is poor, the transformed integral may be no easier than the original one.

A common heuristic is LIATE, which suggests prioritizing:

The rule is not absolute, but it often leads to effective choices in routine problems. Related mnemonics and local conventions serve the same purpose: to encourage a choice that makes differentiation favorable.

3.3 When the method is effective

Integration by parts is most useful when the integrand is a product and one factor has a derivative of lower complexity. Typical examples include polynomials multiplied by exponentials or trigonometric functions, logarithmic integrals, and inverse trigonometric expressions. It is less helpful when both factors become more complicated under differentiation or when no simpler antiderivative emerges.

4 Applications

4.1 Polynomial times exponential functions

Integrals such as \[ \int x^n e^x\,dx \] are classic applications. Differentiating the polynomial eventually eliminates it, leaving an elementary exponential integral. This makes the method efficient and often repetitive in a predictable pattern.

4.2 Polynomial times trigonometric functions

Expressions like \[ \int x^n\sin x\,dx \quad \text{or} \quad \int x^n\cos x\,dx \] are also well suited to repeated parts integration. The polynomial is reduced step by step, while the trigonometric factor remains easy to integrate at each stage.

4.3 Logarithmic integrals

Logarithms are often integrated by parts because their derivatives simplify significantly. For instance, \[ \int \ln x\,dx \] can be treated by writing \(\ln x\) as \(u\) and \(dx\) as \(dv\). More complicated logarithmic products can often be handled in the same way.

4.4 Inverse trigonometric integrals

Inverse trigonometric functions such as \(\arctan x\) and \(\arcsin x\) frequently appear in integrals where they are best chosen as \(u\). Their derivatives are rational expressions that may combine well with a simple \(dv\), leading to manageable remaining integrals.

4.5 Definite integrals involving boundary terms

For definite integrals, the boundary term may simplify the computation dramatically. In some cases, the product \(uv\) vanishes at one or both endpoints, leaving only a simpler integral to evaluate. This is especially useful in problems involving powers, logarithms, or trigonometric factors on bounded intervals.

5 Advanced techniques

5.1 Reduction formulas

Reduction formulas express a family of integrals in terms of another member of the same family with lower index or degree. Integration by parts is a common way to derive these formulas. They are particularly useful for powers of trigonometric functions, polynomial-exponential products, and similar structured integrals.

5.2 Recursive integrals

Some integrals generate a recurrence relation after one application of the method. By repeatedly substituting the relation, one can obtain a closed form or reduce the problem to a base case. This approach is common in symbolic integration and manual computation.

5.3 Integration by parts with substitutions

In some problems, substitution is used first to reshape the integral, after which integration by parts becomes effective. In other cases, parts integration creates an expression that is easiest to finish by substitution. The two methods often complement each other rather than competing.

5.4 Improper integrals

For improper integrals, integration by parts may be used with limits taken carefully. The boundary term must be evaluated as a limit, and convergence must be checked before assuming the formula applies straightforwardly. In many classical problems, parts integration helps determine whether the integral converges and what value it has.

6 Special cases and pitfalls

6.1 Integrals that do not simplify

Not every product becomes easier after the transformation. Some choices of \(u\) and \(dv\) return an integral of comparable difficulty or even greater complexity. In such situations, another method may be preferable.

6.2 Sign errors and algebraic mistakes

A frequent source of error is mishandling the minus sign in \[ \int u\,dv = uv-\int v\,du. \] Careful bookkeeping is essential, especially in repeated applications where multiple alternating signs appear. Small algebraic slips can change the result completely.

6.3 Choosing poor variables

If \(u\) is chosen so that its derivative is more complicated than the original function, the method may become unproductive. Likewise, selecting a \(dv\) that is hard to integrate defeats the purpose. Good judgment in selecting the components is often more important than the mechanics of the formula.

6.4 Constant of integration issues

For indefinite integrals, the final answer must include the constant of integration. When the method is applied repeatedly, intermediate constants can be absorbed into a single final constant. In definite integrals, no such constant appears, since endpoint evaluation determines the result.

7 Extensions

7.1 Multivariable integration by parts

In several variables, integration by parts generalizes through divergence-like identities and boundary integrals. These formulas connect derivatives over a region with contributions from its boundary. They are central in vector calculus and partial differential equations.

7.2 Integration by parts in differential equations

The technique appears in the derivation of weak formulations and energy estimates for differential equations. It allows derivatives to be shifted from one function to another, often reducing the required smoothness or revealing conserved quantities. This role makes it indispensable in analysis and applied mathematics.

7.3 Connection to Fourier analysis

Integration by parts is frequently used to study Fourier coefficients and transform behavior. By moving derivatives onto oscillatory factors, one can show decay properties and establish convergence results. The method is especially helpful when estimating integrals involving sines, cosines, and complex exponentials.

8 Examples

8.1 Worked indefinite integral examples

A standard example is \[ \int x e^x\,dx. \] Let \(u=x\) and \(dv=e^x\,dx\). Then \(du=dx\) and \(v=e^x\), so \[ \int x e^x\,dx = xe^x-\int e^x\,dx = xe^x-e^x+C. \] This illustrates the typical pattern: one factor simplifies under differentiation, and the remaining integral is elementary.

8.2 Worked definite integral examples

Consider \[ \int_0^1 x\ln x\,dx. \] Take \(u=\ln x\) and \(dv=x\,dx\), or alternatively \(u=x\) and \(dv=\ln x\,dx\) after rewriting. A convenient choice gives a boundary term and a manageable remainder, leading to a finite value. Such examples show how endpoint behavior and logarithmic terms interact.

8.3 Mixed-function examples

Integrals combining algebraic, exponential, and trigonometric terms often require more than one step. For example, \(\int x^2\cos x\,dx\) may need repeated parts integration, while \(\int e^x\sin x\,dx\) typically leads to a cyclic system that is solved algebraically. These cases demonstrate the versatility of the method.

8.4 Challenge problems

Typical challenge problems include integrals with nested products, logarithms multiplied by powers, and expressions that reappear after one application of the formula. Problems of this kind often test both method selection and algebraic accuracy. They also highlight when integration by parts is the natural tool and when another technique may be more efficient.