1.4: Partial Fractions
- Last updated
- Aug 16, 2023
- Save as PDF
- Page ID
- 138035
- Gilbert Strang & Edwin “Jed” Herman
- University of Iowa
( \newcommand{\kernel}{\mathrm{null}\,}\)
Learning Objectives
- Integrate a rational function using the method of partial fractions.
- Recognize simple linear factors in a rational function.
- Recognize repeated linear factors in a rational function.
- Recognize quadratic factors in a rational function.
We have seen some techniques that allow us to integrate specific rational functions. For example, we know that
∫duu=ln|u|+C
and
∫duu2+a2=1atan−1(ua)+C.
However, we do not yet have a technique that allows us to tackle arbitrary quotients of this type. Thus, it is not immediately obvious how to go about evaluating
∫3xx2−x−2dx.
However, we know from material previously developed that
∫(1x+1+2x−2)dx=ln|x+1|+2ln|x−2|+C.
In fact, by getting a common denominator, we see that
1x+1+2x−2=3xx2−x−2.
Consequently,
∫3xx2−x−2dx=∫(1x+1+2x−2)dx.
In this section, we examine the method of partial fraction decomposition, which allows us to decompose rational functions into sums of simpler, more easily integrated rational functions. Using this method, we can rewrite an expression such as:
3xx2−x−2
as an expression such as
1x+1+2x−2.
The key to the method of partial fraction decomposition is being able to anticipate the form that the decomposition of a rational function will take. As we shall see, this form is both predictable and highly dependent on the factorization of the denominator of the rational function. It is also extremely important to keep in mind that partial fraction decomposition can be applied to a rational function P(x)Q(x) only if deg(P(x))<deg(Q(x)). In the case when deg(P(x))≥deg(Q(x)), we must first perform long division to rewrite the quotient P(x)Q(x) in the form A(x)+R(x)Q(x), where deg(R(x))<deg(Q(x)). We then do a partial fraction decomposition on R(x)Q(x). The following example, although not requiring partial fraction decomposition, illustrates our approach to integrals of rational functions of the form ∫P(x)Q(x)dx, where deg(P(x))≥deg(Q(x)).
Example 1.4.1: Integrating ∫P(x)Q(x)dx, where deg(P(x))≥deg(Q(x))
Evaluate
∫x2+3x+5x+1dx.
Solution
Since deg(x2+3x+5)≥deg(x+1), we perform long division to obtain
x2+3x+5x+1=x+2+3x+1.
Thus,
∫x2+3x+5x+1dx=∫(x+2+3x+1)dx=12x2+2x+3ln|x+1|+C.
Visit this website for a review of long division of polynomials.
Exercise 1.4.1
Evaluate
∫x−3x+2dx.
- Hint
-
Use long division to obtain x−3x+2=1−5x+2.
- Answer
-
x−5ln|x+2|+C
To integrate ∫P(x)Q(x)dx, where deg(P(x))<deg(Q(x)), we must begin by factoring Q(x).
Nonrepeated Linear Factors
If Q(x) can be factored as (a1x+b1)(a2x+b2)…(anx+bn), where each linear factor is distinct, then it is possible to find constants A1,A2,…An satisfying
P(x)Q(x)=A1a1x+b1+A2a2x+b2+⋯+Ananx+bn.
The proof that such constants exist is beyond the scope of this course.
In this next example, we see how to use partial fractions to integrate a rational function of this type.
Example 1.4.2: Partial Fractions with Nonrepeated Linear Factors
Evaluate ∫3x+2x3−x2−2xdx.
Solution
Since deg(3x+2)<deg(x3−x2−2x), we begin by factoring the denominator of 3x+2x3−x2−2x. We can see that x3−x2−2x=x(x−2)(x+1). Thus, there are constants A, B, and C satisfying Equation ??? such that
3x+2x(x−2)(x+1)=Ax+Bx−2+Cx+1.
We must now find these constants. To do so, we begin by getting a common denominator on the right. Thus,
3x+2x(x−2)(x+1)=A(x−2)(x+1)+Bx(x+1)+Cx(x−2)x(x−2)(x+1).
Now, we set the numerators equal to each other, obtaining
3x+2=A(x−2)(x+1)+Bx(x+1)+Cx(x−2).
There are two different strategies for finding the coefficients A, B, and C. We refer to these as the method of equating coefficients and the method of strategic substitution.
Strategy one: Method of Equating Coefficients
Rewrite Equation ??? in the form
3x+2=(A+B+C)x2+(−A+B−2C)x+(−2A).
Equating coefficients produces the system of equations
A+B+C=0−A+B−2C=3−2A=2.
To solve this system, we first observe that −2A=2⇒A=−1. Substituting this value into the first two equations gives us the system
B+C=1
B−2C=2.
Multiplying the second equation by −1 and adding the resulting equation to the first produces
−3C=1,
which in turn implies that C=−13. Substituting this value into the equation B+C=1 yields B=43. Thus, solving these equations yields A=−1,B=43, and C=−13.
It is important to note that the system produced by this method is consistent if and only if we have set up the decomposition correctly. If the system is inconsistent, there is an error in our decomposition.
Strategy two: Method of Strategic Substitution
The method of strategic substitution is based on the assumption that we have set up the decomposition correctly. If the decomposition is set up correctly, then there must be values of A,B, and C that satisfy Equation ??? for all values of x. That is, this equation must be true for any value of x we care to substitute into it. Therefore, by choosing values of x carefully and substituting them into the equation, we may find A,B, and C easily. For example, if we substitute x=0, the equation reduces to 2=A(−2)(1). Solving for A yields A=−1. Next, by substituting x=2, the equation reduces to 8=B(2)(3), or equivalently B=4/3. Last, we substitute x=−1 into the equation and obtain −1=C(−1)(−3). Solving, we have C=−13.
It is important to keep in mind that if we attempt to use this method with a decomposition that has not been set up correctly, we are still able to find values for the constants, but these constants are meaningless. If we do opt to use the method of strategic substitution, then it is a good idea to check the result by recombining the terms algebraically.
Now that we have the values of A,B, and C, we rewrite the original integral:
∫3x+2x3−x2−2xdx=∫(−1x+43⋅1x−2−13⋅1x+1)dx.
Evaluating the integral gives us
∫3x+2x3−x2−2xdx=−ln|x|+43ln|x−2|−13ln|x+1|+C.
In the next example, we integrate a rational function in which the degree of the numerator is not less than the degree of the denominator.
Example 1.4.3: Dividing before Applying Partial Fractions
Evaluate ∫x2+3x+1x2−4dx.
Solution
Since deg(x2+3x+1)≥deg(x2−4), we must perform long division of polynomials. This results in
x2+3x+1x2−4=1+3x+5x2−4
Next, we perform partial fraction decomposition on 3x+5x2−4=3x+5(x+2)(x−2). We have
3x+5(x−2)(x+2)=Ax−2+Bx+2.
Thus,
3x+5=A(x+2)+B(x−2).
Solving for A and B using either method, we obtain A=11/4 and B=1/4.
Rewriting the original integral, we have
∫x2+3x+1x2−4dx=∫(1+114⋅1x−2+14⋅1x+2)dx.
Evaluating the integral produces
∫x2+3x+1x2−4dx=x+114ln|x−2|+14ln|x+2|+C.
As we see in the next example, it may be possible to apply the technique of partial fraction decomposition to a nonrational function. The trick is to convert the nonrational function to a rational function through a substitution.
Example 1.4.4: Applying Partial Fractions after a Substitution
Evaluate ∫cosxsin2x−sinxdx.
Solution
Let’s begin by letting u=sinx. Consequently, du=cosxdx. After making these substitutions, we have
∫cosxsin2x−sinxdx=∫duu2−u=∫duu(u−1).
Applying partial fraction decomposition to 1u(u−1) gives 1u(u−1)=−1u+1u−1.
Thus,
∫cosxsin2x−sinxdx=−ln|u|+ln|u−1|+C=−ln|sinx|+ln|sinx−1|+C.
Exercise 1.4.2
Evaluate ∫x+1(x+3)(x−2)dx.
- Hint
-
x+1(x+3)(x−2)=Ax+3+Bx−2
- Answer
-
25ln|x+3|+35ln|x−2|+C
Repeated Linear Factors
For some applications, we need to integrate rational expressions that have denominators with repeated linear factors—that is, rational functions with at least one factor of the form (ax+b)n, where n is a positive integer greater than or equal to 2. If the denominator contains the repeated linear factor (ax+b)n, then the decomposition must contain
A1ax+b+A2(ax+b)2+⋯+An(ax+b)n.
As we see in our next example, the basic technique used for solving for the coefficients is the same, but it requires more algebra to determine the numerators of the partial fractions.
Example 1.4.5: Partial Fractions with Repeated Linear Factors
Evaluate ∫x−2(2x−1)2(x−1)dx.
Solution
We have deg(x−2)<deg((2x−1)2(x−1)), so we can proceed with the decomposition. Since (2x−1)2 is a repeated linear factor, include
A2x−1+B(2x−1)2
in the decomposition in Equation ???. Thus,
x−2(2x−1)2(x−1)=A2x−1+B(2x−1)2+Cx−1.
After getting a common denominator and equating the numerators, we have
x−2=A(2x−1)(x−1)+B(x−1)+C(2x−1)2.
We then use the method of equating coefficients to find the values of A,B, and C.
x−2=(2A+4C)x2+(−3A+B−4C)x+(A−B+C).
Equating coefficients yields 2A+4C=0, −3A+B−4C=1, and A−B+C=−2. Solving this system yields A=2,B=3, and C=−1.
Alternatively, we can use the method of strategic substitution. In this case, substituting x=1 and x=1/2 into Equation ??? easily produces the values B=3 and C=−1. At this point, it may seem that we have run out of good choices for x, however, since we already have values for B and C, we can substitute in these values and choose any value for x not previously used. The value x=0 is a good option. In this case, we obtain the equation −2=A(−1)(−1)+3(−1)+(−1)(−1)2 or, equivalently, A=2.
Now that we have the values for A,B, and C, we rewrite the original integral and evaluate it:
∫x−2(2x−1)2(x−1)dx=∫(22x−1+3(2x−1)2−1x−1)dx=ln|2x−1|−32(2x−1)−ln|x−1|+C.
Exercise 1.4.3
Set up the partial fraction decomposition for
∫x+2(x+3)3(x−4)2dx.
(Do not solve for the coefficients or complete the integration.)
- Hint
-
Use the problem-solving method of Example 1.4.5 for guidance.
- Answer
-
x+2(x+3)3(x−4)2=Ax+3+B(x+3)2+C(x+3)3+D(x−4)+E(x−4)2
The General Method
Now that we are beginning to get the idea of how the technique of partial fraction decomposition works, let’s outline the basic method in the following problem-solving strategy.
Problem-Solving Strategy: Partial Fraction Decomposition
To decompose the rational function P(x)/Q(x), use the following steps:
- Make sure that deg(P(x))<deg(Q(x)). If not, perform long division of polynomials.
- Factor Q(x) into the product of linear and irreducible quadratic factors. An irreducible quadratic is a quadratic that has no real zeros.
- Assuming that deg(P(x))<deg(Q(x), the factors of Q(x) determine the form of the decomposition of P(x)/Q(x).
- If Q(x) can be factored as (a1x+b1)(a2x+b2)…(anx+bn), where each linear factor is distinct, then it is possible to find constants A1,A2,...An satisfying P(x)Q(x)=A1a1x+b1+A2a2x+b2+⋯+Ananx+bn.
- If Q(x) contains the repeated linear factor (ax+b)n, then the decomposition must contain A1ax+b+A2(ax+b)2+⋯+An(ax+b)n.
- For each irreducible quadratic factor ax2+bx+c that Q(x) contains, the decomposition must include Ax+Bax2+bx+c.
- For each repeated irreducible quadratic factor (ax2+bx+c)n, the decomposition must include A1x+B1ax2+bx+c+A2x+B2(ax2+bx+c)2+⋯+Anx+Bn(ax2+bx+c)n.
- After the appropriate decomposition is determined, solve for the constants.
- Last, rewrite the integral in its decomposed form and evaluate it using previously developed techniques or integration formulas.
Simple Quadratic Factors
Now let’s look at integrating a rational expression in which the denominator contains an irreducible quadratic factor. Recall that the quadratic ax2+bx+c is irreducible if ax2+bx+c=0 has no real zeros—that is, if b2−4ac<0.
Example 1.4.6: Rational Expressions with an Irreducible Quadratic Factor
Evaluate
∫2x−3x3+xdx.
Solution
Since deg(2x−3)<deg(x3+x), factor the denominator and proceed with partial fraction decomposition. Since x^3+x=x(x^2+1) contains the irreducible quadratic factor x^2+1, include \dfrac{Ax+B}{x^2+1} as part of the decomposition, along with \dfrac{C}{x} for the linear term x. Thus, the decomposition has the form
\dfrac{2x−3}{x(x^2+1)}=\dfrac{Ax+B}{x^2+1}+\dfrac{C}{x}.\nonumber
After getting a common denominator and equating the numerators, we obtain the equation
2x−3=(Ax+B)x+C(x^2+1).\nonumber
Solving for A,B, and C, we get A=3, B=2, and C=−3.
Thus,
\dfrac{2x−3}{x^3+x}=\dfrac{3x+2}{x^2+1}−\dfrac{3}{x}.\nonumber
Substituting back into the integral, we obtain
\begin{align*} \int \dfrac{2x−3}{x^3+x}\,dx &=\int \left(\dfrac{3x+2}{x^2+1}−\dfrac{3}{x}\right)\,dx \nonumber \\[4pt] &=3\int \dfrac{x}{x^2+1}\,dx+2\int \dfrac{1}{x^2+1}\,dx−3\int \dfrac{1}{x}\,dx & & \text{Split up the integral} \\[4pt] &=\dfrac{3}{2}\ln ∣x^2+1∣+2 \tan^{−1}x−3\ln |x|+C. & & \text{Evaluate each integral} \end{align*}
Note: We may rewrite \ln ∣x^2+1∣=\ln (x^2+1), if we wish to do so, since x^2+1>0.
Example \PageIndex{7}: Partial Fractions with an Irreducible Quadratic Factor
Evaluate \displaystyle \int \dfrac{\,dx}{x^3−8}.
Solution: We can start by factoring x^3−8=(x−2)(x^2+2x+4). We see that the quadratic factor x^2+2x+4 is irreducible since 2^2−4(1)(4)=−12<0. Using the decomposition described in the problem-solving strategy, we get
\dfrac{1}{(x−2)(x^2+2x+4)}=\dfrac{A}{x−2}+\dfrac{Bx+C}{x^2+2x+4}. \nonumber
After obtaining a common denominator and equating the numerators, this becomes
1=A(x^2+2x+4)+(Bx+C)(x−2). \nonumber
Applying either method, we get A=\dfrac{1}{12},B=−\dfrac{1}{12}, and C=−\dfrac{1}{3}.
Rewriting \int \dfrac{\,dx}{x^3−8}, we have
\int \dfrac{\,dx}{x^3−8}=\dfrac{1}{12}\int \dfrac{1}{x−2}\,dx−\dfrac{1}{12}\int \dfrac{x+4}{x^2+2x+4}\,dx. \nonumber
We can see that
\int \dfrac{1}{x−2}\,dx=\ln |x−2|+C,\nonumber
but
\int \dfrac{x+4}{x^2+2x+4}\,dx \nonumber
requires a bit more effort. Let’s begin by completing the square on x^2+2x+4 to obtain
x^2+2x+4=(x+1)^2+3. \nonumber
By letting u=x+1 and consequently du=\,dx, we see that
\begin{align*} \int \dfrac{x+4}{x^2+2x+4}\,dx &=\int \dfrac{x+4}{(x+1)^2+3}\,dx & & \text{Complete the square on the denominator} \\[4pt] &=\int \dfrac{u+3}{u^2+3}\,du & & \text{Substitute }u=x+1,\,x=u−1,\text{ and } du=dx \\[4pt] &=\int \dfrac{u}{u^2+3}du+\int \dfrac{3}{u^2+3}du & & \text{Split the numerator apart} \\[4pt] &=\dfrac{1}{2}\ln ∣u^2+3∣+\dfrac{3}{\sqrt{3}}\tan^{−1}\dfrac{u}{\sqrt{3}}+C & & \text{Evaluate each integral} \\[4pt] &=\dfrac{1}{2}\ln ∣x^2+2x+4∣+\sqrt{3}\tan^{−1}\left(\dfrac{x+1}{\sqrt{3}}\right)+C & & \text{Rewrite in terms of }x\text{ and simplify} \end{align*}
Substituting back into the original integral and simplifying gives
\int \dfrac{ \,dx}{x^3−8}=\dfrac{1}{12}\ln |x−2|−\dfrac{1}{24}\ln |x^2+2x+4|−\dfrac{\sqrt{3}}{12}\tan^{−1}\left(\dfrac{x+1}{\sqrt{3}}\right)+C. \nonumber
Here again, we can drop the absolute value if we wish to do so, since x^2+2x+4>0 for all x.
Example \PageIndex{8}: Finding a Volume
Find the volume of the solid of revolution obtained by revolving the region enclosed by the graph of f(x)=\dfrac{x^2}{(x^2+1)^2} and the x-axis over the interval [0,1] about the y-axis.
Solution
Let’s begin by sketching the region to be revolved (see Figure \PageIndex{1}). From the sketch, we see that the shell method is a good choice for solving this problem.

The volume is given by
V=2π\int ^1_0x⋅\dfrac{x^2}{(x^2+1)^2}\,dx=2π\int ^1_0\dfrac{x^3}{(x^2+1)^2}\,dx. \nonumber
Since deg((x^2+1)^2)=4>3=deg(x^3), we can proceed with partial fraction decomposition. Note that (x^2+1)^2 is a repeated irreducible quadratic. Using the decomposition described in the problem-solving strategy, we get
\dfrac{x^3}{(x^2+1)^2}=\dfrac{Ax+B}{x^2+1}+\dfrac{Cx+D}{(x^2+1)^2}. \nonumber
Finding a common denominator and equating the numerators gives
x^3=(Ax+B)(x^2+1)+Cx+D. \nonumber
Solving, we obtain A=1, B=0, C=−1, and D=0. Substituting back into the integral, we have
V=2π\int _0^1\dfrac{x^3}{(x^2+1)^2}\,dx=2π\int _0^1\left(\dfrac{x}{x^2+1}−\dfrac{x}{(x^2+1)^2}\right)\,dx=2π\left(\dfrac{1}{2}\ln (x^2+1)+\dfrac{1}{2}⋅\dfrac{1}{x^2+1}\right)\Big|^1_0=π\left(\ln 2−\tfrac{1}{2}\right). \nonumber
Exercise \PageIndex{4}
Set up the partial fraction decomposition for \int \dfrac{x^2+3x+1}{(x+2)(x−3)^2(x^2+4)^2}\,dx. \nonumber
- Hint
-
Use the problem-solving strategy.
- Answer
-
\dfrac{x^2+3x+1}{(x+2)(x−3)^2(x^2+4)^2}=\dfrac{A}{x+2}+\dfrac{B}{x−3}+\dfrac{C}{(x−3)^2}+\dfrac{Dx+E}{x^2+4}+\dfrac{Fx+G}{(x^2+4)^2} \nonumber
Key Concepts
- Partial fraction decomposition is a technique used to break down a rational function into a sum of simple rational functions that can be integrated using previously learned techniques.
- When applying partial fraction decomposition, we must make sure that the degree of the numerator is less than the degree of the denominator. If not, we need to perform long division before attempting partial fraction decomposition.
- The form the decomposition takes depends on the type of factors in the denominator. The types of factors include nonrepeated linear factors, repeated linear factors, nonrepeated irreducible quadratic factors, and repeated irreducible quadratic factors.
Glossary
- partial fraction decomposition
- a technique used to break down a rational function into the sum of simple rational functions