1.02y - Partial fractions with linear factors

1.02y - Partial fractions with linear factors

Split a complicated fraction into simple pieces, including repeated linear factors, and check the result by recombining. Then use the split with supplied expansion and integration rules to make longer problems manageable.

Splitting two distinct linear factors

Can making one fraction into several actually make a problem easier? Compare 1/((1−x)(1−2x))1/((1-x)(1-2x)) with −1/(1−x)+2/(1−2x)-1/(1-x)+2/(1-2x). At x=0x=0, both give 11; at x=1/4x=1/4, both give 8/38/3. The second form will make its power series almost immediate.

You already know

Adding fractions uses a common denominator. A polynomial identity holds at every real input; a fraction equation still excludes every zero of its denominator.

We are going to reverse that common-denominator calculation.

Partial fractions

A partial-fraction decomposition expresses a rational expression as a sum of simpler fractions with the same values on its domain. Distinct linear denominator factors each receive one constant numerator. For example, a denominator (x−1)(x+2)(x-1)(x+2) uses A/(x−1)+B/(x+2)A/(x-1)+B/(x+2).

The constants are unknown until the numerator identity determines them.

Worked example 1: two monic factors

Question. Express (5x+1)/((x−1)(x+2))(5x+1)/((x-1)(x+2)) in partial fractions.

Plan. Clear denominators and substitute the zeros of the factors into the resulting polynomial identity.

5x+1≡A(x+2)+B(x−1)clear denominatorsx=1:6=3A⇒A=2remove Bx=−2:−9=−3B⇒B=3remove A\begin{aligned} 5x+1&\equiv A(x+2)+B(x-1) && \text{clear denominators}\\ x=1:\quad6&=3A\quad\Rightarrow\quad A=2 && \text{remove }B\\ x=-2:\quad-9&=-3B\quad\Rightarrow\quad B=3 && \text{remove }A \end{aligned}

Why this step? Although the fractions are undefined at 1,−21,-2, the cleared polynomial identity holds there. We substitute into the identity, not into an undefined fraction.

Answer. 2/(x−1)+3/(x+2)2/(x-1)+3/(x+2), for x≠1,−2x\ne1,-2.

Check. The recombined numerator is 2(x+2)+3(x−1)=5x+12(x+2)+3(x-1)=5x+1.

Change one feature: one factor has leading coefficient 22. Its zero and its multiplier must both be respected.

Worked example 2: a non-monic factor

Question. Express (7x+4)/((2x−1)(x+2))(7x+4)/((2x-1)(x+2)) in partial fractions.

Plan. Keep the factors as written and use x=1/2,−2x=1/2,-2.

7x+4≡A(x+2)+B(2x−1)clear denominatorsx=12:152=52A⇒A=3first constantx=−2:−10=−5B⇒B=2second constant\begin{aligned} 7x+4&\equiv A(x+2)+B(2x-1) && \text{clear denominators}\\ x=\tfrac12:\quad\tfrac{15}2&=\tfrac52 A\quad\Rightarrow\quad A=3 && \text{first constant}\\ x=-2:\quad-10&=-5B\quad\Rightarrow\quad B=2 && \text{second constant} \end{aligned}

Why this step? Setting x=1x=1 would not make 2x−12x-1 zero; 1/21/2 does.

Answer. 3/(2x−1)+2/(x+2)3/(2x-1)+2/(x+2), for x≠1/2,−2x\ne1/2,-2.

Check. 3(x+2)+2(2x−1)=7x+43(x+2)+2(2x-1)=7x+4.

You can instead compare coefficients. In the last identity, A+2B=7A+2B=7 and 2A−B=42A-B=4. Both routes give the same constants. The decomposition method is not in the formula sheet: learn it.

Watch out

Wrong: 1/(2x−1)=1/(x−1/2)1/(2x-1)=1/(x-1/2). Right: 1/(2x−1)=(1/2)/(x−1/2)1/(2x-1)=(1/2)/(x-1/2). If you rescale a denominator, rescale its numerator too.

Keep the non-monic factor but change the numerator.

Handling three distinct factors

Three distinct linear factors need three simple fractions. Clearing the denominator gives a quadratic identity even when the starting numerator is only constant or linear.

Worked example 3: a constant numerator

Question. Decompose 6/((x−1)(x+1)(x+2))6/((x-1)(x+1)(x+2)).

Plan. Put constants A,B,CA,B,C over the three factors and isolate each in turn.

6≡A(x+1)(x+2)+B(x−1)(x+2)+C(x−1)(x+1)clear factorsx=1:6=6A⇒A=1isolate Ax=−1:6=−2B⇒B=−3isolate Bx=−2:6=3C⇒C=2isolate C\begin{aligned} 6&\equiv A(x+1)(x+2)+B(x-1)(x+2)+C(x-1)(x+1) && \text{clear factors}\\ x=1:\quad6&=6A\quad\Rightarrow\quad A=1 && \text{isolate }A\\ x=-1:\quad6&=-2B\quad\Rightarrow\quad B=-3 && \text{isolate }B\\ x=-2:\quad6&=3C\quad\Rightarrow\quad C=2 && \text{isolate }C \end{aligned}

Why this step? At each chosen input two products vanish, leaving just one constant to find.

Answer. 1/(x−1)−3/(x+1)+2/(x+2)1/(x-1)-3/(x+1)+2/(x+2), with x≠1,−1,−2x\ne1,-1,-2.

Check. (x2+3x+2)−3(x2+x−2)+2(x2−1)=6(x^2+3x+2)-3(x^2+x-2)+2(x^2-1)=6. Both higher-power coefficients cancel.

Change one feature: make the numerator linear. The denominator and setup stay the same.

Worked example 4: a linear numerator

Question. Decompose (x+5)/((x−1)(x+1)(x+2))(x+5)/((x-1)(x+1)(x+2)).

Plan. Use the same identity with x+5x+5 on its left.

x=1:6=6A⇒A=1first constantx=−1:4=−2B⇒B=−2second constantx=−2:3=3C⇒C=1third constant\begin{aligned} x=1:\quad6&=6A\quad\Rightarrow\quad A=1 && \text{first constant}\\ x=-1:\quad4&=-2B\quad\Rightarrow\quad B=-2 && \text{second constant}\\ x=-2:\quad3&=3C\quad\Rightarrow\quad C=1 && \text{third constant} \end{aligned}

Why this step? The new numerator changes the values substituted on the left; reusing the previous constants would give the wrong identity.

Answer. 1/(x−1)−2/(x+1)+1/(x+2)1/(x-1)-2/(x+1)+1/(x+2), for x≠1,−1,−2x\ne1,-1,-2.

Check. (x2+3x+2)−2(x2+x−2)+(x2−1)=x+5(x^2+3x+2)-2(x^2+x-2)+(x^2-1)=x+5.

Our denominator is no more complicated than three distinct linear factors or one linear factor times a squared linear factor. Numerators here are constant or linear, and the split uses no more than three terms. This setup is not in the formula sheet: learn it. An irreducible quadratic such as x2+1x^2+1 needs a different setup, beyond this lesson's scope.

Problem-solving

Check the recombined numerator, including coefficients that should be zero. With a constant starting numerator, all x2x^2 and xx terms must cancel.

Keep a linear numerator and change its constant.

Including every power of a repeated factor

A squared factor needs two available terms: one over its first power and one over its square. Without both, the numerator usually cannot be reconstructed.

A squared linear factor

For distinct factor zeros, the complete form is

px+q(ax+b)(cx+d)2=Aax+b+Bcx+d+C(cx+d)2,\frac{px+q}{(ax+b)(cx+d)^2} =\frac A{ax+b}+\frac B{cx+d}+\frac C{(cx+d)^2},

where a,c≠0a,c\ne0, and all denominator zeros remain excluded. Some solved coefficients may be zero.

Not in the formula sheet: learn it.

Use substitution for the constants it isolates, then compare a coefficient for the remaining one.

Worked example 5: a squared monic factor

Question. Decompose (3x+1)/((x−1)(x+1)2)(3x+1)/((x-1)(x+1)^2).

Plan. Include both powers of x+1x+1, clear the denominator and use the x2x^2 coefficient last.

3x+1≡A(x+1)2+B(x−1)(x+1)+C(x−1)complete identityx=1:4=4A⇒A=1isolate Ax=−1:−2=−2C⇒C=1isolate C0=A+B⇒B=−1compare x2\begin{aligned} 3x+1&\equiv A(x+1)^2+B(x-1)(x+1)+C(x-1) && \text{complete identity}\\ x=1:\quad4&=4A\quad\Rightarrow\quad A=1 && \text{isolate }A\\ x=-1:\quad-2&=-2C\quad\Rightarrow\quad C=1 && \text{isolate }C\\ 0&=A+B\quad\Rightarrow\quad B=-1 && \text{compare }x^2 \end{aligned}

Why this step? Substitution at the repeated zero removes both terms involving A,BA,B, so a different equation is needed for BB.

Answer. 1/(x−1)−1/(x+1)+1/(x+1)21/(x-1)-1/(x+1)+1/(x+1)^2, for x≠1,−1x\ne1,-1.

Check. (x+1)2−(x−1)(x+1)+(x−1)=3x+1(x+1)^2-(x-1)(x+1)+(x-1)=3x+1.

Change one feature: replace the repeated monic factor by 2x+12x+1. Its zero is now −1/2-1/2.

Worked example 6: a squared non-monic factor

Question. Decompose (7x+2)/((x−1)(2x+1)2)(7x+2)/((x-1)(2x+1)^2).

Plan. Keep both powers of 2x+12x+1 and compare the quadratic coefficient.

7x+2≡A(2x+1)2+B(x−1)(2x+1)+C(x−1)clear denominatorx=1:9=9A⇒A=1first constantx=−12:−32=−32C⇒C=1third constant0=4A+2B⇒B=−2compare x2\begin{aligned} 7x+2&\equiv A(2x+1)^2+B(x-1)(2x+1)+C(x-1) && \text{clear denominator}\\ x=1:\quad9&=9A\quad\Rightarrow\quad A=1 && \text{first constant}\\ x=-\tfrac12:\quad-\tfrac32&=-\tfrac32 C\quad\Rightarrow\quad C=1 && \text{third constant}\\ 0&=4A+2B\quad\Rightarrow\quad B=-2 && \text{compare }x^2 \end{aligned}

Why this step? Squaring 2x+12x+1 makes its leading coefficient 44, which matters in the coefficient equation.

Answer. 1/(x−1)−2/(2x+1)+1/(2x+1)21/(x-1)-2/(2x+1)+1/(2x+1)^2, with x≠1,−1/2x\ne1,-1/2.

Check. (2x+1)2−2(x−1)(2x+1)+(x−1)=7x+2(2x+1)^2-2(x-1)(2x+1)+(x-1)=7x+2.

Leaving out a possible term can accidentally work in a special case, but it is not a reliable general method. Set up every allowed power before finding the constants.

Watch out

Wrong: use only A/(x−1)+C/(x+1)2A/(x-1)+C/(x+1)^2 for a general denominator (x−1)(x+1)2(x-1)(x+1)^2. Right: include B/(x+1)B/(x+1) too. Its coefficient may be non-zero.

Keep the non-monic repeated factor and change the numerator.

Using the split in a power series

Partial fractions let you expand one simple factor at a time. We supply the needed rules here; the series lesson develops them further.

General binomial expansion

For real nn and ∣u∣<1|u|<1,

(1+u)n=1+nu+n(n−1)2!u2+⋯ .(1+u)^n=1+nu+\frac{n(n-1)}{2!}u^2+\cdots.

For non-negative integer nn, the terminating expansion is valid for every real uu.

In the formula sheet.

Set n=−1,−2n=-1,-2 and replace uu by −u-u. The useful cases, not printed separately on the sheet, are

11−u=1+u+u2+⋯ ,1(1−u)2=1+2u+3u2+⋯ ,∣u∣<1.\frac1{1-u}=1+u+u^2+\cdots,\qquad \frac1{(1-u)^2}=1+2u+3u^2+\cdots,\qquad |u|<1.

A power series is an infinite sum of powers; writing terms through x2x^2 truncates it and is not an exact polynomial identity.

Worked example 7: two simple factors

Question. Find the expansion through x2x^2 of 1/((1−x)(1−2x))1/((1-x)(1-2x)), using its split −1/(1−x)+2/(1−2x)-1/(1-x)+2/(1-2x). State validity.

Plan. Expand each fraction separately and intersect their validity conditions.

−11−x+21−2x=−(1+x+x2+⋯ )+2(1+2x+4x2+⋯ )use u=x,2x=1+3x+7x2+⋯collect coefficients\begin{aligned} -\frac1{1-x}+\frac2{1-2x} &=-(1+x+x^2+\cdots)+2(1+2x+4x^2+\cdots) && \text{use }u=x,2x\\ &=1+3x+7x^2+\cdots && \text{collect coefficients} \end{aligned}

Why this step? Both expansions must work together: ∣x∣<1|x|<1 and ∣2x∣<1|2x|<1.

Answer. 1+3x+7x21+3x+7x^2 through x2x^2, valid for ∣x∣<1/2|x|<1/2.

Check. Multiplying the displayed terms by 1−3x+2x21-3x+2x^2 gives constant 11 and zero coefficients of x,x2x,x^2.

Change one feature: square the factor 1−2x1-2x. Its coefficients change, but its zero and the strict validity bound do not.

Worked example 8: a repeated factor in a series

Question. Expand 1/((1−x)(1−2x)2)1/((1-x)(1-2x)^2) through x2x^2, using

1(1−x)(1−2x)2=11−x−21−2x+2(1−2x)2.\frac1{(1-x)(1-2x)^2}=\frac1{1-x}-\frac2{1-2x}+\frac2{(1-2x)^2}.

Plan. Use the squared-factor rule for the last term and collect powers.

(1+x+x2+⋯ )−2(1+2x+4x2+⋯ )simple factors+2(1+4x+12x2+⋯ )squared factor=1+5x+17x2+⋯collect powers\begin{aligned} & (1+x+x^2+\cdots)-2(1+2x+4x^2+\cdots) && \text{simple factors}\\ &\quad+2(1+4x+12x^2+\cdots) && \text{squared factor}\\ &=1+5x+17x^2+\cdots && \text{collect powers} \end{aligned}

Why this step? The squared rule contributes 3(2x)2=12x23(2x)^2=12x^2, not 4x24x^2.

Answer. 1+5x+17x21+5x+17x^2 through x2x^2, for ∣x∣<1/2|x|<1/2.

Check. The split recombines because (1−2x)2−2(1−x)(1−2x)+2(1−x)=1(1-2x)^2-2(1-x)(1-2x)+2(1-x)=1. Its denominator is 1−5x+8x2−4x31-5x+8x^2-4x^3; the product with the displayed series has no xx or x2x^2 term.

The opening fractions really were easier to use separately. The striking coefficients 1,3,71,3,7 are 21−1,22−1,23−12^1-1,2^2-1,2^3-1; in fact its coefficient of xnx^n is 2n+1−12^{n+1}-1, directly from the two geometric expansions.

Watch out

Wrong: use the rational expression's whole domain as the series interval. Right: intersect the expansion conditions. A fraction may exist at an input where this power series does not converge.

Change the repeated factor's scale, using the supplied split.

Using the split in an integration step

An antiderivative differentiates back to the integrand. Splitting a fraction lets you apply familiar rules to each piece; we supply those rules here so the focus stays on using the decomposition.

Integration rules

For a differentiable function f\mathrm f, on an interval where f(x)≠0\mathrm f(x)\ne0,

∫f′(x)f(x) dx=ln⁡∣f(x)∣+C,\int\frac{\mathrm f'(x)}{\mathrm f(x)}\,\mathrm dx=\ln|\mathrm f(x)|+C,

and, for n≠−1n\ne-1,

∫f′(x)(f(x))n dx=(f(x))n+1n+1+C.\int\mathrm f'(x)(\mathrm f(x))^n\,\mathrm dx=\frac{(\mathrm f(x))^{n+1}}{n+1}+C.

Use real-valued powers on their domain.

In the formula sheet.

For a≠0a\ne0, these give the following useful cases, not printed separately on the sheet. We supply them for this application:

∫1ax+b dx=1aln⁡∣ax+b∣+C,∫1(ax+b)2 dx=−1a(ax+b)+C.\int\frac1{ax+b}\,\mathrm dx=\frac1a\ln|ax+b|+C, \qquad \int\frac1{(ax+b)^2}\,\mathrm dx=-\frac1{a(ax+b)}+C.

Work on an interval not crossing ax+b=0ax+b=0.

The factor 1/a1/a compensates for the derivative of ax+bax+b. One arbitrary constant covers the sum on each allowed interval.

Worked example 9: integrate simple factors

Question. Integrate 3x/((x−1)(2x+1))3x/((x-1)(2x+1)), given its split 1/(x−1)+1/(2x+1)1/(x-1)+1/(2x+1).

Plan. Apply the logarithm rule to each term, including its denominator coefficient.

∫3x(x−1)(2x+1) dx=∫(1x−1+12x+1)dxuse the split=ln⁡∣x−1∣+12ln⁡∣2x+1∣+Cintegrate terms\begin{aligned} \int\frac{3x}{(x-1)(2x+1)}\,\mathrm dx &=\int\left(\frac1{x-1}+\frac1{2x+1}\right)\mathrm dx && \text{use the split}\\ &=\ln|x-1|+\tfrac12\ln|2x+1|+C && \text{integrate terms} \end{aligned}

Why this step? Differentiating ln⁡∣2x+1∣\ln|2x+1| introduces 22, so its coefficient must be 1/21/2.

Answer. ln⁡∣x−1∣+12ln⁡∣2x+1∣+C\ln|x-1|+\frac12\ln|2x+1|+C, on an interval avoiding 1,−1/21,-1/2.

Check. Differentiating gives 1/(x−1)+1/(2x+1)1/(x-1)+1/(2x+1); recombining gives numerator (2x+1)+(x−1)=3x(2x+1)+(x-1)=3x.

Change one feature: the second factor is squared. It now needs a reciprocal antiderivative as well as a logarithm.

Worked example 10: integrate a repeated factor

Question. Integrate (7x+2)/((x−1)(2x+1)2)(7x+2)/((x-1)(2x+1)^2), using the earlier split.

Plan. Integrate 1/(x−1)−2/(2x+1)+1/(2x+1)21/(x-1)-2/(2x+1)+1/(2x+1)^2 term by term.

∫(1x−1−22x+1+1(2x+1)2)dx=ln⁡∣x−1∣−ln⁡∣2x+1∣−12(2x+1)+Capply both rules\begin{aligned} \int\left(\frac1{x-1}-\frac2{2x+1}+\frac1{(2x+1)^2}\right)\mathrm dx &=\ln|x-1|-\ln|2x+1|\\ &\quad-\frac1{2(2x+1)}+C && \text{apply both rules} \end{aligned}

Why this step? A squared denominator uses the reciprocal rule, not the logarithm rule.

Answer. ln⁡∣x−1∣−ln⁡∣2x+1∣−1/(2(2x+1))+C\ln|x-1|-\ln|2x+1|-1/(2(2x+1))+C, on an interval avoiding 1,−1/21,-1/2.

Check. The derivative of the last term is 1/(2x+1)21/(2x+1)^2. The two logarithms give the other two fractions, already verified by recombination.

For comparison, the monic split from Worked example 1 gives

∫5x+1(x−1)(x+2) dx=2ln⁡∣x−1∣+3ln⁡∣x+2∣+C.\int\frac{5x+1}{(x-1)(x+2)}\,\mathrm dx=2\ln|x-1|+3\ln|x+2|+C.

The absolute values allow either sign of each non-zero argument; an integral is taken on an interval without a denominator zero. Constants can differ on disconnected intervals.

Watch out

Wrong: integrate every denominator using a logarithm. Right: a first power produces a logarithm; a squared linear factor produces a reciprocal. Differentiate your answer to catch missing factors and signs.

Keep the repeated-factor split, with a new coefficient in the last term.

Quick Recap

  • Distinct linear factors each need one constant-numerator fraction.
  • A squared factor needs both its first and second powers; include all possible terms before solving.
  • Clear denominators to obtain a polynomial identity, then substitute factor zeros or compare coefficients and recombine to check.
  • A split makes expansions and integrals simpler; keep the correct series interval and integration domain.

Yes, several fractions can be easier than one: the opening split gives 1+3x+7x2+⋯1+3x+7x^2+\cdots immediately, for ∣x∣<1/2|x|<1/2.