8.

1 L’Hôpital’s Rule
Actually, L’Hôpital’s Rule
was developed by his teacher
Johann Bernoulli. De
L’Hôpital paid Bernoulli for
private lessons, and then
published the first Calculus
book based on those lessons.
Guillaume De l'Hôpital
1661 - 1704
 8.1 L’Hôpital’s Rule
x2  4
Consider: lim If we try to evaluate this by 0
x2 x  2
direct substitution, we get: 0
Zero divided by zero can not be evaluated, and is an example
of indeterminate form.
In this case, we can evaluate this limit by factoring and
canceling:

lim
x2  4
 lim
 x  2  x  2  lim  x  2   4
x2 x  2 x 2
x2 x2
 8.1 L’Hôpital’s Rule
f  x x2  4 The limit is the ratio of
lim  lim
xa g  x x2 x  2
the numerator over the
denominator as x approaches 2.
x2  4

If we zoom in far enough,

the curves will appear as
straight lines.

x2
8.1 L’Hôpital’s Rule
f  x x2  4
lim  lim As x2
xa g  x  x2 x  2

df f  x
becomes:
g  x
dg
dx
df
df  dx
dg dg
dx
 8.1 L’Hôpital’s Rule
d 2
f  x x 4
2
dx
 x  4
2x
lim  lim  lim  lim 4
xa g  x x2 x  2 x 2 d x 2 1
 x  2
dx
8.1 L’Hôpital’s Rule

L’Hôpital’s Rule:
f  x
If lim
xa g  x 
is indeterminate, then:

f  x f  x
lim  lim
x a g  x  x a g  x
 8.1 L’Hôpital’s Rule
We can confirm L’Hôpital’s rule by working backwards, and
using the definition of derivative:

f  x  f  a f  x  f  a
f  a lim

xa xa  lim xa
g a  g  x  g  a x a g  x   g  a 
lim
x a xa xa

f  x  f  a f  x  0 f  x
 lim  lim  lim
x a g  x   g  a  xa g  x   0 xa g  x 
 8.1 L’Hôpital’s Rule
Example:

1  cos x  lim sin x  0

lim x 0 1  2 x
x 0 x  x 2

If it’s no longer
indeterminate, then STOP!

If we try to continue with L’Hôpital’s rule:

sin x cos x 1 which is wrong,

 lim  lim 
x 0 1  2 x x 0 2 2 wrong, wrong!
 8.1 L’Hôpital’s Rule
On the other hand, you can apply L’Hôpital’s rule as many
times as necessary as long as the fraction is still
indeterminate:
x 1 1
1
 1 x 

1  x 1  0 2
0
lim 2  lim 2 2
x 0 x2 0 x 0 2x 0
1
1 1 3

 1 x 1 x   1 x 2

2
0
lim 2  lim 4 not
x 0 x2 x 0 2 0
1
 1
 4 
2 8
 8.1 L’Hôpital’s Rule
L’Hôpital’s rule can be used to evaluate other indeterminate
0
forms besides .
0
The following are also considered indeterminate:

0  1
0 0
0

 0
The first one, , can be evaluated just like .
 0

The others must be changed to fractions first.

 8.1 L’Hôpital’s Rule

lim  x sin 
1
This approaches   0
x 
 x
1
sin
lim x 0
x  1 This approaches
x 0
1 1  1 
sin cos      2 
x  x   x   lim cos  1 
lim  lim    cos  0   1
x  1 x  1 x 
 x
 2
x x
 8.1 L’Hôpital’s Rule
Indeterminate Forms: 1 00 0

lim f  x   L lim f  x   lim e

ln f  x 
  eL
xa x a x a

1/ x 1 1
lim x  0 lim ln  x  L’Hôpital
x  e x x lim x applied
e x 1


lim ln x1/ x  ln  x 
e x
e
lim
x x e 0 1


 8.2 Relative Rates of Growth
The function y  e x

grows very fast. We have

At 64 gonethe
inches, lessy-value
than half-
way
wouldacross
be atthe
theboard
edge of the
If x is 3 inches,  3, 20  horizontally,
known universe!
value
and already the y-
(10.5 would
billion reach the
light-years)
y is about 20 Andromeda Galaxy!
inches:

At x  10 inches,
At x  44 inches,
1
y mile y  2 million light-years
3
 8.2 Relative Rates of Growth
The function
y = ln x grows By the time we reach the edge
very slowly. of the universe again (10.5
billion light-years) the chalk
We would have to
move 2.6 miles to the line will only have reached 64
right before the line inches!
moves a foot above
the x-axis!

The function y = ln x increases everywhere, even though it

increases extremely slowly.
 8.2 Relative Rates of Growth
Let f (x) and g(x) be positive for x sufficiently large.

x
Definitions: Faster, Slower, Same-rate Growth as
1. f grows faster than g (and g grows slower than f )
as x  if
f  x g  x
lim  or lim 0
x  g  x x  f  x
2. f and g grow at the same rate as x if 
f  x
lim L0
x  g  x
 8.2 Relative Rates of Growth

WARNING
Please temporarily suspend your common sense.
 8.2 Relative Rates of Growth
According to this definition, y = 2x does not grow faster than
yx 2x
lim  lim 2  2
x  x x 

Since this is a finite non-zero

limit, the functions grow at the
same rate!

The book says that “f grows faster

yx than g” means that for large x
values, g is negligible compared to f.
y  2x
 8.2 Relative Rates of Growth
x 2
Which grows faster, eor x?

ex  Weiscan
This confirm thissographically:
indeterminate, we
lim 2 
x  x  apply L’Hôpital’s rule.

ex
 ex
lim  Still indeterminate. y 2
x  2 x  x

ex
lim 
x  2

x 2
e grows faster than x.
 8.2 Relative Rates of Growth

“Growing at the same rate” is transitive.

In other words, if two functions grow at the same rate as a

third function, then the first two functions grow at the same
rate.
 8.2 Relative Rates of Growth

Show that f  x   x  5 and g  x    2 x  1 grow

2
2

at the same rate as x  .

 8.2 Relative Rates of Growth
Let h  x   x

x  5  lim
2 x2  5 x2 5 50
lim  lim 2  2  lim 1  2  1
x  x x 
x2 x  x x x  x

   
2 2
2
2 x 1 2 x 1 2 x 1 0
lim  lim  lim     4
 x
2
x  x x  x 
 x x

f  f h 1 1 f and g grow at the

lim  lim     1  
x  g x  h g
  4 4 same rate.
 8.2 Relative Rates of Growth
Definition f of Smaller Order than g

Let f and g be positive for x sufficiently large. Then f

is of smaller order than g as x   if
f  x
lim 0
x  g  x
We write f  o  g  and say “f is little-oh of g.”

Saying f  o  g  is another way to say that f grows

slower than g.
 8.2 Relative Rates of Growth
Definition f of at Most the Order of g

Let f and g be positive for x sufficiently large. Then f

is of at most the order of g as x   if there is a
positive integer M for which
f  x
M for x sufficiently large
g  x
We write f  O  g  and say “f is big-oh of g.”
Saying f  O  g  is another way to say that f grows
no faster than g.
 8.3 Improper Integrals

Until now we have been finding integrals of continuous

functions over closed intervals.

Sometimes we can find integrals for functions where

the function or the limits are infinite. These are called
improper integrals.
 8.3 Improper Integrals

1 1 x The function is

0 1 x
dx
undefined at x = 1 . Can we find
the area under
Since x = 1 is an asymptote, the an infinitely
function has no maximum. high curve?
We could define this integral as:

b 1 x
lim  dx (left hand limit)
b 1 0 1 x We must approach the limit from
inside the interval.
 8.3 Improper Integrals
b 1 x 1 x
lim 
b 1 0 1 x
dx  1  x2
dx  
1  x2
dx

1 x 1 x
 1 x 1 x
dx
u  1  x2
du  2 x dx
1+x
 1  x2
dx 1
 du  x dx
2
1
1 
sin 1 x   u 2 du
2
 8.3 Improper Integrals
1 x u  1  x2
 1  x2
dx  
1  x2
dx
du  2 x dx
1 
1
1
sin x   u du
1 2  du  x dx
21 2
sin 1 x  u 2
This integral converges
b
lim sin 1 x  1  x 2 because it approaches a
b 1 solution.
 0

0 0
 2
lim sin b  1  b 2  sin 1 0  1
b 1
1
   

2
1
 8.3 Improper Integrals
dx 1
1
lim 
1
0 x b 0 b x
dx

1
lim ln x b lim ln1  ln b
b 0 b 0

1
lim ln 
b 0 b
This integral diverges.
 8.3 Improper Integrals
3 dx The function
0 2 approaches 
 x  1 3
when x  1 .
3 2

  x  1

3 dx
0

b 2 3 2
lim   x  1 dx  lim   x  1
 
3 3 dx
b 1 0 c 1 c

1 b 1 3
lim 3  x  1 3  lim 3  x  1 3
b 1 c 1
0 c
 8.3 Improper Integrals
b 2 3 2
lim   x  1 dx  lim   x  1
 
3 3 dx
b 1 0 c 1 c

1 b 1 3
lim 3  x  1 3  lim 3  x  1 3
b 1 c 1
0 c

0  1

0
 1
 1 1
lim 3  b  1  3  1   lim 3  2  3  c  1 
3 3 3 3
b 1   c 1  

3  33 2
 8.3 Improper Integrals
b  P 1 1 P 1
 dx P0 lim 
 1 x P
b   P  1 P  1
(P is a constant.) What happens here?

1
x  P dx If P  1 then b  P 1
gets bigger
and bigger as b   , therefore
b the integral diverges.
lim  x P
dx
b  1
b
1 If P  1 then b has a negative
 P 1
lim x exponent and b  P 1  0 ,
b   P  1
1 therefore the integral converges.
8.3 Improper Integrals


x
e dx Converges
1

b
x b
lim  e dx x
lim e
b  1 b  1

0
lim  e  e
b 
b
 1
 lim 
b 
1 1
e b

e

1
e
8.3 Improper Integrals


 x2
Does e dx converge?
1

Compare:
1 1
to for positive values of x.
2 x
e x
e

x2 1 1
For x  1, e e   x
x
x2
e e
 8.3 Improper Integrals
x2 1 1
For x  1, e e   x
x
x2
e e
1 1
Since x2 is always below x ,
e e
we say that it is “bounded above” by
1
x
e
1 1
Since x converges to a finite number, x2 must also converge!
e e
 8.3 Improper Integrals
Direct Comparison Test:

Let f and g be continuous on  a,   with 0  f  x   g  x 

for all xa , then:

 
1  f  x
a
dx converges if  g  x
a
dx converges.

 
2  g  x
a
dx diverges if  f  x
a
dx diverges.
 8.3 Improper Integrals

 sin 2 x The maximum value of sin x  1 so:


1 x 2
dx
2
sin x 1
0 2
 2 on
on  1,  
x x

1 sin 2 x
Since 2 converges, 2 converges.
x x
8.3 Improper Integrals
 1

1
x 2  0.1
dx

x 2  0.1  x for positive values of x, so:

1 1
 on  1,  
x  0.1 x
2

1 1
Since diverges, diverges.
x x  0.1
2
 8.3 Improper Integrals
If functions grow at the same
 dx rate, then either they both
Does 
1 1 x 2 converge? converge or both diverge.

As x   the “1” in the denominator becomes

1
insignificant, so we compare to 2 .
x
1 1
1  x 2 Since 2 converges,
lim x
2
 lim 2  lim 2 x  1 1 x
x  1 x  x x  2 x converges.
1  x2 1  x2
 8.3 Improper Integrals
 dx dx

b
 lim 
1 1  x2 b  1 1 x 
2

1
b 2 4
 lim tan x  lim tan b  tan 1 1 1
b  1 b 
 
4 2
  
  
2 4 4

As y  , x 
2 y  tan x
 8.3 Improper Integrals
 dx  1
1 1 x 2 
1 x 2
dx

b dx b
 lim   lim  x 2 dx
b  1 1  x2 b  1

b b
1
 lim tan x  lim  x 1
b  1 b  1

 lim tan 1 b  tan 1 1 0

b   1 1
    lim      1
  
b  b  1
2 4 4
 8.4 Partial Fractions
5x  3 This would be a lot easier if we could
 x 2  2 x  3 dx re-write it as two separate terms.

5x  3 A B
  Multiply by the common
 x  3  x  1 x  3 x 1 denominator.
5 x  3  A  x  1  B  x  3

5 x  3  Ax  A  Bx  B  3 Set like-terms equal to

each other.
5x  Ax  Bx 3  A  B  3
3  A  3B Solve two equations with
5  A B two unknowns.
 8.4 Partial Fractions
5x  3 5  A B 3  A  3B
 x 2  2 x  3 dx 3   A  3B
5x  3 A B 8  4B
  5  A 2
 x  3  x  1 x  3 x  1 2B
3 2 3 A
5 x  3  A  x  1  B  x  3  x  3  x  1 dx
5 x  3  Ax  A  Bx  B  3
5x  Ax  Bx 3  A  B  3 3ln x  3  2 ln x  1  C
Solve two equations with
5  A B 3  A  3B This technique is called
two unknowns.
Partial Fractions
 8.4 Partial Fractions

Good News!

The AP Exam only requires non-repeating linear factors!

The more complicated methods of partial fractions are good

to know, and you might see them in college, but they will
not be on the AP exam or on my exam.
 8.4 Partial Fractions
6x  7 A B Repeated roots: we must use
 
 x  2 x  2  x  2 2 two terms for partial
2

fractions.
6x  7  A  x  2  B 7  12  B

6 x  7  Ax  2 A  B
6 5
6x  Ax 7  2A  B 
x  2  x  2 2
6 A 7  26  B

5  B
 8.4 Partial Fractions
2 x3  4 x 2  x  3 If the degree of the numerator is
higher than the degree of the
x2  2x  3 denominator, use long division first.
5x  3
2x 2x 
x 2  2 x  3 2 x3  4 x 2  x  3  x  3  x  1
2 x3  4 x 2  6 x 3 2
 2x  
5x  3  x  3  x  1
5x  3
2x  2
x  2x  3
(from example one)
 8.4 Partial Fractions
2 x  4 Ax  B C D
 2  
x 2
 1  x  1
2
x  1 x  1  x  1 2

irreducible
repeated root
quadratic
factor

 
2 x  4   Ax  B   x  1  C x 2  1  x  1  D x 2  1
2
 
   
2 x  4   Ax  B  x 2  2 x  1  C x 3  x 2  x  1  Dx 2  D

2 x  4  Ax3  2 Ax 2  Ax  Bx 2  2 Bx  B  Cx3  Cx 2  Cx  C  Dx 2  D
 8.4 Partial Fractions
dx These are in
42

a xx 22

 4  x2
the same
form. x

2sec  d
2

 2sec 2a

4  x2 x
 sec d sec 
2
tan  
2
ln sec  tan   C
2 tan   x
2sec  4  x 2

4 x 2
x
ln  C 2sec 2  d  dx
2 2
 8.4 Partial Fractions
dx
 4 x 2 ln
4  x2  x
C
2
2sec 2  d
 2sec
ln 4  x 2  x  ln 2  C

 sec d
This is a constant.

ln sec  tan   C ln 4  x2  x  C

4  x2 x
ln  C
2 2
 8.4 Partial Fractions
a2  x2
If the integral contains a 2  x 2, x

we use the triangle at right.
a
If we need a 2  x,2 we If we need x 2  a,2 we

move a to the hypotenuse. move x to the hypotenuse.

a x
x
 x2  a2

a2  x2 a
 8.4 Partial Fractions
x 2 dx 3 x 9  x2
 9  x2 x sin   3
cos  
3

9  x2 3sin   x 3cos   9  x 2
9sin 2   3cos  d 3cos  d  dx
 3cos  x 1 x
sin     sin
1  cos 2 9 3 3
9
2
d
2  1  cos 2 d
9 1 x 9 x 9  x 2
9 9 1 sin    C
   sin 2  C 2 3 2 3 3
2 2 2
 8.4 Partial Fractions
x 2 dx 3
 9 x 2

x

9  x2
9 9 1
   sin 2  C
2 2 2

9 1 x 9
sin   2sin  cos   C
9 1  x  x 2 3 4
sin    9  x2  C
2 3 2 9 1 x 9 x 9  x 2
sin    C
2 3 2 3 3
 8.4 Partial Fractions
dx We can get 2x  x 2 into the necessary form
 2x  x2 by completing the square.

2x  x 2 dx sin   u

1   x  1
 
2
 x  2x 2
cos  d  du

 x2  2x  1  1 
  x  1  1
2

1   x  1
2
 8.4 Partial Fractions
dx dx
 
1   x  1
2
2x  x 2

du 1
 1 u2 Let u  x  1 u

du  dx
cos  d 1 u2
 cos
1 u2
cos   1 u2
1
 d
1
 C  sin u C
sin   u cos  d  du
 sin 1  x  1  C
 8.4 Partial Fractions
dx Complete the square:
 4x2  4x  2
4 x2  4 x  2
dx 4x2  4x  1  1
  2 x  1 2
1
 2 x  1
2
1
 8.4 Partial Fractions
dx u2 1
 4x2  4x  2 Let u  2 x  1
u
du  2 dx 
dx
  2 x  1 2
1
1
du  dx 1
2 tan   u
1 du
2  u2 1
sec 2  d  du
sec  u 2  1
1 sec 2  d 1 1
2
   
2  sec 2 
d C sec 2
  u 2
1
2
1 1
 1
tan u  C  tan 1  2 x  1  C
2 2
 8.4 Partial Fractions
Here are a couple of shortcuts that are result from
Trigonometric Substitution:

du 1 1 u
 u 2  a 2  a tan a  C
These are on your list of
formulas. They are not
du u really new.

1
 sin C
a u
2 2 a