Chapter 6

Transcendental functions

Contents
6.1 Logarithmic and exponential functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.2 Hyperbolic functions and their inverses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6.3 Relative rates of growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

6.1 Logarithmic and exponential functions

Definition 6.1. The natural logarithm is the function ln : (0, 1) ! R defined by
Z x
1
ln x := dt.
1 t

Theorem 6.1 (Properties of natural logarithm). We have:

1. ln x is continuous and differentiable on (0, 1).

d 1
2. ln |x| = for x 6= 0.
dx x
3. ln x is strictly increasing and concave down on (0, 1).
4. For a, b > 0 and r 2 Q, we have

(a) ln ab = ln a + ln b
a
(b) ln = ln a ln b
b
1
(c) ln = ln a
a
(d) ln ar = r ln a

5. lim+ ln x = 1 and lim ln x = 1.

x!0 x!1

6. ln : (0, 1) ! R is surjective.
Proof.
1
1. Follows from the fundamental theorem of calculus upon noting that f (t) = is continuous on (0, 1).
t

61
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 62

2. By Leibniz’s rule, for any x 6= 0,

"Z #
|x|
d d 1 1 1
ln |x| = dt = · (|x|)0 = .
dx dx 1 t |x| x

d 1 d2 1
3. Since ln x = > 0 and 2 ln x = < 0 on (0, 1), the result follows.
dx x dx x2
a 1
4. (a) The function f (x) = ln ax ln x is constant on (0, 1) because f 0 (x) = = 0. Since
ax x
f (1) = ln a and f (b) = ln ab ln b, we get ln a = ln ab ln b.
x 1/b 1
(b) The function f (x) = ln ln x is constant on (0, 1) because f 0 (x) = = 0. Since
b x/b x
a a
f (a) = ln ln a and f (b) = ln 1 ln b = ln b, we get ln ln a = ln b.
b b
1
(c) By the preceding item ln = ln 1 ln b = ln b.
b
rxr 1 r
(d) For r 6= 0, the function f (x) = ln xr r ln x is constant because f 0 (x) = r
= 0. Since
x x
f (1) = 0, we get f (x) = ln x r
r ln x = 0 and the result follows. The case r = 0 is obvious.
1
5. Since ln x is strictly increasing, we have ln < ln 1 = 0 < ln 2. But then
2
✓ ◆n
1 1
lim ln = lim n ln = 1 and lim ln 2n = lim n ln 2 = 1.
n!1 2 n!1 2 n!1 n!1

Since ln x is increasing, these imply

lim ln x = 1 and lim ln x = 1.

x!0+ x!1

6. Since ln x is continuous, intermediate value theorem implies with the aid of 5. that it is surjective.
So the proof is complete. ⇤

d 1
Conclusion 6.1. ln x : (0, 1) ! R is a bijective, strictly increasing, continuous and ln x = 6= 0 on (0, 1).
dx x
Since ln x is a continuous strictly increasing function with ln 1 = 0 and lim ln x = 1, the intermediate
x!1
value theorem implies that the equation ln x = 1 has exactly one solution in the interval (1, 1).
Definition 6.2. We define the number e (Euler’s number) to be the unique solution of the equation ln x = 1 in the
interval (1, 1).
Definition 6.3. The inverse of the natural logarithm function ln x : (0, 1) ! R denoted by exp : R ! (0, 1) is
called the natural exponential function.
Remark 6.1. It is customary to use the notation ex (x 2 R) for exp(x) since, for any rational number r, we have

ln er = r ln e = r · 1 = r =) er = exp(ln er ) = exp(r).

Theorem 6.2 (Properties of the natural exponential function). We have:

d x
1. ex : R ! (0, 1) is a bijective, continuous, strictly increasing, differentiable function with e = ex whose
dx
graph is concave up.
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 63

2. lim ex = 0 and lim ex = 1.

x! 1 x!1

Proof. Since ln x : (0, 1) ! R is a bijective, continuous, strictly increasing function, its inverse ex : R !
d 1
(0, 1) is also a bijective, continuous, strictly increasing function by Theorem ??. Also, since ln x = 6= 0
dx x
on (0, 1), setting f (x) = ln x and f 1 (x) = ex , Theorem 3.6 entails
d x d 1 1 1 1
e = f (x) = = =f (x) = ex .
dx dx f 0 (f 1 (x)) 1
f 1 (x)
This, in turn, implies that
d2 x
e = ex > 0 on R
dx2
so that the graph of ex is concave up.
Since e > 1, in light of Remark 6.1, we have

lim exp( n) = lim e n

= 0 and lim exp(n) = lim en = 1.
n!1 n!1 n!1 n!1

Since exp(x) is increasing, we therefore deduce

lim exp(x) = 0 and lim exp(x) = 1.

x! 1 x!1

This finishes the proof. ⇤

Definition 6.4 (General exponential function). For a > 0, the exponential function with base a is defined as

ax = ex ln a , x 2 R.

Theorem 6.3 (Laws of exponents). For a, b > 0 and p, q 2 R, we have:

1. ap aq = ap+q
2. ap /aq = ap q

3. (ap )q = apq
4. ap bp = (ab)p
Proof.
1. We have
ln(ap aq ) = ln ap + ln aq = ln ep ln a + ln eq ln a = p ln a + q ln a = (p + q) ln a
so that p q
ap aq = eln(a a )
= e(p+q) ln a = ap+q .

2. Similar.
3. This follows from p p ln a
(ap )q = eq ln a = eq ln e = epq ln a = apq .

4. We have

ln(ap bp ) = ln ap + ln bp = ln ep ln a + ln ep ln b = p ln a + p ln b = p (ln a + ln b) = p ln(ab)

so that p p
ap bp = eln(a b )
= ep ln(ab) = (ab)p .
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 64

The proof is thus complete. ⇤

Theorem 6.4. For a > 0, we have:

d x
1. a = ax ln a.
dx
2. If a > 1, ax : R ! (0, 1) is a bijective, continuous, strictly increasing function with

lim ax = 0 and lim ax = 1.

x! 1 x!1

3. If 0 < a < 1, ax : R ! (0, 1) is a bijective, continuous, strictly decreasing function with

lim ax = 1 and lim ax = 0.

x! 1 x!1

4. For a 2 (0, 1) {1}, the graph of ax is concave up.

Proof. Item 1. follows from a direct calculation:

d x d x ln a
a = e = ex ln a ln a = ax ln a.
dx dx
The remaining items are left as exercises. ⇤

Theorem 6.5. For a 2 (0, 1) {1}, the base a exponential function ax : R ! (0, 1) is invertible with a differen-
d x
tiable inverse (since a = ax ln a = ex ln a ln a 6= 0 on R). The inverse function, denoted by loga : (0, 1) ! R,
dx
is called the base a logarithm function and enjoys the following properties for any x, y > 0:
ln x
1. loga x = (in particular loge x = ln x)
ln a
2. loga xy = loga x + loga y
x
3. loga = loga x loga y
y
4. loga 1 = 0
1
5. loga = loga y
y
6. loga xy = y loga x (here y 2 R)
Moreover:
7. If a > 1, then loga x is strictly increasing, its graph is concave down, and

lim loga x = 1 and lim loga x = 1.

x!0+ x!1

8. If 0 < a < 1, then loga x is strictly decreasing, its graph is concave up, and

lim loga x = 1 and lim loga x = 1.

x!0+ x!1

Finally:
9. aloga x = x (x > 0)
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 65

10. loga ax = x (x 2 R)
d 1
11. loga x =
dx x ln a
Proof. We have
ln x ln x
y = loga x =) ay = x =) y ln a = ln x =) y = =) loga x = .
ln a ln a
This establishes 1. The remaining items are left as exercises. ⇤

Definition 6.5. Given r 2 R, the function

xr := er ln x , x > 0,

is called the power function with exponent r.

Theorem 6.6 (Exponentiation rules). For any r, s 2 R and x, y > 0, we have:
1. xr xs = xr+s
xr
2. = xr s
xs
3. (xr )s = xrs
4. xr y r = (xy)r
5. x1 = x
6. x0 = 1
1
7. x r
=
xr
Proof. Exercise. ⇤

Theorem 6.7 (General power rule). For any r 2 R, we have

d r
x = r xr 1
, x > 0.
dx
Proof. By the chain rule
d r d r ln x r r
x = e = er ln x = xr = r xr 1
. = r xr 1
dx dx x x
⇤

Example 6.1. Here are some examples on differentiation.

1. y = ln(sec(ln x)) =) y 0 = ?
Repeated application of the chain rule entails
1 d 1 d
y0 = (sec(ln x)) = sec(ln x) tan(ln x) (ln x)
sec(ln x) dx sec(ln x) dx
1 1 tan(ln x)
= sec(ln x) tan(ln x) = .
sec(ln x) x x
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 66
s
(x + 1)5
2. y = =) y 0 = ?
(x + 2)20

Since
1
ln y = [5 ln(x + 1) 20 ln(x + 2)] ,
2
we have
y0 5 10
= .
y 2(x + 1) x+2
Z p
3 x
3. y = p
ln t dt =) y 0 = ?
x

By Leibniz’s rule
p d p p d p
y 0 = ln( 3 x) ( 3 x) ln x ( x) = . . .
dx dx
4. y = x(x + 1)(x + 2) · · · (x + 100) =) y 0 = ?
Since
ln y = ln x + ln(x + 1) + ln(x + 2) + · · · + ln(x + 100)
we have
y0 1 1 1 1
= + + + ··· +
y x x+1 x+2 x + 100
p
x x2 + 1
5. y = =) y 0 = ?
(x + 1)2/3
Since
1 2
ln y = ln x + ln(x2 + 1) ln(x + 1),
2 3
we have
y0 1 x 2 1
= + 2 .
y x x +1 3x+1
6. y = xx =) y 0 = ?
Since
ln y = x ln x,
we have
y0
= 1 + ln x.
y
Alternatively, using the chain rule

d
y = xx = ex ln x =) y 0 = ex ln x (x ln x) = ex ln x (1 + ln x).
dx

7. y = (ln x)ln x =) y 0 = ?

We have
y0 ln(ln x) 1/x 1 + ln(ln x)
ln y = ln x ln(ln x) =) = + ln x =
y x ln x x
8. y = 2sin 3t =) y 0 = ?
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 67

We have
y0
ln y = sin 3t ln 2 =) = 3 cos 3t ln 2 = ln 8 cos 3t
y
dy
9. ln xy = ex+y =) =?
dx
By implicit differentiation, we have
✓ ◆
y + xy 0 1 1
= ex+y (1 + y 0 ) =) ex+y y 0 = ex+y =) . . .
xy y x

10. Locate and identify the absolute extrema of f (x) = ln(cos x) on the interval [ ⇡/4, ⇡/3].
On the given interval, we have
sin x
f 0 (x) = = tan x = 0 , x = 0.
cos x
Next, note that
p 1
f ( ⇡/4) = ln(cos( ⇡/4)) = ln(1/ 2) = ln 2
2
f (⇡/3) = ln(cos(⇡/3)) = ln(1/2) = ln 2
f (0) = ln(cos 0) = ln 1 = 0.

Example 6.2. Evaluate the following integrals.

Z
1. tan ✓ d✓

We have Z Z
sin ✓
tan ✓ d✓ = d✓
cos ✓
so that the substitution u = cos ✓ [du = sin ✓ d✓] yields
Z Z Z
sin ✓ du
tan ✓ d✓ = d✓ = = ln |u| + C = ln | cos ✓| + C = ln | sec ✓| + C.
cos ✓ u
Z
2. sec ✓ d✓

We have Z Z Z
sec ✓ + tan ✓ sec2 ✓ + sec ✓ tan ✓
sec ✓ d✓ = sec ✓ d✓ = d✓
sec ✓ + tan ✓ sec ✓ + tan ✓
so that the substitution u = tan ✓ + sec ✓ [du = (sec2 ✓ + sec ✓ tan ✓)d✓] yields
Z Z Z
sec2 ✓ + sec ✓ tan ✓ du
sec ✓ d✓ = d✓ = = ln |u| + C = ln | tan ✓ + sec ✓| + C.
sec ✓ + tan ✓ u
Z
sec ✓
3. d✓
ln(sec ✓ + tan ✓)
sec ✓ tan ✓ + sec2 ✓
For u = ln(sec ✓ + tan ✓), we have du = d✓ = sec ✓ d✓ so that
sec ✓ + tan ✓
Z Z
sec ✓ du
d✓ = = ln |u| + C = ln | ln(sec ✓ + tan ✓)| + C.
ln(sec ✓ + tan ✓) u
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 68
Z
dx
4. p
x+x
p
1 dx
The substitution u = 1 + p ] yields
x [du =
2 x
Z Z Z
dx dx du p
p = p p =2 = 2 ln |u| + C = 2 ln |1 + x| + C.
x+x x(1 + x) u
Z
dx
5.
x1/4 +x
3 dx
The substitution u = 1 + x3/4 [du = ] yields
4 x1/4
Z Z Z
dx dx 4 du 4 4
= = = ln |u| + C = ln(1 + x3/4 ) + C.
x1/4 + x x1/4 (1 + x3/4 ) 3 u 3 3
Z ⇡
sin t
6. dt
0 2 cos t
The substitution u = 2 cos t [du = sin t dt] yields
Z ⇡ Z 3 u=3
sin t du
dt = = ln |u| = ln 3.
0 2 cos t 1 u u=1

Z ln(⇡/2)
7. 2 ex cos ex dx
ln(⇡/6)

The substitution u = ex [du = ex dx] yields

Z ln(⇡/2) Z ⇡/2
2 ex cos ex dx = 2 cos u du = . . .
ln(⇡/6) ⇡/6

Z ⇡/4
8. (1 + etan ✓ ) sec2 ✓ d✓
0

The substitution u = tan ✓ [du = sec2 ✓ d✓] yields

Z ⇡/4 Z 1 u=1
(1 + etan ✓ ) sec2 ✓ d✓ = (1 + eu ) du = u + eu = (1 + e) (0 + 1) = e.
0 0 u=0

Z
dx
9.
1 + ex
The substitution u = 1 + ex [du = ex dx = (u 1) dx] yields
Z Z Z ✓ ◆
dx 1 1 1 1
= du = du = ln |u 1| ln |u| + C
1 + ex u u 1 u 1 u
= ln |ex | ln |1 + ex | + C = x ln(1 + ex ) + C.
Z 2
x 2x
10. dx
1 + 2x 2
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 69
2 2 2 2
The substitution u = 1 + 2x = 1 + eln 2 x [du = eln 2 x ln 2 2x dx = ln 4(x 2x ) dx] yields
Z 2 Z
x 2x 1 du 1 1 2 2

x 2 dx = = ln |u| + C = ln |1 + 2x | + C = log4 (1 + 2x ) + C.
1+2 ln 4 u ln 4 ln 4
Z ⇡/4 ✓ ◆tan ✓
1
11. sec2 ✓ d✓
0 3
The substitution u = tan ✓ [du = sec2 ✓ d✓] yields
Z ⇡/4 ✓ ◆tan ✓ Z 1 ✓ ◆u Z 1
1 2 1
sec ✓ d✓ = du = eu ln(1/3) du
0 3 0 3 0
1 u=1 1 2
u ln(1/3)
= e = (eln(1/3) e0 ) = .
ln(1/3) u=0 ln(1/3) 3 ln 3
Z
4x + 1
12. dx
2x2 + 4x + 10
Completing the square in the denominator
4x + 1 4x + 1 4x + 1 1 4x + 1
= = = 2
2x2 + 4x + 10 2
2(x + 2x + 1) + 8 2
2(x + 1) + 8 8 1 + x+1
2

x+1
and then substituting u = , we obtain
2
Z Z Z Z
4x + 1 1 4x + 1 1 4(2u 1) + 1 1 8u 3
2
dx = 2 dx = 2
du = du
2x + 4x + 10 8 1+ 2 x+1 4 1+u 4 1 + u2
Z Z
2u 3 1 3
= du du = ln |1 + u2 | arctan u + C
1 + u2 4 1 + u2 4
✓ ◆2 !
x+1 3 x+1
= ln 1 + arctan + C.
2 4 2

Lemma 6.1 (An important limit). We have

⇣ x ⌘t
lim 1+ = ex for all x 2 R.
t!±1 t
Proof. Fix x 2 R, and define
⇣ x ⌘t
y(t) = 1 + .
t
Then
x/t2
⇣ x⌘ x
⇣ x⌘ ln 1 + 1+ x
lim ln y = lim t ln 1 + = lim t = lim t = lim
t!±1 t!±1 t t!±1 1 L0 H t!±1 1/t 2 t!±1 x =x
1+
t t
so that
lim y(t) = lim eln y(t) = elimt!±1 ln y(t) = ex .
t!±1 t!±1

This completes the proof. ⇤

CHAPTER 6. TRANSCENDENTAL FUNCTIONS 70

Example 6.3. Here are some examples in the same vein.

1. Evaluate lim x1/x .
x!1

We have
ln x 1/x
y = x1/x =) lim ln y = lim = lim =0
x!1 x!1 x LH0 x!1 1
so that
lim y = lim eln y = elimx!1 ln y = e0 = 1.
x!1 x!1

2. Evaluate lim (ln x)1/x .

x!1

We have
  
ln(ln x) ln(ln x) 1
lim (ln x)1/x = lim exp = exp lim = exp lim = exp(0) = 1.
x!1 x!1 x x!1 x 0
LH x!1 x ln x

3. Evaluate lim x1/(x 1)

.
x!1

We have
  
ln x ln x 1/x
lim x1/(x 1)
= lim exp = exp lim = exp lim = exp(1) = e.
x!1 x!1 x 1 x!1 x 1 0
LH x!1 1

4. Evaluate lim (1 + x)1/x .

x!0

Exercise.

6.2 Hyperbolic functions and their inverses

Definition 6.6. The functions

ex e x
sinh x 1
sinh x = tanh x = sech x =
2 cosh x cosh x
ex + e x
cosh x 1
cosh x = coth x = csch x =
2 sinh x sinh x
are called hyperbolic functions.
Remark 6.2 (Why are they called hyperbolic functions?). The parametric equations

x = cos t, y = sin t, 0  t  2⇡,

represent the unit circle

x2 + y 2 = 1.
For 0  t  2⇡, we have

t = angle in radians from the positive x-axis to the point (cos t, sin t)
= length of the arc on the unit circle from (1, 0) to (cos t, sin t)
= twice the area of the sector in the unit circle formed by (0, 0), (1, 0) and (cos t, sin t).

Since
cosh2 t sinh2 t = 1 and cosh t 1,
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 71

the parametric equations

x = cosh t, y = sinh t, x 2 R,
represent the right half of the curve
x2 y2 = 1
called the unit hyperbola. It can be shown that if t 0, then t is equal to twice the area of the region in the first
quadrant bounded by the x-axis, the curve x2 y 2 = 1, and the the line from the origin to (cosh t, sinh t).

3 3

1 1

-1 -1

-3 -3

-4 -2 0 2 4 -3 -1 1 3
sinh x cosh x

-1

-2

-2 -1 0 1 2
tanh x (blue) & coth x (green)

2 2

1 1

0 0

-1 -1

-2 -2

-2 -1 0 1 2 -2 -1 0 1 2
sech x csch x

Figure 6.1: Hyperbolic functions

CHAPTER 6. TRANSCENDENTAL FUNCTIONS 72

Theorem 6.8 (Hyperbolic identities). The following identities hold:

sinh( x) = sinh x sinh(x ± y) = sinh x cosh y ± cosh x sinh y

cosh( x) = cosh x cosh(x ± y) = cosh x cosh y ± sinh x sinh y
cosh x + sinh x = ex sinh 2x = 2 sinh x cosh x
cosh x sinh x = e x cosh 2x = cosh2 x + sinh2 x = 2 sinh2 x + 1 = 2 cosh2 x 1
cosh2 x sinh2 x = 1 1 tanh2 x = sech2 x and coth2 x 1 = csch2 x
cosh 2x 1
sinh2 x =
2
cosh 2x + 1
cosh2 x =
2
Theorem 6.9 (Derivatives of hyperbolic functions). The following identities hold:

d d d
sinh x = cosh x cosh x = sinh x tanh x = sech2 x
dx dx dx
d d d
coth x = csch2 x sech x = sech x tanh x csch x = csch x coth x
dx dx dx
Remark 6.3. The domains and ranges of hyperbolic functions are as follows:

sinh : R ! R cosh : R ! [1, 1) tanh : R ! [ 1, 1]

coth : R ! R [ 1, 1] sech : R ! (0, 1] csch : R {0} ! R {0}

When properly restricted, they are bijections:

sinh : R ! R cosh : [0, 1) ! [1, 1) tanh : R ! ( 1, 1)

1
coth :R {0} ! R [ 1, 1] sech : [0, 1) ! (0, 1] csch : R {0} ! R {0}

Definition 6.7. The inverses of these functions

1 1 1
sinh :R!R cosh : [1, 1) ! [0, 1) tanh : ( 1, 1) ! R

1 1 1
coth :R [ 1, 1] ! R {0} sech : (0, 1] ! [0, 1) csch :R {0} ! R {0}

are called inverse hyperbolic functions.

Theorem 6.10. We have
1 1
1. sech x = cosh (1/x), 0 < x  1
1 1
2. csch x = sinh (1/x), x 6= 0
1 1
3. coth x = tanh (1/x), x 2 R [ 1, 1]
Proof. These follow from the definitions. For instance
✓ ◆
1 1 1 1
sech cosh = 1 1 = 1 =x
x cosh cosh x x

proves 1. ⇤
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 73

8
5

0 4

-5
0

-5 0 5 0 4 8
sinh-1 x cosh -1 x

2 2

1 1

0 0

-1 -1

-2 -2

-2 -1 0 1 2 -2 -1 0 1 2
-1 -1
tanh x coth x

2
2

1 0

0
-2

0 1 2 -2 0 2
sech-1 x csch -1 x

Figure 6.2: Inverse hyperbolic functions

Theorem 6.11 (Derivatives of inverse hyperbolic functions). We have:

d 1 1 d 1 1
sinh x= p (x 2 R) cosh x= p (x > 1)
dx 1 + x2 dx x2 1

d 1 1 d 1 1
tanh x= (|x| < 1) coth x= (|x| > 1)
dx 1 x2 dx 1 x2
d 1 1 d 1 1
sech x= p (0 < x < 1) csch x= p (x 6= 0)
dx x 1 x2 dx |x| 1 + x2
Proof.
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 74
1
1. Setting f (x) = sinh x and f 1
(x) = sinh x, we have
d 1 d 1 1 1
sinh x= f (x) = = 1 .
dx dx f 0 (f 1 (x))
cosh(sinh x)
p
Since cosh2 x sinh2 x = 1, we have cosh x = 1 + sinh2 x so that
q p
cosh(sinh 1 x) = 1 + sinh2 (sinh 1 x) = 1 + x2 .

2. Similar.
1
3. Setting f (x) = tanh x and f 1
(x) = tanh x, we have
d 1 d 1 1 1
tanh x= f (x) = = .
dx dx f 0 (f 1 (x))
sech2 (tanh 1
x)

Since sech2 x = 1 tanh2 x, we have

sech2 (tanh 1
x) = 1 tanh2 (tanh 1
x) = 1 x2 .

4. Similar.
1
5. Setting f (x) = sech x and f 1
(x) = sech x, we have
d 1 d 1 1 1 1
sech x= f (x) = = 1 1 = 1 .
dx dx f 0 (f 1 (x))
sech(sech x) tanh(sech x) x tanh(sech x)
p
Since 1 tanh2 x = sech2 x, we have tanh x = 1 sech2 x so that
q p
tanh(sech 1 x) = 1 sech2 (sech 1 x) = 1 x2 .

6. Similar.
So the proof is complete. ⇤

6.3 Relative rates of growth

Definition 6.8. Let f (x) and g(x) be positive for all sufficiently large x.
1. f grows faster than g as x ! 1 if

f (x) g(x)
lim =1 or equivalently lim = 0.
x!1 g(x) x!1 f (x)
In this case we also say that g grows slower than g as x ! 1.
2. f and g grow at the same rate as x ! 1 if

f (x)
lim =L with 0 < L < 1.
x!1 g(x)

Remark 6.4. The followings are important.

1. ex grows faster than xc for any c 2 R.
CHAPTER 6. TRANSCENDENTAL FUNCTIONS 75

2. ex grows faster than any polynomial.

3. xc grows faster than ln x for any c > 0.
4. a > b > 0 implies ax grows faster than bx .

5. a, b 2 (0, 1) {1} implies | loga | and | logb x| grow at the same rate.
6. Polynomials of the same power grow at the same rate.
p p
Example 6.4. f (x) = x2 + 5 and g(x) = (2 X 1)2 grow at the same rate.