Cosh

cosh and sinh
The hyperbolic functions cosh and sinh are defined by

ex + e−x
(1) cosh x =
2
ex − e−x
(2) sinh x =
2
ex +e−x x −x ex −e−x
We compute that the derivative of 2 is e −e 2
and the derivative of 2
is
ex +e−x
2
, i.e.
d
(3) cosh x = sinh x
dx
d
(4) sinh x = cosh x
dx
Note that sinh x > 0 for x > 0, and sinh x < 0 for x < 0. However cosh x ≥ 0
for all x (strictly positive away from 0). sinh x is increasing for all x. cosh x is
increasing for all x > 0 (and decreasing for x < 0). Note that cosh(x) = cosh(−x)
and sinh(−x) = − sinh(x). The minimum of cosh x is attained at x = 0 where
cosh(0) = 1, thus cosh(x) ≥ 1 for all x.
Draw pictures:
1
2
The inverse of sinh

sinh x is (strictly) increasing and limx→∞ sinh(x) = ∞ and limx→−∞ sinh(x) = −∞.
We see that the range of sinh is (−∞, ∞) and sinh is invertible. Then
sinh : (−∞, ∞) → (−∞, ∞)
sinh−1 : (−∞, ∞) → (−∞, ∞)
Let us compute the inverse. That is, we consider the equation sinh(x) = y, and
x −x
express x in terms of y. This means we need to solve for x in e −e 2
= y. To do
this we first set w = ex , determine w and then take the natural logarithm of w. The
equation for w becomes w − w−1 = 2y or w2 − 2yw −p 1 = 0. By the quadraticpformula
there are two possibilities for w, namely w = y + y 2 + 1 and w = y − y 2 + 1.
The first solution for w is positive, the second is negative, and since w =p ex has to be
x
positive we can discard the negative solution for w.pHence e = w = y + y 2 + 1 and
after taking the logarithm we see that x = ln(y + y 2 + 1). We have thus computed
the inverse function for sinh and it is given by
p
(5) sinh−1 (y) = ln(y + y 2 + 1).
In some of the European
p literature this inverse function is denoted by Arsinh, hence
Arsinh(y) = ln(y + y 2 + 1) . 1
The derivative of sinh−1

We could use the general formula for the derivative of inverse functions, or just the
above formula for sinh−1 , let’s do the latter.
I am now writing x for the independent variable. Using the chain rule we get
d¡ √ ¢ 1 d¡ √ ¢
ln(x + x2 + 1) = √ x + x2 + 1
dx x + x + 1 dx
2
d
¡ ¢
We note that dx x + (x2 + 1)1/2 = 1 + 12 (x2 + 1)−1/2 2x and get that the last displayed
expression is equal to
³ ´ √
1 x 1 x2 + 1 + x 1
√ 1+ √ = √ √ =√
x + x2 + 1 x2 + 1 x + x2 + 1 x2 + 1 x2 + 1
Therefore we get the rule
d 1
(6) Arsinh0 (x) ≡ sinh−1 (x) = √ .
dx x2 + 1
For the integral this is
Z √
1
√ dx = ln(x + x2 + 1) + C .
x2 + 1
1Arsinhstands for Area sinus hyperbolicus. I have not seen this in any American textbook. Use
sinh−1 instead but then make sure that you do not confuse it with the reciprocal of sinh y.
3
The inverse of cosh

As a function on the real line cosh does not have an inverse (note that cosh(x) =
cosh(−x) so that two different points in x correspond to the same value of cosh).
However if we restrict the domain to [0, ∞) then cosh is strictly increasing and
invertible. The range of cosh is [1, ∞) so that we have
cosh : [0, ∞) → [1, ∞)
cosh−1 : [1, ∞) → [0, ∞)
We compute cosh−1 (y) for y ≥ 1. Thus, for each y ≥ 1 we wish to determine
an x ≥ 0 so that cosh x = y, or equivalently (ex + e−x )/2 = y. To determine x
we again first determine w = ex from the equation w + w−1 = 2y, or equivalently,
w2 −p2yw + 1 = 0. This quadratic equation has two solutions, namely w could be
2
y ± y 2 − 1. Thep possibility of the minus sign can be discarded since a calculation
shows that y − y 2 − 1 is p < 1 for all y ≥ 1, and therefore it can not be an ex for
some x > 0. Thus w = y + y 2 − 1 and ex = w, so cosh−1 y = x = ln w. We get the
formula
p
(7) cosh−1 (y) = ln(y + y 2 − 1).
Again this inverse function is
p occasionally denoted by Arcosh, so we may write
sometimes Arcosh(y) = ln(y + y 2 + 1).
The derivative of cosh−1

Again I am now writing x for the independent variable (it is now restricted to
x > 1). The calculation is completely analogous to the calculation for the derivative
of sinh−1 . By the chain rule we get
d¡ √ ¢ 1 d¡ √ ¢
ln(x + x2 − 1) = √ x + x2 − 1
dx x + x2 − 1 dx
and we now calculate that this yields √ 1 . Hence we get the rule
x2 −1
d 1
(8) Arcosh0 (x) ≡ cosh−1 (x) = √ .
dx x2 − 1
This also yields (for x > 1)
Z √
1
√ dx = ln(x + x2 − 1) + C .
x2 − 1
p √ 2 2
y 2 −(
2Indeed y− y2 − 1 = √y2 −1) = √1 2 < 1 for y ≥ 1 .
y+ y −1 y+ y −1
4
Problems
1. Prove the following identities.

(i) (cosh x)2 − (sinh x)2 = 1 .
(ii) cosh(2x) = (cosh x)2 + (sinh x)2 .
(iii) sinh(2x) = 2 sinh x cosh x .
2. Compute
Rx the integrals
(i) R0 (a2 − t2 )−1/2 dt for |x| < a.
x
(ii) R0 (a2 + t2 )−1/2 dt for all x.
x
(iii) R0 (a2 − t2 )1/2 dt for |x| < a.
x
(iv) 0 (a2 + t2 )1/2 dt for all x.
3. The function tanh is defined by

sinh x
tanh x =
cosh x
(i) Show that tanh is defined and differentiable for all x and show that its derivative
is given by
1
tanh0 (x) = .
cosh2 x
(ii) Show that the range of tanh is the interval (−1, 1) and that
tanh : (−∞, ∞) → (−1, 1)
is invertible.
(iii)Prove that the inverse function
tanh−1 : (−1, 1) → (−∞, ∞)
¡ ¢
is given by tanh−1 (y) = 12 ln(1 + y) − ln(1 − y) .
(iv) Sketch the graph of tanh and the graph of tanh−1 .

Cosh

Uploaded by

Copyright:

Available Formats

Cosh

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cosh

Uploaded by

Copyright:

Available Formats

cosh and sinh

The hyperbolic functions cosh and sinh are defined by

The inverse of sinh

The derivative of sinh−1

The inverse of cosh

The derivative of cosh−1

1. Prove the following identities.

3. The function tanh is defined by

You might also like