We saw that (i) dxd(sin−1x)=1−x21
(ii) dxd(tan−1x)=1+x21
We know that sin−1x+cos−1x=2π
tan−1x+cot−1x=2π
sec−1x+cosec−1x=2π
∴cos−1x=2π−sin−1x
⇒dxd(cos−1x)=0−dxd(sin−1x)=1−x2−1
Il y, dxd(cot−1x)=−dxd(tan−1x)=−1+x21
Let's calculate dxd(sec−1x).
Let y=sec−1x Then x=secy
Diff. w.r.t. x, we get
1=dxd(secy)=dyd(secy)⋅dxdy
(by chain rule)
=secytanydxdy
⇒dxdy=secytany1
Since, secy=x,tan2y=sec2y−1=x2−1
⇒tany=±x2−1
Recall : sec−1x∈(0,2π) if x∈(1,∞)
2sec−1x∈(2π,π) if x∈(−∞,−1)
( sec−1x is not defined if −1<x<1 )
∴y=sec−1x∈(0,2π) if x∈(1,∞)
∈(2π,π] if x∈(−∞,−1)
⇒tany⩾0 if x∈(1,∞)
⩽0 if x∈(−∞,−1)
tany={x2−1 if x⩾1x2−1 if x⩽−1
tan(sec−1x)
∴dxd(sec−1x)=secytany1={xx2−11 if x⩾1, −xx2−11 if x⩽1
Since ∣x∣=x if x⩾0or−x if x<0
Sometimes we have am equation relating y to x , but it is difficult to write explicitly y as a fr. of x'
For Example: y+sin=x
Find =dxdy
Here instead of trying to write y as a function of x, we will do implicit differentiation as follows.
y + sin y =x
⇒dxd=(y+siny)=dxd(x)
⇒dxdy+dxd(siny)=1
⇒dxdy+dydsinydxdy=1
⇒dxdy[1+cosy]=1
⇒dxdy=1+cosy1 , provided cosy=−1
Note that by doing have, we need not get dxdy as a function of x only.
y=exp(x)
Putting x=1, we get
exp(1)=1+1!1+2!1+⋯+n!1+⋯
=∑k=0∞k!1
We know that exp(1) is a real, and we denote this by e (Euler's constant).
So, e=exp(1)=∑k=0∞k!1
We can show that 2<e<3 (In fact, e≈2.718…
e=1+1!1+2!1+⋯>1+1!1=2 .
Also, e=1+1!1+2!1+3!1+4!1+⋯
<1+1+21+221+231+⋯
(because (n+1)!1<2n1 for n⩾2 )
Recall: Geometric series
a+ar+ar2+⋯=1−ra if ∣r∣<1
Putting a=1,r=21 gives
1+21+221+⋯=1−1/21=2
∴e<1+2=3
Fact: It can be proved that e is an irrational number.
Also,
limn→∞(1+n1)n=e.
limx→0(1+x)1/x=e.
Let’s calculate limh→0hexp(h)−1
Notation: We write exp(x) as ex.
eh=exp(h)=1+1!h+2!h2+⋯+n!hh+⋯
⇒eh−1=h+2!h2+3!h3+⋯+n!hh+⋯
⇒heh−1=1+2!h+3!h2+⋯+n!hh−1+⋯ for h=0.
limh→0heh−1=1
Derivative of ex=exp(x)
f(x)=ex
f′(x)=limh→0hf(x+h)−f(x)
=limh→0hex+h−ex=limh→0hexeh−ex
=exlimh→0(heh−1)=ex.
(1) dxd(e5x)=d(5x)d(e5x)⋅dxd(5x)=e5x⋅5
(2) dxd(ex2)=ex2⋅dxd(x2)=2xex2.
(3) dxdtan−1(ex)=1+(ex)21⋅dxd(ex)=1+e2xex