How do you evaluate the integral #int 1/(x-1)^(2/3)dx# from 0 to 2?

2 Answers
Jan 22, 2017

To evaluate #int_0^2 1/(x-1)^(2/3)dx#, the integrand, i.e.,

#1/(x-1)^(2/3)# has to be continuous over the interval #[0,2]#.

Here, it is not so at #x=1 in [0,2]#.

Jan 23, 2017

The integrand is not defined at #x = 1#, so this is an improper integral.

Explanation:

We need to break this into two improper integrals and try to evaluate each of them. If both integrals converge, then we can add the values to get the integral on #[0,2]#

#int_0^2 1/(x-1)^(2/3) dx = int_0^1 1/(x-1)^(2/3) dx+int_1^2 1/(x-1)^(2/3) dx#

provided that both integrals on the right converge,

# = lim_(brarr1^-)int_0^b 1/(x-1)^(2/3) dx + lim_(ararr1^+)int_a^2 1/(x-1)^(2/3) dx#

provided that both integrals on the right converge.

#lim_(brarr1^-)int_0^b 1/(x-1)^(2/3) dx ={: lim_(brarr1^-) 3(x-1)^(1/3) ]_0^b = 3#

and

#lim_(ararr1^+)int_a^1 1/(x-1)^(2/3) dx ={: lim_(ararr1^+) 3(x-1)^(1/3) ]_a^2 = 3#

We conclude:

#int_0^2 1/(x-1)^(2/3) dx = int_0^1 1/(x-1)^(2/3) dx+int_1^2 1/(x-1)^(2/3) dx#

# = 3+3=6#