The Student Room Group

Finding all values in C of arcsech(i)

Hello, I was wondering if anyone could check my solution for this problem, The solution follows as:

First note that sech1(i)=log(1z+(1z21)12)sech^{-1}(i)=\log(\frac{1}{z}+(\frac{1}{z^2}-1)^\frac{1}{2}). So sech1(i)=log(i(1±2)sech^{-1}(i)=\log(i(-1\pm \sqrt{2}), this means that we have two cases due to the ±\pm so taking the positive case first we have that sech1(i)=log(i(1+2))sech^{-1}(i)=\log(i(-1+\sqrt{2})), now recall that log(z)=Logz+i(arg(z)+2kπ)\log(z)=Log|z|+i(\arg(z)+2k\pi) in this case z=i+2iz=-i+\sqrt{2}i which implies that z=1+2|z|=-1+\sqrt{2} and arg(i+2i)=π2\arg(-i+\sqrt{2}i)=\frac{\pi}{2} since there is no real component of zz and the imaginary part is greater than 0, so we must have that sech1(i)=log(1+2)+i(π2+2kπ)sech^{-1}(i)=\log(-1+\sqrt{2})+i(\frac{\pi}{2}+2k\pi) where kZ k \in \Bbb Z.
My main concern is the argument i calculated i'm unsure if its correct , i assume if my case is correct for the positive then it will be the same method for the negative. Any help would be highly appreciated.
(edited 6 years ago)
Reply 1
Pricipal argument:
Yes, your principal argument of π/2 \pi / 2 is fine as (1+2)i (-1 + \sqrt{2} ) i is purely imaginary and is of form ai, where aR+ a \in \mathbb{R}^{+} .

As a check, try sech(z) \mathrm{sech}(z) , where z z is the answer in your first post.

Though for part of your last equation, instead of 2ki 2ki , you need 2kπ 2 k \pi .

Method for negative second case:
Yes, similar method for the second negative case.

For the real case, as sech(x) \mathrm{sech}(x) is an even function, the positive case is usually used of arsech(x) \mathrm{arsech}(x) so to return positive values (depending on context of course).
(edited 6 years ago)
Reply 2
Original post by simon0
Pricipal argument:
Yes, your principal argument of π/2 \pi / 2 is fine as (1+2)i (-1 + \sqrt{2} ) i is purely imaginary and is of form ai, where aR+ a \in \mathbb{R}^{+} .

As a check, try sech(z) \mathrm{sech}(z) , where z z is the answer in your first post.

Though for part of your last equation, instead of 2ki 2ki , you need 2kπ 2 k \pi .

Method for negative second case:
Yes, similar method for the second negative case.

For the real case, as sech(x) \mathrm{sech}(x) is an even function, the positive case is usually used of arsech(x) \mathrm{arsech}(x) so to return positive values (depending on context of course).

Hi thanks for your response I understand most of what you are saying I just don't see how there can be a real case
Reply 3
Original post by simon0
Pricipal argument:
Yes, your principal argument of π/2 \pi / 2 is fine as (1+2)i (-1 + \sqrt{2} ) i is purely imaginary and is of form ai, where aR+ a \in \mathbb{R}^{+} .

As a check, try sech(z) \mathrm{sech}(z) , where z z is the answer in your first post.

Though for part of your last equation, instead of 2ki 2ki , you need 2kπ 2 k \pi .

Method for negative second case:
Yes, similar method for the second negative case.

For the real case, as sech(x) \mathrm{sech}(x) is an even function, the positive case is usually used of arsech(x) \mathrm{arsech}(x) so to return positive values (depending on context of course).

Also what would the Argand diagram look like for the positive case ?
Reply 4
Original post by Scary
Hi thanks for your response I understand most of what you are saying I just don't see how there can be a real case


I meant of sech(x) \mathrm{sech}(x) where xR x \in \mathbb{R} .
Reply 5
Original post by Scary
Also what would the Argand diagram look like for the positive case ?


Your solutions for the positive case are of form:

log(1+2)+i((π/2)+2kπ), \displaystyle \log ( -1 + \sqrt{2} ) + i( (\pi/2) + 2k \pi ), where kZ k \in \mathbb{Z} .

This is just a complex value (of form a + bi) and using the pricipal argument, can you now see how we can create an argand diagram?
Reply 6
Original post by simon0
Your solutions for the positive case are of form:

log(1+2)+i((π/2)+2kπ), \displaystyle \log ( -1 + \sqrt{2} ) + i( (\pi/2) + 2k \pi ), where kZ k \in \mathbb{Z} .

This is just a complex value (of form a + bi) and using the pricipal argument, can you now see how we can create an argand diagram?

Yeah the a value is constant so it will just be a set of solutions on that vertical line repeating up and down the imaginary axis
(edited 6 years ago)
Reply 7
Original post by Scary
Yeah the a value is constant so it will just be a set of solutions on that vertical line repeating up and down the imaginary axis


Yes and of course the set of solutions for the negative case.
Reply 8
Original post by simon0
Yes and of course the set of solutions for the negative case.


Brilliant, thank you for your help, i was just unsure about the diagrams, also a quick question if we had iiπ i^{i\pi} and wanted to find all of it's values in C\Bbb C when we plot an argand diagram for these values just all lie on the negative real axis? using the fact that zα=eαlog(z)z^{\alpha} =e^{\alpha \log(z)} using this i calculated iiπ=eπ(π2+2kπ),kZ i^{i\pi} =e^{-\pi (\frac{\pi}{2}+2k\pi)},k \in \Bbb Z.
Reply 9
Original post by Scary
Brilliant, thank you for your help, i was just unsure about the diagrams, also a quick question if we had iiπ i^{i\pi} and wanted to find all of it's values in C\Bbb C when we plot an argand diagram for these values just all lie on the negative real axis? using the fact that zα=eαlog(z)z^{\alpha} =e^{\alpha \log(z)} using this i calculated iiπ=eπ(π2+2kπ),kZ i^{i\pi} =e^{-\pi (\frac{\pi}{2}+2k\pi)},k \in \Bbb Z.


Looks good to me.

Quick Reply

Latest