How to prove that...
$begingroup$
How to prove that $$int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx=frac{pi^{2}}{16}-frac{ln^{2}left(sqrt{2}+1right)}{4}?$$
I have encountered this integral recently, and I am aware that one can show how this is true with a substitution $u=ln{x}$ and then expanding it into a series of integrals of the form $int_{0}^{ln{(sqrt{2}+1)}}ue^{-nu}du$ where each integral is then calculated individually, but it feels pretty brute-forced. Is there any other way to do this? Thanks.
real-analysis calculus integration definite-integrals
$endgroup$
add a comment |
$begingroup$
How to prove that $$int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx=frac{pi^{2}}{16}-frac{ln^{2}left(sqrt{2}+1right)}{4}?$$
I have encountered this integral recently, and I am aware that one can show how this is true with a substitution $u=ln{x}$ and then expanding it into a series of integrals of the form $int_{0}^{ln{(sqrt{2}+1)}}ue^{-nu}du$ where each integral is then calculated individually, but it feels pretty brute-forced. Is there any other way to do this? Thanks.
real-analysis calculus integration definite-integrals
$endgroup$
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
1
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
1
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
3
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43
add a comment |
$begingroup$
How to prove that $$int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx=frac{pi^{2}}{16}-frac{ln^{2}left(sqrt{2}+1right)}{4}?$$
I have encountered this integral recently, and I am aware that one can show how this is true with a substitution $u=ln{x}$ and then expanding it into a series of integrals of the form $int_{0}^{ln{(sqrt{2}+1)}}ue^{-nu}du$ where each integral is then calculated individually, but it feels pretty brute-forced. Is there any other way to do this? Thanks.
real-analysis calculus integration definite-integrals
$endgroup$
How to prove that $$int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx=frac{pi^{2}}{16}-frac{ln^{2}left(sqrt{2}+1right)}{4}?$$
I have encountered this integral recently, and I am aware that one can show how this is true with a substitution $u=ln{x}$ and then expanding it into a series of integrals of the form $int_{0}^{ln{(sqrt{2}+1)}}ue^{-nu}du$ where each integral is then calculated individually, but it feels pretty brute-forced. Is there any other way to do this? Thanks.
real-analysis calculus integration definite-integrals
real-analysis calculus integration definite-integrals
asked Jan 10 at 2:13
Edward H.Edward H.
1439
1439
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
1
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
1
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
3
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43
add a comment |
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
1
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
1
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
3
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
1
1
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
1
1
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
3
3
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43
add a comment |
3 Answers
3
active
oldest
votes
$begingroup$
Put
begin{equation*}
I=int_{1}^{sqrt{2}+1}dfrac{ln x}{x^2-1}, mathrm{d}x =[x=1/y] = int_{sqrt{2}-1}^{1}dfrac{ln y}{y^2-1}, mathrm{d}y.tag{1}
end{equation*}
After the substitution $ y=frac{1-z}{1+z}$ and integration by parts we have
begin{gather*}
I = int_{0}^{sqrt{2}-1}dfrac{lnleft(frac{1-z}{1+z}right)}{-2z}, mathrm{d}z = left[-dfrac{1}{2}ln(z)lnleft(frac{1-z}{1+z}right)right]_{0}^{sqrt{2}-1}+int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z =\[2ex]-dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z.tag{2}
end{gather*}
If we add (1) and (2) we get
begin{equation*}
2I = -dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{1}dfrac{ln z}{z^2-1}, mathrm{d}z.
end{equation*}
Consequently
begin{gather*}
I = -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}int_{0}^{1}left(sum_{k=0}^{infty}ln(z)z^{2k}right), mathrm{d}z =\[2ex] -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}sum_{k=0}^{infty}int_{0}^{1}ln(z)z^{2k}, mathrm{d}z =\[2ex]
-dfrac{1}{4}ln^2(sqrt{2}+1) +dfrac{1}{2}sum_{k=0}^{infty}dfrac{1}{(2k+1)^2} = dfrac{pi^2}{16} -dfrac{1}{4}ln^2(sqrt{2}+1).
end{gather*}
$endgroup$
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
add a comment |
$begingroup$
A Complete Solution Now
Consider
$$F(s)=int_1^sfrac{log x}{x^2-1}mathrm dx$$
Like before,
$$F(s)=frac12int_1^{s}frac{log x}{x-1}mathrm dx-frac12int_1^{s}frac{log x}{x+1}mathrm dx$$
$$F(s)=-frac12mathrm{Li}_2(1-s)-frac12J(s)$$
For $J(s)$, we integrate by parts with $mathrm dv=frac{mathrm dx}{1+x}$ to get
$$J(s)=log(s)log(s+1)-int_1^sfrac{log(1+x)}{x}mathrm dx$$
$$J(s)=log(s)log(s+1)+int_2^{1+s}frac{log x}{1-x}mathrm dx$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(1-x)bigg|_2^{1+s}$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)-mathrm{Li}_2(-1)$$
Using $mathrm{Li}_k(-1)=(2^{1-k}-1)zeta(k)$,
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)+frac{pi^2}{12}$$
Plugging in $s=1+sqrt2$,
$$J(1+sqrt2)=log(1+sqrt2)log(2+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
Which, as The OP noted, becomes
$$J(1+sqrt2)=log^2(1+sqrt2)+frac12log(2)log(1+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
And again as the OP noted,
$$F(1+sqrt2)=-frac{pi^2}{48}-frac14log^2(1+sqrt2)+frac14log2log(1+sqrt2)+frac1{16}log^22+frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)$$
And since
$$frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)=frac{pi^2}{12}-frac14log2log(1+sqrt2)-frac1{16}log^22$$
We have
$$F(1+sqrt2)=frac{pi^2}{16}-frac14log^2(1+sqrt2)$$
$endgroup$
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
|
show 2 more comments
$begingroup$
Here is one possible approach.
Enforcing a substitution of $x mapsto 1/x$ to begin with gives
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -int_{sqrt{2}-1}^1 frac{ln x}{1 - x^2} , dx = -sum_{n = 0}^infty int_{sqrt{2} - 1}^1 x^{2n} ln x , dx,
end{align}
where we have taken advantage of the geometric sum for $1/(1 - x^2)$.
Integrating by parts then leads to
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= - ln (1 + sqrt{2}) sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} + sum_{n = 0}^infty frac{1}{(2n + 1)^2}\
&qquad - sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2}. tag1
end{align}
Observing that
$$tanh^{-1} z = sum_{n = 0}^infty frac{z^{2n + 1}}{2n + 1}, qquad |z| < 1, qquad (*)$$
the first of the sums appearing in (1) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} = tanh^{-1} (sqrt{2} - 1) = frac{1}{2} ln (1 + sqrt{2}).$$
To find the second and third sums in (1), divide ($*$) by $x$ before integrating up from $0$ to $x$. Thus
begin{align}
sum_{n = 0}^infty frac{1}{2n + 1} int_0^x t^{2n} , dt &= int_0^x frac{tanh^{-1} t}{t} , dt\
sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} &= frac{1}{2} int_0^x ln left (frac{1 + t}{1 - t} right ) frac{dt}{t}\
&= frac{1}{2} int_0^{-x} frac{ln (1 - t)}{t} , dt - frac{1}{2} int_0^x frac{ln (1 - t)}{t} , dt
end{align}
or
$$sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (x) - operatorname{Li}_2 (-x) right ], qquad (**)$$
where $operatorname{Li}_2 (x)$ is the dilogarithm function.
The second and third sums appearing in (1) can now be found using ($**$). For the second sum, setting $x = 1$ gives
$$sum_{n = 1}^infty frac{1}{(2n + 1)^2} = frac{1}{2} operatorname{Li}_2 (1) - frac{1}{2} operatorname{Li}_2 (-1) = frac{1}{2} left (frac{pi^2}{6} + frac{pi^2}{12} right ) = frac{pi^2}{8},$$
where the well-known special values for the dilogarithm function at 1 and $-1$ have been used.
For the third sum, setting $x = sqrt{2} - 1$ gives
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) right ]. qquad (dagger)$$
It now remains to express the difference between the two dilogarithms is terms of elementary constants.
To do this, we will make use of the following identities for the dilogarithm:
begin{align}
operatorname{Li}_2 (1-x) + operatorname{Li}_2 left (1 - frac{1}{x} right ) &= -frac{1}{2} ln^2 x tag2\
operatorname{Li}_2 (x) + operatorname{Li}_2 (1-x) &= frac{pi^2}{6} - ln x ln (1 - x) tag3\
operatorname{Li}_2 (x) + operatorname{Li}_2 (-x) &= frac{1}{2} operatorname{Li}_2 (x^2) tag4
end{align}
So here we go.
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) &= operatorname{Li}_2 [1 - (2 - sqrt{2})]\
&= -operatorname{Li}_2 left (1 - frac{1}{2 - sqrt{2}} right ) - frac{1}{2} ln^2 (2 - sqrt{2}) qquad text{(by (2))}\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 left (frac{sqrt{2}}{1 + sqrt{2}} right )\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) -frac{1}{2} left [frac{1}{2} ln 2 - ln (1 + sqrt{2}) right ]^2\
&= -operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 + frac{1}{2} ln 2 ln (1 + sqrt{2}) - frac{1}{2} ln^2 (1 + sqrt{2})
end{align}
And
begin{align}
operatorname{Li}_2 (1 - sqrt{2}) &= -operatorname{Li}_2 left (1 - frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 sqrt{2} qquad text{(by (2))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + ln left (frac{1}{sqrt{2}} right ) ln left (1 - frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 qquad text{(by (3))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} -frac{1}{2} ln 2 ln left (frac{1}{sqrt{2}(1 + sqrt{2})} right ) - frac{1}{8} ln^2 2\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 left [frac{1}{2} ln 2 + ln (1 + sqrt{2}) right ] - frac{1}{8} ln^2 2\
&=operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 ln (1 + sqrt{2}) + frac{1}{8} ln^2 2.
end{align}
Thus
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) &= - left [operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) + operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) right ] +frac{pi^2}{6} - frac{1}{2} ln^2 (1 + sqrt{2}) - frac{1}{4} ln^2 2\
&= -frac{1}{2} operatorname{Li}_2 left (frac{1}{2} right ) + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) - frac{1}{4} ln^2 2 qquad text{(by (4))} \
&= -frac{1}{2} left [frac{pi^2}{12} - frac{1}{2} ln^2 2 right ] + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) -frac{1}{4} ln^2 2\
&= frac{pi^2}{8} - frac{1}{2} ln^2 (1 + sqrt{2}).
end{align}
Here the well-known value for the dilogarithm function at $x = 1/2$ has been used.
So the sum in ($dagger$) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}).$$
Putting the final pieces of the puzzle together, (1) becomes
begin{align}
int_{1}^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -frac{1}{2} ln^2 (1 + sqrt{2}) + frac{pi^2}{8} - left [frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}) right ]\
&= frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}),
end{align}
as required.
$endgroup$
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3068155%2fhow-to-prove-that-int-1-sqrt21-frac-lnxx2-1dx-frac-pi2%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
3 Answers
3
active
oldest
votes
3 Answers
3
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
Put
begin{equation*}
I=int_{1}^{sqrt{2}+1}dfrac{ln x}{x^2-1}, mathrm{d}x =[x=1/y] = int_{sqrt{2}-1}^{1}dfrac{ln y}{y^2-1}, mathrm{d}y.tag{1}
end{equation*}
After the substitution $ y=frac{1-z}{1+z}$ and integration by parts we have
begin{gather*}
I = int_{0}^{sqrt{2}-1}dfrac{lnleft(frac{1-z}{1+z}right)}{-2z}, mathrm{d}z = left[-dfrac{1}{2}ln(z)lnleft(frac{1-z}{1+z}right)right]_{0}^{sqrt{2}-1}+int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z =\[2ex]-dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z.tag{2}
end{gather*}
If we add (1) and (2) we get
begin{equation*}
2I = -dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{1}dfrac{ln z}{z^2-1}, mathrm{d}z.
end{equation*}
Consequently
begin{gather*}
I = -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}int_{0}^{1}left(sum_{k=0}^{infty}ln(z)z^{2k}right), mathrm{d}z =\[2ex] -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}sum_{k=0}^{infty}int_{0}^{1}ln(z)z^{2k}, mathrm{d}z =\[2ex]
-dfrac{1}{4}ln^2(sqrt{2}+1) +dfrac{1}{2}sum_{k=0}^{infty}dfrac{1}{(2k+1)^2} = dfrac{pi^2}{16} -dfrac{1}{4}ln^2(sqrt{2}+1).
end{gather*}
$endgroup$
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
add a comment |
$begingroup$
Put
begin{equation*}
I=int_{1}^{sqrt{2}+1}dfrac{ln x}{x^2-1}, mathrm{d}x =[x=1/y] = int_{sqrt{2}-1}^{1}dfrac{ln y}{y^2-1}, mathrm{d}y.tag{1}
end{equation*}
After the substitution $ y=frac{1-z}{1+z}$ and integration by parts we have
begin{gather*}
I = int_{0}^{sqrt{2}-1}dfrac{lnleft(frac{1-z}{1+z}right)}{-2z}, mathrm{d}z = left[-dfrac{1}{2}ln(z)lnleft(frac{1-z}{1+z}right)right]_{0}^{sqrt{2}-1}+int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z =\[2ex]-dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z.tag{2}
end{gather*}
If we add (1) and (2) we get
begin{equation*}
2I = -dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{1}dfrac{ln z}{z^2-1}, mathrm{d}z.
end{equation*}
Consequently
begin{gather*}
I = -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}int_{0}^{1}left(sum_{k=0}^{infty}ln(z)z^{2k}right), mathrm{d}z =\[2ex] -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}sum_{k=0}^{infty}int_{0}^{1}ln(z)z^{2k}, mathrm{d}z =\[2ex]
-dfrac{1}{4}ln^2(sqrt{2}+1) +dfrac{1}{2}sum_{k=0}^{infty}dfrac{1}{(2k+1)^2} = dfrac{pi^2}{16} -dfrac{1}{4}ln^2(sqrt{2}+1).
end{gather*}
$endgroup$
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
add a comment |
$begingroup$
Put
begin{equation*}
I=int_{1}^{sqrt{2}+1}dfrac{ln x}{x^2-1}, mathrm{d}x =[x=1/y] = int_{sqrt{2}-1}^{1}dfrac{ln y}{y^2-1}, mathrm{d}y.tag{1}
end{equation*}
After the substitution $ y=frac{1-z}{1+z}$ and integration by parts we have
begin{gather*}
I = int_{0}^{sqrt{2}-1}dfrac{lnleft(frac{1-z}{1+z}right)}{-2z}, mathrm{d}z = left[-dfrac{1}{2}ln(z)lnleft(frac{1-z}{1+z}right)right]_{0}^{sqrt{2}-1}+int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z =\[2ex]-dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z.tag{2}
end{gather*}
If we add (1) and (2) we get
begin{equation*}
2I = -dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{1}dfrac{ln z}{z^2-1}, mathrm{d}z.
end{equation*}
Consequently
begin{gather*}
I = -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}int_{0}^{1}left(sum_{k=0}^{infty}ln(z)z^{2k}right), mathrm{d}z =\[2ex] -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}sum_{k=0}^{infty}int_{0}^{1}ln(z)z^{2k}, mathrm{d}z =\[2ex]
-dfrac{1}{4}ln^2(sqrt{2}+1) +dfrac{1}{2}sum_{k=0}^{infty}dfrac{1}{(2k+1)^2} = dfrac{pi^2}{16} -dfrac{1}{4}ln^2(sqrt{2}+1).
end{gather*}
$endgroup$
Put
begin{equation*}
I=int_{1}^{sqrt{2}+1}dfrac{ln x}{x^2-1}, mathrm{d}x =[x=1/y] = int_{sqrt{2}-1}^{1}dfrac{ln y}{y^2-1}, mathrm{d}y.tag{1}
end{equation*}
After the substitution $ y=frac{1-z}{1+z}$ and integration by parts we have
begin{gather*}
I = int_{0}^{sqrt{2}-1}dfrac{lnleft(frac{1-z}{1+z}right)}{-2z}, mathrm{d}z = left[-dfrac{1}{2}ln(z)lnleft(frac{1-z}{1+z}right)right]_{0}^{sqrt{2}-1}+int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z =\[2ex]-dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{sqrt{2}-1}dfrac{ln z}{z^2-1}, mathrm{d}z.tag{2}
end{gather*}
If we add (1) and (2) we get
begin{equation*}
2I = -dfrac{1}{2}ln^2(sqrt{2}+1) + int_{0}^{1}dfrac{ln z}{z^2-1}, mathrm{d}z.
end{equation*}
Consequently
begin{gather*}
I = -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}int_{0}^{1}left(sum_{k=0}^{infty}ln(z)z^{2k}right), mathrm{d}z =\[2ex] -dfrac{1}{4}ln^2(sqrt{2}+1) -dfrac{1}{2}sum_{k=0}^{infty}int_{0}^{1}ln(z)z^{2k}, mathrm{d}z =\[2ex]
-dfrac{1}{4}ln^2(sqrt{2}+1) +dfrac{1}{2}sum_{k=0}^{infty}dfrac{1}{(2k+1)^2} = dfrac{pi^2}{16} -dfrac{1}{4}ln^2(sqrt{2}+1).
end{gather*}
answered Jan 10 at 21:58
JanGJanG
2,802514
2,802514
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
add a comment |
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
2
2
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
$begingroup$
Now that is a solution! Very nice (+1)
$endgroup$
– omegadot
Jan 10 at 22:06
1
1
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
$begingroup$
@ omegadot$ Thank you.
$endgroup$
– JanG
Jan 11 at 5:56
add a comment |
$begingroup$
A Complete Solution Now
Consider
$$F(s)=int_1^sfrac{log x}{x^2-1}mathrm dx$$
Like before,
$$F(s)=frac12int_1^{s}frac{log x}{x-1}mathrm dx-frac12int_1^{s}frac{log x}{x+1}mathrm dx$$
$$F(s)=-frac12mathrm{Li}_2(1-s)-frac12J(s)$$
For $J(s)$, we integrate by parts with $mathrm dv=frac{mathrm dx}{1+x}$ to get
$$J(s)=log(s)log(s+1)-int_1^sfrac{log(1+x)}{x}mathrm dx$$
$$J(s)=log(s)log(s+1)+int_2^{1+s}frac{log x}{1-x}mathrm dx$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(1-x)bigg|_2^{1+s}$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)-mathrm{Li}_2(-1)$$
Using $mathrm{Li}_k(-1)=(2^{1-k}-1)zeta(k)$,
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)+frac{pi^2}{12}$$
Plugging in $s=1+sqrt2$,
$$J(1+sqrt2)=log(1+sqrt2)log(2+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
Which, as The OP noted, becomes
$$J(1+sqrt2)=log^2(1+sqrt2)+frac12log(2)log(1+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
And again as the OP noted,
$$F(1+sqrt2)=-frac{pi^2}{48}-frac14log^2(1+sqrt2)+frac14log2log(1+sqrt2)+frac1{16}log^22+frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)$$
And since
$$frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)=frac{pi^2}{12}-frac14log2log(1+sqrt2)-frac1{16}log^22$$
We have
$$F(1+sqrt2)=frac{pi^2}{16}-frac14log^2(1+sqrt2)$$
$endgroup$
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
|
show 2 more comments
$begingroup$
A Complete Solution Now
Consider
$$F(s)=int_1^sfrac{log x}{x^2-1}mathrm dx$$
Like before,
$$F(s)=frac12int_1^{s}frac{log x}{x-1}mathrm dx-frac12int_1^{s}frac{log x}{x+1}mathrm dx$$
$$F(s)=-frac12mathrm{Li}_2(1-s)-frac12J(s)$$
For $J(s)$, we integrate by parts with $mathrm dv=frac{mathrm dx}{1+x}$ to get
$$J(s)=log(s)log(s+1)-int_1^sfrac{log(1+x)}{x}mathrm dx$$
$$J(s)=log(s)log(s+1)+int_2^{1+s}frac{log x}{1-x}mathrm dx$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(1-x)bigg|_2^{1+s}$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)-mathrm{Li}_2(-1)$$
Using $mathrm{Li}_k(-1)=(2^{1-k}-1)zeta(k)$,
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)+frac{pi^2}{12}$$
Plugging in $s=1+sqrt2$,
$$J(1+sqrt2)=log(1+sqrt2)log(2+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
Which, as The OP noted, becomes
$$J(1+sqrt2)=log^2(1+sqrt2)+frac12log(2)log(1+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
And again as the OP noted,
$$F(1+sqrt2)=-frac{pi^2}{48}-frac14log^2(1+sqrt2)+frac14log2log(1+sqrt2)+frac1{16}log^22+frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)$$
And since
$$frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)=frac{pi^2}{12}-frac14log2log(1+sqrt2)-frac1{16}log^22$$
We have
$$F(1+sqrt2)=frac{pi^2}{16}-frac14log^2(1+sqrt2)$$
$endgroup$
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
|
show 2 more comments
$begingroup$
A Complete Solution Now
Consider
$$F(s)=int_1^sfrac{log x}{x^2-1}mathrm dx$$
Like before,
$$F(s)=frac12int_1^{s}frac{log x}{x-1}mathrm dx-frac12int_1^{s}frac{log x}{x+1}mathrm dx$$
$$F(s)=-frac12mathrm{Li}_2(1-s)-frac12J(s)$$
For $J(s)$, we integrate by parts with $mathrm dv=frac{mathrm dx}{1+x}$ to get
$$J(s)=log(s)log(s+1)-int_1^sfrac{log(1+x)}{x}mathrm dx$$
$$J(s)=log(s)log(s+1)+int_2^{1+s}frac{log x}{1-x}mathrm dx$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(1-x)bigg|_2^{1+s}$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)-mathrm{Li}_2(-1)$$
Using $mathrm{Li}_k(-1)=(2^{1-k}-1)zeta(k)$,
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)+frac{pi^2}{12}$$
Plugging in $s=1+sqrt2$,
$$J(1+sqrt2)=log(1+sqrt2)log(2+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
Which, as The OP noted, becomes
$$J(1+sqrt2)=log^2(1+sqrt2)+frac12log(2)log(1+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
And again as the OP noted,
$$F(1+sqrt2)=-frac{pi^2}{48}-frac14log^2(1+sqrt2)+frac14log2log(1+sqrt2)+frac1{16}log^22+frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)$$
And since
$$frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)=frac{pi^2}{12}-frac14log2log(1+sqrt2)-frac1{16}log^22$$
We have
$$F(1+sqrt2)=frac{pi^2}{16}-frac14log^2(1+sqrt2)$$
$endgroup$
A Complete Solution Now
Consider
$$F(s)=int_1^sfrac{log x}{x^2-1}mathrm dx$$
Like before,
$$F(s)=frac12int_1^{s}frac{log x}{x-1}mathrm dx-frac12int_1^{s}frac{log x}{x+1}mathrm dx$$
$$F(s)=-frac12mathrm{Li}_2(1-s)-frac12J(s)$$
For $J(s)$, we integrate by parts with $mathrm dv=frac{mathrm dx}{1+x}$ to get
$$J(s)=log(s)log(s+1)-int_1^sfrac{log(1+x)}{x}mathrm dx$$
$$J(s)=log(s)log(s+1)+int_2^{1+s}frac{log x}{1-x}mathrm dx$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(1-x)bigg|_2^{1+s}$$
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)-mathrm{Li}_2(-1)$$
Using $mathrm{Li}_k(-1)=(2^{1-k}-1)zeta(k)$,
$$J(s)=log(s)log(s+1)+mathrm{Li}_2(-s)+frac{pi^2}{12}$$
Plugging in $s=1+sqrt2$,
$$J(1+sqrt2)=log(1+sqrt2)log(2+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
Which, as The OP noted, becomes
$$J(1+sqrt2)=log^2(1+sqrt2)+frac12log(2)log(1+sqrt2)+mathrm{Li}_2(-1-sqrt2)+frac{pi^2}{12}$$
And again as the OP noted,
$$F(1+sqrt2)=-frac{pi^2}{48}-frac14log^2(1+sqrt2)+frac14log2log(1+sqrt2)+frac1{16}log^22+frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)$$
And since
$$frac12mathrm{Li}_2bigg(frac1{sqrt2}bigg)-frac12mathrm{Li}_2(1-sqrt2)=frac{pi^2}{12}-frac14log2log(1+sqrt2)-frac1{16}log^22$$
We have
$$F(1+sqrt2)=frac{pi^2}{16}-frac14log^2(1+sqrt2)$$
edited Jan 10 at 19:32
answered Jan 10 at 3:02
clathratusclathratus
4,234336
4,234336
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
|
show 2 more comments
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
1
1
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
$begingroup$
Evaluating $frac14mathrm{Li}_2(1-x^2)-mathrm{Li}_2(1-x)$ at $x=1$ gives $0$, right?
$endgroup$
– Ant
Jan 10 at 3:07
1
1
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
$begingroup$
@Ant Right... I'm on it
$endgroup$
– clathratus
Jan 10 at 3:08
1
1
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
$begingroup$
I think the WolframAlpha result is using equality (6~7) from this MathWorld page
$endgroup$
– Edward H.
Jan 10 at 5:24
1
1
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
$begingroup$
Seeing @omegadot 's answer I'm thinking that yours can also work, like this: begin{align*} &quadfrac{1}{4}text{Li}_2(-2-2sqrt{2})-text{Li}_2(-sqrt{2})\ &=frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2}+1)ln(sqrt{2}+2)-text{Li}_2(-sqrt{2}-1)+text{Li}_2(-sqrt{2})right)-text{Li}_2(-sqrt{2})\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}text{Li}_2(-sqrt{2}-1)-frac{1}{2}text{Li}_2(-sqrt{2})\ end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:14
1
1
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
$begingroup$
begin{align*} &quaddots\ &=-frac{pi^2}{24}-frac{1}{2}ln^2(sqrt{2}+1)-frac{1}{4}ln{2}ln(sqrt{2}+1)-frac{1}{2}left(-frac{1}{2}ln^2(sqrt{2}+2)-text{Li}_2(frac{1}{sqrt{2}})right)\ &quad-frac{1}{2}left(-frac{pi^2}{12}-ln(sqrt{2})ln(sqrt{2}+1)+text{Li}_2(1-sqrt{2})-frac{1}{2}text{Li}_2(-1)right)\ &=-frac{pi^2}{48}-frac{1}{4}ln^2(sqrt{2}+1)+frac{1}{4}ln{2}ln(sqrt{2}+1)+frac{1}{16}ln^2{2}+frac{1}{2}text{Li}_2(frac{1}{sqrt{2}})-frac{1}{2}text{Li}_2(1-sqrt{2}) end{align*} (tbc)
$endgroup$
– Edward H.
Jan 10 at 18:22
|
show 2 more comments
$begingroup$
Here is one possible approach.
Enforcing a substitution of $x mapsto 1/x$ to begin with gives
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -int_{sqrt{2}-1}^1 frac{ln x}{1 - x^2} , dx = -sum_{n = 0}^infty int_{sqrt{2} - 1}^1 x^{2n} ln x , dx,
end{align}
where we have taken advantage of the geometric sum for $1/(1 - x^2)$.
Integrating by parts then leads to
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= - ln (1 + sqrt{2}) sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} + sum_{n = 0}^infty frac{1}{(2n + 1)^2}\
&qquad - sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2}. tag1
end{align}
Observing that
$$tanh^{-1} z = sum_{n = 0}^infty frac{z^{2n + 1}}{2n + 1}, qquad |z| < 1, qquad (*)$$
the first of the sums appearing in (1) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} = tanh^{-1} (sqrt{2} - 1) = frac{1}{2} ln (1 + sqrt{2}).$$
To find the second and third sums in (1), divide ($*$) by $x$ before integrating up from $0$ to $x$. Thus
begin{align}
sum_{n = 0}^infty frac{1}{2n + 1} int_0^x t^{2n} , dt &= int_0^x frac{tanh^{-1} t}{t} , dt\
sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} &= frac{1}{2} int_0^x ln left (frac{1 + t}{1 - t} right ) frac{dt}{t}\
&= frac{1}{2} int_0^{-x} frac{ln (1 - t)}{t} , dt - frac{1}{2} int_0^x frac{ln (1 - t)}{t} , dt
end{align}
or
$$sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (x) - operatorname{Li}_2 (-x) right ], qquad (**)$$
where $operatorname{Li}_2 (x)$ is the dilogarithm function.
The second and third sums appearing in (1) can now be found using ($**$). For the second sum, setting $x = 1$ gives
$$sum_{n = 1}^infty frac{1}{(2n + 1)^2} = frac{1}{2} operatorname{Li}_2 (1) - frac{1}{2} operatorname{Li}_2 (-1) = frac{1}{2} left (frac{pi^2}{6} + frac{pi^2}{12} right ) = frac{pi^2}{8},$$
where the well-known special values for the dilogarithm function at 1 and $-1$ have been used.
For the third sum, setting $x = sqrt{2} - 1$ gives
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) right ]. qquad (dagger)$$
It now remains to express the difference between the two dilogarithms is terms of elementary constants.
To do this, we will make use of the following identities for the dilogarithm:
begin{align}
operatorname{Li}_2 (1-x) + operatorname{Li}_2 left (1 - frac{1}{x} right ) &= -frac{1}{2} ln^2 x tag2\
operatorname{Li}_2 (x) + operatorname{Li}_2 (1-x) &= frac{pi^2}{6} - ln x ln (1 - x) tag3\
operatorname{Li}_2 (x) + operatorname{Li}_2 (-x) &= frac{1}{2} operatorname{Li}_2 (x^2) tag4
end{align}
So here we go.
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) &= operatorname{Li}_2 [1 - (2 - sqrt{2})]\
&= -operatorname{Li}_2 left (1 - frac{1}{2 - sqrt{2}} right ) - frac{1}{2} ln^2 (2 - sqrt{2}) qquad text{(by (2))}\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 left (frac{sqrt{2}}{1 + sqrt{2}} right )\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) -frac{1}{2} left [frac{1}{2} ln 2 - ln (1 + sqrt{2}) right ]^2\
&= -operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 + frac{1}{2} ln 2 ln (1 + sqrt{2}) - frac{1}{2} ln^2 (1 + sqrt{2})
end{align}
And
begin{align}
operatorname{Li}_2 (1 - sqrt{2}) &= -operatorname{Li}_2 left (1 - frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 sqrt{2} qquad text{(by (2))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + ln left (frac{1}{sqrt{2}} right ) ln left (1 - frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 qquad text{(by (3))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} -frac{1}{2} ln 2 ln left (frac{1}{sqrt{2}(1 + sqrt{2})} right ) - frac{1}{8} ln^2 2\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 left [frac{1}{2} ln 2 + ln (1 + sqrt{2}) right ] - frac{1}{8} ln^2 2\
&=operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 ln (1 + sqrt{2}) + frac{1}{8} ln^2 2.
end{align}
Thus
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) &= - left [operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) + operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) right ] +frac{pi^2}{6} - frac{1}{2} ln^2 (1 + sqrt{2}) - frac{1}{4} ln^2 2\
&= -frac{1}{2} operatorname{Li}_2 left (frac{1}{2} right ) + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) - frac{1}{4} ln^2 2 qquad text{(by (4))} \
&= -frac{1}{2} left [frac{pi^2}{12} - frac{1}{2} ln^2 2 right ] + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) -frac{1}{4} ln^2 2\
&= frac{pi^2}{8} - frac{1}{2} ln^2 (1 + sqrt{2}).
end{align}
Here the well-known value for the dilogarithm function at $x = 1/2$ has been used.
So the sum in ($dagger$) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}).$$
Putting the final pieces of the puzzle together, (1) becomes
begin{align}
int_{1}^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -frac{1}{2} ln^2 (1 + sqrt{2}) + frac{pi^2}{8} - left [frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}) right ]\
&= frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}),
end{align}
as required.
$endgroup$
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
add a comment |
$begingroup$
Here is one possible approach.
Enforcing a substitution of $x mapsto 1/x$ to begin with gives
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -int_{sqrt{2}-1}^1 frac{ln x}{1 - x^2} , dx = -sum_{n = 0}^infty int_{sqrt{2} - 1}^1 x^{2n} ln x , dx,
end{align}
where we have taken advantage of the geometric sum for $1/(1 - x^2)$.
Integrating by parts then leads to
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= - ln (1 + sqrt{2}) sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} + sum_{n = 0}^infty frac{1}{(2n + 1)^2}\
&qquad - sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2}. tag1
end{align}
Observing that
$$tanh^{-1} z = sum_{n = 0}^infty frac{z^{2n + 1}}{2n + 1}, qquad |z| < 1, qquad (*)$$
the first of the sums appearing in (1) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} = tanh^{-1} (sqrt{2} - 1) = frac{1}{2} ln (1 + sqrt{2}).$$
To find the second and third sums in (1), divide ($*$) by $x$ before integrating up from $0$ to $x$. Thus
begin{align}
sum_{n = 0}^infty frac{1}{2n + 1} int_0^x t^{2n} , dt &= int_0^x frac{tanh^{-1} t}{t} , dt\
sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} &= frac{1}{2} int_0^x ln left (frac{1 + t}{1 - t} right ) frac{dt}{t}\
&= frac{1}{2} int_0^{-x} frac{ln (1 - t)}{t} , dt - frac{1}{2} int_0^x frac{ln (1 - t)}{t} , dt
end{align}
or
$$sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (x) - operatorname{Li}_2 (-x) right ], qquad (**)$$
where $operatorname{Li}_2 (x)$ is the dilogarithm function.
The second and third sums appearing in (1) can now be found using ($**$). For the second sum, setting $x = 1$ gives
$$sum_{n = 1}^infty frac{1}{(2n + 1)^2} = frac{1}{2} operatorname{Li}_2 (1) - frac{1}{2} operatorname{Li}_2 (-1) = frac{1}{2} left (frac{pi^2}{6} + frac{pi^2}{12} right ) = frac{pi^2}{8},$$
where the well-known special values for the dilogarithm function at 1 and $-1$ have been used.
For the third sum, setting $x = sqrt{2} - 1$ gives
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) right ]. qquad (dagger)$$
It now remains to express the difference between the two dilogarithms is terms of elementary constants.
To do this, we will make use of the following identities for the dilogarithm:
begin{align}
operatorname{Li}_2 (1-x) + operatorname{Li}_2 left (1 - frac{1}{x} right ) &= -frac{1}{2} ln^2 x tag2\
operatorname{Li}_2 (x) + operatorname{Li}_2 (1-x) &= frac{pi^2}{6} - ln x ln (1 - x) tag3\
operatorname{Li}_2 (x) + operatorname{Li}_2 (-x) &= frac{1}{2} operatorname{Li}_2 (x^2) tag4
end{align}
So here we go.
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) &= operatorname{Li}_2 [1 - (2 - sqrt{2})]\
&= -operatorname{Li}_2 left (1 - frac{1}{2 - sqrt{2}} right ) - frac{1}{2} ln^2 (2 - sqrt{2}) qquad text{(by (2))}\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 left (frac{sqrt{2}}{1 + sqrt{2}} right )\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) -frac{1}{2} left [frac{1}{2} ln 2 - ln (1 + sqrt{2}) right ]^2\
&= -operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 + frac{1}{2} ln 2 ln (1 + sqrt{2}) - frac{1}{2} ln^2 (1 + sqrt{2})
end{align}
And
begin{align}
operatorname{Li}_2 (1 - sqrt{2}) &= -operatorname{Li}_2 left (1 - frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 sqrt{2} qquad text{(by (2))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + ln left (frac{1}{sqrt{2}} right ) ln left (1 - frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 qquad text{(by (3))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} -frac{1}{2} ln 2 ln left (frac{1}{sqrt{2}(1 + sqrt{2})} right ) - frac{1}{8} ln^2 2\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 left [frac{1}{2} ln 2 + ln (1 + sqrt{2}) right ] - frac{1}{8} ln^2 2\
&=operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 ln (1 + sqrt{2}) + frac{1}{8} ln^2 2.
end{align}
Thus
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) &= - left [operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) + operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) right ] +frac{pi^2}{6} - frac{1}{2} ln^2 (1 + sqrt{2}) - frac{1}{4} ln^2 2\
&= -frac{1}{2} operatorname{Li}_2 left (frac{1}{2} right ) + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) - frac{1}{4} ln^2 2 qquad text{(by (4))} \
&= -frac{1}{2} left [frac{pi^2}{12} - frac{1}{2} ln^2 2 right ] + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) -frac{1}{4} ln^2 2\
&= frac{pi^2}{8} - frac{1}{2} ln^2 (1 + sqrt{2}).
end{align}
Here the well-known value for the dilogarithm function at $x = 1/2$ has been used.
So the sum in ($dagger$) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}).$$
Putting the final pieces of the puzzle together, (1) becomes
begin{align}
int_{1}^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -frac{1}{2} ln^2 (1 + sqrt{2}) + frac{pi^2}{8} - left [frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}) right ]\
&= frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}),
end{align}
as required.
$endgroup$
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
add a comment |
$begingroup$
Here is one possible approach.
Enforcing a substitution of $x mapsto 1/x$ to begin with gives
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -int_{sqrt{2}-1}^1 frac{ln x}{1 - x^2} , dx = -sum_{n = 0}^infty int_{sqrt{2} - 1}^1 x^{2n} ln x , dx,
end{align}
where we have taken advantage of the geometric sum for $1/(1 - x^2)$.
Integrating by parts then leads to
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= - ln (1 + sqrt{2}) sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} + sum_{n = 0}^infty frac{1}{(2n + 1)^2}\
&qquad - sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2}. tag1
end{align}
Observing that
$$tanh^{-1} z = sum_{n = 0}^infty frac{z^{2n + 1}}{2n + 1}, qquad |z| < 1, qquad (*)$$
the first of the sums appearing in (1) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} = tanh^{-1} (sqrt{2} - 1) = frac{1}{2} ln (1 + sqrt{2}).$$
To find the second and third sums in (1), divide ($*$) by $x$ before integrating up from $0$ to $x$. Thus
begin{align}
sum_{n = 0}^infty frac{1}{2n + 1} int_0^x t^{2n} , dt &= int_0^x frac{tanh^{-1} t}{t} , dt\
sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} &= frac{1}{2} int_0^x ln left (frac{1 + t}{1 - t} right ) frac{dt}{t}\
&= frac{1}{2} int_0^{-x} frac{ln (1 - t)}{t} , dt - frac{1}{2} int_0^x frac{ln (1 - t)}{t} , dt
end{align}
or
$$sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (x) - operatorname{Li}_2 (-x) right ], qquad (**)$$
where $operatorname{Li}_2 (x)$ is the dilogarithm function.
The second and third sums appearing in (1) can now be found using ($**$). For the second sum, setting $x = 1$ gives
$$sum_{n = 1}^infty frac{1}{(2n + 1)^2} = frac{1}{2} operatorname{Li}_2 (1) - frac{1}{2} operatorname{Li}_2 (-1) = frac{1}{2} left (frac{pi^2}{6} + frac{pi^2}{12} right ) = frac{pi^2}{8},$$
where the well-known special values for the dilogarithm function at 1 and $-1$ have been used.
For the third sum, setting $x = sqrt{2} - 1$ gives
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) right ]. qquad (dagger)$$
It now remains to express the difference between the two dilogarithms is terms of elementary constants.
To do this, we will make use of the following identities for the dilogarithm:
begin{align}
operatorname{Li}_2 (1-x) + operatorname{Li}_2 left (1 - frac{1}{x} right ) &= -frac{1}{2} ln^2 x tag2\
operatorname{Li}_2 (x) + operatorname{Li}_2 (1-x) &= frac{pi^2}{6} - ln x ln (1 - x) tag3\
operatorname{Li}_2 (x) + operatorname{Li}_2 (-x) &= frac{1}{2} operatorname{Li}_2 (x^2) tag4
end{align}
So here we go.
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) &= operatorname{Li}_2 [1 - (2 - sqrt{2})]\
&= -operatorname{Li}_2 left (1 - frac{1}{2 - sqrt{2}} right ) - frac{1}{2} ln^2 (2 - sqrt{2}) qquad text{(by (2))}\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 left (frac{sqrt{2}}{1 + sqrt{2}} right )\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) -frac{1}{2} left [frac{1}{2} ln 2 - ln (1 + sqrt{2}) right ]^2\
&= -operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 + frac{1}{2} ln 2 ln (1 + sqrt{2}) - frac{1}{2} ln^2 (1 + sqrt{2})
end{align}
And
begin{align}
operatorname{Li}_2 (1 - sqrt{2}) &= -operatorname{Li}_2 left (1 - frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 sqrt{2} qquad text{(by (2))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + ln left (frac{1}{sqrt{2}} right ) ln left (1 - frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 qquad text{(by (3))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} -frac{1}{2} ln 2 ln left (frac{1}{sqrt{2}(1 + sqrt{2})} right ) - frac{1}{8} ln^2 2\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 left [frac{1}{2} ln 2 + ln (1 + sqrt{2}) right ] - frac{1}{8} ln^2 2\
&=operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 ln (1 + sqrt{2}) + frac{1}{8} ln^2 2.
end{align}
Thus
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) &= - left [operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) + operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) right ] +frac{pi^2}{6} - frac{1}{2} ln^2 (1 + sqrt{2}) - frac{1}{4} ln^2 2\
&= -frac{1}{2} operatorname{Li}_2 left (frac{1}{2} right ) + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) - frac{1}{4} ln^2 2 qquad text{(by (4))} \
&= -frac{1}{2} left [frac{pi^2}{12} - frac{1}{2} ln^2 2 right ] + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) -frac{1}{4} ln^2 2\
&= frac{pi^2}{8} - frac{1}{2} ln^2 (1 + sqrt{2}).
end{align}
Here the well-known value for the dilogarithm function at $x = 1/2$ has been used.
So the sum in ($dagger$) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}).$$
Putting the final pieces of the puzzle together, (1) becomes
begin{align}
int_{1}^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -frac{1}{2} ln^2 (1 + sqrt{2}) + frac{pi^2}{8} - left [frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}) right ]\
&= frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}),
end{align}
as required.
$endgroup$
Here is one possible approach.
Enforcing a substitution of $x mapsto 1/x$ to begin with gives
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -int_{sqrt{2}-1}^1 frac{ln x}{1 - x^2} , dx = -sum_{n = 0}^infty int_{sqrt{2} - 1}^1 x^{2n} ln x , dx,
end{align}
where we have taken advantage of the geometric sum for $1/(1 - x^2)$.
Integrating by parts then leads to
begin{align}
int_1^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= - ln (1 + sqrt{2}) sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} + sum_{n = 0}^infty frac{1}{(2n + 1)^2}\
&qquad - sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2}. tag1
end{align}
Observing that
$$tanh^{-1} z = sum_{n = 0}^infty frac{z^{2n + 1}}{2n + 1}, qquad |z| < 1, qquad (*)$$
the first of the sums appearing in (1) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{2n+1} = tanh^{-1} (sqrt{2} - 1) = frac{1}{2} ln (1 + sqrt{2}).$$
To find the second and third sums in (1), divide ($*$) by $x$ before integrating up from $0$ to $x$. Thus
begin{align}
sum_{n = 0}^infty frac{1}{2n + 1} int_0^x t^{2n} , dt &= int_0^x frac{tanh^{-1} t}{t} , dt\
sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} &= frac{1}{2} int_0^x ln left (frac{1 + t}{1 - t} right ) frac{dt}{t}\
&= frac{1}{2} int_0^{-x} frac{ln (1 - t)}{t} , dt - frac{1}{2} int_0^x frac{ln (1 - t)}{t} , dt
end{align}
or
$$sum_{n = 0}^infty frac{x^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (x) - operatorname{Li}_2 (-x) right ], qquad (**)$$
where $operatorname{Li}_2 (x)$ is the dilogarithm function.
The second and third sums appearing in (1) can now be found using ($**$). For the second sum, setting $x = 1$ gives
$$sum_{n = 1}^infty frac{1}{(2n + 1)^2} = frac{1}{2} operatorname{Li}_2 (1) - frac{1}{2} operatorname{Li}_2 (-1) = frac{1}{2} left (frac{pi^2}{6} + frac{pi^2}{12} right ) = frac{pi^2}{8},$$
where the well-known special values for the dilogarithm function at 1 and $-1$ have been used.
For the third sum, setting $x = sqrt{2} - 1$ gives
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{1}{2} left [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) right ]. qquad (dagger)$$
It now remains to express the difference between the two dilogarithms is terms of elementary constants.
To do this, we will make use of the following identities for the dilogarithm:
begin{align}
operatorname{Li}_2 (1-x) + operatorname{Li}_2 left (1 - frac{1}{x} right ) &= -frac{1}{2} ln^2 x tag2\
operatorname{Li}_2 (x) + operatorname{Li}_2 (1-x) &= frac{pi^2}{6} - ln x ln (1 - x) tag3\
operatorname{Li}_2 (x) + operatorname{Li}_2 (-x) &= frac{1}{2} operatorname{Li}_2 (x^2) tag4
end{align}
So here we go.
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) &= operatorname{Li}_2 [1 - (2 - sqrt{2})]\
&= -operatorname{Li}_2 left (1 - frac{1}{2 - sqrt{2}} right ) - frac{1}{2} ln^2 (2 - sqrt{2}) qquad text{(by (2))}\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 left (frac{sqrt{2}}{1 + sqrt{2}} right )\
&= - operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) -frac{1}{2} left [frac{1}{2} ln 2 - ln (1 + sqrt{2}) right ]^2\
&= -operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 + frac{1}{2} ln 2 ln (1 + sqrt{2}) - frac{1}{2} ln^2 (1 + sqrt{2})
end{align}
And
begin{align}
operatorname{Li}_2 (1 - sqrt{2}) &= -operatorname{Li}_2 left (1 - frac{1}{sqrt{2}} right ) - frac{1}{2} ln^2 sqrt{2} qquad text{(by (2))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + ln left (frac{1}{sqrt{2}} right ) ln left (1 - frac{1}{sqrt{2}} right ) - frac{1}{8} ln^2 2 qquad text{(by (3))}\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} -frac{1}{2} ln 2 ln left (frac{1}{sqrt{2}(1 + sqrt{2})} right ) - frac{1}{8} ln^2 2\
&= operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 left [frac{1}{2} ln 2 + ln (1 + sqrt{2}) right ] - frac{1}{8} ln^2 2\
&=operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) - frac{pi^2}{6} + frac{1}{2} ln 2 ln (1 + sqrt{2}) + frac{1}{8} ln^2 2.
end{align}
Thus
begin{align}
operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2}) &= - left [operatorname{Li}_2 left (frac{1}{sqrt{2}} right ) + operatorname{Li}_2 left (-frac{1}{sqrt{2}} right ) right ] +frac{pi^2}{6} - frac{1}{2} ln^2 (1 + sqrt{2}) - frac{1}{4} ln^2 2\
&= -frac{1}{2} operatorname{Li}_2 left (frac{1}{2} right ) + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) - frac{1}{4} ln^2 2 qquad text{(by (4))} \
&= -frac{1}{2} left [frac{pi^2}{12} - frac{1}{2} ln^2 2 right ] + frac{pi^2}{6} - frac{1}{2} ln^2(1 + sqrt{2}) -frac{1}{4} ln^2 2\
&= frac{pi^2}{8} - frac{1}{2} ln^2 (1 + sqrt{2}).
end{align}
Here the well-known value for the dilogarithm function at $x = 1/2$ has been used.
So the sum in ($dagger$) can be expressed as
$$sum_{n = 0}^infty frac{(sqrt{2} - 1)^{2n + 1}}{(2n + 1)^2} = frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}).$$
Putting the final pieces of the puzzle together, (1) becomes
begin{align}
int_{1}^{1 + sqrt{2}} frac{ln x}{x^2 - 1} , dx &= -frac{1}{2} ln^2 (1 + sqrt{2}) + frac{pi^2}{8} - left [frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}) right ]\
&= frac{pi^2}{16} - frac{1}{4} ln^2 (1 + sqrt{2}),
end{align}
as required.
edited Jan 10 at 12:54
answered Jan 10 at 7:39
omegadotomegadot
5,5402728
5,5402728
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
add a comment |
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
Actually, the solution can be simplified by using the chi function's known value which is re-proved by you.
$endgroup$
– Kemono Chen
Jan 10 at 8:12
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
$begingroup$
That is correct, but I thought perhaps the dilogarithm function and some of its special values would be more familiar to work with rather than $chi_2 (x)$, the Legendre chi function of order 2. Some special values for this function are also known and can be found here, one of which is the value I found for ($dagger$), namely $chi_2 (sqrt{2} - 1) = frac{1}{2} [operatorname{Li}_2 (sqrt{2} - 1) - operatorname{Li}_2 (1 - sqrt{2})] = frac{pi^2}{16} - frac{1}{4} ln^2(1 + sqrt{2})$.
$endgroup$
– omegadot
Jan 10 at 9:52
1
1
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
$begingroup$
Really neat! (+1) Good work
$endgroup$
– clathratus
Jan 10 at 15:19
add a comment |
Thanks for contributing an answer to Mathematics Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3068155%2fhow-to-prove-that-int-1-sqrt21-frac-lnxx2-1dx-frac-pi2%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
$begingroup$
The substitution $u=x^2$ turns it into the dilogarithm integral, right?
$endgroup$
– Ant
Jan 10 at 2:29
1
$begingroup$
Oops, my bad. The anti-derivative is $frac{1}{4}textrm{Li}_2(1-x^2)-textrm{Li}_2(1-x)$, though, so maybe there's another way...
$endgroup$
– Ant
Jan 10 at 2:46
1
$begingroup$
It's straightforward to see how that anti-derivative differentiates to your integrand; so reversing that says that you can re-write your integrand as the sum of two terms (and you need to use $u=x^2$ on one of them). But using the various dilogarithm identities to evaluate at your bounds is tricky.
$endgroup$
– Ant
Jan 10 at 3:01
3
$begingroup$
A possible approach $$I=int_{1}^{sqrt{2}+1}frac{ln{x}}{x^{2}-1}dx\ =int_{pi/4}^{3pi/8}frac{lntan x}{tan^2x-1}sec^2xdxtext{ (tangent substitution)}\ =int_{pi/4}^{3pi/8}-lntan xsec2xdxtext{ (trig identity)}\ =int_{pi/4}^{3pi/8}-csc2xleft(lncscleft(x-fracpi4right)+lnsinleft(fracpi4+xright)right)dx-frac12ln^2(1+sqrt2)text{ (IBP)}\ =int_{0}^{pi/8}sec2xlncot xdx-frac12ln^2(1+sqrt2)text{ (linear substitution)}\ =int_{0}^{pi/8}sum_{n=0}^inftyfrac{cos(4n+2)x}{2n+1}sec 2xdx-frac12ln^2(1+sqrt2)text{ (Fourier series)}$$
$endgroup$
– Kemono Chen
Jan 10 at 3:43