Find the solution of the problem $$u_t(x, t)-u_{xx}(x, t)=0, 0<x<1, t>0 \tag {*} \\ u(0, t)=0, t>0 \\ u_x(1,t)+u_t(1,t)=0, t>0$$

I have done the following:

We are looking for solutions of the form $$u(x, t)=X(x) \cdot T(t)$$

$$u(0, t)=X(0) \cdot T(t)=0 \Rightarrow X(0)=0 \\ X'(1) \cdot T(t)+X(1) \cdot T'(t)=0 \Rightarrow \frac{X'(1)}{X(1)}=-\frac{T'(t)}{T(t)}$$

$$(*) \Rightarrow X(x) \cdot T'(t)-X''(x) \cdot T(t)=0 \\ \Rightarrow \frac{X(x) \cdot T'(t)}{X(x) \cdot T(t)}-\frac{X''(x) \cdot T(t)}{X(x) \cdot T(t)}=0 \\ \Rightarrow \frac{T'(t)}{T(t)}=\frac{X''(x)}{X(x)}=-\lambda$$

So, we get the following two problems: $$\left.\begin{matrix} X''(x)+\lambda X(x)=0, 0<x<1\\ X(0)=0 \\ \frac{X'(1)}{X(1)}=\lambda \Rightarrow X'(1)-\lambda X(1)=0 \end{matrix}\right\}(1) $$

$$\left.\begin{matrix} T'(t)+\lambda T(t)=0, t>0 \end{matrix}\right\}(2)$$

For the problem $(1)$ we do the following:

The characteristic polynomial is $d^2+\lambda=0$.

  • $\lambda<0$:

    General solution: $X(x)=c_1 e^{\sqrt{-\lambda }x}+c_2e^{-\sqrt{-\lambda}x}$

$$X(0)=0 \Rightarrow c_1+c_2=0 \Rightarrow c_1=-c_2$$


$$X'(x)=\sqrt{-\lambda}c_1e^{\sqrt{-\lambda }x}-\sqrt{-\lambda}c_2 e^{-\sqrt{-\lambda}x} \\ X'(1)=\sqrt{-\lambda}c_1e^{\sqrt{-\lambda }}-\sqrt{-\lambda}c_2 e^{-\sqrt{-\lambda}}=\sqrt{-\lambda}c_1(e^{-\lambda}+e^{-\sqrt{-\lambda}}$$

$$X'(1)-\lambda X(1)=0 \Rightarrow \sqrt{-\lambda}c_1(e^{\sqrt{-\lambda}}+e^{-\sqrt{-\lambda}})-\lambda c_1(e^{\sqrt{-\lambda}}-e^{-\sqrt{-\lambda}})=0 \Rightarrow c_1 [ e^{\sqrt{-\lambda}}(\sqrt{-\lambda}-\lambda)+e^{-\sqrt{-\lambda}}(\sqrt{-\lambda}+\lambda)]=0$$

How could we continue to show that for $\lambda <0$ we get the trivial solution $X(x)=0$ ??



  • $\lambda <0$ :

    $X(x)=c_1 \sinh (\sqrt{-\lambda} x)+c_2 \cosh (\sqrt{-\lambda}x)$

    Using the initial values we get that $X(x)=0$, trivial solution.

  • $\lambda=0$ :

    $X(x)=c_1 x+c_2$

    Using the initial values we get that $X(x)=0$, trivial solution.

  • $\lambda >0$ :

    $X(x)=c_1 cos (\sqrt{\lambda}x)+c_2 \sin (\sqrt{\lambda}x)$

    $X(0)=0 \Rightarrow c_1=0 \Rightarrow X(x)=c_2=\sin (\sqrt{\lambda}x)$

    $X'(1)-\lambda X(1)=0 \Rightarrow \tan (\sqrt{\lambda})=\frac{1}{\sqrt{\lambda}}$

That means that the eigenvalue problem $(1)$ has only positive eigenvalues $0<\lambda_1 < \lambda_2 < \dots < \lambda_k < \dots $ that are the positive roots of the equation $\tan \sqrt{x}=\frac{1}{\sqrt{x}}$.

Is this correct??

Why can we say that the number of the eigenvalues is countable ??

How can we show that $$\lim_{k \rightarrow +\infty} \frac{\sqrt{\lambda_k}}{k \pi}=1$$ ??


It may be easier in this case to take the general solution as $$ X(x)=c_1\sinh \mu x+c_2\cosh \mu x,\quad \mu:=\sqrt{-\lambda}. $$ Then the first boundary condition implies that $c_2=0$ and the second is that either $$ c_1=0 $$ or $$ \tanh \mu =-\frac{1}{\mu}. $$ But the last equation has no solutions, hence $c_1=0$.

  • $\begingroup$ How do we know that the equation $$\tanh \mu=-\frac{1}{\mu}$$ has no solutions ?? @Artem $\endgroup$ – Mary Star Apr 6 '15 at 1:42
  • $\begingroup$ I have edited my initial post... Could you take a look at it ?? @Artem $\endgroup$ – Mary Star Apr 6 '15 at 2:59
  • $\begingroup$ @MaryStar It looks fine to me. To answer your two last questions try to sketch the graphs of $\tan h$ and $1/h$. $\endgroup$ – Artem Apr 6 '15 at 3:14

Using an integrating factor, we can show that the solution to $$ X''+\lambda X=0\tag{1} $$ for $\lambda\gt0$, is $$ X(x)=a\cos(\sqrt\lambda\,x)+b\sin(\sqrt\lambda\,x)\tag{2} $$ for $\lambda=0$, is $$ X(x)=a+bx\tag{3} $$ and for $\lambda\lt0$, is $$ X(x)=a\cosh(\sqrt{-\lambda}\,x)+b\sinh(\sqrt{-\lambda}\,x)\tag{4} $$ Furthermore, we also have $$ X(0)=0\tag{5} $$ and $$ X'(1)-\lambda X(1)=0\tag{6} $$

If $\lambda\gt0$, $(2)$ and $(5)$ say that $a=0$. Then $(2)$ and $(6)$ say that $$ b(\sqrt\lambda\cos(\sqrt\lambda)-\lambda\sin(\sqrt\lambda))=0\tag{7} $$ if $b=0$, then $X(x)=0$, otherwise, we need $$ \tan(\sqrt\lambda)=\frac1{\sqrt\lambda}\tag{8} $$ which has an infinite set of solutions, the smallest being $\lambda=0.7401738843949670422$.

If $\lambda=0$, $(3)$ and $(5)$ say that $a=0$. Then $(3)$ and $(6)$ say that $b=0$. Therefore, $X(x)=0$.

If $\lambda\lt0$, $(4)$ and $(5)$ say that $a=0$. Then $(4)$ and $(6)$ say that $$ b(\sqrt{-\lambda}\cosh(\sqrt{-\lambda})-\lambda\sinh(\sqrt{-\lambda}))=0\tag{9} $$ if $b=0$, then $X(x)=0$, otherwise, we need $$ \tanh(\sqrt{-\lambda})=-\frac1{\sqrt{-\lambda}}\tag{10} $$ which has no solutions since $\tanh(x)\gt0$ when $x\gt0$ and $\tanh(x)\lt0$ for $x\lt0$.

Thus, as you have added since I started writing this up, there is a sequence of solutions with $\lambda\gt0$, where $\lambda$ satisfies $(8)$.

As $\lambda$ gets larger, $\frac1{\sqrt\lambda}\to0$. The places were $\tan(x)$ is positive and near $0$ are slightly greater than multiples of $\pi$. Thus, $\sqrt\lambda$ must be slightly greater than $k\pi$ for some non-negative $k\in\mathbb{Z}$. Certainly, we can label each such $\lambda$ as $\lambda_k$. In that case, we have $$ \lim_{k\to\infty}\frac{\sqrt{\lambda_k}}{k\pi}=1\tag{11} $$

  • $\begingroup$ To divide with $X(1)$ are we sure that it isn't zero?? $$X'(1) \cdot T(t)+X(1) \cdot T'(t)=0 \Rightarrow \frac{X'(1)}{X(1)}=-\frac{T'(t)}{T(t)}$$ ?? @robjohn $\endgroup$ – Mary Star Apr 6 '15 at 17:28
  • $\begingroup$ Indeed. I am not sure why you are bringing this up. Are you worried that we get that $a=0$ and $b=0$ imply that $X(x)=0$? Well, $X(x)=0$ is a solution, so we don't need to worry about the division by $0$. However, if you want to worry about it, you can assume that $X(x)\ne0$ for some $x$ and then when we derive $X(x)=0$, it is a contradiction; that is $b\ne0$. $\endgroup$ – robjohn Apr 6 '15 at 17:58
  • $\begingroup$ Ok... When we have an other problem and we end up having $$ \tanh(\sqrt{-\lambda}\pi)=-\frac1{\sqrt{-\lambda}}\tag{10} $$ would we say again that it has no solutions since $\tanh(x)\gt0$ when $x\gt0$ and $\tanh(x)\lt0$ for $x\lt0$ ?? @robjohn $\endgroup$ – Mary Star Apr 6 '15 at 23:27
  • $\begingroup$ Yes. That is just what I wrote. To be inclusive, I wanted to also knock out the negative $\sqrt{-\lambda}$, too. If we use $\sqrt{-\lambda}\gt0$, the left hand side is positive and the right hand side is negative. If we use $\sqrt{-\lambda}\lt0$, the left hand side is negative and the right hand side is positive. $\endgroup$ – robjohn Apr 6 '15 at 23:56
  • $\begingroup$ So can we write it in that way although at tan we have $\sqrt{-\lambda }\cdot \pi $ and at the other side of the equation just $\sqrt{-\lambda}$ ?? @robjohn $\endgroup$ – Mary Star Apr 7 '15 at 0:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.