Electrostatics: Permanent multipole moment interactions and atomic dipole polarization.
@Michele_Laino
we have to approximate, using the Taylor expansion, this quantity: \[\int {d{\mathbf{r}}'} \frac{{{{\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|}^2}}}{{\left| {{\mathbf{r}} - {\mathbf{r}}'} \right|}}\] which compares into the formula for the potential at point \(r\): \[\phi = \sum\limits_i {\frac{{{Z_i}}}{{\left| {{\mathbf{r}} - {{\mathbf{R}}_i}} \right|}}} - \int {d{\mathbf{r}}'} \frac{{{{\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|}^2}}}{{\left| {{\mathbf{r}} - {\mathbf{r}}'} \right|}}\]
such potential is written in units \(e=1\), wherein \(e\) is the electronic charge
now, using the Taylor expansion, I approximate this quantity: \[\frac{1}{{\left| {{\mathbf{r}} - {\mathbf{r}}'} \right|}}\] before integrating
so I can write this: \[\frac{1}{{\left| {{\mathbf{r}} - {\mathbf{r}}'} \right|}} \simeq \frac{1}{r} + \frac{{r'}}{{{r^2}}}\cos \theta + \frac{1}{{{r^3}}}\frac{{3{{\left( {\cos \theta } \right)}^2} - 1}}{2}r{'^2}\]
wherein \(\theta\) is the angle between the vectors \(r\) and \(r'\)
such expansion is up to the second order, of course we can go beyond, namely we can compute the term corresponding to the third order, and so on Now I substitute into such integral, so I get:
\[\begin{gathered} \phi = \sum\limits_i {\frac{{{Z_i}}}{{\left| {{\mathbf{r}} - {{\mathbf{R}}_i}} \right|}}} - \int {dr'} \frac{{{{\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|}^2}}}{{\left| {{\mathbf{r}} - {\mathbf{r}}'} \right|}} \simeq \hfill \\ \hfill \\ \simeq \sum\limits_i {\frac{{{Z_i}}}{{\left| {{\mathbf{r}} - {{\mathbf{R}}_i}} \right|}}} - \frac{1}{r}\int {dr'} {\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|^2} - \frac{1}{{{r^2}}}\int {dr'} {\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|^2}r'\cos \theta + \hfill \\ \hfill \\ - \frac{1}{{{r^3}}}\int {dr'} {\left| {\Psi \left( {{\mathbf{r}}'} \right)} \right|^2}\frac{{3{{\left( {\cos \theta } \right)}^2} - 1}}{2}r{'^2} \hfill \\ \hfill \\ \end{gathered} \]
the second term of our expansion, is like a dipole moment
now, it is more easy to compute the electric field, due to such potential, by means of this equation: \[\Large {\mathbf{E}} = - \nabla \phi \]
maybe, we have, in the similar way, to expand this part also: \[\Large \sum\limits_i {\frac{{{Z_i}}}{{\left| {{\mathbf{r}} - {{\mathbf{R}}_i}} \right|}}} \]
I was just about to ask, why wouldn't we expand the sum of nuclei?
you question is right! I think that better is if we expand the first contribution
here is the corresponding formula, up to the second term:
\[\begin{gathered} \sum\limits_i {\frac{{{Z_i}}}{{\left| {{\mathbf{r}} - {{\mathbf{R}}_i}} \right|}}} \simeq \frac{1}{r}\sum\limits_i {{Z_i} + \frac{1}{{{r^2}}}} \sum\limits_i {{Z_i}{R_i}\cos \theta } + \hfill \\ \hfill \\ + \frac{1}{{{r^3}}}\sum\limits_i {{Z_i}R_i^2\frac{{3{{\left( {\cos \theta } \right)}^2} - 1}}{2}} \hfill \\ \end{gathered} \]
now, you have to put together all such terms at the right side, and then to compute the electrostatic field using this formula: \[{\mathbf{E}} = - \nabla \phi \]
technically speaking, what we have done is the multipole expansion of the electric potential, due to the nuclear charges and all the electrons charges
oops.. of course, into the second expantion, \(\theta\) is the angle between the vectors \(r\) and \(R_i\)
expansion*
Ahh I see. But this is an expansion around the electric field as I also tried to follow a little bit in a tutorial video. What confused me was when they went to matrix notation in the book I sent you and evaluate the energy of the different multipoles. Any way you could explain that?
Or rather multipole interactions.
I'm looking at your textbook
sorry, where is such matrix?
sorry, it is the polarizability tensor
I was specifically thinking what was happening in equation 2.27 and afterwards.
Cause I can see that must be the way he derive the expression: \[E_{el}=q_0\phi -\bf \mu_0 F - \frac{ 1 }{ 2 } \Theta_0 F'-\frac{ 1 }{ 2 } \alpha F^2\] Where it should include the first 3 terms of the multipole expansion and the fifth term.
here we have an interaction between other molecules, located at othr sites, and a specific molecule located at a specific site . Such molecules create an electric field \(E^fic molecule. Such induced electric dipole moment \(p^{ind}\), is given by the subsequent tensor equation: \[\Huge {p^{ind}}_i = {\alpha _{ij}}E_j^0\] wherein the quantities: \[\huge {\alpha _{ij}}\] are the so called polarizability tensor
The polarizability happens from an induced dipole right?
Ah what you wrote. Sorry my bad,
that's right! oops.. I meant this: Such molecules create an electric field \(E^0\). Such induced electric dipole moment\(p^{ind}\)
we can think the induced electric dipole oment as an answer from a specific molecule under the action of an external electric field. In our case such external electric field is the electric field due to the other molecules
moment*
I see.That explains why would should do external potential in eq. 2.27, but because my linear algebra sucks I must ask: Why do we do both with and without an external potential?
Ah charge transfer due to external potential and then solved by Coulombs law.
in order to check such formula, you have to perform this computation: \[\Large {\mathbf{E}} = - \nabla \phi \] when we want to compute the induced electric dipole moment by a specific molecule, we need to know the electric potential creates by other molecules at the site at which is our molecule
please keep in mind that electric fields have the property of superposition
Oh, I actually didn't think about that, nor knew really.
here is a simple example: |dw:1451498394432:dw| each of the four water molecules create an electric field, described by the formula above and the total electric field at position of molecule A, is given by the vector sum of each vector field from the single molecules 1, 2, 3 or 4, as if each molecule 1, 2 3 or 4 is present alone without the others
namely, we can write: \[{{\mathbf{E}}_A} = {{\mathbf{E}}_1} + {{\mathbf{E}}_2} + {{\mathbf{E}}_3} + {{\mathbf{E}}_4}\] wherein each of four fields \(E_i,\;i=1,2,3,4\) is given by the computation above
I see. That does indeed make sense yes.
such field \(E_A\) is the field which produces the induced electric dipole moment of molecule A
Right.
Another I'd like to ask is: why is eq. 2.30 right? I feel I am missing the link between the energy and the electric field.
It is the energy needed to produce such dipole inside an external electric field. Such energy \(W\) is given by the subsequent formula: \[\huge W = \frac{1}{2}{\mathbf{P}} \cdot {\mathbf{E}}\] wherein \(P\) is the induced electric dipole moment, and \(E\) is the external field
from the theory of electricity, and in particular from dielectrics, we can write this little proof: \[\begin{gathered} W = \int\limits_V {dV} \int\limits_0^{{E_0}} {PdE} = \int\limits_V {dV} \int\limits_0^{{E_0}} {\chi EdE} = \int\limits_V {\frac{{\chi E_0^2}}{2}dV} \hfill \\ \hfill \\ \frac{W}{V} = \frac{{\chi E_0^2}}{2} = \frac{1}{2}PE \hfill \\ \end{gathered} \] wherein \(W\) is the needed energy to polarize a volume \(V\)
oops.. we can write this \(simple\)* proof
e \(P=\alpha F\) and \(E_0=F\), so, we get: \[\huge \frac{W}{V} = \frac{{\chi E_0^2}}{2} = \frac{1}{2}P{E_0} = \frac{1}{2}\alpha {F^2}\]
furthermore \(\chi\) is the electrical susceptibility of the medium
that is a simple reasoning in order to understand the factor \(1/2\)
nevertheless, what is important, is that it represents the needed energy to make such induced electric dipole moment
of course, the susceptibility \(\chi\) is proportional to the polarizability \(\alpha\)
I see. That explains eq. 2.30 much more!
Last thing I wanted to ask is: Do you have any idea what the expansion is in the picture I send you and how to derive it? I guess it must be a continuation of the expansion we have talked about? The very last equation in the picture with the missing symbols unfortunately. It is a lecture given by the same person who wrote this book.
Like: \(q_0\phi \) is the monopole interaction? \( -\bf \mu_0 F\) is the dipole interaction? And so on?
While: \(-\frac{ 1 }{ 2 } \alpha F^2\) is the first order polarizability?
I thik that we have to continue my procedure above, namely after the expansion of the integral and the summation, we have to apply this formula: \[\Large {\mathbf{E}} = - \nabla \phi \] Please try to perform such computation, and let me know what you get. I will try to do such computation on my own and I will tag you when I got the expected result Furthermore, if you need to know some basic of electromagnetism, please refer to this textbook: \[\begin{gathered} {\text{E}}{\text{. M}}{\text{. Purcell}} \hfill \\ {\mathbf{Electricity and Magnetism}} \hfill \\ {\text{The Berkeley Physics Course}}{\text{. Vol}}{\text{.II}} \hfill \\ \end{gathered} \]
at first sight I think you are right, and now we can understand the reason of such terms
or, you can ask me for electromagnetism, and I will be happy to help you
Any mathematical fundament required to read the book besides calculus and linear algebra?
This topic is basically what I plan on doing for my master degree so I bet there will come many funny questions :)
only calculus, since it is a very simple, and well written textbook
ok! :)
And thank you so much for your help so far it helped quite a lot. :)
Join our real-time social learning platform and learn together with your friends!