LECTURE NOTE   2013.10.29          to index               2013.10.17 - previous lecture note               2013.11.05 - next lecture note


Interaction of Atoms with Light - Part 3

The PDF-file of the presentation : Light_Atom-2013.10.16.pdf            SCF-exercise COMMENT at the end

1. Time-dependence in quantum mechanics - probability rate - short comment
2. Eigenmodes / Normal modes
3. Operator method for harmonic oscillations - creation/annihilation operators

We started with a small review of the time-dependence - Golden Rule - And why the "constant rate" case
depends on "how much is left"  - Illustration by analogy of a flow out of a jar

      010_constant_rate_depends_on_content.png

       010_constant_rate_depends_on_content.png

Operator method for harmonic oscillations  - transforming to dimensionless Q, P - energy h omega outside
Q,P can be combined to a and a+     - commutation relations - and suddenly we can construct the "NUMBER OPERATOR"
H is proportional to N, thus we only need to work with the combination   a ,   a+ and N= a+ a

Provided we know an eigenstate - apply  a+  or a to it  - and it turns you get a NEW eigenstate of N with value n+1 or n-1
    020_operator_method_for_Harmonic_oscillator.png

       020_operator_method_for_Harmonic_oscillator.png

Operator method for harmonic oscillations  - continue
Thus   a and  a+ are LADDER¤ operators
AT the end we obtain a MUCH simpler differential equation for the ground state

Algebraic method for Harmonic oscillator is elegant, but also very important
 ---   it provides the creation - anihilation operator formalism
      030_operator_method_for_Harmonic_oscillator.png

       030_operator_method_for_Harmonic_oscillator.png

Operator method for harmonic oscillations  - normal modes - QUANTA
creation operator "creates a quantum" of energy of the "mode"
PHONON    -   or perhaps also PHOTON ?

Towards the end we started to discuss the spatial distribution of a normal mode "wavefunction"
in real space coordinates, not in the normal coordinates
      040_creation-annihilation_phonon_and_photon.png

       040_creation-annihilation_phonon_and_photon.png

Here are normal coordinates demonstrated in MATLAB. Matlab finds the eigenvalues and eigenvectors
for a coupled string. IT TURNS OUT THESE ARE KNOWN ANALYTICALLY (Toeplitz Matrix, see later)
- but Matlab just does it quickly. There are "quasicontinuously" distributed eigenfrequency values OMEGA Squared
- and corresponding eigenvectors  - the eigenvectors are plotted. They show the eigenmodes as distributed
over "the string"  - it really resembels the eigenmodes of a string (gitar or piano string ...)
N=100 in both cases below, i.e. string of 100 balls
      205_n_3_Oscillator.png

       205_n_3_Oscillator.png

Above and under:   above - the third lowest frequency    n=3
under the third highest frequency - the lines are drawn to show the neighboring displacements     n=98
N=100 in both cases below, i.e. string of 100 balls
      210_n_98_Oscillator.png

       210_n_98_Oscillator.png

Discussion of how the vibration's - photon's wavefunction looks like in the normal space,
i.e. in terms of the INDIVIDUAL DISPLACEMENTS (classically - simply what shown above, MATLAB)
Quantum case - more complicated? Wavefunctions?

These board-shots are from a solid state lecture - spring 2010
      110-normal-modes-eigenmodes.png

       110-normal-modes-eigenmodes.png

These board-shots are from a solid state lecture - spring 2010
It turns out that these drawings are in fact not quite correct - useful for discussion anyway
      120-Harm.Osc.States-and-normal-modes.png

       120-Harm.Osc.States-and-normal-modes.png

These board-shots are from a solid state lecture - spring 2010
It turns out that these drawings are in fact not quite correct - useful for discussion anyway
see a correct description below
      130-Harm.Osc.States-and-normal-modes.png

       130-Harm.Osc.States-and-normal-modes.png

Here is the lecturer's HOMEWORK  (not yet at the lecture)
first slide - the formal operations clarified, The eigenvectors arranget in matrix S,
the transposed matrix is also an INVERSE matrix - from the definition of normalized
eigenvectors

      140_Harm_osc_2013_explanation.png

       140_Harm_osc_2013_explanation.png
Ground state of all the normal modes - and then written in terms of the individual displacements
 (blue-outlined parts)
What are the "individual" a+i   - creations of vibrations with the MODE FREQUENCY but displacement in the given point
THEY ARE NOT real vibrations of the given ball - those would have all the independent vibration frequency.
They "do not exist" - system is strongly coupled, no uncoupled are reasonable

Here is the correct explanation of how the 1 (static) phonon state looks like
alpha-th eigenmode, all other modes in their ground state
Then use the transformation compare with definition above - and read the explanation below
      150_Harm_osc_2013_explanation.png

       150_Harm_osc_2013_explanation.png
The "photon" is smeared out - it is a sum of many terms, in each term the oscillation is concentrated in one "ball" displacement
- with amplitude following from the eigenmode.
In terms of probabilities: we would measure distribution of values of the displacement of the i-th ball
to be just like the width of the 1. excited state with probability amplitude squared - as shown - call it P(i).
It will be found to correspond to its  "ground state"  with the probabilty of squared sums remaining - which is 1- i.e.
1-P(i)          ----  a bit difficult to plot 
----- see also above --------
What are the "individual" a+i   - creations of vibrations with the MODE FREQUENCY but displacement in the given point
THEY ARE NOT real vibrations of the given ball - those would have all the independent vibration frequency.
They "do not exist" - system is strongly coupled, no uncoupled are reasonable.

The last/colored part: about coherent states; see lasers later on



COMPUTATIONAL EXERCISE DISCUSSION

A more detailed project is described here
../2011_09_27/SCF-Herman-Skillman/index.html
in particular the fitting (guessing) of the Z-dependence

There are many tools and simple methods to do such "variational" procedure
We have looked at how to do it in a SPREADSHEET - using a function definition
This is what remained as a note:
      300_how_to_fit_z-dependence_spreadsheet.png

       300_how_to_fit_z-dependence_spreadsheet.png

LECTURE NOTE   2013.10.29          to index               2013.10.17 - previous lecture note               2013.11.05 - next lecture note