Monday, March 11, 2013

Final preparation problems.

Please keep posting more suggestions for problems. If you collaborate and participate then it is more likely that the final will not be incomprehensible and problems that are difficult to understand or pitched at the wrong level.

I'll post some easier problems later. I am still waiting for more feedback in the post "What would you like to see on the final."

1) Consider a graphene sheet that is 2 mm long and 4 nm wide. Suppose that you dope the right half (Let's call that x>0) with donors to get a carrier density of 10^12 carriers/cm^2. Now suppose you role that into a 2mm long, narrow tube.
a) what is the radius of the tube?
b) what is the linear carrier density
c) If you apply an electric field that gives an (average) speed of 10^7 cm/sec to the carriers moving to the left, how many will arrive at the x=0 boundary per second?

2) With regard to the same half-doped graphene tube:
a) can the moving electron coming from the doped x>0 side cross the x=0 boundary? Why or why not? (the left half, x < 0, is undoped.)
b) will they fall into the valence band? why or why not?
c) If not, what do you need to do to enable them to drop into the valence band?
d) assuming you have solved part c) and that each one dropping into the valence band emits a photon,
how much radiated power would you get from one such "nanotube"... if the band gap were engineered to be 0.5 eV.

3. Previously we discussed that in the presence of an oscillating E-field of just the right frequency, the probability of a transition from one quantum state to another is proportional to the electric field squared times an integral of the wave-function of the initial state, the wave-function of the final state and x (where the x factor comes from the interaction potential $e E_x x$ (for an electric field polarized in the x direction).  Starting from this, discuss how a laser works. Compare and contrast a laser and and LED made with the same semiconductor. What are the similarities and differences?
============

4. extra credit: a) Write the equation for a crystal state of wave-vector k (assuming you know the atom state that the crystal state evolves from). For a bandwidth of 4 eV, what is the velocity associated with theis crystal state? (Do you need to be given anything else to calculate that?)
c)  Discuss what the animation below shows. [Hint: it is a Gaussian wave packet of crystal states center NOT at k=0. The packet is pretty narrow. This is for a 1D crystal with b=0.1 nm. Discuss: Around what k-value is this packet centered. How wide is it (in k and in x)? and other relevant things. Explain your reasoning. Please feel free and encouraged to discuss this here (in the comments).



 5. Consider an electron in a harmonic oscillator potential. Suppose that at t=0 it is in the state
$\Psi (x,0) = (c_o + c_1 x + c_2 x^2) e^{-x^2/2a^2}$.
a) Discuss the time dependence of this state? Is it time dependent?
b) How would you calculate $\bar{x}$? Would it depend on time?

6. Referring to hydrogen atom states,
a) write the state $\psi_{2,1,3/2,1/2}$ in terms of $\psi_{n.l.m}$ states (with spin).
b) extra credit: write the state $\psi_{2,1,3/2,1/2}$ in terms of $\psi_{2x}$ type states (with spin).

7. Suppose an electron in a hydrogen atom is in a state that is an equal mix of the ground state and 1st excited state. Pick a particular 1st excited state and for your choice:
a) Describe/discuss what you would expect to get if you calculated $\bar{x}$,  $\bar{y}$ and $\bar{z}$ (include a graph or graphs) in your description.
b) Calculate $\bar{x}$,  $\bar{y}$ and $\bar{z}$ (include a graph or graphs).

8. Describe a p-n junction and how it can be used to create an LED or laser. Which way do the electrons flow? what happens? etc  extra credit for elucidating the difference between an LED and a Laser.

9. For an electron in a harmonic oscillator potential that is in a state that is an equal mix of the ground state and 1st excited state,
a) calculate and graph $\bar{x}$? Would it depend on time?

10. Same as 9, but instead of an equal mix suppose that the coefficient of the g.s. term is 0.9.
a) calculate and graph $\bar{x}$? Compare your results from 9 and 10.

10.1  Same as 9, but instead of an equal mix suppose that the coefficient of the g.s. term is 0.99.
a) What is the coefficient of the 1st-excited state?
b) calculate and graph $\bar{x}$? Compare your results from 9 and 10.

11. Suppose you have a biased (forward) p-n junction with a junction area of 0.1 cm^2. If the average electron speed is 5 x 10^6 cm/sec and the electron density is 10^17/cm^3, how much power would be radiated in either an LED or Laser (assume 100% efficiency). You can assume an equal and opposite flow of holes from the p side to the n side.

12. For an electron in the ground state of an infinite square well of width 0.5 nm, what is the expectation value in eV of:
a) the kinetic energy,
b) the potential energy.
c) why does an electron in a well have kinetic energy?

13. Consider a finite square well of depth 20 eV that has exactly (only) 3 bound states. One is 2.5 eV above the energy of the bottom of the well, the next 9 eV (same reference) and the 3rd at 17 eV.
a) sketch the states.
b) Describe and sketch the optical absorption spectrum that you would predict for this system. Label all sharp lines corresponding to transitions with their initial and final states and indicate their energy.
c) Assuming the strength of any transition is related to x integrated with the initial and final states, which transitions would be unobservable?

35 comments:

  1. Whats the band gap for the graphene?

    ReplyDelete
    Replies
    1. Yes, that is a key question. A lot of people, not just in our class, are thinking about that.

      Delete
    2. So we learned from our substitute today that there is no band gap for graphene?.... And people want to try to create one?

      Delete
  2. Don't we need a band gap in order to calculate the power radiated by the tube?

    ReplyDelete
    Replies
    1. Yes. So really, one can't answer the question, at least not in any simple way. It leads to other questions, primarily about the gap.

      See for example:

      http://en.wikipedia.org/wiki/Graphene_nanoribbons

      http://arxiv.org/abs/cond-mat/0702511

      http://www-als.lbl.gov/index.php/holding/218-controlling-graphenes-electronic-structure.html

      Delete
  3. Regarding problem 5, time dependence arises when we have a mixed state, i.e., the sum of the ground and first excited states of a 1DHO, right? Here it looks like the ground, first, and second excited states are summed. So I supposed my question is that; are these three states summed here? If they are, then is this state time dependent?

    ReplyDelete
    Replies
    1. This is a difficult state to analyze. Best to go step by step. After some trial and error one realizes:
      1) the x^2 term implies some 2nd excited state, and that also takes some of the constant term.
      The x term is 1st excited state related.
      Unless co is exactly c2/2 there is also some ground state. Does that make any sense? It make take some time. If you can understand this state, and how to normalize it, then you are ready for anything.

      Delete
  4. regarding number 11, arent we supposed to be given the gap energy to find the power?

    ReplyDelete
    Replies
    1. actually, if we calculate the amount of electrons/sec, could we count that as the frequency? thus, for each electron, the energy would be E=hf? but that doesnt sound right to me, because we want joules per second (watts), and if we just multiply the amlount of electrons per second times that energy we'd get right back to where we started wouldnt we? or perhaps not...

      Delete
    2. You are right the first time. You definitely need the gap. Let's say 0.8 eV.

      The other frequency is interesting but not really related, as you say.

      Delete
    3. okay, i think i got it now, now it makes so much more sense, thanks! should i get something to the power of 10^-7? in watts?

      Delete
    4. Hmm. I get close to 10^4 watts. (more than 1000 watts)

      Delete
    5. we got 6400 watts. is that right?

      Delete
    6. I'm also getting ~ 6400 at 6408.

      Delete
    7. 6408 watts ***

      Delete
  5. for number 9 and ten, we would need to normalize our resulting combined wavefunction right? so for number 9, the normalized wavefunction would be multiplied by a factor of 1/sqrt(2), and for number nine the wavefunction would be multipled by a factor of 1/(sqrt(.9^2 + 1^2))?

    ReplyDelete
    Replies
    1. For number 9, I agree.

      For number 10, the coefficients would be 0.9, which is given, and sqrt[1-.9^2], which is implied by the normalization constraint. Does that make sense??

      Delete
    2. I just added another version of that problem, 10.1. For 10.1 I get a coefficient of 0.14 for the first excited state.

      For problem 10 I got a coefficient that was around 0.44. Does that seem correct?

      Delete
    3. why are we subtracting the .9^2 instead of adding it to 1? for number ten when normalizing?

      Delete
    4. for number 10, the coefficients squared and then added together has to be one.
      try it for these:
      .9, .44.
      and then try:
      .99, .14

      Delete
  6. This comment has been removed by the author.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. lattice constant. (distance between atoms in our 1D model). for us, 0.1 nm.

      Delete
  8. for #7, if you use $\psi _{2r}$ state for the first excited state with the ground state, since theres equal dependence on x, y, and z, the expectation value goes to zero. Is this intended or was this one that you didnt want us to really use?

    ReplyDelete
    Replies
    1. I wanted you to think about all of them. On the actual final questions would be more specific.

      Delete
  9. In fine structure for the Hydrogen atom, do we need to include the relativistic term for momemtum (p^4 term) due to the velocity involved in the electrons total angular momentum? Otherwise what velocity would we be considering?

    ReplyDelete
    Replies
    1. well p, in quantum physics, is NOT the velocity of a dot, it comes from i hbar d/dx. So as I understand it, p^4 can is non-zero for all states, including l=0 states that have zero angular momentum.

      Delete
    2. Okay, so when we apply the momentum operator, ihbar d/dx, to the wave function of the electron, we find the need to include relativistic corrections?

      Delete
    3. well the p^2/2m term is the regular term, and the p^4/8m^3c^2 is the 1st relativistic correction.

      The p^2/2m is the $-\hbar^2/(2m) d^2/dx^2$ term.

      Delete
    4. Oh, right! The Schrodinger equation is about energy, and that's how Kinetic energy and momentum relate. Then the relatively large Kinetic energy for the electron is responsible for the large momentum, and in turn makes it necessary to include relativity on that order of binomial expansion. Because the standing and symmetric waveforms have Kinetic energy, then the p^4 term must be included for states with zero angular momentum as well, correct?

      Delete
  10. Also, for the Gaussian in number 4, are we looking at a leftward electric field applied to a conduction band? With a narrow wave packet (I'm assuming in x?) there would have to be a broad contribution from k, correct? Could this be a very strong electric field?

    ReplyDelete
    Replies
    1. yes, that all make sense. Although i would kind of say it is broad in x and narrow in k, but that is all relative...

      Delete
    2. Oh, yeah, I got it backwards, it would make more sense to be wide in x and have a very narrow k, since all the conducting electrons are sitting just a little above the Fermi level, so only a small number of k values would add to the total rightward momentum?

      Delete
  11. Will you be posting solutions to these problems?

    ReplyDelete