GoConqr hat ein neues Design! Sag uns, wie es dir gefällt.

Karteikarten

Week 6 LNs

R S C ICMP Year 4 Englisch 29 Karten

Briefly state the differences between geometrical and phonon phonon scattering

Karten in diesem Stapel (29)

  1. 1Briefly state the differences between geometrical and phonon phonon scatteringBriefly state the differences between geometrical and phonon phonon scattering
  2. 2Do phonons carry energyYes both in wave and particles
  3. 3Describe the umklapp scattering processIn an umklapp∗
    (“U”) scattering process, the combined wave vector lies outside
    the first Brillouin zone. When reduced to the first Brillouin zone, the resulting
    vector 𝑲3 = 𝑲1 + 𝑲2 − 𝑮 generally points in a direction that is different from
    the initial propagation directions. This change in direction of the energy flow
    reduces the thermal conductivity very effectively. However, it requires phonons
    of sufficiently large wave vector — zone-boundary phonons — which are hardly
    populated at low temperatures because they also tend to have higher energies
    than the acoustic modes near the Brillouin zone centre.
  4. 4Describe normal scattering processIn a
    normal (“N”) scattering process, two phonons with relatively small wave vectors
    are combined to a third phonon, whose wave vector still lies in the first Brillouin
    zone. The overall direction of phonon propagation, and hence energy flow, is
    hardly affected. In a hypothetical crystal with only N scattering, the heat flow
    would be unhindered and the thermal conductivity infinite. Normal scattering
    does not contribute to limiting the thermal conductivity
  5. 5Explain briefly why the K3 wave vector goes in another direction during Umklapp scatteringNow, suppose the associated wave vectors of the two initial rightward-moving waves,
    k
    1
    and
    k
    2
    , are large i.e. we are no longer in the long wavelength limit. Then, the wave resulting from their merger has a wave vector of
    k
    3
    =
    k
    1
    +
    k
    2
    . However,
    k
    3
    is larger than the span of the Brillouin zone. What happens? This new wave now becomes leftward-moving! It's 'flipped' its direction or in German, is 'Umklapp'.
    Isn't it strange? Two rightward-moving waves merged to give you a... leftward-moving wave when you should have had a rightward-moving one. It's going the 'wrong' way!
    By going the wrong way, you've lost part of your initial (crystal) momentum. This momentum dissipating process explains why phonons move diffusively i.e. they cannot maintain their momentum forever.
    In a sense, Umklapp scattering is just the Wagon-wheel effect but in real space.
    When extending the dispersion relation for either wave vectors outside the 1st BZ they have a negative velocity. This makes the wave vector K3 pointing the other way.
  6. 6Describe the three characteristic temperature regimes for temperature dependence on thermal conductivityThere are three characteristic temperature regimes for the temperature dependence of the thermal conductivity. We use the Debye temperature,
    𝜃Debye = ℏ𝜔𝐷/𝑘𝐵
    with the Debye frequency 𝜔𝐷, as a characteristic temperature given by the phonon
    spectrum of a particular material.
    Describe the three characteristic temperature regimes for temperature dependence on thermal conductivity
  7. 7Describe the thermal conductivity of the following diagramshows the thermal conductivities of a wider range of substances between
    1 K and room temperature. They all show an initial increase in thermal conductivity with increasing temperature. The crystalline materials all exhibit a maximum of the thermal conductivity, but in some cases it occurs above room
    temperature. At room temperature, isotopically pure 12C diamond crystals have
    the highest known thermal conductivity. The slight variation in atomic mass in
    natural crystals that contain about 1% of 13C represents a subtle defect, which
    reduces the thermal conductivity.
    Describe the thermal conductivity of the following diagram
  8. 8Describe the following graphThe increasing slope comes from the intermediate temperature regions up to the maximum of the curve. Following is the region at very low temperatures, the non linear appearance is from the T cubed law.Describe the following graph
  9. 9Describe a metals valence and core electrons and what leads to treating the conduction electrons in a metal as a gas of free particles that are subject to the Pauli exclusion principleWe distinguish between core and valence electrons. The latter have binding
    energies in the free atoms of at most a few eV, whereas the core electrons are
    bound to the nuclei much more strongly. In a metal, the valence electrons become
    detached from the ions and can move freely through the crystal. The ion cores
    occupy only in the order of 10% of the crystal volume†. Attempts to treat the
    conduction electrons in a metal as a gas of classical particles (Drude model) failed
    — the heat capacity due to the electrons was overestimated by two orders of
    magnitude. The fermionic nature of the electrons needs to be taken into account.
    This motivates the attempt of treating the conduction electrons in a metal as a
    gas of free particles that are subject to the Pauli exclusion principle. This is a
    brave approximation, and really only justified by its success.
  10. 10Briefly define independent electron approximation and free electron approximation from the free electron fermi gasAs in a classical (ideal) gas, collisions between particles are required so that that
    energy and momentum can be exchanged and a thermal equilibrium be obtained.
    Between the instantaneous collisions, the electrons are assumed to be completely
    free. They are assumed not to interact with the other electrons, which is the
    independent electron approximation. They are also assumed not to interact with
    the ion cores, which is the free electron approximation.
    With no collisions there are no mechanism to exchange energy.
  11. 11From the electrons in a box analogy, describe what leads to discrete electron energiesThe approximation of free and independent electrons means that we should start
    with a single particle in a box potential. The potential is given by all the ions and
    all other electrons. We replace the periodic potential of the ions with a constant
    potential and increase that potential to infinity to confine the electrons to the
    crystal. In reality, the potential height would be a few electron volts – the work
    function of a metal. We can now use the known solutions for the Schrödinger
    equation for a single electron in a box. The key point is that the boundary
    conditions lead to a spectrum of discrete electron energies.
    The equivalent calculation could now be performed for an electron in a threedimensional potential well, yielding again standing-wave solutions. In practice,
    it is more convenient to consider travelling-wave solutions together with the
    Born–von Karman periodic boundary conditions that we used previously for the
    phonons. As before, this leads to a discrete set of allowed wave vectors and hence a
    spectrum of discrete electron energies.
  12. 12What are wave functions 𝜓𝒌(𝒓) corresponding to the eigenstates of the Schrödinger equation also known as?Orbitals
  13. 13Briefly describe the 3D free electron fermi gasIn a 3D, there are BC, confinement to a box gives standing wave solutions. The travelling waves are generally easier and convenient.
    When adding a length to the crystal in the wave function the same wave function is produced. The Born Von Karman BC considers removing the surface from the crystal which is of finite size. There exists a discrete set of allowed wave vectors.
    For the solutions to the 3D Schrodinger equation, the 1/sqrtV is a normalisation factor. By integrating the probability density over the entire crystal, where probability density is a function where the density of a continuous random variable (in this case the electron) lies somewhere in this integral.
    Energies depend on the wave vector, where the energy is the kinetic energy of the travelling electrons.
  14. 14Describe populating the energy levels in a three dimensional electron gasRecall that every allowed wave vector in reciprocal
    space has a reciprocal volume of 8𝜋
    3/𝑉 attached, where 𝑉 is the volume of the
    crystal. When we populate the energy levels by increasing energy, we populate
    them simultaneously by increasing wave vector |𝒌|, starting from the Brillouin
    zone centre (𝒌 = 𝟎), just as in the one-dimensional case. The yields the Fermi
    wave vector 𝑘𝐹 as a function of the crystal volume, 𝑉 , and the total number of
    free electrons, 𝑁. The Fermi sphere encloses all occupied states. Again, all occupied
    states have energies 𝜖 ≤ 𝜖𝐹 and wave vectors |𝒌| ≤ 𝑘F
  15. 15Define Fermi velocity and fermi temperature Define Fermi energyThe Fermi energy and wave vector are related to a few other useful quantities:
    The Fermi velocity 𝑣𝐹 is the velocity of the highest-energy electrons, and the
    temperature equivalent of the Fermi energy is the Fermi temperature
    𝑇𝐹 = 𝜖𝐹/𝑘𝐵
    The Fermi energy 𝜖𝐹 is the kinetic energy of the highest-energy electrons in the
    Fermi gas at 0 K. In a classical gas the kinetic energy would be zero at 0 K. In
    sodium, the highest-energy electrons have a kinetic energy of 3.1 eV per electron.
    This is a direct consequence of Pauli’s exclusion principle, which forced us to
    populate higher-energy states. A classical gas would have to have a temperature
    of the order of 40 000 K to reach such kinetic energies — the quantum gas of
    fermions does that at zero Kelvin. It becomes clear now why classical models
    failed to describe the heat capacity of metals correctly
  16. 16Define the DOSThe derivative of the number of electrons up to a certain energy. Prop to the square root of the energy.
  17. 17Describe populating the available electron states at finite temperature Describe finding the chemical potential. In order to populate the available electron states at finite
    temperature, we need to use the Fermi-Dirac distribution function 𝑓 (𝜖), or more
    precisely 𝑓 (𝜖, 𝜇, 𝑇 ). Here, 𝜖 denotes the energy of an electron state, 𝜇 the chemical
    potential and 𝑇 the temperature of the system. In solid state physics, in particular
    semiconductor physics, the chemical potential is often called Fermi level — not
    to be confused with the Fermi energy, 𝜖𝐹 . In metals, they differ very little, but
    they are exactly the same only at 𝑇 = 0 K. In semiconductors, they are often very
    different.
    The chemical potential 𝜇 and its variation with temperature depend on the specific
    system, i.e., the density of states 𝐷(𝜖). 𝜇 needs to be chosen such that the correct
    number of electrons, 𝑁, is obtained when the number of populated states is
    integrated over all energies,
    𝑁 = ∫𝐷(𝜖) 𝑓 (𝜖, 𝜇, 𝑇 ) d𝜖. Even though the chemical
    potential appeared here through the Fermi-Dirac distribution, it is the same as the
    chemical potential used in the thermodynamics of classical gases and chemical
    reactions
  18. 18Define chemical potentialThe change in free energy when electrons are added or removed from the system.
  19. 19Describe the heat capacity plot of the following experimental dataThe low-temperature (few K) heat capacity 𝐶 of metals is non-linear. By plotting
    𝐶(𝑇 )/𝑇 as a function of 𝑇
    2
    , one usually obtains a linear relation, which means
    that the heat capacity can be described as 𝐶(𝑇 ) = 𝛾𝑇 + 𝐴𝑇 3.
    We recognise the
    cubic term as originating from the phonon heat capacity — Debye’s 𝑇 ^3
    law. The linear term must then originate from the electrons. The material constant 𝛾 that
    quantifies the electronic contribution is known as the Sommerfeld constant.
    The linear relation for the electronic heat capacity implies 𝐶el → 0 for 𝑇 → 0 K,
    whereas the heat capacity of a classical ideal gas is constant. This discrepancy
    is closely related to a similar discrepancy observed for the phonon-related, lowtemperature heat capacity in insulators, as discussed previously. In addition,
    experimental values for the electronic heat capacity at room temperatures are
    about two magnitudes lower than the classical prediction.
    Describe the heat capacity plot of the following experimental data
  20. 20For the Heat Capacity of the Free Electron Fermi Gas, describe electrons with energies well below the Fermi energyelectrons with energies well below the Fermi energy (which is
    in the order of several eV) cannot be excited thermally (energy gains of the order
    of tens of meV) because the final states are still well below the Fermi energy and
    therefore fully occupied! A thermal excitation is possible only for electrons in
    states close to the Fermi energy, where the occupation probability, 𝑓 (𝜖, 𝜇, 𝑇 ) is less
    than 1. Only electrons within a few 𝑘𝐵𝑇 around the Fermi level can be thermally
    excited, and one needs to reduce the classical result accordingly. At room temperature, this reduces the heat capacity by a about two orders of magnitude
    (∼ 300 K/40 000 K for sodium) and produces the observed linear temperature
    dependence
  21. 21For the fermi electron gas describe determining the change Δ𝑈 in internal energywe determine the change Δ𝑈 in internal
    energy when the electron gas temperature is increased from 0 K to a finite temperature 𝑇 . The temperature dependence of the internal energy comes solely
    from the Fermi-Dirac distribution 𝑓 . The electronic heat capacity 𝐶el is now the
    temperature derivative of the either internal energy or the temperature dependent
    change in energy, Δ𝑈, because 𝑈el(0) is a constant.
  22. 22For the graph, plot df/dT vs e/kb for 300K and 2000K (double peak)we see that d𝑓 /d𝑇 is a sign-changing double peak centred at
    𝜖𝐹 and zero elsewhere if 𝑘𝐵𝑇 ≪ 𝜖𝐹 . Note that we are considering the derivative
    with respect to temperature, not energy (d𝑓 /d𝑇 rather than d𝑓 /d𝜖). Together
    with the equally sign-changing factor (𝜖 − 𝜖𝐹 ), the integrand is a positive double
    peak around 𝜖𝐹 . As the electronic density of states of a metal usually varies
    slowly around the Fermi energy, we can replace 𝐷(𝜖) with a constant 𝐷(𝜖𝐹 ) in
    the integral.
    For the graph, plot df/dT vs e/kb for 300K and 2000K (double peak)For the graph, plot df/dT vs e/kb for 300K and 2000K (double peak)
  23. 23State the assumptions to consider for the Heat Capacity of an Electron Fermi Gaselectronic heat capacity of a metal is expected (i) to vary linearly
    with temperature and (ii) to scale with the electronic density of states at the Fermi
    energy.
    State the assumptions to consider for the Heat Capacity of an Electron Fermi Gas
  24. 24Discuss the difference between the experimental Sommerfeld parameter and the electrongas prediction values being out by a factor of 2The main reason is that the electronic density of states in some metals
    is not well described with free-electron gas model with 𝐷(𝜖) ∝ √
    𝜖. Systems with
    𝑑 or 𝑓 conduction electrons tend to have an enhanced density of states at the
    Fermi level, 𝐷(𝜖𝐹 ), which gives directly an equally enhanced heat capacity.
    The Fermi temperature 𝑇𝐹 depends on the electron density (𝑁/𝑉 ) and electron
    mass 𝑚. The experimental heat capacity data suggest that electrons in a metal can
    behave as if they have a larger mass than a single electron in free space. We will see
    in later lectures that this is indeed the case and that this effect originates from the
    interaction of the electrons with the periodic Coulomb potential of the ion cores
    in the crystal. In addition, the motion of electrons is hindered by scattering with phonons or other electrons, which also contributes to a heavier apparent mass.
  25. 25Define the thermal effective massThe thermal effective mass is a measure of how much the apparent electron mass
    needs to be increased so that the measured electronic heat capacity of a metal
    matches that of the free-electron Fermi gas with with the same electron density
  26. 26Describe this graph of the Electronic Resistivity of metalsAt temperatures around room temperature and above, most metals exhibit an
    electrical resistivity that increases linearly with increasing temperature. We will
    see later that the resistance levels off at very low temperature (not shown here).
    Note how the transition from solid to liquid aluminium is associated with a jump
    in resistivity
    Describe this graph of the Electronic Resistivity of metals
  27. 27Use the Drude model to model the electrical conductivity to find out details of the electrons acceleration in the presence of collisionsIn order to model the electrical conductivity, we start from the electron Fermi
    gas and take into account the scattering of electrons. That collisions with defects,
    phonons and other electrons are important follows from the following argument.
    Suppose there was no electron scattering: The electrons would then accelerate
    continuously in an applied electric field, and their wave vectors/velocities would
    increase linearly with time. All electrons are accelerated by the same amount,
    ℏ𝛿𝒌(𝑡)/𝑚, which corresponds to a shift of the Fermi sphere and all the states
    it contains by 𝛿𝒌(𝑡). While this is reasonable at short time scales, it does not
    yield a steady-state solution, and the ever increasing velocity is neither physically
    plausible nor consistent with the experiment (Ohm’s law). We need to include the
    collisions in order to reach a steady state. However, the result that, on average,
    all electrons are accelerated by the same amount, still holds in the presence of
    scattering.
  28. 28In regards to the Drude models electrical conductivity, describe; the time between collisions acceleration of electrons and electric field States in Fermi Sphere Drift VelocityWith an average time 𝜏 between collisions, the electrons are on average accelerated
    during this time 𝜏. There is a balance between the acceleration of the electrons
    by the electric field and the collisions redistributing the electrons back towards
    their initial distribution in reciprocal (wave vector) space. As a result, all occupied
    states in the Fermi sphere are shifted by a constant 𝛿𝒌(𝜏), which is given by the
    strength of the applied field and the scattering time. This yields an average drift
    velocity 𝑣𝑑 , which we can relate to the applied field and thereby obtain Ohm’s law
    and an expression for the electrical conductivity. The conductivity is given solely
    by properties of the electron gas: the electron density 𝑛, time 𝜏 between collisions
    and electron mass 𝑚. Bearing in mind the heat capacity results suggested that
    electrons in a metal can behave as if they have a mass that is different from the
    free-electron mass, we will keep the electron mass as a material parameter
  29. 29Where does the temperature dependence of the electrical resistivity come fromThe temperature dependence of the electrical resistivity originates from the
    temperature dependence of the scattering rate 1/𝜏. In the high-temperature
    limit, electrons are predominantly scattered by phonons, which will become clear
    once we have considered the low-temperature behaviour. At temperatures well
    above the Debye temperature, the number of phonons ⟨𝑛ph⟩ increases linearly
    with temperature 𝑇 (from the Planck distribution in the classical limit). It is
    reasonable to assume that the scattering rate is proportional to the number of
    scattering phonons, which yields the observed linear temperature dependence of
    the resistivity
    Where does the temperature dependence of the electrical resistivity come from

Informationen

Beschreibung
Year 4 ICMP Karteikarten am Week 6 LNs, erstellt von R S C am 07/07/2024.
Erstellt von
R S C
Aufrufe
1
Bewertung

Feedback

Worum geht es?

Wie gefällt dir das neue Design?

Dein Eindruck