Karteikarten
Week 8 Lns
Karten in diesem Stapel (42)
- 1For Chapter 16 Can you recall the following
- 2Explain how one obtains the central equation for the Schrodingers Periodic PotentialWe start here with an ansatz for the wave function that is more general than
those following from the Bloch’s theorem: We are summing over many more Fourier components than in a regular Fourier series for a periodic function, where
𝐺 = 𝑠
2𝜋
𝑎 with arbitrary integers 𝑠; note the absent 𝑁 in the denominator. The
Fourier sum for 𝜓(𝑥) is essentially a Fourier transform because the allowed wave
vectors 𝑘 = 𝑠
2𝜋
𝐿
are dense in reciprocal space. The wave vectors outside the 1st
Brillouin zone are also included here. This more general ansatz simplifies the
discussion, and we will later recover Bloch’s theorem from this treatment.
As explained in detail in Appendix 15.A, by inserting the expansions of the
potential and the wave functions into the Schrödinger equation, one obtains a set
of homogeneous linear equations for the expansion coefficients 𝐶𝑘 of the wave
functions.
The result is the central equation - 3Explain what the central equation doesFor every wave vector 𝑘, this equation puts a constraint on the
relation between the expansion coefficients 𝐶𝑘 for the wave function and couples
the expansion coefficients for the wave function to the Fourier coefficients 𝑈𝐺,
which define the problem in the first place. For every wave vector 𝑘, one obtains
one equation. The set of all equations can be expressed as a matrix equation,
which has the structure of a eigenvalue problem.
In summary, we have transformed the familiar differential form of the Schrödinger
into a set of simultaneous linear equations for the wave function expansion
coefficients. There is one equation for every expansion coefficient 𝐶𝑘, which links
it to all other 𝐶𝑘−𝐺. The determination of the wave function expansion coefficients
𝐶𝑘 and thus the wave function has become a linear algebra or matrix inversion
problem - 4For the Schrodingers periodic potential, the determination of the wave function expansion coefficients 𝐶𝑘 and the wave function has become a linear algebra or matrix inversion problem, how is this usually solved?Solving this matrix inversion problem is at the core of every modern computer
program for the calculation of the electronic band structure of a crystalline solid.
As stated above, the sum over 𝐺 runs from −∞ to +∞ in one dimension, and in
three dimensions one would used an equivalent triple sum over all reciprocal
lattice points 𝑮. In practice, the Fourier sums can be truncated to finite sums,
where one needs to find an acceptable balance between accuracy and computing
time required. Similarly, only a finite set of electron wave vectors 𝑘 needs to be
considered. - 5In the calculated band structure for sodium metal, a series of wave vectors 𝑘 along a certain line in the Brillouin zone is plotted along the horizontal axis. For each of these wave vectors, there is a set of energy levels. If one follows a particular energy level as a function of wave vector what is obtainedone obtains
one of the electron energy bands - 6What is the challenge in practical electronic structure calculationsThe largest challenge is to go
beyond the independent electron approximation. Direct interactions between
electrons can to some extent be included in the crystal potential 𝑈(𝒓), so that it
describes (i) the Coulomb interaction between an electron and all nuclei (or ion
cores), (ii) the Coulomb interactions between a single electron and the background
charge from all other electrons and (iii) the quantum-mechanical exchange and
correlation energy. For many types of crystal, this can be achieved to a very good
approximation in the framework of Density Functional Theory - 7Describe the central equations wave function expansion coefficients How are the solutions to the central equation therefore foundThe central equation couples all wave function
expansion coefficients 𝐶𝑘, 𝐶𝑘−𝐺 with wave vectors that are related by reciprocal
lattice vectors 𝐺, and only these. The solutions to the central equation can therefore
be expressed using only such a subset of wave vectors. This reduces the sum over
all allowed wave vectors 𝑘 to one over only the reciprocal lattice vectors/points 𝐺. - 8What is the final step of transforming the central equation to Bloch theoremRearranging the expression for the constrained 𝜓𝑘(𝑥) brings it into the form of a
Fourier series (now summed only over all 𝐺) multiplied by a plane-wave part, 𝑒^i𝑘𝑥.
The Fourier series yields a function 𝑢𝑘(𝑥) that has the periodicity of the crystal
lattice, and the wave function 𝜓𝑘 thus takes the form stated in Bloch’s theorem - 9Define and describe the Fermi surfaces in terms of the density of statesThe Fermi surface is a surface of constant energy in reciprocal space. All states on
the Fermi surface, identified by wave vectors 𝒌, have the energy 𝜖(𝒌) = 𝜖𝐹 , where
𝜖𝐹 is the Fermi energy. We have seen that only electrons in states near the Fermi
surface contribute to the electronic heat capacity of a metal, which we expressed
in terms of the density of states at the Fermi energy, 𝐷(𝜖𝐹 ). The density of states
can be expressed as an integral of the Fermi surface, and the same is true for other
physical properties, e.g. the electrical conductivity. - 10Give examples of nearly spherical fermi surfaces materialsNearly spherical Fermi surfaces are observed only for monovalent metals such as
sodium and the other alkali metals as well as copper, silver and gold. Polyvalent
metals have much more complicated Fermi surfaces - 11Describe Fermi surfaces; 1D, 2DIn the simplest case, the Fermi surface separates occupied from unoccupied states
in the Brillouin zone. In one dimension, one fills the available states from low to
high energies until all electrons in the system are distributed. This determines the
Fermy energy 𝜖𝐹 and Fermi wave vector 𝑘𝐹 . The one-dimensional “Fermi surface”
comprises just the two points at either end of the range of wave vectors with
occupied states (blue line in the slide above).
In two dimensions, we have an energy dispersion relation in the form of a paraboloid. Its intersection with the plane of constant energy, 𝜖(𝒌) = 𝜖𝐹 , yields the
Fermi circle. The size (diameter) the circle depends on the number of electrons
per primitive unit cell - 12How are Fermi surfaces observed in experimentsthe parts of the
Fermi surface outside the Brillouin zone can be projected inside so as to obtain
the Fermi surface that would be observed in an experiment, this from the fact that every Bloch state with a wave vector 𝒌 outside the
Brillouin zone corresponds to an equivalent state inside, where 𝒌 → 𝒌 + 𝑮 with a
suitably chosen reciprocal lattice vector G - 13Describe mapping the dispersion from the extended zone scheme into the central Brillouin zoneThis is done by shifting the dispersion segments by suitably chosen
multiples of 2𝜋/𝑎, left or right along the wave vector axis - 14Briefly how are higher order BZ made in 2 and 3DIn two and three dimensions, we can construct higher-order Brillouin zones by
repeating the Wigner-Seitz construction with the reciprocal lattice points that
are further away from Γ than the ones used for the first Brillouin zone, which we
have so far simply called the Brillouin zone. Every higher Brillouin zone covers the same wave
vectors as the first Brillouin zone, but its segments are shifted by various different
reciprocal lattice vectors. - 15Describe Harrisons method for Fermi SurfacesWe can reduce the Fermi surface (the blue circle in the slide above) that is cutting
through higher Brillouin zones to the first Brillouin zone using Harrison’s method:
In the 2D case, the Fermi circle is drawn around every reciprocal lattice point
(black dots on the right). This produces the full Fermi surface in the first zone,
and we can now decompose it into the parts (often called sheets) that originate
from the higher Brillouin zones 1, 2 and 3 as shown on the left. - 16Describe what rounds the edges of the free electron fermi surface for tetravalent metals with fcc What causes the real fermi surface to cross the bragg planes at right anglesthe
modifications of the electron bands due to the crystal potential round them.
The opening of band gaps at the Brillouin zone boundaries (at Bragg planes 1/2𝑮,
where 𝑮 is a reciprocal lattice vector) distorts the free-electron Fermi sphere. - 17In context of rounding of the free electron fermi surface, describe its method in terms of the fermi surface cuspA physical argument why the Fermi surface needs to be rounded in this way is
the following. The group velocity of an electron of wave vector 𝒌 and energy 𝜖(𝒌)
is given by the gradient of the dispersion, 𝑣𝑔 =
1
ℏ
∇𝒌𝜖. We also know that the Fermi
surface is a surface of constant energy, 𝜖(𝒌) = 𝜖𝐹 . The gradient ∇𝒌𝜖 is therefore
perpendicular to the Fermi surface. If a Fermi surface has a cusp as in the pure
free-electron picture, then the gradient at that cusp is undefined. Consequently,
the group velocity of the electron(s) in the state at the cusp is undefined, which
cannot be correct. The cusps disappear if a Fermi surface always cuts through a
Brillouin zone face at right angle - 18what results in one to conclude that the Fermi surface of a divalent metal becomes a combination of hole and electron pocketsThis can be seen if we we combine Harrison’s construction of the Fermi surface with the rounding of
the edges from the electron-ion interaction - 19Briefly describe the importance of Bloch theoremThe Bloch theorem introduced in the previous lecture is one of the most important
factors for understanding the electronic structure of crystals. Firstly, it constrains
the mathematical form of the wave function solutions of Schrödinger’s equation
for any quantum mechanical system with a periodic potential. Secondly, it follows
from the Bloch theorem that Bloch states with wave vectors 𝑘 outside the 1st
Brillouin zone are equivalent to states inside the 1st Brillouin zone. This allows
us, for example, to plot the electronic band structure (i.e., the energy dispersion
relations for the electrons) only along lines in the first Brillouin zone. - 20Describe populating the lowest band of a monovalent metal and electrons near the fermi energy What characterises a metal. In a monovalent metal (e.g. an
alkali metal) with one atom and therefore one conduction electron per primitive
unit cell, one half of the 2𝑁 states of the lowest band will be populated with
electrons. The electrons near the Fermi energy can be accelerated by arbitrarily
small amounts in an electric field, because states with matching energy and wave
vector are available in the same band. A metal is characterised by a partially filled
electronic band. - 21Describe the bands of an insulatorIn an insulator, all states in one band are filled with electrons, and all states of the
next band at higher energy are unoccupied. There are no free states in the lower
band so that an electric field cannot accelerate any of the electrons, and there will
be no electric current. This cannot be understood in the picture of an electron
gas. It is necessary to take into account the band gap, which originates from the
periodic crystal potential. - 22How can electrons be excited across the gapIn insulators at sufficiently high temperatures (relative to the size of the band gap) electrons
can be excited thermally across the band gap, which results in some electrical conductivity. This thermally activated conductivity is usually very low in comparison
to that of a metal - 23Why is there no fundamental difference between an insulator and a semiconductorThe band gap of an insulator is so large (≳ 5 eV) that the electrical conductivity
at around room temperature is vanishingly small. Semiconductors have smaller
band gaps so that they exhibit a small but finite electrical conductivity at around
room temperature. As the temperature is reduced to zero, the conductivity also
decreases to zero because it is no longer possible to excite electrons thermally
across the band gap. - 24Explain why divalent alkaline earth elements are not semiconductorsTheir
experimentally determined electrical conductivities are characteristic of a metal.
The reason is that the simple band-filling argument works in one dimension, but
not necessarily in two and three dimensions (cf. workshop problem 9.3). The
alkaline earth metals have somewhat more complicated band structures, where
the maximum of the first (lower-energy) band is above the minimum of the second
(higher-energy band). The result is that the highest-energy electrons occupy states
in the upper band so that there are two partially filled bands rather than one
fully filled band. A small band overlap along with a low electrical conductivity is
characteristic of a semimetal such as bismuth. The electrical conductivity of the
alkaline earth metals is significantly larger than that of typical semimetals, and
they are not generally regarded as such - 25For Semi C; Describe the important bands for electrical conductivity Define Valence and Conduction bandsFor the electrical conductivity, the bands near the Fermi energy are the most
important. One distinguishes between the valence band and the conduction band.
The valence band is the highest-energy band that is completely filled at zero kelvin.
The conduction band is the next higher-energy band. In a semiconductor/insulator,
the conduction band is completely empty at 0 K, whereas it is partially filled in metals and semimetals. - 26Describe electrical conductivity of a perfect semiconductorIn a perfect semiconductor, the electrical conductivity at 0 K is zero because
there are no free charge carriers. The conductivity then increases with increasing
temperature as the number of electrons in conduction band states increases. In a
metal, the electrical conductivity decreases with increasing temperature because
the electron–phonon and electron–electron scattering rates increase, - 27Describe the difference between the direct and indirect band gapsD: extrema of valence and conduction band are at same wave vector
ID; extrema of valence and CB are at different wave vectors - 28Define absorption and abdosrbanceAbsorption is the process of annihilating a photon by exciting an electron to a
higher-energy state, whereas absorbance is a measure of how strong the absorption
is - 29Describe the terms found in the absorbance equation𝐼in and 𝐼out are the intensities of the light when it enters and leaves the
sample, respectively. Note that the absorbance depends on the sample thickness. - 30Describe the wave vector of optical phonons What does this mean for the wave vector of an electron in the photon absorption processoptical photons have a wave vector that is negligible compared
to a wave vector at the Brillouin zone boundary. Therefore, the wave vector of an
electron does not change noticeably in a simple photon absorption process. The
electron must be excited to a state with the same wave vector, i.e. straight up in
the band scheme - 31In an indirect semiconductor, describe the lowest energy onsent of optical absorption causedue
to an absorption process where an electron is excited from the valence to the
conduction band with the simultaneous annihilation of a phonon. The phonon
“donates” its wave vector and energy ℏΩ to the electron, in addition to the energy
ℏ𝜔 that comes from the photon. This three-particle indirect absorption process
occurs less likely than the two-particle (electron & photon) direct absorption
process. - 32Why can we identify electron wave vector with linear momentum for electrons in a periodic crystal potentialstart from the electron energy dispersion 𝜖(𝒌). The effect of the crystal
potential is contained in the 𝜖(𝒌) dispersion relation, which will differ from the
free-electron case. From there, we obtain the group velocity 𝑣𝑔 for an electron
(wave packet) with wave vectors ≈ 𝒌. From the energy gain of an electron in the
electric field, we obtain an equation of motion in terms of the wave vector.
Equation F shows that ℏ𝒌 plays the role of the linear momentum 𝒑 in Newton’s second
law, as for free electrons - 33State the difference between group and phase velocityWaves can be in a group and such groups are called wave packets, so the velocity with which a wave packet travels is called group velocity. The velocity with which the phase of a wave travels is called phase velocity.
- 34Define holes and how theyre createdA hole is a vacant orbital/state in an otherwise filled band. We can create such
a vacant state in the valence band of a semiconductor by exciting an electron
optically or thermally to a conduction band state. One has now the choice between
describing the valence band properties in terms of the ∼1020−1 populated valence
band (electron) states, or in terms of the one vacant state. - 35Explain the origin of kh = -keIf an electron with wave vector 𝒌𝑒
is removed from a band, the total wave vector of
all remaining electrons is −𝒌𝑒
. Therefore, we can describe all remaining electrons
with a hole quasiparticle that has a wave vector 𝒌ℎ = −𝒌e - 36Describe the energy of the holeThe energy of the hole is the missing energy of the removed electron: 𝜖ℎ(𝒌ℎ) =
−𝜖𝑒(𝒌𝑒). At 0 K, the Fermi energy is at the top of the valence band. Removing an
electron (= creating a hole) requires the more energy the further below Fermi
energy the state is. This relation between wave vectors and energies of the initial
electron state and created hole can be described with a “flipped-up valence band”. - 37explain e(k) = e(-k) What is the result (as well as with k)e electron bands in non-magnetic materials are symmetric under
inversion
The sign change in both the energy and the wave vector gives the result that the
initial electron state and the created hole have the same group velocity, both in
magnitude and direction - 38Explain where the wave vector is negative with positive group velocity for the valence band electron state in 1D can be observedhappens whenever the curvature of the
energy dispersion is negative, i.e. in the vicinity of band maxima - 39Describe the mass of the holeThe mass of a hole is the missing mass of the removed electron. By “mass” the band effective mass 𝑚∗
is meant. the effective mass
of a hole is the missing effective mass of the removed electron - 40Describe the charges of the hole and electronThe positive charge of a hole is the missing negative charge of the removed
electron. This can be confirmed formally by comparing the hole and electron
equations of motion with the velocity and wave vector relations - 41Describe what results in the hole being acceleratedAt time 𝑡 = 0, an electric field 𝐸𝑥 is applied in the positive 𝑥 (≡ 𝑘𝑥 ) direction, to the right. As long as no scattering occurs, the negative electrons are all accelerated to
the left: The electron in state G transfers to the vacant state F, leaving G vacant.
The electron in state H transfers to the now vacant state G, and so on. The same
happens on the left half of the diagram. What is not shown is that the electron in
the state at the 𝑘 = −𝜋/𝑎 zone edge transfers to the equivalent state 𝑘
′ = 𝑘+2𝜋/𝑎
at the right Brillouin zone edge. Remember that these two states are in fact the
same state, coupled by electrons being “diffracted” forth and back. In the end, all
electron wave vectors have changed by the same amount 𝛿𝑘. Consequently, the
wave vector of the vacant state also changes by the same amount — the hole is
accelerated. - 42For electrons and holes define the total currentThe total current is given by the current arising from the single unpaired electron
in state 𝐺 in panel (b). For all electron pairs, the two electrons move in opposite
directions and the net current is therefore zero. The electric current 𝒋
‡ from
the single unpaired negative electron 𝐺 is the same as that of a single hole with
opposite charge and velocity. Here, 𝑣{𝑋} denotes the velocity of an electron in
state 𝑋, and 𝒗{𝐺} = −𝒗{𝐸} because the derivatives 𝑣𝑔 =
1
ℏ
d𝜖(𝑘)/d𝑘 have opposite
signs for states 𝐺 and 𝐸. Note that when we concluded earlier that 𝑣ℎ = 𝑣𝑒
, we
had considered an electron that was removed and a hole that was left behind in
its place. This is different from the case here, where we considered an electron in
state 𝐺 and a hole created from state 𝐸.
The main point is that we can describe the current from the nearly-filled valence
band in terms of either electrons or holes, and the result is the same. Switching
between electrons and holes changes the sign of the velocity as well as that of
the charge, so that the current remains unchanged.
Informationen
- Beschreibung
- Year 4 ICMP Karteikarten am Week 8 Lns, erstellt von R S C am 15/07/2024.
- Erstellt von
- R S C
- Aufrufe
- 0
- Bewertung