Kittel4 4
Kittel4 4
Kittel4 4
Solution The force constant between the p and p + s planes of atoms is given by Cp = A sin(pk0 a) pa ,
and so we generalise the dispersion equation in Kittel Equation (4.16a) viz. 2 to give 2 = 2 M A
p>0
2 M
Cp (1 cos(pka)) ,
p>0
(1)
sin(pk0 a) pa
(1 cos(pka)) .
(2)
The derviative is given by dierentiating through the sum d 2 dk When k = k0 then the sum becomes d 2 dk = 2A sin(kpa)2 , M p>0 (4) = 2A [sin(pak) sin(pak0 )] . M p>0 (3)
which for ka = m diverges. For ka = m, the sin terms vanish and the sum gives exactly zero. 2. Kittel Problem 4.5
Solution This problem is calculated largely analogously to the two atom lattice example in Kittel, Chapter 4. The equations of motion are given by M d2 us dt2 d vs dt2
2
We look for plane wave solutions us vs The equations of motion (5) and (6) become, M 2 u = C 10v0 + v0 eiKa 11u0 M v
2
= u0 ei(tsKa) , = v0 ei(tsKa) .
(8)
(9) (10)
= C 10u0 + u0 e
iKa
11v0 .
We can rewrite this as a matrix equation, M 2 + 11C 10C CeiKa 10C CeiKa M 2 + 11C = 0 0 (11)
This is a simple eigenvalue equation which we solve in the usual way. That is, we rst require that the determinant vanish which will give us the dispersion equation. This gives us M 2 4 22M 2 C + 20C 2 (1 cos Ka) = 0. The dispersion equation is then given by 11C = M
2
(12)
(13)
The eigenvectors can be extracted in the usual manner, though this is rather tedious to do and for our purposes not needed. However, we can see by inspection that the two modes given by the above correspond to the acoustic and optical modes respectively. This can be seen by plotting the dispersion equation below. At K = 0, the cuto frequencies are = 0, and = At K = /a the modes reduce to = 22C/M for the acoustic and optical modes respectively. 2C/m. 20C/M , and =
The dispersion equation is shown below in Figure 1, which illustrate the acoustic and optical branches.
Figure 1: Dispersion equation for Problem 4.5 3. Kittel Problem 4.6 (a) This is a problem in electrostatics, but is good to get an idea of lattice vibrations in a rather crude way. We imagine at each lattice point there is an ion and there is a cloud of negative charge surrounding each ion. If the ion moves a distance r from the lattice point, then the force on the ion is due to the electric charge enclosed in a sphere of radius r centred at the lattice point. Hence, if is the density of electric charge then the electric eld a distance r from the lattice point is the same as that in a uniformly charged sphere of radius R. A simple calculation using Gausss law gives the electric eld in a uniformly charged sphere is (in SI units): er . r E= 40 R3 The force on the ion is then F= e2 r . 40 R3
This is like the force due to that on a spring, with spring constant k = e2 /(r0 R3 ) with natural frequency given by = e2 . 40 mR3
(b) For sodium the mass of a single atom is m 22.9/6.02 1023 3.8 1023 g. The lattice radius of a sodium atom R 3.7 1010 (Ashcroft and Mermin), and this gives a frequency of 1.1 1013 Hz. (14)
(c) A variety of estimates can be used here, so in this context most are equally valid. An example is to take the wavelength of the sound wave to be of the order the length of the ion. Then, K = /R and the velocity of sound becomes vs = 103 m/s. (15) K 4. Kittel Problem 5.1 (a) The dispersion equation from Chapter 4 of Kittel is given by = m | sin Ka/2|, with m = (4C/m)1/2 . The group velocity is given by d m a cos(Ka/2) = , dK 2 and so the density of states D() is given by D() = = = (b) The dispersion equation L 1 , 2 |d/dK| 2L 1 , 2 m a 1 2 /m N
2 m
(16)
(17)
1 . 2
(18)
= 0 AK 2 ,
implies that for all K, we have 0 and hence there are no states for > 0 . Therefore, D() = 0 for > 0 . For 0 we have grad = 2A(kx , ky , kz ) and hence the density of states becomes D() = = V K2 1 , 2 2 |grad | V 0 . 2 A3/2 4
(19)