




Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Questions from the part ib & ii tripos examination for the natural sciences, focusing on mathematics and physics. The questions cover topics such as curvilinear coordinates, temperature distribution, fourier transforms, matrix algebra, and differential equations. Students are required to find solutions, express functions in terms of derivatives, prove theorems, and verify solutions.
Typology: Exams
1 / 8
This page cannot be seen from the preview
Don't miss anything!





Monday 27 May 2002 1.30 to 4.
You may submit answers to no more than six questions. All questions carry the same number of marks.
The approximate number of marks allocated to a part of a question will be indicated in the right hand margin.
Write on one side of the paper only and begin each answer on a separate sheet.
Each question has a number and a letter (for example, 6C).
Answers must be tied up in separate bundles, marked A, B or C according to the letter affixed to each question.
Do not join the bundles together.
For each bundle, a blue cover sheet must be completed and attached to the bundle.
A separate yellow master cover sheet listing all the questions attempted must also be completed.
Every cover sheet must bear your examination number and desk number.
1A In a curvilinear system a point P has coordinates (u, v, w) such that
u^2 =
(s + x), v^2 =
(s − x) and w = z,
where (x, y, z) are the rectangular Cartesian coordinates of P and s =
x^2 + y^2 is its distance from the z-axis. Describe the surfaces u = const, v = const and w = const and sketch the loci of intersections of u^2 = 0, 1, 2 and v^2 = 0, 1, 2 with the x, y plane. (^) [5]
Express x, y and z explicitly in terms of u, v and w in such a way that, when u is defined so that u ≥ 0, y and v have the same sign and the point P is uniquely determined by u, v and w. (^) [4]
Show that u, v and w are orthogonal curvilinear coordinates and find the coeffi- cients hu, hv and hw such that dl, the distance between points (u, v, w) and (u + du, v + dv, w + dw), is given by dl^2 = h^2 udu^2 + h^2 v dv^2 + h^2 wdw^2
in the limit dl → 0. [7]
If φ = φ(u, v) only express ∇^2 φ in terms of derivatives with respect to u and v. (^) [4]
[You may use the following formulae
∇φ =
hu
∂φ ∂u
eu +
hv
∂φ ∂v
ev +
hw
∂φ ∂w
ew
and
∇.A =
huhv hw
∂u
(hv hwAu) +
∂v
(hwhuAv ) +
∂w
(huhv Aw)
Paper 1
4B Suppose A is an n × n matrix:
Prove that An^ = 0 if and only if all eigenvalues of A vanish.
[You may use the fact that a matrix satisfies its own characteristic equation] (^) [5]
By considering the quadratic form Q = xT^ Ax, or otherwise, prove that if the matrix A is symmetric then An^ = 0 implies A = 0. (^) [5]
For the following matrix
1 1 α 1 1 β
find the values of α and β for which all the eigenvalues of T vanish and verify that in that case T^3 = 0. [5]
Consider the transformation x′^ = Tx for three dimensional vectors x and x′. Show that all x′^ are confined to a single plane for any given x.
What is the effect of T^2 and T^3 operating on x? (^) [5]
5B (a) Let A and B be n × n Hermitian matrices (A = A†, B = B†^ ) with distinct eigenvalues. Show that:
(i) H = i (AB − BA) is Hermitian, (^) [4]
(ii) the eigenvectors of A and B are identical if and only if AB = BA, (^) [6]
(iii) the matrix N = A + iB can be diagonalised if and only if NN†^ = N†N. (^) [5]
(b) If C is a unitary matrix and A is Hermitian show that
)n has real eigenvalues if n is a positive integer. (^) [5]
[You may quote the properties of the eigenvalue and eigenvectors of Hermitian matrices without proof.]
Paper 1
6C The differential equation
(1 − x^2 )y′′^ − xy′^ + m^2 y = 0 , (∗)
has two linearly independent solutions about the origin of the form y = xσ^
n=0 anx
n
(with a 0 6 = 0).
(a) Is the origin x = 0 an ordinary or singular point of this differential equation? Determine the two appropriate values of σ and find recurrence relations between the an’s for the two cases. [8]
(b) Show that if m is an integer then there always exists a polynomial solution, denoted by Tm(x). How many non-zero terms do these polynomials contain? Find the first four polynomials Tm(x) (i.e. m = 0, 1 , 2 , 3) given the normalization Tm(1) = 1. (^) [7]
(c) Use the ratio test to discuss the convergence of the non-polynomial solutions on the interval − 1 ≤ x ≤ 1. Comment on the relationship of the radius of convergence R to the location of the singular points of the differential equation (∗). (^) [5]
7C (a) Consider the general eigenvalue equation
y′′^ − b(x)y′^ + c(x)y = −λd(x)y ,
subject to homogeneous boundary conditions y(0) = y(1) = 0. Find suitable functions p(x), q(x) and w(x) which enable this equation to be re-expressed in Sturm-Liouville form
− (p(x)y′)′^ + q(x)y = λw(x)y. (^) [4]
(b) Find eigenfunctions and eigenvalues for the equation
y′′^ + λy = 0 , (†)
subject to the boundary conditions y(0) = 0 and y′(π/2) = 0. Determine an appropriate normalization for these eigenfunctions. (^) [5]
Use these to obtain an eigenfunction expansion as a solution of the inhomogeneous equation y′′^ + κy = x ,
subject to the same boundary conditions as in (†) and where κ is a constant (not an eigenvalue). (^) [7]
Hence, by choosing appropriate values for κ and x (or otherwise), show that
π^4 96
Paper 1 [TURN OVER
9C Consider a set of eigenfunctions yn(x), (with n = 0, 1 , 2.. .) and corresponding eigenvalues λn which satisfy the Sturm-Liouville equation
Ly ≡ − [p(x) y′]′^ + q(x)y = λw(x)y ,
with boundary conditions y(0) = y(1) = 0. Assume that the eigenfunctions are unit
normalized, that is,
0 y
(^2) nwdx = 1.
(a) Given an arbitrary function ˜y satisfying the same boundary conditions, we can define
Fn(x) ≡ Ly˜ − λnw(x)˜y.
Show that for every n we must have
∫ (^1)
0
yn(x)Fn(x)dx = 0. [6]
(b) Consider an approximate trial function ˜y (with ˜y(0) = ˜y(1) = 0) which is close to the lowest eigenfunction y 0 so that we can expand the difference in terms of the remaining eigenfunctions as
y˜ = y 0 +
n=
anyn.
where the an’s can be assumed small. Use the Rayleigh-Ritz method with the trial function y˜ to show that the lowest eigenvalue λ 0 can be approximated as
λ˜ 0 = λ^0 +^
n=1 a
2 nλn 1 +
n=1 a^2 n
≈ λ 0 +
n=
a^2 n(λn − λ 0 ) ,
where terms of order O(a^4 n) have been neglected. (^) [10]
Discuss the relative error in the approximation ˜y to the lowest eigenfunction y 0 compared with the error in the approximation λ˜ 0 to the lowest eigenvalue λ 0. Is λ˜ 0 lower or higher than λ 0? (^) [4]
Paper 1 [TURN OVER
10C (a) Derive the Euler-Lagrange equation satisfied by the function y(x) which makes the integral
I =
∫ (^) b
a
F [x, y(x), y′(x)]dx
stationary subject to the boundary conditions y(a) = y 1 and y(b) = y 2. (^) [7]
(b) Efficient international airline routes must minimize the distance between two locations on the globe.
Show that the path length between two points A(φ 1 , θ 1 ) and B(φ 2 , θ 2 ) on a unit sphere can be expressed in polar coordinates (φ, θ) as
∫ (^) θ 2
θ 1
1 + sin^2 θ φ′^2 dθ.
where φ′^ = dφdθ. (^) [4]
Hence, use the Euler-Lagrange equation to show that a stationary path satisfies
φ(θ) = ±
∫ (^) θ 2
θ 1
dθ sin θ
k^2 sin^2 θ − 1
where k is a constant. [4]
Integrate this expression to find the extremal solution. Use this to specify the shortest route for a flight setting out from Rome (θ ≈ 45 ◦, φ ≈ 15 ◦) and heading for Singapore (θ ≈ 90 ◦, φ ≈ 105 ◦).
[Hint: In the integration consider the substitution t = cot θ.] (^) [5]
Paper 1