Non-Abelian Statistics for six quasihole

states in Pfaffian Quantum Hall State

Project report submitted by

Adarsh Anand
13PH20004

Supervisor: Prof. Sudhansu Sekhar Mandal

Department of Physics

Indian Institute of Technology Kharagpur 721 302, India

August, 2017
Contents
.

2
1 Introduction

Figure 1: Fractional quantum Hall Effect

In the fractional quantum hall effect, since interaction between electrons

is not negligible, it breaks the degeneracy of the ground state. Since, the
degeneracy is extraordinarily large we can not use perturbation theory to
solve the problem. Hence, we study the problem of fractional quantum hall
effect by many different approaches by guessing some approximate solutions
to gain more insight into the problem.
The first approach to the fractional quantum Hall effect was due to Laughlin,
who described the physics at flling fractions = m1 , with m an odd integer.
The vast majority of the observed quantum Hall plateaux sit at fractions
with odd denominator. But there are some exceptions. Most prominent
among them is the very clear quantum Hall state observed at = 52 as
shown in figure ??. The = 52 state is thought to consist of fully filled

3
lowest Landau levels for both spin up and spin down electrons, followed by a
spin-polarised Landau level at half filling. The best candidate for this state,
the Pfaffian state, turns out to have a number of extraordinary properties
that opens up a whole new world of interesting physics involving non-Abelian
anyons.
The quasihole excitations of this state can show non-Abelian statistics
as they have a degenerate set of states of quasiholes at fixed positions and
braiding operations cause these states to rotate into each other. Here, we
study the number of degenerate states for six quasihole states in Pfaffian
state.

4
2 The Pfaffian State
The Pfaffian state or Moore-Read state describes an even number of particles,
?
N, with filling fraction = 1/m. It is given by
Y Y |zj |2 1
(z) = (zj zk )m exp ( )P f ( ) (1)
j<k j
4 zj zk

In the above equation the last factor is the Pfaffian described below.

Pfaffian
Consider a n n matrix M. The Pfaffian simply is a polynomial of degree
n/2 in the elements of the matrix. The simplest way to find the Pfaffian is
to partition n into n/2 pairs of numbers. For example, (1,2),(3,4),...,(n-1,n)
is one such partition.
The Pfaffian then takes the form,

P f (M ) = A[M12 M34 ...Mn1,n ] (2)

Here A means anti-symmetrize over the indices i.e. sum over all partitions
with signs to form a totally anti-symmetric function.
For example, for four electrons the Pfaffian takes the form,
1 1 1 1 1 1 1
Pf( )= + +
zj zk z1 z2 z3 z4 z1 z3 z4 z2 z1 z4 z4 z3

In particular, the case of m=2 describes a half-filled Landau level of

fermions. There is some numerical evidence that this particular state is
a good description of the ground state at = 52 . This wavefunction may be
considered as a variational ansatz for electrons at filling fraction = 12 in the
first excited Landau level (with both spins of the lowest Landau level filled)
interacting through Coulomb interactions.
These trial wavefunctions are known to be the exact zero-energy states of
the three-body toy Hamiltonian :
XX X 0 0
H= (zi zj ) (zj zk ) (3)
i j6=i k6=i,j

5
2.1 Quasi-Holes
The Moore-Read state also shows quasi-hole excitations. One can include 2n
quasiholes at points by modifying the Pfaffian in the manner,

1 (zj )(zj )...(zk )(zk )... + (j k)

Pf( ) Pf( )
zj zk zj zk
(4)
In this expression, the 2n quasiholes have been divided into two groups of
n each (here , , ...and, , ...), such that the quasiholes within each group
always act on the same electron coordinates within an electron pair.
(2n)!
There are apparently 2(n!)(n!) ways of making such a division. But, not all
of these wavefunctions are linearly independent; The true dimension of this
space of wavefunctions is actually 2n1 .
In next sections we study this linearly dependent behaviour for specific cases
of four and six quasiholes.

6
3 Case of Four Quasiholes ?
It is convenient to introduce some notations as given below,
(a1 a2 ...an )(b1 b2 ...bn ) (zj a1 )(zj a2 )...(zj an )(zk b1 )(zk b2 )...(zk bn )+(zj zk )
(5)
ij i j (6)
Now, we consider the basic identity,
(z1 1 )(z1 2 )(z2 3 )(z2 4 ) (z1 1 )(z1 3 )(z2 2 )(z2 4 ) + (z1 z2 )
= (z1 z2 )2 (1 4 )(2 3 )
(7)
In notation introduced above, it becomes,
(12)(34) (13)(24) = (z1 z2 )2 14 23 (8)
Similarly,
(12)(34) (14)(23) = (z1 z2 )2 13 24 (9)

From eq. (??) and eq. (??) we get,

(12)(34) (13)(24) 14 23
= (10)
(12)(34) (14)(23) 13 24
(12)(34)(12 )(34 ) + (13)(24)(13 )(42 ) + (14)(23)(14 )(23 ) = 0 (11)

Hence, for two electrons and four quasiholes the three apparently different
ways of constructing quasihole states are reduced to two through eq. (11);So,
there are only two four quasihole states.
Now, we generalize it for any even number of electrons, Ne . To see this
we insert eq. (8) into Pfaffian of eq. (4) :
(13)(24) (13)(24) (13)(24)
P f(13)(24) = A( ...)
z1 z2 z3 z4 z5 z6
(12)(34) (z1 z2 )2 14 23 (12)(34) (z3 z4 )2 14 23
= A( ...)
z1 z2 z3 z4
(12)(34) (12)(34) (12)(34)
= A( ...) A((z1 z2 )14 23 ...)
z1 z2 z3 z4 z3 z4
(12)(34)
+ A((z1 z2 )14 23 (z3 z4 )14 23 ...) + ...
z5 z6
(12)

7
There will be terms on the right hand side of (12) with zero, one, two, ...,
Ne factors of (zi zj ). Upon antisymmetrization, a term with k factors of
(zi zj ) would have to antisymmetrize 2k variables with a polynomial that
is linear in each. Since this is impossible for k1, such terms vanish. Hence,

(12)(34) (12)(34) (12)(34)

P f(13)(24) = A( ...) A((z1 z2 )14 23 ...) (13)
z1 z2 z3 z4 z3 z4
Similarly,

(12)(34) (12)(34) (12)(34)

P f(14)(23) = A( ...) + A((z1 z2 )13 24 ...) (14)
z1 z2 z3 z4 z3 z4
From these we get,
14 23
P f(12)(34) P f(14)(23) = (P f(12)(34) P f(13)(24) ) (15)
13 24

Hence, we get a linear relation among the three pairing possibilities for two
quasiholes for any even number of electrons. Thus, we deduce that there are
two four-quasihole states.

8
4 Case of Six Quasiholes
For six quasiholes, there can be ten different permutations. We show that
only four of them are linearly independent.
We take four permutations, (235)(146), (135)(246), (145)(236), (245)(136),
and show that any other of the can be written as a linear combination of
these four. To show that we take the permutation, (123)(456), and show its
linear dependence with above four.
Now we have,

(123)(456) (235)(146) = (z1 z2 )2 (15 )[2 3 4 + 2 3 6 2 3 z1 2 3 z2

2 4 6 + 2 z1 z2 3 4 6 + 3 z1 z2
+ 4 6 z1 + 4 6 z2 4 z1 z2 6 z1 z2 ]
(16)

(123)(456) (135)(246) = (z1 z2 )2 (25 )[1 3 4 + 1 3 6 1 3 z1 1 3 z2

1 4 6 + 1 z1 z2 3 4 6 + 3 z1 z2
+ 4 6 z1 + 4 6 z2 4 z1 z2 6 z1 z2 ]
(17)

(123)(456) (145)(236) = (z1 z2 )2 (16 )[2 3 4 + 2 3 5 2 3 z1 2 3 z2

2 4 5 + 2 z1 z2 3 4 5 + 3 z1 z2
+ 4 5 z1 + 4 5 z2 4 z1 z2 5 z1 z2 ]
(18)

(123)(456) (245)(136) = (z1 z2 )2 (26 )[1 3 4 + 1 3 5 1 3 z1 1 3 z2

1 4 5 + 1 z1 z2 3 4 5 + 3 z1 z2
+ 4 5 z1 + 4 5 z2 4 z1 z2 5 z1 z2 ]
(19)
14 24
Now, 15
(eq.(16)) 25
(eq.(17)) gives us,
14 24
[(123)(456) (235)(146)] [(123)(456) (135)(246)]
15 25 (20)
= (z1 z2 )2 (21 )(36 )(4 z1 + 4 z2 z1 z2 42 )

9
 14 24
Similarly, 16
(eq.(18)) 26
(eq.(19)) gives us,
14 24
[(123)(456) (145)(236)] [(123)(456) (245)(136)]
16 26 (21)
= (z1 z2 )2 (21 )(35 )(4 z1 + 4 z2 z1 z2 42 )
35 36
Now, 21
(eq.(20)) 21
(eq.(21)) gives us,
35 14 35 24
[(123)(456) (235)(146)] [(123)(456) (135)(246)]
21 15 21 25
36 14 36 24
+ [(123)(456) (145)(236)] [(123)(456) (245)(136)] = 0
21 16 21 26
(22)

Hence,
35 14 35 24
[P f(123)(456) P f(235)(146) ] [P f(123)(456) P f(135)(246) ]
21 15 21 25
36 14 36 24
+ [P f(123)(456) P f(145)(236) ] [P f(123)(456) P f(245)(136) ] = 0
21 16 21 26
(23)

Similarly, for any five permutations, we can find an equation of the form
of eq.(23). Therefore, we conclude that only four of the ten states are linearly
independent. Thus, there are only four six-quasihole states.

10
References
[1] NonAbelions in the Fractional Quantum Hall Effect, Nucl.
Phys B360 362 (1991) by Greg Moore and Nick Read.

[2] Lecture notes on The Quantum Hall effect by David Tong.

[3] Lecture notes on Quantum Condensed Matter Physics by

Chetan Nayak.

11