QM1 Chapter 3
QM1 Chapter 3
Wenhui Li
Department of Physics
National University of Singapore
3 Description of Quantum Systems
To study a physical system of your interest:
4. Predict the state of the system in the future (or under certain action), verify it by
measurements.
For a classical system, measurements do not disturb the system (the system is unaf-
fected), the outcomes of measurements, or the measured values, can be used to describe the
state of the system,
|x(t = 0), v(t = 0)i |x(t), v(t)i . (3.0.1)
For a quantum system, measurements destroy the system (most of the time), measured values
can not directly define the system being measured. Then how do we find out the state and
the evolution of a system by doing measurements?
1. Which particular Hilbert space to be used usually depends on what is under study.
Hermitian operators:
The physical system is described by a state vector in such a space. Suppose an operator
A has eigenvectors (eigenstates) {|1 i , |2 i , ..., |n i} and all the eigenvectors form an
orthonormal basis of an n-dimensional Hilbert space H , where A |i i = ai |i i. So, in
this orthonormal basis, we have
a1 0 . . . 0
0 a2 .
0 . .
A= .
. (3.1.1)
. .
. . .
0 . . . an
2
For convenience, we can label an eigenstate (eigenvector) by its eigenvalue, i.e., |i i
|ai i. Recall identity operator I, in this n-dimensional orthonormal basis as
1 0 . . . 0
0 1 .
X 0 . .
I= | ai ihai |=
.
. (3.1.2)
. .
i
. . .
0 . . . 1
P
Hence we reach A = i ai | ai ihai | .
2. An arbitrary vector in this Hilbert space corresponds to a possible state of the physical
system.
(a) Generally, only normalized vectors are used to describe the quantum system, i.e.,
|V | = 1.
3
(c) Digression: global phase vs. relative phase
|V i = ei |V i (3.1.6)
i
|W i + |V i =
6 |W i + e |V i (3.1.7)
i
|W i + |V i = e (|W i + |V i) (3.1.8)
Borns rule is one of the key principles of quantum mechanics, and is particularly relev-
ant to physical interpretation of measurements in quantum mechanics.
1. Measurements of a system
Recall X
A= ai | ai ihai | . (3.1.10)
i
Suppose the state of the system before measurement is |i, the state after measurement
is NO LONGER |i, NOR
X
| 0 i = A |i = ai | ai ihai | i, (3.1.11)
i
but one of the eigenstates. Measurement is not a smooth evolution of the system, but
quantum jump!!!
4
Pn
i=1 ni = M , we find that
M n
1 X X ni
a|i = a(k) = ( )ai . (3.1.12)
M k=1 i=1
M
hAi = h| A |i (3.1.15)
a1 0 . . . 0 ha1 |i
0 a2 . ha2 |i
0 . . .
= (h|a1 i h|a2 i ... h|an i) . .
(3.1.16)
. . .
. . . .
0 . . . an han |i
It is very important not to confuse expectation values with eigenvalues. For example,
the expectation value of Sz in Stern-Gerlach experiment can be any real value between
~/2 and ~/2, but the eigenvalue of Sz can only have two values, ~/2 and ~/2.
5
done is shown in the following figure. From the measured values, one can easily figure out
the amplitude and the orientation of the spin, that is, the state of the spin.
No matter what the input state is, there are only two measured values, S+ and
If either |+Zi or |Zi state is input into SGZ device, the outcomes are:
Lets investigate for an arbitrary state |i through SGz, we have the following scenario,
|i = a |+Zi + b |Zi
a
= in {|+Zi , |Zi} basis. (3.2.2)
b
6
aa + bb = 1 a = cos ei 0
2
i
b = sin e 0 , 2
2
cos2 = | h+Z|i |2 (3.2.3)
2
sin2 = | hZ|i |2 (3.2.4)
2
can be measured by measuring probability after SGZ . However, it is not enough to
determine and , additional measurements along x and y directions are needed.
|i = cos ei |+Zi + sin ei |Zi
2 2
= ei (cos |+Zi + sin ei() |Zi) Let =
2 2
= e (cos |+Zi + sin ei |Zi)
i
2 2
cos 2
= in {|+Zi , |Zi} basis, (3.2.5)
sin 2 ei
Before we continue on about the measurements of state |i, we need to know a few
additional things. First of all, how does SX look like in {|+Zi , |Zi} basis?
h+Z| SX |+Zi h+Z| SX |Zi
SX = in {|+Zi , |Zi} basis. (3.2.7)
hZ| SX |+Zi hZ| SX |Zi
Need to know h+Z|+Xi, h+Z|Xi, hZ|+Xi and hZ|Xi. |+Xi and |Xi can be
7
written as superpositions of |+Zi and |Zi as in the general formula shown previously:
+ +
|+Xi = cos |+Zi + sin ei+ |Zi ; |Xi = cos |+Zi + sin ei |Zi .
2 2 2 2
(3.2.8)
1 1
|+Xi = (|+Zi + ei+ |Zi), |Xi = (|+Zi + ei |Zi). (3.2.9)
2 2
And moreover, since |+Xi and |Xi are two eigenstates of operator SX ,
1
hX|+Xi = 0 = (h+Z| + ei hZ|)(|+Zi + ei+ |Zi)
2
1
= (h+Z|+Zi + ei hZ|+Zi +ei+ h+Z|Zi +ei(+ ) hZ|Zi).
2 | {z } | {z }
0 0
(3.2.10)
We have ei(+ ) = 1 + = .
We also need to know about operator SY and its eigenstates. Similar to the case of
SX ,
where
1 1
|+Y i = (|+Zi + ei+ |Zi), |Y i = (|+Zi + ei |Zi). (3.2.13)
2 2
8
Here, + = . To further fix + , , + and . We need to know | h+Y |+Xi |2 .
1
| h+Y |+Xi |2 =
2
1
= (1 + ei(+ + ) )(1 + ei(+ + ) )
4
1
= (1 + cos(+ + )), (3.2.14)
2
noting the followings.
+ =
+ = (3.2.15)
+ + = .
2
By convention, we take + = 0 and + = 2
and we have
ei+ = 1 ei = 1 (3.2.16)
i+ i
e =i e = i (3.2.17)
To summarize, we have
1 1 1
|+Xi = (|+Zi + |Zi) = (3.2.18)
2 2 1
1 1 1
|Xi = (|+Zi |Zi) = (3.2.19)
2 2 1
1 1 1
|+Y i = (|+Zi + i |Zi) = (3.2.20)
2 2 i
1 1 1
|Y i = (|+Zi i |Zi) = . (3.2.21)
2 2 i
where the matrices are expressed in {|+Zi , |Zi} basis. Moreover, the operators SX
SY in {|+Zi , |Zi} basis are
0 1
SX = S , (3.2.22)
1 0
0 i
SY = S . (3.2.23)
i 0
Now back to our original task, how do we figure out an arbitrary state |i (Recall
from Eq(3.2.5)) by measurements? After performing a series measurements shown in
the following figure, we could get the following probabilities:
9
1
| h+Z|i |2 = cos2 ( ) = (1 + cos ); (3.2.24)
2 2
1
| hZ|i |2 = sin2 ( ) = (1 cos ); (3.2.25)
2 2
1
| h+X|i |2 = (1 + sin cos ); (3.2.26)
2
1
| hX|i |2 = (1 sin cos ); (3.2.27)
2
1
| h+Y |i |2 = (1 + sin sin ); (3.2.28)
2
1
| hY |i |2 = (1 sin sin ); (3.2.29)
2
From these equations, one can figure out the value of and , therefore state |i. Now
cos 2
let us look at a few special cases for |i = , = {0, } and = {0, 2}.
sin 2 ei
(Note: all the matrix representations given below are in {|+Zi , |Zi} basis.)
Note: the quantum states are in Hilbert space, which is different from real 3-
dimensional space where the spin directions are in. For example:
But,
h+Z|Zi = 0. (3.2.32)
10
state of the spin is |Xi, we have
1 1 1
|+Xi = (|+Zi + |Zi) = (3.2.33)
2 2 1
1 1 1
|Xi = (|+Zi |Zi) = (3.2.34)
2 2 1
(3.2.35)
Note: Again, the quantum states are in Hilbert space, which is different from real
3-dimensional space where the spin directions are in. For example:
But,
1
h+Z|+Xi = . (3.2.37)
2
(c) When S ~ is along +y ( = /2 and = /2), that is, S ~ = +S y, the quantum
~ is along y ( = /2 and = 3/2), the
state of the spin is |+Y i; and when S
quantum state of the spin is |Y i, we have
1 1 1
|+Y i = (|+Zi + i |Zi) = (3.2.38)
2 2 i
1 1 1
|Y i = (|+Zi i |Zi) = . (3.2.39)
2 2 i
cos 4
i 1 1
|+nxy i = cos |+Zi + sin e |Zi = i = i (3.2.40)
4 4 sin 4 e 2 e
sin 4
i(+) 1 1
|nxy i = sin |+Zi + cos e |Zi = = i(+) .
4 4 cos 4 ei(+) 2 e
(3.2.41)
(e) For the most general case, S ~ is along +n = sin cos x + sin sin y + cos z
~ = +S n), the quantum state is |+ni; and S
(S ~ is along n = sin cos x
sin sin y cos z (S~ = S n), the quantum state is |ni, where = {0, } and
= {0, 2}, we have
cos 2
i
|+ni = cos |+Zi + sin e |Zi = , (3.2.42)
2 2 sin 2 ei
sin 2
i(+)
|ni = sin |+Zi + cos e |Zi = . (3.2.43)
2 2 cos 2 ei(+)
11
Operator A ai |ai i |ai i in {|ai i} |ai i in Z basis basis-independent A A in eigenbasis
A in Z
basis
1 1
1 1 0 0 1
SX S |+Xi 0 1 SX = S |+Xi h+X| + (S) |Xi hX| SX = S SX = S
2
0 1 1 0
-S |Xi 0 1 1
1 2 1
1 1
1 1 0 0 i
SY S |+Y i 0 i SY = S |+Y i h+Y | + (S) |Y i hY | SY = S SY = S
2
0 1 i 0
-S |Y i 0 1 1
1 2 i
1
1 1 0 1 0
SZ S |+Zi 0 0 SZ = S |+Zi h+Z| + (S) |Zi hZ| SZ = S SZ = S
0 1 0 1
-S |Zi 0 0
1 1
Note:
1. S = ~/2.
2. Z basis is {|+Zi,|Zi} basis.
12
3.3 Eigenvalue Problems
Now let us look at things from a different angle, that is, if we know the observable in the
matrix representation, how do we find out eigenvalues and eigenstates. This is to solve
eigenvalue problem.
0 1
Solve eigenvalues of SX = S and find out eigenvectors SX |i = |i.
1 0
0 S a a S a
= =0
S 0 b b S b
S
det = 2 S 2 = 0
S
Thus, = S .
For = +S,
0 S a a
=S
S 0 b b
Sb = Sa a=b
|a|2 + |b|2 = 2|a|2 = 1
Sa = Sb
|a| = 12 a = 12 = b
1
1
|+Si = |+Xi = .
2 1
For = S,
0 S a a
= S
S 0 b b
Sb = Sa a = b
|a|2 + |b|2 = 2|a|2 = 1
Sa = Sb
|a| = 12 a = 12 = b
1
1
|Si = |Xi = .
2 1
For a Stern-Gerlach apparatus aligned along +n = sin cos x + sin sin y + cos z,
the operator of the observable is Sn = S |+ni h+n| + (S) |ni hn|. Please work out the
following exercise: (1) find out the matrix representation of Sn in Z basis; (2) verify that
Sn does have eigenvalues +S and S, by solving eigenvalue problem as shown here. (If you
are realy ambitious, work out the eigenstates this way, and compare with equations ( 3.2.42)
and ( 3.2.43)).
13
3.4 Compatible and incompatible observables
If [A, B] = 0, A and B commute. Then, A and B are called compatible observables (quant-
ities). It means that the system can be in a state that is both an eigenstate of A and an
eigenstate of B. The common eigenstate can be labelled as |ai , bj i where {ai }, {bj } are the
eigenvalues for A and B respectively.
If [A, B] 6= 0, A and B are called incompatible observables.
Example:
0 1 ~
SX = S S=
1 0 2
1 0
SZ = S
0 1
[SZ , SX ] = SZ SX SX SZ
2 1 0 0 1 0 1 1 0
=S
0 1 1 0 1 0 0 1
0 1 0 1
= S2
1 0 1 0
~ 0 2
= ( )2 = i~SY (3.4.3)
2 2 0
SG experiment
Another way of saying that two observables are incompatible is to phrase it as the famous
Heisenberg uncertainty relations.
14
Variance of measured quantity: Suppose quantum system is in state |i.
Variance (uncertainty):
A)2
P
2 i (a(i)
(A) =
M
P 2 2
i (a(i) ) 2a(i) A + A
=
M
2 2
=A A
(A )2 = A2 hAi2
(3.4.11)
Note: If A = 0,
all the measurements give the same result, |i is an eigenstate of A.
Example:
1.
[X, PX ] = i~ (3.4.13)
~
X PX (3.4.14)
2
2.
~
[SZ , SX ] = i~SY , SZ SX | hSY i | (3.4.15)
2
If |i = |+Y i
1 1
|+Y i = in Z basis.
2 i
hSZ i+Y = h+Y | SZ |+Y i
1 1 0 1
= (1, i)
2 0 1 i
1 1
= (1, i) =0 (3.4.16)
2 i
15
hSX i = h+Y | SX |+Y i = 0 (3.4.17)
2 +Y
SZ +Y = h+Y | SZ2 |+Y i
~2 1 ~2
1 0 1
= (1, i) = (3.4.18)
4 2 0 1 i 4
2 ~2
SX = (3.4.19)
4
~
SZ2 = SZ2 hSZ i2
SZ =
2
~2
=
4
~
2 2
hSX i2
SX = SX SX =
2
~2
=
4
~
hSY i+Y = h+Y | SY |+Y i =
2
~ ~ ~ ~ ~
SZ SX = , | hSY i | = (3.4.20)
2 2 2 2 2
3. Energy-time uncertainty
E t h, (3.4.21)
where t is not an operator. A state that only exists for a short time cannot have
definite energy.
16
hSZ i+n = h+n| SZ |+ni
cos 2
i ~ 1 0
= cos sin e
2 2 2 0 1 sin 2 ei
~ ~
= (cos2 sin2 ) = cos (3.5.1)
2 2 2 2
17