Quantum Mechanics Formulas: by R.L.

Griffith@ UCB

Physical Constants

Name Symbol Value Unit

Number π π 3.14159265358979323846
Number e e  n 2.71828182845904523536

P
Euler’s constant γ = lim 1/k − ln(n) = 0.5772156649
n→∞ k=1

Elementary charge e 1.60217733 · 10−19 C

Gravitational constant G, κ 6.67259 · 10−11 m3 kg−1 s−2
Fine-structure constant α = e2 /2hcε0 ≈ 1/137
Speed of light in vacuum c 2.99792458 · 108 m/s (def)
Permittivity of the vacuum ε0 8.854187 · 10−12 F/m
Permeability of the vacuum µ0 4π · 10−7 H/m
(4πε0 )−1 8.9876 · 109 Nm2 C−2
Planck’s constant h 6.6260755 · 10−34 Js
Dirac’s constant ~ = h/2π 1.0545727 · 10−34 Js
Bohr magneton µB = e~/2me 9.2741 · 10−24 Am2
Bohr radius a0 0.52918 Å
Rydberg’s constant Ry 13.595 eV
Electron Compton wavelength λCe = h/c 2.2463 · 10−12 m
Proton Compton wavelength λCp = h/mp c 1.3214 · 10−15 m
Reduced mass of the H-atom µH 9.1045755 · 10−31 kg
Stefan-Boltzmann’s constant σ 5.67032 · 10−8 Wm−2 K−4
Wien’s constant kW 2.8978 · 10−3 mK
Molar gasconstant R 8.31441 J·mol−1 ·K−1
Avogadro’s constant NA 6.0221367 · 1023 mol−1
Boltzmann’s constant k = R/NA 1.380658 · 10−23 J/K
Electron mass me 9.1093897 · 10−31 kg
Proton mass mp 1.6726231 · 10−27 kg
Neutron mass mn 1.674954 · 10−27 kg
1 12
Elementary mass unit mu = 12 m( 6 C) 1.6605656 · 10−27 kg
Nuclear magneton µN 5.0508 · 10−27 J/T
Diameter of the Sun D⊙ 1392 · 106 m
Mass of the Sun M⊙ 1.989 · 1030 kg
Rotational period of the Sun T⊙ 25.38 days
Radius of Earth RA 6.378 · 106 m
Mass of Earth MA 5.976 · 1024 kg
Rotational period of Earth TA 23.96 hours
Earth orbital period Tropical year 365.24219879 days
Astronomical unit AU 1.4959787066 · 1011 m
Light year lj 9.4605 · 1015 m
Parsec pc 3.0857 · 1016 m
Hubble constant H ≈ (75 ± 25) km·s−1 ·Mpc−1

1
Chapter 1: The wave Function Chapter 2: Stationary States

Schrodinger Equation seperation of variables

∂ψ ~2 ∂ 2 ψ
i~ =− +Vψ (1) Ψ(x, t) = ψ(x)ϕ(t) = ψ(x)e−iEt/~ (12)
∂t 2m ∂x2
Probability
dϕ iE
=− ϕ (13)
Z b dt ~
Pab = |ψ(x, t)|2 dx (2)
a
ϕ(t) = e−iEt/~ (14)
For Normalization

Ψ(x, t) = ψ(x)e−iEt/~ (15)

Z ∞
1= |Ψ(x, t)|2 dx, where |Ψ(x, t)|2 = Ψ∗ Ψ Time Independent Schrodinger equation
−∞
(3)
~2 d2 ψ
once Ψ is normalized it stays normalized for − + V ψ = Eψ (16)
2m dx2
all time
Hamiltonian
∞
d
Z
|Ψ(x, t)|2 dx = 0 (4) p2
dt −∞ H(x, p) = + V (x) (17)
2m
Operators for momentum and position in gen-
substitute p = (~/i)(∂/∂x) and the corre-
eral form
sponding Hamiltonian operator is

Z ~2 d2 ψ
hxi = Ψ∗ xΨdx (5) Ĥ = − + V (x) = Eψ (18)
2m dx2
 
~ ∂
Z
∗
hpi = Ψ
i ∂x
Ψdx (6) hHi = ψ ∗ Ĥψdx = E (19)
 
~ ∂
Z
∗ You can always write the general solution of
hQ(x, p)i = Ψ Q x, Ψdx (7)
i ∂x the Schrodinger equation as a linear combination
of seperable solutions
Expectation value of kinetic energy
∞
X
2 2 Ψ(x, 0) = cn ψn (x) (20)
∂ Ψ
Z
~
hT i = − Ψ∗ dx (8) n=1
2m ∂x2

de Broglie formula ∞
X ∞
X
Ψ(x, t) = cn ψn (x)e−iEn t/~ = cn Ψn (x, t)
n=1 n=1
h 2π~
p= = (9) (21)
λ λ
The Infinite Square Well
uncertainty principle n
0, if 0≤x≤a
V (x) = ∞, otherwise (22)
~
σx σp ≥ (10)
2 plugging this into the TISE and we get

probability current d2 ψ
= −k 2 ψ (23)
dx2
∂Ψ∗ √
 
i~ ∗ ∂Ψ
J(x, t) ≡ {Ψ −Ψ (11) 2mE
2m ∂x ∂x k≡ (24)
~

2
 The harmonic Oscillator
ψ(x) = A sin kx (25)
nπ 1 2
kn = (26) V (x) = kx (38)
a 2
n2 π 2 ~2
r
k
En = (27) ω≡ (39)
2ma2 m
after normalization, the general TISE solu- 1
V (x) = mω 2 x2 (40)
tions for the infinite square well is 2
r rewriting equation 18 in a more suggestive
2  nπ 
form
ψn (x) = sin x (28)
a a
1 2
as a collection, the functions ψn (x) have some [p + (mωx)2 ]ψ = Eψ (41)
interesting properties 2m
1. They are alternately even and odd, with we can use raising and lowering operators to
respect to the center of the well. solve this problem
2. As you go up in energy, each succesive
state has one more node (zero crossing) 1
a+ ≡ √ (−ip + mωx) (42)
3. They are mutually orthogonal, in the 2~mω
sense that 1
a− ≡ √ (ip + mωx) (43)
Z 2~mω
ψm (x)∗ ψn (x)dx = 0 (29) 1
a− a+ = [p2 + (mωx)2 − imw(xp − px)]
2m~ω
Kronecker delta (44)
Z commutator
ψm (x)∗ ψn (x)dx = δmn (30)
n [A, B] ≡ AB − BA (45)
0, if m6=n
δmn = 1, if m=n (31)

we say that ψ’s are orthonormal

[x, p]f (x) = i~f (x) (46)
4. They are complete, in the sense that any
[x, p] = i~ (47)
other function, f (x), can be expressed as a linear
combination of them:
1 1
∞
2X
∞r  nπ  a− a+ = H+ (48)
f (x) =
X
cn ψn (x) = cn sin x ~ω  2 
a n=1 a 1
n=1 H = ~ω a− a+ − (49)
(32) 2
the cn ’s can fe found using Fourier’s trick 1 1
a+ a− = H− (50)
~ω 2
Z [a− , a+ ] = 1 (51)
cn = ψn (x)∗ f (x)dx (33) 
1

H = ~ω a+ a− + (52)
2
r Z a
2  nπ 
= sin x Ψ(x, 0)dx (34)
a 0 a In terms of a± , then the Schrodinger equation
of the harmonic oscillator takes the form
Pψn = |cn |2 (35)  
1
∞ ~ω a± a∓ ± ψ = Eψ (53)
2
X
|cn |2 = 1 (36)
n=1
∞
X H(a+ ψ) = (E + ~ω)(a+ ψ) (54)
hHi = |cn |2 En (37)
n=1 H(a− ψ) = (E − ~ω)(a− ψ) (55)

3
 there must be a lowest state and this occurs
when
p = ~k (70)
r
a− ψ0 = 0 (56) E
vquantum = (71)
2m
this means vclassical = 2vquantum (72)
 
1 d
√ ~ + mωx ψ0 = 0 (57)
2~mω dx ∞
1
Z
~k2
Ψ(x, t) = √ φ(k)ei(kx− 2m t) dk (73)
the genral solution for the ground state of the 2π −∞
harmonic oscillator is
to determine φ(k) we can use the TISE which
 mω 
− mω
2~ x
2 is
ψ0 (x) = e (58)
π~
∞
1
Z
Ψ(x, 0) = √ φ(k)eikx dk (74)
2π −∞
1
E0 = ~ω (59) this is a classic problem in Fourier analysis;
2

1
 the answer is provided by Plancherel’s theo-
En = n + ~ω (60) rem
2
ψn = An (a+ )n ψ0 (x) (61) Z ∞
1
f (x) = √ F (k)eikx dk (75)
2π −∞
Z ∞
1
a+ a− ψn = nψn , a− a+ ψn = (n + 1)ψn (62) F (k) = √ f (x)e−ikx dx (76)
2π −∞

so the solution to the generic quantum prob-

√ √ lem, for the free particle is equation 74, with
a+ ψn = n + 1ψn+1 , a− ψn = nψn−1 (63)
∞
1
Z
1 φ(k) = √ Ψ(x, 0)e−ikx dx (77)
ψn = √ (a+ )n ψ0 (64) 2π −∞
n!
The Delta-Function Potential
r
~
x= (a+ + a− ) (65) E<0 ⇒ bound state
(78)
2mω E>0 ⇒ scattering state

r
~mω Z ∞
p=i (a+ − a− ) (66)
n o
0, if x6=0
2 δ(x) ≡ ∞, if x=0 , with δ(x)dx = 1
−∞

The Free particle (79)

V (x) = 0 everywhere Z ∞ Z ∞
f (x)δ(x−a)dx = f (a) δ(x−a)dx = f (a)
−∞ −∞
~k ~k (80)
Ψ(x, t) = Aeik(x− 2m t + Be−ik(x+ 2m t) (67)
thats the most important property of the
delta function: Under the integral sign it serves
~k2 to “pick out” the value of f (x) at the point a.
Ψk (x, t) = Aei(kx− 2m t (68)
Let’s consider a potential of the form

√ V (x) = −αδ(x) (81)

2mE n
k≡± , with k>0⇒ travelingtotheright
k<0⇒ travelingtotheleft plugging this into equation 18 in the region
~
(69) x < 0, V (x) = 0 so

4
 where
d2 ψ 2mE
2
= − 2 ψ = k2 ψ (82) √
dx ~ 2mE
k≡ (92)
where ~
√ is real and positive. The general solution is
−2mE for x < 0
k≡ (83)
~
for bound states E < 0 the general solution is ψ(x) = Aeikx + Be−ikx (93)
when x < 0 and this time we cannot rule out either term,
since neither blows up. and similarly for x > 0
ψ(x) = Ae−kx + Bekx (84)

the first term blows up as x → −∞ so we ψ(x) = F eikx + Ge−ikx (94)

must choose A=0 for boundary condition number 1 this implies

ψ(x) = Bekx (85) F +G=A+B (95)

for bound states E < 0 the general solution is the derivatives are
when x > 0
dψ/dx=ik(F eikx +Ge−ikx ), for(x>0), so dψ/dx|+ =ik(F −G)
n
ψ(x) = F e−kx + Gekx (86) dψ/dx=ik(Aeikx +Be−ikx ), for(x<0), so dψ/dx|− =ik(A−B)
(96)
the second term blows up as x → ∞ so we
and hence ∆(dψ/dx) = ik(F − G − A + B),
must choose G=0
meanwhile ψ(0) = (A+B), so the second bound-
ery condition says
ψ(x) = F e−kx (87)

It remains only to stitch these two functions 2mα

together, using the appropriate boundery condi- ik(F − G − A + B) = − (A + B) (97)
~2
tions at x = 0, the standard boundary conditions
for ψ or more compactly
1. ψ is always continuous 2. dψ/dx is con-
tinuous except at points where the potential is mα
F −G = A(1+2iβ)−B(1−2iβ), where β ≡
infinite ~2 k
(98)
in this case the first boundary condition tells
us that F = B, so having imposed both boundary condition we
are left with two equations and four unknowns.
Bekx , it follows A is the amplitude of a wave coming
n
(x≤0)
ψ(x) = (88)
Be−kx , (x≥0) from the right, B is the amplitude of the wave
the second boundary condition gives returning to the left, F is the amplitude of the
wave traveling of to the right, and G is the am-

dψ

2mα plitude of the wave coming in from the right.
∆ =− ψ(0) (89)
dx ~2
Evidently the delat-function well, regardless G = 0, for scattering from the left (99)
of its “strength” α, has exactly one bound state A = 0, for scattering from the right (100)

√ for scattering from the left, A is the amplitude

mα −mα|x|~2 mα2
ψ(x) = e ; E=− 2
~ 2~
What about scattering states, with E > 0?For
x < 0 the Schrodinger equation read
of the incident wave, B is the amplitude of the
(90)
reflected wave, and F is the amplitude of the
transmitted wave solving equations 95 and 98
for B and F , we find

d2 ψ 2mE iβ 1
= − 2 ψ = −k 2 ψ (91) B= A, F = A (101)
dx 2 ~ 1 − iβ 1 − iβ

5
 The reflection coefficient is and the general solution is

|B|2 β2
R≡ = (102)
|A|2 1 + β2 ψ(x) = C sin(lx) + D cos(lx), for − a < x < a
(114)
and the transmission coefficient is
when x > a the genral solution is
|F |2 1
T ≡ = (103) ψ(x) = F e−kx + Gekx (115)
|A|2 1 + β2
and the sum should be one the first term blows up as x → −∞ so we
must choose G=0
R+T =1 (104)
ψ(x) = F e−kx , for x > a (116)
R and T are functions of β and hence func-
tions of E The next step is to impose boundary condi-
tions: ψ and dψ/dx continuous at −a and a.
but we can save a little time by noting that this
1 potential is an even function, so we can assume
R= (105)
1 + (2~2 E/mα2 ) with no loss of generality that the solutions are
1 either even or odd. since ψ(−x) = ±ψ(x).
T = (106)
1 + (mα2 /2~2 E) for even solutions use D cos(lx) and for odd
solutions use C sin(lx). I will show the even so-
The Finite Square Well lutions
n
−V0 , for −a≤x≤a
The continuity of ψ(x), at x = a, says
V (x) = 0, for |x|>a (107)
F e−ka = D cos(la) (117)
This potential admits both bound states and
scattering states. we will look at the bound and the continuity of dψ/dx, says
states first. in the region x < −a the potential
is zero, so the Schrodinger equation reads −kF e−ka = −lD sin(la) (118)
d2 ψ 2mE deviding equation 118 by 117 , we find that
= − 2 ψ = k2 ψ (108)
dx2 ~
where k = l tan(la) (119)
√ This is a formula for the allowed energies,
−2mE
k≡ (109) since k and l are both functions of E. To solve
~ for E, we first adopt a nicer notation: Let
the general solution as before is
ap
ψ(x) = Ae−kx + Bekx (110) z ≡ la, and z0 ≡ 2mV0 (120)
~
the first term blows up as x → −∞ so we
must choose A=0
2mV0
q
(k 2 + l2 ) = , so ka = z02 − z 2 (121)
~2
ψ(x) = Bekx , for x < −a (111)
and now equation 119 read
in the region −a < x < a,V (x) = −V0 , and
the Schrodinger equation reads p
tan(z) = (z0 /z)2 − 1 (122)
This is a trancendental equation and can olny
~2 d2 ψ d2 ψ
− 2
− V0 ψ = Eψ, or = −l2 ψ be solved numerically or graphically. Two limit-
2m dx dx2 ing cases are of special interest.
(112)
where 1. Wide, deep well. if z0 is very large,
the intersections occur just slightly below zn =
p
2m(E + V0 )
l≡ (113) nπ/2, with n odd it follows that
~

6
 continuity of dψ/dx at +a gives
n2 π 2 ~2
En + V0 ∼
= (123)
2m(2a)2
l[C cos(la) − D sin(la)] = iF eika (132)
2. Shallow, narrow well. As z0 decreases,
there are fewer and fewer bound states, until fi-
nally (for z0 < π/2,where the lowest odd state We can use two of these to eliminate C and
disappears) only one remains. It is interesting D, and solve the remaining two for B and F :
to note that there is always one bound state, no
sin(2la) 2
matter how weak the well becomes. B=i (l − k 2 )F (133)
2kl
Now moving on to scattering states where
E>0
the general solution as before is e−2ika A
F = 2 2 (134)
cos(2la) − i (k2kl
+l
sin(2la)
ψ(x) = Aeikx + Be−ikx for (x < −a) (124)
The transmission coefficient (T = |F |2 /|A|2 )
expressed in terms of the original variables, is
where(as usual) given by
√
2mE
k≡ (125)
V02
 
~ 2a p
T−1
= 1+ sin2 2m(E + V0 )
inside the well, where V (x) = −V0 and the 4E(E + V0 ) ~
general solution is (135)
Notice that T = 1 (the well becomes “trans-
parent”) whenever the sine is zero, which is to
ψ(x) = C sin(lx) + D cos(lx), for − a < x < a say, when
(126)
2a p
where as before 2m(En + V0 ) = nπ (136)
~
p
2m(E + V0 ) where n is an integer. The energies for perfect
l≡ (127) transmission, then are given by
~
To the right, assuming there is no incoming n2 π 2 ~2
wave in this region, we have En + V0 = (137)
2m(2a)2
ψ(x) = F eikx (128) which happens to be the allowed energies for
the infinite square well
Here A is the incident amplitude, B is the re-
flected amplitude, and F is the transmitted am- Chapter 3: Formalism
plitude.
There are four boundary conditions: Conti- The collection of all functions of x constitues
nuity of ψ(x) at −a says a vector space. To represent a possible physical
state, the wave functionΨ must be normalized :
Z
Ae−ika + Beika = −C sin(la) + D cos(la) (129) |Ψ|2 dx = 1 (138)

continuity of dψ/dx at −a gives The set of all square-integrable functions,

on a specified interval

ik[Ae−ika − Beika ] = l[C cos(la) + D sin(la)] b

Z
(130) f (x) such that |f (x)|2 dx < ∞ (139)
a
continuity of ψ(x) at +a yields
this constitutes a (much smaller) vector space.
mathematicians call it L2 (a, b); physicist call it
C sin(la) + D cos(la) = F eika (131) Hilbert Space. In quantum mechanics, then

7
 Wave functions live in Hilbert Space now the outcome of a measurement has got to
We define the inner product of two func- be real, and so, a fortiori, is the average of many
tions, f (x) and g(x), as follows measurements:

b
hQi = hQi∗ (148)
Z
hf |gi ≡ f (x)∗ g(x)dx (140)
a
But the complex conjugate of an inner prod-
If f and g are both square-integrable (that uct reverses the order
is, if they are both in Hilbert space),their inner
product is guaranteed to exist (the integral in hΨ|Q̂Ψi = hQ̂Ψ|Ψi (149)
equation 140 converges to a finite number), This
follows from the integral Schwarz inequality and this must hold true for any wave function
Ψ. Thus operators representing observables have
Z  sZ the verty special property that
 b  b Z b
f (x)∗ g(x)dx ≤ |f (x)|2 dx |g(x)|2 dx
 

 a  a a
(141) hf |Q̂f i = hQ̂f |f i for all f (x) (150)
Notice in particular that
we call such operator hermitian
hg|f i = hf |gi∗ (142) Observables are represented by hermitian operators
Moreover, the inner product of f (x) with it- Determinate States
self,
Q̂Ψ = qΨ (151)
Z b
hf |f i = |f (x)|2 dx ≥ 0 (143) This is the eigenvalue equation for the op-
a
erator Q̂;Ψ is an eigenfunction of Q̂, and q is
is real and non-negative; it’s zero only if the corresponding eigenvalue; thus
f (x) = 0
Determinate states are eigenfunctions of Q̂
A function is said to be normalized if its
inner product with itself is 1; two functions are Measurement of q on such a state is certain
orthogonal if their inner product is 0; and a set to yield the eigenvalue q. The collection of all
of functions, {fn }, is orthonormal if they are eigenvalues of an operator is called its spec-
normalized and mutually orthogonal; trum. Sometimes two (or more) linearly in-
dependent eigenfunctions share the same eigen-
value; in that case the spectrum is said to be
hfm |fn i = δmn (144)
degenerate.
Finally, a set of functions is complete if any Eigenfunctions of Hermitian Operators
other function (in Hilbert space) can be ex- Discrete Spectra
pressed as a linear combination of them
Mathematically, the normalized eigenfunc-
∞
X tions of a hermitian operato have two important
f (x) = cn fn (x) (145) properties.
n=1 Theorem 1: Their eigenvalues are real.
if the functions {fn (x)} are orthonormal, the
coefficients are given by Fourier’s trick Q̂f = qf (152)
(i.e.,f (x) is an eigenfunction of Q̂, with eigen-
cn = hfn |f i (146) value q), and
Observables
The expectation value of an observable Q(x, p) hf |Q̂f i = hQ̂f |f i (153)
can be expressed very neatly in inner-product
then
notation:

Z qhf |f i = q ∗ hf |f i (154)
∗
hQi = Ψ Q̂Ψdx = hΨ|Q̂Ψi (147) and q must be real

8
 Theorem 2: Eigenfunctions belonging to Similarly, the expectation value Q should be
distinct eigenvalues are orthogonal. the sum over all possible outcomes of the eigen-
value times the probablity of getting that eigen-
Q̂f = qf, and Q̂g = q ′ g (155) value

and Q̂ is hermitian. Then hf |Q̂gi = hQ̂f |gi

X
hQi = qn |cn |2 (162)
so n

The momentum space wave function,

q ′ hf |gi = q ∗ hf |gi (156) Φ(p, t) is essentially the Fourier transform of the
again the inner product exist because the position space wave function Ψ(x, t), which by
eigenfunctions are in Hilbert space by assump- Plancherel’s theorem is its inverse Fourier trans-
tion. but q is real so if q ′ 6= q it must be that form.
hf |gi = 0.
Z ∞
Continuous Spectrum 1
Φ(p, t) = √ e−ipx/~ Ψ(x, t)dx (163)
If the spectrum of a hermitian operator is con- 2π~ −∞
tinuous, the eigenfunctions are not normalizable, 1
Z ∞
and the proofs of Theorems 1 and 2 fail, because Ψ(x, t) = √ eipx/~ Φ(p, t)dp (164)
2π~ −∞
the inner products may not exist.
(165)
Generalized Statistical Interpretation
If you measure an observable Q(x, p) on a par- According to the generalized statistical inter-
ticle in the state Ψ(x, t), you are certain to get pretation, the probability that a measurement of
one of the eigenvalues of the hermitian operator. the momentum would yield a result in the range
If the spectrum is dicrete, the probability of get- dp is
ting the particular eigenvalue qn associated with
the orthonormalized eigenfunction fn (x) is
|Φ(p, t)|2 dp (166)

|cn |2 , where cn = hfn |Ψi (157) The Uncertainty Principle

The generalized uncertainty principle is
If the spectrum is continuous, with real eigen-
values q(z) and associated Dirac orthonormal- 
1
2
2 2
ized eigenfunctions fz (x), the probability of get- σA σB ≥ h[Â, B̂]i (167)
2i
ting result in the range dz is
The energy-time uncertainty principle

|c(z)|2 dz, where c(z) = hfz |Ψi (158) * +

d i ∂ Q̂
hQi = h[Ĥ, Q̂]i + (168)
The eigenfunctions of a hermitian operator dt ~ dt
are complete, so that the wave function can be
written as a linear combination of them Chapter 4: Quantum Mechanics in Three
X Dimensions
Ψ(x, t) = cn fn (x) (159)
n
The generalization to three dimensions is
straightforward. Schrodinger’s equation says
Because the eigenfunctions are orthonormal,
the coefficients are given by Fourier’s trick ∂Ψ
i~ = HΨ (169)
∂t
Z the Hamiltonian operator H is obtained from the
cn = hfn |Ψi = fn (x)∗ Ψ(x, t)dx (160) classical energy

Of course, the total probability (summed over 1 1 2

all possible outcomes) has got to equal one mv 2 + V = (p + p2y + p2z ) + V (170)
2 2m x
by the standard prescription (applied now to
X
|cn |2 = 1 (161)
n
y and z, as well as x)

9
 ~ ∂ ~ ∂ ~ ∂ ψ(r, θ, φ) = R(r)Y (θ, φ) (180)
px → , py → , pz → (171)
i ∂x i ∂y i ∂z ,. The Angular Equation
or
∂2Y
 
∂ ∂Y
~ sin θ sin θ + = −l(l + 1) sin2 θY
p→ ∇ (172) ∂θ ∂θ ∂φ2
i
(181)
thus
you might recognize this equation, it occurs
in the solution to Laplace’s equation in classical
∂Ψ ~2 2
i~ =− ∇ Ψ+VΨ (173) electrodynamics. as always, we try seperation of
∂t 2m variables
where
Y (θ, φ) = Θ(θ)Φ(φ) (182)
∂2
2 ∂2 ∂2
∇ ≡ + + (174)
∂x2 ∂y 2 ∂z 2 plugging this in and deviding by ΘΦ gives us
two solutions
is the Laplacian, in cartesian coordinates.
The potential V and the wave function Ψ are
now functions of r = (x, y, z) and t. 1

d

dΘ

the normalization condition reads sin θ sin θ + l(l + 1) sin2 θ = m2
Θ dθ dθ
Z (183)
|Ψ|2 d3 r = 1 (175)
1 d2 Φ
= −m2 (184)
with the integral taken over all space. if the Φ dφ2
potential is independent of time, there will be a
The φ equation is easy
complete set of stationary states

Ψn (r, t) = ψn (r)e−iEn t/~ (176) d2 Φ

= −m2 Φ ⇒ Φ(φ) = e±imφ (185)
dφ2
where the spatial wave function ψn satisfies
the time-independent Schrodinger
Now when φ advances by 2π, we return to the
same point in space, so it is natural to require
~2 2
− ∇ ψ + V ψ = Eψ (177) that
2m
The general solution to the time-dependent Φ(φ + 2π) = Φ(φ) (186)
Schrodinger equation is
from this it follows that m must be an integer
X
Ψ(r, t) = cn ψn (r)e−iEn t/~ (178) m = 0, ±1, ±2, .... (187)
Seperation Of variables Typically, the po- The θ equation is not so simple, the solution
tential is a function only of the distance from is
the origin. In that case it is natural to adopt
spherical coordinates, (r, θ, φ). In spherical Θ(θ) = APlm (cos θ) (188)
coordinates the Laplacian takes the form
where Plm is the associated Legendre func-
    tion, defined by
1 ∂ ∂ 1 ∂ ∂
∇2 = r2 + 2 sin θ
r2 ∂r ∂r r sin θ ∂θ ∂θ  |m|
1
 2 
∂ d
+ 2 2 (179) Plm (x) 2 |m|/2
≡ (1 − x ) Pl (x) (189)
r sin θ ∂φ2 dx

We begin by looking for solutions that are and Pl (x) is the lth Legendre polynomial,
seperable by products defined by the Rodrigues formula

10
 this is as far as we can go until a specific po-
 l tential V (r) is provided.
1 d
Pl (x) ≡ (x2 − 1)l (190) The infinite spherical well
2l l! dx
n
0, ifr≤a
Now, the volume element in spherical coordi- V (r) = ∞, ifr>a (199)
nates is
Outside the well, the wave function is zero;
3 2 inside the well, the radial equation says
d r = r sin θdrdθdφ (191)
d2 u
 
so the normalization condition becomes l(l + 1) 2
= − k u (200)
dr2 r2
Z where
|ψ|2 r2 drdθdφ =
√
Z Z 2mE
|R|2 r2 dr |Y |2 sin θdθdφ = 1 (192) k≡ (201)
~
as usual. Our problem is to solve this equa-
It is convenient to normalize R and Y sepa- tion, subject to the boundary condition u(a) =
rately 0. The case l = 0 is easy

Z ∞
|R|2 r2 dr = 1 (193) d2 u
= −k 2 u ⇒ u(r) = A sin(kr) + B cos(kr)
0 dr2
Z 2π Z π (202)
|Y |2 sin θdθdφ = 1 (194)
0 0
we must choose B = 0 because r → 0 the
radial wave function blows up. The boundary
The normalized angular wave functions are condition then requires sin(ka) = 0, and hence
called spherical harmonics ka = nπ, for some integer n. The allowed ener-
gies are evidently
s
(2l + 1) (l − |m|)! imφ m
Ylm (θ, φ) =ǫ e Pl (cos θ) n2 π 2 ~2
4π (l + |m|)! En0 = , (n = 1, 2, 3, ...) (203)
(195) 2ma2
m
where ǫ = (−1) for m ≥ 0 and ǫ = 1 for the same as for the one-dimensional infinite
m ≤ 0. As we shall prove later on, they are square well. The general solution to equation
automatically orthogonal. 200 (for an arbitrary integer l) is not as familiar
The Radial Equation Notice that the an-
gular part of the wave function, Y (θ, φ), is the u(r) = Arjl (kr) + Brnl (kr) (204)
same for all spherically symmetric potentials; the
actual shape of the potential ,V (r), affects only where jl (x) is the spherical Bessel function
the radial part of the wave function,R(r), which pf order l, and nl (x) is the spherical Neumann
is determined by function of order l. They are defined as follows

u(r) ≡ rR(r) (196)  l

1 d sin x
using this relationship we can now write the jl (x) ≡ (−x)l (205)
x dx x
radial equation as

 l
~2 d2 u

~2 l(l + 1)
 1 d cos x
− + V + u = Eu (197) nl (x) ≡ −(−x)l (206)
2m dr2 2m r2 x dx x

the normalization condition becomes Notice that the Bessel functions are finite at
the origin, but Neumann functions blow up at
Z ∞ the origin. Accordingly, we must have Bl = 0,
|u|2 dr = 1 (198) and hence
0

11
 so that
R(r) = Ajl (kr) (207)
There remains the boundary condition, R(a) = d2 u
 
ρ0 l(l + 1)
0. Evidently k must be chosen such that = 1− + u (217)
dρ2 ρ ρ2

jl (ka) = 0 (208) Next we examine the asymptotic form of the

solutions. As ρ → ∞, the constant term in the
the boundary condition requires that brackets dominate, so approximately
1 d2 u
k= βnl (209) =u (218)
a dρ2
where βnl is the nth zero of the lth spherical The general solution is
Bessel function. The allowed energies then, are
given by
u(ρ) = Ae−ρ + Beρ (219)
2
~ but eρ blows up ( as ρ → ∞), so B = 0,
Enl = β2 (210)
2ma2 nl evidently
and the wave functions are
u(ρ) ∼ Ae−ρ (220)

ψnlm (r, θ, φ) = Anl (βnl r/a)Ylm (θ, φ) (211) for large ρ. On the other hand, as ρ → 0 the
centrifugal term dominates, approximately then
with the constant Anl to be detemined by nor-
malization. d2 u l(l + 1)
= u (221)
dρ2 ρ2
The Hydrogen Atom the general solution is
From Coulomb’s law, the potential energy is
u(ρ) = Cρl+1 + Dρ−l (222)
2
e 1 let D = 0, because ρ−l blows up as ρ → 0,
V (r) = − (212)
4πǫ0 r thus
and the radial equation says
u(ρ) ∼ Cρl+1 (223)

~2 d2 u e2 1 ~2 l(l + 1) The next step is to peel off the asymptotic

 
− + − + u = Eu behaviour, introducing the new function v(ρ)
2m dr2 4πǫ0 r 2m r2
(213)
our problem is to solve this equation for u(r), u(ρ) = ρl+1 e−ρ v(ρ) (224)
and determine the allowed energies. Our first Finally, we assume the solution, v(ρ), can be
task is to tidy up the notation. Let expressed as a power series in ρ

√ ∞
−2mE k 2 ~2
X
k≡ ⇒E=− (214) v(ρ) = cj ρ j (225)
~ 2m j=0

dividing equation 213 by E gives  

2(j + l + 1) − ρ0
cj+1 = cj (226)
(j + 1)(j + 2l + 2)
2 2
 
1 d u me 1 l(l + 1)
= 1− + u
2
k dr 2 2
2πǫ0 ~ k (kr) (kr)2 This recursion formula determines the coef-
(215) ficients, and hence the function v(ρ). Now lets
This suggest that we introduce see what the coefficients look like for large j (this
corresponds to large ρ, where the higher power
me2 dominate . In this regime the recursion formula
ρ ≡ kr and ρ0 ≡ (216) says
2πǫ0 ~2 k

12
 2j 2 4πǫ0 ~2
cj+1 ∼
= cj = cj (227) a≡ = 0.529x10−10m (238)
j(j + 1) j+1 me2

suppose for a moment that this were exact. is called the Bohr radius. it follows that
Then r
ρ= (239)
2j an
cj = c0 (228) The spatial wave functions for hydrogen are
j!
labeled by three quantum numbers (n, l, andm)
so
∞
X 2j ψnlm (r, θ, φ) = Rnl (r)Ylm (θ, φ) (240)
v(ρ) = c0 ρj (229)
j=0
j!
and
and hence
1 l+1 −ρ
l+1 ρ
Rnl (r) = ρ e v(ρ) (241)
u(ρ) = c0 ρ e (230) r
and v(ρ) is a polynomial of degree jmax =
which blows up at large ρ. The series must
n − l − 1 in ρ, whose coefficients are determined
terminate. There must occur some maximal in-
by the recursion formula
teger, jmax such that
2(j + l + 1 − n
cjmax +1 = 0 (231) cj+1 = cj (242)
(j + 1)(j + 2l + 2)
(and beyond which all coefficients vanish au- The polynomial v(ρ) is a function well known
tomatically). equation 226 becomes to mathematicians; apart from normalization, it
can be written as
2(jmax + l + 1) − ρ0 = 0 (232)
Defining v(ρ) = L2l+1
n−l−1 (2ρ) (243)
where
n ≡ jmax + l + 1 (233)
p
(the so-called principal quantum num-

d
ber), we have Lpq−p (x) ≡ (−1)p Lq (x) (244)
dx

ρ0 = n (234) is an associated Laguerre polynomial,

and
But ρ0 determines E
 q
x d
k ~2 2
me 4 Lq (x) ≡ e (e−x xq ) (245)
E=− =− 2 2 2 2 (235) dx
2m 8π ǫ ~ ρ0
is the qth Laguerre polynomial, the nor-
so the allowed energies are malized hydrogen wave functions are

" 2 #
e2

m 1 E1
s 3  l
En = − = 2 n = 1, 2, .. 2 (n − l − 1)! − 2r
2~2 4πǫ n 2 n ψnlm = e r/na
na 2n[(n + l)!]3 na
(236)  2l+1
Ln−l−1 (2r/na) Ylm (θ, φ) (246)

This is the famous Bohr Formula. combin-
ing equation 216 and 234 we get
The spectrum of Hydrogen

me 2

1 1 An electron may undergo a transition to
k= = (237) some other energy state, either by absorbing en-
4πǫ0 ~2 n an
ergy or emmiting energy, the difference in energy
where between the initial and final states are

13
 !
1 1 [L2 , Lx ] = 0, [L2 , Ly ] = 0, [L2 , Lz ] = 0 (255)
Yγ = Ei − Ef == 13.6eV 2 − 2 (247)
ni nf
or, more compactly,
Now, according to the Planck formula, the
energy of a photon is proportional to its fre- [L2 , L] = 0 (256)
quency So L2 is compatibale with each component of
textbfL, and we can hope to find simultaneous
Eγ = hv (248) eigenstates of L2 and (say) Lz
Meanwhile, the wavelength is given by λ =
c/v, so L2 f = λf and Lz f = µf (257)
! We’ll use a ladder operator technique, Let
1 1 1
=R − 2 (249)
λ n2f ni L± ≡ Lx ± iLy (258)
where The commutator with Lz is

2 [Lz , L± ] = ±~(Lx ± iLy ) (259)

e2

m 7 −1
R≡ = 1.097 × 10 m (250) so
4πc~3 4πǫ0

is know as the Rydberg constant [Lz , L± ] = ±~L± (260)

Angular Momentum
And of course,
Classically, the angular momentum of a par-
ticle (with respect to the origin) is given by the
[L2 , L± ] = 0 (261)
formula
I claim that if f is an eigenfunction of L2 and
L=r×p (251) Lz , so also is L± f equation 261 says

which is to say
Lx = ypz −zpy , Ly = zpx −xpz , Lz = xpy −ypx
(252)
The corresponding quantum operators are
obtained by the standard prescription px →
−i~∂/∂x etc.
Eigenvalues
The operators Lx and Ly do not commute; in
fact
[Lx , Ly ] = i~Lz , [Ly , Lz ] = i~Lx , [Lz , Lx ] = i~Ly
(253)
Notice that Lx , Ly , and Lz are incompatible
observables and it would therefore be futile to
look for states that are simultaneously eigenfunc-
tions of Lx and Ly . On the other hand, the
square of the total angular momentum
L2 ≡ L2x + L2y + L2z (254)
does commute with Lx
L2 (L± f ) = L± (L2 f ) = L± (λf ) = λ(L± f )
(262)
2
so if L± f is an eigenfunction of L , with the
same eigenvalue λ, and equation 260 says
Lz (L± f ) = (Lz L± − L± Lz ) + L± Lz f = ±~L± + L± (µf )
(263)
= (µ ± ~)(L± f ) (264)
so L± f is an eigenfunction of Lz , with the
new eigenvalue µ ± ~. we call L+ the raising
operator, because it increases the eigenvalue of
Lz by ~, and L− the lowering operator, because
it lowers the eigenvalue by ~. When using the
raising operator we will eventually reach a state
for which the z-component exceeds the total
L ± ft = 0 (265)
let ~l be thje eigenvalue of Lz at this top rung.
Lz ft = ~lft ; L2 ft = λft (266)

14
 Now,

θ̂ = (cos θ cos φ)î + (cos θ sin φ)ĵ − (sin θ)k̂

L± L∓ = L2 − L2z ∓ i(i~Lz ) (267)
(276)
or putting it the other way around φ̂ = −(sin φ)î + (cos φ)ĵ (277)

L2 = L± L∓ + L2z ∓ ~Lz (268) Evidently

it follows that
 
~ ∂ ∂
Lx = − sin φ − cos φ cot θ (278)
i ∂θ ∂φ
2
L ft = (L− L+ + L2z 2
+ ~Lz )ft = ~ l(l + 1)ft  
~ ∂ ∂
(269) Ly = + cos φ − sin φ cot θ (279)
i ∂θ ∂φ
and nence
and
λ = ~2 l(l + 1) (270)
~ ∂
This tells us the eigenvalue of L2 in terms of Lz = (280)
i ∂φ
the maximum eigenvalue of Lz . Evidently the
eigenvalues of Lz are m~, where m goes from −l using the raising and lowering operators we
to l in N integer steps. In particular, it follows find
that l = −l+N , and hence l = N/2, so l must be
an integer or a half-integer. The eigenfunctions
1 ∂2
   
1 ∂ ∂
are characterized by the numbers l and m L2 = −~2 sin θ +
sin θ ∂θ ∂θ sin2 θ ∂φ2
(281)
L2 flm = ~2 l(l + 1)flm ; Lz flm = ~mflm (271) conclusion: Spherical harmonics are eigen-
functions of L2 and Lz . When we solved the
where Schrodinger equation by seperation of variables,
we were inadvertently constructing simultaneous
eigenfunctions of the three commuting operators
l = 0, 1/2, 1, 3/2, ...; m = −l, −l + 1, ..., l − 1, l H, L2 , and Lz
(272)
Eigenfunctions
Hψ = Eψ, L2 ψ = ~2 l(l + 1)ψ, Lz ψ = ~mψ
First of all we need to rewrite Lx , Ly , and Lz (282)
in spherical coordinates. Now, L=(~/i)(r× ∇) ,
incidentally, we can use equation 281 to
and the gradient, in spherical coordinates, is
rewrite the Schrodinger equation

∂ 1 ∂ 1 ∂    
∇ = r̂ + θ̂ + φ̂ (273) 1 2 ∂ 2 ∂ 2
∂r r ∂θ r sin θ ∂φ −~ r + L ψ + V ψ = Eψ
2mr2 ∂r ∂r
(283)
meanwhile, r=rr̂, so
Spin
  In classical mechanics, a rigid object admits
~ ∂ ∂ 1 ∂ two kinds of angular momentum: orbital (L =
L= r(r̂ × r̂) + (r̂ × θ̂) + (r̂ × φ̂)
i ∂r ∂θ sin θ ∂φ r × p), associated with the motion of the center
(274) of mass, and spin (S = Iω) , associated with
But (r̂ × r̂) = 0, (r̂ × θ̂) = φ̂, and (r̂ × φ̂) = −θ motion about the center of mass. The algebraic
and hence theory of spin is a carbon copy of the theory of
orbital angular momentum, beggining with the
fundamental commutation relations
 
~ ∂ 1 ∂
L= φ̂ − θ̂ (275)
i ∂θ sin θ ∂φ

The unit vectors θ̂ and φ̂ can be resolved into [Sx , Sy ] = i~Sz , [Sy , Sz ] = i~Sx , [Sz , Sx ] = i~Sy
their cartesian components (284)

15
 it follows that the eigenvectors of S 2 and Sz Similarly
satisfy
~ ~
Sz χ+ = χ+ , Sz χ− = − χ− (293)
2 2
S |smi = ~ s(s + 1)|smi; Sz |smi = ~m|smi 2 2
(285) from which it follows that
and  
~ 1 0
Sz = (294)
p 2 0 −1
S± |smi = ~ s(s + 1) − m(m ± 1)|s(m ± 1)i
(286) Meanwhile, equation 286 says
where S± ≡ Sx ±iSy . But this time the eigen-
functions are not spherical harmonics , and there S+ χ− = ~χ+ , S− χ+ = ~χ− , Sχ+ = Sχ− = 0
is no apriori reason to exclude the half-integer (295)
values of s and m:
so

1 3    
s = 0, , 1, , ...; m = −s, −s + 1, ..., s − 1, s. 0 1 0 0
2 2 S+ = ~ , S− = ~ (296)
(287) 0 0 1 0
Pi mesons have spin 1/2; photons have spin 1;
Now S± = Sx ± iSy , so Sx = (1/2)(S+ + S− )
deltas have spin 3/2; gravitons have spin 2; and
and Sy = (1/2i)(S+ − S− ), and hence
so on.
Spin 1/2    
By far the most important case is s = 1/2, for ~ 0 1 ~ 0 −i
Sx = , Sy = (297)
this is the spin of the particles that make up ordi- 2 1 0 2 i 0
nary matter( protons, neutrons, and electrons),
as well as quark and leptons. The general state Since Sx , Sy , and Sz all carry a factor of ~/2,
of a spin-1/2 particle can be expressed as a two- it is tidier to write S=(~/2)σ, where
element column matrix (or spinor);
     
  0 1 0 −i 1 0
a σx ≡ , σy ≡ , σz ≡
χ= = aχ+ + bχ− (288) 1 0 i 0 0 −1
b (298)
with These are the famous Pauli spin matrices.
  The eigenspinors of Sz are (of course)
1
χ+ = (289)
0    
1 ~
χ+ = , eigenvalue + (299)
representing spin up, and 0 2
   
  0 ~
0 χ− = , eigenvalue − (300)
χ− = (290) 1 2
1
for spin down. Meanwhile, the spin operators If you measure Sz on a particle in the general
become 2×2 matrices, which we can work out by state χ (equation 288), you could get +~/2, with
noting their effect on χ+ and χ− .Equation 285 probability |a|2 , or −~/2, with probability |b|2 .
says Since they are the only possibilities

|a|2 + |b|2 = 1 (301)

3 3
S2 χ+ = ~2 χ+ and S2 χ− = ~2 χ− (291) (i.e, the spinor must be normalized ). But
4 4
what if instead, you chose to measure Sx ? What
we can write S2 in matrix for as are the possible results. According to the gener-
alized statistical interpretation, we nee dto know
the eigenvalues and eigenspinors of Sx , the char-
 
3 2 1 0
S2 = ~ (292) acteristic equation is
4 0 1

16
  
 −λ ~/2 2 ~ 2

~ The Hamiltonian, in matrix form, is
~/2 −λ  = 0 ⇒ λ = 2 ⇒ λ ± 2
 

Not suprisingly, the possible values for Sx are  

γB0 ~ 1 0
the same as those for Sz . The normalized eigen- H = −γB0 Sz = − (309)
spinors of Sx are 2 0 −1

The eigenstates of H are the same as those of

1 Sz
√    (
(x) 2 , eigenvalue + ~ χ+ , with energy E+ = −(γB0 ~)/2
χ+ = (302)
1 2 χ− , with energy E− = +(γB0 ~)/2
√
2 since the Hamiltonian is time-independent,
1 the general solution to the time-dependent
√   
(x) 2 , eigenvalue − ~ Schrodinger equation is
χ− = (303)
−1 2
√ ∂χ
2 i~
= Hχ (310)
∂t
As the eigenvectors of a hermitian matrix, can be expressed in terms of the stationary
they span the space; the generic spinor χ (equa- states
tion 288) can be expressed as a linear combina-
tion of them
aeiγB0 t/2
 
−iE+ t/~ −iE− t/~
χ(t) = aχ+ e +bχ− e =
    be−iγB0 t/2
a+b (x) a−b (x) (311)
χ= √ χ+ + √ χ− (304)
2 2 The constants a and b are determined by the
initial conditions
If you measure Sx , the probability of getting
+~/2 is (1/2)|a + b|2, and the probability of get-  
a
ting −~/2 is (1/2)|a − b|2 . χ(0) = (312)
b
Electron in a Magnetic Field
With no essential loss of generality I will write
A spinning charged particle constitutes a
a = cos(α/2) and b = sin(α/2), Thus
magnetic dipole. Its magnetic dipole mo-
ment,µ, is proportional to its spin angular mo-
cos(α/2)eiγB0 t/2
 
mentum, S: χ(t) = (313)
sin(α/2)e−iγB0 t/2
µ = γS (305) To get a feel for whats happening here, let’s
calculate the expectation value of S, as a func-
the proportionality constant, γ, is called the tion of time
gyromagnetic ratio. When a magnetic dipole
is placed in a magnetic field B, it experiences a
tourqe, µ x B which tends to line it up parallel to hSx i = χ(t)† Sx χ(t) = (cos(α/2)e−iγB0 t/2 sin(α/2)eiγB0 t/2 )
the field. The energy associated with this tourqe
cos(α/2)eiγB0 t/2
  
~ 0 1
is ×
2 1 0 sin(α/2)e−iγB0 t/2
~
H = −µ · B (306) = sin α cos(γB0 t) (314)
2
so the Hamiltonian of a spinning charged par-
similarly,
ticle, at rest in a magnetic field B is

H = −γB · S (307) ~
hSy i = χ(t)† Sy χ(t) = − sin α sin(γB0 t)
2
Larmor preccesion: Imagine a particle of (315)
spin 1/2 at rest in a uniform magnetic field, and
which points in the z direction
~
B = B0 k̂ (308) hSz i = χ(t)† Sy χ(t) = cos α (316)
2

17
 
Evidently hSi is tilted at a constant angle α |11i =↑↑

to the z axis, and precesses about the field at the |10i = √12 (↑↓ + ↓↑) s = 1 (triplet)
Larmor frequency 
|1 − 1i =↓↓


This is called the triplet combination, for the

ω = γB0
Just as it would classically
Addition of Angular Momenta
Suppose now that we have twofor e spin-1/2
particles–for example, the electron and the pro-
ton in the ground state of hydrogen. Each can
have spin up or spin down so there are four pos-
sibilities in all
↑↑, ↑↓, ↓↑, ↓↓
where the first row refers to the electron and
the second row to the proton. Question: What
is the total angular momentum of the atom? Let
(317)
obvious reason. Meanwhile, the orthogonal state
with m = 0 carries s = 0;
 
1
|00i = √ (↑↓ − ↓↑) s = 0 (singlet) (324)
2
(if you apply the raisong or lowering operator
to this state, you’ll get zero
I need to prove that the triplet states are
(318) eigenvectors of S 2 with eigenvalue 2~2 , and the
singlet is an eigenvector of S 2 with eigenvalue 0.
Now,

(1) (2)
S 2 = (S(1) + S(2) ) · (S(1) + S(2) (325)
S≡S +S (319)
= (S (1) )2 + (S (2) )2 + 2S(1) · S(2) (326)
Each of these four composite states is an
eigenstate of Sz –the z-components simply add Using equations 294 and 297, we have

Sz χ1 χ2 = (Sz(1) +Sz(2) )χ1 χ2 = (Sz(1) chi1 )χ2 +χ1 (Sz(2) χ2 ) S(1) + S(2) (↑↓) = (Sx(1) ↑)(Sx(2) ↓)
= (~m1 χ1 )χ2 +χ1 (~m2 χ2 ) = ~(m1 +m2 )χ1 χ2 + (Sy(1) ↑)(Sy(2) ↓) + (Sz(1) ↑)(Sz(2) ↓)
(320)      
~ ~ i~ −i~
= ↓ ↑ + ↓ ↑)
So m (the quantum number of the composite 2 2 2 2
  
system) is just m1 + m2 ~ −~
+ ↑) ↓)
2 2
~2
↑↑: m = 1; = (2 ↓↑ − ↑↓) (327)
4
↑↓: m = 0;
similarly
↓↑: m = 0;
↓↓: m = −1
~2
S(1) + S(2) (↓↑) = (2 ↑↓ − ↓↑) (328)
At first galnce, this doesn’t look right: m is 4
supposed to advance in integer steps , from −s
to +s, so it appears that s = 1– but there is an It follows that
extra state with m = 0. One way to untangle
this problem is to apply the lowering operator, ~2 1 ~2
(1) (2)
S− = S− + S− to the state ↑↑, using equation S(1) ·S(2) |10i = √ (2 ↓↑ − ↑↓ +2 ↑↓ − ↓↑) = |10i
4 2 4
296 (329)
and
(1) (2)
S− (↑↑) = (S− ↑) ↑ + ↑ (S− ↑) (321)
= (~ ↓) ↑ + ↑ (~ ↓) (322) ~2 1 3~2
S(1) ·S(2) |00i = √ (2 ↓↑ − ↑↓ −2 ↑↓ + ↓↑) = − |00i
= ~(↓↑ + ↑↓) (323) 4 2 4
(330)
Evidently the three states with s = 1 are (in Returning to equation 325-326 ( and using
the notation |s mi): equation 291), we conclude that

18
 The statistical interpretation carries over in
the obvious way
3~2 3~2 2
 
~
S 2 |10i = + +2 |10i = 2~2 |10i
4 4 4
(331) |Ψ(r1 , r2 , t)|2 d3 r1 d3 r2 (338)
is the probability of finding particle 1 in the
volume d3 r1 and particle 2 in the volume d3 r2 ;
3~2 3~2 2
 
3~
S 2 |00i = + −2 |00i = 0 (332) evidently Ψ must be normalized in such a way
4 4 4 that
What we have just done (combining spin 1/2 Z
with spin 1/2 to get spin 1 and spin 0) is the sim-
|Ψ(r1 , r2 , t)|2 d3 r1 d3 r2 = 1 (339)
plest example of a larger problem: if you com-
bine spin s1 with spin s2 , what total spins s can
you get? The answer is that you get every spin For time-independent potentials, we obtain a
from (s+ s2 ) down to (s1 − s2 ) or (s2 − s1 ), if complete set of solutions by seperation of vari-
s2 > s1 in integer steps ables

s = (s1 +s2 ), (s1 +s2 −1), (s1 +s2 −2), ..., |s1 −s2 | Ψ(r1 , r2 , t) = ψ(r1 , r2 )e−iEt /~ (340)
(333)
where the spatial wave function (ψ) satisfies
The combined state |smi with total spin s and the time-independent Schrodinger equation
z-component m will be some linear combination
of the composite states |s1 m1 i|s2 m2 i
~2 2 ~2 2
− ∇1 ψ − ∇ ψ + V ψ = Eψ (341)
X 2m1 2m2 2
s1 s2 s
|smi = Cm 1 m2 m
|s1 m1 i|s2 m2 i (334)
m1 +m2 =m and E is the total energy of the system
Bosons and Fermions
(because the z components add, the only com- Suposse particle 1 is in the (one-particle) state
posite states that contribute are those for which ψa (r), and particle 2 is in the state ψb (r). (ignor-
m1 + m2 = m). The triplet combination and ing spin at the moment). In that case ψ(r1 , r2 )
the singlet are special cases of this general form, is a simple product
with s1 = s2 = 1/2 I used the notation ↑= | 12 21 i,
↓= | 12 (− 12 )i. The constants Cm
s1 s2 s
1 m2 m
are called
ψ(r1 , r2 ) = ψa (r)ψb (r) (342)
Clebsch-Gordan coefficients.
Quantum mechanics neatly accomidates the
Chapter 5: Identical Particles existance of particles that are indistinguishable
in principle: We simply construct a wave func-
Two-Particle System tion that is non committal as to which particle
The state of a two-particle system is a func- is in which state. there are actually two ways to
tion of the coordinates of particle one (r1 ), the do it
coordinates of particle two (r2 ), and the time

Ψ(r1 , r2 , t) (335) ψ± (r1 , r2 ) = A[ψa (r)ψb (r) ± ψb (r)ψa (r)] (343)

Its time evolution is deterimed (as always) by Thus the theory admits to kinds of identical
the Schrodinger equation: particles: bosons, for which we use the plus
sign , and fermions, for which we use the minus
∂Ψ sign. Photons and mesons are bosons; protons
i~ = HΨ (336)
∂t and electrons are fermions. It so happebns that
(
Where H is the Hamiltonian for the whole all particles with integer spin are bosons, and
system all particles with half-integer spin are fermions
It follows, in paticular, that two identical
~ 2
~ 2 fermions cannot occupy the same state. For if
H =− ∇2 − ∇2 + V (r1 , r2 , t) (337) ψa = ψb , then
2m1 1 2m2 2

19

ψ− (r1 , r2 ) = A[ψa (r)ψa (r) − ψa (r)ψa (r)] = 0
(344)
This is the famous Pauli exclusion princi-
ple. I assumed, for the sake of argument, that
one particle was in the state ψa and the other in
state ψb , but there is a more general way to for-
mulate the problem. Let us define the exchange
operator, P , which interchanges the two parti-
cles
P f (r1 , r2 ) = f (r1 , r2 ) (345)
1
ψ− (x1 , x2 ) = √ [ψa (x1 )ψb (x2 ) − ψb (x1 )ψa (x2 )]
2
(350)
Let’s calculate the expectation value of the
square of the separation distance between the
two particles
h(x1 − x2 )2 i = hx21 i + hx22 i − 2hx1 x2 i (351)
Case 1: Distinguishable particles. For
the wave function in equation 348

Clearly P 2 = 1, and it follows that the eigen- Z Z

values of P are ±1. Now, if the two particles are hx21 i = x21 |ψa (x1 )|2 dx1 |ψb (x2 )|2 dx2 = hx2 ia
identical, the Hamiltonian must treat them the
same: m1 = m2 and V (r1 , r2 ) = (r2 , r1 ) . It fol- (352)
lows that P and H are compatible observables, and

[P, H] = 0 (346)
Z Z
hx22 i = |ψa (x1 )|2 dx1 x22 |ψb (x2 )|2 dx2 = hx2 ib
and hence we can find a complete set of func- (353)
tions that are simultaneously eigenstates of both.
and
That is to say, we can find solutions to the
Schrodinger equation that are either symmetric
(eigenvalue + 1) or antisymmetric (eigenvalue -
Z Z
1) under exchange hx1 x2 i = x1 |ψa (x1 )|2 dx1 x2 |ψb (x2 )|2 dx2 = hxia hxib
(354)
ψ(r1 , r2 ) = ±ψ(r1 , r2 ) (347) in this case then

The symmetrization requirement states

that, for identical particles, the wave function is h(x1 − x2 )2 id = hx2 ia + hx2 ib − 2hxia hxib (355)
not merely alllowed, but required to satisfy equa-
tion 347. This is the general statement, of which Case 2: Identical particles. For the wave
equation 343 is a special case. function in equations 349 and 350
Exchange Forces
To give some sense of what the symmetriza- 1
Z Z
tion requirement does, we can suppose that one hx21 i =( x21 |ψa (x1 )|2 dx1 |ψb (x2 )|2 dx2
2
particle is in state ψa (x), and the other is in Z Z
state ψb (x), and these two states are orthogonal + x21 |ψb (x1 )|2 dx1 |ψa (x2 )|2 dx2
and normalized. If the two particles are distin- Z Z
guishable, and number 1 is in state ψa , then the ± x21 ψa (x1 )∗ ψb (x1 )dx1 ψb (x2 )∗ ψa (x2 )dx2
combined wave function is Z Z
± x21 ψb (x1 )∗ ψa (x1 )dx1 ψa (x2 )∗ ψb (x2 )dx2 )
ψ(x1 , x2 ) = ψa (x1 )ψb (x2 ) (348)
1 2  1
hx ia + hx2 ib ± 0 ± 0 = hx2 ia + hx2 ib


if they are identical bosons, the composite =
2 2
wave function is
(356)
1
ψ+ (x1 , x2 ) = √ [ψa (x1 )ψb (x2 ) + ψb (x1 )ψa (x2 )] similarly
2
(349)
1
hx22 i = hx2 ib + hx2 ia


and if they are identical fermions (357)
2

20
 (naturally, hx22 i = hx21 i, since you can’t tell (naturally, hx22 i = hx21 i, since you can’t tell
them apart.) But them apart.) But

1
Z Z
1
Z Z
2
hx1 x2 i = ( x1 |ψa (x1 )| dx1 x2 |ψb (x2 )|2 dx2 2
hx1 x2 i = ( x1 |ψa (x1 )| dx1 x2 |ψb (x2 )|2 dx2
2 2
Z Z Z Z
+ x1 |ψb (x1 )|2 dx1 x2 |ψa (x2 )|2 dx2 + x1 |ψb (x1 )|2 dx1 x2 |ψa (x2 )|2 dx2
Z Z Z Z
∗
± x1 ψa (x1 ) ψb (x1 )dx1 x2 ψb (x2 )∗ ψa (x2 )dx2 ∗
± x1 ψa (x1 ) ψb (x1 )dx1 x2 ψb (x2 )∗ ψa (x2 )dx2
Z Z Z Z
± x1 ψb (x1 )∗ ψa (x1 )dx1 x2 ψa (x2 )∗ ψb (x2 )dx2 ) ± x1 ψb (x1 )∗ ψa (x1 )dx1 x2 ψa (x2 )∗ ψb (x2 )dx2 )
1 1
= (hxia hxib + hxib hxia ± hxiab hxiba ± hxiba hxiab ) = (hxia hxib + hxib hxia ± hxiab hxiba ± hxiba hxiab )
2 2
= hxia hxib ± |hxiab |2 (358) = hxia hxib ± |hxiab |2 (364)

where where
Z
Z hxiab ≡ xψa (x)∗ ψb (x)dx (365)
2
psia (x1 )| dx1 x22 |ψb (x2 )|2 dx2 2
= hx ib (359)
Evidently
and

h(x1 −x2 )2 i± = hx2 ia +hx2 ib −2hxia hxib ∓2|hxiab |2

(366)
Z Z
hx1 x2 i = x1 |ψa (x1 )|2 dx1 x2 |ψb (x2 )|2 dx2 = hxia hxib
comparing equations 355 and 360, we see that
(360) the difference resides in the final term
in this case then

h(∆x)2 i± = h(∆x)2 id ∓ 2|hxiab |2 (367)

2 2 2
h(x1 − x2 ) id = hx ia + hx ib − 2hxia hxib (361)
Atoms
Case 2: Identical particles. For the wave A neutral atom, of atomic number Z, consists
function in equations 349 and 350 of a heavy nucleus, with electric charge Ze, sor-
rounded by Z electrons (mass m and charge −e).
The Hamiltonian for this system is
1
Z Z
hx21 i 2 2
= ( x1 |ψa (x1 )| dx1 |ψb (x2 )|2 dx2
2
Z 
Z Z
~2 2 Ze2
  
+ x21 |ψb (x1 )|2 dx1 |ψa (x2 )|2 dx2
X 1
H= − ∇ −
Z Z j=1
2m j 4πǫ0 rj
2 ∗
± x1 ψa (x1 ) ψb (x1 )dx1 ψb (x2 )∗ ψa (x2 )dx2 Z
e2
 X
1 1
+ (368)
2 4πǫ0 |rj − rk |
Z Z
± x21 ψb (x1 )∗ ψa (x1 )dx1 ψa (x2 )∗ ψb (x2 )dx2 ) j6=k

1 2  1 The term in the curly brackets represents the

hx ia + hx2 ib ± 0 ± 0 = hx2 ia + hx2 ib


=
2 2 kinetic plus potential of the jth electron, in the
electric field of the nucleus; the secon sum (which
(362) runs over all values of j and k except j = k) is the
potential associated with the multual repulsion
similarly of the electrons. (the factor of 1/2 in front cor-
responds that the summation counts each pair
1 twice. The problem is to solve Schrodinger’s
hx22 i = hx2 ib + hx2 ia


(363) equation
2

21
 The Free Electron Gas
Hψ = Eψ (369) Suppose the object in question is a rectangu-
lar solid, with dimensions lx , ly , lz , and imagine
for the wave function ψ(r1 , r2 , ...rZ ). Because
that an electron inside experiences no force at
electrons are identical fermions, however, not all
all, except at the impenetrable walls
solutions are acceptable;only those for which the (
complete state (position and spin) 0, if0 < x < lx , 0 < y < ly , and0 < z < lz
V (x, y, z) =
∞, otherwise
ψ(r1 , r2 , ...rZ )χ(s1 , s2 , ...sZ ) (370) The Schrodinger equation,

is antisymmetric with respect to interchange ~2 2

of any two electrons. In particular, no two elec- − ∇ ψ = Eψ (376)
2m
trons can occupy the same state
separates, in cartesian coordinates: ψ(x, y, z) =
Helium
X(x)Y (y)Z(z), with
After hydrogen, the simplest atom is helium
(Z = 2). The Hamiltonian,
~2 dX 2
− = Ex X (377)
2m dx2
~2 2 1 2e2 2 2
 
~ dY
H= − ∇ − − = Ey Y (378)
2m 1 4πǫ0 r1 2m dy 2
~2 2 1 2e2 ~2 dZ 2
 
+ − ∇ − − = Ez Z (379)
2m 2 4πǫ0 r2 2m dz 2
1 e2 and E = Ex + Ey + Ez . Letting
+ (371)
4πǫ0 |r1 − r2 |
√ p √
consists of two hydrogenic Hamiltonians (with 2mEx 2mEy 2mEz
nuclear charge 2e), one for each electron 1 and 2, kx ≡ , ky ≡ , kz ≡
~ ~ ~
together with the final term describing the repul- (380)
sion of the two electrons. It is this last term that we obtain the general solutions
causes all the the trouble. If we simply ignore it,
the Schrodinger equation sperates, and the so-
lutions can be written as products of hydrogen X(x) = Ax sin(kx x) + Bx cos(kx x) (381)
wave functions Y (y) = Ay sin(ky y) + Bx cos(ky y) (382)
Z(z) = Az sin(kz z) + Bx cos(kz z) (383)
ψ(r1 , r2 ) = ψnlm (r1 )ψn′ l′ m′ (r2 ) (372) (384)

The boundery conditions require X(0) =

only with half the Bohr radius and four times
Y (0) = Z(0) = 0, so Bx = By = B + z = 0,
the Bohr energies. The total energy would be
and X(lx ) = Y (ly ) = Z(lz ) = 0, so that
E = 4(En + En′ ) (373)
kx lx = nx π, ky ly = ny π, kz lz = nz π (385)
where En = −13.6/n2 eV. In particular, the
ground state would be
where each n is a positive integer

8 −2(r1 +r2 )/a

ψ0 (r1 , r2 ) = ψ100 (r1 )ψ100 (r2 ) = e nx = 1, 2, 3, .., ny = 1, 2, 3, .., nz = 1, 2, 3, ..
πa3
(374) (386)
and its energy would be The normalized wave functions are

s      
E0 = 8(−13.6 eV) = −109 eV (375) 8 nx π ny π nz π
ψnx ny nz = sin x sin y sin z
lx ly lz lx ly lz
Solids (387)

22
 Z
hxiab ≡ xψa (x)∗ ψb (x)dx (388)
~2 2 1 2e2
 
H= − ∇ −
2m 1 4πǫ0 r1
Evidently
~2 2 1 2e2
 
+ − ∇ −
2m 2 4πǫ0 r2
h(x1 −x2 )2 i± = hx2 ia +hx2 ib −2hxia hxib ∓2|hxiab |2 1 e2
+ (394)
(389) 4πǫ0 |r1 − r2 |
comparing equations 355 and 360, we see that
the difference resides in the final term consists of two hydrogenic Hamiltonians (with
nuclear charge 2e), one for each electron 1 and 2,
together with the final term describing the repul-
h(∆x)2 i± = h(∆x)2 id ∓ 2|hxiab |2 (390) sion of the two electrons. It is this last term that
causes all the the trouble. If we simply ignore it,
the Schrodinger equation sperates, and the so-
Atoms
lutions can be written as products of hydrogen
A neutral atom, of atomic number Z, consists wave functions
of a heavy nucleus, with electric charge Ze, sor-
rounded by Z electrons (mass m and charge −e).
The Hamiltonian for this system is
ψ(r1 , r2 ) = ψnlm (r1 )ψn′ l′ m′ (r2 ) (395)

Z 
only with half the Bohr radius and four times
~2 2 Ze2
  
X 1 the Bohr energies. The total energy would be
H= − ∇ −
j=1
2m j 4πǫ0 rj
Z E = 4(En + En′ ) (396)
e2
 X
1 1
+ (391)
2 4πǫ0 |rj − rk | where En = −13.6/n2 eV. In particular, the
j6=k
ground state would be
The term in the curly brackets represents the
kinetic plus potential of the jth electron, in the
electric field of the nucleus; the secon sum (which 8 −2(r1 +r2 )/a
ψ0 (r1 , r2 ) = ψ100 (r1 )ψ100 (r2 ) = e
runs over all values of j and k except j = k) is the πa3
(397)
potential associated with the multual repulsion
of the electrons. (the factor of 1/2 in front cor- and its energy would be
responds that the summation counts each pair
twice. The problem is to solve Schrodinger’s
equation E0 = 8(−13.6 eV) = −109 eV (398)

Solids
Hψ = Eψ (392)
The Free Electron Gas
for the wave function ψ(r1 , r2 , ...rZ ). Because Suppose the object in question is a rectangu-
electrons are identical fermions, however, not all lar solid, with dimensions lx , ly , lz , and imagine
solutions are acceptable;only those for which the that an electron inside experiences no force at
complete state (position and spin) all, except at the impenetrable walls
(
0, if 0 < x < lx , 0 < y < ly , and 0 < z < lz
V (x, y, z) =
ψ(r1 , r2 , ...rZ )χ(s1 , s2 , ...sZ ) (393) ∞, otherwise
The Schrodinger equation,
is antisymmetric with respect to interchange
of any two electrons. In particular, no two elec-
trons can occupy the same state ~2 2
− ∇ ψ = Eψ (399)
2m
Helium
After hydrogen, the simplest atom is helium separates, in cartesian coordinates: ψ(x, y, z) =
(Z = 2). The Hamiltonian, X(x)Y (y)Z(z), with

23
 π3 π3
2
~ dX 2 = (411)
− = Ex X (400) lx ly lz V
2m dx2
2 2 of ’k-space,’ where V = lx ly lz is the volume of
~ dY
− = Ey Y (401) the object itself. Suppose the sample contains N
2m dy 2
atoms , and each contribute q electrons. They
~2 dZ 2 will fill up one octant of a sphere in k-space,
− = Ez Z (402)
2m dz 2 whose radius ,kF is determined by the fact that
each pair of electrons require a volume π 3 /V
and E = Ex + Ey + Ez . Letting
π3
   
1 4 3 Nq
√ √ πk = (412)
p
2mEy 8 3 F 2 V
2mEx 2mEz
kx ≡ , ky ≡ , kz ≡
~ ~ ~ thus
(403)
we obtain the general solutions kF = (3ρπ 2 )1/3 (413)
where
X(x) = Ax sin(kx x) + Bx cos(kx x) (404)
Nq
Y (y) = Ay sin(ky y) + Bx cos(ky y) (405) ρ≡ (414)
V
Z(z) = Az sin(kz z) + Bx cos(kz z) (406) is the free electron density ( the number of
free electrons per unti volume)
The boundery conditions require X(0) =
The boundery seperating occupied and unoc-
Y (0) = Z(0) = 0, so Bx = By = B + z = 0,
cupied states, in k-space, is called the Fermi
and X(lx ) = Y (ly ) = Z(lz ) = 0, so that
surface. The corresponding energy is called the
Fermi energy, EF ; for a free electron gas,
kx lx = nx π, ky ly = ny π, kz lz = nz π (407)
~2
EF = (3ρπ 2 )2/3 (415)
2m
where each n is a positive integer
The total energy of the electron gas can be
calculated as follows; Each of these states carries
nx = 1, 2, 3, .., ny = 1, 2, 3, .., nz = 1, 2, 3, .. an energy ~2 k 2 /2m, so the energy of the shell is
(408)
~2 k 2 V 2
The normalized wave functions are dE = k dk (416)
2m π 2
s and hence the total energy is
     
8 nx π ny π nz π
ψnx ny nz = sin x sin y sin z
lx ly lz lx ly lz Z kF
~2 V 4 ~2 kF5 V
(409) Etot = k dk =
2mπ 2 0 10π 2 m
and the allowed energies are
~2 (3π 2 N q)5/3 −2/3
= V (417)
10π 2 m
~2 π 2 n2x n2x n2x ~2 k 2
 
Enx ny nz = + + = This quantum mechanical energy plays a role
2m lx2 lx2 lx2 2m rather analogous to the internal thermal energy
(410) (U ) of an ordinary gas. In particular, it exerts a
where k is the magnitude of the wave vector, pressure on the walls, for if the box expands by
k≡ (kx , ky , kz ). an amount dV , the total energy decreases
If you imagine a three-dimensional space,
with axes kx , ky , kz , and planes drawn in at kx =
(π/lx ), (2π/lx ), (3π/lx ), ..., at ky = (π/ly ), (2π/ly ), (3π/ly ), ..., 2 ~2 (3π 2 N q)5/3 −5/3
dEtot = − V dV
and at kz = (π/lz ), (2π/lz ), (3π/lz ), .... Each 3 10π 2 m
block in this grid, and hence each state, occu- 2 dV
= − Etot (418)
pies a volume 3 V

24
 and this shows up as work done on the out-
side (dW = P dV ) by the quantum pressure P .
Evidently

2 Etot 2 ~2 kF5 (3π 2 )2/3 ~2 5/3

P = = 2
= ρ
3 V 3 10π m 5m
(419)
this is sometimes called degeneracy pres-
sure strictly due to quantum effects.

25