134 Chap.

4 Quantum Mechanics in Three Dimensions

4.2.1 The Radial Wave Function

Our first task is to tidy up the notation. Let

J-2mE
K=--- [4.54)
n
(For bound states, E < 0, so K is real.) Dividing Equation 4.53 by E, we have
2 2
1 d u [
1
me 1 l(l+ 1)]
K2 dr 2 = 2
- 2JTEoh K (Kr) + (Kr) 2 u.

This suggests that we let

me 2
p =Kr, and Po =2 JTEo
n2 K ' [4.55)

so that
2
d u =
dp2
[i _ Po
p
+ l(l + l)J u.
p2
[4.56)

Next we examine the asymptotic form of the solutions. Asp --+ oo, the constant
term in the brackets dominates, so (approximately)

The general solution is

u(p) = Ae-P + BeP, [4.57]
but eP blows up (asp--+ oo), so B = 0. Evidently,

u(p)"' Ae-P [4.58]

for large p. On the other hand, asp --+ 0 the centrifugal term dominates 12 ; approxi-
mately, then,
d2 u l(l + 1)
dp2 p2 u.
The general solution (check it!) is

u(p) = Cp1+1 + Dp-1,

but p- 1 blows up (asp --+ 0), so D = 0. Thus

u(p) "'Cp1+1 [4.59]

12 This argument does not apply when l = 0 (although the conclusion, Equation 4.59, is in fact
valid for that case too). But never mind: All I am trying to do is provide some motivation for a change of
variables (Equation 4.60.)
 Sec. 4.2: The Hydrogen Atom 135

for small p.
The next step is to peel off the asymptotic behavior, introducing the new function
v(p):
[4.60]
in the hope that v (p) will tum out to be simpler than u (p). The first indications are
not auspicious:
du
-=/e-P [ (l+l-p)v+p-,
dv]
dp dp
and
2
d u
-=/e-p
dp 2
{[ -2l-2+p+ l(l +
p
l)] dv
dp
d v} .
v+2(l+I-p)-+p-
dp 2
2

In terms of v(p), then, the radial equation (Equation 4.56) reads

d2v dv
Pd 2 +2(l+l-p)-+[p0 -2(l+l)]v=0. [4.61]
p dp

Finally, we assume the solution, v(p ), can be expressed as a power series in p:

00

v(p) = Laipi. [4.62]

)=0

Our problem is to determine the coefficients (a 0 , a 1 , a2 , .•• ). Differentiating term by

term,
d 00 00

___!: = L jaipJ-I =LU+ l)a1+1p 1.

dp )=0 J=O

[In the second summation I have renamed the "dummy index": j -+ j + 1. If this
troubles you, write out the first few terms explicitly, and check it. You might say that
the sum should now begin at j = -1, but the factor U + 1) kills that term anyway,
so we might as well start at zero.] Differentiating again,

Inserting these into Equation 4.61, we have

00 00

L jU + l)a1+1P 1 + 2(/ + 1) LU+ I)a1+1P 1

J=O j=O

00 00

-2 Lja1 pi + [p0 - 2(1 + l)] L:a1 pi = 0.

136 Chap. 4 Quantum Mechanics in Three Dimensions

Equating the coefficients of like powers yields

j(j + l)a1+1 + 2(1 + l)(j + l)aJ+l - 2ja1 +[po - 2(1+l)]a1 =0,

or
2(} + l + 1) - Po }
aJ+I = (j + I)(j + 21+2) ai.
{ [4.63 J

This recursion formula determines the coefficients, and hence the function v (p J:
We start with a0 = A (this becomes an overall constant, to be fixed eventually b:
normalization), and Equation 4.63 gives us a 1 ; putting this back in, we obtain a 2 , and
so on. 13
Now let's see what the coefficients look like for large j (this corresponds to
large p, where the higher powers dominate). In this regime the recursion formula
says
2·
J 2
a1+1= j(j+l)a1= j+Iai,
so
2J
a1· .,
1·
[4.64]

Suppose for a moment that this were the exact result. Then

v(p) =A .L:-.J.21 pi = Ae
00

j=O
1
2
P,

and hence
[4.65]
which blows up at large p. The positive exponential is precisely the asymptotic behav-
ior we didn't want in Equation 4.57. (It's no accident that it reappears here; after all.
it does represent the asymptotic form of some solutions to the radial equation-they
just don't happen to be the ones we're interested in, because they aren't normalizable.)
There is only one way out of this dilemma: The series must terminate. There must
occur some maximal integer, }max• such that

[4.66]

(and beyond which all coefficients vanish automatically). Evidently (Equation 4.63)

2Urnax + l + 1) - Po = 0.

13
You might wonder why I didn't use the series method directly on u(p)-why factor out the
asymptotic behavior before applying this procedure? The reason for peeling off /+ 1 is largely aesthetic:
Without this, the sequence would begin with a Jong string of zeroes (the first nonzero coefficient being
a1+ 1 ); by factoring out p 1+ 1 we obtain a series that starts out with p 0 . The e-P factor is more critical-if
you don't pull that out, you get a three-term recursion formula involving a1+2, a1+ 1 , and ai (try it!), and
that is enormously more difficult to work with.
 Sec. 4.2: The Hydrogen Atom 137

Defining
n- jmax +l+ 1 [4.67]
(the so-called principal quantum number), we have

Po= 2n. [4.68]

But p 0 determines E (Equations 4.54 and 4.55):

[4.69]

so the allowed energies are

n=l,2,3, .... [4.70]

This is the famous Bohr formula-by any measure the most important result in
all of quantum mechanics. Bohr obtained it in 1913 by a serendipitous mixture
of inapplicable classical physics and premature quantum theory (the Schrodinger
equation did not come until 1924 ).
Combining Equations 4.55 and 4.68, we find that
2
K-
me ) 1 - 1 [4.71]
- ( 4rrE0ii 2 ;; - an'

where

4JrEoiz 2
a- 2
= 0.529 x 10- 10 m [4.72]
me

is the so-called Bohr radius. It follows (again, from Equation 4.55) that
r
p=-. [4.73]
an

Evidently the spatial wave functions for hydrogen are labeled by three quantum num-
bers (n, l, and m ):
Yt
1/!ntm (r, e, </J) = Rnt (r) (e, </J ), [4.74]

where (referring back to Equations 4.36 and 4.60)

[4.75]
138 Chap. 4 Quantum Mechanics in Three Dimensions

and v(p) is a polynomial of degree }max = n - l - 1 in p, whose coefficients are

determined (up to an overall normalization factor) by the recursion formula

2(j + l + 1 - n)
[4.76]
aJ+I = (j + l)(j + 21+2) ai.

The ground state (that is, the state of lowest energy) is the case n = 1; putting
in the accepted values for the physical constants, we get

Ei = -[.!!!__2 = -13.6 eV. [4.77]

21i 4JrEo

Evidently the binding energy of hydrogen (the amount of energy you would have to
impart to the electron in order to ionize the atom) is 13.6 eV. Equation 4.67 forces
l = 0, whence also m = 0 (see Equation 4.29), so

1/!wo(r, e, </J) = Rw(r)Yg(e, </J). [4.78]

The recursion formula truncates after the first term (Equation 4.76 with j = 0 yields
a1 = 0), so v(p) is a constant (ao) and

ao r/a .
R10(r) = -e- [4.79]
a
Normalizing it, in accordance with Equation 4.31,

la 12100 2 a
l oo
o
IR101 2r 2dr = - 0 -
a2
e- r/a r 2 dr = laol 2- = 1,
o 4

so a0 = 2/ ../Ci. Meanwhile, Yg = 1/ .J4Jr, so

[4.80]

If n = 2 the energy is
-13.6 eV
E2 = = -3.4 eV; [4.81]
4

this is the first excited state--or rather, states, since we can have either l = 0 (in
which case m = 0) or l = 1 (with m = -1, 0, or +1), so there are actually four
different states that share this energy. If l = 0, the recursion relation (Equation 4.76)
gives
a 1 = -a 0 (using j = 0), and a 2 = 0 (using j = 1),
 Sec. 4.2: The Hydrogen Atom 139

so v(p) = a0 (l - p), and hence

Rzo(r) = ao
2a
(1 - 2a e-r/2.a. [4.82]

If l = 1 the recursion formula terminates the series after a single term, so v(p) is a
constant, and we find
ao r/2.a
R21(r) = - 2 re- . [4.83]
4a
(In each case the constant a0 is to be determined by normalization-see Problem
4.11.)
For arbitrary n, the possible values of l (consistent with Equation 4.67) are

l = 0, 1, 2, ... , n - 1. [4.84]
For each l, there are (2/ + 1) possible values of m (Equation 4.29), so the total
degeneracy of the energy level En is
n-1
d(n) = L)2l + 1) = n 2 . [4.85]
1=0

The polynomial v(p) (defined by the recursion formula, Equation 4.76) is a function
well known to applied mathematicians; apart from normalization, it can be written as
v(p) = (2p), [4.86]
where
p
Lq-p(x) - (-1) p ( dx
= d )p Lq(x) [4.87]

is an associated Laguerre polynomial, and

Lq(x) - q (e-xxq) [4.88]

is the qth Laguerre polynomial. 14 (The first few Laguerre polynomials are listed in
Table 4.4; some associated Laguerre polynomials are given in Table 4.5. The first
few radial wave functions are listed in Table 4.6 and plotted in Figure 4.4.) The
normalized hydrogen wave functions are 15

1/!nlm = (n-/-1)! e-r/na(2r)1L21+1 (2r)Ym(e </J). [4.89]

na 2n[(n + /)!]3 na n-l-I na 1 '

14 As usual, there are rival normalization conventions in the literature; I have adopted the most nearly
standard one.
15
If you want to see how the normalization factor is calculated, study (for example), L. Schiff,
Quantum Mechanics, 2nd ed. (New York: McGraw-Hill, 1968), page 93.
140 Chap. 4 Quantum Mechanics in Three Dimensions

Table 4.4: The first few Laguerre polynomials, Lq(x).

Lo= 1
L1=-x+1
L2 = x 2 -4x +2
L3 = -x 3 +9x 2 -18x +6
L4 = x 4 - 16x 3 + 72x - 96x + 24
2

L 5 = -x 5 + 25x 4 - 200x 3 + 600x 2 - 600x + 120

L6 = x 6 - 36x 5 + 450x 4 - 2400x 3 + 5400x 2 - 4320x + 720

Table 4.5: Some associated Laguerre polynomials, L:_P(x).

Lg= 1 L6 = 2

L?=-x+l Li=-6x+18

Lg= x 2 -4x +2 = 12x 2 - 96x + 144

Lb= 1 L6 = 6

L: =-2x+4 Li= -24x +96

Li= 3x 2 - 18x + 18 = 60x 2 - 600x + 1200

They are not pretty, but don't complain-this is one of the very few realistic systems
that can be solved at all, in exact closed form. As we will prove later on, they are
mutually orthogonal:

[4.90]

*Problem 4.10 Work out the radial wave functions R30 , R 31 , and R32 , using the
recursion formula (Equation 4.76). Don't bother to normalize them.

*Problem 4.11

(a) Normalize R20 (Equation 4.82), and construct the function 1/1 200 .

(b) Normalize R21 (Equation 4.83), and construct 1/1211, 1/1210, and 1/12 1- 1.

**Problem 4.12

(a) Using Equation 4.88, work out the first four Laguerre polynomials.
(b) Using Equations 4.86, 4.87, and 4.88, find v(p) for the case n = 5, l = 2.
 Sec. 4.2: The Hydrogen Atom 141

Table 4.6: The first few radial wave functions for hydrogen, Rn 1(r).

Rio= 2a- 312 exp(-r/a)

1
R2o = -a-
..J2
3/2( 1 - - - exp (-r/2a)
2 a
Ir)
1 3/2 r
R21 = - a - - exp (-r /2a)
../24 a
2
2
R3o =--a- 312 ( I - -
2r- + -2(r)
- ) exp(-r/3a)
,./fl 3a 27 a
R31 = - 8- a - 312 ( I - -1 -r ) ( -r ) exp (-r/3a)
27,,/6 6 a a
2
4 3 2 r
R32 = - - a - I (
-) exp (-r/3a)
81.J30 a

1 312 ( ) - -3 -r + -I ( -r) - -
1 ( -r)
2 3) exp(-r/4a)
R4o=-a-
4 4a 8 a 192 a
2
JS _ 312 ( 1---+-
R41=--a 1 r 1 ( -r) ) -exp(-r/4a)
r
16,J3 4 a 80 a a
1
R42=--a-
64JS
312 ( 1 -1- -
12 a
r) (-ar) 2 exp(-r/4a)
R43 = 1 a- 3/2 (r)3
- exp (-r /4a)
768v'35 a

I
I
I
0.8 I
:10
I
0.7 I
I
I
I
I
0.6 I
I
120 :
0.5 • II
I
""' I
I
0.4 I
I
I
I
I
I

0.2

0.1
32
30 3
0 rla
12 14 16 18

Figure 4.4: Graphs of the first few hydrogen radial wave functions, Rn 1(r).
142 Chap. 4 Quantum Mechanics in Three Dimensions

( c) Again, find v (p) for the case n = 5, l = 2, but this time get it from the recursion
formula (Equation 4.76).

*Problem 4.13

(a) Find (r) and (r 2 ) for an electron in the ground state of hydrogen. Express your
answers in terms of the Bohr radius a.
(b) Find (x) and (x 2 ) for an electron in the ground state of hydrogen. Hint: This
requires no new integration-note that r 2 = x 2 + y2 + z 2 , and exploit the
symmetry of the ground state.
(c) Find (x 2 ) in the state n = 2, l = 1, m = 1. Hint: This state is not symmetrical
in x, y, z. Use x = r sine cos</J.

Problem 4.14 What is the probability that an electron in the ground state of hy-
drogen will be found inside the nucleus?

(a) First calculate the exact answer, assuming that the wave function (Equation
4.80) is correct all the way down tor = 0. Let b be the radius of the nucleus.
(b) Expand your result as a power series in the small number 8 = 2b/a, and show
that the lowest-order term is the cubic: P (4/3)(b/a) 3 . This should be a
suitable approximation, provided that b << a (which it is).
(c) Alternatively, we might assume that 1/1 (r) is essentially constant over the (tiny)
volume of the nucleus, so that P (4/3)rrb 3 11/f (O) 12 . Check that you get the
same answer this way.
(d) Use b 10- 15 m and a 0.5 x 10- 10 m to get a numerical estimate for
P. Roughly speaking, this represents the "fraction of its time that the electron
spends inside the nucleus".

Problem 4.15

(a) Use the recursion formula (Equation 4.76) to confirm that when l =n - 1 the
radial wave function takes the form

Rn(n-1) = N nr n-l e -r/na '

and determine the normalization constant Nn by direct integration.
(b) Calculate (r) and (r 2 ) for states of the form 1/!n(n-l)m·

(c) Show that ar = (r) / ./2n + 1 for such states. Note that the fractional spread in
r decreases with increasing n (in this sense the system "begins to look classical"
for large n ). Sketch the radial wave functions for several values of n to illustrate
this point.
 Sec. 4.2: The Hydrogen Atom 143

4.2.2 The Spectrum of Hydrogen

In principle, if you put a hydrogen atom into some stationary state \IInlm, it should
stay there forever. However, if you tickle it slightly (by collision with another atom,
say, or by shining light on it), then the atom may undergo a transition to some other
stationary state--either by absorbing energy and moving up to a higher-energy state,
or by giving off energy (typically in the form of electromagnetic radiation) and moving
down. 16 In practice such perturbations are always present; transitions (or, as they are
sometimes called, "quantum jumps") are constantly occurring, and the result is that
a container of hydrogen gives off light (photons), whose energy corresponds to the
difference in energy between the initial and final states:

1
Ey = E; - E1 = -13.6eV ( - [4.91]
n;2 n1

Now, according to the Planck formula, 17 the energy of a photon is proportional

to its frequency:
Ey = hv. [4.92]
Meanwhile, the wavelength is given by A= c/v, so

_!_ = R (_!__ - __!_), [4.93]

A n}
where
2
R _ _ m_
4ncii 3
(__!__)
4rrEo
= 1.097 x 107 m- 1 . [4.94]

R is known as the Rydberg constant, and Equation 4. 93 is the Rydberg formula for
the spectrum of hydrogen. It was discovered empirically in the nineteenth century,
and the greatest triumph of Bohr's theory was its ability to account for this result-
and to calculate R in terms of the fundamental constants of nature. Transitions to the
ground state (n f = 1) lie in the ultraviolet; they are known to spectroscopists as the
Lyman series. Transitions to the first excited state (n f = 2) fall in the visible region;
they constitute the Balmer series. Transitions to n1 = 3 (the Paschen series) are in
the infrared, and so on (see Figure 4.5). (At room temperature, most hydrogen atoms
are in the ground state; to obtain the emission spectrum, you must first pump them
up into the various excited states; typically this is done by passing an electric spark
through the gas.)

16
By its nature, this involves a time-dependent interaction, and the details will have to wait for
Chapter 9; for our present purposes the actual mechanism involved is immaterial.
17
The photon is a quantum of electromagnetic radiation; it's a relativistic object if there ever was
one, and therefore outside the scope of nonrelativistic quantum mechanics. It will be useful in a few places
to speak of photons and to invoke the Planck formula for their energy, but please bear in mind that this is
external to the theory we are developing.