One electron Atoms

The Hydrogen Atom

In a hydrogenic atom an electron of charge −e and mass m is held together with a nucleus of
charge +Ze and mass M by Coulomb interaction. By using CM frame, we can reduce the problem
to the dynamics of a single particle whose mass µ = mM /(m + M ) is the reduced mass and
whose coordinate r is the relative coordinate of the two particle.
mM m
Now, µ = =
m+M 1 + m/M
m 1
since, ≃ ≪ 1; ∴ µ≈m
M 2000
From Coulomb’s law, the potential energy is given by

Ze2 1
V (r) = − (1)
4πε0 r

If ψ(⃗r) describe the wavefunction for the one-body system, then the time-independent Schrödinger
equation in the position representation is given by
[ ]
ℏ2 2
− ∇ + V (r) ψ(⃗r) = E ψ(⃗r) (2)
2m

Here the potential (equation 1) is a function only of the distance from the origin. In that case
to exploit the symmetry of the problem we use ‘spherical coordinates’ (r, θ, ϕ). In spherical
coordinates, the Laplacian is:
( ) ( )
2 1 ∂ 2 ∂ 1 ∂ ∂ 1 ∂2
∇≡ 2 r + 2 sin θ + 2
r ∂r ∂r r sin θ ∂θ ∂θ r2 sin θ ∂ϕ2
Therefore, the time-independent Schrödinger equation is spherical coordinates reads
[ ( ) ( ) ]
ℏ2 1 ∂ 2 ∂ψ 1 ∂ ∂ψ 1 ∂2ψ
− r + 2 sin θ + 2 + V (r) ψ = E ψ (3)
2m r2 ∂r ∂r r sin θ ∂θ ∂θ r2 sin θ ∂ϕ2

Separation of variables
We solve the equation using the method of separation of variables, where ψ is taken as the
product of functions of r, θ and ϕ, that is

ψ(r, θ, ϕ) = R(r)Y (θ, ϕ) (4)

so equation (3) can be written as

[ ( ) ( ) ]
ℏ2 Y d 2 dR R ∂ ∂Y R ∂2Y
− r + sin θ + 2 + V RY = E RY
2m r2 dr dr r2 sin θ ∂θ ∂θ r2 sin θ ∂ϕ2
2
dividing by RY and multiplying by − 2mr
ℏ2

[ ( ) ] [ ( ) ]
1 d dR 2mr2 1 1 ∂ ∂Y 1 ∂2Y
r2 − (V − E) + sin θ + =0
R dr dr ℏ2 Y sin θ ∂θ ∂θ 2
sin θ ∂ϕ2
The terms in the first bracket depends only on r, whereas the remainder is a function of (θ, ϕ);
therefore each must be equal to constant. The ‘separation constant’ in this case is taken as

1
 One electron Atoms

l(l + 1). The motive behind such choice is to exploit the fact that the equations reduces to some
of the well known differential equations whose solutions we already know.
[ ( ) ]
1 d dR 2mr2
r2 − (V − E) = l(l + 1); (5)
R dr dr ℏ2
[ ( ) ]
1 1 ∂ ∂Y 1 ∂2Y
sin θ + 2 = −l(l + 1) (6)
Y sin θ ∂θ ∂θ sin θ ∂ϕ2

The angular equation

2
Equation (6) is the angular equation; we multiply both side by Y sin θ; it becomes
( )
∂ ∂Y 2 ∂2Y
sin θ sin θ + l(l + 1) sin θ Y + =0 (7)
∂θ ∂θ ∂ϕ2

Here, Y is a function of and ϕ; we use the separation of variables method again as

Y (θ, ϕ) = Θ(θ) Φ(ϕ) (8)

plugging this in and dividing by Θ Φ

{ [ ( )] }
1 d dΘ 2 1 d2 Φ
sin θ sin θ + l(l + 1) sin θ + =0
Θ dθ dθ Φ dϕ2

The first term is a function of only θ and the second term is a function of ϕ only; therefore they
must be equal to a constant. The separation constant is taken to be m2l .
[ ( )]
1 d dΘ 2
sin θ sin θ + l(l + 1) sin θ = m2l ; (9)
Θ dθ dθ

1 d2 Φ
= −m2l (10)
Φ dϕ2
The ϕ equation (10) yields
Φ(ϕ) = eiml ϕ (11)
The ϕ equation, being a second order differential equation should have two solutions, namely,
eiml ϕ and e−iml ϕ . We cover the later by allowing ml to have negative values. There could a
constant at the front, but we might absorbed that in Θ. Now we have to satisfy the azimuthal
symmetry:
Φ(ϕ + 2π) = Φ(ϕ) (12)
In other words, eiml (ϕ+2π) = eiml ϕ → e2πiml = 1. Therefore, it follows that ml can only take
integer values.
ml = 0, ±1, ±2, ±3, · · · (13)
The θ equation
( ) [ ]
d dΘ 2
sin θ sin θ + l(l + 1) sin θ − m2l Θ = 0 (14)
dθ dθ

The solution of the above equation is given by ‘associated Legendre polynomial’

Θ(θ) = APlml (cos θ) (15)

Here l is a nonnegative integer (the solution with negative l blows up at θ = 0 and/ or θ = π)

and |ml | < l, therefore,

2
 One electron Atoms

l = 0, 1, 2, 3, · · · ; ml = −l, − (l − 1), · · · , − 1, 0, 1, · · · , (l − 1), l (16)

Therefore, the total solution of the angular equation is of the form:

Ymml (θ, ϕ) = APlml (cos θ) eiml ϕ (17)

Now, the volume element in spherical coordinate is

d3 r = r2 sin θ dr dθ dϕ

So, the normalization condition demands

∫
|ψ|2 r2 sin θ dr dθ dϕ = 1

∫ ∫
i.e. |R| r dr 2 2
|Y |2 sin θ dθ dϕ = 1

It is convenient to normalize R and Y individually;

∫ ∞
|R|2 r2 dr = 1
0
∫ π∫ 2π
|Y |2 sin θ dθ dϕ = 1
0 0

The normalized angular wavefunctions are called spherical harmonics:

√
(2l + 1) (l − |ml |)! iml ϕ ml
Ylml (θ, ϕ) =ε e Pl (cos θ) (18)
4π (l + |ml |)!

where, ε = (−1)m for ml ≥ 0 and ε = 1 for ml ≤ 0. The spherical harmonics are also orthogonal.
In addition, as Plm = Pl−m and e−imϕ = (−1)m eimϕ , therefore, Yl−m = (−1)m Ylm .
Note that the angular part of the wavefunction, Ylml (θ, ϕ), is same for all spherically symmetric
potentials; the actual shape of the potential, V (r), affects only the radial part of the wavefunc-
tion, R(r).

The Radial Equation

( )
d dR 2mr2
r2 − (V − E) R = l(l + 1) R (19)
dr dr ℏ2
The equation simplifies with the change of variables:

u(r) ≡ rR(r) (20)

( )
dR u 1 du 1 du
=− 2 + = 2 r −u
dr r r dr r dr
( ) ( )
d 2 dR d du d2 u
r = r −u =r 2
dr dr dr dr dr
Inserting these into the radial equation we get

3
 One electron Atoms

Figure 1: The effective potential (Veff ) as a function of r for l = 0, 1, 2, 3. With increasing l, the
mimima of Veff shifts vertically up in this schematics. The horizontal dotted lines represents
the total energy. For hydrogen(1/r potential) it is found that energy levels of different l exactly
equal one another, so for the n’th level (starting from n = 1 as the lowest) there are n different
radial wavefunctions all of the same energy.

d2 u 2mr2 u u
r − − 2 (V − E) = l(l + 1)
dr 2 ℏ r r
simplification yields
[ ]
ℏ2 d2 u ℏ2 l(l + 1)
− + V (r) + u=Eu (21)
2m dr2 2m r2
This is the radial equation, it is identical in form to the one dimensional Schrödinger equation
with an effective potential,
ℏ2 l(l + 1)
Veff = V (r) + (22)
2m r2
contains an extra term, known as the centrifugal term. It tends to throw the electron outward
away from the origin, just like the centrifugal force in classical mechanics. Writing all the ex-
plicitly, the radial equation reads:
[ ]
ℏ2 d2 u Ze2 1 ℏ2 l(l + 1)
− + − + u=Eu (23)
2m dr2 4πε0 r 2m r2

So the problem of solving the Schrödinger equation reduces to solving the one dimensional
radial equation corresponding to a mass m (ideally µ) moving in an effective potential, Veff con-
sists of the Coulomb potential plus the centrifugal potential.

The Effective potential

The effective potential as a function of the radial distance is shown in figure 1 for three different
l. The centrifugal barrier arises as a result of the orbital motion of the electron around the
nucleus and is non-zero for all electrons other than those with l = 0 (s-states).

4
 One electron Atoms

Also notice that the Coulomb potential is attractive while the centrifugal barrier is repulsive
for all r. The effect of the centrifugal barrier is to prevent the system reaching r = 0 when it
has non-zero angular momentum. It reduces the effective depth of the potential well, an effect
which increases as l increases. Combining the centrifugal barrier with the Coulomb attraction,
we see that the net ‘potential energy’ for each value of l, yields an infinite number of solutions.
The potential energy curves imply that the wavefunctions with larger l will have more energy,
and will be associated with a larger mean value of r, i.e. the electron is on average further away
from the proton.
Consider the electron bound in the well with a negative total energy (bound states). As a
result of the effective potential it moves in an orbit. As is seen in the Kepler’s planetary orbits,
in genera,l the classical orbits are elliptical with a closest (‘perigee’) and furthest (’apogee’)
classical turning points from the nucleus. With increasing l the orbits become circular from
being an elliptical.
These general features will be true no matter what the detailed form of V (r), (as long as it
roughly resembles a 1/r curve), but for the case where V (r) varies exactly as 1/r there is a set
of coincidences: the k’th member of the set of energy levels for a given value of l (counting up
from the one of least energy) has exactly the same energy as the k + 1’th member of the set
associated with the next smaller angular momentum, l−1.

Solution of the radial equation

lim Veff (r) = 0

r→∞

Therefore for large r, the radial equation becomes

d2 u 2mE
+ 2 u=0 (24)
dr2 ℏ
It is clear for E > 0,
2mE
u ∼ eiκr ; κ2 =
ℏ2
the reduced mass system behaves as a free particle at infinity and results in oscillatory solutions
and we will have a continuous spectrum.
We will now look for a bound state (E < 0) solutions for the radial equation. In order to
proceed, we introduce the dimensional quantities

Zme2
ρ = κr and ρ0 = (25)
2πϵ0 ℏ2 κ

so that the Schrödinger equation (23) reduces to

[ ]
d2 u ρ0 l(l + 1)
= 1− + u (26)
dr2 ρ ρ2

We begin by examining the asymptotic behaviour of the solution. As ρ → ∞, only the constant
term on the RHS dominates. So (approximately)

d2 u
=u
dr2
Possible solutions are eρ and e−ρ , but eρ diverges as ρ → ∞. Therefore for large ρ

lim u(ρ) ∼ e−ρ (27)

r→∞

On the other hand as ρ → 0, the centrifugal term dominates, i.e.

5
 One electron Atoms

d2 u l(l + 1)
≈ u
dr2 ρ2
The general solution is
u(ρ) = U1 ρl+1 + U2 ρ−l
but ρ−l blows up at ρ → 0. Thus,
lim u(ρ) ∼ ρl+1 (28)
r→0

The next step is to peel off the asymptotic behaviour by introducing another function v(ρ), such
that the solution is given by
u(ρ) = ρl+1 e−ρ v(ρ) (29)
There is no assumption behind ansatz. We are just expressing our knowledge about the be-
haviour of u(ρ). This is an hope that v(ρ) may turn out to be simpler than u(ρ). e−ρ is the form
at ρ → ∞ and ρl+1 is the form at ρ → 0, but in between we have v(ρ). From the ansatz
[ ]
du l −ρ dv
=ρe (l + 1 − ρ) v + ρ
dρ dρ
[{ } ]
d2 u l −ρ l(l + 1) dv d2 v
=ρe −2l − 2 + ρ + v + 2(l + 1 − ρ) +ρ 2
dρ2 ρ dρ dρ
We plug in the ansatz in the differential equation for u(ρ) (equation 26)

d2 v dv
ρ 2
+ 2(l + 1 − ρ) + {ρ0 − 2(l + 1)} v = 0 (30)
dρ dρ

This equation is solvable, and we express v(ρ) as a power series in ρ as

∞
∑
v(ρ) = aj ρj (31)
j=0

∞ ∞
dv ∑ ∑
= jaj ρj−1 = (j + 1) aj+1 ρj
dρ j=0 j=0
∞
d2 v ∑
= j(j + 1) aj+1 ρj−1
dρ2 j=0

Inserting the above into (30) we get

∞
∑ ∞
∑ ∞
∑
j(j + 1) aj+1 ρj + 2(l + 1) (j + 1) aj+1 ρj − 2 j aj+1 ρj
j=0 j=0 j=0

∞
∑
+ {ρ0 − 2(l + 1)} aj ρj = 0
j=0

Equating the coefficients of like power in ρ we get the recursion relation as,

j(j + 1) aj+1 + 2(l + 1)(j + 1) aj+1 − 2j aj + ρ0 aj − 2(l + 1) aj = 0

i.e.
2(j + l + 1) − ρ0
aj+1 = aj (32)
(j + 1)(j + 2l + 2)
This recursion formula determines the coefficients and hence the function v(ρ). Now, let us
examine what the coefficients look like at j → ∞

6
 One electron Atoms

aj+1 2j 2
lim = 2 ≈ (33)
j→∞ aj j j+1
Therefore,
2
a1 =
a0
1
2 2 2
a2 = a1 = · a0
2 1 2
2 2 2 2
a3 = a2 = · · a0
3 1 2 3
2 2 2 2 2
a4 = a3 = · · · a0
4 1 2 3 4
.. .. ..
. . .
2 2 2 2 2
aj = · · · · · · a0
1 2 3 4 k
2j
i.e. aj = a0
j!
Therefore,
∞
∑ ∞
∑
j 2j j
v(ρ) = aj ρ = a0 ρ = a0 e2ρ
j=0 j=0
j!

Hence,
u(ρ) = ρ(l+1) e−ρ · a0 e2ρ = a0 ρ(l+1) eρ
The solution diverges at ρ → ∞. Therefore, v(ρ) can not be a infinite series, it must be a polyno-
mial of order N ; and beyond the N th term other coefficients vanish automatically; it happens
if
aN +1 = 0
⇒ 2(N + l + 1) − ρ0 = 0
defining,
n=N +l+1 (34)
the so called principal quantum number, we have

ρ0 = 2n (35)

Energy Levels
Plugging in the value of ρ0 from equation (25)
( )2
m Ze2 1
E=− 2
2ℏ 4πϵ0 n2

Using the fine-structure constant

e2 1
α=
4πϵ0 ℏc

1 Z2
En = − mc2 α2 2 (36)
2 n

7
 One electron Atoms

Figure 2: Energy level diagram of atomic hydrogen.

From equation (16) we see that l is an integer, so as N . They can take values from 0 to
infinity and are related to each other by equation (34). Therefore, lmax = n − 1 (corresponding
to Nmin = 0). Hence, for each n, possible values of l are

l = 0, 1, 2, 3, · · · , (n − 1)

From the angular equation we have for each l

ml = −l, − (l − 1), · · · , − 1, 0, 1, · · · , (l − 1), l

there are (2l + 1) possible value of ml . So, the total degeneracy of the energy level En is

∑
n−1 ∑
n−1 ∑
n−1
(n − 1)n
d(n) = (2l + 1) = 2 l+ 1=2· + n = n2 (37)
2
l=0 l=0 l=0

8
 One electron Atoms

The Wavefunctions of the bound states

Combining equation (25) and (35) and using the fact that the Bohr radius is given by

4πε0 ℏ2
a=
Zme2
so from equation (31) it can be written that
1
κ2 =
a2 n2
(r) 1
∴ ρ = κr = (38)
a n
Therefore the radial wavefunction can be written as
1
Rn,l = u(ρ)
r
1 l+1 −ρ
i.e. Rn,l = ρ e v(ρ)
r
The polynomial v(ρ) is well known in mathematics. It is known as associated Laguerre polyno-
mial and written as
v(ρ) = L2l+1
n−l−1 (2ρ)

So,
ρl e−ρ 2l+1
Rn,l (r) = A L (2ρ) (39)
n a n−l−1
Therefore, total wavefunction is given by
√( ) ( )l ( )
3
2Z (n − l − 1)! −r/na 2r 2l+1 2r
ψnlm = e Ln−l−1 Ylm (θ, ϕ) (40)
na 2n[(n + l)!]3 na na

Features of radial wavefunctions

Spherical harmonics are normalized on an unit sphere. Therefore, the probability of finding
an electron between r and r + dr from the nucleus is given by

Dnl (r) dr = r2 |Rnl (r)|2 dr (41)

Several interesting features emerges from examining the radial wavefunction Rnl (r) and the
radial distribution function Dnl (r)
• For l = 0 (s-states), the radial wavefunction does not vanish at r = 0, however because
of the r2 factor in Dnl (r), the electron distribution function vanishes at r = 0. Moreover,
for each of the s-state wavefunction, ψn00 ≡ ψns is such that dRn0 /dr ̸= 0 at r = 0. This
arises from the fact that the potential energy at r = 0 is infinite. In addition, note that
Y00 = (4π)−1/2 is independent of θ and ϕ,

1 1 4Z 3
|ψn00 (0)|2 = |Rn0 (0)|2 = (42)
4π 4π a3 n3
This result play an important role in the theory of hyperfine structure.

9
 One electron Atoms

Figure 3: Radial distribution function Dnl (r) in atomic hydrogen.

• For l ̸= 0, Rnl is proportional to rl for small r. So the wavefunction remains small in the
vicinity of the nucleus. This is because the centrifugal barrier prevents the electron from
approaching the nucleus. For a fixed n, the wavefunction which has the lowest l has the
largest amplitude in the vicinity of the nucleus.

• The associated Legendre function appraring in the radial wavefunction is a polynomial of

degree N = (n − l − 1), which have N nodes (zeros). Thus, the radial distribution function,
Dnl (r) will have n − l maxima. Hence, for the highest l (= n − 1), there exist only one
maxima. In this case
Rn,n−1 (r) ∼ rn−1 e−Zr/na
Therefore, the radial distribution function will have maxima at a value of r given by

Dn,n−1 d ( 2 ) d ( 2n −2Zr/na )
= r |Rn,n−1 |2 = r e =0
dr dr dr
( )
2n 2Z
i.e. − r2n e−2Zr/na = 0
r na
n2
i.e. r = a
Z
therefore, the probable distance is proportional to n2 and inversely proportional to Z;
this is the same value which appears in Bohr model. In other words, the maxima of Dnl (r)
recedes from the nucleus with increasing n and becomes closer with to the nucleus with
increasing Z.

10