Chapter 3.

The Structure of the Atom

Notes:
Most of the material in this chapter is taken from Thornton and Rex, Chapter 4.

3.1 The Atomic Models of Thomson and Rutherford

Determining the structure of the atom was the next logical question to address following
the discovery of the electron by J. J. Thomson. It was already established the number of
electrons within an atom was essentially half of the atoms mass number (i.e., the ratio of
the atoms mass to that of hydrogen). Since the electron was also known and measured to
be much less massive than the atom, it was expected that mass of the positively charged
component of the atom would be significant (relatively speaking; the atoms were known
to be electrically neutral). Understanding the structure of the atom was also desirable
when considering that, at that time, chemists had identified more than 70 different kinds
of atoms. One would then be hopeful that these different atomic realisations could be
explained within the context of a theory resting on simple rules, which would define
the structure of atoms.

Thomson proposed a model where the positive charge within an atom is uniformly
distributed within a sphere of appropriate size, with the electrons somehow localized
within this volume; this is the so-called plum-pudding model. Thomson then theorized
that when the atom is heated up the electrons are randomly accelerated and responsible
for the emission of radiation. The electrons lower mass implied that their motion would
be more important than the heavier positive charges and therefore more apt to produce

Figure 1 Schematic of the

scattering experience performed
in Rutherfords laboratory to
investigate the atomic structure.

- 38 -
radiation. Despite the apparent simplicity of his model, Thomson never was able to
calculate the spectrum of hydrogen with it. This model eventually had to be abandoned.

This would be made evident from the work of Ernest Rutherford (1871-1937) who
would use his newly established experimental method of bombarding particles (i.e.,
helium nuclei) onto thin target materials. The main evidence came from the fact that his
teams experiment on gold-leaf targets revealed the backward scattering (i.e., at more
than 90 ; see Figure 1) for some incident particles. Since Thomsons model would
imply that the particles should go through the positive charge of the atom basically
unimpeded (that component corresponds to the pudding in the model) the scattering
could only be the result of collisions between the particles and the electrons
populating the atoms. The impossibility that the backward scattering resulted from such
collisions can be verified as follows.

We calculate the maximum scattering angle stemming from the elastic collision between
one particle and an electron, choosing a reference frame where the electron is initially
at rest. Of course, we must impose the conservations of energy and linear momentum
before and after the collision

1 1 1
m v2 = m v 2 + me ve 2
2 2 2 (3.1)
m v = m v + me v e .

initial final

It can be shown (and you have seen in your first-year physics course) that the solution to
this problem, when combining these two equations, is

m me
v = v
m + me
(3.2)
2m
v e = v .
m + me

For this case we have m me 4 1837 1 and equations (3.2) simplify to

v v
(3.3)
v e 2v .

That is, the particle basically overruns the electron as if nothing was standing in its
way. More precisely, we calculate for the change in momentum for each particle

- 39 -
 p = m ( v v )
2m me
= v
m + me (3.4)
2me v
p e = p .

The absolute maximum scattering angle (as defined in Figure 1) will happen when these
changes in momenta are all oriented in a direction normal to the initial velocity of the
particle. We then find that

p
= tan 1
m v
2me
tan 1 (3.5)
m
0.016.

We must however consider the fact that one particle will collide with more than one
electron as it scatters against the thin gold-leaf target. But these collisions are all
statistically independent, i.e., they will all be randomly oriented on a cone of opening
angle 2 . The angle resulting from one scattering event can therefore be decomposed
into two components about perpendicular axes. For example, if we choose the direction
of the incident particle as along the z -axis, i.e., v = v e z , then we have

= x e x + ye y . (3.6)

The random composition of the different angles implies that the mean angle about a given
axis is on average zero, but the standard deviation will be determined as follows, say, for
the x -axis,

x,N
2
= N x 2 , (3.7)

after N statistically independent collisions (this is an example of a random walk). The

same rule applies for the y -axis, such that the standard deviation of the total angle is

(
N2 = N x2 + y2 ) (3.8)
= N . 2

For Rutherfords experiment the gold-leaf was = 6 10 7 m thick; we must use this
information to approximately calculate N . Using the atomic data for gold we can
determine the density of atoms n in the leaf as

- 40 -
 atoms 1 mol 19.3g
n = 6.02 10 23
mol 197g cm 3 (3.9)
= 5.9 10 29 atoms/m 3 ,

(
which implies that one atom occupies on average a volume of 1 5.9 10 29 m 3 . The)
linear distance between atoms is therefore approximately the cubic root of this value, i.e.,
d 2.6 10 10 m . It then follows that

N = 2, 300 atoms, (3.10)
d

and

N = N
2, 300 0.016 (3.11)
0.8.

This value is evidently completely inconsistent with backward scattering. This prompted
Rutherford to conclude that his results were not in agreement with Thomsons atomic
model. In fact, looking at the first of equations (3.2) we find that a backward scattering
will be more likely to happen if we replace the mass of the electron by a larger one. It is
straightforward to verify that backward scattering is very likely if the mass of the target is
49m (i.e., approximately the atomic mass of gold). For example, the final speed of the
particle is v 0.85v for a one-dimensional collision. Rutherford then proposed a
new model where the atom is mostly empty space with the charge (positive or negative;
his experiment could not determine the position of the electrons) concentrated at the
centre: the nucleus.

3.1.1 Rutherford Scattering

Rutherford could quantitatively follow up on his proposition on the presence of an atomic
central charge by making further scattering experiments using his particles
bombarding technique and making the following assumptions:

1. The targets (initially at rest) are much more massive than the particles and,
therefore, do not recoil; this implies that the magnitude of the initial and final
momenta of the incident particles are the same.
2. The particles scatter off only one target (i.e., the target foil is very thin).
3. The incident and target particles can be treated as point masses.
4. The only force involved in the scattering is the (electrostatic) Coulomb force.

A diagram for a given collision is shown in Figure 2; such process is called Coulomb or
Rutherford scattering. The important parameters are shown in the figure: b is the impact

- 41 -
 Figure 2 Rutherford scattering of a
particle off a massive target.
parameter, is the scattering angle, Z1e and Z 2 e are the charges of the incident and
target particles, respectively, and the instantaneous distance r of the incident particle to
the target can be parameterised with the angle , which is set to zero at the collisional
symmetry axis (the z -axis). Because the angular momentum is also conserved (and the
target does not move or recoil) the magnitude of the angular momentum of the particle
is the same before and after the scattering at

L = mv0b, (3.12)

with v0 is the initial speed of the particle, and it also implies that the scattering event
takes place in a plane. Rutherford was able to derive an equation relating the impact
parameter b and the scattering angle , which we demonstrate here.

The change in momentum of the particle is given by

p = p f p i (3.13)

where the only difference between the initial and final vectors is in their direction, as they
both have a magnitude of mv0 . We then write

{ }
p = mv0 cos ( ) 1 e x + sin ( ) e y , (3.14)

with the x and y axes being horizontal and vertical in Figure 2, respectively. It follows
that

p = mv0 cos ( ) 1 + sin 2 ( )

2

= mv0 2 1 cos ( )
(3.15)
= mv0 2 2sin ( 2 )
2

= 2mv0 sin ( 2 ) ,

- 42 -
since more generally

1
sin ( a ) sin ( b ) = cos ( a b ) cos ( a + b )
2
1
cos ( a ) cos ( b ) = cos ( a b ) + cos ( a + b ) (3.16)
2
1
sin ( a ) cos ( b ) = sin ( a b ) + sin ( a + b ) .
2

We can further use equations (3.14)-(3.16) to write

p = mv0 2sin 2 ( 2 ) e x + 2sin ( 2 ) cos ( 2 ) e y

= p sin ( 2 ) e x + cos ( 2 ) e y (3.17)
= pe z

The last equation follows from the fact that, according to Figure 2, the z -axis makes an
angle of ( ) 2 relative to the negative (horizontal) x -axis, which then means that

e z = cos ( ) 2 e x + sin ( ) 2 e y
(3.18)
= sin ( 2 ) e x + cos ( 2 ) e y .

The instantaneous Coulomb force felt by the particle is

1 Z1Z 2 e2
F= er , (3.19)
4 0 r 2

which is related to p in the last of equations (3.17) through

p = F cos ( ) e z dt , (3.20)

from the definition of the angle . We can therefore write

p = 2mv0 sin ( 2 )
Z1Z 2 e2 cos ( )
(3.21)
=
4 0
r2
dt .

But the magnitude of the angular momentum of a point mass m equals its moment of
inertia mr 2 times its angular speed d dt , and from equation (3.12) we write

- 43 -
 d
mr 2 = mv0b, (3.22)
dt

which can be inserted in equation (3.21) to yield

Z1Z 2 e2 cos ( )
2mv0 sin ( 2 ) = dt
4 0 v b ( dt d )
0
(3.23)
Z Z e2 f
= 1 2 cos ( ) d .
4 0 v0b i

The limits integration are simply related through i = f , since the z -axis is the axis of
symmetry, and from our previous discussion (following equation (3.17)) we know that

i = . (3.24)
2

Equation (3.23) now transforms to

Z1Z 2 e2
2mv0 sin ( 2 ) = sin ( )
f

4 0 v0b f

Z1Z 2 e2
= 2sin (3.25)

4 0 v0b 2
Z1Z 2 e2
= 2 cos ,
4 0 v0b 2

or

Z1Z 2 e2
b= cot
4 0 mv0 2
2
(3.26)
Z1Z 2 e2
= cot .
8 0 K 2

with K the kinetic energy of the particle. But in order to compare with experiments it
is desirable to transform this equation to determine the number of particles N ( )
scattered at angle . To do so we first consider N s the number of targets of volume
density n contained with a foil of thickness t and area A

N s = ntA. (3.27)

The probability of scattering f for an incident particle equals this number of targets

- 44 -
times their cross-section (i.e., their subtended area) divided by the total area. That is,

ntA
f= = nt . (3.28)
A

We now associate the cross-section to the area b 2 , i.e., = b 2 , for scattering

through an angle or more, since this angle increases with a smaller impact parameter
(see Figure 2 and equation (3.26)). The probability of scattering then becomes

2
Z Z e2
f = nt 1 2 cot 2 . (3.29)
8 0 K 2

But since a detector located at an angle will have a finite resolution (or width) d
associated with a set of impact parameters contained within the range [ b,b + db ] we need
to calculate f ( + d ) f ( ) f ( ) + ( df d ) d f ( ) = df , which is

2
Z Z e2
df = nt 1 2 cot sin 2 d . (3.30)
8 0 K 2 2

The number of scatterings per unit area at is then simply the product of incident
particles N i of impact parameters in the range [ b,b + db ] and df divided by the area
containing these particles (see Figure 3)

N i df
N ( ) =
dA
2
Z1Z 2 e2
N i nt cot sin 2 d
8 0 K 2 2
= (3.31)
2 r sin ( ) d
2

2
N nt e2 Z12 Z 22
= i ,
16 4 0 r 2 K 2 sin 4 ( 2 )

since, from the last of equations (3.16), 2sin ( ) = sin ( 2 ) cos ( 2 ) . This the Rutherford
scattering equation, which was found to perfectly match the experimental data obtained
by his research team. This confirmed his proposition that the atom contained a compact
central charge.

Exercises
1. (Ch. 4, Prob. 5 in Thornton and Rex.) Calculate the impact parameter for scattering a
7.7-MeV particle from gold at an angle of (a) 1 and (b) 90 .

- 45 -
 Figure 3 The area covered by incident
particles with impact parameters contained
within .

Solution.

The particle and gold nucleus have Z1 = 2 and Z 2 = 79 , respectively, to be inserted in

equation (3.26) for = 1

Z1Z 2 e2
b= cot
8 0 K 2
(3.32)
( 2 )( 79 )(1.44 10 9 eV m )
= cot ( 0.5 ) = 1.69 10
12
m
2 ( 7.7 10 6 eV)

and for = 90

( 2 )( 79 )(1.44 10 9 eV m )
b= cot ( 45 ) = 1.48 10 14 m. (3.33)
2 ( 7.7 10 eV)
6

We used e2 ( 4 0 ) = 1.44 10 9 eV m to obtain these equations.

2. (Ch. 4, Prob. 7 in Thornton and Rex.) For aluminum ( Z 2 = 13 ) and gold ( Z 2 = 79 )

targets, what is the ratio of particle scattering at any angle for equal numbers of
scattering nuclei per unit area?

Solution.

The number of scatterings at an angle can be obtain from equation (3.31) with

- 46 -
 2
N i nt e2 Z12 Z 22
N ( ) = , (3.34)
16 4 0 r 2 K 2 sin 4 ( 2 )

where nt represents the number of targets per unit area. Then the ratio of particle
scattering at any angle is

2
N Au nAut Au Z Au 79 2
= 2
= = 36.93, (3.35)
N Al nAlt Al Z Al 132

since it is assumed that the numbers of scattering nuclei per unit area are the same
nAut Au = nAlt Al .

3. (Ch. 4, Prob. 12 in Thornton and Rex.) Consider the scattering of an particle from
the positively charged part of the Thomson plum-pudding model. Let the kinetic energy
from the particle be K (nonrelativistic) and let the atomic radius be R . (a) Assuming
that the maximum Coulomb force acts on the particle for a time t = 2R v (where v
is the initial speed of the particle), show that the largest scattering angle we can expect
from a single atom is

e2 Z1Z 2
. (3.36)
4 0 KR

(b) Evaluate for an 8.0-MeV particle scattering from a gold atom of radius 0.135
nm.

Solution.

(a) The maximum Coulomb force is at the surface can be calculated by putting all the
charge at the centre of the sphere and is therefore equal to Z1Z 2 e2 4 0 R 2 , with Z1 = 2 .
Then

p = Ft
Z1Z 2 e2 2R
= (3.37)
4 0 R 2 v
e2 2Z1Z 2
= .
4 0 Rv

Using the same argument as that leading to equation (3.5) we find the maximum angular
deflection, for small angles, with

- 47 -
 p
tan =
p
1 e2 2Z1Z 2
(3.38)
mv 4 0 Rv
e2 Z1Z 2
.
4 0 KR

(b) For an 8.0-MeV particle scattering from a gold atom ( Z 2 = 79 ) of radius 0.135 nm
we find that

=
( 2)(79)(1.44 eV nm ) = 2.11 10 4
rad = 0.012. (3.39)
(8 MeV )(0.135 nm )
3.2 The Classical Atomic Model
Rutherfords theory of particle scattering and the corresponding experimental results
allowed physicists to establish that the atom was composed of a nucleus harbouring the
positive charge of the atom and electrons surrounding it (Rutherford scattering equation
(3.31) yields information on Z 2 the charge of the nucleus, while the experimental
determination of the scattering angles confirmed that it is massive and concentrated).
Although these results clearly excluded Thomsons plum-pudding model, it did not
completely rule out a classical model.

Indeed, Thomson had previously considered a planetary model with a positively

charged nucleus at the centre with orbiting electrons. As we will later see this is very
close in structure to early quantum mechanical models, but it could easily be shown to be
untenable according to classical physics. If we take for example the hydrogen atom
composed of a proton nucleus and a single orbiting electron, then we can calculate the
Coulomb force binding the electron to the proton with

e2
F= er , (3.40)
4 0 r 2

where e r is the unit vector directed from the proton to the electron. For the atom to be in
equilibrium, this force must be counteracted by the centrifugal acceleration (times the
electron mass m ) due to the orbital motion of the electron

e2 mv 2
+ = 0, (3.41)
4 0 r 2 r

or, for the electron kinetic energy,

- 48 -
 1 2
K= mv
2
1 e2
= (3.42)
2 4 0 r
1
= U,
2

with U the potential energy due to the Coulomb interaction. It therefore follows that the
total energy of the electron is negative with

E = K +U
1 e2 (3.43)
= < 0.
2 4 0 r

This negative value indicates that the electron is on a bound orbit about the nucleus.

But we also know from Maxwells laws that the electron must radiate electromagnetic
energy as it is accelerating on its orbit (to constantly change its direction of motion and
stay on a circular orbit). From the conservation of energy its total orbital energy must
become more negative, and we see from equation (3.43) that the size of its orbit r
decreases accordingly. This is shown schematically in Figure 4. This scenario is evidently
flawed as the electron would eventually collapse on the proton and the atom would cease
to exist. This was another spectacular failure for classical physics

Exercises

4. (Ch. 4, Prob. 14 in Thornton and Rex.) The radius of a hydrogen nucleus is believed to
be about 1.2 10 15 m . (a) If an electron rotates around the nucleus at that radius, what
would be its speed according to the planetary model? (b) What would be the total

Figure 4 The orbit of the

electron about a proton for a
classical hydrogen atom.

- 49 -
mechanical energy? (c) Are these reasonable?

Solution.

(a) According to equation (3.42) we have

e2 e2 c
v= =
m4 0 r 4 0 mc 2 r
c 1.44 10 9 eV m
= (3.44)
( 511 keV)(1.2 10 15 m)
= 1.53c.

(b) According to equation (3.43) we write

1 e2
E=
2 4 0 r
1.44 10 9 eV m
= (3.45)
2 1.2 10 15 m
= 600 keV.

(c) Clearly the speed calculated in (a) is impossible as it is greater than the speed of light,
and the energy in (b) is far too great compared to the ionization potential of hydrogen
(i.e., 13.6 eV).

3.3 The Bohr Model of the Hydrogen Atom

Niels Bohr (1885-1962) was certainly one of the greatest physicists of the twentieth
century, and arguably one of the best ever. Like Einstein he was one of the first to
recognize the importance of Plancks quantum hypothesis, but he also pushed it further
and with more far-reaching consequences than anybody else. He was a bold thinker who
could see through the challenges classical physics faced and came up with imaginative
solutions based on the nascent quantum theory. When considering the failure of the
atomic classical model Bohr proposed four general assumptions or postulates, which he
could then use to explain much of the experimental data known at the time:

1. To avoid the in-spiral motions of electrons orbiting a nucleus (see Figure 4), Bohr
postulated that stationary states existed in atoms where electrons could stay on
stable orbits. The electrons would therefore not radiate electromagnetic energy
when occupying a stationary state.
2. The emission or absorption of radiation can only occur when an electron makes a
transition between two stationary states. If we denote these states by the
subscripts 1 and 2, then Bohr asserted that the difference in the energy of the
two states is quantized with

- 50 -
 E = E1 E2 = hf , (3.46)

where h is the Planck constant and f the frequency of the radiation emitted or
absorbed. This hypothesis is then consistent with Einsteins earlier work on the
quantization of radiation (i.e., the notion of the photon) and the photoelectric
effect.
3. The classical laws of physics only apply to stationary states, not to the transitions
between them.
4. Finally, not only is the energy between stationary states quantized but so is the
angular momentum of the electron-nucleus system, with levels that are multiples
of h 2 .

We can follow Bohr and calculate the following fundamental relations and quantities for
the hydrogen atom. We first consider the angular momentum of the orbiting electron and
equate it to the quantized level of the systems stationary state (the nucleus is much more
massive than the electron and we will neglect its contribution at present)

L = mrv = n, (3.47)

with r the radius of the orbit (assumed circular), v the speed of the electron, and
n = 1, 2, 3, the principal quantum number. Using this relation and equation (3.42) we
find that

e2
v2 =
4 0 mr
(3.48)
n22
= 2 2,
mr

This relation, in turn, allows us the define the Bohr radius when n = 1

4 0 2
a0 =
me2 (3.49)
= 0.53 10 10 m,

from which we determine the other permitted orbital radii for stationary states

4 0 n 2 2
rn = 2
= n 2 a0 . (3.50)
me

Bohr was thus able to determine a fundamental a0 for the hydrogen atom, entirely based
on fundamental constants, that was in excellent agreement with the experimental data
already existent at the time (in 1913). This size, i.e., the Bohr radius, corresponds to the
ground state of the hydrogen atom (when n = 1 ), while stationary states for n > 1 are

- 51 -
called excited states. The energy associated with stationary states are given, from
equations (3.43) and (3.50),

e2
En =
8 0 rn
(3.51)
e2 E
= 20 ,
8 0 a0 n 2
n

with the energy of the ground state

e2
E0 =
8 0 a0
2 (3.52)
m e2
= 2 = 13.6 eV.
2 4 0

We can also use the second of Bohrs four assumptions to calculate the frequency or the
wavelength of the radiation emitted as a result of a transition between two stationary
states of principal quantum numbers nu and nl with

1 Eu El
=
hc
E 1 1
= 0 2 2 (3.53)
hc nu nl
1 1
= R 2 2 ,
nl nu

with

E0
R =
hc
2 (3.54)
m e2
= 3 = 1.097373 10 7 m 1 .
4 c 4 0

Bohr had thus succeeded in theoretically deriving the Rydberg equation, established
experimentally in 1890, and confirming the value of the Rydberg constant R (see
equation (2.11) of Chapter 2 of the lecture notes). The small discrepancy between Bohrs
theoretical value and the experimental determination of Rydberg can be explained by
taking into account the mass of the proton. Furthermore, Bohrs theory could be used to
predict the presence of then unknown spectral lines, which were eventually verified
experimentally.

- 52 -
Finally, another fundamental parameter can be established by considering the orbital
speed of the electron. Using equations (3.48) and (3.50) we find that

1
vn =
n ma0
(3.55)
1 e2
= .
n 4 0

Focusing on the speed of the ground state at n = 1 and dividing it by the speed of light we
get a unit-less parameter

v1
=
c ma0 c
(3.56)
e2 1
= ,
4 0 c 137

which is the so-called fine structure constant.

3.3.1 The Correspondence Principle

It was highly difficult for several physicists at the start of the twentieth century to accept
the new quantum physics at face value and equally troubling that it should be assigned
with any kind of physical reality. On the one hand, classical physics was incredibly
successful in describing the macroscopic world as we see and feel it, while on the other it
appeared that the new experiments could only be explained within the context of the
theories of Planck, Einstein, and Bohr.1 This apparent paradox was lessened by the fact
that the classical and quantum realms dealt with scales that were worlds apart. For
example, consider the scale of the Bohr radius (i.e., on the order of 10 10 m ) to those we
experience on a daily basis. Although such could be the hope that both worlds would
coexist independently in this manner, common sense would also dictate that there must
exist a domain where the two realities must merge. After all, nature cannot be both
classical and quantum at the same time, in view of their very different characters.

Bohr was acutely aware of this issue and proposed The Correspondence Principle as a
guideline:

In the limits where classical and quantum theories should agree, the quantum theory
must reduce to the classical result.

This is not to be interpreted as a concession made by the new generation of physicists

1
Incidentally, we now know, as physicists of that generation were about to realise, that
Bohrs theory of the hydrogen atom was not correct. But it provided the basics elements
that would guide the elaboration of the successful, modern theory of quantum mechanics.

- 53 -
to make the quantum theory more palatable But it should be understood in the context
where one transitions from the quantum to the classical domain. Alternatively the
Correspondence Principle could be stated such that the behaviour of quantum systems
must reproduce the results of classical physics in the large-quantum-number limit. This
will be made clearer with the following example.

According to equations (3.46) and (3.51) the angular frequency of the radiation resulting
from an atomic transition between two adjacent stationary states can be calculated with

E0 1 1
q = 2 2
( n + 1) n
(3.57)
E 2n + 1
= 0 2 2 .
n ( n + 1)

To find the large-quantum-number limit, we choose an arbitrarily large value for n such
that we can approximate equation (3.57) to

2E0
q
n 3
2 (3.58)
m e2
3 3 .
n 4 0

This equation should reproduce the corresponding classical result as for such a large
value for the principal quantum number the difference in energy associated with the two
stationary states is much smaller than that of the ground state (i.e., En+1 En E0 ).

Classically, the frequency of the radiation c is the same as that of the acceleration of
the electron responsible for the emission of electromagnetic waves. We therefore write

v
c = , (3.59)
r

which is simply the reciprocal of the time taken to complete one orbit (i.e., the period).
The classical relation for the orbital speed is obtained with equation (3.42) (or
equivalently through the first of (3.48))
12
e2
v= (3.60)
4 mr 0

such that

- 54 -
 12
e2
c = 3
. (3.61)
4 0 mr

As can be verified by examining the different quantities composing it, this equation is
entirely classical in nature, i.e., the Planck constant is not part of it.

To establish the correspondence between equations (3.58) and (3.61), we note that the
quantity n present (to the third power) in the quantum mechanical equation is nothing
more than the angular momentum of the electron. We therefore first establish the
following correspondence
12
mre2
n mvr = . (3.62)
4 0

Inserting this result in equation (3.58) we have

2
4 0 e2
32

q m
mre2 4
0
12
(3.63)
e 2

3
= c,
4 0 mr

which is what we sought to establish in agreement with the Correspondence Principle.

3.3.2 The Limitations of the Bohr Model

As we stated before, the Bohr model of the hydrogen atom was not the final answer to the
problem of atomic structure but an important, successful, and fruitful first step. His main
successes were covered in the previous sections, but there were also some shortcomings
that made it clear that further refinements were needed. First, the wavelength for the
different spectral lines that could be experimentally verified for the hydrogen atom, it was
found that the previous equation (3.53) was not perfect but that small discrepancies
existed between theory and experiment. However, these could easily be corrected by
taking the finite mass of the proton into account; Bohrs theory effectively approximated
the protons mass as infinite (see the Second Problem List). More importantly, the
following limitations of the Bohr model were eventually recognised:

1. Although it could be successfully applied to other one-electron atoms (e.g., He + ,

Li2+ , etc.), it did not work for more complicated atoms, not even the next
simplest atom, helium.
2. It was not able to account for the intensity of spectral lines, or the fact they could
split into sets of so-called fine structure lines when subjected to external magnetic
or electric fields, for example.
3. It could not explain the binding of atoms into molecules.

- 55 -
 4. Finally, it was eventually discovered that the atomic ground state has an orbital
angular momentum of zero and not , as assumed by Bohr.

These difficulties were eventually conquered, but only when the full quantum theory was
established in the decade or two that followed.

Exercises

5. (Ch. 4, Prob. 19 in Thornton and Rex.) The Ritz combination rules express the
relationships between the observed frequencies of the optical emission spectra. Prove one
of the more important ones:

( )
f ( K ) + f ( L ) = f K , (3.64)

where K and K refer to the Lyman series and L to the Balmer series of hydrogen
(see Figure 5).

Solution.

According to equation (3.53) the frequency of a transition for a one-electron atom is

given by

c 1 1
f = = Z 2 Rc 2 2 , (3.65)
nl nu

where Z is the charge of the nucleus and R the Rydberg constant associated to the atom.

Figure 5 The different

shells (i.e., K, L, M, etc.) of
the hydrogen atom and their
associated transitions.

- 56 -
Referring to the figure we have

1
f (K ) = Z 2 Rc 1
4
1
f (K ) = Z 2 Rc 1 (3.66)
9
1 1
f (L ) = Z 2 Rc ,
4 9

which yield equation (3.65) since (1 1 4 ) + (1 4 1 9) = 1 1 9 .

6. (Ch. 4, Prob. 29 in Thornton and Rex.) A hydrogen atom exists in an excited state for
typically 10 8 s . How many revolutions would an electron make in a n = 3 state before
decaying?

Solution.

We know from equation (3.55) that

vn = . (3.67)
nma0

But the number of revolutions N in a time interval t is given by

vt
N=
2 rn
t
= (3.68)
nma0 2 n 2 a0
t
= ,
2 mn 3a02

which for t = 10 8 s yields N = 2.38 10 6 .

- 57 -