C.

THE HYDROGEN ATOM

Relations (C-12a) and (C-12b) can be written in the form:

1 2 2
= (C-40a)
2
1-
0 = (C-40b)

where is the fine structure constant, a dimensionless constant which plays a very
important role in physics:
2 2
1
= = (C-41)
~ 4 0~ 137

and where - is defined by:

- = ~ (C-42)

Since is almost the same as , the rest mass of the electron, - is practically equal to
the Compton wavelength of the electron, which is given by:

~ 3
38 10 Å (C-43)

Relation (C-40b) therefore indicates that 0 is on the order of one hundred times
the Compton wavelength of the electron. Relation (C-40a) shows that the order of
magnitude of the binding energy of the electron is between 10 4 2 and 10 5 2 , where
2
is practically equal to the rest energy of the electron:
2
0 51 106 eV (C-44)

It follows that:
2
(C-45)

This justifies our choice of the non-relativistic Schrödinger equation to describe the hy-
drogen atom. Of course, relativistic effects, although small, do exist; nevertheless, their
smallness allows them to be studied by perturbation theory (cf. Chap. XI and XII).

C-4-b. Energy levels

. Possible values of the quantum numbers; degeneracies

For fixed , there exists an infinite number of possible energy values [formula (C-
36)], corresponding to = 1 2 3, ... Each of them is at least (2 + 1)-fold degenerate:
this is an essential degeneracy related to the fact that the radial equation depends only
on the quantum number and not on (§ A-3). But, in addition, there exist accidental
degeneracies: equation (C-36) indicates that two eigenvalues and correspond-
ing to different radial equations ( = ) are equal if + = + . Figure 4, in which the
first eigenvalues associated with = 0 1 2 and 3 are shown on a common energy scale,
clearly reveals several accidental degeneracies.

825
CHAPTER VII PARTICLE IN A CENTRAL POTENTIAL. THE HYDROGEN ATOM

0
(n = 4) 4s 4p 4d 4f
(n = 3) 3s 3p 3d

(n = 2) 2s 2p

1s
(n = 1)
– EI
l=0 l=1 l=2 l=3
(s) (p) (d) (f)

Figure 4: Energy levels of the hydrogen atom. The energy of each level depends only
on . If is fixed, several values of are possible: = 0 1 2 1. To each of these
values of correspond (2 + 1) possible values for :

= +1
2
Consequently, the level is -fold degenerate.

In the special case of the hydrogen atom, does not depend on and separately,
but only on their sum. We set:

= + (C-46)

The various energy states are labeled by the integer (greater than or equal to 1), and
(C-36) becomes:
1
= 2
(C-47)

According to (C-46), it is equivalent to specify and or and to determine the

eigenfunctions. Following convention, from now on we shall use the quantum numbers

826
 C. THE HYDROGEN ATOM

and . The energy is fixed by , which is called the principal quantum number; a given
value of characterizes what is called an electron shell.
Since is necessarily an integer which is greater than or equal to 1 (§ 3-c above),
there is only a finite number of values of associated with the same value of . According
to (C-46), if is fixed, one can have:

=0 1 2 1 (C-48)

The shell characterized by is said to contain sub-shells 11 , each one corresponding to

one of the values of given in (C-48). Finally, each sub-shell contains (2 + 1) distinct
states, associated with the (2 + 1) possible values of for fixed .
The total degeneracy of the energy level is therefore:
1
( 1) 2
= (2 + 1) = 2 + = (C-49)
2
=0

We shall see in Chapter IX that the existence of electron spin multiplies this number by
2 (if we also take into account the proton spin, which is equal to that of the electron, we
obtain another factor of 2).

. Spectroscopic notation
For historical reasons (dating from the period, before the development of quantum
mechanics, in which the study of spectra resulted in an empirical classification of the
numerous lines observed), letters of the alphabet are associated with the various values
of . The correspondence is as follows:

= 0
= 1
= 2
= 3
= 4
.. ..
. .

alphabetical order (C-50)

Therefore, spectroscopic notation labels a sub-shell by the corresponding number fol-

lowed by the letter that characterizes the value of . Thus, the ground level [which is
non-degenerate, according to (C-49)], sometimes called the “ shell”, includes only the
1 sub-shell; the first excited level, or “ shell”, includes the 2 and 2 sub-shells; the
second excited level (“ shell”) includes the 3 , 3 and 3 sub-shells, etc. (The capi-
tal letters sometimes associated with the successive shells follow an alphabetical order,
starting with the letter .)
11 The concept of a sub-shell can even be found in the semi-classical model of Sommerfeld. This model

assigns, to each value of Bohr’s quantum number, elliptical orbits of the same energy and different
angular momenta. One of these orbits is circular; it is the one that corresponds to the maximum value
of the angular momentum.

827
CHAPTER VII PARTICLE IN A CENTRAL POTENTIAL. THE HYDROGEN ATOM

C-4-c. Wave functions

The wave functions associated with the eigenstates common to L2 , and the
Hamiltonian of the hydrogen atom are generally labeled, not by the three quantum
numbers , , , as we have done until now, but by , and [passage from one set
to the other simply involves use of relation (C-46)]. Since the operators L2 and
constitute a C.S.C.O. (cf. § A-3), specification of the three integers and , which
is equivalent to that of the eigenvalues of , L2 and , unambiguously determines the
corresponding eigenfunction (r).

. Angular dependence

As is the case for any central potential, the functions (r) are products of
a radial function and a spherical harmonic ( ). To visualize their angular de-
pendence on the axis characterized by the polar angles and , we can measure off a
2
distance that is proportional to ( ) for any fixed , that is, proportional to
2
( ) . Thus, we obtain a surface of revolution about the axis, since we know
that ( ) depends on only through the factor e (§ D-1-b of Chapter VI); con-
2
sequently, ( ) is independent of . We can therefore represent its cross-section
by a plane containing . This is what is done in Figure 5, for = 0 and = 0, 1
and 2 [the corresponding spherical harmonics are given in Complement AVI , formulas
(31), (32) and (33)]: 00 is a constant, and is therefore spherically symmetric; 10 2 is
proportional to cos2 , and 20 2 to (3 cos2 1)2 .

z z

O O
O

l=0 l=1 l=2

m=0 m=0 m=0

Figure 5: Angular dependence, ( ), of some stationary wave functions of the hy-

drogen atom, corresponding to well-defined values of and . For each direction of polar
2
angles , , the value of ( ) is recorded; a surface of revolution about the axis
is thus obtained. When = 0, this surface is a sphere centered at ; it becomes more
complicated for higher values of .

828
 C. THE HYDROGEN ATOM

Figure 6: Radial dependence ( ) of wave

functions associated with the first few levels
of the hydrogen atom. When 0, ( )
behaves like ; only the states (for which
= 0) have a non-zero position probability
at the origin.

. Radial dependence
The radial functions ( ), each of which characterizes a sub-shell, can be cal-
culated from the results of § C-3-c, paying attention, however, to the change in notation
introduced by formula (C-46). Figure 6 represents the variation with respect to of the
three radial functions given in (C-39):

=1 =0 =1 =0 ; =2 =0 =2 =0 ; =1 =1 =2 =1 (C-51)

The behavior of ( ) in the neighborhood of = 0 is that of (see discussion

of § A-2-c). Consequently, only states belonging to sub-shells ( = 0) give a non-zero
probability of presence at the origin. The greater , the larger the region around the
proton in which the position probability of the electron is negligible. This has a certain
number of physical consequences, particularly in the phenomenon of electron capture by
certain nuclei, and in the hyperfine structure of lines (cf. Chap. XII, § B-2).
Finally, we can derive formula (C-11b) for the successive Bohr radii. To do so,
consider the various states for which = 112 . We calculate the variation with of
the probability density for each of the preceding levels in an infinitesimal solid angle dΩ
about a fixed direction of polar angles and . In general, the position probability for
the electron in the volume element d3 = 2 d dΩ situated at the point ( ) is given
by:
2 2
d3 ( )= ( ) d dΩ
2 2 2
= ( ) d ( ) dΩ (C-52)

Here, we have fixed , and dΩ. The probability of finding the electron between and
2
+ d , inside the solid angle under consideration, is then proportional to 2 ( ) d .
2
The corresponding density is therefore, to within a constant factor, 2 ( ) (the
12 These states correspond to the circular orbits of the Sommerfeld theory (cf. note 11).

829
CHAPTER VII PARTICLE IN A CENTRAL POTENTIAL. THE HYDROGEN ATOM

factor 2 arises from the expression for the volume element in spherical coordinates). We
are interested in the case where = 1, that is, = = 1; § C-3-c then indicates
1
that the polynomial which enters into ( ) contains only one term in ( 0) . The
desired probability density is therefore proportional to:
1 2
2
( )= 2 e 0

0 0
2
2
= e 0
(C-53)
0

This function has a maximum for:

2
= = 0 (C-54)
which is the radius of the Bohr orbit corresponding to the energy .
Finally, the table below gives the expressions for the wave functions of the first
energy levels:

1
1 level =1 =0 =0 = 3
e 0

1 2
2 level =2 =0 =0 = 1 e 0

8 3
0
2 0

1 2
=2 =1 =1 = 3
e 0
sin e
8 0 0

1 2
2 level =2 =1 =0 = 3
e 0
cos
4 2 0 0

1 2
=2 =1 = 1 = 3
e 0
sin e
8 0 0

References and suggestions for further reading:

Particle in a central potential: Messiah (1.17), Chap. IX; Schiff (1.18), § 16.
The Bohr-Sommerfeld atom and the old quantum therory: Cagnac and Pebay-
Peyroula (11.2), Chaps. V, VI and XIII; Born (11.4), Chap. V, §§ 1 and 2; Pauling
and Wilson (1.9), Chap. II; Tomonaga (1.8), Vol. I; Eisberg and Resnick (1.3), Chap. 4.
Hydrogen-like wave functions: Levine (12.3), § 6.5; Karplus and Porter (12.1),
§§ 3.8 and 3.10; Eisberg and Resnick (1.3), §§ 7.6 and 7.7.
Degeneracy related to a 1 potential (dynamical group): Borowitz (1.7), § 13.7;
Schiff (1.18), § 30; Bacry (10.31), § 6.11.
Mathematical treatment of differential equations: Morse and Feshbach (10.13),
Chaps. 5 and 6; Courant and Hilbert (10.11), Vol. I, § V-11.

830
COMPLEMENTS OF CHAPTER VII, READER’S GUIDE

AVII : HYDROGEN-LIKE SYSTEMS Presentation of various hydrogen-like systems to

which the calculations of Chapter VII can be
applied directly. The accent is placed on physical
discussions and on the influence of the masses
of the particles involved in the system. Simple,
advised for a first reading.

BVII : A SOLUBLE EXAMPLE OF A CEN- Study of another case (a three-dimensional

TRAL POTENTIAL: THE ISOTROPIC THREE- harmonic oscillator) in which it is possible to
DIMENSIONAL HARMONIC OSCILLATOR calculate exactly the energy levels of a particle in
a central potential by the method of Chapter VII
(solution of the radial equation). Not theoret-
ically difficult; may be considered as a worked
example.

CVII : PROBABILITY CURRENTS ASSOCIATED Completes the results of § C-4-c of Chapter VII
WITH THE STATIONARY STATES OF THE concerning the properties of the stationary
HYDROGEN ATOM states of the hydrogen atom by calculating their
probability currents. Short and simple, useful for
Complement DVII .

Discussion of a certain number of physical phe-

nomena, using the results of Chapter VII.
DVII : THE HYDROGEN ATOM PLACED IN A DVII : the properties of an atom in a magnetic
UNIFORM MAGNETIC FIELD. PARAMAGNETISM field (diamagnetism, paramagnetism, Zeeman
AND DIAMAGNETISM. THE ZEEMAN EFFECT effect). Moderately difficult, important because
of the numerous applications.
EVII : complement intended to introduce the
EVII : STUDY OF SOME ATOMIC ORBITALS. concept of an atomic hybrid orbital, essential for
HYBRID ORBITALS understanding certain properties of the chemical
bond. No theoretical difficulty. Stresses the
geometrical aspect of wave functions.
FVII : direct application of the theory of Chap-
FVII : VIBRATIONAL-ROTATIONAL LEVELS ter VII to the study of the vibrational-rotational
OF DIATOMIC MOLECULES spectrum of heteropolar diatomic molecules.
Sequel to Complements AV (§ 1) and CVI ;
moderately difficult.

GVII : EXERCISES Exercise 2 studies the influence of a uniform

magnetic field on the levels of a simple physical
system, in an exactly soluble case. It then
provides a concrete illustration of the general
considerations of Complements CVII and DVII
concerning the influence of the paramagnetic and
diamagnetic terms of the Hamiltonian.

831
 • HYDROGEN-LIKE SYSTEMS

Complement AVII
Hydrogen-like systems

1 Hydrogen-like systems with one electron . . . . . . . . . . . 834

1-a Electrically neutral systems . . . . . . . . . . . . . . . . . . . 834
1-b Hydrogen-like ions . . . . . . . . . . . . . . . . . . . . . . . . 838
2 Hydrogen-like systems without an electron . . . . . . . . . . 839
2-a Muonic atoms . . . . . . . . . . . . . . . . . . . . . . . . . . . 839
2-b Hadronic atoms . . . . . . . . . . . . . . . . . . . . . . . . . . 840

The calculations of Chapter VII, which enabled us to determine numerous physical

properties of the hydrogen atom (energy levels, spatial distribution of the wave functions,
etc.) are based on the fact that the system under study is composed of two particles (an
electron and a proton) whose mutual attraction energy is inversely proportional to the
distance between them. There exist numerous other systems in physics that fulfil this
condition: deuterium or tritium, muonium, positronium, muonic atoms, etc. The results
obtained in Chapter VII are therefore directly applicable to these examples. All we need
to do is change the constants introduced in the calculations (masses and charges of the
two particles). This is what we shall do in this complement, in which we shall study,
in particular, how the Bohr radius and ionization energy are modified in each of the
systems to be considered. The wave functions associated with their stationary states and
the corresponding energies will then be obtained by replacing, in formulas (C-39) and
(C-47) of Chapter VII, 0 and by their new values, which give the order of magnitude
of the spatial extension of the wave functions and the binding energies of these systems.
We recall the expressions for 0 and :
1 ~2
0 =- = 2
(1)
4
1 2 2
= = (2)
2 2~2
where is the reduced mass of the electron-proton system:

= (H) = 1 (3)
+
2
and characterizes the intensity of the attractive potential ( ):
2
( )= (4)
r1 r2
In the case of hydrogen, we have seen that:

0 (H) 0 52 Å (5a)
18
13 6 eV 22 10 J (5b)

833
COMPLEMENT AVII •

How can we obtain the corresponding values for a system of two arbitrary particles,
of masses 1 and 2 , whose attraction energy is:
2
( )= (6)
r1 r2
(where is a dimensionless parameter)? All we must do is calculate the reduced mass
of the system by replacing and by 1 and 2 in (3):
1 2
= (7a)
1+ 2
and substitute the result obtained into (1) and (2), being careful to perform the substi-
tution:
2 2
= (7b)
This is what we shall do in a certain number of physical examples.

1. Hydrogen-like systems with one electron

1-a. Electrically neutral systems

. Heavy isotopes of hydrogen

The physical systems closest to the hydrogen atom are its two isotopes, deuterium
and tritium. In these atoms, the proton is replaced by a nucleus having the same charge
but possessing either one or two neutrons in addition to the proton. The mass of the
deuterium nucleus is approximately 2 , and that of the tritium nucleus, 3 . The
reduced masses therefore become, in these two cases:

(deuterium) 1 (8a)
2

(tritium) 1 (8b)
3
Since:
1
1 (9)
1 836
it is clear that the reduced masses of deuterium and tritium are very close to that of
hydrogen, and that they can be replaced by without great inaccuracy.
If we substitute either (3), or (8a), or (8b) into formulas (1) and (2), we see that the
Bohr radii and energies of the hydrogen, deuterium and tritium atoms are practically the
same. Nevertheless, there are slight differences, of the order of a thousandth in relative
value. These differences can be detected experimentally. For example, with an optical
spectrograph of sufficient resolution, it can be observed that the wavelengths of the
lines emitted by hydrogen atoms are slightly greater than those emitted by deuterium,
which are in turn greater than those emitted by tritium. This slight shift in the emitted
wavelengths is related to the fact that the nucleus is not infinitely heavy, and does not
remain fixed while the electron moves; this is called the “nuclear finite mass effect”.
Experiments have verified that formulas (7a), (1) and (2) account very precisely for this
effect.

834
 • HYDROGEN-LIKE SYSTEMS

. Muonium
The muon is a particle whose fundamental properties are the same as those of the
electron, except for a difference in mass (the mass of the muon is approximately equal
to 207 ). In particular, the muon is not sensitive to nuclear forces (strong interactions).
There are two types of muons, the and the + , whose charges are respectively equal
to those of the electron and the positron1 + . Like all other charged particles, the
muon is sensitive to electromagnetic interactions.
We can therefore consider a physical system formed by a + muon and an electron
in which the electrostatic attraction is the same as for a proton and an electron.
Bound states therefore exist. This is, so to speak, a light isotope of hydrogen, in which
the + muon replaces the proton. This “isotope” is called muonium (its atomic mass is
on the order of 0 1).
It is not difficult to use the results in Chapter VII to find the ionization energy
and Bohr radius associated with muonium; (1), (2) and (7) yield:

1+ 1
0 (muonium) = 0 (H) 0 (H) 1+ (10a)
1+ 200
1+ 1
(muonium) = (H) (H) 1 (10b)
1+ 200

Since the muon is approximately ten times lighter than the proton, the nuclear finite
mass effect is about ten times greater for muonium than for hydrogen; however, since
the electron is distinctly lighter than the muon, this effect remains small (on the order
of 0.5%). For example, the wavelengths of the optical lines emitted by muonium are be
close to those of the corresponding lines for hydrogen. The frequency of the resonance
line has been measured with a very high accuracy in 1999 [cf. ref. (11.26)], and has
shown that the charges of the electron and of the muon are the same within 10 9 .
Experimentally, the existence of muonium was revealed by its instability: the +
muon decays, emitting a positron and two neutrinos, and the lifetime of muonium is
2 2 10 6 s. The positron resulting from this decay can be detected. It is emitted
preferentially in the direction of the + muon spin2 (non-conservation of parity in weak
interactions). Detection of the positrons then leads to the experimental determination
of this direction. Since, in addition, the spin of the + muon of a muonium atom
is coupled to that of the electron (hyperfine structure coupling; cf. Chap. XII and
complements), its precession frequency in a magnetic field is different from that of a free
muon. Measurement of this frequency thus reveals the existence of muonium atoms.
The study of muonium is of very great interest, theoretically as well as experi-
mentally. The two particles which constitute this system are not subjected to strong
interactions, so that its energy levels (in particular, the hyperfine structure of the ground
state 1 ) can be calculated with great precision, without bringing in any “nuclear cor-
rection” (for the hydrogen atom, on the other hand, one must take into account the
internal structure and polarizability of the proton, which are due to strong interactions).
Comparison between theoretical predictions and experimental results provides a very
severe test of quantum electrodynamics. For instance, a measurement of the hyperfine
1 In addition, like and + , and + are antiparticles of each other.
2 Like the electron, the muon has a spin of 1/2 with which is associated a magnetic moment M =
S.

835
COMPLEMENT AVII •

structure of muonium can give an excellent determinations of the fine structure constant
= 2 ~ .

. Positronium
Positronium is a bound system composed of an electron and a positron + .
Like muonium, it can be said by extension to be an isotope of hydrogen, with the proton
being replaced by a positron. However, it must be noted that the situation is not quite
the same: in the hydrogen atom, the proton (which is much heavier than the electron)
remains almost motionless, while in positronium, the positron, the antiparticle of the
electron, has the same mass and consequently the same velocity as the electron when the
center of mass of positronium is fixed (cf. Fig. 1-b).
According to (7a), the reduced mass associated with positronium is:

(positronium) = (11)
2
Therefore:

0 (positronium) 2 0 (H) (12a)

1
(positronium) (H) (12b)
2
Thus, for a given state of positronium, the average electron-positron distance is twice the
electron-proton distance for the corresponding state of the hydrogen atom (cf. Fig. 1).
The differences between the energies of the stationary states, however, are twice as small,
and the optical line spectrum emitted by positronium is obtained by doubling all the
wavelengths of that of hydrogen.

Comment:
One should not conclude from formula (12a) that the radius of positronium is
twice that of the hydrogen atom. The Bohr radius gives an idea of the extension

Figure 1: Schematic representation of the

hydrogen (electron + proton system) and
positronium (electron + positron system)
e– e– atoms. Since the proton is much heavier
than the electron, it is located practically at
the center of mass of the hydrogen atom;
p
the electron “revolves” about the proton at
a distance 0 ( ). On the other hand, the
e+ positron is equal in mass to the electron;
a b these particles therefore both revolve about
their center of mass, the distance between
them being 0 (positronium) = 2 0 ( ).

836
 • HYDROGEN-LIKE SYSTEMS

of the wave functions associated with the “relative particle” (cf. Chap. VII, § B),
whose position r1 r2 is related to the distance between the two particles and not
to the distance between them and the center of mass . Figure 1 clearly shows,
moreover, that the hydrogen and positronium atoms are of equal size. In general,
all hydrogen-like systems for which the attractive potential is given by (6) with
= 1 have exactly the same radius, since formula (B-5) of Chapter VII shows
that:
2
r1 r = r= r (13)
1+ 2 1

Using (1), which gives the order of magnitude of the spatial extension of the wave
function 100 (r) of the ground state, we see that the “radius” of the atom can
be defined by:

~2
= 2
(14)
1

where 1 is the mass of the lighter particle (the heavier particle is found closer to
the center of mass). In all the systems considered until now, = 1 and 1 = ;
their radii are therefore the same. We shall see cases later in which the radius
is smaller, either because 1 = or because = 1.

The optical spectrum of positronium has been observed in 1975 [cf. ref. (11-23)]. As
for the (hyperfine) structure of the ground state (due to the interaction between the magnetic
moments of the electron and the positron), it has been accurately determined (cf. Comple-
ment CXII ).
The fact that positronium, like muonium, is a purely electrodynamic system (neither the
electron nor the positron is sensitive to strong interactions) explains the importance attached
to its theoretical and experimental study.
Let us also point out that positronium is an unstable system. Since the ground state is a 1
state, the electron and positron come into contact and annihilate, yielding two or three photons,
depending on the hyperfine structure level they started from. The study of the corresponding
decay rate is also of great interest in quantum electrodynamics.

. Hydrogen-like systems in solid state physics

Atomic physics is not the only domain of application of the theory presented in
Chapter VII. For example, the “donor atoms” localized in semiconductors constitute
approximately hydrogen-like systems in solid state physics.
Consider a silicon crystal; in the silicon lattice, each atom uses its four valence
electrons to form four tetrahedral bonds with its neighbors. If a pentavalent atom like
phosphorus (a donor atom) is introduced into the lattice in place of a silicon atom, it
must lose a valence electron, and its overall charge becomes positive. It then behaves like
a center which can retain the electron and form a hydrogen-like system with it. Actually,
the force acting on the electron cannot be calculated directly from Coulomb’s law in a
vacuum since silicon has a large dielectric constant 12, so that (4) must be replaced
by:
2
( )= (15)
r1 r2

837
COMPLEMENT AVII •

To be completely rigorous, we should have to replace the electron mass by the “effective
mass” of the electron in silicon, which is different from the free electron mass because
of interactions with the charges of the nuclei in the crystal. Nevertheless, we shall confine
ourselves to a qualitative discussion, noting that the effect of the high value of is to
decrease 2 in (15), that is, according to (1), to increase the Bohr radius by a factor on
the order of 10. The donor atom impurity is therefore similar to a very large hydrogen
atom, whose wave functions are spread over distances much greater than the unit cell
length of the silicon lattice.
Let us briefly describe another hydrogen-like system in solid state physics: the ex-
citon. Consider a semi-conductor crystal. In the absence of external perturbations, the
outside electrons of the atoms forming the crystal are all in states belonging to the “va-
lence band” (the temperature is assumed to be sufficiently low; cf. Complement CXIV ).
By suitably illuminating the crystal, we can, through the absorption of a photon, cause
an electron to go into the “conduction band” (which contains a set of energy levels which
are higher than those of the valence band). There is then one electron missing from the
valence band. We can treat this band as if it contained a particle of charge opposite
to that of an electron, called a “hole”. The hole can attract an electron of the valence
band and form a bound system with it: the exciton. The exciton, like hydrogen, has
a series of energy levels between which it can undergo transitions. Its existence can be
demonstrated by a measurement of light absorption by the crystal.

1-b. Hydrogen-like ions

The neutral helium atom is composed of two electrons and a positively charged
nucleus of charge 2 . Such a system, which consists of three particles, cannot be
studied with the theory of Chapter VII. However, if an electron is somehow removed
from the helium atom, the He+ ion is similar to a hydrogen atom; the only differences
are in the nuclear charge, which is twice that of the proton (the total charge of the ion is
positive and equal to ) and its mass (which, for 4 He, is approximately four times that
of the proton). Of course, there are other hydrogen-like ions: the Li++ ion (the lithium
atom, when it is not ionized, has = 3 electrons), the Be+++ ion ( = 4), etc...
Let us then consider a system formed by a nucleus, of mass and positive charge
, and an electron. If we substitute (7b) in (1) and (2), we obtain:

0( )
0( ) (16)
2
( ) ( ) (17)

(since , we have neglected the difference between the reduced mass of hydrogen
and that of the hydrogen-like ion under study; the consequences of the nuclear finite
mass effect on 0 and are, in effect, negligible compared to those due to the charge
variation). Hydrogen-like ions are therefore all smaller than the hydrogen atom, as one
would expect since the nucleus and the electron are more strongly bound. Moreover,
their energy increases rapidly with (quadratically): for example, the energy which
must be supplied to a Li++ ion to remove its last electron is greater than 100 eV. This is
why the electromagnetic frequencies that can be emitted or absorbed by a hydrogen-like
ion fall into the ultraviolet, and even, when is large enough, into the domain of X-rays.

838
 • HYDROGEN-LIKE SYSTEMS

2. Hydrogen-like systems without an electron

Thus far, the systems we have considered all include an electron. Nevertheless, there
exist numerous other particles having the same charge , able to form a hydrogen-like
system with a nucleus of charge . We shall give a few examples. The “atoms”
we are going to describe here are of course less common than the “usual” atoms which
appear in Mendeleev’s classification. They are unstable, and, in order to study them, it
is necessary to use high-energy particle accelerators to produce the particles needed for
their formation. This is why they are sometimes called “exotic atoms”.

2-a. Muonic atoms

We have already mentioned some essential features of the muon and pointed out
the existence of the muon. When this particle is attracted by a positively charged
atomic nucleus, it can form a bound system called a “muonic atom”3 .
Consider, for example, the simplest muonic atom, which is composed of a muon
and a proton; this is a neutral system whose Bohr radius is:

+ ~2 0( )
0( ) 2
(18)
200

and whose ionization energy is:

4
+
( ) 200 ( ) (19)
2~2
The size of this muonic atom is therefore on the order of several thousandths of an
Angström. Its spectrum is obtained from that of hydrogen by dividing the wavelengths
by 200; it therefore falls into the domain of soft X-rays.
What happens if, instead of revolving about a proton, the is captured by a
nucleus whose charge is times greater, like lead4 for example, for which = 82?
Formulas (1) and (2) then yield:

0( )
0( ) (20)
200
( ) 200 2 ( ) (21)

Setting = 82 in these formulas, we find for the transitions of the lead muonic atom
energies equal to several MeV (1 MeV = 106 electron volts). However, it must be noted
that formulas (1) and (2) are no longer valid in this case, since equation (20) would yield:

5
0( Pb) 3 10 Å = 3 Fermi (22)
3 We could also imagine a bound system composed of a + muon and a muon. However, given
the low intensity of muon beams that can be produced, such an atom is very difficult to create.
4 Such a system can be formed by directing a beam onto a target of lead atoms. When a is
captured by a lead nucleus, it revolves about it at a distance about 200 times smaller than the distance
to the electrons of the innermost shell of the atom. The nuclear charge is therefore practically the only
one to affect the muon. Thus, in studying the states of the muonic atom, we can simply ignore the
electrons.

839
COMPLEMENT AVII •

that is, a radius slightly smaller than the radius of the lead nucleus. The calculations of
Chapter VII are therefore no longer valid. This is because they are based on form (6) of
the potential ( ), which is correct5 only when the particles under study, separated by
distances much greater than their dimensions, can be considered to be point particles.
This hypothesis, very well satisfied for hydrogen, is not valid in the case studied here.
However, (20) and (21) give the correct order of magnitude of the energies and
radius of the lead muonic atom. The physical consequences of the existence of a non-
zero spatial extension of the nucleus (“volume effect”) will be studied in greater detail
in Complement DXI . We take this occasion to note, however, that one of the reasons
for interest in muonic atoms is precisely related to this type of effect: the muon
“explores”, as it were, the internal structure of the nucleus6 , and the energy levels of
muonic atoms depend on the electrical charge distribution and magnetism inside the
nucleus (recall that the muon is not sensitive to nuclear forces). Thus, the study of these
states can furnish information very useful in nuclear physics.

2-b. Hadronic atoms

“Hadrons” are those particles sensitive to strong interactions, as opposed to “lep-

tons”, which are not. Electrons and muons, whose bound states in a Coulomb potential
we have studied thus far, are leptons. Protons, neutrons and mesons such as the meson,
etc... are hadrons. Of the latter, those that are negatively charged can form a hydrogen-
like bound system with an atomic nucleus, called a “hadronic atom”. For example, the
nucleus- meson system yields a “pionic atom”; the nucleus-Σ particle system, a “sig-
maonic atom”7 ; the nucleus- meson system, a “kaonic atom”; the nucleus-antiproton
system, an “antiprotonic atom”, etc... All of the systems just cited have actually been
observed and studied. They are all unstable, but their lifetimes are long enough for us
to observe some of their spectral lines. Hydrogen atom theory, which takes into account
only the electrostatic interaction between the two particles under consideration, does not,
of course, apply to such systems, in which strong interactions play an important role.
However, since they are of very short range, strong interactions can be neglected in the
study of excited states of the hadronic atom (other than the states), in which the two
particles are far apart. The theory of Chapter VII is then applicable, as are formulas
(1) and (2), which, in all these cases, lead to much smaller Bohr radii and much greater
energies than for hydrogen. Thus, measurement of the spectral frequencies emitted by
pionic atoms gives a very precise determination of the mass of the meson.

References and suggestions for further reading:

Exotic atoms: see the subsection “Exotic atoms” of section 11 of the bibliography;
see also Cagnac and Pebay-Peyroula (11.2), Chap. XIX, § 7; Weissenberg (16.19),
Chap. 4, § 2 and Chap. 6.
Excitons: Kittel (13.2), Chap. 17, p. 538; Ziman (13.3), § 6.7.

5 Inside the nucleus, the potential is approximately parabolic (cf. Complements AV , § 4 and DXI ).
6 The concept of the impenetrability of two solid bodies is macroscopic. In quantum mechanics,
nothing prevents the wave functions of two particles of different nature from overlapping.
7 The term “mesic atom” is sometimes used to denote a system involving a meson. Similarly, since

the Σ is a hyperon (a particle heavier than the proton), the sigmaonic atom is sometimes called a
“hyperonic atom”.

840