Contents

3 Atomic Mechanisms 3

3.1 Dipoles and Polarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

3.2 Magnetic Dipole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3.3 Magnetic Dipoles and Magnetization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3.4 Diamagnetism and Paramagnetism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1
PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

PHAS0038: Term 1 Atomic Mechanisms 2

Chapter 3

Atomic Mechanisms
We begin by looking a little more in depth at the necessary tools to describe electromagnetism and its interaction
with matter - part of this investigation is concerned with the multipole representation of the potential arising
from an arbitrary distribution of charge or current. The ‘dipole moment’ of a distribution of charge or current
is an important concept, because it is also used, for example, in characterizing the ‘magnetization’ of a material.

3.1 Dipoles and Polarization

Dipoles

Let’s begin by considering this simple distribution of electrostatic charges:

• A dipole is a pair of equal (magnitude) and opposite (sign) charges separated by a small distance (see Figure
3.1). By ‘small’, we mean small compared to the distances from the dipole at which we are interested in
examining its potential.
• The microscopic effect of an electric field on a dielectric can be modelled with a collection of single dipoles.
• We can also expand the field or potential of an arbitrary charge distribution in terms of multipoles, of
which the dipole ‘moment’ is one term.
• This more general ‘multipole’ concept will be discussed after we have looked at the single dipole case.

Notice that if we have a pair of equal but opposite charges separated by a distance l (Figure 3.1), we can introduce
the vector displacement l which points from the negative to the positive charge position, and has magnitude l.
Using this definition, the dipole moment is p = ql.

Dipole Field and Potential

Let’s choose an origin, for convenience, halfway between the charges of our dipole. Our ‘test point’ for evaluating
field and potential has vector position r with respect to this origin. The charges have positions l/2 (positive
charge) and −l/2 (negative charge). The corresponding expressions for the electric field E and the electrostatic
potential V become:

{\bf E} ({\bf r}) &= \frac {q}{4\pi \epsilon _0} \biggl ( \frac {\hat {\bm r}_+}{r^2 _+} - \frac {\hat {\bm r}_-}{r^2 _-} \biggr ) \\ V ({\bf r}) &= \frac {q}{4\pi \epsilon _0} \biggl ( \frac {1}{r _+} - \frac {1}{r _-} \biggr )

(3.1b)

3
PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Figure 3.1: Geometry of a simple dipole .

Note that the positions of the test point with respect to each charge are:

{\bm r}_+ &= {\bm r} - {\bm l}/2 \\ {\bm r}_- &= {\bm r} + {\bm l}/2
(3.1d)

And we adopt the common convention of using the same symbols in regular font to represent magnitude, and
with the ‘hat’ to represent a unit vector in the same direction.
We seek the electric field at position r due to a dipole (length l) at the origin, oriented along the z axis for
convenience. Using the equations above, we can calculate the distances from each charge to the test point (note
that θ denotes the angle between the dipole direction and the position vector r):

|{\bf r}_+| &= \sqrt {({\bm r} - {\bm l}/2) \cdot ({\bm r} - {\bm l}/2)}\\ {\bm r}_+ \cdot {\bm r}_+ &= r^2-rl \cos \theta +\frac {l^2}{4} \\ &= r^2 \biggl ( 1-\frac {l}{r} \cos \theta +\frac {l^2}{4r^2} \biggr ) \\ |{\bf r}_+| &= r\sqrt {1-\frac {l}{r}\cos \theta +\frac {l^2}{4r^2} }\\

(3.2e)

Similarly:

|{\bf r}_-| &= r\sqrt {1+\frac {l}{r}\cos \theta +\frac {l^2}{4r^2} } (3.2f)

PHAS0038: Term 1 Atomic Mechanisms 4

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

We can now write down 1/|r+ | and 1/|r− | and expand them to first order in l/r:

\frac {1}{|{\bf r}_+|} &= \frac {1}{r} \biggl ( 1-\frac {l}{r} \cos \theta +\frac {l^2}{4r^2} \biggr )^{-\frac {1}{2}} \\ &\approx \frac {1}{r} \biggl ( 1+\frac {l}{2r}\cos \theta \biggr ) \\ \frac {1}{|{\bf r}_-|} &\approx \frac {1}{r} \biggl ( 1- \frac {l}{2r}\cos \theta \biggr )

(3.3c)

where we have used (1 + δ)n ≃ 1 + nδ to first order in δ. Note that this is only valid when r >> l. Now,
\frac {1}{|{\bf r}_+|} - \frac {1}{|{\bf r}_-|} = \frac {l}{r^2}\cos \theta ,
(3.4)

so the potential at the test point is finally given by:

Dipole potential:
\textcolor {purple}{V({\bm r}) = \frac {ql \cos \theta }{4\pi \epsilon _0 r^2} = \frac {{\bm p} \cdot \hat {{\bm r}}}{4\pi \epsilon _0 r^2} }, (3.5)

where p is the dipole moment, defined as:

{\bm p}&= q{\bm l} \\ {\bm p}&= \int _\Vol {\bm r} \, {\rho }({\bm r})\, d\Vol

(3.6b)

where the second form is for the dipole moment of a charge density distribution in a small volume V. Now that
we have the potential we can calculate the field. Notice that we have naturally ended up working in spherical
polar coordinates (with θ defined as the angle from the z-axis), and that there is no dependence on azimuth φ
for the single dipole.

Now let’s go back to the single dipole potential we derived, and work in spherical polar coordinates to obtain
the corresponding components of the electric field:

{\bf E} ({\bf r}) &= -{\bf \nabla } V = -{\bf \hat {r}}\frac {\partial V}{\partial r} -{\bf \hat {\theta }}\frac {1}{r}\frac {\partial V}{\partial \theta } -{\bf \hat {\phi }}\frac {1}{r \sin \theta }\frac {\partial V}{\partial \phi } \\ \frac {\partial V}{\partial r} &= -2 \frac {ql \cos \theta }{4\pi \epsilon _0 r^3} \\ \frac {\partial V}{\partial \theta } &= -\frac {ql \sin \theta }{4 \pi \epsilon _0 r^2} \\ \frac {\partial V}{\partial \phi } &= 0

(3.7d)

Since φ here is an azimuthal angle with respect to the natural axis of symmetry of the dipole charge pair,
derivatives with respect to φ must vanish.

Single Dipole Field and Potential

E_r (r,\theta ,\phi ) &= \frac {ql \cos \theta }{2\pi \epsilon _0 r^3} \\ E_{\theta } (r,\theta ,\phi ) &= \frac {ql \sin \theta }{4\pi \epsilon _0 r^3}

(3.8b)

PHAS0038: Term 1 Atomic Mechanisms 5

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

We note that:
• The potential is zero for θ = π/2 (xy plane)
• The field decays as 1/r3 , potential decays as 1/r2
• As always, the electric field lines are always perpendicular to equipotential curves (see Figure below)

Figure 3.2: Electric field and equipotentials for a single dipole.

Multipole Potential Expansion

Let’s now the more general case of the electrostatic potential at a point r due to an arbitrary charge distribution,
with density ρ(r ′ ). This can be written as:

V({\bm r}) = \frac {1}{4\pi \epsilon _0} \int \frac {\rho ({\bm r}')}{|{\bm R}|}d{\Vol ^{\prime }}. (3.9)

where R = r − r ′ is the vector running from the variable source point to our test point for evaluating the
potential. In order to make this simpler, we can derive the magnitude of R from a dot product:

{\bm R} \cdot {\bm R} &= ({\bm r} - {\bm r}') \cdot ({\bm r} - {\bm r}') \\ &= r^2 + (r')^2 -2rr' \cos \theta
(3.10b)

where θ is the variable angle between position vectors r and r ′ . By taking a factor of r2 outside, we see that
we can write:

R^2 &= r^2 \biggl ( 1+ \biggl ( \frac {r'}{r} \biggr )^2 - 2\biggl ( \frac {r'}{r} \biggr ) \cos \theta \biggr ) \\ \Longrightarrow R &= r\sqrt {1+\epsilon } \\ \epsilon &= \biggl (\frac {r'}{r} \biggr ) \biggl ( \biggl ( \frac {r'}{r}\biggr ) - 2\cos \theta \biggr )

(3.11c)

PHAS0038: Term 1 Atomic Mechanisms 6

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

So we can expand 1/R using the binomial expansion, as follows:

\frac {1}{R} &= \frac {1}{r}(1+\epsilon )^{-1/2}=\frac {1}{r} \biggl [ 1-\frac {1}{2}\epsilon + \frac {3}{8} \epsilon ^2 - \frac {5}{16} \epsilon ^3 + ... \biggr ] \\ &= \frac {1}{r} \biggl [ 1-\frac {1}{2}\biggl ( \frac {r^{\prime }}{r} \biggr ) \biggl ( \biggl ( \frac {r^{\prime }}{r} \biggr ) -2 \cos \theta \biggr ) + \frac {3}{8} \biggl ( \frac {r^{\prime }}{r} \biggr )^2 \biggl ( \biggl ( \frac {r^{\prime }}{r} \biggr ) -2\cos \theta \biggr )^2 + ... \biggr ]

(3.12b)

r′


Now gathering terms which have common powers in r , this can be written:

\frac {1}{R} &= \frac {1}{r} \biggl [ 1+\biggl ( \frac {r^{\prime }}{r} \biggr )\, \cos \theta + {\biggl ( \frac {r^{\prime }}{r} \biggr )^2} \, \frac {3 \cos ^2 \theta -1}{2} + \biggl ( \frac {r^{\prime }}{r} \biggr )^3 \, \frac {5 \cos ^3 \theta -3\cos \theta }{2} + ... \biggr ] \\ &= \frac {1}{r} \sum ^{\infty } _0 \biggl ( \frac {r^{\prime }}{r} \biggr )^n P_n (\cos \theta )

(3.13b)

where Pn (cos θ) represent the Legendre polynomials. Substituting this expression into the integral for the
potential, we find:

V({\bm r}) &= \frac {1}{4\pi \epsilon _0} \biggl [ \frac {1}{r} \int \rho ({\bm r}^{\prime }) \, d\Vol ^{\prime } + \frac {1}{r^2} \int r^{\prime } \cos \theta \rho ({\bm r}^{\prime }) \, d\Vol ^{\prime } + \frac {1}{r^3} \int (r^{\prime })^2 \biggl ( \frac {3}{2} \cos ^2 \theta -\frac {1}{2}\biggr ) \rho ({\bm r}^{\prime }) \, d\Vol ^{\prime } + ... \biggr ] \\ &= \frac {1}{4\pi \epsilon _0} \sum _0 ^{\infty } \frac {1}{r^{n+1}} \int (r^{\prime })^n P_n (\cos \theta ) \rho ({\bm r}^{\prime }) \, d\Vol ^{\prime }

(3.14b)

This shows that for large distances, an arbitrary charge distribution behaves approximately like the total charge
concentrated at a point (the first term, which falls off with 1/r, is known as the monopole term). At shorter
and shorter distances, the other terms become important and can be brought in to improve the approximation.
The dipole term in this expansion of the potential scales with 1/r2 , the quadrupole term scales with 1/r3 , the
octupole term with 1/r4 etc.
• Ultimately, we want to understand the response of matter to applied fields, i.e. how microscopic charges
and currents ‘reconfigure’.
• We derived the potential and electric field for a single dipole
• The dipole potential falls off as 1/r2 , the dipole field as 1/r3
• We have looked at the potential of an arbitrary charge distribution
• Expanding 1/R we then made the multipole expansion (general case)

Microscopic Dipoles

• Dielectrics in their original form have no free charges

• Atoms consist of nuclei (positive ‘core’) and electrons (negative ‘cloud’) which can respond to an applied
external E field
• Positive charge shifts in the direction of the field (by definition), negative charge moves against it
• This microscopic displacement is ultimately limited by a restoring force due to the charge separation
• This results in a material which has a net non-zero dipole moment, but which is electrically neutral on
adequately large scale

PHAS0038: Term 1 Atomic Mechanisms 7

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Polarized Dielectric

• Consider a small volume of a dielectric

Figure 3.3: A piece of unpolarized dielectric.

• Apply a field - there is a net displacement of distance a (see Figure below):

Figure 3.4: A piece of polarized dielectric.

Polarization per unit volume

• For a distribution of charge density ρ, an infinitesimal charge element within that distribution can be
written as dq = ρ(r) dV

• For a piece of dielectric material of volume ∆V ′ , the total dipole moment is ∆p = ∆V ′ P(r′ ) dV ′ , where
R

P represents a macroscopic dipole moment per unit volume.

• How can we relate this quantity P to the pattern of induced charge densities?

• What is the electric field which arises from these charge distributions in / on a polarized dielectric?

• There are two approaches: one relatively simple (e.g. Grant & Phillips ‘Electromagnetism’), and one more
rigorous. We consider the rigorous one first. In both cases, we must take care about using the expression for
potential outside a dielectric and inside the dielectric material. Specifically, we must average over some volume
which is:

• (i) large enough that the effects (‘fluctuations’) of individual electrons and ions are ‘smoothed out’; and

• (ii) small compared to the typical scales of the object considered.

PHAS0038: Term 1 Atomic Mechanisms 8

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

This type of ‘averaging’ approach gives the macroscopic field.

• Note that the potential and field outside a polarized dielectric (which we derived earlier) also rely on this
averaging, but we are far enough away that the ‘fine details’ of the all the microscopic dipoles don’t need to be
considered. It turns out that this same potential can be used, and we will outline a justification here (see Griffiths
section 4.2.3 for a detailed discussion). Along the way, we need to provide some results related to averaged fields
(given here without proof). In both cases we need the averaged potential over an appropriately sized sphere of
radius R.

Figure 3.5: Looking at a Sphere inside a Dielectric.

The macroscopic field at the centre of our imaginary sphere is the sum of: (i) The averaged field due to all of
the dipoles outside the sphere; and (ii) The averaged field due to all of the dipoles inside the sphere.

1. The potential, averaged over a spherical volume, due to a polarized dielectric outside the sphere is identical
to the total potential produced at the centre of the sphere by those outside charges 1 (the same is true for
the field).
Thus we can use a formula for this potential, similar to the one we have already seen:

{\langle V_{out} ({\bf r_C}) \rangle } = \frac {1}{4\pi \epsilon _0} \int _{\Vol , outside} \frac { ({\bf r_C}-{\bf r'}) \cdot {\bf P}({\bf r'})}{|{\bf r_C}-{\bf r'}|^3} \, d\Vol \label {eq:Vout} (3.15)

Here, rC is the position of the centre of the sphere. The radius R of the sphere is chosen to be large enough
so that we can use the ‘far field’ approximation for the potential due to all the individual dipoles lying
outside the sphere.

2. What about the field from the charge inside the sphere? Many of the dipoles inside our sphere will be too
close to the sphere’s centre to use the ‘dipole far-field’ potential, but fortunately all we need is the average
field over the sphere, due to the inside dipoles.
Here we can make use of the following useful result (given without proof): the averaged field, over the
sphere, due to all of the inside charges qj (see also Griffiths’ discussion, sec. 4.2.3), is given by:

{\langle {\bf E} _{in}\rangle } = -\frac {1}{4\pi \epsilon _0 R^3} \sum _j q_j {\bf r}_j, \label {eq:Ein} (3.16)

where rj denotes charge position relative to the centre of the sphere.

We can recognize that the summation in this expression is related to the electric dipole moment generated
by the entire collection of interior charges qj , so that:

{\langle {\bf E} _{in}\rangle } \approx -\frac {1}{4\pi \epsilon _0 R^3} \, {\bf P} \, \Vol _S, \label {eq:Ein2} (3.17)

1 See Griffiths, section 3.1.4 - excerpt pages are on Moodle

PHAS0038: Term 1 Atomic Mechanisms 9

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

where P VS denotes the total dipole moment of all the charge inside the sphere, which is the product of
the polarization (dipole moment per unit volume), P, and the volume of the sphere, VS = 43 πR3 . The
assumption we are making here is that R is large enough for us to calculate meaningful averages, but small
compared to the length scale over which P varies.
Developing this a little more:
{\langle {\bf E} _{in}\rangle } \approx -\frac {1}{4\pi \epsilon _0 R^3} \, {\bf P} \, \frac {4}{3} \pi R^3 = \frac {-{\bf P}}{3\epsilon _0}. \label {eq:Ein3} (3.18)

This final expression is actually equal to the electric field at the centre of an isolated, uniformly polarized
sphere, due to the charges that reside on that sphere.

• Macroscopic Field: Now, the term which is omitted from the integration in equation 3.15 does, in fact,
correspond to the field at the centre of our uniformly polarized, small spherical volume of radius R. But this is
exactly the field represented in equation 3.18. Thus it follows that the total macroscopic, or volume-averaged,
field, for our spherical volume, can be calculated from an integral of the same general form as equation 3.15,
but which runs over the entire volume filled with dielectric material. Thus we can identify the macroscopic
(averaged) field over our sphere with the full integral expression which we would use to compute field at the
centre of the sphere due to all of the charges.
This is a comforting result, because this is precisely the integral we have used previously to consider the interior
field of a dielectric medium.
• Now for a more simple demonstration. Consider a small block of material of size δx δy δz, located at (x, y, z).
Since the polarization, P, is defined as the dipole moment per unit volume, the amount of charge which accumu-
lates at x − δx/2 must be −Px (x − δx/2) δy δz, and at x + δx/2 it is Px (x + δx/2) δy δz. The net charge which
must reside inside our small cube, due to the polarization in the x-direction, must therefore be:

-(P_x(x+\delta x / 2)\delta y \delta z - P_x(x - \delta x / 2)\delta y \delta z ) = -\frac {\partial P_x}{\partial x} \delta x \delta y \delta z (3.19)

We can write similar equations for the y and z directions, and find that the total charge is given by:
\biggl ( -\frac {\partial P_x}{\partial x} -\frac {\partial P_y}{\partial y} -\frac {\partial P_z}{\partial z} \biggr ) \delta x \delta y \delta z
(3.20)

If we divide this charge by the volume element δxδyδz then we find an effective polarization charge density:
\rho _P = - {\bf \nabla } \cdot {\bf P} (3.21)

In uniformly polarized, bulk dielectrics this will tend to zero as the number of charges entering and leaving will be
the same; near discontinuous surfaces or areas where the polarization varies rapidly, a significantly non-zero net
charge density will develop. These two effects are now described further. This derivation has already appeared
in ‘Macroscopic Fields’, but is simply repeated here for convenience. For the more rigorous demonstration, we
start by finding the potential at a point r due to a small volume of polarized material at a point r′ . We will then
integrate this over the entire piece of dielectric material (we are justified in doing this because of the ‘macroscopic’
description).

\delta \varphi ({\bf r}) &= \frac {\delta {\bf p} ({\bf r}^{\prime })\cdot ({\bf r-r}^{\prime })}{4\pi \epsilon _0 |{\bf r-r}^{\prime }|^3} \\ &= \frac {\delta \Vol ^{\prime } \, {\bf P} ({\bf r}^{\prime })\cdot ({\bf r-r}^{\prime })}{4\pi \epsilon _0 |{\bf r-r}^{\prime }|^3}

(3.22b)

When we take the limit δV ′ → 0 and sum over the elements, we get an expression for the total potential:

\varphi ({\bf r}) = \int _{\Vol '} \frac {d\Vol ^{\prime }\, {\bf P} ({\bf r}^{\prime })\cdot ({\bf r-r}^{\prime })}{4\pi \epsilon _0 |{\bf r-r}^{\prime }|^3} (3.23)

We use the gradient of 1/|r − r′ | to transform this:

\varphi ({\bf r}) = \frac {1}{4\pi \epsilon _0} \int _{\Vol '}{\bf P} ({\bf r}^{\prime })\cdot {\bf \nabla } ^{\prime } \biggl ( \frac {1}{|{\bf r-r}^{\prime }|} \biggr ) d\Vol ^{\prime } (3.24)

PHAS0038: Term 1 Atomic Mechanisms 10

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Using the formula for ∇ · (φF) from the Mathematical Identities, and rearranging (we want F · ∇φ) we can write:

\varphi ({\bf r}) = \frac {1}{4\pi \epsilon _0} \int _{\Vol '} \biggl [ {\bf \nabla } \cdot \biggl ( \frac {\bf P({\bf r}^{\prime })}{|{\bf r-r}^{\prime }|} \biggr ) - \frac {1}{|{\bf r-r}^{\prime }|} {\bf \nabla \cdot P}({\bf r}^{\prime }) \biggr ] d\Vol ^{\prime } (3.25)

Finally, we use the divergence theorem on the first term to give the potential outside a polarized dielectric object:

\varphi ({\bf r}) = \frac {1}{4\pi \epsilon _0} \oint _{S'} \frac {\bf P({\bf r}^{\prime })\cdot {\bf n}}{|{\bf r-r}^{\prime }|} dS^{\prime } + \frac {1}{4\pi \epsilon _0} \int _{\Vol '} \frac { - {\bf \nabla \cdot P}({\bf r}^{\prime }) }{|{\bf r-r}^{\prime }|} d\Vol ^{\prime } (3.26)

Polarization Charge Densities

• The surface polarization charge density is defined:

\sigma _P = {\bf P \cdot n} (3.27)

• The volume polarization charge density is defined:

\rho _P = - {\bf \nabla \cdot P} (3.28)

• We can write the potential as:

\varphi ({\bf r}) = \frac {1}{4\pi \epsilon _0} \biggl ( \oint _{S'} \frac {\sigma _P}{|{\bf r-r}^{\prime }|} dS^{\prime } + \int _{\Vol '} \frac { \rho _P}{|{\bf r-r}^{\prime }|} d\Vol ^{\prime }\biggr ) = \frac {1}{4\pi \epsilon _0} \int \frac {dq_P}{|{\bf r-r}^{\prime }|} (3.29)

Does this mean that our dielectric is now charged? To understand this, consider a polarized dielectric, with
volume V0 . We must also consider the outside surface S0 , and its associated polarization surface charge density.
Using the divergence theorem we can write for the total bound charge QP :

Q_P &= \int _{V_0} \rho _P \, d\Vol + \oint _{S_0} \sigma _P \, dS \\ &= \int _{V_0} - {\bf \nabla \cdot P}\,d\Vol + \oint _{S_0} {\bf P \cdot n} \, dS \\ &= -\oint _{S_0} {\bf P \cdot n} \, dS + \oint _{S_0} {\bf P \cdot n} \, dS = 0

(3.30c)

So, when only bound charge (dipoles) is present, the overall dielectric is electrically neutral (which we assumed
at the start). However, the field inside the dielectric can, of course, be non-zero: in particular, if there is an
applied external field inducing the polarization, then the dielectric itself will affect that field. For completeness,
note that we can write:

{\bf E} ({\bf r}) = \frac {1}{4\pi \epsilon _0} \biggl ( \oint _S \frac {\sigma _P(\bf {r^{\prime }})\,({\bf r -r}^{\prime })}{|{\bf r-r}^{\prime }|^3} \, dS^{\prime } + \int _V \frac {\rho _P(\bf {r^{\prime }})\,({\bf r -r}^{\prime })}{|{\bf r-r}^{\prime }|^3} \, d\Vol ^{\prime } \biggr ) (3.31)

• Polarization arises from alignment of microscopic dipoles

• These behave collectively to give us surface and volume polarization charge densities

• We have considered the form of potential both inside and outside dielectric media

PHAS0038: Term 1 Atomic Mechanisms 11

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

P
z axis
rP
z

r’
x axis P’ R
!′
!

3.2 Magnetic Dipole

Geometry
• Current loop, radius a, carrying current I, centred at the origin of our coordinate system
• We consider the magnetic induction (B field) at a test point, P
• Using cylindrical polar coordinates, P is located at (R, φ, z) (see Figure above)
• The loop can be considered as the superposition of many small elements, each of length dl′ . A single
element is located at a variable point P ′ , with coordinates (a, φ′ , 0), where the azimuth φ′ runs from 0 to
2π thus ‘tracing out’ the entire loop.
• The vector running from the ‘source’ element dl′ to the test point P is denoted rP − r ′ .

There are two ways to derive the magnetic field produced our current loop:

• Use a multipole expansion and approximate the loop by its dipole moment ;
• A ‘fuller’ derivation which develops the vector potential of the loop in terms of elliptic integrals. An
approximate form of this result is identical to that from the ‘dipole’ approach.

Multipole Expansion for Arbitrary Current Loop

In the Macroscopic Fields part of the course, we showed that, using the Coulomb gauge, the vector potential at
a point r due to a current density J (r ′ ) distributed over a volume V ′ could be written:

{\bm A} ({\bm r}) = \frac {\mu _0}{4\pi } \int _{\Vol } \frac { {\bm J} ({\bm r}^{\prime })}{|{\bf r-r}^{\prime }|} \, d\Vol ^{\prime } (3.32)

For a constant current I in an arbitrary loop, we can make the substitution J dV ⇒ I dl, and thus rewrite the
volume integral as a line integral. Re-introducing now the source-to-test-point vector R = r − r′ , we write:

{\bm A} ({\bf r}) = \frac {\mu _0 I}{4\pi } \int _C \frac { d {\bm l}^{\prime }}{|{\bm R}|} (3.33)

As with the potential of a charge distribution, we now need to write the distance R in terms of r and r′ . Using
Legendre polynomials again, and recalling that the angle between r and r′ is θ, we find:

{\bm A} ({\bm r}) &= \frac {\mu _0 I}{4\pi } \sum ^{\infty } _0 \frac {1}{r^{n+1}} \oint ({\bm r}^{\prime })^n P_n (\cos \theta ) \, d{\bm l}^{\prime } \\ &= \frac {\mu _0 I}{4\pi } \biggl [ \frac {1}{r} \oint d{\bm l}^{\prime }+ \frac {1}{r^2} \oint r^{\prime }\cos \theta \, d{\bm l}^{\prime }+ \frac {1}{r^3} \oint r^{\prime ^2} \biggl ( \frac {3}{2} \cos ^2 \theta - \frac {1}{2} \biggr )\, d{\bm l}^{\prime } + ...\biggr ]

(3.34b)

PHAS0038: Term 1 Atomic Mechanisms 12

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Now notice that the first term (the monopole term) is multiplied by a closed loop integral with integrand 1,
which is identically zero (it is not surprising that the monopole term disappears as our theory does not support
magnetic monopoles). So the first non-zero term in the expansion is our dipole term; now use the identity
\oint r ^{\prime } \cos \theta \, d{\bm l} ^{\prime } = \oint ({\bm \hat {{\bm r}} \cdot {\bm r}}^{\prime }) d{\bm l} ^{\prime } = - {\bm \hat {{\bm r}}} \times \int d{\bm S}^{\prime }
(3.35)

∇′ T × dS ′ = − T dl′ using the

R H
to simplify this. (You can derive this identity from the more general form S′ C
scalar function T = r̂ · r ′ ).
Then the dipole term only becomes:

{\bf A}_{dip} ({\bf r}) = \frac {\mu _0 I}{4\pi r^2} \oint r^{\prime } \cos \theta \, d{\bf l}^{\prime } = \frac {\mu _0 I}{4\pi } \frac {\int d{\bf S}^{\prime } \times {\bf \hat {r}}}{r^2} = \frac {\mu _0}{4\pi } \frac {{\bf m} \times {\bf \hat {r}}}{r^2} (3.36)

where m = I dS ′ = I S ′ is the magnetic dipole moment (note that the current flow direction around the loop
R

and the direction of the surface vector S ′ are defined in a ‘right-handed’ sense). Our derivation now allows a full
expansion to be made for an arbitrary current loop; but sufficiently far from our loop, it is the dipole behaviour
which dominates the magnetic potential. We will encounter this form of the vector potential again.
We will next consider another derivation of the magnetic potential due to a circular current loop at the origin.

Let us continue with a specific derivation of the vector potential for our circular current loop. We can write the
vector representing any small loop element, dl′ (see diagram above):

d{\bm l'} = d\phi ^{\prime } \, a \, (-\sin \phi ^{\prime } {\bm i} + \cos \phi ^{\prime } {\bm j}) (3.37)

Here we’ve used the standard two-dimensional formula for arc length, and projected it onto Cartesian vectors.
Note that we want the direction of dl′ to be tangential to the current loop; so at φ′ = 0 it lies along the +y-
direction, at φ′ = π/2 it lies along the −x-direction, etc. We’ve basically taken the differential of the vector
position on the unit circle. We now consider the magnetic induction due to the current loop. One approach
would be to use the Biot-Savart Law, and integrate around the loop, but this quickly becomes very complicated.
Instead, we will use the concept of the vector potential; this isn’t trivial, but it’s easier, and allows us to get
further with our derivation. First we notice that, as before, we can substitute elements J dV ′ ⇒ Idl′ for a
constant current through the loop.

Vector Potential of Current Loop

• We can write for the vector potential at point P (position rP ):

{\bf A}({\bm r_P}) &= \frac {\mu _0 }{4\pi } \int _{\Vol '} \frac {{\bm J({\bm r'})}}{|{\bm r_P}-{\bm r'}|}\,d{\Vol '} \\ &= \frac {\mu _0 I}{4\pi } \oint \frac {d{\bm l'}}{|{\bm r_P}-{\bm r'}|}.

(3.38b)

• The contribution to the vector potential is in the direction of the current element.

Using our knowledge of the geometry and symmetry of the system, we can deduce that the vector potential
only has an azimuthal component at the test point P (i.e. the component which follows the local direction of
increasing azimuth φ), Aφ . (Think about ‘pairs’ of loop elements, where each member of the pair is at the
same distance from P - and consider how the vector contributions from each one would add together.) We can
therefore now write Aφ = A · iφ with:
{\bm i}_{\phi } = -\sin \phi {\bm i} +\cos \phi {\bm j}. (3.39)

PHAS0038: Term 1 Atomic Mechanisms 13

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

So we can write:

{\bf A}_{\phi } &= {\bm A} \cdot {\bm i}_{\phi } \\ &= \frac {\mu _0 I}{4\pi } \oint \frac {d{\bm l'} \cdot {\bf i}_{\phi }}{|{\bm r_P}-{\bm r'}|}.

(3.40b)

Note now that dl′ · iφ = a δφ′ (sin φ sin φ′ + cos φ cos φ′ ) = a δφ′ cos(φ′ − φ). Hence:

A_{\phi } &= \frac {\mu _0 I\,a}{4\pi } \int _0 ^{2\pi } \frac {\cos (\phi ^{\prime } - \phi )}{|{\bf r_P}-{\bm r'}|}\,d\phi ^{\prime } (3.41a)

This now seems to be quite an easy integral to evaluate. Using the cylindrical symmetry of the system, we know
that Aφ must be independent of the value of φ, so we can evaluate the integral for any convenient choice of φ;
we’ll take φ = 0. However, we mustn’t forget that |rP − r ′ | must be written in terms of the quantities R (here
we now use the symbol R to represent the cylindrical radial distance to the observer / field point), z, a, φ and
φ′ :

|{\bm r}| = |{\bm r_P}-{\bm r'}| &= \sqrt { ({\bm r_P}-{\bm r'}) \cdot ({\bm r_P}-{\bm r'}) } \\ &= \sqrt {(R\cos \phi - a\cos \phi ^{\prime })^2 +(R \sin \phi - a \sin \phi ^{\prime })^2+z^2} \\ A_{\phi } &= \frac {\mu _0 Ia}{4\pi } \int _0 ^{2\pi } \frac {\cos (\phi ^{\prime } - \phi )\, d\phi ^{\prime }} {[(R\cos \phi - a\cos \phi ^{\prime })^2 +(R \sin \phi - a \sin \phi ^{\prime })^2+z^2]^{1/2}}

(3.42c)

Setting φ = 0, we can simplify somewhat:

|{\bm r}| &= \sqrt {(R - a\cos \phi ^{\prime })^2 +( - a \sin \phi ^{\prime })^2+z^2} \\ &= \sqrt {R^2-2aR\cos \phi ^{\prime } +a^2 \cos ^2 \phi ^{\prime } +a^2 \sin ^2 \phi ^{\prime } +z^2} \\ &= \sqrt {R^2 +a^2 +z^2 -2aR\cos \phi ^{\prime }} \\ A_{\phi } &= \frac {\mu _0 Ia}{4\pi } \int _0 ^{2\pi } \frac {\cos \phi ^{\prime } d\phi ^{\prime }} {[R^2 +a^2 +z^2 -2aR\cos \phi ^{\prime }]^{1/2}}

(3.43d)

This expression, while somewhat complex, can be written in terms of special mathematical functions called
elliptical integrals which have been tabulated, and are implemented in packages such as Mathematica. However,
with one approximation we can find simpler expressions.

|{\bf r}| = \sqrt {R^2 +a^2 +z^2} \biggl ( 1- \frac {2aR\cos \phi ^{\prime }}{R^2+a^2+z^2} \biggr ) ^{1/2} (3.44)

Let us make the approximation that 2aR | cos φ′ | << R2 + a2 + z 2 . This is fullfilled if

• R2 + z 2 >> a2 : far from the current loop

• a2 + z 2 >> R2 : close to the axis

These are two important cases. Now we can write:

|{\bf r}|^{-1} &= (R^2+a^2+z^2)^{-1/2} \biggl ( 1- \frac {2aR\cos \phi ^{\prime }}{R^2+a^2+z^2} \biggr ) ^{-1/2} \\ &\approx (R^2+a^2+z^2)^{-1/2} \biggl ( 1+ \frac {aR\cos \phi ^{\prime }}{R^2+a^2+z^2} \biggr ) \\ A_{\phi } &\approx \frac {\mu _0 Ia}{4\pi } \int _0 ^{2\pi } \frac {\cos \phi ^{\prime } \, d\phi ^{\prime }} {(R^2 +a^2 +z^2)^{1/2}} \biggl ( 1+ \frac {aR\cos \phi ^{\prime }}{R^2+a^2+z^2} \biggr )

(3.45c)

PHAS0038: Term 1 Atomic Mechanisms 14

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

But we can simplify this, using some basic integrals:

\int _0 ^{2\pi } \cos \phi ' \, d \phi ' &= 0 \\ \int _0 ^{2\pi } \cos ^2 \phi ' \, d \phi ' &= \int _0 ^{2\pi } \frac {1}{2} (1+\cos 2 \phi ' \, d \phi ' ) = \biggl [ \frac {1}{2} \phi ' \biggr ]^{2\pi } _0 = \pi

(3.46b)

Form of A
• Far from the dipole (or near the axis), we can write:

A_{\phi } = \frac {\mu _0 I}{4} \frac {a^2 R}{(R^2+a^2+z^2)^{3/2}} (3.47)

• But spherical polars are also convenient.

√
• For the case R2 + z 2 >> a2 , we have |r| ≈ R2 + z 2 and sin θ ≈ R/r. Hence:

A_{\phi } &= \frac {\mu _0 I a^2}{4} \frac { R}{(R^2+z^2)^{3/2}} \\ &= \frac {\mu _0 I a^2}{4} \frac { \sin \theta }{r^2}

(3.48b)

Now that we have the vector potential, we can recover the magnetic induction, B. We write:

{\bf B} ={\bf \nabla \times A} = \frac {1}{ r^2 \sin \theta } \Biggl | \begin {matrix} {\bf i}_r & r{\bf i}_{\theta } & r\sin \theta {\bf i}_{\phi } \\ \frac {\partial }{\partial r} & \frac {\partial }{\partial \theta } & \frac {\partial }{\partial \phi } \\ A_r & rA_{\theta } & r\sin \theta A_{\phi } \end {matrix} \Biggr | (3.49)

But we know that A only has a component in the φ direction, which simplifies things considerably! We can
write the different components of B as follows:

B_r &= \frac {1}{r^2 \sin \theta } \biggl [ \frac {\partial }{\partial \theta } (r \sin \theta A_{\phi }) \biggr ] \\ &= \frac {1}{r}\frac {\partial A_{\phi }}{\partial \theta } + \frac {1}{r^2 \sin \theta } r\cos \theta A_{\phi } \\ &= \frac {\mu _0 Ia^2}{4 } \biggl [ \frac {\cos \theta }{r^3} +\frac {\cos \theta }{\sin \theta }\frac {\sin \theta }{r^3} \biggr ] \\ &= \frac {\mu _0 Ia^2}{4 } \biggl [ \frac {2 \cos \theta }{r^3} \biggr ]

(3.50d)

B_{\phi } &= 0 \\
(3.50f)

B_{\theta } &= \frac {r}{r^2 \sin \theta } \biggl [ - \frac {\partial }{\partial r} (r \sin \theta A_{\phi }) \biggr ] \\ &= \frac {1}{r \sin \theta } \biggl [ -\sin \theta A_{\phi } -r \sin \theta \frac {\partial A_{\phi }}{\partial r} \biggr ] \\ &= \frac {-\mu _0 Ia^2}{4} \biggl [ \frac {\sin \theta }{r^3} -2 \frac {\sin \theta }{r^3} \biggr ] \\ &= \frac {\mu _0 Ia^2}{4} \biggl [ \frac {\sin \theta }{r^3} \biggr ] \\

(3.50k)

PHAS0038: Term 1 Atomic Mechanisms 15

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Form of Magnetic Field

• We find for the components of B far from the dipole:

Components of B:

B_r &= \frac {\mu _0 Ia^2}{2 } \frac {\cos \theta }{r^3} \\ B_{\phi } &= 0 \\ B_{\theta } &= \frac {\mu _0 Ia^2}{4} \frac {\sin \theta }{r^3}

(3.51c)

• These have identical dependence on θ to the electric dipole in the ‘far field’ region.

• Close to the electric and magnetic dipole, however, the respective fields strongly differ.

Figure 3.6: Illustration of field lines for an electric dipole (left) and for a magnetic current loop (right). The
plane of the current loop is perpendicular to the page, so that it would be coming out and going into the page.

• The electric dipole consists of two charge aligned along the dipole axis, while the magnetic dipole consists of a
current loop lying in the plane perpendicular to the dipole axis. Field lines for the electric dipole start and end
on the charges, while the field lines for the magnetic dipole form closed loops.

3.3 Magnetic Dipoles and Magnetization

As we saw with the electric field, E, the introduction of a medium other than vacuum results in ‘modifications’
to Maxwell’s equations. These are implemented using alternative vector fields, which implicitly include the
effects of the medium. We derived various equations for the electric displacement D, using a dipole model of the
polarizability of the atoms (or molecules) making up the medium. We will now apply the same ideas to magnetic
media. A word of caution: non-linear magnetic media are much more common than non-linear electric media;
we will deal with these rather interesting materials in more detail in our next section, ‘Ferromagnetism’.

Magnetization

• We introduced the polarizability of a dielectric material, P ∝ E

• Similarly, we introduce a quantity, proportional to the magnetic induction B

PHAS0038: Term 1 Atomic Mechanisms 16

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

• This is the magnetization, M (magnetic dipole moment per unit volume)

• It describes the response of a material to the magnetic induction
• Electrons move in loops around atoms in a ‘classical’ picture: we can use the magnetic dipole generated
by this kind of motion to model the response

Microscopic Origin

• Consider a small piece of a material of volume ∆V

• If mi is the magnetic dipole moment due to the ith atom in ∆V , we define:

{\bf M} = \lim _{\Delta \Vol \rightarrow 0} \frac {1}{\Delta \Vol } \sum _i {\bf m}_i (3.52)

• This is analogous to the polarization (electric dipole moment per unit volume)
• With no external field, the directions of the mi are random and so M ≈ 0
• We will now consider the magnetization currents arising from dipoles

Approach

• We will consider a given body made up from adjacent current loops

• We find three simple limits:
1. Uniformly magnetised bulk: No net magnetization current
2. Non-uniformly magnetised bulk: Non-zero volume magnetization current
3. Uniformly magnetised slab: Surface magnetization current

Surface Magnetization Current

• Consider a slab as shown in Figure 3.7 of thickness t and cross-section area A.

• The slab has uniformly distributed (‘microscopic’) magnetic dipoles.

• If there is a net magnetization M, then, given that the volume of the sample is St, the expected magnetic
field (at large distances) is the same as would arise from a dipole of magnitude M (S t). Let the magnetization
be perpendicular to the main surface of the slab. Now consider a small strip of current loops, ABCD, shown in
the bottom of Fig. 3.7. For each loop with a component to the right, there is an equal and opposite loop with
a component to the left; for the strip PQRS, for each current going up there is an equal and opposite current
going down. From a distance which is large compared to the current loops, these will cancel out (this can be
shown with the Biot-Savart Law).
• However, at the ‘exposed’ boundary of the material there are no loops to ‘cancel out’ the edge loops. This will
give us a surface magnetization current density jM , analogous to surface polarization charge density in a dielectric.
We know that the total dipole moment is (S t) M. The surface magnetization current has a magnitude of |jM | t.

PHAS0038: Term 1 Atomic Mechanisms 17

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

But a current loop has dipole moment of magnitude IS, so we know that StM = IS, and |jM | = I/t = M . As
jM is perpendicular to the magnetization, we can write (using the outward unit normal n):

{\bf j}_M = {\bf M} \times {\bm n} (3.53)

Figure 3.7: Uniformly magnetised slab.

Volume Magnetization Current

• Consider the strip shown with dashed lines in Figure 3.8

• The net downward current is larger than the net upward current (due to the non-uniform magnetization)

• There will thus be a net downward magnetization current Jm

• Let us now make a more rigorous derivation of the relationship between the magnetization and the volume
magnetization current density. Consider two small, adjacent volumes in a piece of non-uniformly magnetised
material. They have dimensions dx, dy, dz and are located at (x, y, z) and (x, y + dy, z). The magnetization
in the first element will be taken as M(x, y, z), and we assume that the material behaves linearly. Then the

PHAS0038: Term 1 Atomic Mechanisms 18

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

magnetization in the second material can be written:

{\bf M}(x,y+dy,z) = {\bf M}(x,y,z)+\frac {\partial {\bf M}}{\partial y}dy+... (3.54)

• As indicated in the picture below, we will concentrate on the x-component of magnetization, Mx , which arises
from small circulating currents Ic and Ic′ , flowing parallel to the yz plane. Now, Rthe magnetic moment of the
first small element is M dx dy dz, which Hcan be written as Ic dy dz (since m = 12 I r × dl, we need the current
multiplied by the surface area and 2a = r × dl). So we can write for both elements:

M_x \,dx \,dy \, dz &= I_c \, dy \, dz \\ \biggl ( M_x + \frac {\partial M_x}{\partial y} dy \biggr ) \, dx \, dy \, dz &= I^{\prime }_c \, dy \, dz

(3.55b)

Figure 3.8: Non-uniform magnetic material.

Figure 3.9: Two small volumes of magnetised material.

Now the net current flowing in the z-direction on the boundary between the small elements is Ic − Ic′ so:

I_c - I^{\prime }_c &= M_x dx - \biggl ( M_x + \frac {\partial M_x}{\partial y} dy \biggr ) dx \\ &= - \frac {\partial M_x}{\partial y} \, dy \, dx

(3.56b)

PHAS0038: Term 1 Atomic Mechanisms 19

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

There is another contribution to the current flowing in the z-direction in the first element (i.e. the one at (x, y, z))
that comes from a similar consideration involving another small volume element, this time at (x + dx, y, z); this
would be in front of the first element in the figure above. Here we write:

M_y \, dx \, dy \, dz &= I^{\prime \prime }_c \, dx \, dz \\ \biggl ( M_y + \frac {\partial M_y}{\partial x} \, dx \biggr ) \, dx \, dy \, dz &= I^{\prime \prime \prime }_c dx dz

(3.57b)

This time we’re considering the current flowing around the face perpendicular to the y-direction. The net current
in the z-direction this time is given by Ic′′′ − Ic′′ , which gives

I^{\prime \prime \prime }_c - I^{\prime \prime }_c = \frac {\partial M_y}{\partial x} \, dy \, dx (3.58)

From these two contributions to the current, and the knowledge of the surface area of the small element, we can
write the net current density:
J_{M,z} = \frac {\partial M_y}{\partial x} - \frac {\partial M_x}{\partial y} (3.59)

Now this is easily recognised as part of the formula for a curl. We can perform similar calculations for the other
two directions.

Full Result for Volume Magnetization Current Density

• The final result we obtain is, as before:

{\bf J}_M = \biggl ( \frac {\partial M_z}{\partial y} - \frac {\partial M_y}{\partial z} \biggr ) {\bf i} + \biggl ( \frac {\partial M_x}{\partial z} - \frac {\partial M_z}{\partial x} \biggr ) {\bf j} + \biggl ( \frac {\partial M_y}{\partial x} - \frac {\partial M_x}{\partial y} \biggr ) {\bf k} (3.60)

{\bf J}_M = {\bf \nabla \times M} (3.61)

• Again, notice this is the magnetic ‘counterpart’ to ρP = −∇ · P

• We have already considered another way of reaching this same result, using the vector potential, which is
repeated below for convenience.

Let us consider the vector potential at a point r1 due to a small volume of magnetised material at a point r2 .
This small volume will have magnetic moment δm = M (r2 )δV2 . Then we can write:

{\bm A} ({\bm r}_1) &= \frac {\mu _0}{4\pi } \int _V \frac {\delta {\bm m} \times {\bm r}_{12}}{|{\bm r}_{12}|^3} \\ &= \frac {\mu _0}{4\pi } \int _V \frac {{\bm M}({\bm r}_2)\times {\bm r}_{12}}{|{\bm r}_{12}|^3}\,dV_2 \\ &= \frac {\mu _0}{4\pi } \int _V {\bm M}({\bm r}_2)\times {\bf \nabla }_2 \frac {1}{{r}_{12}}\,dV_2

(3.62c)

where we’ve used a standard result in the last passage. Now we use the expansion of ∇ × (φF), with F = M
and φ = r112 to write:

{\bf F} \times {\bf \nabla } \phi &= \phi {\bf \nabla } \times {\bf F} - {\bf \nabla } \times (\phi {\bf F}) \\ {\bm A}({\bm r}_1) &= \frac {\mu _0}{4\pi } \int _V \biggl [ \frac { {\bf \nabla }_2 \times {\bm M}({\bm r}_2)}{r_{12}} - {\bf \nabla }_2 \times \biggl ( \frac {{\bm M}({\bm r}_2)}{r_{12}} \biggr ) \biggr ] dV_2

(3.63b)

PHAS0038: Term 1 Atomic Mechanisms 20

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

R R
Now we use the theorem V
∇ × FdV = S
n × Fda to write:

{\bm A}({\bf r}_1) &= \frac {\mu _0}{4\pi } \int _V \frac { {\bf \nabla }_2 \times {\bm M}({\bm r}_2)}{r_{12}} \, dV_2 - \frac {\mu _0}{4\pi } \int _S \frac { {\bm n} \times {\bm M}({\bm r}_2)}{r_{12}} \, dS_2 \\ {\bm A}({\bf r}_1) &= \frac {\mu _0}{4\pi } \int _V \frac { {\bf \nabla }_2 \times {\bm M}({\bm r}_2)}{r_{12}} \, dV_2 + \frac {\mu _0}{4\pi } \int _S \frac { {\bm M}({\bf r}_2) \times {\bm n} }{r_{12}} da_2

(3.64b)

Magnetization current densities

• We formally define:

Magnetization Current Densities

{\bf J}_M &= {\bf \nabla \times M} \\ {\bf j}_M &= {\bf M \times n}
(3.65b)

• JM is the volume magnetization current density

• jM is the surface magnetization current density

• We now move on to considering how linear magnetic media behave

3.4 Diamagnetism and Paramagnetism

Magnetic Susceptibility

• For a linear, isotropic material, we assert (based on experimental observations):

{\bf M} = \chi _m {\bf H} (3.66)

• χM is the magnetic susceptibility

• We can write B = µ0 (1 + χm )H = µ0 µr H

• If χm > 0 we have a paramagnetic material

• If χm < 0 we have a diamagnetic material

• Note that χm can depend on temperature, but is generally small for these materials (magnitude . 10−5 ,
with exceptions)

PHAS0038: Term 1 Atomic Mechanisms 21

PHAS0038: Electromagnetic Theory Chapter 3 - Atomic Mechanisms

Diamagnetism

• No intrinsic moments (no unpaired electrons): current loops

• With H = 0, the loops are ‘unexcited’ and M = 0
• As |H| increases, |B| increases, so there is more flux through each loop

• From Faraday’s and Lenz’s laws, we can show that a voltage develops to oppose the change in flux
• This results in χm < 0 and B < µ0 H
• These arguments are purely electromagnetic: no T (temperature) dependence

Note: magnetism is subject to quantum mechanical effects

Figure 3.10: M as a function of H and χ as a function of T for a diamagnet and paramagnet.

Paramagnetism

• Some free intrinsic moments (e.g. unpaired electrons)

• With no external field, these are randomly aligned, M = 0 (thermal fluctuation)

• As H increases, they experience a torque aligning them with H

• So χm > 0 and B > µ0 H
• However, this alignment is opposed by the ‘misalignment’ due to random thermal fluctuations
• Thermodynamic analysis gives Curie’s law:

\chi _m = \frac {Nm_0^2 \mu _0}{3kT} (3.67)

(m0 is the magnetic moment on each component of the system)

• There is also often competition between diamagnetic and paramagnetic effects, so χm = N m20 µ0 /3kT +χdia
• At low T , paramagnetic effects dominate

PHAS0038: Term 1 Atomic Mechanisms 22