Chapter 4 Electric Fields in Matter In short, these two types of mechanisms lead to the same result → material becomes

polarized (dipoles point in the direction of field). We use the polarization 𝑃 (≡ dipole
4.1 Polarization moment per unit volume) to measure the effect.
Matter can be divided into two large classes: conductor and insulator (dielectrics).
What are dielectrics? All dielectrics (insulators) charges are attached to specific atoms or 4.2 The Field of a Polarized Object

molecules. → They can just move a bit within the atoms or molecules. But the cumulative What is the field produced by the polarized object? To solve this problem, we can start

effects lead to two kinds of mechanisms: stretching and rotating. 1 ̂

𝑷(𝑟′)∙𝓻 1 ̂
𝓻
with evaluate the potential 𝑉(𝒓) = 4𝜋𝜖 ∫𝜈 𝑑𝜏′ . Using the ∇′ (𝓇 ) = 𝓇 2 notation →
0 𝓇2

Stretching (Induced Dipole) happens when the dielectric is placed in an electric field 𝑬. Atom
1 1 1 1
𝑉 = 4𝜋𝜖 ∫𝜈 𝑷 ∙ ∇′ (𝓇 ) 𝑑𝜏′ − integration by part + divergence theorem→ 𝑉 = 4𝜋𝜖 ∮𝑠 𝓇 𝑷 ∙
as a whole is electrically neutral, consisting of the nucleus and electron. 𝑬 pushed the nucleus 0 0

in the direction of the field and the electron in the opposite direction. There are two kinds
1 1
𝑑𝒂′ − 4𝜋𝜖 ∫𝜈 𝓇 (∇′ ∙ 𝑷)𝑑𝜏′. It just looks like the potential of a surface charge (𝜎𝑏 ≡ 𝑷 ∙ 𝒏
̂)
of results (i) ionizing: the atom being pulled apart completely (ii) reaching the equilibrium: 0

external field = field induced by the apart nucleus and electron. In the second situation, the plus the volume charge (𝜌𝑏 ≡ −𝛁 ∙ 𝑷). To find the field we can first look at those bound
atom is been polarized and has a dipole moment 𝒑 = 𝛼𝑬 in the same direction as 𝑬 . charges (𝜎𝑏 and 𝜌𝑏 ).
Constant 𝛼 is called atomic polarizability, which depends on the structure of the atom. Ex. 4.2. Find the electric field produced by a uniformly polarized sphere (radius = R)
Ex. 4.1. The atomic polarizability of an atom (+q nucleus, -q (radius = a) uniformly charged spherical cloud) 1
Sol) 𝜌𝑏 = 0 since it is uniform. 𝑬 = −∇𝑉 = − 3𝜖 𝑷 inside the sphere. 𝑬 = 4𝜋𝜖
1 𝒑∙𝒓̂
outside the sphere
0 0 𝑟2
1 𝑝
Sol) The field produced by the electron cloud 𝐸𝑒 = 𝐸𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙 → 𝐸𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙 = 4𝜋𝜖 → 𝛼 = 4𝜋𝜖0 𝑎 3
0 𝑎3 → just like a perfect dipole at the origin.
In the case of the molecules, the dipole moment might not be the same direction as 𝑬. → What is bound charge? The bound charges represent the accumulation of charges. Take a
𝑷𝒙 𝛼𝑥𝑥 𝛼𝑥𝑦 𝛼𝑥𝑧 𝑬𝒙
𝒑 = 𝛼⊥ 𝑬⊥ + 𝛼∥𝑬∥ → (𝑷𝒚 ) = (𝛼𝑦𝑥 𝛼𝑦𝑦 𝛼𝑦𝑧 ) (𝑬𝒚 ). Where 𝛼𝑖𝑗 forms the polarizability tensor. long string of dipole for example, the dipole cancels out with each
𝑷𝒛 𝛼𝑧𝑥 𝛼𝑧𝑦 𝛼𝑧𝑧 𝑬𝒛
other except at the ends. The net charge at the ends is the bound
charge. If to calculate the bound charge of a dielectric tube:𝜎𝑏 =
Rotating (Alignment of Polar Molecules) happens on some molecules (polar molecules) which
𝑞
̂ . Just as we expected the accumulation of charges at the surface is the
= 𝑃𝑐𝑜𝑠𝜃 = 𝑷 ∙ 𝒏
have the built-in dipole moment. When it is placed in a uniform electric field 𝑬, the force on 𝐴𝑒𝑛𝑑

bound charges. If the polarization is nonuniform there will be bound charges accumulate
the positive end 𝑭+ cancels the negative end 𝑭− , but there is torque leading to rotation.
𝒅 𝒅 within the dielectric: ∫𝜈 𝜌𝑏 𝑑𝜏 = − ∮𝑠𝑷 ∙ 𝑑𝒂 = − ∫𝜈 (𝛁 ∙ 𝑷)𝑑𝜏 (the negative sign represents
Torque 𝑵 = 2 × 𝑭+ + × 𝑭− = 𝒑 × 𝑬 is in the direction to line 𝒑 up parallel to 𝑬. If 𝑬
2
the pileup of negative charge) → 𝜌𝑏 = −𝛁 ∙ 𝑷. The result is the same as we deduced earlier,
isn’t uniform, there will be a net force 𝑭 =
which means the accumulation of charges within is the bound charges too.
𝑞(∆𝑬) = 𝑞[(𝑑 ∙ ∇)𝑬] = (𝒑 ∙ ∇)𝑬. Where ∆𝑬 is
Why we can see the physical dipole as the pure dipole in the previous discussion? It’s because
the field difference between +𝑞 and −𝑞. Thus,
outside the dielectric 𝓇 is greater than the separation distance between charges. As for the
we can obtain the torque about the dipole in a
nonuniform field 𝑵 = 𝒑 × 𝑬 + 𝒓 × 𝑭.
inside of the dielectric when we are looking for the macroscopic field, we can replace the since it can be deduced directly from the free charge distribution. In linear dielectric 𝑫 =
actual microscopic charge configuration with smooth charge distribution. 𝜖0 𝑬 + 𝜖0 𝜒𝑒 𝑬. From this equation we can find out that 𝑫 is also proportional to 𝑬 (𝑫 =
𝜖𝑬), we call this constant permittivity. Divide the permittivity by 𝜖0 to obtain the relative
4.3 The Electric Displacement permittivity 𝜖𝑟 (dielectric constant) of the dielectric.
What is the total field? To calculate it we need to put the field produced by the bound charges Ex. 4.5. A metal sphere (r = a) with charge Q surrounded by linear dielectric (r = b, permittivity = 𝜖). Find the

and all other fields produced by the free charges 𝜌𝑓 together. From Gauss’s law, we know potential at the center.
𝑄 1
that 𝜖0 𝛁 ∙ 𝑬𝒕𝒐𝒕 = 𝜌 = 𝜌𝑏 + 𝜌𝑓 = −𝛁 ∙ 𝑷 + 𝜌𝑓 → ∇ ∙ (𝜖0 𝑬 + 𝑷) = 𝜌𝑓 We use the Sol) use ∮ 𝐷 ∙ 𝑑𝑎 = 𝑄𝑓𝑒𝑛𝑐 and 𝑫 = 𝜖𝑬, the field can be deduced → calculate the potential 𝑉 = 4𝜋 (𝜖
0𝑏
+

electric displacement 𝑫 ≡ 𝜖0 𝑬 + 𝑷 to substitute the parentheses’ part → 𝛁 ∙ 𝑫 = 𝜌𝑓 . 1 1

− 𝜖𝑏 ) → Dielectric is like imperfect conductor.
𝜖𝑎

Most of the time we only know the free charge, so this equation enables us to calculate the Is 𝑬 and 𝑫 alike in the linear dielectrics? The answer is no, even the 𝑷 and 𝑫 is now
𝑫 directly. Or using the integral form: ∮ 𝑫 ∙ 𝑑𝒂 = 𝑄𝑓𝑒𝑛𝑐 . proportional to the 𝑬. The factor 𝜖0 𝜒𝑒 isn’t the same at the vacuum (= 0) and dielectric →
Ex. 4.4. A wire with uniform line charge 𝜆 surrounded by a dielectric material curl 𝑷 ≠ 𝟎. But if the space is filled with homogeneous linear dielectric (𝜖0 𝜒𝑒 is constant
(radius = a), find 𝑫. everywhere) 𝑫 can be found just from the free charge (as the dielectric doesn’t exist): 𝑫 =
𝜆
Sol) use ∮ 𝐷 ∙ 𝑑𝑎 = 𝑄𝑓𝑒𝑛𝑐 → 𝐷 =
2𝜋𝑠
𝑠̂ 𝜖0 𝑬𝒗𝒂𝒄 . Using the relation between 𝑫 and 𝑬 and the relative permittivity, we can obtain
The equation 𝛁 ∙ 𝑫 = 𝜌𝑓 just looks like Gauss’s law except for 1
𝑬 = 𝜖 𝑬𝒗𝒂𝒄 . This equation points out that in homogeneous linear dielectric, the field is the
some details. However, it is not that much alike, there is no “Coulomb’s law” for 𝑫 and it 𝑟

cannot be expressed as the gradient of a scalar (it doesn’t have “potential”). Because 1
same as there is no dielectric but reduced by . For example, a free charge inside the
𝜖𝑟
𝛁 × 𝑫 = 𝛁 × 𝑷 ≠ 𝟎. When the problem exhibits symmetry, 𝑫 can be get directly by using
1 𝑞
the usual Gauss’s law methods (∵ 𝛁 × 𝑷 = 0). homogeneous linear dielectric, 𝑬 = 4𝜋𝜖 𝑟 2 𝒓̂ (it is 𝜖 not 𝜖0 → shielding effect)
⊥
What are the boundary conditions of 𝑫? (i) 𝐷𝑎𝑏𝑜𝑣𝑒 ⊥
− 𝐷𝑏𝑒𝑙𝑜𝑤 = 𝜎𝑓 (ii) 𝑫∥𝑎𝑏𝑜𝑣𝑒 − 𝑫∥𝑏𝑒𝑙𝑜𝑤 = Ex. 4.6.

𝑷∥𝑎𝑏𝑜𝑣𝑒 − ∥
𝑷𝑏𝑒𝑙𝑜𝑤 Most of the time crystal is easier to polarize in some directions, it can be denoted like the
polarization tensor but using the susceptibility tensor.
4.4 Linear Dielectrics
In homogeneous linear dielectric bound charge density is proportional to the free charge
What are linear dielectrics? We call the material whose polarization is proportional to the
𝜒
field (if external 𝑬 isn’t too strong) and obey the following equation 𝑷 = 𝜖0 𝜒𝑒 𝑬 linear density 𝜌𝑏 = − (1+𝜒𝑒 ) 𝜌𝑓 . Except the free charge is embedded in the material (really), the
𝑒

dielectrics. The constant 𝜒𝑒 is the electric susceptibility of the substance. The value of 𝜒𝑒
potential obeys Laplace’s equation.
depends on the external condition and its own structure. We cannot evaluate the 𝑷 directly
Ex. 4.7.
from the linear dielectrics’ equation, because the 𝑬 in the equation represents the total field.
Ex. 4.8.
Because the external field → polarization induced field → contribute to the total field →
What is the energy in dielectric system? From the formula we get in previous chapter add
impacts the induced field → …, so we need another method. It can start at evaluate the 𝑫,
some little adjustment, we obtain the energy stored in the linear dielectric system 𝑊 =
𝜖0 1
∫ 𝜖𝑟 𝐸 2 𝑑𝜏 = 2 ∫ 𝑫 ∙ 𝑬𝑑𝜏. The energy of a system implies the work require to assemble the
2
system, but in dielectric system it is not just bring in all the free charge but also the work
involved in stretching and twisting. This equation only applies on the linear dielectric system,
cause the energy of dissipative system depends on not only the final configuration but the
processes.
Ex. 4.9.

The bound charge tends to accumulate near the free

charge → the electric field attracts the dielectric (has
force), so what’s the force on dielectrics? It is entirely
determined by the distribution of charges (both free
and bound). Assume a linear dielectric material is
partially inserted between the capacitor plate. There is
a fringing field that pulls the dielectric into the capacitor not as we always suppose (the field
is uniformly inside the plates). The force caused by the fringing field is hard to evaluate, but
we can use another way to calculate it by supposing I pull the dielectric out a little bit and
1
with the help of the equation about capacitance. Deducing the force 𝐹 = 2 𝑉 2 .