Dalat University

Faculty of Physics & Nuclear Engineering

Nguyễn Đăng Chiến, Ph.D.

Associate Professor

2026

Contents

Chapter 1: Introduction to the Course

Chapter 2: Review on Derivatives and Integrals

Chapter 3: Vector Analysis

Chapter 4: Wave Equations

Chapter 5: Heat Conduction Equations

Chapter 6: Laplace and Poisson Equations

1
3 Chapter 5: Heat Conduction Equations

Contents of Chapter 5

I. Derivation of Heat Conduction Equations

II. Initial and Boundary Conditions

III. The Cauchy Problem of an Infinite Rod

IV. Homogeneous Heat Conduction Equations

V. Inhomogeneous Heat Conduction Equations

4 Chapter 5: Heat Conduction Equations

2
 I. Derivation of Heat Conduction Equations

 Physical observation

• Direction: Heat moves from the hot end to the cold end.

• Intensity: The greater the temperature difference between the ends,

the faster the heat flows.

• Material: A copper rod conducts heat much faster than a wooden

stick of the same size.

5 Chapter 5: Heat Conduction Equations

I. Derivation of Heat Conduction Equations

 Why does heat conduct from a hotter place to a

colder place?

• Microscopic perspective: When high-speed molecules from the hot zone

collide with the slower ones in the cold zone, according to the laws of
conservation of momentum and energy, the faster molecules transfer a
portion of their kinetic energy to the slower ones.

• Macroscopic perspective: Having a hot zone next to a cold zone

represents a state of certain "order". Nature dislikes this concentration; it has
a tendency to disperse energy uniformly across space to reach the highest
state of chaos. (Nature likes simplicity & harmony)
• The higher the chaos, the more the simplicity is because simplicity is
actually as freedom. (complexity needs the control  less freedom).
• The higher the chaos, the more the harmony is because hamorny is
actually as equality.

6 Chapter 5: Heat Conduction Equations

3
 I. Derivation of Heat Conduction Equations

 1D Fourier's law of heat conduction

• For a 1D rod, let ( , ) be the temperature. Experiments show that

the amount of heat energy conducted per unit time ∆ /∆ is:
o Proportional to the cross-sectional area
∆ ∆
o Proportional to the temperature difference ∆ ⟹ ~−
∆ ∆
o Inversely proportional to the distance ∆

• To characterize the rate of heat conduction, we define the heat flux

– the heat energy conducted per unit time per unit area:
1∆
=
∆
• Taking the limit ∆ → 0, we get the 1D Fourier’s law:
=−

7 Chapter 5: Heat Conduction Equations

I. Derivation of Heat Conduction Equations

 3D Fourier's law of heat conduction

• In a 3D medium, heat flows in the direction where the temp.

decreases most rapidly:
⟹ =− .

• : the heat flux vector. =

.

• : the thermal conductivity, It is a measure of the material's ability to

. .
conduct heat. =
.
o In isotropic materials, is a constant scalar.
o In anisotropic materials (like wood or crystals), can be a
tensor, meaning heat flows more easily in one direction than
another.
• : The direction of steepest temperature increase.

8 Chapter 5: Heat Conduction Equations

4
 I. Derivation of Heat Conduction Equations

 1D heat conduction eq.: Without heat exchange on the side

• Consider a differential element , from to + . Within the
differential time interval , the heat energy entering the element is:
= . − .

• If there is a heat source inside the rod with a heat generation rate
per unit volume of ( , ), the heat generated in within is:
= ,
• From the definition of specific heat capacity , this total heat energy
increases the temperate of the element by an amount :
+ | − |
= = + ,
( )
−
( , ) ⟺ = + ( , )
⟹ = +

9 Chapter 5: Heat Conduction Equations

I. Derivation of Heat Conduction Equations

 3D heat conduction equation

• Consider a differential volume enclosed by the surface . Within
the differential time interval , the heat energy entering is:

= − ⃗ =− =

• The heat generated by the heat source ( , , , ) in within is:

= , , ,

• The total heat energy increases the temperate of by :

+
= = + , , ,
( )
( , , , )
⟹ = + ⟺ = + ( , , , )

10 Chapter 5: Heat Conduction Equations

5
 I. Derivation of Heat Conduction Equations

 Heat conduction equation: Cause & effect perspective

• If relying on the physical reasoning used to derive the equation:
= + , , , ,
it is obvious that:
o is the exogenous cause that makes change. (Do not interpret
“exogenous” spatially; it refers to factors outside the kinematic structure).
o is the thermodynamic cause that makes change. (Being
intrinsic to the kinematic structure).

• From the view point of dynamic nature:

o : primary / direct cause: The active injection of energy.
o : secondary / indirect cause: internal reactive mechanism.
 These causes work together to dictate the future of rod’s temp.

11 Chapter 5: Heat Conduction Equations

I. Derivation of Heat Conduction Equations

 Heat conduction equation: Effects of on the temp evolution

• Direct effect:
o The source directly injects energy into / drains energy from a
local point, independent of the rod's current state.
o This energy input/output forces an immediate rate of change in
temp. ( ). In a split second, the point gets hotter/colder simply
because it is being "loaded" with power.

• Indirect effect:
o is non-uniform  does not merely alter the local temp, but
also distorts the temp. distribution (bending the temp. field).
o This distortion induces the curvature (thermodynamic stress,
activating conduction mechanisms) to drive further temperature
changes through the redistribution of heat.

12 Chapter 5: Heat Conduction Equations

6
 II. Initial and Boundary Conditions

 Initial condition
• In physics, the first-order time derivative implies that the future state
of the system depends directly on its current state.
• The initial condition (IC) describes the temperature distribution
throughout the entire object at the exact moment you begin your
observation (usually defined as = 0).
• Some cases of ICs:
o Uniform distribution: The object starts at a const. temp. throughout
,0 =
o Non-uniform distribution: Temp. varies depending on the location.
,0 = ( )
o Impulse distribution: A singular temp. spike at a specific location
,0 = ( − )

13 Chapter 5: Heat Conduction Equations

II. Initial and Boundary Conditions

 Boundary conditions: Dirichlet boundary condition

• If the two ends of the rod are in contact with heat reservoirs & their
temps. are kept at the reservoirs’ temps. (external law of constraint):
0, = ( ); , = ( )

• The reservoirs is so large that they can provide / absorb any amount
of heat energy necessary to keep those point at exactly , .

• Therefore, the heat flowing through the boundaries is a result of the

solution (i.e. the governing law), not a separate constraint, to
maintain that equilibrium.

• Note: The heat transfer at boundaries is not affected by external

dynamics; otherwise the boundary temp. cannot be maintained properly.
 External constraint-driven boundary temperature
 kinematic structure-driven exchange of heat

14 Chapter 5: Heat Conduction Equations

7
 II. Initial and Boundary Conditions

 Boundary conditions: Neumann boundary condition

• If the two ends are subjected to a specific heat supply & the rate of
heat flow is prescribed (external law of constraint):
− (0, ) = ; − , = ( )
• The “smart” heater/cooler must dynamically adjust its own intensity
or temp. to compensate for whatever the rod is doing internally.

• Consequently, the temp. at boundaries is governed by the system's

internal kinematics so that the resulting spatial gradient precisely
matches the required flux.

• Note: The boundary temperature is "sacrificed" or forced to evolve in

such a way that the energy flow rule is never violated.
 External constraint-driven heat flux,
 kinematic structure-governed boundary temperature

15 Chapter 5: Heat Conduction Equations

II. Initial and Boundary Conditions

 Boundary conditions: Isolated Neumann BC

• If the two ends of the rod are perfectly wrapped in an ideal insulating
material, no heat exchange can occur with the external environment:
0, = 0 ; , =0
• The temp. at the boundaries is a result of the governing law, not a
separate constraint, to maintain that state of zero energy leakage.
• The kinematic structure of the system must redistribute internal heat
such that the temperature gradient at the ends remains exactly zero
at all times.
• Note: The kinematic structure must "flatten" the gradient at the
boundary to ensure not a single Joule of heat escapes. (while in the
Neumann condition with flux, the system must "strain" to accept energy)
 External constraint-driven zero heat flux;
 kinematic structure-driven internal thermal redistribution.

16 Chapter 5: Heat Conduction Equations

8
 II. Initial and Boundary Conditions

 Heat transfer at the boundaries: Newton’s law of cooling

• The heat conduction by Fourier’s law is not applied at boundaries.
• Dirichlet & Neumann: The heat exchange at the boundaries is non-
spontaneous (i.e. not follow the natural heat transfer laws of the rod-
environment interaction), but is subjected to certain constraints.

• Now, we consider the spontaneous exchange of heat. When in

contact with a fluid, heat energy is exchanged through the thermal
interaction between the solid surface and the fluid particles.
• Newton found that the exchange rate is governed by convection, and
the heat flux ℎ is proportional to the temp. difference:
= , −
quantifies how "easily" energy can cross the interface  called
convective heat transfer coefficient, also thermal “conductance”.

17 Chapter 5: Heat Conduction Equations

II. Initial and Boundary Conditions

 Newton’s law of cooling: Application to solids

• When two solids pressed together, they are never perfectly flat at a
microscopic level. "air gaps" & "contact points" exist simultaneously.
 This interface behaves exactly like a fluid layer with very high
resistance. Therefore, we can apply the same mathematical form:
= , −
• Here, (contact conductance) represents the "quality
of the handshake" between the two solid structures.
 If the contact pressure of or the surface flatness
increases, increases (the handshake gets tighter).
 The heat transfer is still driven by the temp. gap, just
like in a fluid, but the "buffer" is now the microscopic
interface resistance.
• Note: (fluid) or (solid) defines the "transparency" of boundary.

18 Chapter 5: Heat Conduction Equations

9
 II. Initial and Boundary Conditions

 Boundary conditions: Robin boundary condition

• If the two ends of the rod are exposed to a fluid at a temp. & the
heat exchange occurs naturally (no external or internal constraint).

• The rate of heat flow is governed by the Newton’s law of cooling,

where the heat flux is proportional to the local temp. difference:
0, = 0, − ; − ( , ) = , −

• Both the temp. & heat flux are "free“, and depend on each other
through the system's internal kinematic structure and the
environment's thermal resistance ( = 1/ ).

• Note: The heat transfer is self-regulated; it is neither forced

(Neumann) nor is the temperature fixed (Dirichlet).

 Temperature-difference driven, self-regulating heat exchange

19 Chapter 5: Heat Conduction Equations

II. Initial and Boundary Conditions

 Boundary conditions: The Spectrum of Connectivity

• From Neumann  Robin  Dirichlet, discuss the spectrum of
thermal connectivity between the rod & the external world.

• Neumann ( = 0): Isolated system. The gate is completely closed. The

external world has zero influence on the energy flow. The internal kinematic
structure is "sovereign," redistributing its own heat without any leakage.

• Robin (0 < < ∞): Negotiated system. The gate is partially open. Heat
flow depends on the "Thermal Gap". The internal kinematics & the external
environment reach a compromise based on the resistance of the interface.

• Dirichlet ( → ∞): Enslaved system. The gate is wide open. The interface
is transparent. The internal state at the boundary is instantly forced to match
the external source. The "negotiation" ends in total submission to the
constraint.

 From Isolation to Submission: The evolution of thermal interaction

20 Chapter 5: Heat Conduction Equations

10
 II. Initial and Boundary Conditions

 Boundary conditions: Robin to Dirichlet transition

• For a perfect contact, the microscopic gaps between the two solids
disappear, thus the thermal contact resistance drops to zero: →∞

• From the Robin equation:

0, = 0, − ⟹ 0, =

 The Robin boundary condition collapses into a Dirichlet boundary

condition.
• Note: In this state, the “kinematic structure" of the rod is now directly
enslaved to the temperature of the external solid. There is no longer
a "bridge" or "filter"; The boundary temperature is instantly forced to
match the external source.
 Robin logic vanishes, Dirichlet constraint dominates

21 Chapter 5: Heat Conduction Equations

III. The Cauchy Problem of an Infinite Rod

 The purity of interior kinematics: The Cauchy problem

• We study an infinitely long rod (−∞ < < ∞) to remove the "noise"
of the boundaries.
• Without Dirichlet or Robin conditions, the heat flow is not
"negotiated" with the environment but is governed solely by the
material's internal structure ( ) and the initial “memory” :

= ; ,0 = ( )

• This is a study of pure diffusion. We observe how energy naturally

seeks equilibrium by spreading into the infinite void without any
external constraints.
• Note: In solids, heat doesn't "flow" like a liquid; it "diffuses" through
the lattice. The atoms don't move places, but their vibrational state
(phonons) diffuses through the structure.

22 Chapter 5: Heat Conduction Equations

11
 III. The Cauchy Problem of an Infinite Rod

 Separation of Variables: Is it workable?

• In wave equations, space & time are locked ( ± ) to create a
“running" shape. Velocity is a “complex” kinematic attribute - a blend
of & . (If you separate them, you "kill" the motion)  Unworkable
• In heat conduction eqs., the shape stays at its location and only
"melts" or "fades" in place.  No coupling of space & time (to form
complex kinematic dimensions like velocity, acceleration,…).  Workable
− Space ( ): Defines the "geometry" of the heat packet (the static statue).
− Time ( ): Acts as a simple "volume knob" (the fading timer).

 Since there is no “running“, space & time are decoupled. This

allows us to solve for the "Shape" and the "Fading" independently.
• Note: A propagating wave is a marriage of space & time to create
speed. But heat is a divorce. The heat peak doesn't travel to a new
city; it just dies in its own bed.

23 Chapter 5: Heat Conduction Equations

III. The Cauchy Problem of an Infinite Rod

 Separation of Variables: Modeling the “decoupled" flow

• Set ( , ) = ( ) ( ) & insert into the heat conduction eq. to get:
1 ′ − =0
− =− ⟺
+ =0
• Solve the ODEs to obtain the eigen-solutions or thermal modes:
( , )= ( ) ( )
= cos + sin ; =

• Kinematic interpretation:
− represents the “static geometry" (the standing waves).
− : “decay engine“ to turn down the intensity of wave over .

 are “simple static & fading waves of heat”, forming a basis of the
entire temp. field, the fundamental DNAs of the equation.

24 Chapter 5: Heat Conduction Equations

12
 III. The Cauchy Problem of an Infinite Rod

 Role of the initial memory: The mode mixing formula

• The general solution is the linear sum of thermal modes:

( , )= cos + sin

• Since & represent how much of each wavevector is

needed to build the shape of temp. profile, their maps serve as a
“color mixing formula” for painting the temp. distribution.
• Since the formula remains unchanged, we can determine it at = 0:
,0 = ⟹ cos + sin = ( )
1 1
⟹ = cos ; = ( ) sin
2 2
 The initial memory not only provides the information but also
establishes the mode mixing rule.

25 Chapter 5: Heat Conduction Equations

III. The Cauchy Problem of an Infinite Rod

 Redistribution: From memory to blur

• Because each mode ( ) fades at a different rate, their sum ( ) will

change shape over time. Insert , into to get:
1
( , )= ( ) cos ( − )
2
Memory Distributor

• Note that the memory ( ) contained in , has been moved

out of the square bracket to make it an impartial distributor.
• The distributor redistributes the initial temp. (initial memory) to each
location over time. (through the “seeping” driven by gradient change)
− = 0, the distributor must be ( − ): the initial information ( )
at is completely distributed at .
− > 0, the distributor lowers but widens, blurring the memory.

26 Chapter 5: Heat Conduction Equations

13
 III. The Cauchy Problem of an Infinite Rod

 Distribution function: A Gaussian bell

• To understand the distributor, we calculate its integral by setting:

= ; − =
1
⟹ Distributor = cos ≜ ( , )
2

• Take the derivative both sides & perform

+ =0
the integration by parts, we obtain: 2
⟹ , = ,0
1 1 1
where ,0 = = =
2 2 2

1 1
⟹ , = = ≜ ( , ; )
2 2
 This is a Gaussian function which is symmetric with respect to .

27 Chapter 5: Heat Conduction Equations

III. The Cauchy Problem of an Infinite Rod

 The origin of memory blurring: Diffusion of a heat impulse

1
, = , ; =
2
• Assume , = 0 = 0. At = 0 , the heat energy is conducted
into segment ( − ℎ, + ℎ), causing its temp. to increase to:
1
= ⟹ , =
2ℎ 2 2ℎ

• Let ℎ → 0, we obtain: , = = ( , ; )
2

⟺ , ≡ , = , ; =
2
 Time pulls the peak downward A heat impulse is diffused
 Time forces the energy to spread outward equally in both sides.

28 Chapter 5: Heat Conduction Equations

14
 III. The Cauchy Problem of an Infinite Rod

 Re-derivation of ( , ): Collection of heat impulses

From the distribution function of heat energy in a
differential element , re-derive the temp. profile ( , ).

29 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs: Zero temperature ( / = )

• We want to learn the heat conduction in a 1D rod of length :
− Without heat exchange on the side & heat source;
− The temperature at both ends is kept at zero.

• The heat conduction eq. that needs to be solved is: =

with the boundary conditions: , = , =

and the initial condition: , = ( )

• Notes: − Using the separation of variables;

− Exist an internal constraint that allows flexible thermal exchanges
at the boundaries as long as the ends are instantly & permanently
frozen at zero (e.g., in contact with an "infinite ice bath“).

30 Chapter 5: Heat Conduction Equations

15
 IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs ( / = ): Dynamics vs. Geometry

• The term represents a future impulse, i.e., it tells us how fast

the memory is changing at this very moment.

• The term presents a local curvature (stress). The 2nd

derivative in space detects "bumps" or "valleys" in the temp. profile.

• Nature hates sharp edges:

− If there is a "peak" (negative curvature), the first derivative in time
must be negative  Heat flows out to lower the peak.
− If there is a "valley" (positive curvature), the first derivative in time
must be positive  Heat flows in to fill the valley.
 The equation is a “smoothing machine" that uses time to flatten
the geometry of space.

31 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs ( / = ): From PDE to ODEs

• Let , = . ( ), substitute into = to obtain:

1 + =0
= ⟺ = = =− ⟺
+ =0
• Note: A negative constant − is required for heat to
fade (not explode). [Exercise: Discuss why this is the case]
• > 0 ( ′ & : same sign): If > 0  > 0 thus must keep increasing.
The hotter it is, the faster it heats up, leading to an infinite thermal explosion.
• < 0 ( ′ & : opposite signs): If > 0  < 0 thus the change of must
be negative. The system is "self-correcting," forcing the heat to fade away to
reach equilibrium.
Spatial: = cos + sin (The shape)
• Solve the ODEs:
Temporal: = (The life span)

32 Chapter 5: Heat Conduction Equations

16
 IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs ( / = ): Spatial finiteness & quantization

• In infinite world: no constraints  Waves ( ) can take any shape.
• In a finite space, waves are "caged"  They must obey the specific
"posture" dictated by the given boundaries (i.e. by constraints).
• Only waves that conform to the boundary constraints will survive
within the finite cage:
− At = 0, the boundary demands 0 = . 1 + . 0 = 0. The “free"
Cosine is instantly killed or removed out of the cage ( = 0).
− At = , the boundary demands ( ) = sin = 0 . Any
wave that does not perfectly match this posture cannot exist
inside the cage. To avoid a dead/void system ( = 0), the wave
vector must sacrifice its freedom:
= ⟹ ≡ = ( = 1, 2, 3, … )

33 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs ( / = ): Mode distributor

• Because each ( , ) serves as a fundamental structural form that
the system is allowed to exist (eigenstate, fundamental/basis state),
all eigenstates have equal rights (democracy) in determining the total
state:
, = sin

• serves as a “color mixing formula” or “mode mixing rule”, which is

unchanged over time, determined through the initial thermal state:
2
,0 = sin = ⟹ = ( ) sin

• also acts as the mode distributor which measures & distributes

the contribution of basis states, passed down from the memory ( ).

34 Chapter 5: Heat Conduction Equations

17
 IV. Homogeneous Heat Conduction Equations

 Homo. Dirichlet BCs ( / = ): Spatial distributor

• Insert into , , we have:
2
, = ( ) sin sin

• If a thermal impulse is introduced into the frozen rod at :

= ( − )

the patial distribution of temperature at > 0 will be:

2
, = sin sin ≜ , ;

• , ; acts as the spatial distributor, which is the "machine" that

receives heat at & disperses it into the patial structure of the rod.

35 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Homo. Neumann BCs: Zero heat flux ( ′ / = )

This problem is for students to practice at home!

(Please perform the step-by-step investigations as presented previously)
……………………..
…………………
……………..
…………
…….
…
.

36 Chapter 5: Heat Conduction Equations

18
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: A fixed temperature at both ends

• Solving the heat conduction equation: =

with the boundary conditions: , = , = ≠

and the initial condition: , = ( )

• Geometrically, we can visualize as follows:

− = 0: The cage placed on the ground (where its boundaries are
anchored to the absolute zero level).
− ≠ 0: The elevated cage (the entire spatial structure is "lifted" to a
higher pedestal). Elevated Cage

Grounded Cage

0 0
Ground
=0 ≠0

37 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: A fixed temperature at both ends

• Since the thermal distribution is uniformly elevated across the cage,
the internal thermal echoes preserve their fundamental geometry.
 Their form remains undistorted, dancing with the same rhythm as
they would on the ground, only now upon a higher stage.

• Physically, performing the thermal rhythm does not depend on the

height of the thermal stage  step down to the thermal ground.

• Mathermatically, it can be converted into the problem of the

homogeneous BCs by setting , = , + , we have:

= ; = ⟹ =

with the boundary conditions: 0, = , =0

and the initial condition: ,0 = − ≜ ( )

38 Chapter 5: Heat Conduction Equations

19
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Different const. temps. at the two ends

• Cosidering the boundary conditions: 0, = ; , =

and the initial condition: ,0 = ( )

• Geometrically, we can visualize as follows:

− The cage is no longer lifted uniformly. One side is raised to a
height of , while the other reaches , creating an asymmetric
foundation for the heat to evolve.

Uniformly lifted cage

−
Thermal stage platform
0 (Single layer) 0
Ground

39 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Different const. temps. at the two ends

• The internal thermal echoes (modes) now perform their "dance"
upon a slanted thermal stage.
 Despite the tilt, the rhythm of these echoes remains pure and
undistorted relative to this new surface.

• This stage is built from two layers: a flat base with a constant
thickness of & a wedge with a varying thickness of − .

• By removing the stage platform, , − − − ≜ , ,

new thermal rhythm , is performed on the thermal ground, i.e.
0, = , =0

• Since the stage platform is also removed at the initial state, we have:
,0 = − − − ≜ ( )

40 Chapter 5: Heat Conduction Equations

20
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Varying temperatures at both ends

• In the previous cases, the outside world was static, so the

boundaries were kept silent  The thermal stage also remained still.
• When the boundaries move (caused by the shaking foundation):
, = ; , = ,
they are now shaking the entire platform, making the stage to shift
up, down, and tilt as time flows.

move
 The thermal stage is no longer a
fixed platform. It becomes a
moving / floating stage.

move
 The thermal modes perform
their dance upon the moving
stage. 0 Ground

41 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Varying temperatures at both ends

• The rhythm of the thermal echoes is the sum of:
− The “internal” rhythm on a static stage (without external movement);
− The “external” rhythm, caused & followed by the stage's shaking.

• The external rhythm is quasi-static (no echoes), described by:

, = + −

• To learn the “internal” rhythm, the “forced” one (determined by the BCs)
should be removed: (mathematically, changing the frame of reference)
, − , ≜ ,

• It is easy to find that the internal rhythm ( , ) is governed by:

− = , ≜ ( , )

42 Chapter 5: Heat Conduction Equations

21
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Varying temperatures at both ends

• Discuss the ( , ): − What does it originate from?
− What is its operation mechanism?
− Is it a real source of heat?
• = ⁄  it originates from the shaking of stage.
• = ⁄  a change in temperature over time at point creates a
thermal gradient that forces heat to flow. The rod’s heat feels a “thermal
inertia force” because it cannot keep up with the shaking stage.
• ( , ) is the energy portal that pipes heat from the outside into point (or
vice versa) via stage movement  effective heat source.
 This heat is delivered to or withdrawn from the modes, forcing them to
vibrate in sync with the shaking stage instead of fading away.
 The internal thermal echoes (modes) no longer only remember the past
( ), but also respond to the present ( , ).

43 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Dirichlet BCs: Varying temperatures at both ends

• Finally, the boundary conditions of the internal rhythm , can be
derived:
, = , + + −

⟹ = , + + − ; = ,
⟹ 0, = ⟹ 0, =0
, = ⟹ , =0

• Since the stage begins shaking from the very first moment, the
external quasi-static rhythm at = 0 must also be removed to purely
examine the internal rhythm:
,0 = ⟹ ,0 = − 0 − 0 − 0 ≜ ( )

• Note: By removing the stage , , we bring the internal rhythm , to

dance in a ground-placed cage  The thermal modes are Sine functions.

44 Chapter 5: Heat Conduction Equations

22
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Constant and equal flux at both ends

• Cosidering the boundary conditions: , = , = ≠ 0

and the initial condition: ,0 = ( )

• Physically, the BCs represents a steady stream of heat (i.e., heat

enters at one end and leaves at the other at the exact same rate).

• Geometrically, we can visualize as follows:

− = 0: The cage placed on a flat stage.
− ≠ 0: The cage placed on a uniformly tilted stage.
Insulated cage

The flat stage

0 0
Ground
=0 ≠0

45 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Constant and equal flux at both ends

• The quasi-static stage: Mathematically, the tilted stage is a straight
line with a constant tilt:
( ) =

• This external state is static, which acts as a silent base (or stage) for
the internal rhythm to “dance”, and does not generate heat.

• By removing the external state from the total rhythm :

, − ≜ , ,
only the internal rhythm , confined in a perfectly insulated cage
is investigated:
, = , ; 0, = , =0

• Similar procedure needs to be done for the internal memory ,0 :

,0 = ,0 − = − ( ) ≜ ( )

46 Chapter 5: Heat Conduction Equations

23
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Different constant fluxes at the two ends

• Cosidering the boundary conditions: , = ; , =

and the initial condition: ,0 = ( )

• The stage is no longer tilted uniformly. One side is bent to a slope of

, while the other reaches a slope of , creating a curved
(parabolic) foundation for the heat to evolve.

Why is a parabolic stage?

How to use it? The mechanism?

0 0
Ground

47 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Different constant fluxes at the two ends

The Curved Stage
• The challenge: A straight line can only represent a single slope. To
satisfy different slopes at the ends, we need a "bent" / curved stage.
• Why parabolic? We choose a 2nd-order polynomial because it is the
simplest curve that bridges and .
• Constructing the stage: We build the external rhythm ( ), i.e. the
curved stage, such that its slopes match the boundaries:
= + + ; =2 +
−
0 = ⟹ = ; = ⟹ =
2
• The baseline height of the stage is not determined by the BCs, but
by the initial state.
−
 Freely choose: = 0 ⟹ = +
2

48 Chapter 5: Heat Conduction Equations

24
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Different constant fluxes at the two ends

Shifting the Stage
• From the flat ground to a curved stage: Removing the external
rhythm from the total one: − ≜ : the internal rhythm dancing on
the curved stage & confined in an isolated cage:
0, = , =0

• Substituting , = , − into the Heat Conduction

Equation, we get: −
, − , = ≜ ( )

• Question: The total rhythm exchanges heat with the outside world
through the thermal flux at the boundaries. However, for the internal
rhythm , the boundaries are locked (insulated). So, how does it
exchange heat, and through what mechanism?

49 Chapter 5: Heat Conduction Equations

IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Different constant fluxes at the two ends

The Effective Heat Source
• Thermodynamic force: Wherever the temp. field exhibits
“curvature” (tension / stress), a thermodynamic force naturally
emerges to redistribute energy, smoothing out the peaks and filling
the valleys until the tension is dissolved.
• Because the internal rhythm must cling to the floor of the stage ,
it is forced to bend exactly to the stage's shape. This forced
curvature creates an internal thermal stress, generating a
thermodynamic force proportional to that curvature:
−
= =

• is the energy portal piping heat from the outside into every point
(or vice versa) via stage curvature  Effective heat source.

50 Chapter 5: Heat Conduction Equations

25
 IV. Homogeneous Heat Conduction Equations

 Inhomo. Neumann BCs: Varying fluxes at both ends

• Cosider the boundary conditions: , = ( ); , = ( )

and the initial condition: ,0 = ( )

This problem is for students to practice at home!

(Please perform the step-by-step investigations as presented previously)
…………………
The curvature
……………..
varies over time
…………
…….
( )
• Note: By removing the stage , , we
bring the internal rhythm , to dance in
a flat cage (but not ground-placed).
0
 The thermal modes are Cosin functions. Ground

51 Chapter 5: Heat Conduction Equations

V. Inhomogeneous Heat Conduction Equations

 Two types of heat sources

• The heat conduction equation that needs to be solved is:

− = , ≠ 0

with the general boundary conditions:

0, = ; , =
and the initial condition: ,0 = ( )

• The heat source can be physical (real) or effective factor or both:

, = , + , , .

, : physical sources (electro-thermal chemo-thermal, radiant heat).

, : effective sources (in curved and/or moving frames of reference).

52 Chapter 5: Heat Conduction Equations

26
 V. Inhomogeneous Heat Conduction Equations

 Corresponding homo. equation: The method of thermal stage

• Find the solution of the corresponding homogeneous equation

− = 0 satisfying the Dirichlet, Neumann, or Robin BCs.

• The method of thermal stage is applicable to every types of BCs:

− Build the simplest stage ( , ) that satisfies the inhomo. BCs of
. (To shift the boundary complexities into the ( , ) & into the Heat
Conduction Equation as an effective source).
− Find the thermal modes “standing” on this stage, i.e.,
, = , − ( , )
that satisfies a new inhomogeneous Heat Conduction Equation
, − , = , − , ≜ ( , )

53 Chapter 5: Heat Conduction Equations

V. Inhomogeneous Heat Conduction Equations

 The method of thermal stage: Locking boundaries

• The main purpose of the method is to create a safe space for the
internal rhythm to exist in its purest, homogeneous form.
• To do so, the chaos at the boundaries must be completely absorbed
so that no external noise can disturb the dance of thermal echoes.

Boundary Absorbed Boundary Locked Effective Heat Source

Type Complexity (Chaos) Boundary Generation

Non-zero temps.: =0 − (stage movement)

Dirichlet
(0, ) & ( , ) (Frozen) (Thermal inertia force)

Non-zero heat fluxes: =0 (stage curvature)

Neumann
(0, ) & ( , ) (Insulated) (Thermal tension force)

Boundary energy imbalance: ± + ℎ = 0 − (Combined

Robin
± +ℎ = ( ) (Balanced) interaction stress)

54 Chapter 5: Heat Conduction Equations

27
 V. Inhomogeneous Heat Conduction Equations

 The method of thermal stage: Solution of the homo. equation

• The general form of the solution of the homogeneous equation for :

, = ( )

• The set of in combination with ( ) forms the thermal

configuration determined by the kinematic structure of the rod.

Boundary Eigenfunctions
Physical Meaning
Type (Natural thermal modes)

Dirichlet = sin Ends are "frozen"

Neumann = cos Ends are “insulated"

Robin = cos + sin Ends are “breathing"

55 Chapter 5: Heat Conduction Equations

V. Inhomogeneous Heat Conduction Equations

 Method of variation of constants: Presevation of configuration

• We only study the thermal evolution in a rod which its kinematic

structure is preserved. (otherwise, its nature will change)
 The thermal configuration, or the “internal DNA”, is preserved:
, = ( ) ( )

• The heat source (external force) only redistributes the importance of

the thermal modes.  ( ) evolve to carry its effect.

• If the external force wants to influence the system, it must "speak the
language" of the rod's natural modes.  projecting ( , ) onto basis
to see how much energy is being "fed" into each mode:

, = ( ) ( ), where ( )=

56 Chapter 5: Heat Conduction Equations

28
 V. Inhomogeneous Heat Conduction Equations

 Method of variation of constants: Heat absorbtion of mode

• Substituting , & , into the inhomo. heat conduction eq.

with note that =− , the PDE collapses into a set of
independent ODEs
+ = ( )

 Describe how the natural mode absorb heat from the source.

• These 1st-order inhomogeneous linear ODEs are solved [using (1) the
method of variation of constants or (2) reduction to exact ODEs by an
integrating factor. See Ch. 4 in Math 2, p. 37] to obtain:

= 0 +

Physically: Natural Decay (Memory) + Accumulated Driving Force

Mathematically: Response to Initial Data + Response to Input

57 Chapter 5: Heat Conduction Equations

V. Inhomogeneous Heat Conduction Equations

 The initial memory of the thermal configuration

• The total solution is the sum of the stage evolution (boundary
response) and the internal evolution (structure-driven response):

, = , + ( ) ( )

⟹ ,0 = ,0 + 0 ⟹ ,0 = (0) ( )

,0 ∫ ,0
⟹ 0 = =
∫
Conclusion:
 The stage absorbs the boundary's turbulence to create a silent cage,
allowing the system’s internal behavior to evolve through its own
natural DNA − the eigenfunctions.
 While the real & effective heat sources fuel the energy, they never
destroy the system's soul (internal kinematic structure).

58 Chapter 5: Heat Conduction Equations

29
 V. Inhomogeneous Heat Conduction Equations

 Illustration: An example

• Solve the heat conduction equation:

, − , = 1− 1− cos − sin

with BCs: 0, = sin ; , =1−

and the initial condition: ,0 = 0

……………………..
…………………
……………..
…………
…….
…

59 Chapter 5: Heat Conduction Equations

60 Chapter 5: Heat Conduction Equations

30