Question 11.8: A Peltier generator is made up of two thermoelectric element...

A Peltier generator is made up of two thermoelectric elements connected in series (Fig. 11.1). One side of the Peltier generator is maintained at temperature T^+  and the other temperature T^− . The electric current I generated by the Peltier generator flows through the thermoelectric materials labelled 1 and 2. The plate which is heated up to temperature T^+ connects electrically these two materials but is not electrically available to the user.
Its electric potential is V^+ . The other ends of the thermoelectric materials are on the cold side, at a temperature T^− . They are connected to the electric leads of the device. A load resistance R_0 is connected to these leads. The voltage V designates the electric potential difference between the leads.
Analyse the operation of this generator using the electric charge and heat transport
equations,

j_{q1} = −σ_1 ε_1 ∇ T_1 − σ_1∇ \varphi _1        and      j_{Q1} = −κ_1∇ T_1 + T_1 ε_1 j_{q1}

j_{q2} = −σ_2 ε_2 ∇ T_2 − σ_2∇ \varphi _2        and      j_{Q2} = −κ_2∇ T_2 + T_2 ε_2 j_{q2} .

The thermoelectric materials 1 and 2 have a length d and a cross-section surface area A, which can be written as,

d = \int_{0}^{d}{dr} .\hat{x}                                             A = \int_{S}{dS} .\hat{x}

where \hat{x} is a unit vector oriented clockwise along the electric current density j_q , and the infinitesimal length and surface vectors dr and dS are oriented in the same direction. The temperature difference between the hot and cold ends is given by,

\Delta T= T^+ - T^- = \int_{0}^{d}{dr} .\nabla T_1 = \int_{0}^{d}{dr}.(-\nabla T_2 )

Likewise, the electric potential differences \Delta \varphi _1 and \Delta \varphi _2 between the hot and cold ends are written as,

\Delta \varphi _1 = V^+ = \int_{0}^{d}{dr}.\nabla \varphi _1 .

\Delta \varphi _2 = V^+ − V = \int_{0}^{d}{dr}.(-\nabla \varphi _2)

The electric charge conservation implies that the electric current densities are the same in each material, i.e. j_{q1} = j_{q2} . The electric current I flowing through materials 1 and 2 is the integral of the electric current densities j_{q1} and j_{q2} over the cross-section area A,

I = \int_{S} {j_{q1}} .dS =\int_{S} {j_{q2}} .dS

According the relation (10.104), the thermal powers P_{Q1} and P_{Q2} are the integrals of the heat current densities j_{Q1} and j_{Q2} flowing through materials 1 and 2 over the cross-section area A,

τ ≡ T\frac{∂ε}{∂T} .       (11.104)

P_{Q1} =  \int_{S} {(-j_{Q1)}}.dS                                       P_{Q2} =  \int_{S} {j_{Q2}}.dS

Determine:
a) the thermal power \acute{P}_Q applied on the hot side of the device when no electric current flows through the device.
b) the effective electric resistance R of the two thermoelectric materials when the temperatures are equal, i.e. T^+ = T^− , and no electric current flows through the resistance R_0 , i.e. when R_0 = ∞. Instead, an electric current flows through the thermoelectric materials.
c) the electric current I as a function of the temperature difference ΔT.

d) the thermodynamic efficiency of the generator defined as,

η =\frac{R_0 I^2}{P_Q}

where here, P_Q is the thermal power at the hot side when the electric current is flowing through the device. Show that the optimum load resistance is given by

\frac{R_0}{R} = \sqrt{1+\zeta }

where \zeta is a dimensionless parameter given by,

\zeta = \frac{T^+ (ε_1 − ε_2)^2}{(κ_1 + κ_2) \Bigl(\frac{1}{\sigma _1} + \frac{1}{\sigma _2} \Bigr) }
The Blue Check Mark means that this solution has been answered and checked by an expert. This guarantees that the final answer is accurate.
Learn more on how we answer questions.

a) When no electric current flows through the device, i.e. j_{q1} = j_{q2} = 0 , in order to determine the thermal power \acute{a}_Q, we integrate the heat transport equations over the volume V. The integrals over the volume are the product of integrals over the crosssection area A times integrals over the length d of the thermoelectric materials,

\int_{S}{(-\acute{j}_{Q1} )}.dS \int_{0}^{d}{dr} .\hat{x} = \kappa _1\int_{0}^{d}{dr} .\nabla T_1 \int_{S}{dS}.\hat{x}

 

\int_{S}{\acute{j}_{Q2}}.dS \int_{0}^{d}{dr} .\hat{x} = \kappa _2\int_{0}^{d}{dr} .(-\nabla T_2 )\int_{S}{dS}.\hat{x}

which reduce to,

\acute{P}_{Q1} = \kappa _1 \frac{A}{d} ΔT

 

\acute{P}_{Q2} = \kappa _2 \frac{A}{d} ΔT

Thus, the total thermal power is given by,

\acute{P}_{Q} = \acute{P}_{Q1} +\acute{P}_{Q2} = (κ_1 + κ_2)\frac{A}{d}ΔT

b) When the temperatures of the hot and cold sources are equal, i.e. T^+ = T^−, the temperature gradients vanish, i.e. ∇T_1 = ∇T_2, which implies that there is no thermoelectric effect. The integrals of the electric charge transport equations over the volume are the product of integrals over the cross-section area A times integrals over the length d of the thermoelectric materials,

\int_{S}{(-{j}_{q1} )}.dS \int_{0}^{d}{dr} .\hat{x} = \sigma _1\int_{0}^{d}{dr} .\nabla \varphi _1 \int_{S}{dS}.\hat{x}

 

\int_{S}{(-{j}_{q2} )}.dS \int_{0}^{d}{dr} .\hat{x} = -\sigma _2\int_{0}^{d}{dr} .(-\nabla \varphi _2) \int_{S}{dS}.\hat{x}

Since, the load resistance is infinite, i.e. R_0 = ∞, the electric current used for the measurement flows in the opposite direction, I→−I under the condition that V^+ < V.
In this case, the electric charge transport equations integrated over the volume reduce to,

I = σ_1 \frac{A}{d}Δ\varphi _1 = σ_1 \frac{A}{d} V^+

 

I = – σ_2 \frac{A}{d}Δ\varphi _2 = – σ_1 \frac{A}{d} (V^+ – V)

The potential difference across the thermoelectric materials 1 and 2 connected in series is given by,

V = Δ\varphi _1 – Δ\varphi _2 = I \frac{A}{d} \Bigl(\frac{1}{\sigma _1} +\frac{1}{\sigma _2} \Bigr) =I \frac{2 d}{A} \frac{1}{σ}

where σ is the effective conductivity of both thermoelectric materials. Since the effective electric resistivity ρ is the inverse of the effective electric conductivity ρ,

ρ = \frac{1}{σ} = \frac{1}{2 }\Bigl(\frac{1}{\sigma _1} +\frac{1}{\sigma _2} \Bigr)

Thus, the potential difference across both thermoelectric materials is written as,

V = ρ \frac{2 d}{A} I= R I

where 2 d is the effective length of both materials of length d each, section area A, connected in series, and R is their effective resistance. Hence,

R = ρ\frac{2 d}{A} = \frac{1 }{2} \Bigl(\frac{1}{\sigma _1} +\frac{1}{\sigma _2} \Bigr) \frac{2 d}{A}

c) The integrals of the electric charge transport equations over the volume are the product of integrals over the cross-section area A times integrals over the length d of the thermoelectric materials,

\int_{S}{j_{q1}}.dS \int_{0}^{d}{dr} .\hat{x} = -\sigma _1 \varepsilon _1\int_{0}^{d}{dr} .\nabla T_1 \int_{S}{dS} .\hat{x} – \sigma _1 \int_{0}^{d}{dr} .\nabla \varphi _1 \int_{S}{dS}. \hat{x}

 

\int_{S}{j_{q2}}.dS \int_{0}^{d}{dr} .\hat{x} = \sigma _2 \varepsilon _2\int_{0}^{d}{dr} .(-\nabla T_2) \int_{S}{dS} .\hat{x} + \sigma _2 \int_{0}^{d}{dr} .(-\nabla \varphi _2) \int_{S}{dS}. \hat{x}

The electric charge transport equations integrated over the volume reduce to,

I = −σ_1 ε_1 \frac{A}{d} ΔT − σ_1 \frac{A}{d}V^+

 

I = σ_2 ε_2 \frac{A}{d} ΔT − σ_2 \frac{A}{d}(V^+ – V)

Ohm’s law for the load resistance is written as,

V = R_0 I

In view of this relation that charaterises the electric properties of the load, the charge transport equations can be recast as,

V^+ = −\frac{1}{\sigma _1}\Bigl(\frac{d}{A}I – σ_1 ε_1 ΔT \Bigr)

 

V^+ = \frac{1}{\sigma _2}\Bigl( \Bigl(\frac{d}{A} + σ_2 R_0\Bigr)I − σ_2 ε_2 ΔT\Bigr)

which implies that the electric current is given by,

I = \frac{ε_2  ε_1}{\Bigl(\frac{1}{\sigma _1} +\frac{1}{\sigma _2} \Bigr) \frac{d}{A} +R_0} ΔT = \frac{ε_2  ε_1}{R + R_0} ΔT

This result for the current I is consistent with the current obtained in the analysis of the Seebeck loop (§ 11.6.2), which is equivalent to a Peltier generator (Fig. 11.1) in which the load resistance is set at zero, i.e. R_0 = 0.

d) In order to determine the thermal power P_Q entering through the hot plate at temperature T^+ , we integrate the heat transport equations over the volume V. The integrals over the volume are the product of integrals over the cross-section area A times integrals over the length d of the thermoelectric materials,

\int_{S}{(-j_{Q1})}.dS \int_{0}^{d}{dr} .\hat{x} =\kappa _1 \int_{0}^{d}{dr} .\nabla T_1 \int_{S}{dS} .\hat{x} – T^+ \varepsilon _1 \int_{S}{j_{q1}}.dS \int_{0}^{d}{dr} .\hat{x}

 

\int_{S}{j_{Q2})}.dS \int_{0}^{d}{dr} .\hat{x} =\kappa _2 \int_{0}^{d}{dr} .(-\nabla T_2) \int_{S}{dS} .\hat{x} + T^+ \varepsilon _2 \int_{S}{j_{q2}}.dS \int_{0}^{d}{dr} .\hat{x}

which reduce to,

P_{Q1} = κ_1 \frac{A}{d} ΔT − T^+ ε_1 I

 

P_{Q2} = κ_2 \frac{A}{d} ΔT + T^+ ε_2 I

and implies that,

P_Q = P_{Q1} + P_{Q2} = (κ_1 + κ_2) \frac{A}{d} ΔT + T^+ (ε_2 − ε_1) I

Therefore, the efficiency η for any load resistance R_0 is given by,

η = \frac{R_0 I^2}{P_Q} = \frac{R_0 \frac{(ε_2 − ε_1)^2 ΔT^2}{(R + R_0)^2}}{(κ_1 + κ_2) \frac{A}{d} ΔT + T^+ (ε_2 − ε_1)^2 \frac{ΔT}{R + R_0}}

which can be recast as,

η = \frac{ΔT}{T^+} \frac{\frac{R_0}{R}}{\frac{(κ_1 + κ_2)}{T^+ (ε_2 − ε_1)^2} \Bigl(\frac{1}{\sigma _1} +\frac{1}{\sigma _2} \Bigr) \Bigl(1+\frac{R_0}{R}\Bigr)^2 + \Bigl(1+\frac{R_0}{R}\Bigr) }

Using the definition of the coefficient  \zeta > 0, the temperature difference ΔT = T^+ − T^− > 0 and the ratio r = 1 + R_0/R > 0 , the efficiency is reduced to,

η = \Bigl(1 -\frac{T^−}{T^+} \Bigr) \frac{r − 1}{\zeta ^{-1} r^2 +r}

To find the optimal load resistance, we have to optimise the efficiency η with respect to the ratio r,

\frac{dη}{dr} = \Bigl(1 -\frac{T^−}{T^+} \Bigr) \frac {\zeta ^{-1} r^2 +r -(r-1) \bigl(\zeta ^{-1} r^2 +r\bigr)}{\bigl(\zeta ^{-1} r^2 +r\bigr)^2} =0

which implies that,

r^2 − 2 r − \zeta = 0

Thus, the optimum ratio r > 0, is given by,

r = 1+\sqrt{1+\zeta }

Hence, for an optimal resistance load, the efficiency is given by,

\eta = \Bigl(1- \frac{T^-}{T^+} \Bigr)\frac{\zeta \sqrt{1+\zeta } }{(1+\sqrt{1+\zeta } )^2 +\zeta (1+\sqrt{1+\zeta } )} \leq 1-\frac{T^-}{T^+}

In the limit \zeta → ∞ , the efficiency of the Peltier generator η tends towards the Carnot efficiency η_C (7.46),

 

lim_{\zeta → ∞}η =1- \frac{T^-}{T^+}= η_C

 

η_C = \frac{T^+ -T^-}{T^+} = 1- \frac{T^-}{T^+}<1                 (7.46)

Related Answered Questions