跳转至

热学

Thermal Physics

Basic Concepts

The Laws of Thermodynamics

The Zeroth Law of Thermodynamics: If two systems are in thermal equilibrium with a third system, then they must be in thermal equilibrium with each other.

The Third Law of Thermodynamics: It is impossible for any procedure to lead to the isotherm \(T= 0\) in a finite number of steps.


Temperature Scales

Absolute temperature (Kelvin) scale: $\displaystyle T = T_{\text{triple}} \frac{p}{p_{\text{triple}}} \approx \frac{273.16}{610} p $

\(273.16\) is the difference between absolute zero and the temperature of the triple point of water .

Celsius scale: \(T_C = T - 273.15\)

Fahrenheit scale: \(\displaystyle T_F = \frac{9}{5}T_C + 32^{\circ}F\)


Thermal Expansion

Linear expansion: \(\Delta L = \alpha L_0 \Delta T\) Volume expansion: \(\Delta V = \beta V_0 \Delta T\)

\[ \begin{gathered} L^3 = [L_0(1 + \alpha \Delta T)]^3 = L_0^3(1 + 3\alpha \Delta T + O(\Delta T)^2) \\\\ V = V_0(1 + \beta \Delta T) \\\\ \Rightarrow \beta = 3\alpha \end{gathered} \]


Ideal Gases

Equation of state for an ideal gas:

\[ \begin{gathered} PV = nRT \\\\ PV = \frac{N}{N_A}RT = Nk_BT \end{gathered} \]

where \(R = 8.315 \text{ J/}(\text{mol}\cdot\text{K})\), \(\displaystyle k_B = \frac{R}{N_A} = 1.38\times 10^{-23} \text{ J/K}\).

Therefore \(\(\ln V = \ln T + \ln(nR / P)\)\)

Then we can get \(\(\beta = \left( \frac{1}{V} \frac{dV}{dT} \right)_p = \left( \frac{d(\ln V)}{dT} \right)_p = \frac{d(\ln T)}{dT} = \frac{1}{T}\)\)


Real Gases

The van der Waals equation of state is \(\((P + \frac{an^2}{V^2})(V - nb) = nRT\)\)

\(a\) is about the attractive forces between the gas molecules. \(b\) is about the volume of gas molecule.


Microscopic Model of Ideal Gases

Pressure on the Wall

The total momentum generated by molecules colliding with wall during \(dt\) is \(\(dP_x = \frac{1}{2} \cdot \frac{N}{V} \cdot Av_xdt \cdot 2mv_x = \frac{NAmv_x^2}{V} dt\)\)

So the pressure on the wall is \(\(p = \frac{1}{A}\frac{dP_x}{dt} = \frac{Nmx_v^2}{V} = \rho v_x^2\)\)

Therefore, we can get

\[ \begin{gathered} \overline{v^2} = \overline{v_x^2} + \overline{v_y^2} + \overline{v_z^2} \\\\ p = \frac{1}{3} \rho \overline{v^2} \end{gathered} \]

Furthermore, we can get

\[ \begin{gathered} pV = Nm \overline{v_x^2} = Nk_BT \\\\ \overline{\frac{1}{2} mv_x^2} = \overline{\frac{1}{2} mv_y^2} = \overline{\frac{1}{2} mv_z^2} = \frac{1}{2} k_BT \end{gathered} \]

So for ideal gas, temperature is a measure of internal energy. It measures the average energy per degree of freedom per molecule/atom.


Heat Capacity

Equipartition Theorem

The energy of each molecule is \(\(E = \sum_i \frac{p_i^2}{2m} + \sum_i \frac{1}{2}I_i\omega_i^2 + \sum_i \frac{1}{2}k_iq_i^2 + \cdots\)\)

The mean value of each independent quadratic term in the energy is equal to \(k_BT/2\). So the average energy of each molecule is \(\(f\cdot \frac{k_BT}{2}\)\)

where \(f\) is the degrees of freedom of the molecule. Vibration has both kinetic energy and potential energy, so it should be counted twice.

Internal energy: \(\displaystyle U = \frac{f}{2} Nk_BT = \frac{f}{2} nRT = nC_VT\)

Heat capacity: \(\displaystyle c_V = \left( \frac{\Delta U}{\Delta T} \right)_V = \frac{f}{2}Nk_B\)

Molar specific heat: \(\displaystyle C_V = \frac{c_V}{n} = \frac{f}{2}R\)


Mode Counting
  • \(N\)-atom linear molecule
    • Translation: \(3\)
    • Rotation: \(2\)
    • Vibration: \(3N - 5\)
  • \(N\)-atom nonlinear molecule
    • Translation: \(3\)
    • Rotation: \(3\)
    • Vibration: \(3N - 6\)

For single atom, it only has three degrees of freedom. For solid, Each atom vibrates around its equilibrium position and has three degrees of freedom, so \(f=6\).


Maxwell Distribution

Define the distribution function \(f(v)\) as \(\(f(v) = \frac{1}{N} \frac{dN}{dv}\)\)

Then we have

\[ \begin{gathered} dN = Nf(v)dv \\\\ \int f(v)dv = 1 \end{gathered} \]

The Maxwell distribution function is \(\(f(v) = 4\pi \left( \frac{m}{2\pi k_BT} \right)^{3/2} v^2 e^{-mv^2/2k_BT}\)\)

Characteristic Speed

Gauss’s Probability Integral

\[ \begin{gathered} I_{2n} = (-1)^n \frac{d^n}{da^n} I_0 \\\\ I_{2n+1} = (-1)^n \frac{d^n}{da^n} I_1 \end{gathered} \]

So we can get

\[ \begin{gathered} I_0 = \int_0^{\infty} e^{-ax^2} = \frac{1}{2} \sqrt{\frac{\pi}{a}}, \quad I_1 = \int_0^{\infty} xe^{-ax^2} = \frac{1}{2a} \\\\ I_2 = \int_0^{\infty} x^2e^{-ax^2} = \frac{1}{4} \sqrt{\frac{\pi}{a^3}}, \quad I_3 = \int_0^{\infty} x^3e^{-ax^2} = \frac{1}{2a^2} \\\\ I_4 = \int_0^{\infty} x^4e^{-ax^2} = \frac{3}{8} \sqrt{\frac{\pi}{a^5}}, \quad I_5 = \int_0^{\infty} x^5e^{-ax^2} = \frac{1}{a^3} \\\\ \end{gathered} \]

Most Probable Speed

\[ \begin{gathered} \frac{df(v)}{dv} = 0 \\\\ \Rightarrow (2v - \frac{mv^3}{k_BT}) e^{-mv^2/2k_BT} = 0 \\\\ \Rightarrow v_p = \sqrt{\frac{2k_BT}{m}} \end{gathered} \]

Average Speed

\[ \begin{aligned} \bar{v} &= \frac{1}{N} \int vdN = \int vf(v)dv \\ &= \int_0^{\infty} 4\pi \left( \frac{m}{2\pi k_BT} \right)^{3/2} v^3 e^{-mv^2/2k_BT} \\ &= \sqrt{\frac{8k_BT}{\pi m}} \end{aligned} \]

Root Mean Sqaure Speed

\[ \begin{aligned} v_{\text{rms}} &= \sqrt{\overline{v^2}} = \sqrt{\frac{1}{N} \int v^2dN} = \sqrt{\int v^2f(v)dv} \\ &= \int_0^{\infty} 4\pi \left( \frac{m}{2\pi k_BT} \right)^{3/2} v^4 e^{-mv^2/2k_BT} \\ &= \sqrt{\frac{3k_BT}{m}} \end{aligned} \]


Heat and The 1st Law of Thermodynamics

Heat ,Latent Heat, and Specific Heat

Definition of Heat Heat is defined as the transfer of energy across the boundary of a system due to a temperature difference between the system and its surroundings.

The ways of heat transfer are conduction, convection, and radiation.

Unit of heat includes \(\mathrm{J}\) and \(\mathrm{cal}\), where \(1 \mathrm{cal} = 4.186\mathrm{J}\).


Latent Heat Latent heat is energy released or absorbed during a constant-temperature process.

The latent heat of vaporization for a given substance is usually somewhat higher than the latent heat of fusion.

The specific latent heat is defined as \(\(L = \frac{Q}{m}\)\)


Specific Heat Specific heat capacity is defined as \(\(c = \frac{1}{M} \cdot \frac{dQ}{dT}\)\)

Molar heat capacity is defined as \(\(C = \frac{1}{n} \cdot \frac{dQ}{dT}\)\)

Molar heat capacity at constant volume is denoted by \(C_V\). Molar heat capacity at constant pressure is denoted by \(C_p\).


Heat Conduction

Fourier heat conduction law: $$\frac{Q}{\Delta t} = -\kappa_t A \frac{dT}{dx} $$

where \(\lambda\) or \(\kappa\) is coefficient of thermal conductivity.


Mean Free Path

Assume there is only one moving molecule. It will collide with the stationary molecules whose centers of mass are located within a cylinder of diameter \(2d\).

So during time interval \(t\), the average number of collisions is \(\(z = n_V\pi d^2vt\)\)

where \(\displaystyle n_V = \frac{p}{k_B T}\) is the density of the molecules.

Therefore, the mean free path is \(\(l = \frac{vt}{n_V\pi d^2vt} = \frac{1}{n_V\pi d^2} = \frac{k_BT}{\pi d^2p}\)\)

Because all molecules are actually moving, we must consider their relative motion when counting the number of collisions. By \(\(|\mathbf{v}_1 - \mathbf{v}_2|^2 = |\mathbf{v}_1|^2 + |\mathbf{v}_2|^2 - 2\mathbf{v}_1\cdot \mathbf{v}_2\)\)

We can conclude that the average relative speed is \(\sqrt{2}v\). Hence, \(\(l = \frac{vt}{n_V\pi d^2(\sqrt{2}v)t} = \frac{1}{\sqrt{2}n_V\pi d^2} = \frac{k_BT}{\sqrt{2}\pi d^2p}\)\)

And the average molecular separation is \(\(d = \frac{1}{n_V^{1/3}}\)\)


The 1st Law of Thermodynamics

The change in the internal energy \(U\) of the system can be expressed as: \(\(\Delta U = Q - W\)\)

The work done by the gas is

\[ \begin{gathered} dW = Fdy = PAdy = PdV \\\\ W = \int_{V_i}^{V_f}PdV \end{gathered} \]


Thermodynamic Processes

A quasi-static process is one that is performed sufficiently slowly so that the system is always close to equilibrium. A reversible process is one that can be run backward in time by simply reversing its inputs and outputs.

A reversible process must be quasi-static. A quasi-static process is not necessarily reversible (e.g., energy dissipation)


Isothermal vs Free Expansion

In isothermal expansion,

\[ \begin{gathered} W = \int_{V_i}^{V_f} \frac{Nk_BT}{V} dV = Nk_BT \ln \frac{V_f}{V_i} \\\\ Q = W \end{gathered} \]

In free expansion, \(\(P= 0, W = 0\)\)


Isobaric vs Isovolumetric Processes

In isovolumetric (isochoric) process

\[ W = 0 \\\\ Q = nC_V\Delta T \\\\ \Delta U = Q = nC_V\Delta T \]

In isobaric process,

\[ W = P\Delta V \\\\ Q = nC_P\Delta T \\\\ \Delta U = Q - W = n(C_P - R)\Delta T \]

So

\[ C_P = C_V + R \\\\ C_V = \frac{f}{2}R \\\\ C_P = \frac{f+2}{2}R \\\\ \gamma = \frac{C_P}{C_V} = \frac{f+2}{f} \]


Adiabatic Equation

In adiabatic process, no heat transfer into or out of the system, that is,

\[ \begin{gathered} Q = 0 \\\\ \Delta U = -W \end{gathered} \]

So

\[ \begin{gathered} dU = nC_VdT = -PdV = -\frac{nRT}{V} dV \\\\ \Rightarrow \frac{dT}{T} = -\frac{R}{C_V} \frac{dV}{V} \\\\ \Rightarrow TV^{\gamma-1} = \text{const} \\\\ \Rightarrow PV^{\gamma} = \text{const} \end{gathered} \]


Engines and The 2nd Law of Thermodynamics

Otto Cycle

Otto cycle is a four-stroke cycle consisting of two upstrokes and two downstrokes. Two of the strokes are adiabatic processes and the others are isovolumetric processes. It represents the gasoline engine.

\[ \begin{gathered} W = Q_h - Q_c \\\\ Q_h = nC_V(T_C - T_B) \\\\ Q_c = nC_V(T_D - T_A) \\\\ e = \frac{W}{Q_h} = 1 - \frac{Q_c}{Q_h} = 1 - \frac{T_D - T_A}{T_C - T_B} \end{gathered} \]

Because \(\(\left( \frac{V_2}{V_1} \right)^{\gamma-1} = \frac{T_A}{T_B} = \frac{T_D}{T_C}\)\)

We can get \(\(e = 1 - \frac{T_A}{T_B} = 1 - \frac{T_D}{T_C} = 1 - \frac{1}{(V_1/V_2)^{\gamma-1}} = 1 - \frac{1}{r^{\gamma-1}}\)\)


Diesel Engine

\[ Q_h = nC_p(T_C - T_B) \\\\ Q_c = nC_V(T_D - T_A) \\\\ W_{\text{cycle}} = n[C_p(T_C - T_B) - C_V(T_D - T_A)] \]


Carnot Cycle

Carnot cycle provides an upper limit on the efficiency of any classical thermodynamic engine, which consists of four steps: isothermal expansion, adiabatic expansion, isothermal compression, adiabatic compression.

In isothermal processes,

\[ \begin{gathered} Q_h = W_{AB} = nRT_h \ln \frac{V_B}{V_A} \\\\ Q_c = |W_{CD}| = nRT_c \ln \frac{V_C}{V_D} \\\\ \Rightarrow \frac{Q_c}{Q_h} = \frac{T_c}{T_h} \frac{\ln(V_C/V_D)}{\ln(V_B/V_A)} \end{gathered} \]

In adiabatic processes,

\[ \begin{gathered} T_hV_B^{\gamma-1} = T_cV_C^{\gamma-1} \\\\ T_hV_A^{\gamma-1} = T_cV_D^{\gamma-1} \\\\ \Rightarrow \frac{Q_c}{Q_h} = \frac{T_c}{T_h} \end{gathered} \]

So the efficiency of Carnot engine is \(\(e_C = 1 - \frac{T_c}{T_h}\)\)


Heat Pumps

The coefficient of performance at heating mode is \(\(\frac{Q_h}{W} = \frac{Q_h}{Q_h - Q_c} = \frac{1}{1 - \dfrac{Q_c}{Q_h}} = \frac{1}{1 - \dfrac{T_c}{T_h}} = \frac{T_h}{T_h - T_c}\)\)

The coefficient of performance at cooling modet is \(\(\frac{Q_c}{W} = \frac{T_c}{T_h - T_c}\)\)


The 2nd Law of Thermodynamics

Clausius Statement It is impossible to construct a cyclical machine whose sole effect is the continuous transfer of heat from one object to another object at a higher temperature without the input of energy by work.

Kelvin Statement It is impossible to construct a heat engine that, operating in a cycle, produces no effect other than the absorption of heat from a reservoir and the performance of an equal amount of work.


Entropy and Information

The Function of Entropy

In a Carnot cycle, \(\displaystyle \frac{Q_c}{Q_h} = \frac{T_c}{T_h}\), now put in the proper signs, we can get \(\(\frac{Q_h}{T_h} + \frac{Q_c}{T_c} = 0\)\)

Any reversible process can be approximated by a sum of Carnot cycles, hence \(\(\oint_{\text{reversible } C} \frac{dQ}{T} = 0\)\)

Therefore, we can define entropy as a state function.

The entropy of an ideal gas is

\[ \begin{gathered} dQ = dU + Pdv \\\\ \Rightarrow TdS = dU + PdV \\\\ \Rightarrow dS = \frac{nC_V^{mol}dT}{T} + \frac{nRdV}{V} \\\\ \Rightarrow S(T, V) = S_0 + nC_V^{mol}\ln \frac{T}{T_0} + nR\ln \frac{V}{V_0} \end{gathered} \]


The Second Law in terms of Entropy

If the path is arbitrary,

\[ \begin{gathered} 1 - \frac{Q_c}{Q_h} \le 1 - \frac{T_c}{T_h} \\\\ \Rightarrow \frac{Q_h}{T_h} + \frac{Q_c}{T_c} \le 0 \end{gathered} \]

Then we can get

\[ \begin{gathered} \oint_C \frac{dQ}{T} = \oint_{C_1} \frac{dQ}{T} + \oint_{C_2} \frac{dQ}{T} = \oint_{C_1} \frac{dQ}{T} + S(1) - S(2) \le 0 \\\\ \Rightarrow \oint_{C_1} \frac{dQ}{T} \le S(2) - S(1) \\\\ \Rightarrow dS \ge \frac{dQ}{T} \end{gathered} \]

So the change in entropy of an isolated system must be greater than zero for an irreversible process and equal to zero for a reversible process.


Entropy: A Measure of Disorder

Assume that each molecule occupies some microscopic volume \(V_m\), and the number of molecules is \(N\), then the number of microstates is

\[ \begin{gathered} W_i = \left( \frac{V_i}{V_m} \right)^N \\\\ W_f = \left( \frac{V_f}{V_m} \right)^N \\\\ \Rightarrow W = \frac{W_f}{W_i} = \left( \frac{V_f}{V_i} \right)^N \end{gathered} \]

Hence we can suggest \(\(S = k_B \ln W\)\)


Shannon Entropy

Given a random variable \(X \in \{ x_1, x_2, \cdots, x_n \}\), and its probability \(P_x\), we can define Shannon entropy as \(\(S = -\sum_x P_x\log_2 P_x\)\)

Shannon entropy quantifies the average information/uncertainty of the random variable. Shannon entropy also quantifies the resources needed to store the information.