4: Thermodynamics and Statistical Mechanics- From Static Response to Dynamic Fluctuations

Fluctuations Near Equilibrium · LibreTexts

0%

Contents

    Text

    Size
    100%
    Theme
    Typeface
    Line length
    Spacing

    Read aloud

    How

    Speed: 1×
    Light up
    Highlight

    Symbols: Mulberry Symbols, © 2018–2026 Steve Lee, CC BY-SA 4.0 (licence, the set), the site's own copies.

    From LibreTexts, under CC BY-NC-SA 4.0, as the source states it; its equations are written in TeX, which the PDF shows typeset. This page at LibreTexts.

    4: Thermodynamics and Statistical Mechanics- From Static Response to Dynamic Fluctuations

    4.1 The Statistical Basis of Response

    In thermodynamics, the response of a system to an external field is deeply connected to the spontaneous fluctuations of the system at equilibrium. This relationship allows us to relate macroscopic response coefficients, such as susceptibility, to microscopic statistical variances [7, [6, 3].

    4.2 Linear Response to a Conjugate Field

    Consider a system in thermal equilibrium at temperature \(T\) with an unperturbed Hamiltonian \(H_{0}\). When an external field \(h\) is applied to an observable \(A\), the total Hamiltonian becomes \(H=H_{0}-h A\).

    The partition function \(Z\) in the canonical ensemble is:

    \[ Z(h, \beta)=\operatorname{Tr}\left[e^{-\beta\left(H_{0}-h A\right)}\right] \tag{4.1} \label{4.1} \]

    where \(\beta=1 /\left(k_{B} T\right)\). The equilibrium expectation value \(\langle A\rangle\) is related to the derivative of the partition function:

    \[ \langle A\rangle=\frac{1}{\beta} \frac{\partial \ln Z}{\partial h} \tag{4.2} \label{4.2} \]

    The static susceptibility \(\chi\), defined as the rate of change of the average value \(\langle A\rangle\) with respect to the field \(h\), is derived from Eq. 4.2 as:

    \[ \chi=\left.\frac{\partial\langle A\rangle}{\partial h}\right|_{h \rightarrow 0}=\beta\left(\left\langle A^{2}\right\rangle-\langle A\rangle^{2}\right)=\frac{\operatorname{Var}(A)}{k_{B} T} \tag{4.3} \label{4.3} \]

    As shown in Eq. 4.3, the macroscopic response is determined entirely by the equilibrium variance [?].

    4.3 Cross-Correlations and Generalized Response

    If we measure the response of a different observable, \(B\), to the field \(h\) conjugate to \(A\), the thermal average is governed by the covariance \(C_{B A}\) :

    \[ \begin{gather*} \langle B\rangle=\frac{1}{Z} \operatorname{Tr}\left(B e^{-\beta\left(H_{0}-h A\right)}\right) \tag{4.4} \label{4.4}\\[4pt] \frac{\partial\langle B\rangle}{\partial h}=\beta C_{B A} \tag{4.5} \end{gather*} \]

    Here,

    \[ C_{B A}=\langle B A\rangle-\langle B\rangle\langle A\rangle \tag{4.6} \label{4.6} \]

    is the cross correlation.

    4.4 Cross-Correlations and Generalized Response

    If we measure the response of a different observable \(B\) to the field \(h\) conjugate to \(A\), the thermal average is:

    \[ \langle B\rangle=\frac{1}{Z} \operatorname{Tr}\left(B e^{-\beta\left(H_{0}-h A\right)}\right) \tag{4.7} \label{4.7} \]

    To find the response, we differentiate with respect to \(h\) :

    \[ \frac{\partial\langle B\rangle}{\partial h}=\frac{\beta \operatorname{Tr}\left(B A e^{-\beta H}\right)}{Z}-\frac{\operatorname{Tr}\left(B e^{-\beta H}\right)}{Z} \frac{\beta \operatorname{Tr}\left(A e^{-\beta H}\right)}{Z}, \tag{4.8} \label{4.8} \]

    remembering \(H=H_{0}-h A\). Simplifying the terms into expectation values, we obtain:

    \[ \frac{\partial\langle B\rangle}{\partial h}=\beta(\langle B A\rangle-\langle B\rangle\langle A\rangle)=\beta C_{B A} \tag{4.9} \label{4.9} \]

    where \(C_{B A}\) is the cross-correlation (or covariance) between \(A\) and \(B\). \(A, B\) are classical observable and therefore they commute.

    4.5 Physical Applications

    4.5.1 Curie's Law in Paramagnets

    Equation (9) above gives, for a magnetic response in an external field, \(H_{\text {ext }}\)

    \[ \chi=\frac{\partial\langle M\rangle}{\partial H_{e x t}}=\beta\left(\left\langle M^{2}\right\rangle-\langle M\rangle^{2}\right) \tag{4.10} \label{4.10} \]

    For a system where the mean magnetization is zero:

    \[ \chi=\beta\left\langle M^{2}\right\rangle \tag{4.11} \label{4.11} \]

    where the variance \(\left\langle M^{2}\right\rangle=N \mu^{2}\) drives the response.Thus, for a paramagnetic material, like oxygen, with \(N\) non-interacting spins and magnetic moment \(\mu\), the perturbation is \(-\mu \cdot \mathbf{H}\). The magnetization \(M_{z}\) is conjugate to the external field \(H_{\text {ext }, z}\) [5]. In the limit \(H_{\text {ext }, z} \rightarrow 0\), the susceptibility follows Curie's Law [4]:

    \[ \chi_{z z}=\frac{N \mu^{2}}{3 k_{B} T}=\frac{C}{T} \tag{4.12} \label{4.12} \]

    The factor of \(1 / 3\) comes from averaging \(\cos ^{2}(\theta)\), where the angle between each magnetic moment and the field is assumed to be uniformly distributed.

    4.5.2 Quantum Treatment: The Brillouin Function

    In the quantum mechanical regime[5, 6], the magnetic moment is quantized. The magnetization \(M\) for \(N\) spins with total angular momentum \(J\) is given by:

    \[ M=N g_{J} \mu_{B} J B_{J}(x), \quad x=\frac{g_{J} \mu_{B} J H}{k_{B} T} \tag{4.13} \label{4.13} \]

    where \(B_{J}(x)\) is the Brillouin function:

    \[ B_{J}(x)=\frac{2 J+1}{2 J} \operatorname{coth}\left(\frac{2 J+1}{2 J} x\right)-\frac{1}{2 J} \operatorname{coth}\left(\frac{1}{2 J} x\right) \tag{4.14} \label{4.14} \]

    In the limit of high temperature \((x \ll 1)\), the susceptibility \(\chi=M / H\) reduces to Curie's Law:

    \[ \chi=\frac{N g_{J}^{2} \mu_{B}^{2} J(J+1)}{3 k_{B} T} \tag{4.15} \label{4.15} \]

    where the effective moment \(\mu_{e f f}^{2}=g_{J}^{2} \mu_{B}^{2} J(J+1)\) replaces the classical \(\mu^{2}\).

    In the quantum model, the magnetization is derived by taking the thermal average of the discrete magnetic moments \(g_{J} \mu_{B} m_{J}\). In the high-temperature (low-field) limit, the susceptibility \(\chi\) is proportional to the average of the squares of the magnetic quantum numbers \(m_{J}\) :

    \[ \left\langle m_{J}^{2}\right\rangle=\frac{1}{2 J+1} \sum_{m_{J}=-J}^{J} m_{J}^{2} \tag{4.16} \label{4.16} \]

    Using the identity for the sum of squares \(\sum_{k=1}^{n} k^{2}=\frac{n(n+1)(2 n+1)}{6}\), the summation over the \(2 J+1\) states yields:

    \[ \sum_{m_{J}=-J}^{J} m_{J}^{2}=\frac{1}{3} J(J+1)(2 J+1) \tag{4.17} \label{4.17} \]

    Normalizing by the number of states, we find \(\left\langle m_{J}^{2}\right\rangle=\frac{1}{3} J(J+1)\). This demonstrates that the factor of 1/3 persists in the quantum regime due to the isotropic distribution of the total angular momentum across the three spatial dimensions.

    4.5.3 Heat Capacity and Energy Fluctuations

    In the canonical ensemble, the internal energy \(E\) fluctuates. The average energy \(\langle E\rangle\) is given by \(-\frac{\partial \ln Z}{\partial \beta}\). The constant-volume heat capacity \(C_{V}\) is defined as the rate of change of energy with respect to temperature [?]:

    \[ C_{V}=\left(\frac{\partial\langle E\rangle}{\partial T}\right)_{V}=\left(-\frac{1}{k_{B} T^{2}}\right) \frac{\partial\langle E\rangle}{\partial \beta} \tag{4.18} \label{4.18} \]

    Differentiating the average energy with respect to \(\beta\) using the quotient rule:

    \[ \frac{\partial\langle E\rangle}{\partial \beta}=-\left[\frac{\sum E_{i}^{2} e^{-\beta E_{i}}}{Z}-\left(\frac{\sum E_{i} e^{-\beta E_{i}}}{Z}\right)^{2}\right]=-\left(\left\langle E^{2}\right\rangle-\langle E\rangle^{2}\right) \tag{4.19} \label{4.19} \]

    Substituting this back, we arrive at the energy fluctuation relation:

    \[ C_{V}=\frac{\left\langle E^{2}\right\rangle-\langle E\rangle^{2}}{k_{B} T^{2}}=\frac{\operatorname{Var}(E)}{k_{B} T^{2}} \tag{4.20} \label{4.20} \]

    Equation 4.20 demonstrates that a system with high thermal inertia also exhibits large spontaneous energy fluctuations.

    4.5.4 Analogous Equilibrium Effects

    The following table summarizes the most significant macroscopic response functions and the microscopic fluctuations that drive them:

    Property Symbol Response to... Underlying Fluctuation
    Heat Capacity \(C_{V}\) Temperature \((T)\) Energy \((E)\)
    Isothermal Compressibility \(\kappa_{T}\) Pressure \((P)\) Particle Number \((N)\) or Volume \((V)\)
    Magnetic Susceptibility \(\chi\) Magnetic Field \((H)\) Magnetization \((M)\)
    Electric Polarizability \(\alpha\) Electric Field \((E)\) Dipole Moment \((P)\)
    Table 4.1: Analogies between response functions and spontaneous fluctuations.

    4.6 Transition to Dynamic Response

    The Dynamic Fluctuation-Dissipation Theorem (FDT) is the time-dependent evolution of the static identities discussed above. While the static case considers the variance, the dynamic case utilizes the Time-Correlation Function, \(C(t)=\langle A(t) A(0)\rangle\), which describes how a fluctuation persists or decays over time. The FDT states that the way a system dissipates energy from an external field is identical to how it spontaneously relaxes from a natural thermal fluctuation. This will be discussed later in detail.

    Bibliography

    [1] Landau, L. D., and Lifshitz, E. M. (2013). Statistical Physics. Elsevier.

    [2] Chandler, D. (1987). Introduction to Modern Statistical Mechanics. Oxford University Press.

    [3] Pathria, R. K., and Beale, P. D. (2011). Statistical Mechanics. Academic Press.

    [4] Curie, P. (1895). Proprietes magnetiques des corps a diverses temperatures. Annales de Chimie et de Physique.

    [5] Kittel, C. (2004). Introduction to Solid State Physics. Wiley.

    [6] Brillouin, L, (1926). Les problèmes de quanta et la théorie cinétique du magnétisme, Journal de Physique et le Radium, 7, (5),135-160,


    This page titled 4: Thermodynamics and Statistical Mechanics- From Static Response to Dynamic Fluctuations was last modified on Mon, 02 Nov 2026 06:09:25 GMT and is shared under a CC BY-NC-SA 4.0 license and was authored, remixed, and/or curated by Pabitra N. Sen.