fermi liquid non fermi liquid marginal fermi liquid transport

# Fermi Liquid Quasiparticle Kinetics and Luttinger Liquid Breakdown in Nanoscale Systems

## 1. Introduction: From Fermi Gas to Fermi Liquid

In metals at low temperature, electrons do not behave as independent particles (Fermi gas), but rather as quasiparticles—collective excitations that retain many properties of individual electrons but with modified mass and lifetime. Landau Fermi liquid theory provides the theoretical framework for understanding this behavior in 3D systems.

However, at reduced dimensionality—particularly in 1D nanowires—the Fermi liquid picture breaks down completely. Instead, charge and spin excitations decouple and propagate independently, a phenomenon known as spin-charge separation. This leads to the emergence of a Luttinger liquid, a fundamentally non-Fermi liquid state.

Understanding this transition is crucial for sub-2nm carbon nanotube and semiconductor nanowire devices, where quantum confinement forces quasi-1D geometry and novel transport phenomena emerge.

## 2. Landau Fermi Liquid Theory: Quasiparticle Renormalization

In a Fermi liquid at zero temperature, the low-energy excitations are quasiparticles with dispersion:

$$\epsilon_\mathbf{k} = \frac{v_F(k - k_F)}{1 + F_s/3}$$

where:
- $v_F$ is the bare Fermi velocity
- $F_s$ is the Landau parameter (dimensionless interaction strength)
- The denominator $(1 + F_s/3)$ is the renormalization factor

The effective mass enhancement:
$$m^* = m \left(1 + \frac{F_s}{3} ight)$$

In a strongly interacting system, $F_s$ can be large, leading to dramatic mass enhancement. For example, in heavy fermion systems, $m^*/m > 100$.

## 3. Quasiparticle Lifetime and the $T^2$ Law

The fundamental property of a Fermi liquid is that quasiparticles near the Fermi surface have a lifetime diverging as temperature decreases:

$$ au^{-1} \propto \left(\frac{T}{T_F} ight)^2 + \left(\frac{\epsilon - \mu}{E_F} ight)^2$$

where $T_F \sim E_F / k_B$ is the Fermi temperature.

This quadratic ($T^2$) scaling arises from phase space restrictions: at low T, only quasiparticles within $k_B T$ of the Fermi surface can scatter, and the density of available final states (for three-body scattering: particle + hole + hole) also scales as $T^2$.

### Electrical Resistivity

The $T^2$ law directly impacts transport:
$$ ho(T) = ho_0 + A T^2$$

The $T^2$ coefficient $A$ is directly related to the quasiparticle scattering rate and Fermi liquid interactions. Deviation from $T^2$ behavior signals the breakdown of Fermi liquid physics.

## 4. Self-Energy and Spectral Function

The self-energy $\Sigma(\mathbf{k}, \omega)$ encodes all interactions:

$$\Sigma(\mathbf{k}, \omega) = \Sigma'(\mathbf{k}, \omega) + i \Sigma''(\mathbf{k}, \omega)$$

For a Fermi liquid at low energy:
- Real part: $\Sigma'(\omega) \approx m(\omega) v_F^2 (k - k_F)$ (renormalizes dispersion)
- Imaginary part: $\Sigma''(\omega) \propto \omega^2$ (determines lifetime)

The spectral function:
$$A(\mathbf{k}, \omega) = -\frac{1}{\pi} ext{Im}[G(\mathbf{k}, \omega)]$$

where $G$ is the Green's function. For a Fermi liquid, $A(\mathbf{k}, \omega)$ shows a sharp quasiparticle peak at $\omega = \epsilon_\mathbf{k}$, with width $\propto \Sigma''(\omega)$.

## 5. The Luttinger Liquid: 1D Deconfinement of Charge and Spin

In one dimension, the Fermi liquid picture collapses due to a crucial fact: all forward-scattering processes involve the same pair of states—no other states available at the Fermi surface.

This leads to perfect nesting and dramatic renormalization group flow. The elementary excitations are no longer electrons, but rather:
- Charge excitations (holons): carry charge $+e$, spin 0, velocity $v_c$
- Spin excitations (spinons): carry spin 1/2, charge 0, velocity $v_s$

These propagate independently—a phenomenon called spin-charge separation.

### Luttinger Parameter and Interaction

The Tomonaga-Luttinger parameter $K$ characterizes the strength of interactions:

$$K = \frac{\pi v_F}{\pi v_F + \int_0^\infty dq \, f(q)}$$

where $f(q)$ is the forward-scattering amplitude.

• $K = 1$: Non-interacting system (free fermions)
• $K < 1$: Repulsive interactions (typical for electrons)
• $K > 1$: Attractive interactions

The parameter $K$ determines all low-energy properties of the Luttinger liquid.

## 6. Spectral Function in Luttinger Liquid: Power-Law Singularities

Unlike the Fermi liquid's sharp quasiparticle peak, the Luttinger liquid spectral function exhibits power-law singularities:

$$A(\omega) \propto |\omega|^{\alpha(K) - 1}$$

where the exponent $\alpha(K)$ depends on $K$. For repulsive interactions ($K < 1$), there is no quasiparticle pole—the singularity is broader and non-singular.

This has dramatic experimental consequences: no Fermi surface exists in the traditional sense. Instead, the momentum distribution shows a power-law divergence at the boundary.

## 7. Conductance Quantization and the Quantum Wire

In a quantum wire (quasi-1D conductor), the conductance is quantized in units of $2e^2/h$:

$$G = N \frac{2e^2}{h} (1 - R)$$

where:
- $N$ is the number of propagating modes
- $R$ is the reflection probability

The quantization arises because in 1D, each mode has exactly two degrees of freedom (spin up/down), and each contributes $e^2/h$ to conductance (ignoring interactions).

However, interactions modify this. In a Luttinger liquid, the conductance is:

$$G = K \frac{2e^2}{h}$$

The factor $K$ represents the renormalization of conductance by interactions. For repulsive interactions ($K < 1$), the conductance is suppressed below the non-interacting value.

## 8. Marginal Fermi Liquid and Strange Metals

Some systems (notably cuprate superconductors in the normal state) exhibit marginal Fermi liquid behavior, intermediate between Fermi and non-Fermi liquid:

$$\Sigma(\omega, T) \approx -i \frac{\pi \Gamma}{2} \max(|\omega|, \pi k_B T)$$

where $\Gamma$ is a coupling constant.

This yields linear-in-$T$ resistivity ($
ho \propto T$) rather than the Fermi liquid $T^2$:

$$ ho(T) = ho_0 + B T$$

The $T$-linear resistivity persists up to room temperature in cuprates, suggesting the presence of marginal Fermi liquid scaling throughout a broad temperature regime.

## 9. Python Implementation: Fermi Liquid to Luttinger Liquid Transition

import numpy as np
import matplotlib.pyplot as plt
from scipy.integrate import quad
from scipy.optimize import fsolve

def fermi_dirac(E, mu, T):
    """Fermi-Dirac distribution."""
    k_B = 8.617e-5  # eV/K
    if T == 0:
        return 1.0 if E < mu else 0.0
    return 1.0 / (1.0 + np.exp((E - mu) / (k_B * T)))

def fermi_liquid_self_energy(omega, T, tau0=1e-12):
    """
    Fermi liquid self-energy: Σ''(ω) ∝ (T/T_F)² + (ω/E_F)²
    """
    E_F = 1.0  # eV (reference)
    T_F = E_F / 8.617e-5  # K
    
    # Imaginary part (lifetime broadening)
    Sigma_imag = -(np.abs(omega)**2 + (T/T_F)**2) / (2 * tau0)
    
    # Real part (small contribution)
    Sigma_real = 0.1 * np.sign(omega) * np.abs(omega)
    
    return Sigma_real, Sigma_imag

def luttinger_liquid_spectral_function(omega, K, cutoff=1e-3):
    """
    Luttinger liquid spectral function with power-law singularity.
    
    A(ω) ∝ |ω|^(α(K) - 1)
    """
    
    # Exponent (depends on K)
    alpha_K = 0.5 * (K + 1.0/K)
    
    # Avoid singularities
    omega_reg = np.abs(omega) + cutoff
    
    # Spectral function (normalized)
    A = np.power(omega_reg, alpha_K - 1.0) / (1 + omega_reg**2)
    
    return A

def fermi_liquid_spectral_function(omega, broadening=0.01):
    """
    Fermi liquid spectral function: sharp quasiparticle peak.
    A(ω) = (Γ/π) / [(ω - ε_k)² + Γ²]
    """
    
    epsilon_k = 0.0  # Quasiparticle energy
    Gamma = broadening
    
    A = (Gamma / np.pi) / ((omega - epsilon_k)**2 + Gamma**2)
    
    return A

def resistivity_fermi_liquid(T, rho0=1e-6, A_coeff=1e-8):
    """
    Fermi liquid resistivity: ρ(T) = ρ₀ + A T²
    """
    return rho0 + A_coeff * T**2

def resistivity_luttinger_liquid(T, rho0=1e-6, B_coeff=1e-7):
    """
    Luttinger liquid (marginal FL) resistivity: ρ(T) = ρ₀ + B T
    """
    return rho0 + B_coeff * T

def conductance_luttinger(K, N_modes=1):
    """
    Quantized conductance in Luttinger liquid.
    G = K × N × (2e²/h)
    """
    G_quantum = 2 * 7.748e-5  # 2e²/h in Siemens (approximate)
    return K * N_modes * G_quantum

# Energy/frequency array
omega = np.linspace(-0.5, 0.5, 1000)  # eV

# Spectral functions
A_FL = fermi_liquid_spectral_function(omega, broadening=0.05)
K_values = [0.5, 1.0, 2.0]  # Various Luttinger parameters
A_LL = [luttinger_liquid_spectral_function(omega, K) for K in K_values]

# Resistivity vs temperature
T_array = np.linspace(1, 300, 200)  # K
rho_FL = resistivity_fermi_liquid(T_array, rho0=1e-6, A_coeff=1e-8)
rho_MFL = resistivity_luttinger_liquid(T_array, rho0=1e-6, B_coeff=1e-7)

# Conductance for different K values
K_array = np.linspace(0.3, 2.0, 100)
G_norm = np.array([conductance_luttinger(K, N_modes=1) for K in K_array])

# Plotting
fig, axes = plt.subplots(2, 2, figsize=(13, 10))

# Panel A: Spectral function comparison
axes[0, 0].plot(omega, A_FL, 'b-', linewidth=2.5, label='Fermi Liquid')
for K, A_L in zip(K_values, A_LL):
    axes[0, 0].plot(omega, A_L / np.max(A_L) * np.max(A_FL), 
                   linewidth=2.5, label=f'Luttinger (K={K})')
axes[0, 0].set_xlabel('Energy ω (eV)', fontsize=11)
axes[0, 0].set_ylabel('Spectral Function A(ω)', fontsize=11)
axes[0, 0].set_title('FL vs LL Spectral Functions', fontsize=12, fontweight='bold')
axes[0, 0].legend(fontsize=9)
axes[0, 0].grid(True, alpha=0.3)

# Panel B: Resistivity vs temperature
axes[0, 1].plot(T_array, rho_FL*1e6, 'b-', linewidth=2.5, label='Fermi Liquid (T²)')
axes[0, 1].plot(T_array, rho_MFL*1e6, 'r--', linewidth=2.5, label='Marginal FL (T)')
axes[0, 1].set_xlabel('Temperature (K)', fontsize=11)
axes[0, 1].set_ylabel('Resistivity (μΩ·cm)', fontsize=11)
axes[0, 1].set_title('Resistivity: FL vs MFL scaling', fontsize=12, fontweight='bold')
axes[0, 1].legend(fontsize=10)
axes[0, 1].grid(True, alpha=0.3)

# Panel C: Self-energy (Fermi liquid)
T_test = 100  # K
omega_range = np.linspace(-0.2, 0.2, 200)
Sigma_real, Sigma_imag = fermi_liquid_self_energy(omega_range, T_test)
axes[1, 0].plot(omega_range, Sigma_real*1e12, 'b-', linewidth=2, label="Re[Σ]")
axes[1, 0].plot(omega_range, -Sigma_imag*1e12, 'r--', linewidth=2, label="-Im[Σ]")
axes[1, 0].set_xlabel('Frequency ω (eV)', fontsize=11)
axes[1, 0].set_ylabel('Self-Energy (10⁻¹² eV)', fontsize=11)
axes[1, 0].set_title('Fermi Liquid Self-Energy', fontsize=12, fontweight='bold')
axes[1, 0].legend(fontsize=10)
axes[1, 0].grid(True, alpha=0.3)

# Panel D: Conductance renormalization by interactions (Luttinger)
axes[1, 1].plot(K_array, G_norm, 'purple', linewidth=2.5)
axes[1, 1].axhline(y=conductance_luttinger(1.0), color='k', linestyle='--', 
                  alpha=0.5, label='Non-interacting (K=1)')
axes[1, 1].fill_between(K_array, 0, G_norm, alpha=0.3, color='purple')
axes[1, 1].set_xlabel('Luttinger Parameter K', fontsize=11)
axes[1, 1].set_ylabel('Conductance G/(2e²/h)', fontsize=11)
axes[1, 1].set_title('Conductance Renormalization in LL', fontsize=12, fontweight='bold')
axes[1, 1].legend(fontsize=10)
axes[1, 1].grid(True, alpha=0.3)

plt.tight_layout()
plt.savefig('fermi_luttinger_liquid.png', dpi=150, bbox_inches='tight')
plt.show()

print("=== Fermi Liquid to Luttinger Liquid Analysis Complete ===")
print(f"Fermi liquid resistivity @ 100K: {resistivity_fermi_liquid(100)*1e6:.2e} μΩ·cm")
print(f"Marginal FL resistivity @ 100K: {resistivity_luttinger_liquid(100)*1e6:.2e} μΩ·cm")
print(f"Conductance renormalization (K=0.5): {conductance_luttinger(0.5)*1e5:.2f} × 10⁻⁵ S")
print(f"Conductance renormalization (K=1.0): {conductance_luttinger(1.0)*1e5:.2f} × 10⁻⁵ S")

## 10. Charge Density Wave and Spin Density Wave Instabilities

At very low temperature or high density, Luttinger liquids can undergo spontaneous ordering:
- Charge density wave (CDW): Periodic modulation of charge density
- Spin density wave (SDW): Periodic modulation of spin density

These emerge from divergent charge/spin susceptibilities at certain wavevectors, driven by forward-scattering interactions. The instability occurs when the interaction becomes sufficiently strong.

## 11. Bosonization: Mathematical Framework

The bosonization technique is the mathematical tool for solving Luttinger liquid problems. The key idea:
- Fermionic excitations → replaced by bosonic plasmon modes (charge/spin density oscillations)
- All correlation functions → computed from bosonic Lagrangian

This reduces a complicated fermionic many-body problem to manageable boson algebra.

## 12. Experimental Signatures in Carbon Nanotubes

Carbon nanotubes exhibit strong 1D confinement effects. Signatures of Luttinger liquid behavior:
- Tunneling conductance $G(\omega) \propto |\omega|^{\alpha}$ with $\alpha
eq 0$ (non-Fermi liquid power law)
- Suppressed low-energy conductance ($G < 2e^2/h$)
- Modified Coulomb blockade oscillations (fractional periodicity)

## 13. Renormalization Group Flow and Scaling

The renormalization group describes how interactions flow to longer distances/lower energies. In 1D systems with perfect nesting, the forward-scattering interaction diverges logarithmically at low energy—a hallmark of instability toward ordered phases.

## 14. Comparison: Fermi Liquid vs. Luttinger Liquid vs. Marginal FL

## 15. Future Directions: Quantum Liquids and Quantum Criticality

Emerging research areas:
- Quantum critical points: Phase transitions at $T = 0$ drive non-Fermi liquid scaling
- Topological Luttinger liquids: Edge states of topological insulators
- Kondo lattices: Competition between local moment screening and RKKY interactions
- Mott insulators and metal-insulator transitions: Interplay of Coulomb repulsion and band structure

## 16. Implication for Sub-2nm Semiconductor Devices

As silicon GAAFETs, sheet-FETs, and carbon nanotube channels scale below 2 nm cross-sections, quantum confinement forces carriers into quasi-1D subbands. Understanding the transition from 3D Fermi liquid kinetics to 1D Luttinger liquid dynamics is crucial for predicting channel resistance, carrier mobilities, and thermal dissipation limits in sub-2nm nodes.

## 17. References & Further Reading

1. Giamarchi, T. (2004). *Quantum Physics in One Dimension*. Oxford University Press.
2. Schulz, H. J. (1995). "Fermi liquids and non-Fermi liquids." *arXiv preprint condensed-matter/9503150*.
3. Voit, J. (1995). "One-dimensional Fermi liquids." *Reports on Progress in Physics*, 58(9), 977.