Model systems with interaction

Slides:



Advertisements
Similar presentations
Dr Roger Bennett Rm. 23 Xtn Lecture 19.
Advertisements

Significance Testing.  A statistical method that uses sample data to evaluate a hypothesis about a population  1. State a hypothesis  2. Use the hypothesis.
Ch1. Statistical Basis of Thermodynamics
Area scaling from entanglement in flat space quantum field theory Introduction Area scaling of quantum fluctuations Unruh radiation and Holography.
Chapter 3 Classical Statistics of Maxwell-Boltzmann
1 Spin Freezing in Geometrically Frustrated Antiferromagnets with Weak Bond Disorder Tim Saunders Supervisor: John Chalker.
MSEG 803 Equilibria in Material Systems 8: Statistical Ensembles Prof. Juejun (JJ) Hu
Statistical Mechanics
Magnetism in Chemistry. General concepts There are three principal origins for the magnetic moment of a free atom: The spins of the electrons. Unpaired.
Introduction to Thermostatics and Statistical Mechanics A typical physical system has N A = X particles. Each particle has 3 positions and.
Magnetic model systems Let’s start with a reminder and a word of caution: Goal: Using magnetic model systems to study interaction phenomena We know from.
Monte Carlo Simulation of Ising Model and Phase Transition Studies
Observables. Molar System The ratio of two extensive variables is independent of the system size.  Denominator N as particle  Denominator N as mole.
Find the numerical value of the expression. sinh ( ln 4 )
Chap.3 A Tour through Critical Phenomena Youjin Deng
The Ising Model of Ferromagnetism by Lukasz Koscielski Chem 444 Fall 2006.
Presentation in course Advanced Solid State Physics By Michael Heß
Monte Carlo Simulation of Ising Model and Phase Transition Studies By Gelman Evgenii.
Relating computational and physical complexity Computational complexity: How the number of computational steps needed to solve a problem scales with problem.
Boltzmann Distribution and Helmholtz Free Energy
Excerpts of Some Statistical Mechanics Lectures Found on the Web.
Spatial Statistics Applied to point data.
31.1 Thermodynamics of Mixing of Ideal Solutions For the process where solute and solvent are mixed to form an ideal solution at constant temperature and.
Outline Review of extended ensemble methods (multi-canonical, Wang-Landau, flat-histogram, simulated tempering) Replica MC Connection to parallel tempering.
Lecture 8. Systems with a “Limited” Energy Spectrum The definition of T in statistical mechanics is consistent with our intuitive idea of the temperature.
1 Worm Algorithms Jian-Sheng Wang National University of Singapore.
The Ising Model Mathematical Biology Lecture 5 James A. Glazier (Partially Based on Koonin and Meredith, Computational Physics, Chapter 8)
Heat Conduction and the Boltzmann Distribution Meredith Silberstein ES.241 Workshop May 21, 2009.
8. Selected Applications. Applications of Monte Carlo Method Structural and thermodynamic properties of matter [gas, liquid, solid, polymers, (bio)-macro-
Entropy and temperature Fundamental assumption : an isolated system (N, V and U and all external parameters constant) is equally likely to be in any of.
Thermodynamics System: Part of Universe to Study. Open or Closed boundaries. Isolated. Equilibrium: Unchanging State. Detailed balance State of System:
13. Extended Ensemble Methods. Slow Dynamics at First- Order Phase Transition At first-order phase transition, the longest time scale is controlled by.
Para, Dia and Ferromagnetism. Magnetization The presence (or absence) of a significant magnetic moment for atoms will govern how they respond to magnetic.
Chapter 7 Bose and Fermi statistics. §7-1 The statistical expressions of thermodynamic quantities 1 、 Bose systems: Define a macrocanonical partition.
An Introduction to Statistical Thermodynamics. ( ) Gas molecules typically collide with a wall or other molecules about once every ns. Each molecule has.
42C.1 Non-Ideal Solutions This development is patterned after that found in Molecular Themodynamics by D. A. McQuarrie and John D. Simon. Consider a molecular.
Eutectic Phase Diagram NOTE: at a given overall composition (say: X), both the relative amounts.
Computational Physics (Lecture 10) PHY4370. Simulation Details To simulate Ising models First step is to choose a lattice. For example, we can us SC,
Monte Carlo Simulation of the Ising Model Consider a system of N classical spins which can be either up or down. The total.
Boltzmann statistics, average values
Thermal Physics Heat Physics.
Applications of the Canonical Ensemble: Simple Models of Paramagnetism
Computational Physics (Lecture 10)
Prepared by: Lakhtartiya Amit A. Guided by: Mr. P. L. Koradiya
Statistical Mechanics
The units of g(): (energy)-1
Dimension Review Many of the geometric structures generated by chaotic map or differential dynamic systems are extremely complex. Fractal : hard to define.
LAWS OF THERMODYNAMICS
MAGNETIC MATERIALS. MAGNETIC MATERIALS – Introduction MAGNETIC MATERIALS.
Heat capacity of the lattice
Applications of the Canonical Ensemble:
Entropy & Energy Quality
Classical Statistical Mechanics in the Canonical Ensemble: Application to the Classical Ideal Gas.
Ising Model of a Ferromagnet
Chapter 14: The Classical Statistical Treatment of an Ideal Gas
Einstein Model of Solid
In “Adaptive Cooperative Systems” Summarized by Ho-Sik Seok
Find the numerical value of the expression. sinh ( ln 6 )
Energy Fluctuations in the Canonical Ensemble
Mixed order phase transitions
Higher National Certificate in Engineering
Presented by Rhee, Je-Keun
Repeating this process 8 more times gives us our slope field.
Advanced Molecular Dynamics
Mean Field Approximation
3.7. Two Theorems: the “Equipartition” & the “Virial”
Adaptive Cooperative Systems Chapter 3 Coperative Lattice Systems
Biointelligence Laboratory, Seoul National University
of the Canonical Ensemble: A Quantum System of Spins J
Thermodynamics and Statistical Physics
Presentation transcript:

Model systems with interaction Ising spin chain (with open boundary conditions) Classical variable 𝑆 𝑖 =±1 J 𝑆 𝑛 𝑆 𝑛+1 𝑆 𝑛−1 𝐻=− 1 2 𝑖,𝑗 𝐽 𝑖𝑗 𝑆 𝑖 𝑆 𝑗 𝐽 𝑖𝑗 = 𝐽 𝑓𝑜𝑟 𝑖,𝑗 𝑛𝑒𝑎𝑟𝑒𝑠𝑡 𝑛𝑒𝑖𝑔ℎ𝑏𝑜𝑟𝑠 0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒 𝐻=−𝐽 𝑆 1 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 𝑍= 𝛼 𝑒 −𝛽 𝐸 𝛼 = 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑒 𝛽𝐽 𝑆 1 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁

𝑍= 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑒 𝛽𝐽 𝑆 1 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 = 𝑆 2 , …, 𝑆 𝑁 𝑒 𝛽𝐽 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 − 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 = 𝑆 3 , …, 𝑆 𝑁 𝑒 𝛽𝐽 1+ 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −1+ 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −1− 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 1− 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 = 𝑆 3 , …, 𝑆 𝑁 (𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 )( 𝑒 𝛽𝐽 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 ) = 2cosh 𝛽𝐽 𝑆 3 , …, 𝑆 𝑁 ( 𝑒 𝛽𝐽 𝑆 3 + 𝑆 3 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −𝑆 3 +𝑆 3 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 ) = 2cosh 𝛽𝐽 𝑆 4 , …, 𝑆 𝑁 ( 𝑒 𝛽𝐽 1+ 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −1+ 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −1− 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 1− 𝑆 4 …+ 𝑆 𝑁−1 𝑆 𝑁 ) = 2cosh 𝛽𝐽 𝑆 4 , …, 𝑆 𝑁 (𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 )( 𝑒 𝛽𝐽 𝑆 4 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −𝑆 4 + …+ 𝑆 𝑁−1 𝑆 𝑁 ) = 2cosh 𝛽𝐽 2 𝑆 4 , …, 𝑆 𝑁 ( 𝑒 𝛽𝐽 𝑆 4 + …+ 𝑆 𝑁−1 𝑆 𝑁 + 𝑒 𝛽𝐽 −𝑆 4 + …+ 𝑆 𝑁−1 𝑆 𝑁 )

𝐺=− 𝑘 𝐵 𝑇 ln 2 − 𝑘 𝐵 𝑇 𝑁−1 ln 2cosh 𝛽𝐽 𝑍= 2cosh 𝛽𝐽 𝑁−2 𝑆 𝑁 𝑒 𝛽𝐽 𝑆 𝑁 + 𝑒 −𝛽𝐽 𝑆 𝑁 = 2cosh 𝛽𝐽 𝑁−2 𝑆 𝑁 2 cosh 𝛽𝐽 𝑆 𝑁 = 2cosh 𝛽𝐽 𝑁−2 𝑆 𝑁 2 cosh 𝛽𝐽 𝑍=2 2cosh 𝛽𝐽 𝑁−1 𝐺=− 𝑘 𝐵 𝑇 ln 2 − 𝑘 𝐵 𝑇 𝑁−1 ln 2cosh 𝛽𝐽 𝑔= 𝐺 𝑁 𝑁→∞ =− 𝑘 𝐵 𝑇 ln 2cosh 𝛽𝐽 Entropy per spin: s= 𝑆 𝑁 =− 𝜕𝑔 𝜕𝑇 = 𝑘 𝐵 ln 2cosh 𝛽𝐽 − 𝑘 𝐵 𝑇 2 sinh 𝛽𝐽 2cosh 𝛽𝐽 𝐽 𝑘 𝐵 𝑇 2 ln 2 𝑠 𝑘 𝐵 𝑇 𝐽 𝑘 𝐵 =1 𝐾 For s= 𝑘 𝐵 ln 2cosh 𝛽𝐽 − 𝐽 𝑇 tanh 𝛽𝐽 T->∞ , 𝛽→0 s= 𝑘 𝐵 ln 2 1 s ≈𝑘 𝐵 ln e 𝛽𝐽 − 𝐽 𝑇 =0 T->0 , 𝛽→∞ 2

Internal energy per spin: We recall from thermodynamics 𝑈=𝐺+𝑇𝑆+ 𝜇 0 𝑉𝑀𝐻 which reduces for H=0 to 𝑈=𝐺+𝑇𝑆 𝑢= 𝑈 𝑁 =𝑔+𝑇𝑠 =− 𝑘 𝐵 𝑇 ln 2cosh 𝛽𝐽 + 𝑘 𝐵 𝑇 ln 2cosh 𝛽𝐽 − J tanh 𝛽𝐽 𝑇 𝑢 𝐽 𝑘 𝐵 =1 𝐾 For 𝑢=− J tanh 𝛽𝐽 T->∞ , 𝛽→0 𝑢=0 1 T->0 , 𝛽→∞ 𝑢=-J 2 𝑐 𝑇 Heat capacity per spin: 𝑐= 𝜕𝑢 𝜕𝑇 =− J 𝜕 𝜕𝑇 tanh 𝛽𝐽 = k B 𝐽 𝑘 𝐵 𝑇 cosh 𝐽 𝑘 𝐵 𝑇 2

Spin-spin correlation function 𝑺 𝒂 𝑺 𝒂+𝒃 measures the statistical correlation between the spins (random variables) at position a and a+b a a+b Intuitive properties 𝑺 𝒂 𝑺 𝒂 =1 No interaction between spins (spins are uncorrelated) 𝑺 𝒂 𝑺 𝒂+𝒃 = 𝑺 𝒂 𝑺 𝒂+𝒃 =0 for 𝑏≠0 Interaction creates finite T-dependent correlation on characteristic length scale 𝜉(𝑇) Let’s calculate 𝑺 𝒂 𝑺 𝒂+𝒃 for chain of nearest neighbor interacting Ising spins 𝑺 𝒂 𝑺 𝒂+𝒃 = 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑆 𝑎 𝑆 𝑎+𝑏 𝑒 𝛽𝐽 𝑆 1 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑒 𝛽𝐽 𝑆 1 𝑆 2 + 𝑆 2 𝑆 3 + …+ 𝑆 𝑁−1 𝑆 𝑁

To calculate 𝑺 𝒂 𝑺 𝒂+𝒃 we use a few tricks Let’s switch from 𝑯=−𝑱 𝒊=𝟏 𝑵−𝟏 𝑺 𝒊 𝑺 𝒊+𝟏 to the more general Hamiltonian 𝑯=− 𝒊=𝟏 𝑵−𝟏 𝑱 𝒊 𝑺 𝒊 𝑺 𝒊+𝟏 The partition function for the generalized Hamiltonian reads 𝑍=2 𝑖=1 𝑁−1 2cosh 𝛽 𝐽 𝑖 Let’s first calculate the correlation function 𝑺 𝒂 𝑺 𝒂+𝟏 S a S a+1 = 1 𝑍 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑆 𝑎 𝑆 𝑎+1 𝑒 𝛽 𝑖=1 𝑁−1 𝐽 𝑖 𝑆 𝑖 𝑆 𝑖+1 = 1 𝑍 1 𝛽 𝜕 𝜕 𝐽 𝑎 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑒 𝛽 𝑖=1 𝑁−1 𝐽 𝑖 𝑆 𝑖 𝑆 𝑖+1 = 1 𝑍 1 𝛽 𝜕𝑍 𝜕 𝐽 𝑎 = 𝛽 sinh 𝛽 𝐽 𝑎 cosh 𝛽 𝐽 𝑎 1 𝛽 =tanh𝛽 𝐽 𝑎 For the special case 𝐽 𝑎 =𝐽 we obtain S a S a+1 =tanh𝛽𝐽

Let’s now calculate the correlation function 𝑺 𝒂 𝑺 𝒂+𝟐 Per definition S a S a+2 = 1 𝑍 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑆 𝑎 𝑆 𝑎+2 𝑒 𝛽 𝑖=1 𝑁−1 𝐽 𝑖 𝑆 𝑖 𝑆 𝑖+1 We use the trick 𝑆 𝑎+1 𝑆 𝑎+1 = 𝑆 𝑎+1 2 =1 𝑆 𝑎 𝑆 𝑎+2 = 𝑆 𝑎 ∗𝟏∗𝑆 𝑎+2 = 𝑆 𝑎 𝑆 𝑎+1 𝑆 𝑎+1 𝑆 𝑎+2 S a S a+2 = 1 𝑍 𝑆 1 , 𝑆 2 , …, 𝑆 𝑁 𝑆 𝑎 𝑆 𝑎+1 𝑆 𝑎+1 𝑆 𝑎+2 𝑒 𝛽 𝑖=1 𝑁−1 𝐽 𝑖 𝑆 𝑖 𝑆 𝑖+1 = 1 𝑍 1 𝛽 2 𝜕 2 𝑍 𝜕 𝐽 𝑎 𝜕 𝐽 𝑎+1 =tanh𝛽 𝐽 𝑎 tanh𝛽 𝐽 𝑎+1 For the special case 𝐽 𝑎 = 𝐽 𝑎+1 =𝐽 T->0 S a S a+𝑏 𝑏 we obtain S a S a+2 = tanh 𝛽𝐽 2 S a S a+𝑏 = tanh 𝛽𝐽 𝑏

Correlation length The last figure suggests: Correlation between spins decays with increasing separation of the spin on a characteristic T-dependent length scale we will call 𝝃(𝑻) =𝑒 − 𝑏 𝜉 𝑇 S a S a+𝑏 = tanh 𝛽𝐽 𝑏 =𝑒 ln tanh 𝛽𝐽 𝑏 =𝑒 𝑏 ln tanh 𝛽𝐽 tanh(x) 𝜉 𝑇 = 1 ln tanh 𝛽𝐽 Note that tanh 𝛽𝐽 <1 and thus ln tanh 𝛽𝐽<0 Discussion T->∞ , 𝛽→0 tanh 𝛽𝐽 →0 ln tanh 𝛽𝐽 →∞ 𝜉 𝑇→∞ →0 1 tanh 𝛽𝐽= 𝑒 𝛽𝐽 − 𝑒 −𝛽𝐽 𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 = 𝑒 𝛽𝐽 − 𝑒 −𝛽𝐽 𝑒 𝛽𝐽 1+ 𝑒 −2𝛽𝐽 = 1− 𝑒 −2𝛽𝐽 1+ 𝑒 −2𝛽𝐽 T->0 , 𝛽→∞ 2 ≈ 1− 𝑒 −2𝛽𝐽 1+ 𝑒 −2𝛽𝐽 ≈ 1− 𝑒 −2𝛽𝐽 2 ≈1− 2𝑒 −2𝛽𝐽 𝜉 𝑇→0 → 1 2 𝑒 2𝛽𝐽 ln tanh 𝛽𝐽≈ ln (1− 2𝑒 −2𝛽𝐽 )≈ 2𝑒 −2𝛽𝐽

At Tc fluctuations on all length scales including Significance of 𝝃(𝑻) is more transparent for higher spatial dimensions (2d, 3d) where 𝝃(𝑻) diverges at finite temperature Tc At Tc fluctuations on all length scales including length scale of the entire sample 𝝃 𝑻= 𝑻 𝑪 →∞ W𝐢𝐭𝐡 𝝃 𝑻 also the susceptibility 𝝌 𝑻 diverges because there is a general relation between 𝝌 𝑻 and the sum of spin-spin correlation functions S 𝐢 S 𝐣 𝜒 𝑇 = lim 𝐻→0 𝜕𝑀 𝜕𝐻 ∝ lim 𝐻→0 𝜕 𝜕𝐻 𝑎 𝑆 𝑎 ∝ lim ℎ→0 𝜕 𝜕ℎ 𝑎 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 + 𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 = lim ℎ→0 𝑎 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍− 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 + 𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍 2 = lim ℎ→0 𝑎 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍 − 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 + 𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍 2

with const= 𝑁 𝑉 𝑔 𝜇 𝐵 2 𝜇 0 𝐽(𝐽+1) 3 = lim ℎ→0 𝑎 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍 − 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝛽 𝑖 𝑁 𝑆 𝑖 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 +𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑆 1 , 𝑆 2 …, 𝑆 𝑁 𝑆 𝑎 𝑒 𝛽𝐽 𝑖 𝑁 𝑆 𝑖 𝑆 𝑖+1 + 𝛽ℎ 𝑖 𝑁 𝑆 𝑖 𝑍 ∗ 𝑍 ∝ 𝑖 𝑁 𝑆 𝑖 ∝ 𝑀 for h->0 𝑆 𝑎 ->0 for h->0 No long range order for the Ising chain 𝜒 𝑇 =const 𝛽 𝑁 𝑎,𝑖 𝑆 𝑎 𝑆 𝑖 with const= 𝑁 𝑉 𝑔 𝜇 𝐵 2 𝜇 0 𝐽(𝐽+1) 3 Let’s see how this expression reproduces paramagnetic susceptibility in the limit J=0 1 𝑁 𝑎,𝑖 𝑆 𝑎 𝑆 𝑖 = 1 𝑁 𝑎=𝑖 𝑆 𝑎 𝑆 𝑖 + 1 𝑁 𝑎 𝑖≠𝑎 𝑆 𝑎 𝑆 𝑖 =1+ 1 N 𝑎 2 𝑖 =1 𝑁 𝑆 𝑎 𝑆 𝑎+𝑗 The 2 originates from the fact that terms such as appear twice in the a,i-double sum 𝑆 1 𝑆 2 = 𝑆 2 𝑆 1 =1+2 𝑗=1 𝑁 tanh 𝛽𝐽 𝑗

Let’s have a look to the limit of a long (relative to the correlation length) Ising chain 𝑁→∞ 1+2 𝑗=1 ∞ tanh 𝛽𝐽 𝑗 =1−2+2 𝑗=0 ∞ tanh 𝛽𝐽 𝑗 1 𝑁 𝑎,𝑖 𝑆 𝑎 𝑆 𝑖 → =1+2 1 1− tanh 𝛽𝐽 −1 =1+2 tanh 𝛽𝐽 1− tanh 𝛽𝐽 =1+2 𝑒 𝛽𝐽 − 𝑒 −𝛽𝐽 𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 1− 𝑒 𝛽𝐽 − 𝑒 −𝛽𝐽 𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 =1+2 𝑒 𝛽𝐽 − 𝑒 −𝛽𝐽 𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 − 𝑒 𝛽𝐽 + 𝑒 −𝛽𝐽 =1+ 𝑒 2𝛽𝐽 −1 𝜒 𝑇 =const 𝛽 𝑁 𝑎,𝑖 𝑆 𝑎 𝑆 𝑖 =const 𝛽 𝑁 𝑒 2𝛽𝐽 𝜒 𝑇 ∞ 1 𝑇 For J=0 this indeed reduced to the Curie-law of interaction free paramagnets

For J0 it is interesting to discuss the cases of ferro and antiferromagnetic coupling 𝜒 𝑇 T J 𝜒∝ 𝑒 2𝐽 𝑘 𝐵 𝑇 𝑇 J<0 antiferromagnetic coupling 𝜒 𝑇 T J=-1/2 J=-1 𝜒∝ 𝑒 − 2 𝐽 𝑘 𝐵 𝑇 𝑇