This action might not be possible to undo. Are you sure you want to continue?

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

In the previous chapter, two new types of nanometer rod-shaped precipitates were observed; they were named QP and QC and seemed to be precursors of the stable QAl5Cu2Mg8Si7. Before a complete TEM and HREM study of these phases, subject of the next chapter, an introduction to ordering mechanisms is required. Order-disorder transitions will be introduced in the global framework of phase transitions (solid-liquid-gas, ferro-para magnetic, ferro-para electric, superfluids, polymers), without enlarging the presentation to critical phenomena. Most of the approach presented in this chapter is based on the simple following thermodynamic concept: for a closed system in thermal equilibrium, the transition is a consequence of a compromise: the energy tends to order and the entropy associated to the temperature tends to break the order. Different classifications of phase transitions will be presented in section 5.1. Phenomenological Landau’s approach by thermodynamics will be treated in section 5.2. A more general approach by using statistical mechanics on an Ising model, as well as Monte Carlo simulations, will be treated in section 5.3. It will help us to introduce the order parameters and approximate methods such as the Bragg-Williams method. Since we are interested in disordered nano-precipitates present in a matrix, the most appropriate observation means, i.e. TEM diffraction and HREM will be treated in section 5.5 to show their potential applications for the study of ordering mechanisms.

77

5. Introduction to Order-Disorder Transitions ______________________________________________________________________

**5.1 Classification of the Phase Transitions
**

5.1.1 Chemistry A microscopic approach by crystal chemistry can provide a basis for the classification of the phase transitions [118]. If a solid undergoes a phase transition at a critical temperature Tc by absorbing thermal energy, the transformed phase possesses higher internal energy, the bonding between neighboring atoms or units are weaker than in the low-temperature phase.This results in a change in the nature of the first and second-nearest neighbor bonds. Phase transition in solids may be classified into three categories: (1) Displacive transitions [119] proceed through a small distortion of the bonds (dilatational or rotational). The atomic displacements are reduced to 0.01-0.1Å and the specific heat is low (few J/g). The main characteristic is the group-subgroup relationship between the phases. This permits for example to clearly define an order parameter used for the thermodynamical description of the transition. These transitions can be of the first or second order (these terms will be explained in the next section). (2) Reconstructive transitions [120] proceed through the breaking of the primary or secondary bonds. These transitions were firstly described by Buerger [121]. They imply large atomic displacements with 10-20% of distortion of the lattice, the specific heat is important (~kJ/g). These transformations are sluggish since the barrier of energy is high. The main characteristic is the absence of any group-subgroup relationship between the phases contrarily to the case of Landau transitions (section 5.2). The transitions can even increase the symmetry of the high temperature phase. This transition occurs in many materials such as ZnS, C, H2O, Am, C, SiO2, TiO2. Bain transitions (BCC-FCC) and Buerger transitions (BCC-HCP) can be described as reconstructive. (3) Order-disorder transitions proceed through substitution between atoms possibly followed by small atomic displacements. They are commonly found in metals and alloys but also in some ceramics. Some of them keep a group-subgroup relationship, as for the CuZn transition (between BCC and simple cubic SC structure), others are also reconstructive as for Am, Fe, Co, ZnS or SiC (FCC-HCP). These transitions can be described with the help of a latent lattice common to the phases [120]. 5.1.2 Thermodynamics Let us consider a closed, isochore and diathermic system in thermal contact with a heat bath. This system is characterized by its free energy F (minimum at equilibrium), given by F = E - T.S (5.1) E is the internal energy and depends on the bonding between the atoms. S is the entropy, characteristic of the disorder by S = kB.Log Ω (Ε), where Ω is the complexion number, i.e. number of configurations of the system for a given energy E. At low temperatures, the

78

S which is maximum for a disordered system.2). The development of the free energy keeps therefore only the even exponents of ξ α(T) 2 β(T) 4 γ (T) 6 F ( T.S component in F).2) Let us assume that β and γ do not depend on the temperature. Higher order transitions would involve further differential quantities. it appears that T is the balancing coefficient between order and disorder: a phase transition must exist at a critical temperature Tc. depending on a parameter called the order parameter. The order of the transition is the same than the order of the derivation of G required to obtain a discontinuity: ∂G ∂G If V = or – S = has a discontinuity. 5. or the percentage of atoms that are on their right sublattice for an order-disorder transition (for this type of transition. heat capacity C p .5.1). Actually the system must be considered as a canonical ensemble (section 5.ξ 4 6 2 (5. It can be the magnetization for ferro-paramagnetic transition.ξ + -----------ξ + ---------. volume V. However. ξ ) = F 0 ( T ) + ----------. we must have γ > 0. The main property of the free energy is to remain unchanged by the symmetry operations of the highest symmetric phase implied in the transition. ∂p T ∂T p ∂ G Cp ∂V If 2 = = – V β . Ehrenfest classified the phase transitions in function of the thermodynamic quantities that present a discontinuity. This parameter is characteristic of the degree of order. Since F is an increasing function with ξ at high temperatures (preponderance of the T. the transition is of first order. the polarization for ferro-paraelectric transition. if β ≤ 0. thermodynamic quantities like entropy S. If β > 0. The theory is based on the assumption that the free energy of the system is a continuous function that can be developed in a Taylor series near the critical temperature Tc. the volume thermal expansivity α or the compressibility β can undergo discontinuity. details will be given in section 5. In this first approach we have voluntarily neglected the fact that the internal energy of the system can fluctuate. the system is driven by T. ∂ G = ∂V = V α or ∂ G = – ∂S = – ----. Therefore. expansible and diathermic systems are characterized by their free energy G which remains continuous during the phase transition.4.has a ∂ p T 2 ∂ T p ∂T p T ∂p T ∂ T ∂p ∂T p 2 2 2 discontinuity. all the terms must be taken into 79 . Closed.3. and noted ξ.2 Landau’s Phenomenological Approach A phenomenological treatment of phase transitions has been given by Landau in 1937 [122]. the transition is of second order. Landau s Phenomenological Approach ______________________________________________________________________ entropic term is negligible and the system is driven by E (negative) which has its maximum absolute value when the bonds of highest energy are formed (the system is ordered). the exponent 6 term can be ignored.2. At high temperatures.

for T ≥ Tc.1a. ξ ) = F 0 ( T ) + -------------------------ξ + -. two minima represented in Fig. 5. it exists only one 80 . These two cases are the conditions of a second and a first order transition respectively. and for ξ › 0 when T < Tc. the sign of α must change at Tc. Similarly to the precedent case. 5. This transition is a second order transition in the Ehrenfest classification. the sign of α changes at Tc and in first approximation α = α0.1c. 5. Since F is minimum for ξ = 0 when T ≥ Tc. In first approximation α = α0.1 Second Order Transitions Case β > 0. Introduction to Order-Disorder Transitions ______________________________________________________________________ consideration. and the expression of F is α0 ( T – Tc ) 2 β 4 F ( T. The stable states are given by ∂F = ξ ( α + βξ 2 + γ ξ 4 ) = 0 ∂ ξ T ∂ F 2 4 2 = α + 3βξ + 5γ ξ > 0 ∂ξ T 2 2 (5.2.5) As shown in Fig.(T − Tc) with α0 > 0.– 0 ∂T ξ ∂ T ξ 2 (5.6) It can be noticed that S and ξ are continuous at T = Tc.9) As shown in Fig. 5.2 First Order Transitions Case γ > 0 and β < 0.5.4) (5. and for T < Tc.ξ 2 4 The stable states are given by ∂F = ξ ( α + βξ 2 ) = 0 ∂ ξ T ∂ F 2 2 = α + 3βξ ≥ 0 ∂ξ T 2 (5. 5.8) (5.2.(T − Tc).∆S (Tc) can be calculated by ∂F α 2 ∂F -ξ S = – = – -.1b given by α0 ( T – Tc ) ξ = ± ------------------------β The specific heat L =Tc.7) (5. the system has one minimum at ξ = 0. it can be noticed that for T > T 2 = T c + β ⁄ ( 4γ α 0 ) .3) (5.

2d order transitions F(ξ)-F0 T > Tc T = Tc T < Tc ξ 1 ξ 0 T Tc (a) 1st order transitions F(ξ)-F0 T > T2 T < T2 T = T1 T < Tc ξ ξ 1 (b) 0 T Tc T1 T2 (c) (d) Fig. This situation is generally observed with a thermal hysteresis. This phase becomes stable as soon as F .10) It can be noticed that a temperature range ∆T = T2-Tc exists where the two phases can coexist.5. Just below T2.2. it appears another metastable phase corresponding to ξ › 0 (phase II).7)) are discontinuous at T1.d) second and first order transitions respectively: (a. 5. the phase I becomes metastable until Tc is reached for T < Tc. The order parameter ξ.c) the free energy curves in function of the order parameter and (b. metastable and unstable. which corresponds to ξ = 0 (phase I).1 Landau’s treatment of phase transitions: (a. as well as the specific heat L (equation (5. This transition is a first order transition in the Ehrenfest classification. Landau s Phenomenological Approach ______________________________________________________________________ phase. 81 .d) the order parameter curves in function of temperature. only phase II exists.1d – β + ( β – 4α 0 ( T – T c ) ) ξ = ± ----------------------------------------------------------------2γ 2 1⁄2 (5. 5. The order parameter corresponding to the metastability (T1 < T < T2) or to the stability (T < T1) of phase II is represented in Fig.b) and (c. Each of them are successively stable. Below T1.F0 = 0 obtained for 2 T = T 1 = T c + 3β ⁄ ( 16γ α 0 ) .

e. i. Indeed.11) The constant of proportionality Z is called the canonical partition function. Introduction to Order-Disorder Transitions ______________________________________________________________________ 5. The <> denote the thermal average.T). 5. up and down. let us imagine a long chain of N ordered spins.3. but the high number of configuration impedes the calculations of an explicit expression of the partition function when the dimension of the system is d ≥ 3. V (5. N.e. the average on all the configurations with their respective probability.13) F = – k B T log Z 〈 E〉 = ∑ Px Ex x ∂ = – ln Z ∂β N. one can easily show that there is no phase transition in the absence of any external field.3. V ) = -------∑ exp – kBxT x E (5.14) where β = 1/(k B . V = cst) and diathermic system in thermal contact with a heat bath is in equilibrium when its free energy F is minimum. defined as a canonical ensemble [125]. However this model is very interesting because it brings most of the important and basic ideas about phase transitions. The simplest coupling is then introduced by considering that the nearest neighbor spins interact: a pair of parallel spins has an energy -J and a pair of antiparallel spins has an energy J. i. it does not depend on the specific state of the system and is determined by the normalization requirement Z ( T.5. Its internal energy can fluctuate and actually the system must be considered as a set of all the microstates. The free energy required to create a simple antiphase boundary. Of course it is quite easy to calculate the energy of each configuration for a finite system. further noted 1 and 1 respectively.exp – -------- k B T Z (5. Let us consider a simple lattice and suppose that there is a magnetic moment at each lattice site which can only have two orientations along a given direction.1 Canonical Ensembles A closed isochore (N. The probability Px that the system has the energy Ex (and is in a configuration x) is given by the Boltzmann distribution law Ex 1 P x = -.3 Statistical Mechanics Approach 5.2 The Ising Model The Ising model is probably the simplest statistical model whose solution is not trivial. For d = 1. the free energy difference between two possible 82 . It was introduced by Lenz and Ising in 1925 [123].12) This partition function is characteristic of the thermodynamics of the system since the free energy and the average energy can be deduced from it by (5.

1 . subject of this section. A trajectory is a chronological sequence of configurations for a system.... For the case d = 2. The exact rigorous solution is far more difficult to obtain and was given only in 1944 by Onsager: kB. 1 . σN. 83 . T T t=1 T (5... σ3.17) This means that trajectories should be ergodic and constructed in such a way that the Boltzmann distribution law is in agreement with the relative frequencies with which the 1.3 Monte Carlo Simulations There are two general classes of simulation. and the trajectory is formed by integrating Newton’s equations of motion. N ). The other. The origin of the name comes from the city in south of France well known for its roulette and other hazard games.5. The usual Ising model can be generalized by considering the nth nearest interaction. σ3. This procedure is more generally applicable than molecular dynamics in that it can be used to study quantum systems and lattice models as well as classical assemblies of molecules.1..1.3.1. one considers a classical dynamical model for atoms and molecules.1)...15) which is always negative for any reasonable value of N and T.. 126]. Since. A phase transition is therefore possible at kBTc of the order of J [118]..3. For more simplicity. This implies that the disordering occurs spontaneously in 1-D system.kBT lnN (5. σN). so that the thermal average value of a property P follows 1 〈 P〉 = lim -. 1 ) is J.J and the entropy of the order of ln( 3 . One is called the molecular dynamics method. the energy required to create an antiphase boundary is of the order of N N . σ2... there are N simple states in antiphase boundary configuration. 1 ..1. Let us call x a point in the N-dimensional configurational space (also called phases space) obtained along the trajectory at time t: x(t) = (σ1. A configuration of a lattice Ising magnet is the list of spin variables σ1. for example. the Monte Carlo method will be discussed on the base of the magnetic Ising model [125... The procedure provides dynamical information as well as equilibrium statistical properties.m) nearest neighbors. Let us denote H({σn}) the energy of a configuration characterized by the spin numbers σn in the presence of a magnetic field h 1 H ( { σ n } ) = – -..∑ P x ( t ) .Tc = 2.1. The aim of the Monte Carlo method is to simulate trajectories representative of the thermal equilibrium state of the system. σ2.1.1) and (1. is called the Monte Carlo method1. Statistical Mechanics Approach ______________________________________________________________________ kinds of configurations: (1. 1 . Here....2692 |J| [124].1.16) (5.. m n where Jnm is the pair energy of the (n .1.. 5.∑ J nm σ n σ m – h ∑ σ n 2 n. x(t) = (1. the free energy change is ∆F = J .....

the move is accepted only with an exponential probability which depends on both temperature and difference of energy. Details are given by Ducastelle [128].= exp -------------..1..26).. which associated to (5. produced by flipping a spin in the x state. 1 . they can never be definitive.19) kB T w x'x p x' Provided a trajectory obeys this condition. the statistics acquired from this trajectory will coincide with those obtained from a canonical ensemble. Usually the variation of energy produced by a spin flipping or by an exchange of atoms is calculated with the multiple interaction energies between the spins with equation (5. because the trajectory can be blocked in a local minimum surrounded by large energy barriers. Let us define a given cluster αi = {n1. given by the Boltzmann’s law (5.1..Ex governs the relative probability of this change through the Boltzmann distribution law.).16) or between the atoms (as detailed in the next section) with equation (5. In the Metropolis algorithm [127]. At T = 0K.18) At equilibrium in the canonical ensemble. The energy associated to a possible change of state ∆Exx’ = Ex’ . respectively px and px’. Monte Carlo method makes possible to predict the different kinds of thermodynamically stable structures in function of temperature. 1 . the following particular values of wxx’ have been chosen 1.. Unfortunately. a certain degree of experimentation is always required to indicate whether the statistical results are reliable..20) That is if ∆Exx’ ≤ 0. the problem is reduced to the minimization of the internal energy (ground state) and can be treated analytically.) and x’ = (. . Lets call wxx’ this probability of change per unit time.1. The two states x and x’ have a probability of existence. nri} a given set of lattice sites (the index i to specify the type of the 84 .4 Phase Diagrams Let us chose the generalized multi-body Ising model.. (5... Depending on the resulted values. d px = 0 . Introduction to Order-Disorder Transitions ______________________________________________________________________ different configurations are visited. the move is accepted and if ∆Exx’ ≥ 0.3. ∆E xx' ≤ 0 = – ∆E xx' exp -------------.11). ∆E xx' > 0 kB T w xx' (5. 5..1.11) results in dt – ∆E xx' w xx' px -------.1.. Let us consider a trajectory going through two states x = (.5. 1 . And while the indications can be convincing. Indeed. The evolution of the states probabilities follows the Master equation d px = dt ∑ [ – wxx' px + wx'x px' ] x' (5.. some problems may arise when the system is sluggish such as in substitutional transitions.= ---.

triangle.[ 1 + x 1 ( σ n + σ m + σ p ) + x 2 ( σ n σ m + σ m σ p + σ n σ p ) + x 3 σ n σ m σ p ] 8 (5. it takes the form 1 ρ { n. The solutions {xi}0 give a configuration of the clusters that constitute the phases. p } = -.vi. These ones were given by Kanamori [129]. For example. and the minimization of the energy. and depend on the Xi values (linked to the bonding energies {Vnm}i). 85 . A general expression of the relationships was given by Ducastelle [128] who exprimed the probability of finding a type-i cluster ρi by 1 ρ i = -----.28). <pn(1-pm)>. being probabilities. Some relationships exist between the xi. <pnpm(1-pl)>. The problem can be solved by linear programming. Equation (5..23) where |α| is the number of sites in the cluster α.24) We must minimize the energy given by equation (5..σnri. A simple geometrical solution was given by Kudo and Katsura in 1976 [130]: the inequalities given by (5.23) looks complicated but its application is easy and direct.2 that an important and interesting problem which occurs for some lattices is the construction of periodic structures with some of the cluster solution {xi}0. <(1-pn)pm>.. m. It will be seen in section 6.23) define a polyhedra in the xi space.4. which is a linear form of the xi. The relationships between the xi can be obtained by expressing the fact that <pnpm>. for triangular clusters. These lattices are called frustrated. Let us define its occupation number σα = σn1. should be positive and lower than 1.5. The construction may be impossible and may imply a degeneration of the solution in infinite approximate solutions.3. .22) where Xi = ri.. The Hamiltonian (5. <(1-pn)(1-pm)>.21) i where νi only depends on the bonding energies {Vnm}i limited to the size of the type-i cluster. pn and σn are linked by equation (5. Statistical Mechanics Approach ______________________________________________________________________ cluster: pair.). where ri is the number of type-i clusters in the system. Noting xi the correlation function xi = < σ αi >. the energy of the system is 〈 E〉 = 〈 H〉 = ∑ Xi xi i (5.16) can be written in a generalized way by H ( { σn } ) = ∑ να σα i i (5. etc.23).22) respecting the linear inequalities given by 0 ≤ ρ i ≤ 1 with ρ i given in equation (5. and <pnpmpl>. 1 + ∑ x j ∑ σ βj α 2 j ≠ ∅ βj ⊂ αi (5. is obtained for one of the vertices of the polyhedra.. Let us note pn the number equalling 1 if σn = 1 and 0 if σn = 1..

µ – -.4 Ordering in Binary Alloys 5. 132]. Introduction to Order-Disorder Transitions ______________________________________________________________________ 5. µ = µ A . which is not the case for the number of up or down spins). Let us define the parameter pn pn = pnA = 1 0 if the site n is occupied by the atom A if the site n is occupied by the atom B (5. The two expressions (5. ferroelectrics. µ i is the chemical potential of the element i.m) (5. superfluids.∑ V nm p n p m – µ ∑ p n 2 n.29) (5.2 Order Parameters During an order-disorder transition in an alloy. It plays a central role in the unification quest (grand unified theory. and V ij= ij ∑ V ( n ) .∑ V ( n ) 2 4 n 1 J nm = – -.4.µ B + V BB .25) The expression of the Hamiltonian is (apart from an irrelevant constant) [128] 1 H ( { p n } ) = -. An A-B change corresponds to a spin flip.5.26) where Vnm = V(n-m) represents the energy of the creation of an A-B pair separated by (n . This equivalence can be generalized to many physical phenomena called critical phenomena1. which would have been subject to a symmetry breaking during its expansion and cooling after the Big Bang [133]. the Ising lattice is the Bravais lattice of the highest symmetry phase. Critical phenomena is certainly one of the most interesting branches of modern physics.16) and (5. m n (5.28) (5.26) are formally equivalent. 86 . superstring theory): our actual universe with 4 forces would come from a hot and condensed universe governed by one force.V nm 4 (5.1 Equivalence with the Ising Model For a AB binary alloy which features an order-disorder phase transition. It gives the same fundamental base of a priori many different phenomena. transitions in polymers.m) nearest neighbors occupied by i and j atoms. It can even be applied to the quarks bond in protons and neutrons [131.2Vnm AB where Vnm ij = V ij (n-m) is the energy of the (n . 5.V AB. such as superconductivity.30) It can be noticed that an order-disorder transition in a binary AB alloy is equivalent to a magnetic transition where h › 0 (because the number of atom A and B must remain constant. with n σ n = 2p n – 1 1 1 h = -. one of the symmetry or translational 1.27) Vnm = Vnm AA + Vnm BB .4.

5. rA + wA = 1.5. 50%) with the same number of sites in each sublattice. Therefore. all the A atoms are on one of the sublattices. elements of the high temperature phase is broken and the Bravais lattice is separated into two sublattices.32) n∈β ∑ pn A where rA denotes the number of A atoms on the right α sublattice divided by the total number of A atoms NA.= 2r – 1 r+w (5. At T = 0K. and wA the number of A atoms on the wrong β sublattice divided by NA. such as the CuZn transition illustrated in Fig. At T = 0K. Indeed. let us introduce the new parameters 1 r A = ----NA 1 w A = ----NA n∈α ∑ pn A (5. For the magnetic Ising 1 model. these two parameters can be used to create an order parameter similar to the magnetization r–w ξ = ----------.28) that its value would be always cA-cB = 0. 5 . Let us illustrate the concepts of short and long order parameters with the example of an AB alloy (50%. rA = 1. cA = cB = 1/2. that will be noted α.33) ξ is called the long range order LRO parameter. 2 D i s o rd e re d ( h i g h t e m p e r a t u re ) a n d o rd e re d structure (low temperature) of CuZn. most of the A atoms remain on their right α sublattice.31) (5. Moreover. the order parameter is the normalized magnetization M/N = 〈 --. rA = wA. it is a measure of the fraction of atoms A 87 . Therefore. it follows that rA = rB + cB Nα/N = rB = r and wA = wB = w. a similar parameter is not relevant because h › 0. Ordering in Binary Alloys ______________________________________________________________________ Cu 50% Zn 50% Cu Zn F i g .∑ σ n〉 .4. By definition. For 0 < T ≤ Tc. and other A atoms can change the sublattice to go on the wrong sublattice (noted β). it can be deduced from equation (5. The broken symmetry c o r re s p o n d s t o a c h a n g e between the BCC and the SC structures. Let us call cA and cB the fractional portions of A and B atoms (cA = cB = 1/2).2. and the number of sites of each sublattice are equal (Nα = Nβ = N/2). For the orderN disorder transition. at T = ∞ .

let us define rn – wn ξ n = ---------------. Similarly to equation (5. Introduction to Order-Disorder Transitions ______________________________________________________________________ sitting on their right sublattice site.33).. Similarly to equation (5.= 2r n – 1 rn + wn (5. and is usually referred to as the B-W method.The model is based on the simplification provided by the assumption that ξ1 = ξ2 =. Since the atoms always tend to have for nearest neighbors f the right ones.34) which corresponds to the number of right nth nearest neighbors divided by 2Nzn the number of pairs formed by the nearest neighbors of the system (independently whether A is in the right or wrong position).. becomes ξ ∞ = 2r ∞ – 1 = ξ 2 2 2 (5.3 Approximate Methods There exist many methods for treating the phase transitions on an Ising lattice. 0 < ξ < 1 for 0 < T ≤ Tc. The first one was introduced by Bragg and Williams in 1934 for the order-disorder transition [134]. 5. The mean number of the 1st nearest neighbor A-B bonds is by definition NAB = (N/2) z r1. It corresponds to the pair correlation coefficient between the nth nearest neighbors. and 0 for T > Tc. Its value is ξ = 1 at T = 0 K. Here again rn + wn = 1. with equation (5.31). It means that all the sites of a sublattice α or β have the same probability to be occupied by A or B independently of their neighbors.5.There is absolutely no correlation length (in magnetic terms it is the maximum distance over which the flip of a spin has no more influence on the other spin). Moreover. for very large distances between sites n → ∞ . This method neglects all the fluctuations of the system.36) (5.. = ξ . let us introduce the parameter 1 r n = ----------2Nz n ∑ (j ∈ α)(j + n ∈ β) (j ∈ β)(j + n ∈ α) pj pj + n A B (5..33). This is similar to the Weiss’s method for the treatment of ferro-paramagnetic transitions [135]. the probability of finding a right pair r n becomes independent of n and can be expressed in function of the probability of finding a pair of atoms both on their right site or both on their wrong site. we have ξ1 ≥ ξ2 ≥ . where z is the 88 . Equation (5.≥ ξn.34) therefore becomes r∞ = r + w .4. Let us illustrate this method with the CuZn phase transition. which.37) The long range order parameter corresponds to the square of the limit of the short range order parameter series.35) ξn is a short range order SRO parameter.

5. the entropy of configuration can be simplified kN S = – ------ ⋅ [ r A ln r A + ( 1 – r A ) ln ( 1 – r A ) + r B ln r B + ( 1 – r B ) ln ( 1 – r B ) ] 2 With equation (5. generally used for the magnetic Ising model. The B-W model is also called the mean-field approximation.42) (5. with V(1) defined in (5. using the Stirling approximation and the fact that NA = NB = N/2.⋅ [ ( 1 + ξ ) ln ( 1 + ξ ) + ( 1 – ξ ) ln ( 1 – ξ ) – 2 ln 2 ] 4 N 2 (5. and the other V(n) = 0. the mixing energy becomes E = . ( ξ + 1 ) + -------. the solution ξ = 0 is stable: the alloy is disordered. even if not completely ordered.40) (5.44) does not have any analytical solution but can be solved numerically. Ordering in Binary Alloys ______________________________________________________________________ number of nearest neighbors (its value is 8 for the CuZn phases). which. the B-W approximation is equivalent to consider that only V(1) › 0. it has always the solution ξ = 0. ⋅ [ ( 1 + ξ ) ln ( 1 + ξ ) + ( 1 – ξ ) ln ( 1 – ξ ) – 2 ln 2 ] 2 Therefore.3.27). the complexion number in the B-W approximation is N B! N A! Ω = ----------------------------------------. and when T < Tc. Apart from a constant term corresponding to the mean field of 89 .41) To find the order parameter for the equilibrium state.= – ----. As shown in Fig. the free energy of mixing is given by kB T Vz 2 F --. Indeed.45) Equation (5. it becomes kB N S = – -------.5. is the same as to replace a fluctuating field by a mean field.33) and the fact that rA = rB = r.44) (5.N V z (ξ2 + 1) /4 Additionally.4. For T < Tc the solution ξ = 0 is unstable and the other solution is stable: the alloy is ordered.38) N = (N/2) z r = N z (ξ2 + 1) /4 AB Calling V = V(1)/2 the mean energy to form an A-B bond.ξ ⇔ ξ = tanh T c ξ ---------------- T ∂ ξ T 1 – ξ Tc where Vz T c = ------2k B (5. it has also a second distinct solution. There is no discontinuity of the order parameter when T goes through Tc: it is a second order transition.39) and therefore.⋅ ----------------------------------------N A r A!N A ( 1 – r A )! N B r B!N B ( 1 – r B )! (5. Under these conditions (5.29). we are looking for the minimum of the F(ξ) obtained for T ∂F = 0 ⇔ ln 1 + ξ = 2 ---. by considering equation (5.43) (5. For T ≥ Tc.

1 1.5. i. For instance. No metastable CuZn structure phase exists at room 1. Introduction to Order-Disorder Transitions ______________________________________________________________________ F(ξ) . 5. the model is more complex and must take into account the 1st-antiliberals the α-α and β-β bonds in addition to the α-β bonds. T Tc the completely disordered state. The first order character of the transition of FCC alloy can also be understood by the the fact that the FCC lattice is a very constained or frustrated lattice. Its results in a prevision of a first order transition1. Actually. Considering that the proportion r/w obeys the Boltzmann law.4 0.= exp -------- = exp -------. not easily ordered [128].44). 50%) alloy with a transition between CC and SC structures.6 0.5 y= ξ ξ .8 1 (c) 1.2 (a) ξ (b) 1 0.2 0.2 0. the CuZn BCC-SC transition is of second order whereas the AuCu3 FCC-SC transition is of first order. The B-W model can also be applied to transitions between FCC and SC structures. the result (5.39). the model can be generalized to any AB (cA.2 1. cB) alloy.e.3 Bragg-Williams treatment of orderdisorder phase transitions: (a) free energy curve in function of the order parameter.2 Fig. The B-W model is generally in good agreement with the observations of order-disorder phase transition. the expression of the equivalent mean field is zVξ h 0 = -------2 (5. 90 .46) This can also be obtained by the differentiation of (5.1 1 0.47) We have introduced the B-W in the example of AB (50%.44) can be found again 2h 0 zVξ 1+ξ r --. In this case.5 ξ 0 1 y=0 1 -2 -1 -1 0.6 0. leading to the drawing of (c) the order parameter curve in function of temperature.8 Tanh ξ 3 2 y= 1.F(0) N kBT 0. = ---------- k B T kB T 1–ξ w (5.8 0.6 1 -2 -3 2 ξ -0.9 0. (b) graphical resolution of equation (5.4 0.

The method is based on finding a link between the coupling parameters (such as J in the simple Ising model) so that the problem remains invariant. Many disordered structures can be generated along a trajectory during Monte Carlo simulations of an ordering mechanism. the diffuse scattering comes from a potential disorder in a crystal: the finite size of a particle breaks the translational and infinite order and makes the spots broader.48) 91 . Nevertheless.4. Other methods usually more effective for ordering characterization [140].2).5. the thermal motions (Debye-Waller) or small displacements of atoms around their mean positions lower the spot intensities for high G vectors. Nevertheless. like X-ray and neutron diffraction will no be presented because they are not appropriate to our study of the nanosized precipitates present with many other phases in a matrix. as it will be shown. Wilson’method is very general and has wide applicability extending well beyond the field of phase transitions.1 Diffuse Scattering and Disorder As introduced in section 3. the correlation length or the heat capacity) are not correct. The method can predict with a very high degree of accuracy the critical exponent (behavior close to Tc). This method has been generalized by considering the polyhedra interactions. It gives very good approximations for the critical exponents. 5. HREM observations and simulations of disordered particles. in contrary to the AuCu3 system. 5. Another very performing method is based on the renormalization group theory developed by Wilson in 1972 [139] (he was award the 1982 Nobel Prize in Physics for his contribution).5.5 Scattering and HREM Images of Disordered Particles This section is dedicated to the diffraction.2.e. Scattering and HREM Images of Disordered Particles ______________________________________________________________________ temperature after quenching.5. there are many inadequacies in the B-W approximation. The critical exponents (representative of the variation with T near Tc of parameters such as the order parameters. A better approximate model has been given by Bethe which considers not only the point correlation function (as the B-W method). this has been done mainly by Kikuchi (1951) with its Cluster Variation method (CV) [137. for a binary substitutional alloy. this diffraction pattern is not enough to completely characterize the ordering mechanisms. but also the pair correlation functions [136]. but if large enough. the partition function must be unchanged during the scaling steps (for example when a group of 4 spins is replaced by one spin according to a certain averaging law). i. Their representations are not easily comparable in the direct space. The diffraction generally comes from the order. 138]. Cowley [141] has expressed the total diffracted intensity by I ( u ) = N ( fA cA + fB cB ) 2 ∑ exp [ 2πiu • Rn ] + N ( fA – fB ) ∑ 〈 s0 sn〉 exp [ 2πiu • Rn ] 2 2 n n (5.4. which. depends on the correlations that exist between the substitutions (as shown in section 5. they exhibit the same diffraction patterns.

51) The parameter αnm =<snsm>/cAcB is called the Warren-Cowley parameter. (5.pnB (5. The superstructure peaks for 0 < ξ < 1 are diffuse due to the small size of the ordered clusters (linked to SRO parameter).3-dibromo-2. they are essential to unambiguously define the complete arrangement of the atoms. In equation (5.are fitted by the program. It can be noticed that this diffuse intensity part produces sharp superstructure Bragg spots when the perfect superstructure is obtained (ξ = 1). more 1. Here the brackets <> represent the spatial average over all sites1. <s n s m > is linked to the probability <P n A P m B > noted P nm AB to find the A and B atoms separated by a vector m .> cannot be obtained by diffraction experiments.25). Welberry [142] has shown that substantial differences in the multi-point lattice averages can exist without affecting the two-point correlations. but can only be used to simulate diffractions when strong physical and chemical arguments govern the model and only when few parameters -in general the interaction constants of the Monte-Carlo process.48) cannot be used to prove the unicity of a model. In a large system in thermal equilibrium. for example for 1. This description is a simplified presentation. and <s0sn> the correlation between the site 0 and n. and a uniform background for disordered structure (ξ = 0). Rn the vectors between the lattice points n. it tends to the square of the long range order ξ when the modulus of m-n tends to ∞ . The other multiple-point correlations <snsmsl.49) with pnΑ defined in (5. Their intensities decrease rapidly with u because the order is only partially obtained inside of each cluster (similar to a Debye-Waller effect).4.5. It is also often used. However. equation (5.48) noted Id(u) 〈 s 0 s n〉 = –1 1 ---------------------------. with m and n respectively on α and β. fA and fB their atomic diffusion factors. sn represents the difference between the real occupation of the site n and the mean occupation sn = pn Α . and thus proving that very different kinds of ordering clearly distinguishable in direct space can exhibit the same diffraction patterns with the same intensities. 92 . and thus to completely determine the order.48).2 because their one-point and two-point correlations can directly be obtained by inversion of the second term of equation (5. this spatial averaging is equivalent to a temporal averaging or to an averaging on all the configurations with their respective probability. Indeed.50) <snsm> are other SRO parameters. diffuse peaks or streaks if the structure just exhibits a ˙tendency towards ordering¨ (0 < ξ < 1). Therefore. Similarly to ξn. They are more often used than the ξn SRO parameters defined in 5.. Introduction to Order-Disorder Transitions ______________________________________________________________________ where N is the number of atoms.ℑ [ I d ( u ) ] 2 N ( fA – fB ) (5. the first term gives the Bragg peaks and the second term gives the diffuse scattering.6dimethylbenzene and for mullite [142].cA = cB .5-diethyl-4.n given by PnmAB = <pn Α pmB> = cAcB + <snsm>..

at least for the precipitates studied in this work.48).5. one can notice that the intensity of the diffraction pattern is proportional to the square of the modulus of the FFT of the projected potential I(u) = A(u) = K 2 2 Ω ∫ V ( r ) exp [ 2πiu • r ] dv 2 2 S 2 = K 2 r ⊥ n r || n ∫ ∫ V( r ) exp [ 2πiu • r ] dv = K ∫ Vp ( r ) exp [ 2πiu • r ] ds 2 (5. Si. These conditions are obvious for translational periodic structures (perfect crystals). 5. with a large difference between the A and B diffusion factors. from equation (3. a direct interpretation of the atomic position remains possible [145]. structures containing defects and disordered structures. it is possible to calculate from the HREM images the multiple point correlations and thus completely define the order.2 Diffuse Scattering Simulations In chapter 6.5.7). 142. 5. If the observed area of the crystal is large enough. the thin parts of the samples 93 . Moreover. However. We will study an order-disorder transition inside a precipitate family containing four elements (Al. To simulate diffuse scattering. projected potential simulations can be performed for comparisons with the experimental HREM images (if the weak phase object approximation is obtained). The advantage of the method presented here is that no periodic conditions are required due to the use of Continuous Fourier Transform to calculate the atomic potential of a single atom. noting n the zone axis.5. Indeed. permitting to distinguish two potential disordering processes producing the same point and pair correlation parameters. diffuse scattering simulations will be performed to confirm the validity of an ordering model developed for the precipitation in the Al-Cu-Mg-Si alloys. Mg. and Cu). with substitutions and atomic displacements. Even if the weak phase object approximation is not obtained.52) The projected potentials are calculated for the atomic species using the method described in annex A and integrated along the observation direction (which can be different from a zone axis). Such potential images are represented in Fig. for simple A-B substitutional transformations. but are arbitrary and require large supercell sizes for quasicrystals. Diffraction and HREM simulations usually require periodic conditions to perform Discrete Fourier Transforms.3 Filtered HREM Images In the weak phase object approximation at Scherzer defocus a HREM image is nearly the representation of the projected potential of the object. Scattering and HREM Images of Disordered Particles ______________________________________________________________________ details are given in [141. Since this transition is more complicated than the substitution ordering with two species.4.5. it will not be possible to apply equation (5. 5. 143. 144]. we have preferred to build an image of the total projected potential of the disordered structure to perform its FFT.

the atoms are disordered around the lattice of the final ordered phase. the latent lattice. Si and Cu) with substitutions and atomic displacements (see chapter 6). i. no clear pattern can be identified in the observed precipitates due to the slight disorder between each atomic plane normal to the electron beam. which means that the 94 . the HREM images were acquired in thicker parts which impeded the weak phase object approximation. and (b) projected potential calculated with annex A. Fig. 5.53) shows that applying a mask filter in the reciprocal space corresponds to average each pattern separated by the periodicity a of the direct space image with the neighboring other patterns on a distance m. The size of the mask (1/m) must be chosen to contain the diffuse intensity.e. Mg. this makes appear a pattern (of size a) only if mean positions (with random disorder around these positions) exist inside this pattern. Thus. are too amorphous to be observed. as usual in ordering phenomena. The filtered image If (u) can be written in one dimension by If ( x ) = ℑ –1 ℑI ( u ) • n = –∞ ∑ ∞ δ ( u – n ⁄ a ) ⊗ exp [ – u ⁄ m ] 2 = I(x) ⊗ ℑ –1 n = –∞ ∑ ∞ δ ( u – n ⁄ a ) ⊗ exp [ – u ⁄ m ] 2 ∞ = I(x) ⊗ πma h = –∞ ∑ δ ( x – ha ) • exp [ – mπ x ] 2 2 (5.4 Disordered structure represented in (a) illustrative model: Cu. Moreover. In addition.5. the filtering corresponds to apply to the FFT of the image (noted ℑ I(u)) a Gaussian mask (size noted 1/m [nm-1]) with the periodicity (noted 1/a [nm-1]).53) Equation (5. Introduction to Order-Disorder Transitions ______________________________________________________________________ 30 25 20 15 10 5 (a) -5 0 5 10 15 20 (b) Si. Mg. Thus. Indeed. Applying a mask filter in the reciprocal space of the HREM images allows to show an image of the mean atomic positions. the transformation is more complex than a simple A-B substitutional transformation since it involves four elements (Al. Nevertheless.

5).5. 150]. Therefore. This type of images is widely used for ordered-disordered structures [146. if it is possible. 147] such as Cu-Zn. DF superstructure images will be used to show different superordered parts in the AlCuMgSi rod-shaped precipitates (section 6. these images can be used to observe the ordered parts of a disordered precipitate.5. with 3 non-colinear spots (atomic column arrangement information) of each superstructure phase. If the information limit is sufficient to transfer the spatial frequencies.1.5. This method was used for comparisons between experimental and simulated HREM images based on a refined model of the crystal structure (section 6.4 Dark Field Superstructure Images Some slightly disordered structures can be mixed. 5. and possibly to determine the correlation length. For example.e.3. In this work. 95 . Cu-Al or Ni-Mo[148]. Dark field superstructure images consist in acquiring dark field images with two diffracted spots (single plane stacking information) or. i. AuCd. Au4Mg. Scattering and HREM Images of Disordered Particles ______________________________________________________________________ average must be performed on the correlation length distance. the disordering must occur two—dimensionally so that the crystal remains perfectly periodic in the third direction (direction of the electron beam) and so that superposition effects are eliminated. To obtain such images. AuMn [149.3).4 nm. Therefore. constituted with two or more phases corresponding to superstructures. This method can be applied to thick regions with a relatively low resolution microscope. a conventional electron microscope not necessarily dedicated to HREM is sufficient for periodicities higher than 0. the lattices of each structure appear by interference. this method is effective only for slightly disordered structures (largely below the critical transition temperature).

5. Scattering and HREM Images of Disordered Particles ______________________________________________________________________ 96 .5.

- ece4990notes2
- ece4990notes2
- Energy Levels in One Dimension
- Biological Physics
- 0812
- viva
- 409857
- ABB Power Course Load Flow
- Fundamentals of Matrix Converter
- Ieee Dictionary Terms
- yagi
- July 28 - Introduction to Structural Dynamics Dr. HW Huang
- Differential Analysis of Fluid Flow 1221
- Statistical mechanics lecture notes (2006), L16
- 04-06!11!08!57!22_DC-Link Voltage Control and Performance Analysis of StatCom
- Lec 14
- sol4_99
- e_057_02_0459
- Elasticity General
- Bohrs Correspondence principle.pdf
- chap1lecture-140418013435-phpapp01
- Lesson 1 -Introduction to Physics
- Noah Graham- An Introduction to Solitons and Oscillons
- TEORI DINAMO
- Chanwoo - Cdr Pdfs Indexed 1 1
- Mechanics
- Alsuwa Etal Jva03
- Chapter 4-Symmetry and Group Theory Applications
- Chapter 5
- Handout_4 Normal Shock Waves

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue listening from where you left off, or restart the preview.

scribd