Quantum Mechanical Addition Of Angular Momenta And Spin

3y ago
34 Views
3 Downloads
350.43 KB
42 Pages
Last View : 2m ago
Last Download : 3m ago
Upload by : Dani Mulvey
Transcription

Chapter 6Quantum Mechanical Addition ofAngular Momenta and SpinIn this section we consider composite systems made up of several particles, each carrying orbitalangular momentum decribed by spherical harmonics Y m (θ, φ) as eigenfunctions and/or spin. Oftenthe socalled total angular momentum, classically speaking the sum of all angular momenta and spinsof the composite system, is the quantity of interest, since related operators, sums of orbital angularmomentum and of spin operators of the particles, commute with the Hamiltonian of the compositesystem and, hence, give rise to good quantum numbers. We like to illustrate this for an exampleinvolving particle motion. Further below we will consider composite systems involving spin states.Example: Three Particle ScatteringConsider the scattering of three particles A, B, C governed by a Hamiltonian H which dependsonly on the internal coordinates of the system, e.g., on the distances between the three particles,but neither on the position of the center of mass of the particles nor on the overall orientation ofthe three particle system with respect to a laboratory–fixed coordinate frame.To specify the dependency of the Hamiltonian on the particle coordinates we start from the ninenumbers which specify the Cartesian components of the three position vectors rA , rB , rC of the particles. Since the Hamiltonian does not depend on the position of the center of mass R(mA rA mB rB mC rC )/(mA mB mC ), six parameters must suffice to describe the interactionof the system. The overall orientation of any three particle configuration can be specified by This eleminates three further parameters fromthree parameters1 , e.g., by a rotational vector ϑ.the dependency of the Hamiltonian on the three particle configuration and one is left with threeparameters. How should they be chosen?Actually there is no unique choice. We like to consider a choice which is physically most reasonablein a situation that the scattering proceeds such that particles A and B are bound, and particle Cimpinges on the compound AB coming from a large distance. In this case a proper choice for adescription of interactions would be to consider the vectors rAB rA rB and ρ C (mA rA mB rB )/(mA mB ) rC , and to express the Hamiltonian in terms of rAB , ρC , and ( rAB , ρ C ).The rotational part of the scattering motion is described then in terms of the unit vectors r̂AB and1We remind the reader that, for example, three Eulerian angles α, β, γ are needed to specify a general rotationaltransformation141

142Addition of Angular Momenta and Spinρ̂C , each of which stands for two angles. One may consider then to describe the motion in termsof products of spherical harmonics Y 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ) describing rotation of the compound ABand the orbital angular momentum of C around AB.One can describe the rotational degrees of freedom of the three-particle scattering process throughthe basisB { Y 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ), 1 0, 1, . . . , 1,max , 1 m1 1 ; 2 0, 1, . . . , 2,max , 2 m1 2 }(6.1)where 1,max and 2,max denote the largest orbital and rotational angular momentum values, thevalues of which are determined by the size of the interaction domain V , by the total energy E,by the masses mA , mB , mC , and by the moment of inertia IA B of the diatomic molecule A–Bapproximately as followsr2 mA mB mC E1p , 2,max 2 IA B E .(6.2) 1,max mA mB mB mC mA mC The dimension d(B) of B isd(B) 1,maxX 1 0(2 1 1) 2,maxX(2 2 1) ( 1,max 1)2 ( 2,max 1)2(6.3) 2 0For rather moderate values 1,max 2,max 10 one obtains d(B) 14 641, a very large number.Such large number of dynamically coupled states would constitute a serious problem in any detaileddescription of the scattering process, in particular, since further important degrees of freedom, i.e.,vibrations and rearrangement of the particles in reactions like AB C A BC, have noteven be considered. The rotational symmetry of the interaction between the particles allows one,however, to separate the 14 641 dimensional space of rotational states Y 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ) intosubspaces Bk , B1 B2 . . . B such, that only states within the subspaces Bk are coupled in thescattering process. In fact, as we will demonstrate below, the dimensions d(Bk ) of these subspacesdoes not exceed 100. Such extremely useful transformation of the problem can be achieved throughthe choice of a new basis setX (n)B0 {c 1,m1 ; 2 ,m2 Y 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ), n 1, 2, . . . 14 641} .(6.4) 1 ,m1 2 ,m2The basis set which provides a maximum degree of decoupling between rotational states is of greatprinciple interest since the new states behave in many respects like states with the attributesof a single angular momentum state: to an observer the three particle system prepared in suchstates my look like a two particle system governed by a single angular momentum state. Obviously,composite systems behaving like elementary objects are common, albeit puzzling, and the followingmathematical description will shed light on their ubiquitous appearence in physics, in fact, will maketheir appearence a natural consequence of the symmetry of the building blocks of matter.There is yet another important reason why the following section is of fundamental importance forthe theory of the microscopic world governed by Quantum Mechanics, rather than by ClassicalMechanics. The latter often arrives at the physical properties of composite systems by adding the

6.0: Addition of Angular Momenta and Spin143corresponding physical properties of the elementary components; examples are the total momentumor the total angular momentum of a composite object which are the sum of the (angular) momentaof the elementary components. Describing quantum mechanically a property of a composite objectas a whole and relating this property to the properties of the elementary building blocks is then thequantum mechanical equivalent of the important operation of addittion. In this sense, the readerwill learn in the following section how to add and subtract in the microscopic world of QuantumPhysics, presumably a facility the reader would like to acquire with great eagerness.Rotational Symmetry of the HamiltonianAs pointed out already, the existence of a basis (6.4) which decouples rotational states is connectedwith the rotational symmetry of the Hamiltonian of the three particle system considered, i.e.,connected with the fact that the Hamiltonian H does not depend on the overall orientation of the of the wave functions ψ( rAB , ρthree interacting particles. Hence, rotations R(ϑ) C ) defined through ψ( rAB , ρ rAB , R 1 (ϑ) ρR(ϑ) C ) ψ(R 1 (ϑ) C )(6.5)do not affect the Hamiltonian. To specify this property mathematically let us denote by H0 theHamiltonian in the rotated frame, assuming presently that H0 might, in fact, be different from H. ψ R(ϑ) H ψ. Since this is true for any ψ( rAB , ρ It holds then H0 R(ϑ) C ) it follows H0 R(ϑ)0 R(ϑ) H, from which follows in turn the well-known result that H is related to H through the H R 1 (ϑ). The invariance of the Hamiltonian under overallsimilarity transformation H0 R(ϑ)rotations of the three particle system implies then H R 1 (ϑ) .H R(ϑ)(6.6)For the following it is essential to note that H is not invariant under rotations of only rAB or ρ C ,but solely under simultaneous and identical rotations of rAB or ρ C .Following our description of rotations of single particle wave functions we express (6.5) accordingto (5.48) · J (1) exp i ϑ · J (2) exp i ϑ(6.7)R(ϑ) where the generators J (k) are differential operators acting on r̂AB (k 1) and on ρ̂C (k 2). Forexample, according to (5.53, 5.55) holdsi (1) J1 zAB yAB; yAB zABObviously, the commutation relationshipshiJp(1) , Jq(2) 0i (2) J3 ρy ρx. ρx ρyfor p, q 1, 2, 3(6.8)(6.9)hold since the components of J (k) are differential operators with respect to different variables. Onecan equivalently express therefore (6.7) i exp ϑ · JR(ϑ)(6.10)

144Addition of Angular Momenta and Spinwhere J J (1) J (2).(6.11)By means of (6.11) we can write the condition (6.6) for rotational invariance of the Hamiltonian inthe form i i H exp ϑ· J H exp ϑ·J.(6.12) 1. To order O( ϑ ) one obtainsWe consider this equation for infinitesimal rotations, i.e. for ϑ i i i i H 11 ϑ · J H 11 ϑ · J H H ϑ·J ϑ· JH .(6.13) it follows H J J H 0 or, componentwise,Since this holds for any ϑ[H, Jk ] 0 , k 1, 2, 3 .(6.14)We will refer in the following to Jk , k 1, 2, 3 as the three components of the total angularmomentum operator.The property (6.14) implies that the total angular momentum is conserved during the scatteringprocess, i.e., that energy, and the eigenvalues of J2 and J3 are good quantum numbers. To describethe scattering process of AB C most concisely one seeks eigenstates YJM of J2 and J3 which canbe expressed in terms of Y 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ).Definition of Total Angular Momentum StatesThe commutation property (6.14) implies that the components of the total angular momentumoperator (6.12) each individually can have simultaneous eigenstates with the Hamiltonian. Wesuspect, of course, that the components Jk , k 1, 2, 3 cannot have simultaneous eigenstates among each other, a supposition which can be tested through the commutation properties of theseoperators. One can show readily that the commutation relationships[Jk , J ] i k m Jm(6.15)are satisfied, i.e., the operators Jk , k 1, 2, 3 do not commute. For a proof one uses (6.9), the(n)(n)(n)properties [Jk , J ] i k m Jm for n 1, 2 together with the property [A, B C] [A, B] [A, C].We recognize, however, the important fact that the Jk obey the Lie algebra of SO(3). Accordingto the theorem above this property implies that one can construct eigenstates YJM of J3 and ofJ2 J21 J22 J23(6.16)following the procedure stated in the theorem above [c.f. Eqs. (5.71–5.81)]. In fact, we will findthat the states yield the basis B0 with the desired property of a maximal uncoupling of rotationalstates.Before we apply the procedure (5.71–5.81) we want to consider the relationship between YJM andY 1 m1 (r̂AB ) Y 2 m2 (ρ̂C ). In the following we will use the notationΩ1 r̂AB ,Ω2 ρ̂C .(6.17)

6.1: Clebsch-Gordan Coefficients6.1145Clebsch-Gordan CoefficientsIn order to determine YJM we notice that the states Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 ) are characterized by four 2 2(1) (2)quantum numbers corresponding to eigenvalues of J (1) , J3 , J (2) , and J3 . Since YJMsofar specifies solely two quantum numbers, two further quantum numbers need to be specifiedfor a complete characterization of the total angular momentum states. The two missing quantum 2 2numbers are 1 and 2 corresponding to the eigenvalues of J (1) and J (2) . We, therefore,assume the expansionXYJM ( 1 , 2 Ω1 , Ω2 ) (J M 1 m1 2 m2 ) Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 )(6.18)m1 ,m2where the states YJM ( 1 , 2 Ω1 , ρ̂C ) are normalized. The expansion coefficients (J M 1 m1 2 m2 )are called the Clebsch-Gordan coefficients which we seek to determine now. These coefficients,or the closely related Wigner 3j-coefficients introduced further below, play a cardinal role in themathematical description of microscopic physical systems. Equivalent coefficients exist for othersymmetry properties of multi–component systems, an important example being the symmetrygroups SU(N) governing elementary particles made up of two quarks, i.e., mesons, and three quarks,i.e., baryons. 2 2Exercise 6.1.1: Show that J2 , J3 , J (1) , J (2) , and J (1) · J (2) commute. Why can states YJMthen not be specified by 5 quantum numbers?Properties of Clebsch-Gordan CoefficientsA few important properties of Clebsch-Gordan coefficients can be derived rather easily. We firstnotice that YJM in (6.18) is an eigenfunction of J3 , the eigenvalue being specified by the quantumnumber M , i.e.J3 YJM M YJM .(6.19)(1)(2)Noting J3 J3 J3 and applying this to the l.h.s. of (6.18) yields using the property(k)J3 Y k mk (Ωk ) mk Y k mk (Ωk ) , k 1, 2M YJM ( 1 , 2 Ω1 , Ω2 ) X(m1 m2 ) (J M 1 m1 2 m2 ) Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 ) .(6.20)m1 ,m2This equation can be satisfied only if the Clebsch-Gordan coefficients satisfy(J M 1 m1 2 m2 ) 0for m1 m2 6 M.One can, hence, restrict the sum in (6.18) to avoid summation of vanishing termsXYJM ( 1 , 2 Ω1 , Ω2 ) (J M 1 m1 2 M m1 ) Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 ) .(6.21)(6.22)m1We will not adopt such explicit summation since it leads to cumbersum notation. However, thereader should always keep in mind that conditions equivalent to (6.21) hold.

146Theory of Angular Momentum and SpinThe expansion (6.18) constitutes a change of an orthonormal basisB( 1 , 2 ) {Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 ), m1 1 , 1 1, . . . , 1 ,m2 2 , 2 1, . . . , 2 } ,(6.23)corresponding to the r.h.s., to a new basis B0 ( 1 , 2 ) corresponding to the l.h.s. The orthonormalityproperty impliesZZdΩ1 dΩ2 Y 1 m1 (Ω1 ) Y 2 m2 (Ω2 )Y 01 m01 Ω1 ) Y 02 m02 (Ω2 ) δ 1 01 δm1 m01 δ 2 02 δm2 m02 .(6.24)The basis B( 1 , 2 ) has (2 1 1)(2 2 1) elements. The basis B0 ( 1 , 2 ) is also orthonormal2 andmust have the same number of elements. For each quantum number J there should be 2J 1elements YJM with M J, J 1, . . . , J. However, it is not immediately obvious what the J–values are. Since YJM represents the total angular momentum state and Y 1 m1 (Ω1 ) and Y 2 m2 (Ω2 )the individual angular momenta one may start from one’s classical notion that these states represent J (1) and J (2) , respectively. In this case the range of J –values angular momentum vectors J,wouldbe the interval [ J (1) J (2) , J (1) J (2) ]. This obviously corresponds quantum mechanicallyto a range of J–values J 1 2 , 1 2 1, . . . 1 2 . In fact, it holds X1 2( 2J 1 ) (2 1 1) (2 2 1),(6.25)J 1 2 i.e., the basis B0 ( 1 , 2 ) should beB2 {YJM ( 1 , 2 Ω1 , Ω2 ); J 1 2 , 1 2 1, 1 2 ,M J, J 1, . . . , J} .(6.26)We will show below in an explicit construction of the Clebsch-Gordan coefficients that, in fact, therange of values assumed for J is correct. Our derivation below will also yield real values for theClebsch-Gordan coefficients.Exercise 6.1.2: Prove Eq. (6.25)We want to state now two summation conditions which follow from the orthonormality of the twobasis sets B( 1 , 2 ) and B0 ( 1 , 2 ). The propertyZZdΩ1 dΩ2 Y JM ( 1 , 2 Ω1 , Ω2 ) YJ 0 M 0 ( 1 , 2 Ω1 , Ω2 ) δJJ 0 δM M 0(6.27)together with (6.18) applied to Y JM and to YJ 0 M 0 and with (6.24) yieldsX(J M 1 m1 2 m2 ) (J 0 M 0 1 m1 2 m2 ) δJJ 0 δM M 0.(6.28)m1 ,m22This property follows from the fact that the basis elements are eigenstates of hermitian operators with differenteigenvalues, and that the states can be normalized.

6.2: Construction of Clebsch-Gordan Coefficients147The second summation condition starts from the fact that the basis sets B( 1 , 2 ) and B0 ( 1 , 2 )span the same function space. Hence, it is possible to expand Y 1 m1 (Ω1 )Y 2 m2 (Ω2 ) in terms ofYJM ( 1 , 2 Ω1 , Ω2 ), i.e., X1 2Y 1 m1 (Ω1 )Y 2 m2 (Ω2 ) JXcJ 0 M 0 YJ 0 M 0 ( 1 , 2 Ω1 , Ω2 ) ,(6.29)J 0 1 2 M 0 Jwhere the expansion coefficients are given by the respective scalar products in function spaceZZcJ 0 M 0 dΩ1 dΩ2 Y J 0 M 0 ( 1 , 2 Ω1 , Ω2 )Y 1 m1 (Ω1 )Y 2 m2 (Ω2 ) .(6.30)The latter property follows from multiplying (6.18) by Y J 0 M 0 ( 1 , 2 Ω1 , Ω2 ) and integrating. Theorthogonality property (6.27) yieldsX(J M 1 m1 2 m2 ) cJ 0 M 0 .(6.31)δJJ 0 δM M 0 m1 ,m2Comparision with (6.28) allows one to conclude that the coefficients cJ 0 M 0 are identical to (J 0 M 0 1 m1 2 m2 ) ,i.e.,Y 1 m1 (Ω1 )Y 2 m2 (Ω2 ) X1 2JX(J 0 M 0 1 m1 2 m2 ) YJ 0 M 0 ( 1 , 2 Ω1 , Ω2 ) ,(6.32)J 0 1 2 M 0 Jwhich complements (6.18). One can show readily using the same reasoning as applied in thederivation of (6.28) from (6.18) that the Clebsch-Gordan coefficients obey the second summationconditionX(6.33)(J M 1 m1 2 m2 ) (J M 1 m01 2 m02 ) δm1 m01 δm2 m02 .JMThe latter summation has not been restricted explicitly to allowed J–values, rather the convention(J M 1 m1 2 m2 ) 0if J 1 2 , or J 1 2(6.34)has been assumed.6.2Construction of Clebsch-Gordan CoefficientsWe will now construct the Clebsch-Gordan coefficients. The result of this construction will includeall the properties previewed above. At this point we like to stress that the construction will be basedon the theorems (5.71–5.81) stated above, i.e., will be based solely on the commutation properties ofthe operators J and J (k) . We can, therefore, also apply the results, and actually also the propertiesof Clebsch-Gordan coefficients stated above, to composite systems involving spin- 21 states. A similarconstruction will also be applied to composite systems governed by other symmetry groups, e.g.,the group SU(3) in case of meson multiplets involving two quarks, or baryons multiplets involvingthree quarks.

148Addition of Angular Momenta and SpinFor the construction of YJM we will need the operatorsJ J1 iJ2 .(6.35)The construction assumes a particular choice of J { 1 2 , 1 2 1, . . . 1 2 } and forsuch J–value seeks an expansion (6.18) which satisfiesJ YJJ ( 1 , 2 Ω1 , Ω2 ) 0(6.36)J3 YJJ ( 1 , 2 Ω1 , Ω2 ) J YJJ ( 1 , 2 Ω1 , Ω2 ) .(6.37)The solution needs to be normalized. Having determined such YJJ we then construct the wholefamily of functions XJ {YJM ( 1 , 2 Ω1 , Ω2 ), M J, J 1, . . . J} by applying repeatedlypJ YJM 1 ( 1 , 2 Ω1 , Ω2 ) (J M 1)(J M )YJM ( 1 , 2 Ω1 , Ω2 ) .(6.38)for M J 1, J 2, . . . , J.We embark on the suggested construction for the choice J 1 2 . We first seek an unnormalizedsolution GJJ and later normalize. To find GJJ we start from the observation that GJJ represents thestate with the largest possible quantum number J 1 2 with the largest possible componentM 1 2 along the z–axis. The corresponding classical total angular momentum vector Jclass(1)(2)would be obtained by aligning both J class and J class also along the z–axis and adding these twovectors. Quantum mechanically this corresponds to a stateG 1 2 , 1 2 ( 1 , 2 Ω1 , Ω2 ) Y 1 1 (Ω1 ) Y 2 2 (Ω2 )(6.39)which we will try for a solution of (6.37). For this purpose we insert (6.39) into (6.37) and replace(1)(2)according to (6.11) J by J J . We obtain using (5.66,5.68) (1)(2)J J Y 1 1 (Ω1 ) Y 2 2 (Ω2 )(6.40) (1)(2) J Y 1 1 (Ω1 ) Y 2 2 (Ω2 ) Y 1 1 (Ω1 ) J Y 2 2 (Ω2 ) 0 .Similarly, we can demonstrate condition (6.25) using (6.11) and (5.64) (1)(2)J3 J3Y 1 1 (Ω1 ) Y 2 2 (Ω2 ) (1)(2) J3 Y 1 1 (Ω1 ) Y 2 2 (Ω2 ) Y 1 1 (Ω1 ) J3 Y 2 2 (Ω2 ) ( 1 2 ) Y 1 1 (Ω1 ) Y 2 2 (Ω2 ) .In fact, we can also demonstrate using (?) that G 1 2 , 1 2 ( 1 , 2 Ω1 , Ω2 ) is normalizedZZdΩ1dΩ2 G 1 2 , 1 2 ( 1 , 2 Ω1 , Ω2 ) Z Z dΩ1 Y 1 1 (Ω1 )dΩ2 Y 2 2 (Ω2 ) 1 .(6.41)(6.42)We, therefore, have shownY 1 2 , 1 2 ( 1 , 2 Ω1 , Ω2 ) Y 1 1 (Ω1 ) Y 2 2 (Ω2 ) .(6.43)

6.2: Construction of Clebsch-Gordan Coefficients149We now employ property (6.38) to construct the family of functions B 1 2 {Y 1 2 M ( 1 , 2 1 , ( 1 2 ), . . . , ( 1 2 )}. We demonstrate the procedure explicitly only for M 1 2 1. (1)(1)The r.h.s. of (6.38) yields with J J J the expression 2 1 Y 1 1 1 (Ω1 )Y 2 2 (Ω2 ) p 2 2 Y 1 1 1 (Ω1 )Y 2 2 1 (Ω2 ). The l.h.s. of (6.38) yields 2( 1 2 )Y 1 2 1 2 1 ( 1 , 2 Ω1 , Ω2 ).One obtains thenq 1 1 2Y 1 2 1 2 1 ( 1 , 2 Ω1 , Ω2 ) q2Y 1 1 1 (Ω1 )Y 2 2 (Ω2 ) 1 Y 1 1 (Ω1 )Y 2 2 1 (Ω2 ) .22 ),M (6.44)This construction can be continued to obta

the socalled total angular momentum, classically speaking the sum of all angular momenta and spins of the composite system, is the quantity of interest, since related operators, sums of orbital angular momentum and of spin operators of the particles, commute with the Hamiltonian of the composite system and, hence, give rise to good quantum numbers.

Related Documents:

it could be considered a new framework. Angular versions 2 and up are backward compatible till the Angular 2, but not with Angular 1. To avoid confusion, Angular 1 is now named Angu-lar JS and Angular versions 2 and higher are named Angular. Angular JS is based on JavaScript while Angular is based on JavaScript superset called TypeScript.

Angular Kinetics similar comparison between linear and angular kinematics Mass Moment of inertia Force Torque Momentum Angular momentum Newton’s Laws Newton’s Laws (angular analogs) Linear Angular resistance to angular motion (like linear motion) dependent on mass however, the more closely mass is distributed to the

Both Angular 2 and 4 are open-source, TypeScript-based front-end web application platforms. is the latest version of Angular. Although Angular 2 was a complete rewrite of AngularJS, there are no major differences between Angular 2 and Angular 4. Angular 4 is only an improvement and is backward compatible with Angular 2.

According to the quantum model, an electron can be given a name with the use of quantum numbers. Four types of quantum numbers are used in this; Principle quantum number, n Angular momentum quantum number, I Magnetic quantum number, m l Spin quantum number, m s The principle quantum

Angular Momentum 1 Angular momentum in Quantum Mechanics As is the case with most operators in quantum mechanics, we start from the clas-sical definition and make the transition to quantum mechanical operators via the standard substitution x x and p i . Be aware that I will not distinguish

0. Since the angular measure q is the angular analog of the linear distance measure, it is natural to define the average angular velocity Xw\ as Xw\ q f-q 0 Dt where q 0 is the initial angular position of the object when Dt 0 and q f is the final angular position of the object after a time Dt of angular motion.

1. Quantum bits In quantum computing, a qubit or quantum bit is the basic unit of quantum information—the quantum version of the classical binary bit physically realized with a two-state device. A qubit is a two-state (or two-level) quantum-mechanical system, one of the simplest quantum systems displaying the peculiarity of quantum mechanics.

dimensions of an angular momentum, and, moreover, the Bohr quantisation hypothesis specified the unit of (orbital) angular momentum to be h/2 π. Angular momentum and quantum physics are thus clearly linked.“ 3 In this passage angular momentum is presented as a physical entity with a classical and a quantum incarnation.