PERTURBATION THEORY 'n the development of quantum mechanics, only very simple physical situations have been treateg with the help of Schroedinger equation, which cin be solved exactly. However, in practice, exactly, solvable problems are rare, i.e, for the majority of systems of physical interest, the exact solution of the Schroedinger equation presents mathematical difficulties. Ih order t0 discuss these systems, ¥arious SPProwimate methods af solving the wave equation have been devised, leading to More oF less accurate csaluation of energy values and wave functions. Out of these methods, a simple and beautiful one ig “ave mechanical perturbation theory, develoepd by Schroedinger in 1926 ‘The basic idea ts taken from the perturbation theory in classical mechanics. The motion of a Planet in the solar system is essentially determined by the gravitational force of the Sun Of course, the SresMational attractions of the other planets, though small, yet influence the motion and as a result, the orbit instead of being a closed ellipse 1s a slowly precessing ellipse, The perturbation theory enables us 0 calculate this change. Similar situations arise in quantum mechanics which can be treated with Perturbation method. Perturbation theories are of two kinds (1) ume-independent perturbation. (1) time-dependent perturbation. Each of these is useful for specific problems. The stationary state perturbation theory concerns with finding the changes in energy levels and eigenfunctions of a system when a small disturbance is applied. In such cases, the Hamiltonian may be regarded as broken up into two paris, one of which is large and characterises a system for which wave “guation can be solved exactly while the other partis small and can be treated as perturbation, Consider 3 physical system which is subjected to 2 perturbation. The effect of this perturbation is to shift the energy levels: of course, the arrangement of energy levels remains the same, Thus the Perturbation introduces some additional terms in the Hamiltonian of unperturbed system. Now the perturbed Hamiltonian can be written as H=HOs HO), -U) where 7°) is the unperturbed Hamiltonian and AU corresponds to perturbation effect. if yal and E,{° be the orthonormal eigenfunctions and eigen values of unperturbed jamiltonian HO), then HOG) = Ey 0) 2) In the perturbed state, the Schroedinger equation can be written as A Ey. (3)7.1. TIME INDEPENDENT PERTURBATION THEORY FOR NON-DEGENERATE LEVELS ; If corresponding to cach energy level or cigeny: known as non-degenerate. jue there is only one wavefunction, the system is Let us write the true wave equation in perturbed state as Hy=Ey, ay inwhich H represents the operator of tel Lt ls H=(|-2 viv > 1 Im Vth @ We assume that it is possible to expand H in terms of some parameter A, yielding the expression H=H) +2HO + 22H) 4 @) where H is large compared with HO), i.c., the energy associated with HW) is large compared with that associated with ACY, Here H©) is unperturbed Hamiltonian and other terms are perturbation terms. Let the nvalues and eigenfunctions of the unperturbed problem be Ey) EOE, oo Egy on 20d Yr}, YO, Ys oon Yl, on Fespectively. Thus HOG = EO yO, 1” (4) This equation is said to be the wave equation for the unperturbed system. E,(°) denotes the energy of the nth level of the system The eigenfunctions y, for the perturbed system satisfy the equation AYn = En Yn (5) where E,, are the energy eigenvalues of the modified Hamiltonian. It ty assumed that it is possible t0 expand the eigenfunctions and eigenvalues of the total Hamiltonian H of equation (5) in a power series in 2: Eq = Eg) +A ED 4 IPE) + (6) Vin = nl + Ay az yA) + ea) where E,), E,@, and vq), yn, .-.. are quantities to be determined. Substituting equations (3), (6) and (7) in equation (5), we have (HO 4 AHO) (yO +A yD +2 yO #0) = (El) A Bgl) #2 Eg 2) +.) Pal) +A pA +22 ys #0) rc) oF HO yp.) + AHO yg) + HO yl) + 22H yy) + HO yD 2) Ha. = E,O yO + AE gO yg) + Eg) yg) + I2EO YA + £2 yO) + EO yD) $e af) In the above substitution we have limited H to nwo terms namely H° the unperturbed and major part and H©) the perturbed and small part. In order that cquation (9) may be satisfied for all values of A, the coefficients of the same powers Of on both sides of equation (9) must equal each other, Thus we have HO YO) = EO yO (10)AD 222 ——=_>_SS se QUANTUM MECHaN @ +E Oy, (Ny HO y 2 +E, (Pg gt + En ya (12) shed, first order perturbation and second orde, (12) represent unpertu equations respectively Evaluation of first order energy E,'!!. The first order pertubation equation 1s FE 2) ET yO = EO yD + EY For the solution of this equation, we ase the expansion theorem. We consider that the unknown can be expanded in terms of the known functions @® O_»¥., Lahn (13) We now solve equation (11) to determine the perturbation of the first order. Substituting equation (13) in equation (11). we have ECHO yO + HD y OM HFC FOy EO yO. (14) a Hy, aE Oy | Hence EE im EC En ig) + HO yy) = EO yO, or ECE — EA] yal) + HO yO) = ED yO). (15) Multiplying equation (15) by y7,(°) * and integrating over the space variables, Se MCE — EO yO dr + f yO" HO py) dr = SEM yg sy, de, (16) yf) der = 0, if tej j AT) | (0) We know that Sv |. if i= Applying equation (17) to equation (16), we have 0+ fy" HY yO dr = £0) | or EO = S96 HO yO) dr, (18) This equation gives the first order perturbation of the energy eigenvalue. The equation (18) can also be written as ED = (19) In indicates that first order energy correction for a non-degenerate system is just the expectation value of first order Hamiltonians 1") over the unperturbed state, here nth state of the system. Evaluation of first order wavefunction y,!") In order to calculate the wave functions, we proceed as follows :PERTURBATION THEORY Multiplying equation (15) by y,,(* and integrating over all space we have SE Cale = En Vn Mare f Applying condition (17), we have CLE ~ Ee} = Sv OHV yd or Cy = — nee On E26! Hence we can denote the first order energy and wave functions by the expressions E, = E043 (9) (22) and - (23) TE = » or Yee va aa SLALOM Ine yo) 4) Eni) -E,) where E' indicates the summation over the values of m from 0 to », excluding m =n. Evaluation of Second order energy E,'*). We consider that the unknown function y,.'*! can be expanded of terms in the known function y,.(°) as nO) = Ekim Yim (25) Substituting equation (25) and (13) in equation (12), we have Ee HO) vi) HE Cy HY Yl) = EG Eg Yr EC En) Ym + Eg? gl” va! (26) 4 £2) yO OTE Bg HO yg ) = E Bog En YY) = E Cog En Wg) — E Co HF m m in 7 OF Eg En Wp) — E Seg Eh Wg) = E Cy Eg?) yg — E pq HOD yg a ED yp (03 OF EgpalE mm — En en) = E Cal Eq? — HOY yg + £2 y, 1, ay Muluplying equation (27) by y.(°)" and integrating over the space variables, FE BelE ms — En] nl Yn de = LE VAMC aE) — HO] yl de + JEP yO (28) Applying condition (7). we have O= EC, fy OH yar + £2) m or Eg? = ECS yg OH yO de 29) Putting the value of C,,, we have gp LP OHA yn de Fl” HO yal a 2 EE)a6 QUANTUM MECHANI¢g Syn OHO y, Ode fy” HO vm) de Sv HO yp de J vg” HO von” ae o 2 - 1H is Hermitian operator, then hn HO si de SY HO yl de E,=5 Lm OHO yf) dt J vg O"HO nae 1) 7 om Eg — Eg ay (mH [nay : 2,0 6,0 ay Evaluation of second order wavefunction y/,) Multiplying equation (27) by y,,{°) *(m = n) and integrating, we have S28 SLE) — En) Yn” yg at = ECS im OE? — HOY YO de + f Eg Yn Vy ae (33) or SrlEm © — Eg} = EC ( En — J YH vl de) ao m 5 0) On BE Eo (Ea? SHO veld) oo =)’ _ ew ) Bm > E-50 (E; ) In this way we can calculate the second order perturbed wavefunctions by substituting the value of gy, from equation (35), in equation (25). 7.2. THE PERTURBED HARMONIC OSCILLATOR (a) Interaction energy is proportional tox Hamiltonian of one dimensional harmonic oscillator is then given by 2 iL =B_yt H=KE. +PE =7 7 +5kx tx (1) 3 ne 1 or H=-Z EV + ke tbe wa(2) --B) wd) The unperturbed Schroedinger eqn. is Hy V9) = En vai?PERTURBATION THEORY 37 we 1 ~ ea + beet = Ey i (6) ‘This gives the cnergy eigenvalue Ey= (143) no n=0,1,2, ) w = Angular frequency of oscillation = "YW k (8) ‘The first order energy correction is given by E,'= (9) =Syq (0? B yp) de (10) with Yai) = Ny Hye 22 al) (12) H,, = Hermite polynomial of degree n a 2 Nj, = Normalisation constant = | 2 wernt From eqn. (10) again Eq! = NpPSe7 @2V2 2b xen (EHV de 3) =N,2 fev? HH, bade (iy As this is an odd function of x, integrand vanishes. It implies that first order energy correction is zero. Second Order Energy Correction Second order energy correction is given by : [sn] Bl m> 2 BE," = “2 Eq) = Eg Summation over prime indicates that n = m is excluded. ‘The term (13) = J" on H gat = 57 vnt ima form=n+1 form 1 = Now form =n+1 for all other values =320 QUANTUM MECHANios b SSentt ae ent) Thus the energy eigenvalue in n" state is given by b E, 25, +5, = (n+ 3] Reo gis 3241) @ But + vie (c) Interaction energy is proportional tox Asan example we shall consider the approximate energy levels of the system whose wave equation is a) In equation (1) ifa and b are zero, this reduces to a wave equation for harmonic oscillator, In case @ and b are small, we can treat these terms as perturbation, HO = a+ ost BE [epee ist] pe . (2) The first order perturbed energy is given by EO =F yO Hyde = 72 yO ad yy Ode + SE yD ya En a FT yO? yg Ode +b S75 yO yO ax, 0) Since x? is an odd function and y,(” y,(°) is an even function, hence the value of the first integral is zero, ie., first order perturbation duc to ax? is zero. Thus Eg) = L7% yeh yd, @) To solve it, let = STE gO yO de (3) We know that in case of linear harmonic oscillator Walt) = Ny Hy () exp|~ 7/2); where & = ex, (6) Substituting equation (6) in equation (5), f= [ Ny? Hye) € *)PERTURBATION THEORY 7 No [t 2 | Hee eae. o Considering equation (15) of § 4 dsishave EHy() = $y 0 18) #0 Hy) -@) or PGE) = 38 Hy 4 18) #08 My 18) ©) From equation (8) substituting n = (n + 1) and n = (n — 1), we have EMy 618) = 4 Hy 428) + 0+ 1) Male) E Hn 16) =F Hn(6) + (2 - D Hy 28 Substituting these values in equation (9), we get P= Pear +e + | tn [3m +(n- Dm, 200] or EH y(8) = 4 Hy 4 28) + (« + 5) Hy(€) + n(n = 1) Hy 28). (10) Substituting equation (10) in equation (7), Ne (t 2 2s [Hest + (n+3) m+n 07%-0)] ei) ‘We know that te, [- ce FH) Hp(G) de = 0 if men fiz) =2ntnl? ifmean Using the results of equation (12), equation (11) becomes 2 visa +2224 (ned) Barenaoreta—a:] -| inl or isin + 2ionts (ne 5) tat ena P2—Fn—2)1| 2 =-Lenrs2 3 = TR en tint. (13) From equation (13) and (4), £, =6 3 2nt+2n 40), (14) 4a SO that the total energy to first order becomes . 1 3bw E, =F, +80 = ned) noy PE one ane (15) where oe VE) stat 7 “4ES = g QUANTUM MECHANicg 73. THE NORMAL HELIUM ATOM Now we shall consider the normal state of helium atom which contains 2 hucleus of charge Ze and two electrons each having a charge — ¢. The Potential energy of such a system 1s given by ve(_Z2_22,¢2)_1 (1) nm 2 Mz | 426 in which ry andr are the distances of electrons 1 and 2, respectively from nucleus, and r)2 is the separation of the two electrons as shown in fig. (7-1). Ze Nuclous Fig. 7.1. Helium atom. Considering the nucleus at rest, the wave equation for the two electrons becomes ey) am[p, 1 (Ze Ze _ 2 22 ara qaol nm |Yo" (2) The variables x).y;,z) are cartesian coordinates of one electron and x>,) 2 those of the other. If the term 7——— is omitted, the wave equation can be solved exactly; hence this term may be regarded Feor2 4s perturbation term ; 21 1) = & He aaxig 3) The unperturbed wave equation can be separated into two equations by substituting eG yi 21229222) = OCH 121) wo (2 9222)- (4) Considering the wavefunctions in polar coordinates 7), 6, $1 and, 8, ¢2 and in normal state, 0) WAM, 100 = #01: 81 61) Violas 82.2) = Per 81,44), t(D (r2. 82, $2) and corresponding renwal is EAN, 100 = £00) +E: =,- 227 Ey a) The first order perturbed energy EC) is the average value of the perturbation function HOD 2 over the unperturbed state of the system. Hence = | OH yOdr=—t iD, © _ 0) @awev |Z # in which p = 2Zr/ap and ag = =~ (42 £0). Tz eori2 E We know that ray a ee? 0) 4 $00, 100 = meee, P1/2) exp(— 7/2). (8)peRTURBATION THEORY t ume element is ‘The vo 41 = 17d sin 8, dO, dey 7:2 sin 0) dO, dey (9) fo that the integral for E) becomes s vraetoeak EE CfA axe F aso tole Jo laly Pe Pr dpsin 8, a0, de, p2* dp sin 0d0,4¢2 (10) inwhich p12 = 2Z72/do. Tne value of the integral can be calculated by considering the electrostatic interaction energy of two spherically symmetrical distributions of electricity, with density functions exp (—p;) and exp (~ #2), Fespectively. Solving the integral, we have EY = (5/4) ZEy (1) Thus, the total energy of the system is E=-(22?-5/4.2Z) Ey Thus, we can calculate the energy of He, Li*, Be+*. (12) etc. with Z = 2,3, 4 etc, respectively. 7.4. THE STARK EFFECT OF THE PLANE ROTATOR In case of the plane rotator under the influence of electric field E’, the wave ‘equation is @y 20 +E +KE' cos 0, ag? 2 Erne oy where /is the moment of inertia and, through its centre of mass. When E’ ~() the electric moment. It rotates ina = 0, the wave equation is py 20 de we Plane about an axis passing Ey Whose normalized solutions are Yon = (viz) em (3) md theenergyvalue (0) = MPH? oy 8x77 7 ‘| The term ~ x E’ cos 6 may be Tegarded as the perturbed term and hence the Perturbation energy Will be la En) = —pE' | Ym "COS $ Yq! dp = — Ae | dO — 9 cos 9 dg o , pa 1 (aa En) = BET em —m+2)9 ag HED im’ — m1) ) ‘m aS J, él dp a |, é dp on(S) i) form’ #m+1 (6) -HE'/2 form’ =m+1 Thus we see that the first order energy correction is ze10, i.e., E_() = 0,eS QUANTUM ME‘ aa 'CHANICS ‘The second order energy is given by 2 Ee [loa OS 9 Ym in| EY + Eg?) = — Eg En — Elm -1) \ vl" H 008 6 VN 4 1) db E, — EO) (in +1) 4x? IE? W(4m* = 1)" So that the total energy to the second order is = (8) = 64 E04 £2 Be Et Et Em = 52] eam? 1) ~@) 7.5. CLASSICAL HAMILTONIAN WHEN ELECTROMAGNETIC FIELD IS PRESENT ‘The force F acting on charge particle in motion with velocity v is given by FaqlE+ xB) xO The potential Vis given by 7 | Vaqe-qa.v 2) 1 | Aa Bxr ~B) | [As BecurlA=VxA | Now = Bx r=VxAxr=(V.r)A-(A.V)r=3A-A=2A] where and A arc scalar and vector potentials. Now Hamiitontan function is given by H=Epq-L Ad) where p, and g, 27 momentum and velocity coordinates respectively. L is Lagrangian and is given by L=T-V = 4mm? —y|p-A- | i) al in(ud + yj? + 02) —G |b — Age —Ay vy —A (6) qm + wy + uz) —4 ee — Ayvy — Az ue) { aL (PDX = By =M et GAy aL Now Oiy= Samy tga, w) aL P)e= Gy = Mert qa, | or p=mu+qga AB) Hence HH = (mv, — q Ay) vy + (mv — Ay) vy + (mv, ~ 4G A,) v, ~ ney 2apahe ZIM Fy) + UL) ~ gp — Ag ug — Ay ¥y — Ar) (9)pERTURBATION THEORY Substituting the value of ¥ from eqn. (8) x BSI) a0 = aq (09a? #49 (00) or n= p-gay+ v (11) V = potential energy of the system = qo 1 Ths zm ip +A —p.A-A.ph +a 2 2 ={2 - LL 42 H (B+) a(p-A+A.p) +3 A H=HO+H' +H" (12) Quantum Mechanical Hamiltonian in Electromagnetic Field Cc a p=-ihV=-it (+85 +ks) A=iA,+jAy+kA, Let us now evaluate (p.A +A. p) by operating it on a quantity y, (p-A+A.p)y=[-iNVA+AiRVy] =—ih[V.(Ay) tAV¥] But V.(Ay)=(V-A)p+A. Vp Hence (p.A+A.p)y=iN[(V.A)y+A.Vy +A. Vy] =—iR[(V.A)y+2A.Vy] Now V.A=v-L@xn) But V.(axb)=b.Vxa—a.Vxb But VxB=OandVxr=0 Hence V.A=0 Now AVysh@xnvy Hence (p.A+A ny=-m[ov2xta@xn,vy| -ih[B.(rx Dy Hence p.A+A.p=B.(rXp) (p.A+A.p)=(B.L) (13) Hence total Hamiltonian is given by -(-_%w @ (Bxe)? n[ wy +) seek x } -(-rwe -( mvtY 2 +B.L+ 2 BP sin?9 (14) 4m ALITER. The final form of total Hamiltonian may also be deduced as follows : ‘The eqn. (10) is =nL pay? H=ZR(P- gay eV—— a QUANTUM MECHA ate it over some operand yy, TO evaluate the first term, we write its x-component, and operate it Ps Y. Thus .) ( 2a % Bur AoA tartar @) From the relation spherical symmetric, then et Be 7-® 22 Bn 3 Now change in energy value of first state due to this perturbation 2 = | vic LEE aye 6) sere Vp r/09 2 Sg 1/072 Gp si = eo "207? dr sin 6 dO d¢ oloJo Vara? Vir a9 : (2a a [I e7 2170074 dr sin 8 dB dp oloto ax etrana 3 pags Q/age 3 Hage 32° ap : Ew “Tee (4) This gives the change in energy value of ground state when second order effect is considered, Note that in general (5)— a peRTURBATION THEORY 329 and second order energy change pe OB 2aent}| 3 K+ 1)- we Se | (6) ne 7.1. FIRST ORDER PERTURBATION THEORY FOR A DEGENERATE LEVEL ‘The first order non-degenerate perturbation method is not applicable when the energy level of unperturbed system is degenerate. This is duc to the fact that in the treatment of non-degencrate perturbation theory it was assumed that the perturbed wavefunction differs slightly from one function Yn) which is the solution of the unperturbed wave equation for a given value whereas in case of degenerate level there are such several functions which correspond to the same energy. An energy level js called a-fold degenerate when there exist @ linearly independent wavefunctions such as Vn ¥m2 ~~ ¥na Stislying the wave equation. Let, in the perturbed state, the Schroedinger wave equation be given by Hy=Ey (1) where H is the perturbed Hamiltonian, E, the perturbed energy and y, perturbed wavefunction. The perturbed Hamiltonian can be expressed in terms of unperturbed Hamiltonian H‘) as H=HO +H) + 2 HO). (2) ‘The wave equation for the unperturbed system is. HO yO = BO yO. =) The solutions of equation (3) are vos vor, vo”, vos - Vout corresponding to energy level Eq) V1 V12 v3 vrs Via corresponding to energy level Ey Vr Vins Vn s Vind) «os Wma”) corresponding to energy level E,,(°) In case of a degenerate system, the problem is to determine the set of unperturbed wavefunctions to which the perturbed functions reduce when the perturbation vanishes, ée., the evaluation of coefficients in the linear transformation converting the initially chosen wavefunction into the correct zeroth order wavefunction. These combinations are ami= ki Ym = kr Ym + kV 2 + hae Ym2 pp se (4) oo Fn © = Kr Yt) + ka Ym2 + + kta Vmal”) Am 2°) = Kay Vm) + 22 Yin 20 4 thea Vma” Provided the function yn, is given by amt = an FAV nO ER? Yi (5) and Eyg1 = El?) +A Ey) +22 Eg _6ee QUANTUM ny Sh Now the perturbed wave equation is given by H¥mt Emi ¥nt =O si - f Sebsututing the values of H, y,,; and £,, from equations (2). (5) and (6) in equation aren Oe ach zy 2 HO apy +) ni tA ved +B ye) +) ~ Em 2A EO BEL) Gal HARM ty, By : sy Solving it, we have (>, Ei) Yn 6) + AO vg) + HO gy) = En Ym OY = Eg Om O +. Q 5 ® a ‘The Bist order perturbation equation can be obtained by equating the coefficient ofa tiny ZEN, f.e., 0 HOY Oe HO, OE Oy OE, Ox, <0 bs 9) Now tet us expand y, ©) as " © =F anim ¥, Ym tO SE Grater Vint (10) mr Substituting equation (10) and (4) in equation (9), we get frm I rm’ HO y,, Oe = ki HO vy ” rat Gm trm t E me O Ying) Skyy HO gy pO —E Oop Em Vg 1 -2 Kir En Ym = 0 rea OPE atm Em) = Eg) v5 O = me r Muluplying equation (11) on both sides by Yj)" and integrating over configuration space we get Ea tre Em) ~ Em) f Yn Om Ode mt E bar Eni — HO) Yq (ll) =, 2 hr [En Fo Ym) dr — Fe! HO ns ar] (2) The left hand side of equation (12) is equal to zero because y, jO) and Ying p ms IF m =m’ and E,,.( — E,,(°) is zero if'm = m'. Now put ir = Sq O* HO y, ar (13) and Sir =S Vin Ym Ode “ Introducing these symbols in eq. (12), we have are orthogonal if E dy rlyy® tPERTURBATION THEORY 33 j=123,..6 (as) Eq, (15) represents a system of @ homogeneous linear simultaneous equations in the « unknown quantities. ky kias Kee ‘The equations are (Hy ~ 4) En) py + Hyg) = 49 Ege) yg tm Hr? = Bi Em Ka =P (Hay ~ 831 Ein) ky + Hag = gy Egg) B12 + one + Hel = B20 Em i) Rta = 0 ~(16) Ha = Mat Em) ry + Had”) = go Emm) Kya + oo + Hal — Sea Emi) bie = 0 The condition which must be satisfied if such a set of homogeneous linear equations is to have non-zero solutions is that the determinant of the coefficients of the unknown quantities vanishes; that is Hy = dy Em Hy ~ 42 Eg i Hy = Ste Emi™ Hay) ~ 8x Em (Hag ~ B29 Eg) «Haq — Ao¢ Emi? . =0 (17) Ha = Be1 Emit) Hea ~ Meo Emi) Hee” — Bee Emi Now applying the condition we get Hn — Eni) Hy) os AO Hy Ca) Eg). Hz (18) Hy He a Hae — Emi) ‘The equations (17) and (18) are often called secular equations. It is interesting to note that in case the secular equation has the form Cy — Eni) 0 os 0 0 (Hx = Emi)... 0 =0, 419) 0 0 aa) ~ Emi) then the initially assumed functions Yn 1"), Ym 2 --- Ym q) are the correct zero order functions for the perturbation H), The secular equation in which all the elements are zero except along the Principal diagonal is said to be in the diagonal form. The roots E,, )) are obtainable from an equation in this form, since the algebraic equation equivalent to it isSee eee os se QUANTUM MECHIe, <= (Ha al!) ~ Em) = 0. (20) i = 6,0 a with the roots E,,) = Hy Oy, Hy N 7.8. FIRST ORDER STARK EFFECT IN HYDRO hydrogen stom due (0 aN extETMAt eect, rk effect. Let us consider the first order change in energy level field of strength F directed along the z-axis. This is known ‘The unperturbed Hamiltonian for the hydrogen atom is given by 1 an” F Taeg ~(l) The perturbation HC), which is the extra energy of Rucleus where « is magnitude it the reduced mass. and electron due to external field F is HO) = ~eF2=-eFroos6, (2) where ¢ is the electronic charge and z = r cos 6 in polar coordinates. Here the potential energy and wavefunctions are spherically symmetric. We know that the waveluncuions, for any spherically symmetric potential energy when expressed in spherical harmonics, have even parity when azimuthal quantum number / is even and odd when / is odd.* Let us consider the ground state (n = 1,/ = 0,m = 0) of hydrogen atom, Since in hydrogen the wavefunction is spherically symmetric and has the same form for all orientations of the polar axis, therefore, there is no degeneracy. For such a non-degenerate state the wavefunction is ¥100 = Ryo(r) Yoo(8. ) --G) 1 = Ry — vay 7 ¥ntm = Rut Yim(O @) Vir, 0) = Nip Pi(CO8 8) Om lO) 1 ne f+y @-Im) - or? andy, =| Glen Coin pt Pn)” Tay a { Tm a [@t+nd-|m) 1 presiuine rine = | OE | gg rrore) ' we also know that om aime dit! p Pray = 0 1" 2 2 Pay °, 072 Fn os cBEURE 10x P.fcone) Po a= -2) wat Po) Polk) = 1 or Pq"(cos 6) = 1 P,%) = 0 eerie =P yx) = ror Py"\c08 6) = cose Pi! = a= 2 Eye = 0 — F200 Py (0080) = sin 21724 page(s) —2)2or pn r, Nay = 1 0 Pye = or Py, (C084) = sing (Continued)pERTURBATION THEORY 333 From equation (2), we see that the perturbation H) has odd parity, therefore, for the ground state of hydrogen Foo, 00 = S72 v 100 RY vron de = 0 or Hino, 100" = 0. (Because when y',, and y;, have the same parity, even or odd and Ht!) Sv in HO yn dt = 0), We can also understand it as the definite parity, then AH) =—cFreosé ie, A=eF,H = —reosd Finn! tm = {ff Wn tm(— F608 8) Yn" fm: Sin 8 dr dO dp The first order perturbation energy in ground state is AH, 100 = -eF {freee ‘Thas we see that for the ground state of hydrogen, there is no first order Stark effect. 7 4/4) 2 sin 8 dr dO dp = 0. (4) Hag Let us consider the excited state (n = 2) of hydrogen atom. Since n = ?, therefore, | = Oor! = 1 and m may have the following values 1,0, — 1. Thus the quantum numbers / and m have the following combinations : (0,0), (1,0), (1, ), 1, = 1) which shows that when n = 2, the state of hydrogen is four-fold degenerate The corresponding wavefunctions are 1 Wom = Rao Yoo = Fa Ravl). even parity ) 10 = Roy Yo=V (35} Ryx(r) cos 8, odd panty. (6) 3 \ - gar =RaYu -v(&) Rar) sind %, odd partiy (7) a Yh, 1 = v(a) Ry(r),sinde"!? odd parity (8) Since there is four-fold degeneracy, therefore, the secular determinant contains 4x 4= 16 elements, ¥299, ¥2115 ¥210: #2, 1, - 1 all have the same energy. Hence Y241-1 2 even purity v2 3 \v2 P, (cos) 6° = (=) cose, odd panty va v2 Pyleoos ye = { singe’? odd parity v2 Jet esse oxld pantySS: : OoW—_h—tXWwO_—_—rls —~ QUANTUM MECHANICS 334 FS, 200 ~ ED HS, on 2 wh, al #8 2m HD a — EHD v0 wt 2-1 ‘baw ay ha Ha AD 20 HDs on Daa HBR, Let us calculate elements of the secular determinant. HSS), 200 = J y 200 HO yop dt = 0 (9) (because ¥ “200 ¥'200 has the even parity), Similarly HSN}, 241. HS%, 210, SD 1 On the same ground HA), 219, Hy, 1, #58), HS. 24, 0 are zero, 1,21, 1 Ate zero. 1 HAY aan and HEP ay a Now consider the integral 18 =~ fff 1 Rasreosory/ (Z 5) nod? Ray x Parsinoaoag (10) 2a But | é% dp =O. 0 Hence 0, 21 Similarly 49), 249 HS), 24, —, and HAD 1, 299 are zero. Thus the remaining elements are 1 1 0,200 and HSL ay Now HG, 210 = Sv "200 HO yryo dr, (11) = Pty = (LY 28) ag (oe where 200 = Tara Raat”) = Viay | 2a ay} °P [~ 2a5 and yay = (4 % cos arn() 3 1)? + r -V (a) (25) aa (-25] where ag is the radius of the first Bohr’s orbit. Substituting these values in €q. (11), we have , ‘honest (a) [lf [a5] "e-8) Jo (a5 sta) [Ln] dt =Psin6.d0 drdp and HO) = — pcos g where_ peRTURBATION THEORY 335 i e020 sinéido ~ — 2 temdaj 2 Again [/ ston 940 = — 2 fos oh = 5 Let r/ag = 7, then 3 =() 1 - 1 a4 nt -(z5) vd oo ag! 9° a9 dn 2| wendy | We" dn ° ° =—“pre = ay5 PFO) - TO} ag V5 =O fag - 120) = - 2% x7 vs = yg iB OO = — Gg = - 3 VS ag Substituting these values in.eq. (12), we have ,, 1 3 2 FM}, 210 = — via) X (= 3 VBay) x 5 x 2 = 3 ay (13) Similarly FA, 100 = 3 a0. (14) | Substituting all these values in the secular determinant we have -EM 0 3ay 0 -EM: 0 0 0 0 0 0 0) Eo) 3a 0 EO) =0. (15) ‘The four roots of this equation are EQ, =0,0, +3 a9, — 3.49 (16) and the corresponding perturbed energy values are AE = 0,0,3ageF, - 3age F. ANT) ‘Thus, we sce that for n = 2 state of hydrogen the energy level is raised by 3 ag e F for one function and is lowered by the same amount for another function. For the remaining two wavefunctions the energy levels are the same.QUANTUM MECH, ‘The splitting of energy level is as shown in the fig. (7.3) =e" (1 _Unpsnutes aE, oe” = dear =” gl Ll peas Es Fig. 7.3. Splitting of an energy level in elect fel. Further, itis known that, when an electric dipole ts Placed in elecinie (esate change in a pte even by = p. Ete hen suggests that hydrogen som inn = 2state (eI BES excited he haves like an electric dipole of dipole moment 3 ¢ ag. The dipole in electric field gets in three different ways rey ate) self orientey (1) parallel to electric field (1) ant-paraliel to electne field. (in) two states perpendicular to the field. Eigenfunctions ‘To evaluate eigenfunctions we use following secular eqn. 0 0 3a, o| | 41 ¥200 9 0 9 0} fava (16) 3a, 0 0 0 439210 0 0 0 Of Jays, - 3.4943 ¥29 = 0 (17) and 3.49 4 Yq = 0 (18) These two give a) = 0,43 =0 As there is no restriction an a, and a4. The linear combination of two wave functions is = eq $4421, -1=0 (19) Using Sytydr=t we pet ay +ae=0 or Hence the wave functions corresponding to E) = Qare 1 5 a BRS vin = Fy lvan + ¥2, -1] 1 oa(20) Wp lan — val for EV) = — 3¢ F ap, we have =3a 0 3ay 0 4, Y20 —3ay 0 0 Ay) i 0 0 zi a 3 ay 0 Bay 0 as Vou 0 0 0 ~ Faq ag ¥24_ ———S peRTURBATION THEORY Ba Solving — 3.49.4) p299 + 3.4349 Y219 41 ¥200 ~ 43 ¥210 = 0 and — 34942 211 =0 a,=0 3.494 ¥209 — 3.49 43 ¥210 = 0 44 ¥200 ~ 43 ¥210 = 0 ~3dgaq yr, -1 =0 ag=0 Normalisation condition gives 2a? a? + a5 or a, =a3= 1 al Yzlv200 —¥20] corresponding 10 E) = 3e ay F. Similarly y = Fzlan0 + Wao] corresponding to EO) = —3eaF. Energy change proportional to F is possible only in the case of degenerate states but if the energy change is proportional to F*, every state will have non-vanishing matrix elements. In the first case the permanent dipole moments are affected but in the latter case dipole moments are induced in the atom in any state, Since the ground states of all atoms and nuclei are very likely to be non-degenerate, it is expected that an atom or a nucleus in its ground state will not possess a permanent electric dipole moment, and none has ever been found experimentally. 7.9. HYDROGEN MOLECULE [HEITLER-LONDON TREATMENT] The hydrogen molecule may be picturised as shown in the figure (7.4) py and py are the positions of two nuclei (protons) and e; and e3 are the positions of two electrons. The separations between them are as marked in the figure. The attractive force between two atoms of molecules is due to sharing of electrons between two atoms. These shared electrons spend more time in between the atoms than anywhere else, thereby producing an attractive force. The covalent bond formation is purely quantum mechanical process and can be explained using perturbation theory together with the Pauli’s exclusion principle. Fig. 7.4. Hydrogen molecule. The electrostatic energy is given by with“Ss a QUANTUM MECHANicg ¢ LL _ 14141) represents the perturbed terms, ofS Ane 12 2 Te 'p “ey Schrocdinger eqn. for hydrogen molecule is now given by 2m al Wet Vy tre -Myn0 ~(6) or Vept vey +R E- Vy- (+ Yay =0 (0) 2 2 gt with we SeS ary? ay? a vie eee ye 2 os ae (1-¥1»21) and (x2, y2, 22) are the coordinates of two electrons. In eqn. (6),£ is the total energy of the system. Leaving in the first instance, the perturbation term, the Schroedinger eqn. can be written as 2 Vibw + Py + SF IE + (V+ aj] =0 (8) E° = unperturbed energy of the system y® = unperturbed state wave function =v") v2 @ (9) Substituting this in eqn. (8) [rv +024 a {E-(V, + | v1 (1) y2 =0 (10) LOVE In? +1 VP vO + BE — (Hy + a1 VO yx =0 “(a ‘The Schroedinger eqn. for wo atoms independently can be written as V7 +E, -K) vi =0 (12) V2? v2) + (Ey — Vo) yo" = 0 (3) Multiplying eqn. (11) and (12) by p20 and y,, respectively and adding, we get V2 Vy? yy + yO Vo? yy + [(Ey + Ey) — Mth) y,Oy=0 — (14) Comparison of eqn. (10) and (13) gives E\ + E,= EO =2Ey (15) As the two electrons one indistinguishable the unperturbed wave function may also be given by v2 = wi Q) y2 (1) (16) ‘The linear combination of wavefunctions given in equal (9) and (16) are also solutions Of Schroedinger eqn. and they are ¥.0) = Nef 1) ¥2 @) + ¥1 @) ys cy au V4) = Nal (1 ¥2Q) — 1 @ yx (ry =)PERTURBATION THEORY a a ; with yan y, are symmetric and antisymmetric wavefunctions with N, and N4 representing corresponding normalisation constant, ‘The normalisation constants come out to be No=— (19) “Vite 1 (20) N,=— t 4° VI=TR with 4= Sy (I yL 2 y1 @) yA @) ary de If now contribution of perturbation is taken, then energy is given by E=EO+E ‘The energy E’ is given by ‘Bie 22) T¥savsa* ae with dr = dry dry Substituting the values of wavefunction, we get ge = LL OO) ¥2R) # vO vO] H wi yA) # v2 wM ae SF vr v2Q) = 2 vIMAI WMD) ¥MR) = v1) va] dt This gives gr =Se4 (23) C= JS vill) v2) H' y(t) v2@) ar A= JJ x2) v2) H yx) v2(I) dr = Exchange integral. With this energy in two cases is given by with C+A Ey=2E,+ Soe (24) a = C-A ?: and E,=2Ey +5 (25) ‘The Coulomb integral ‘C’ is e c= SFay@ [42-4 axe i Te M2 Pot 9) (2) dr (26) Physically, the terms in above integral Tepresent the average potential energy of two Protons, two clectrons, first proton and second electron and second proton and first electron, Tespectively. The exchange integral is A= SQ) yMQ) HMR yM ar w(27) This implies that electron (1) is Sometimes in state (1) and sometimes in state (2). This can be possible when electrons form a Cloud like Structure around the nucleus. For parallel spins energy in eV bhonoaoe Nuclear separation in nm For Ant-parallel spins Fig. 75. Potential energy vs muclear separation.QUANTUM Mi 40 ECHANICS Physical Interpretation The H, molecule contains two electrons instead of single electron of HZ. Both the electrons vay share the same orbital, ‘¢., described by the same wave function when they have antiparallel spins, With two electrons 10 contribute to the bond, Hz appears to be twice as stable as HZ, with a bong energy of 2 x 2.65 = 5-3 eV But due to electron-clectron repulsion, the bond energy comes out tg 4 5eV instead of 5 3.eV The bond length 1s 0-074 nm. Further, note that when total wavefunction 1s antisymmetric (= spatially symmetric and spin any symmetric) the wave function is more stable (Fig, (7.5)}, Physically it is because, when spins are paraltey Heads t more dominating repulsion in accordance with exclusion principle making the state Jey Sable. 7.10. TIME DEPENDENT PERTURBATION THEORY Upto this time we have been concerned with problems in which the Hamiltonian operator H js assumed to be independent of time and energy 1s conserved. In this article we shall consider a differem: type of perturbation theory which is applicd even when H depends explicitly on the time. It permits us to discuss such questions as the probability of transition of the system from one unperturbed stationary state 10 another as a result of the perturbation, In atomic physics, particularly in spectral analysis, the main concern is the quantization of energy, and the time dependent perturbation finds wide application. In nuclear physics, particularly in nuclear reactions and disintegrations, the time dependent method is frequently used. The theory of time dependent perturhation was developed by Dirac and is often called as the theory of the variation of constants. Let us consider an unperturbed system with wave equation including the time yo 0) yy) = — ek a“ nt a a The general solution of eq, (1) is given by $ = Ari EOe, (0 WO) eS gO) + 7 AMEN yO, @ nwo where aq(0) are constants and the functions y,(°) are eigenfunctions of the following time independent Sehroedinger equation HOY) we Ey, ) Here yi.) and £,,© are assumed to be known, Let us consider the Hamiltonian He HO 4.40), (4) where 1A") js the unperturbed Hamiltonian and H js the perturbation depending upon time and is very small, 4e., 1) isa slowly varying function of time. The perturbed equation is 40H) we to Cee xT oi 0) |A general solution of the wave eq. (5) Gin be written as Wer, AD ™ Lag YM), a the quantities a,(/) being functions of Falone.neo ED eee OY Oo “UTTO = M2P (@yih CDH * @y th (yo Fy raz = Ov 10 "PH CoH () wa | OS uz Gn . TUT u 2D (g)" hao) fh On \ FP 7 =P @)"th (DH @) “ih ("0 ty ALY OM ‘soeds UONLINGIUOS sy J9AO Sunesoqur pur ,, 4 hy Sq (g) ‘bo SurktdninW 1a = u are © sh = ("th (pH @)"0 z ¥ “poy[sours ore S10108J Ise] pue is1y oy) BDUDH . we 127 og (ome 4 = (0) # (Q)H ing - uprZ zz _ _4e rae re _ ; (o) fhe rn SP Heme OH LE — = th DH OZ + (0) th (oy "PS “(g) ‘bo out (9) SuNMinsans TT TTT ggg \Wwe XWOSHL NOIWaENLH34 add oe bial “OHM Nea (MH Co + "ek + weet OY SU 7ag ~~ 10 C+ Qrdtt arto > ya om (OL) wba ul (TT) “bo Sunminsgng ave + Deg" gh + Day? V+ Oey"? = O'? oud} ‘sUOHeINI[ed LapIO TUIIAJJIP 105 Y JO sromod snoie, JO swuojouyaco Sunenbs pur y ut soizas somod se “y Surpuedxa ut sisisuoo uoneuixordde uoneganiiad SUL 0 = 718 ones soy) Woy worAIJIpP Yon A19A JOU are (s)“v sapnrt{dure oy} VEY? OS “Tews Anustoyyns SI Uoneqimzed oy) yey paumsse sf 11 axaHy ‘sanbruysar uoneqinzad ay) asn am *(O1) ‘bo aatos gy _ EL umg pue “gq — “I IP (gyth “A | = ()"“H ary ©) * (MH +) q@) u Co)" «(a4 Mey 1) dxo 4444 (i)"” Z Us -=()" 10 u ie - Ma) | oh 0 Bre -=()" 1. , 2p (wa 7 ) asa X @"h (gH (77) ae «on™ | We Zr - = xz= 7 QUANTUM MECHANICg Zero Order Calculation Equaung the coefficient of 2° in eq. (12), we get 2a) or ‘This constant gives the initial state of the system and can be chosen properly if the system is completely Known before the application of perturbation. es dpg(2) = constant, (13) We assume that the system is in a particular state / at r = 0,1.e.,4j(0) = 1 and all other 4,,(0) = 0 or 1 =r (14) First order Calculation : Equating the coefficient of A on both sides of eqn. (12), we get a 2 MO = — FS OH? C9 inn or an (0) =~ ae [ D440 Hina? expe ann | ae BEEP [BbarMin ea (ms or Am) = mf Hn 1) €xpli wp jt) dt (15) Here constant of integration is taken to be zero in order that m0(0) be zero at 1 = — & (before the Perturbation is applied). The eq. (15) can be simp! turned on at one time and off ata respectively. Thus fied if H‘ is independent of time except for being later time; these two instants of time may be taken as 0 and 1 an (9 = = 22ty,, of expli dy t) de o | (16) ‘The probability of finding the particle in mth state can be calculated as follows Probability This is known as first order perturbation theory. Gan 1) Og OO) = | yg (0) |? 7) Le iw, 40 -ia,,)t 1 =e" - Ne -)XS1H, M2 a" Ya Ani |pERTURBATION THEORY ass 1 1 ~ ae a? on? vag = LH)? = gt ~ 205g) =A sin Lag? SEs emit’) 5 41 HE) sin? 1g fy [2 = LEED sit? (1072) Ab Wom? ‘The factor sin (w,» ;/2)/@pm 71S plotted as a function of w,; as shown in figure (7.6). sin? nu 7 ee len an 2g 0 ean Ge BE Oy tt ToT tt Fig. 76, Showing a graph between sn? [nq 1/2)/0% and ty Physical interpretation of the curve. The maximum value of sin? (Wm )!/2)/a_ ;* occurs when Wp, 18 270, i, KING py =X. sin? 1/2) _ 1 (#) ae > oe 2 2) 3287 yt [¥] ; neglecting higher powers x ; (19) | So the highest peak value is 2/4, which can be seen in the graph. | in? ‘The value or will be zero when | sin? (1/2) _ 9 52 or Beans or | Thus, a ban (20)QUANTUM MECHANIog For n=l, It ts quite clear from the figure that the height of the main peak increases in proportion to (equation 19) while its breadth decreases inversely as ¢ (equation 20). Therefore the area under the Curve 1s proportional tor. Thus if there is a group of states that have energies nearly cqual to the initial State, the probability of finding the system in one or another of these states is proportional to 1. This Statement implies that the probability per unit time, that a transition has taken place when the Perturbation has been on for a time r, proportional tor. Second Order Perturbation Equating the coefficients of A? on both sides cf eq. (12), we get nO) Hann eXPCi pn nD) a 5 em) = font 2ni 1, ett 1) gl Omnt “V2 [-s| Gar |] Aaa e Ha Haar 04s ama) _ (nn nr tp Ani Han? 0, Fon, =pl a, lee Hal) Hn [6 io, i, a =5 ae | foto ema ay we ik Integrating above equation with the requirement that ay,2)(f) = Oat Hy? Hn [ demi 7 Fm) Lg Hat Hann) | el @mit — My =b ye Amn ad am 8 2 nt mt This expresion gives the constants in second order perturbation. 7.11. TRANSITION TO THE CONTINUUM (Fermi Golden rule) Now we shall calculate an explicit expression for transition probability per unit time. Let ¥S assume that the system is enclosed in a cubical box of side L that has boundary conditions at its wall ‘The stationary states of the system will be discrete, but separated in energy by an interval which is inversely proportional (0 the volume of the box. If we pass to the limit of a box of inifinite size, tMe levels within a given energy interval increase in number and merge into a continuum, Of course, it being only a formal limiting process, however, permits direct application of the result in calculating *Re probability for transitions that involve states in the continuum.PERTURBATION THEORY 345 ‘The probability of transition of mth state is 4 sin? (aq 1/2) 1 We | a(t) [= 25 | Ami? 1? a) Om Because of the factor 1/w,, ;7, the probability is largest for those states in which the unperturbed energies E,°) are closest to E}). If E,,(°) is far away from E,(°), the probability is small. Since the levels = E}). Since all levels in the cluster are very close to one another, they form a cluster around E,,(” represent nearly the same physical properties, we may add up the probabilities of all levels in the cluster and consider them as a whole, ‘Total probability of transition = Z | a,q(#) |? ‘We define the density of final states p(m) such that p(m) dE is the number of such states in the energy range Ey, and Em + dE. The total probability for transition into these states is obtained multiplying the equation (1) by p(m) aE and integrating with respect to E. Thus we can write E | am(t) [? = S| am( Palen) dE. 2) Rather than considering transition to a particular state, we may consider transition to group of states of neighbouring energies. This is what is always done for transitions of states of continuous spectrum, One may then, under certain limiting conditions, define a probability of transtition per unit time (10 any of groups of states). Asin? (Wp, 11/2) Om, ie transitions) the main contribution comes from this peak and only a small error is involved in extending the limit of this integral to — @ to + =. If the central peak of is the domain of integration (ie., energy conserving +0 4 sin? 1 2 4 sin? (pq 11/2) Ta az! AmE pen) eet But dE =Tidw as Ey, — E) = Noy + 4 sin? Vv? 1 Hn? Poemn | Sr ntl) aw <= nf we Fg nig? 1 IN (Wy 1/2) ” T Fans Poo | ——) dw. @) -o Ont Consider the integral + sin? sin? (Om )t/2} mit Pe £ Jao: pan | 2B) oe mt 2 tee esha ames Bacar St de = 200, wee | ' a? Therefore from (3), we have THE | Hy Poem 2xt in— ps QUANTUM MECHANICg So the transition probability per unit time is 7 = 32 | Hy? Poem) -(3) This formula has wide applications in quantum physics. The theory of scattering can be explaingy with the help of this formula. Equation (5) is the Fermi’s Golden rule and states that the transition probability per unit time (3) ts non-zero only between continuum states of the same energy, (ii) “is proportional to the square of | Hy, (| of the perturbation connecting the states, ang (iit) is proportional to the density of final states 7.12, HARMONIC PERTURBATION ‘The most common way of inducing transitions between stationary states of quantum systems is by applying a harmonic perturbation. For example, the oscillating electric and magnetic fields associated with light can induce transition between stationary states in atoms. Let us consider the situation in which perturbation is an explicit function of time (simple harmonic function) except for being turned on at time = 0 and turned off at a later time f = 1g, we write HY = 2AM sina t* = | YnkO"2AM sin w ty Ode Now An" =2HA) sinwt (1) where — Hy, 9) | Ymi®) +A yi de and is independent of time, 7 = 2 sig Pexpicin rt Q) ls We know that In case of harmonic perturbation, fo Gn CC & tp) = ~ 2a | sinw fexp(i ay, t) dt 0 ; ny det oe = Faun fe) ata o 2x 5H aia a? | 0 [ Ont $2)! Hem - 0) fem tone_ emi 4 ay mi * 0) ~ mr ~ a) |, A perturbation proportional t0 sin a oF cos wf contains the terms e and e~ §@! with equai amplitudes. Only fone term leads 10 absorption or emission of a quantum of energy. Therefore a factor 2 is introduced,- pERTURBATION THEORY wat 10) [ lems ~ 0% Moms to 7 a Mon | ir ia ein ah ) mn (mi - ) Omi +o ‘This shows that the amplitude depends upon perturbation duration f» and does not depend upon time 1. The probability of transition vc,, the probability of finding the system in mth state 1s appreciable only when the denominator of first or the second term in eg. (3) is almost equal to zero. Ifwe consider the mth level above Ith level, then Om 1 = (Em — E{)/0 = positive quantity In this case the second term of eq. (3) will be zero due to larger denominator and the first term is important. Ifwe consider the /th level above mth level, then Om 1 = (Em) — Ej) = Negative quantity In this case the first term of eq. (3) will be zero duc to large denominator and the second term is important. ‘Thus the second term of eq. (3) is important when the effect of the perturbation is to transfer an energy iw to the system on which it acts. The process is known as absorption i., energy is transferred from the perturbing source to the system. The first term will be important if the effect of perturbation is to receive energy from the system. This is known as emission. When E,, > Ey, the first term needs to be considered and the probability is given by | @pq(t = to) r 7 Hn? P a (mie) 07? (4) wv (@m1- ey » When E; > E,, the second term needs to be considered and the probability is given by 4 | Hy PY |? | sin? @n1+ ) to/2 7 | aq) [2 = ALT | St mi) /2 6) 7 mi + ey 7.13. SCATTERING OF FREE PARTICLE BY POTENTIAL ENERGY V(r) Example. The perturbation theory can be applied in a simple way to the theory of scattering if the scattering potential V(r) is regarded as a perturbation of the states of the free particle. We shall | calculate the transition probability when the initial and final states are free particle momentum eigenfunctions (ce, plane waves). We take for initial and final states, the free particle eigenfunctions as -) and where kg and k’ are the initial and final propagation vectors respectively. The perturbation matrix element will be H, | p00 * HO yO de = |——— ——— _— sie QUANTUM MECHANIos 2 P= Ae] feerrenar ~6) where k = (kg — k’), ‘The value of p(E) can be calculated as follows : In case of cubical box of length L, the normalized wave functions and eigenvalues are given by “4 and (5) If we draw a sphere of radius n, all the energy states will be in the octant of this sphere. The number of energy states would be given by volume of the octant divided by the volume of one energy state (unit cell). Thus the number of energy states with energy less than E is given by 37 4x (2mEL? BS | xn The level density is given by 4m age |) 2 Qm)PL d ipsr 3 | aan an dE 372 = vem pa (where V = L3) (6) Similarly in terms of momentum p and velocity v, we have aH |2(ee2)? a2me tm PE) = ae ae as2mE = 2m- mv eet do “Sow ae” aL} , ydp "Eee aE __Bp ( i 5. 2 273 nv The level density corresponding to scattering direction (6, #) per unit solid angle is given by 32 a (E)dw 1 L MEO) Oe ae Tae Pe L3p? 827 Bu Now 7,,.., da is the probability per unit time that a system makes a transition to a final state in (7) de (8) solid angle dw> peATURBATION THEORY 349 2 2 Tr pd = z [erek-n Virydr | p(E.8, 9) dw 32 f esr. Von ae aa 2 2 “aia [enen.nvenar dw, (10) ; v ‘The value obtained in this way is equal to the number of particles scattered into the solid angle dw per unit time when there is one incident particle in the volume L3, If v be the velocity of incident particle then Incident flux = number of particles x velocity per unit volume =to-2, (11) L mL ‘The differential cross-section is defined as Scattered flux _ Tr+14@ 2(8, #) do = cident flux ~ p/m L> 3 pe 00.9) = Say 2 Jener-nra| __p-m any 2 Janer-onne| am} wnt 2 [erex.n roe| 2 = is exp k. 1) Ye) ar |? (12) ‘This is the famous Born approximation formula. 7.14. APPLICATION OF TIME DEPENDENT THEORY TO ALPHA-SCATTERING When alpha particles with charge Z, e and with moderate energy bombard a nucleus with charge Zze, the particles are deflected by mutual Coulomb repulsion, the Rutherford scattering takes place. In this process the nucleus sees a rapidly varying electric field which can cause transition from the ground State to some excited nuclear state. The potential energy in case of alpha scattering is given by 22,2 v= = (CGS. units) ol) ‘The Schroedinger time dependent equation is given by 2 ZZ, -F yy, S28 yi te (2) 2m r 7 ar350 QUANTUM MECHANcs Thus equation can he solved by considering the potential as a perturbation 2 22¢ he. HO = 3) Let us assume that the scattering foil is inserted in position at = 0. Before f = 0, the @-particle jy described by a plane wave along the x-axis, m2 y= Les Mle™) pot] = Ts where p is the magnitude of momentum vector pg of a-particle. AU Expl) pot). (a) 0, the perturbation is switched on. Now the transitions may occur to clusters of the levels-the unperturbed energy level E,,(”) of which are close to £/\"), Let us examine these levels. They are all planc waves (unperturbed eigenfunctions), the momentum vectors of which have nearly the same magnitude but different directions. 1 fen Vn = explo) pr, Lp ep. (8) ‘These levels may be grouped into subclusters and take a typical level yy" for which | p’ | =p For thts level the momentum vector may be denoted by p. Thus wala | expli™) pr} (6) ‘The matrix element of transition is given by Zz. = |{[ eee p)-r] H/(O= dr. AT) ‘The matrix element can be determined graphically. The direction of the vector (py — p) is shown in figure (7.7) and it is = 2p sin @/2, where 0 is the scattering angle. The radius vector r is expressed as 7 c0s 6", so that [Po — P)I = (Po)* + (P)* — 2 po pcos 6 = 2p? - 2p? cose = 2p? [1 - cos 4] = 4p*sin? 2 (Po ~ p) = 2p sin6/2 (Po — p) f= 2prsin@/2 cose (8) Fig. 7.7. Dircetion of Vector 4’ in the angle between (pp — p) andr (Po - P) For finding the volume integral, spherical coordinates (r, 6’, ¢) are employed so that the polar axis is parallel to (po — p) ZyZre (* (* explliD) 2pr sin 8/2 os , Hy = 2 | 7) 2pr sin 6/2 005 8], . Aisin han" V dsl. 7 2xZy2.¢ (* f explw ny 2prsinv/2cosg'y]* ~* sey), =) Iprsino”l 0PERTURBATION THEORY 201 2aZ, Zen (* 6 = sin (22 sin? Vp sin (02) f. vn nia) % 22Z,2,2e0 [* = 1 sine de, r= 22 sin® (9) 2 Vp? sin? (872) J a4 Using the limiting value 1 for the integral Se sin x dr, we have aZ, 2 Vp? sin? ( a 2) ‘We now calculate the level density of the subcluster of level corresponding to scattering in the direction (6, 9). Itis given as Hyp Y= (1) voig E09) das = ED do @ Et ay Iis for an isotropic space and momentum vector pointing into the solid angle do ‘Transition probability per unit time is given by 2 tinh 00m Now 7; ..; is the probability per unit time that a system makes a transition to final state in solid angle de» sin‘ (6/2) sit The differential cross-section is defined as Scattered flux (0. #)de = “Tent Our” “prim L “ i Zezee 4 vy omi? 06.9) =. te 4m sin’ (8/2) P Z2zie or 0(0,9) 22 : (14) “ame sin® 672) This is famous Rutherford’s scattering formula. The expression is the same as obtained by purely classical considerations. This is an important Problem where quantum mechanics and classical mechanics give same results. This is becahse N does Not appear in the expression and when 0 - U, the classical picture corresponds to quantum picture See articte 7.13, fFi. | QUANTUM MECHaNicg 352 APPLICATION OF TIME DEPENDENT THEORY TO IONIZATION OF HYDROGEN ATOM. Here we shall calculate the probability of ionization of hydrogen atom, initially in ground stare when placed in a time varying electric field or by a radiation field. This situation is similar to the case when hydrogen atom is placed between the plates of a capacitor to which an alternating voltage is applied. The electric field is expressed as E(t) = 2 Eysinwr (1) The initial state is the y'499 state of hydrogen atom, expressed as 1 - r/o, ‘The final state is the free electron after ionization has occurred. Thus the final state can be 2) Tepresented by the plane wave (3) where k is the propagation vector of the ejected electron. Let the direction of ejected electron be Z-axis. The z electric field is applied along the direction making polar angles (8. @) with the direction as shown in fig. (7.8). Initially the electron is at a point P(r,6',9'). The P(e) perturbing Hamiltonian H'’ is given by HO) =e (r+E) = 2e (r+Ey)sinwt =2er Ey cos 6" sinwt where 6” is the angle between electric field E(t) and r. Now Ap!) =2 Hp sinwt x’ 90) = fype 6° yydt —_..(4) Fle-7.8. Showing the drection of ejected electron a Arr Sup ¢Eorcos6" wide oO ‘and applied perturbation. The probability of ionisation can be calculated by using the formula of harmonic perturbation i.e., 2. 2 THF | Hef | pF) (8) where p(F) is the density of final state, Hp) = f pp* e Er cos 0" pyr dr sin 8’ d6! dg’ = * 20 ike 41 Fen "0dr J ¢~ 147080" cos 6 sin 6" d0" Sy" a9" = 2mcos fy em "0dr So ew thr cos 6" cos 6’ sin 8’ dé” Consider the integral — Tkr 086" cos 9" sin 8! do” and Lien ier tt fee" *pe dk 4 ewikr =ikr ° ere eit h= rncaso remna|| ehecier- | ° |+apewnenn and E= 1 2uL? In any direction (x, y or 2) in the box, whole number of wavelength should fit Lanien-2% ork 22, an 2a "Tae OL 3 ‘Thus the volume occupied by one state in k-space will be ( a ) Let us consider a small volume in phase space. Allowed states in this volume are given by dk, dk, dk, Ly Ly) p(F) dE = () ake, dy dk, = (4) dk sin 0 d0 dp pate. wk dk AS a, = 7 Pedk _(_L)> Now awe a= (+) 1 dk sin 6 do dp 3 L or PF) = (4) [<] snow A)PERTURBATION THEORY From eqs. (7) and (9) substituting the values of H9(!) and p(F) in eq. (5), we get 2x L\3 (uk). | 1024 1? e? Eo? ag!” k* cos? 6 ¥([(4) (#) sea ES ag? L(1 + ag?)® Thus the probability per unit time that the electron of the hydrogen atom is ejected into solid angle sin @ d0 d@ is given by ml 256K? e? Ey? ay? 2 ei, a ae 0087 8 sin 8.d9 dp --(10) TIL + kag) ) 7.16. THE EMISSION AND ABSORPTION OF RADIATION ‘The most satisfactory treatment of spontaneous emission as well as absorption and induced emission of radiation is due to Dirac. Due to the complexity of the theory, we shall discuss a simpler theory in which absorption and induced emission are related by a general discussion of the Einstein coefficients of emission and absorption of radiation. Absorption of Radiation. Suppose initially an atom is in lower energy state n. Now it is raised to higher energy state m by absorption of a quantum of radiation (photon) of frequency v such that En where E,, and E,, are the energies of the atom in the states m and n respectively. This process is known as absorption of radiation. The probability of occurrence per unit time of this absorption transition n > m depends upon the Properties of the states m and n, and is proportional to the rate at which photons of frequency v fall on the atom and therefore to the spectral energy density. Spontaneous Emission. Suppose the atom is in higher (excited) state m. The life-time of the excited state is usually very small (= 10~ * second), So the atom jumps to the lower energy state n, emitting a photon of frequency. This is known as spontaneous emission of radiation, When there is an assembly of atoms, the radiation emitted spontaneously by each atom has a random direction and fandom phase. So the emission is incoherent. The probability of spontaneous emission is determined by the properties of states m and n. Stimulated or induced Emission. An atom in an excited energy state, clectromagnetic field of a photon of frequency v incident upon it, may jump “mitting an additional photon of the same frequency. Now two photons instead Known as stimulated or induced e1 under the influence of to a lower energy state of one move on. This is ion of radiation. The most important property of this radiation is that it is coherent i, the direction of propagation, energy, phase and state of poloarisation of the Cmitted photon is exactly the same as that of stimulating photon. The probability of induced emission 's proportional to the spectral energy density. ‘The Einstein transition probabilities. According to Bohr’s frequency rule, when a transition takes Place from a nondegenerate stationary state energy E,, to another nondegenerate stationary state _— Snergy E,(E,, > E,), (Fig. (7.9)}, it is accompanied by the emission or absorption of radiation of frequency (1)= we QUANTUM MECN ANG om ts im the kower energy state in a bath of radiation of denstty® 60.) in this my range. then the probability that it will absorb a quantum of energy of radiation and undergo transition to the upper state in unit time is By 22 nnd B, . _ is called Exusein 's coefficient of absorption If at any time there are N,, atoms in the state , the number of such transitions per second ts Nog By =m Pm nm) The probability of emission 1s the sum of two parts, one which ws independent of the radiation density and the other | | | portional to it. The probability that the system in the a pi) upper state m will undergo transition to the lower state with the emission of radiant energy is Ap op + Bm =n Pl mn) where 4... = Einstem s coefficient of spontaneous emission. Fig 79, Einstein transition probabilities Boy aq = Einstein $ coefficeint of induced emission. Ny If \., be the number of atoms in this state, then number of systems undergoing reverse transition Is Nm mn on + Bm = ne mal In thermal equilibrium, emission and absorption must balance so that Ny By -emPCmn) = Nm Am =n + Bm 2nP rnd) Nya _ Amant 8m =nPmn)l rt im B= mPWmn) The equilibrium distribution of atoms among different energy states is given by Boltzmann's law according to which norte EAT by, /kT 7) Substituting in equation (2). brig y’kT _ Amon t Bm nPWmn)] . By =mPVmn) Avpa/kT or By =mPOmne ” — ~ Bm=nP@mn) = Am ~n Ans @ PC mn) = — Tr TZ ~ a Byori tHE Bn § cm., while the wavelength of light in the visible region 1s of the order of 107 S cm., much greater than the dimension of the atom: hence the field be regarded as constant. Thus W E,@) = 2A cos@1=2Acos2ave = Alexp(2 tiv) + exp(- 22 iv D). (9) ‘The density of energy of radiation of frequency v in space, with unit diclectrie constant and magnetic permeability is given by = LPO. (0) where E(¥) represents the average value of the sqaure of the electric field strength corresponding to this radiation. The distribution being isotropic, we can write 1 = Ew (uy) E,(») represents the component of electric field in x-direction ete. 1 3 6 2 No = Py = 20) = 42 2 jow pe) = gz BO) = FEN) =A (22) at Since the average value of cos? 2.2 v ris z Consider two stationary states mandn of an unperturbed system having cnerpies E,, and En(Em > En) and wavefunctions yin) and y,(, respectively. Let at = 0 the system be in the lower State n and come under the perturbing influence of radiation, the electric field strength of each frequency is given by equation (8). Now we shall calculate the probability of the transition to the state 'm. The perturbation energy for a system of electrically charged particles in electric field E,, parallel to “axis is given by HY = EEG 4}, 3)= ms QUANTUM MECHANIog where e, = charge of the jth particle, x-coordinate of the jth particle, E 6); = c-component of electric dipole moment (the sum being taken over all the partictes of yy, 1 5 system). Let us temporarily consider the perturbation due toa single frequency » and then finally summing over all the frequencies. We know that 2ni S a | Wp HO WO de 0 a(t) = — (14) Introducing @,,(0) = 0 and a,,(0) = 1 on the right hand side we have (a,, being the coefficient of Particular state and all other coefficients in the sum being zero) g(t) = = a | Wy," HO WO de ate 2H | yi" TIE nth x Ay) (ert 2ahy Say x yn em ZEA. 45) Let Wma = | Vn OE 6x; vy Ode. i Putting in equation (15), we get de ix i = = anes 2B nn AW) {e271E, Et hah 4 62 HiEm — Eq — hv) hy --(16) Integrating between 0 and 1, 2ni exp{2x i(E,, — E, + hv) t/hy }t Am() = ~~ Usdnn AW) | 2niEm—E,+hv/h [a exp (22 i(E;q — Eq — hv) t/h} | 2xi(E, —E,—hvy/h |g exp2zi, ai E, +hv)t/hy =~ Um n AM) (| 2 iEm E, +hv/h exp{2 (Em, — E, — hv) t/h} -1 2H iE, —E, — hh 1 - exp(22 (En — Ey + hv) t/h} = mn AM) (saa 4 LEO TIE, ~ Ey — hv) /h} En —E, = hv Fora single frequency the term (1,)mqA is always small. The denominator Ey, ~ E — hv causes the influence of the perturbation in changing the system from the state n to a state m; hence neglecting the first term wT) Am) = En Ar) (18) Em —E,—hy 1 = exp{2 iE ~ Ey — hv) t/h) (in case of induced emission the first {erm plays the same role.)PERTURBATION THEORY Z The product dq (1)* dy, (t) i piven by ay (t)" dl) =U mn A) L x Em — En — hv) Ul ~ exp(2.2i(E,, — E, — hv) /h)| [1 = exp{— 22 i(Em 4sin? {(1/h) (Em — Eq — hv) t Eq ~ Eng hy? This expresston is only due to a single frequency and hence in order to include a range of frequencies, we integrate it. Hence hy) i/h}] OF AC)" Olt) = (27) pn ACW) (19) sin?{(/h) (Ep, — Eq — hv) 0} )* a,,(0) = 4,2) 2 tt (20) ml)" drat) = Ate? )m nA nn | Em —E, ah 7 20) ‘The integration can be taken from — © 10 + ©. + sin?{(/h) (Ey, — En — hv) } im(t)™ Am(t) = 4 (1,2 Mma) | ty (21 An" Al) = MIAN) | “Te (21) Consider 7 [- sin? {(t/h) (E,, — En — hv)/t} a Jee En = E, hve : Let {(¢/h) (Em — En — hv) t) =x 27+ gino 6 a2 2 < e(Z sintx (_ de) xr (*@sintx , x? ne r=) [f-ing rte 2 as | Sik ae =x. (22) 2 Substituting the value of this integral in equation (21), we have 4x2 (9 al) = FF oan Amn) 3) + To prove that [1 ~ expi2.xiG,, — E, — hv /h}] x [1 - exp{- 2246, ~ Ey — hy th} = 4sin® {n/h(E,, ~ Ey — hv) th [1 ~ expt2. xi, — E, — hy) 17h}] x [1 ~ exp{— 2246, - E, —hvy/h)] = 1 ~exp(2.1i€,, ~ E, —hvyt/h) ~ exp{-2ni(E,, - E, —hyyt/h) +1 Hexp(2 x iE — Heat xi hvythh] + expt- 22H, hyhyy = 2-202 (Eq — Ey ~ hv) t7h] = 2[1 - cos 2 (Ep, — Ey — hv) t7h] = 21 - (1 - 2s? a(E,, ~ E, — hv) t/h} | = 4 sin? Fen Fe- ho]QUANTUM MECHANICS, The probability of transition per unit time from state n tom Is 4x 2 2 Sn Amn): re ‘Substituting the value of A from equation (12), we get 4 At An (t)* ant) = > 4 mn PU mn) (24) The expression for y and z-directions are similar. From equation (24) we obtain the Einstein coefficient of absorption By, + m aS 3 2, By 2m = SUP nn + 05a + Oma 8) Similarly, Einstein's coefficient of induced emission is 82 2 Bonen = 555 Ue mn + Wy mn + He mnl 26) Using the classical relation between A, ., and B, . -, the Einstein coefficient of spontaneous emission is Wiig? SEE Winn + Onn * Wl 2) Am+n= 7.47. SELECTION RULES FOR SIMPLE HARMONIC OSCILLATOR If a particle executes a hormonic motion along x-axis and possesses a charge +6. the instantancous dipole moment is ex. At the equilibrium, the charge is —e¢ for an ascillator type behaviour. “The transition between two states can be represented by the integral 7 | mlx yg dt = eX an @ This should be nonvanishing for a transition to take place between states m and n. +? Na Nm 2) Now mn = | a Hn Hm Exp(— £7) E a. 2 (in case of inear harmonic oscillato") NaN (*™ = Mate | Hy Hym exp(— &) & dB. = Hy(e)s" We know that exp(-s? +258) = z, 7 : = Hee and an +2 = 5 Aah ; - = 5 expl- 2 +2sdpem(-FH2He= SF fel Hl) Hn) n=O maon!mr ee PERTURBATION THEORY = Multiplying by & exp. (~ &), we get | exp(— 5? + 25 &) exp(— 2 + 218) exp(— £2) Ede = = am r=0 mao” | Hy(B) Hra() exp(— 87) & 48: B) Consider the integral +0 =| exp(= 52 + 2s fexp(—P +208) PEE -| exp(- 2 — P+ 258+ 218 - EYEE =en{ * expl- 5-04 Bab =e exp[- G-5-N'E-s-NdE-s—9 +erero| exp(- (E-s - 97} dE = 5-1) ‘The first integral vanishes and second gives V(t). Thus TaV@) + Ne" evo5oro[tetate sit +4 Ost | | = vay [42 ite. yore l ae Honeewe gtr = v5 [#28 + noes) 7 ~e - 3 3am {" Hiy(G) Hal@) Bexp(~ ab 4) Hence comparing coefficient of #" £”, we See that 2q nS Zero except for m =n * 1. This shows that mr will be equal ton — 1 and n + 1 for nonvanishing integral. Thus ment or (m-ny=el ie, Aneel. Hence the only transitions which this system can undergo with the emission oF absorption of radiations: are those from a given stationary state to the two adjacent states.(QUANTUM MECH: — iANICS, 7.18. ADIABATIC AND SUDDEN PERTURBATIONS (1) General adiabatic perturbation, We have treated this ease ofa Slows WINE potential ony when the potential is so small that perturbation theory can be used. We shall extend this treatment tog more general problem in which the perturbing potential undergoes large changes Now we wish to solve the Schroedinger equation a in 1 =H () when H(0) varies slowly with time. Thus when HG varies slowly enough, we may expect that a good approximation should be given by solving Schroedinger's equation at each instant of time under the assumption that /7 is a constant and equal to its instantancous value H(1'), where 1’ is the value of rat which we wish to evaluate H. The stationary state wavefunctions, obtained by setting ¢ = 1" = constant, would satisfy the equation AC) Wal) = En) vn) (2) ‘One may now expect that if H is a slowly varying function of ,a good approximate solution is if’ "dt" Py.) | a [iz ) ar | (3), So when the wavefunction is known at zero time, at later times we put W = Eay(t) yn(e.t) exp - if E,(t') dr (a) 0 Substituting eq. (4) into eq. (1), we get » a my (invara ee }e0( - fz] Penal a sie n : 1 or my (4 Vn +0) eo ~4 [sre] “pene or ind ( In Vn + Gy cee ) ov(- 2 afar) (3) Multiplying eq. (5) by Yn* exp (*/ Enn(t!) att ‘and integrating over all space, we get ° x D an Vn Vin on = xf eer ae’ o( te [ E,(u') dt’ ar 0 +2 | une “tas on[-a) E, tna aPERTURBATION THEORY Using orthogonality and normalization condition, we have Den vt Sn Spon - if (En — Em) at" dv a : ] Das | vat Seas, on =f, (Eq ~ En) de | 4, or bn Now we shall obtain the value of the integral, £., | Ym" eae as follows : We know that AO Vn) = En Ya: Differentiating eq. (7) with respect to ¢ gives Hy, a Multiplying by y,,* and integrating over the co-ordinates, aE, n [ont Bnet + [vqr Ea = S| vat Yndt +E, [v0 * Now [var waa =0 and remembering that H is Hermitian [vneu Substituting in ¢q. (8), or (Em En) | * or [var Substituting this into eq. (6), of - fw (2) me ]oe(-Sfe-204" an= 2d En — En) nem [x(t] ancl toe E.- Ew abe or an +E (6) (7) (8) (9) (10) We now proceed as in the method of variation of constants. Suppose that the system starts with Landa, = 0 for n # s; then by successive approximations one solves [OF dyy.QUANTUM MI fa IECHANICg (11) where oH ‘To estimate the matrix clement, we can neglect the slow change of (3) We then obtain (2) [or (-46-20") ~en(-406.- £m") | (2) ‘The expoenntial factor in equation (12) is at most of the order of unity. Hence, the total Probability of transmission 10 the mth level is less than 3 2 ane oH = md ms oH 7 7 It (# ) is small enough, a negligible error is made by neglecting | a,,,, |? and saying that the (13) system remains in the state y,(x,1) even though y, itself is changing with time. This is known as adiabatic approximation. For the application of the adiabatic approximation we shall discuss the collisions of gas molecules. If the gas atoms approach each other, an important question arises whethere the foce resulting from their interaction can cause transitions among the electronic states. We know that molecular velocities are ususally low (about 10° cm./sec.) whereas velocities of electrons in atoms are much higher (= 10° cm./sec.) . During an electronic period of rotation in the atom, the molecule does not move very far, so the interaction energy does not change much. Thus the collision may be regarded as adiabatic process. in which the electron remains in its original quantum state. ‘The validity of adiabatic approximation requires not only the smallness of (aH/at), but also that Eym ~ E, Shall not become too small. (2) Sudden approximation. The sudden approximation is used when the change in the Hamiltonian occupies a very short but finite interval of time. To do this, let us suppose that at the time += 0, the Hamiltonian suddenly changes from Ho to Hy and remains constant; therefore, the wavefunctions are given by Yq exp. [(—i/h) E, (n)] where Ho Wn = Eq yy. After ¢= 0, the cigenfunction and Hamiltonian operator are denoted as Yin OME and Hy Yim = Em Yn To find how y changes after r = 0, we adopt the procedure of expanding it in a series solution of Schrocdinger’s equation é.., Yn) =Z Gn Yn). (4) The values of C,,, ure obtained by multiplying equation (14) by y4q*(x) and integrating. We have Crain = S ¥en" (2) Y(t) drTURBATION THEO; PERTURI ay Es After r= 0, the wavefunction becomes ¥n0) = 7 Sinn Vn) €~ OM Em! = Ef on™) Wn () I Yn) 7 VM EmE on( 16) “The above derivation holds for instantaneous changes of the Hamiltonian and must therefore also hold for sufficiently rapid, but not instantaneous changes. The essential simplification resulting from the instantaneous change was that the wavefunction did not change, while the Hamiltonian was changing, SOLVED PROBLEMS ——— 1. A linear harmonic oscillator of charge e is perturbed by an electric field F in the positive edirection. The first order change in the energy level is zero. Why ? Calculate the energy upto second order. ‘The potential energy of the oscillator is (1/2) kcx? plus a small quantity e Fx where ¢ is the charge of the oscillator. The Schroedinger equation is, thus He 42m (eta vere) 920 (I) ‘The additional term — ¢ Fx can be regarded as a perturbation Thus AHO) = -eFx wA2) Setting =-eF (3) then HO =x. ‘The unperturbed wavefunction in a linear harmonic oscillator are given as nl) = Nn Hn(E) exp.(~ 87/2). (A) where Nn {aoa} ee V4) eV (5) ‘When perturbation is applied, the matrix elements Hn, a{l) are given as s(t) ey ; 2 ye ttm = || aaa | teres 8+ | rs} X Hy(€) exp(— £7/2) dr -.(5) +2 v2 1 1 a 2, =| 3 eas Hin) § Banas) exD(— £) 4. Putting Hy(E) = 5 Hn + 16) +" Hn —1@) +00 Ha) = z [ Hn(®) mn) = @) Tom tO mint mis. Hy + 1G) + Hn | exp(— 7) dé (6) RE6 QUANTUM MECHANICS The right hand side vanish except when m =n + Lorm =n — 1 Also [Te even ray? dg = (RYE ti + e Vow rg | tN Layette ot “| Teh} -iv(t}) (a) 1a nar at Form =n — 1, we have 1 We tara Geeta me “V/( LV (23!) mnsee mansl (10, tv(s) seen Using equation (10), E20 The first order energy correction is zero because perturbing Hamiltonian 1"? es an odd fanction oft and hence total integrand giving the value of £,(") is also an odd fanction oft. Henoe tts valee comes out to be zero. Physically, we can also say that as the first order correction depends only oa its expectation value, 1¢., the first order energy correction ts zero. ieabee) Ha Now £, = — xs ay Lv ml HO yO ate fg? HD yp ate 2 S¥m a ee Ey) a > EB (2 1 1 eS ao ml ‘ 1) 2ahe Ime ; am AS E= [n+ 3) nw as) Ey = E42 E,@) os We have E> (n+ t) no - EF (16) 2 Imo Show that the problem in Fx. 1 can be solved exactly, Compare the result with perturbation ‘approximation, HHO eH : (0 2)PERTURBATION THEORY 7 a) Put qer-th (a) mo Le origin ts to be shifted framx = 0 ton = 0. Now we have mo? ooh aa mo? —eF we fl mete at Imai timer ener +3 (6) 2 ‘This expression shows that kinetic energy of the system is reduced by the constant value £5 or m total energy E is reduced by the same value. Thus the sofution will be on the same lines as 1s obtained for one dimensional harmonic oscillator problem (refer art. 4.9), Thus the problem can be solved exactly and the energy value will be 2 Eye (ne })}no- 55 m {tis exactly the same as obtained in second order perturbation theory. A. Use the second order perturbation theory to calculate the change in energy of a linear harmonic oscillator when It is perturbed by a constant force A x. Show that it can be solved exactly: Replace in above two exercises ¢ F by A. 4. When perturbing term in linear harmonic oscillator problem is (1/2) a7, ase perturbation theory to obtain correction in energy eigenvalue upto first order, Show that this problem can also be solved exactly. ‘The Hamiltonian in one dimensional harmonic oscillator is given by 2 HaPiglets tar a rm*2 2 or He HO +H (2) 1 L HOmsar ij First order correction EU a Ey = | yi “ | | ‘The-unperturbed wave function in mth state is Pn = Ny Hytarrye (syi ea area ememee a | cee QUANTUM MECHANIgg ie rm -a Therefore H (cx) € de (7) Fen) en? an (8) putting or (9) using eqn. (10) and (12) of art 7.2. Hence EM = Main? B fe \? or E, =$@n +) (33) (10) Energy corrected upto first order ’ wy 4 2 E,= (n+4}no+teren (3) or E,= (n+) nol i+ (11) n 2 2k Exact solution po 2 =P yt yy (12) HO =f + zhx (12) aL altel ee (13) 2m*2 2 2 Py 1 2 wn(4) Hast gk ta)x ( 12 v2 kta) _(k 2 15) Thus (& )-(4) (1+2) A 2 w= w 1a. (19) =-V{£ where w= VFEI THO! OF PERTURBATION THEORY = The corrected value of energy is ) ho (7) i+ a (18) The energy corrected upto first order is 1 a (n+3)o(t+ 34] ay and upto second order 1) afie-2- (20) 2k sk ‘The expression (19) is exactly the same as derived on the basis of perturbation theory lator whose Calculate the first order energy correction for one dimensional harmonic os Hamiltonian is 2 =f 4,1 4 Hast ke tax and the ground state wavefunction is, ry = |) exp - Yo= lane} P| —2ne Further show that the value of ground state energy can be written as ho Jato B=9 (3%) ‘The perturbing Hamiltonian HY = axt (1) +0 Hence £0= | Yo)” axt yf de 2) Substituting for yo V2 pte =|" xx (az) [o(-88 carton fo}e V2 [+= ia =(s5) of “on(-f ne) te “ k KM pte =2(40) “f on |e 88) 7 Now we make use of the standard integral r (= +1 ) ” - Nem an” | [ive dear me or neO (6)