Baixe o app para aproveitar ainda mais
Prévia do material em texto
1 WKB approximation: particle decay Masatsugu Sei Suzuki Department of Physics, SUNY at Binghamton (Date: November 15, 2011) WKB approximation This method is named after physicists Wentzel, Kramers, and Brillouin, who all developed it in 1926. In 1923, mathematician Harold Jeffreys had developed a general method of approximating solutions to linear, second-order differential equations, which includes the Schrödinger equation. But even though the Schrödinger equation was developed two years later, Wentzel, Kramers, and Brillouin were apparently unaware of this earlier work, so Jeffreys is often neglected credit. Early texts in quantum mechanics contain any number of combinations of their initials, including WBK, BWK, WKBJ, JWKB and BWKJ. The important contribution of Jeffreys, Wentzel, Kramers and Brillouin to the method was the inclusion of the treatment of turning points, connecting the evanescent and oscillatory solutions at either side of the turning point. For example, this may occur in the Schrödinger equation, due to a potential energy hill. (from http://en.wikipedia.org/wiki/WKB_approximation) ________________________________________________________________________ Gregor Wentzel (February 17, 1898 – August 12, 1978) was a German physicist known for development of quantum mechanics. Wentzel, Hendrik Kramers, and Léon Brillouin developed the Wentzel–Kramers–Brillouin approximation in 1926. In his early years, he contributed to X-ray spectroscopy, but then broadened out to make contributions to quantum mechanics, quantum electrodynamics, and meson theory. http://en.wikipedia.org/wiki/Gregor_Wentzel Hendrik Anthony (2 February 1894 – 24 April 1952) was a Dutch physicist who worked with Niels Bohr to understand how electromagnetic waves interact with matter. http://en.wikipedia.org/wiki/Hans_Kramers Léon Nicolas Brillouin (August 7, 1889 – October 4, 1969) was a French physicist. He made contributions to quantum mechanics, radio wave propagation in the atmosphere, solid state physics, and information theory. http://en.wikipedia.org/wiki/L%C3%A9on_Brillouin ________________________________________________________________________ 1. Classical limit Change in the wavelength over the distance x x dx d . When x dx d . 2 In the classical domain, dx d or 1 dx d , which is the criterion of the classical behavior. 2. WKB approximation The quantum wavelength does not change appreciably over the distance of one wavelength. We start with the de Broglie wave length given by h p , )( 2 1 2 xVp m , or )]([2 2 2 xVm h p , or )((2 xVmp . Then we get ] )( [22 32 dx xdV m dx d h , or dx xdV p mh dx xdV p h h m dx xdV h m dx d )()()( 3 3 2 3 2 . When 1 dx d , we have 1 )( 3 dx xdV p mh (classical approximation) 3 If dV/dx is small, the momentum is large, or both, the above inequality is likely to be satisfied. Around the turning point, p(x) = 0. |dV/dx| is very small when V(x) is a slowly changing function of x. Now we consider the WKB approximation, )()( 2 )( 2 22 xxV xm x . When 0V , ipx ikx AeAex )( . If the potential V is slowly varying function of x, we can assume that )( )( xS i Aex , .....)( !3 )( !2 )( !1 )()( 3 3 2 2 10 xSxSxSxSxS . ____________________________________________________________________ ((Mathematica)) 4 WKB approximation eq1 —2 2 m Dx, x, 2 Vx x ∂ x; rule1 Exp — S & ; eq2 eq1 . rule1 Simplify Sx — 2 m ∂ 2 m Vx Sx2 — Sx 2 m rule2 S S0 — S1 — 2 2 S2 — 3 3 S3 — 4 4 S4 & ; eq3 2 ∂ m 2 m Vx Sx2 — Sx; eq4 eq3 . rule2 Expand; list1 Tablen, Coefficienteq4, —, n, n, 0, 6 Simplify; TableForm 0 2 m ∂ 2 m Vx S0x2 1 2 S0x S1x S0x 2 S1x2 S0x S2x S1x 3 S1x S2x 1 3 S0x S3x 1 2 S2x 4 1 12 3 S2x2 4 S1x S3x S0x S4x 2 S3x 5 1 24 4 S2x S3x 2 S1x S4x S4x 6 1 72 2 S3x2 3 S2x S4x _______________________________________________________________________ For each power of ħ, we have 0)]('[)(22 20 xSxmVm , 5 )(")(')('2 010 xiSxSxS , )(")(')(')]('[ 120 2 1 xiSxSxSxS , ..................................................................................................................................... (a) Derivation of )(0 xS )()]([2)]('[ 220 xpxVmxS , where )]([2)(2 xVmxp , or )()('0 xpxS , or x x dxxpxS 0 )()(0 . Since )()( xkxp , x x dxxkxS 0 )()(0 . (b) Derivation of )(1 xS )(")(')('2 010 xiSxSxS , )(' )(' 2)('2 )(" )(' 0 0 0 0 1 xS xS dx d i xS xiS xS , which is independent of sign. )](ln[ 2 )]('ln[ 2 )(')( 011 xk i xS i dxxSxS , or 6 2/1 1 )](ln[)](ln[2 1 )( xkxkxiS , or )( 1)(1 xk e xiS . (c) Derivation of )(2 xS )(")(')(')]('[ 120 2 1 xiSxSxSxS , )(' )]('[)(" )(' 0 2 11 2 xS xSxiS xS . Then the WKB solution is given by ...)](ln[ 2 )( .....)( !3 )( !2 )( !1 )()( 0 3 3 2 2 10 xkidxxk xSxSxSxSxS x x The wave function has the form )]})(exp["])(exp['))]{(ln( 2 1 exp[)( 00 x x x x dxxkiBdxxkiAxkx , or ])(exp[ )( ])(exp[ )( ])(exp[ )( ' ])(exp[ )( ' )( 00 00 x x x x x x x x dxxki xk B dxxki xk A dxxki xk B dxxki xk A x . where we put 'A A , 'B B . 3. The probability current density 7 We now consider the case of B = 0. ])(exp[ )( )( 0 x x dxxki xk A x . The probability is obtained as mv A xk A xxxP 22 )( )()(*)( , where mvxk )( . The probability current density is 2 2 2 A mmv A vvJ . Fig. 2avdtJadt , or 2vJ . 4. WKB approximation near the turning points We consider the potential energy V(x) and the energy shown in the following figure. The inadequacy of the WKB approximation near the turning point is evident, since 0)( xk implies an unphysical divergence of )(x . (a) V(x): increasing function of x around the turning point x = a 8 Vx x aO E (i) For x>>a where V(x)>, ])(exp[ )( ])(exp[ )( )( 11 x a x a I dxx x B dxx x A x , where A1 and B1 are constants, and )(21)( xVmx , (ii) For x<a where V(x)<, ])(sin[ )( ])(cos[ )( )( 22 a x a x II dxxk xk B dxxk xk A x , where )(2 1 )( xVmxk . ________________________________________________________________________ (b) V(x): decreasing function of x around the turning point 9 Vx x bO E (i) For x<<b where V(x)>, ))(exp( )( ))(exp( )( )( b x b x I dxx x B dxx x A x , with )(21)( xVmx , (ii) For x>b where V(x)<, )))(sin( )( ))(cos( )( 2 )( x b x b II dxxk xk B dxxk xk A x where )(2 1 )( xVmxk . _______________________________________________________________________ 5. Exact solution of wave function around the turning point x= a 10 Vx x aO E The Schrödinger equation is given by )()( 2 2 22 xxV dx d m , or 0])([ 2 2 22 xV dx d m , where is the energy of a particle with a mass m. We assume that )()( axgExV , in the vicinity of x =a, where g>0. Then the Schrödinger equation isexpressed by 0)( 2 22 2 axgm dx d . Here we put )( 2 3/1 2 ax mg z . ((Note)) dz dmg dz d dx dz dx d 3/1 2 2 , 11 2 23/2 22 2 2 dz dmg dx d . Then we get 0)( )( 2 2 zz dz zd . The solution of this equation is given by )()(2)( 21 zBCzACz ii , where we use 2C1 instead of C1. The asymptotic form of the Airy function Ai(z) for large |z| is given by ) 4 cos(||)( 4/12/1 zzAi , for z<0 and ezzAi 4/12/1 2 1 )( , for z>0 where 2/3|| 3 2 z 12 -10 -8 -6 -4 -2 2 4 z 0.5 1.0 Aiz Fig. Plot of the Ai(z) (red) and its asymptotic form (blue) as a function of z for z<0. The asymptotic form of the Airy function Bi(z) for large |z|, ) 4 sin(||)( 4/12/1 zzBi , for z<0 ezzBi 4/12/1)( , for z>0 with 2/3|| 3 2 z 13 -10 -8 -6 -4 -2 2 z -0.5 0.5 1.0 1.5 2.0 Biz Fig. Plot of the Bi(z) (red) and its asymptotic form (blue) as a function of z for z<0. Here we note that For z<0, 2/1 3/1 22 || 2 )( 2 )( z mg xag m xk . Then we have 2/3 2/3 2/1 2 2/1 2 || 3 2 )( 2 3 2 2 )( z xa mg dxxa mg dxxk a x a x For z>0, 2/1 3/1 22 || 2 )( 2 )( z mg axg m x , we have 14 2/3 2/3 2/1 2 2/1 2 || 3 2 )( 2 3 2 2 )( z ax mg dxax mg dxx x a x a ______________________________________________________________________ 6. Connection formula (I; upward) Vx x aO E 15 Fig. Connection formula (I; upward). The condition 1 dx d for the WKB approximation is not satisfied in the vicinity of x = a. We need the asymptotic form the exact solution of the Schrödinger equation. (i) Asymptotic form for z<0 (x<a) The asymptotic form of the wave function for z<0 (x<a) can be expressed by )] 4 )(sin( )( 1 ) 4 )(cos( )( 1 2[ 2 ) 4 sin(||) 4 cos(||2)()(2 2 1 6/1 2 2/1 4/12/1 2 4/12/1 121 a x a x ii dxxk xk C dxxk xk C mg zCzCzBCzAC (1) where 2/3|| 3 2 )( zdxxk a x , 2/1 3/1 2 || 2 )( z mg xk . 16 (ii) The asymptotic form for z>0; The asymptotic form of the wave function for z>0 (x>a) can be expressed by )])(exp( )( 1 ))(exp( )( 1 [ 2 )()(2 2 1 6/1 2 2/1 4/12/1 2 4/12/1 121 x a x a ii dxx x C dxx x C mg ezCezCzBCzAC (2) where 2/3|| 3 2 )( zdxx x a , 2/1 3/1 2 || 2 )( z mg x . From Eqs.(1) and (2) we have the connection rule (I; upward) as follows. ])(exp[ )( ])(exp[ )( ] 4 )(sin[ )( ] 4 )(cos[ )( 2 x a x a a x a x dxx x B dxx x A dxxk xk B dxxk xk A (I; upward) at the boundary of x = a. Vx x aO E where C1 = A and C2 = B. 17 ______________________________________________________________________ 7. Exact solution of wave function around the turning point x= b Vx x bO E The Schrödinger equation is given by )()( 2 2 22 xxV dx d m , or 0])([ 2 2 22 xV dx d m , where is the energy of a particle with a mass m. We assume that )()( bxgxV , in the vicinity of x =b, where g>0. The Schrödinger equation is expressed by 0)( 2 22 2 bxgm dx d . Here we put )( 2 3/1 2 bx mg z . Then we get 18 0)( )( 2 2 zz dz zd . The solution of this equation is given by )()(2)( 21 zBCzACz ii . We note the following. (i) For z<0 (x>b) k(x) is expressed by 2/1 3/1 22 || 2 )( 2 )( z mg bxg m xk , 2/32/3 2/1 2 2/1 2 || 3 2 )( 2 3 22 )( zbx mg dxbx mg dxxk x b x b . (ii) For z>0 (x<b), where >V(x) (x) is expressed by 2/1 3/1 2 22 2 )( 2 )]([ 2 )( z mg xbg m xV m x 2/32/3 2/1 2 2/1 2 3 2 )( 2 3 22 )( zxb mg dxxb mg dxx b x b x __________________________________________________________________ 8. Connection formula-II (downward) 19 Fig. Connection formula (II; downward). The condition 1 dx d for the WKB approximation is not satisfied in the vicinity of x = b. We need the asymptotic form the exact solution of the Schrödinger equation. The asymptotic form for z<0; )] 4 )(sin( )( 1 ) 4 )(cos( )( 1 2[ 2 ) 4 sin(||) 4 cos(||2)()(2 2 1 6/1 2 2/1 4/12/1 2 4/12/1 121 x b x b ii dxxk xk C dxxk xk C mg zCzCzBCzAC The asymptotic form for z>0; 20 )])(exp( )( 1 ))(exp( )( 1 [ 2 )()(2 2 1 6/1 2 2/1 4/12/1 2 4/12/1 121 b x b x ii dxx x C dxx x C mg ezCezCzBCzAC Then we have the connection formula (II; downward) as ] 4 )(sin[ )( ] 4 )(cos[ )( 2 ])(exp[ )( ])(exp[ )( x b x b b x b x dxxk xk B dxxk xk A dxx x B dxx x A (II, downward) with Vx x bO E where C1 = A and C2 = B. 9. Tunneling probability We apply the connection formula to find the tunneling probability. In order that the WKB approximation apply within a barrier, it is necessary that the potential V(x) does not change so rapidly. Suppose that a particle (energy and mass m) penetrates into a barrier shown in the figure. There are three regions, I, II, and III. 21 ba I II III Vx E Fig. The connection formula I (upward) is used at x = a and the connection formula II (downward) is used at x = b. For x>b, (region III) ] 4 )(sin[ )( ] 4 )(cos[ )( )] 4 )((exp[ )( 1 1 1 1 1 1 x b x b x b III dxxk xk iA dxxk xk A dxxki xk A (we consider on the wave propagating along the positive x axis), where ))(( 2 )( 21 xV m xk . The connection formula (II, downward) is applied to the boundary between the regions III and II. ] 4 )(sin[ )(2 ] 4 )(cos[ )( ])(exp[ )(2 ])(exp[ )(2 1 1 1 1 x b x b b x b x dxxk xk B dxxk xk A dxx x B dxx x A (II, downward) Here we get 22 iAB 2 . Then we get the wave function of the region II, ] 4 )(sin[ )( ] 4 )(cos[ )( ])(exp[ )( ])(exp[ )(2 1 1 1 1 x b x b III b x b x II dxxk xk iA dxxk xk A dxx x iA dxx x A or ])(exp[ 2)( ])(exp[ 1 )( ])()(exp[ )( ])()(exp[ )(2 x a x a x a b a b a x a II dxx r x A dxx rx iA dxxdxx x iA dxxdxx x A where ])([ 2 )( 2 xVmx , and ])(exp[ b a dxxr , Next, the connection formula (I; upward) is applied to the boundary between the regions II and I. 23 )])(exp( )( ))(exp( )( )] 4 )(sin( )( ) 4 )(cos( )( 2 2 2 2 2 x a x a a x a x dxx x D dxx x C dxxk xk D dxxk xk C (I; upward) Here we get r iAC 1 , r A D 2 . Then we have the wave function of the region I, )]} 4 )((exp[)] 4 )(({exp[ 4)( )]} 4 )((exp[)] 4 )(({exp[ 1 )( ] 4 )(sin[ )(2 ] 4 )(cos[ 1 )( 2 22 2 22 2 2 2 2 2 a x a x a x a x a x a x I dxxkidxxki ir xk A dxxkidxxki rxk iAdxxkr xk A dxxk rxk iA or )]} 4 )((exp[)) 4 1 ()] 4 )((exp[) 4 1 {( )( )]} 4 )((exp[) 1 4 ()] 4 )((exp[) 1 4 {( )( 22 2 22 2 x a x a a x a x I dxxki r r dxxki r rxk iA dxxki r r dxxki r r xk iA The first term corresponds to that of the reflected wave and the second term corresponds to that of the incident wave. Then the tunneling probability is 24 ))(2exp( ) 4 1 ( 1 2 2 b a dxxr r r T , where ))(exp( b a dxxr . 9. -particle decay: quantum tunneling r1 r2 Coulomb repusion Nuclear binding Vr rO E Fig. Gamov's model for the potential energy of an alpha particle in a radioactive nucleus. 25 -2 0 2 4 6 8 10 r nu cl ea r po te nt ia lM eV Re yar Ea Fig. The tunneling of a particle from the 238U (Z = 92). The kinetic energy 4.2 MeV. http://demonstrations.wolfram.com/GamowModelForAlphaDecayTheGeigerNutt allLaw/ For r1<r<r2. )(21)( rVmr . At r = r2, 20 2 1 4 2 r eZ . The tunneling probability is ])(2exp[ 1 2 r r drreP , where 26 ])1([arccos 2 ])(arccos[ 2 1 2 )( 2 )( 2 1 2 1 2 1 2 121 2 1 2 2 2 1 2 1 2 1 r r r r r r r m rrr r r r m dr r rm drrV m drr r r r r r r where m is the mass of -particle (= 4.001506179125 u). fm = 10-15 m (fermi). The quantity P gives the probability that in one trial an particle will penetrate the barrier. The number of trials per second could estimated to be 12r v N , if it were assumed that a particle is bouncing back and forth with velocity v inside the nucleus of diameter 2r1. Then the probability per second that nucleus will decay by emitting a particle, called the decay rate R, would be 2 12 e r v R . ((Example)) We consider the particle emission from 238U nucleus (Z = 92), which emits a K = 4.2 MeV particle. The a particle is contained inside the nuclear radius r1 = 7.0 fm (fm = 10-15 m). (i) The distance r2: From the relation 20 2 4 2 r Ze K , we get r2 = 63.08 fm. (ii) The velocity of a particle inside the nucleus, v: 27 From the relation 2 1 2 1 vmK , where m is the mass of the a particle; m = 4.001506179 u, we get v = 1.42318 x 107 m/s. (iii) The value of : ])(arccos[ 2 121 2 1 2 rrrr r r mK =51.8796. (iv) The decay rate R: 2 12 e r v R = 8.813 x 10-25. ((Mathematica)) 28 Clear"Global`"; rule1 u 1.660538782 1027, eV 1.602176487 1019, qe 1.602176487 1019, c 2.99792458 108, — 1.05457162853 1034, 0 8.854187817 1012, MeV 1.602176487 1013, Ma 4.001506179125 u, fm 1015, Z1 92, r1 7 fm , K1 4.2 MeV; eq0 K1 2 Z1 qe2 4 0 r . rule1 6.729141013 4.245021026 r eq01 Solveeq0, r; r2 r . eq011 6.308421014 r2 fm . rule1 63.0842 eq1 1 2 Ma v2 K1 . rule1; eq2 Solveeq1, v; v1 v . eq22 1.42318107 2 Ma K1 — r2 ArcCos r1 r2 r1 r2 r1 . rule1 51.8796 R1 v1 2 r1 Exp2 . rule1 8.812821025 29 9. Bound state: Bohr-Sommerfeld condition. ab I II III Vx E For x<b (region I), the unnormalized wave function is ])(exp[ )( 1 b x I dxx x , Using the connection rule (II; downward) ] 4 )(sin[ )(2 ] 4 )(cos[ )( ])(exp[ )(2 ])(exp[ )(2 x b x b II b x b x dxxk xk B dxxk xk A dxx x B dxx x A I (II; downward) we get 2A , B = 0. Then we have )] 4 )(cos[ )( 2 x b II dxxk xk . for b<x<a This may also be written as 30 ] 4 )(sin[])(cos[ )( 2 ] 4 )(cos[])(sin[ )( 2 ] 4 )()(sin[ )( 2 ] 24 )()(cos[ )( 2 ] 4 )()(cos[ )( 2 ] 4 )(cos[ )( 2 a x a b a x a b a x a b a x a b a x a b x b II dxxkdxxk xk dxxkdxxk xk dxxkdxxk xk dxxkdxxk xk dxxkdxxk xk dxxk xk Here we use the connection rule (I, upward), ])(exp[ )( ])(exp[ )( ] 4 )(sin[ )( ] 4 )(cos[ )( 2 x a x a a x a x dxx x B dxx x A dxxk xk B dxxk xk A (I; upward) From this we have ] 4 ))(sin[])(cos[ )( 2 ] 4 ))(cos[])(sin[ )( 2 a x a b a x a b II dxxkdxxk xk dxxkdxxk xk with ])(sin[ a b dxxkA , ])(cos[2 a b dxxkB . Since III should have such a form 31 ])(exp[ )( x a III dxx x A . for x>a. Then we need the condition that 0])(cos[2 a b dxxkB , or ) 2 1 ()( ndxxk a b , or ) 2 1 ()( ndxxp a b , where n = 0, 1, 2, ... (Bohr-Sommerfeld condition). 10. Simple harmonics We consider a simple harmonics, 22 00 22 2) 2 1 (2))((2)( xxmxmmxVmxp , where 2 0 0 2 m x . Then we get 0 2 0 0 2 0 0 0 22 00 2 2 1 4 22)( 00 0 m m x mdxxxmdxxp xx x . When ) 2 1 ()( 0 0 ndxxp x x , 32 we have ) 2 1 ( 0 n , or ) 2 1 ( n _______________________________________________________________________ APPENDIX Connection formula )(2 1 )( xVmxk )(21)( xVmx (i) Connection formula at x = a (upward) 33 )])(exp( )( ))(exp( )( )] 4 )(sin( )( ) 4 )(cos( )( 2 1 1 1 1 x a x a a x a x dxx x B dxx x A dxxk xk B dxxk xk A formula I (upward) (ii) Connection formula at x = b (downward) )] 4 )(sin( )(2 ) 4 )(cos( )( )])(exp( )(2 ))(exp( )(2 2 2 2 2 x b x b b x b x dxxk xk D dxxk xk C dxx x D dxx x C formula II (downward) (i) Connection formula at x = b ] 4 )(sin[ )(2 ] 4 )(cos[ )( ])(exp[ )(2 ])(exp[ )(2 1 1 1 1 x b x b b x b x dxxk xk D dxxk xk C dxx x D dxx x C formula II (downward,) (ii) Connection formula at x = a 34 ])(exp[ )( ])(exp[ )( ] 4 )(sin[ )( ] 4 )(cos[ )( 2 2 2 2 2 x a x a a x a x dxx x B dxx x A dxxk xk B dxxk xk A formula I (upward)
Compartilhar