Physics 2D Lecture Slides Lecture 16: Feb 12th Sunil Sinha UCSD Physics P(x,t)= |Ψ(x,t) |2 Wave Function of “Stuff” & Probability Density Probability of a particle to be in an interval a ≤ x ≤b is area under the curve from x=a to a=b x x=a x=b • Although not possible to specify with certainty the location of particle, its possible to assign probability P(x)dx of finding particle between x and x+dx • P(x) dx = | Ψ(x,t)|2 dx • E.g intensity distribution in light diffraction pattern is a measure of the probability that a photon will strike a given point within the pattern Ψ: The Wave function Of A Particle • The particle must be some where +" 2 | ! ( x , t ) | dx = 1 $ #" • • Any Ψ satisfying this condition is NORMALIZED Prob of finding particle in finite interval The Wave Function is a mathematical function that describes a physical object Wave function must have some rigorous properties : • P(a " x " b) = #! * ( x, t ) ! ( x, t )dx • a • Fundamental aim of Quantum Mechanics • – Given the wavefunction at some • instant (say t=0) find Ψ at some b subsequent time t – Ψ(x,t=0) Ψ(x,t) …evolution – Think of a probabilistic view of particle’s “newtonian trajectory” • We are replacing Newton’s 2nd law for subatomic systems Ψ must be finite Ψ must be continuous fn of x,t Ψ must be single-valued Ψ must be smooth fn d! must be continuous dx WHY ? Bad (Mathematical) Wave Functions Of a Physical System : You Decide Why ? A Simple Wave Function : Free Particle • Imagine a free particle of mass m , momentum p and K=p2/2m • Under no force , no attractive or repulsive potential to influence it • Particle is where it wants : can be any where [- ∞ ≤ x ≤ + ∞ ] – Has No relationship, no mortgage , no quiz, no final exam….its essentially a bum ! – how to describe a quantum mechanical bum ? • Ψ(x,t)= Aei(kx-ωt) =A(Cos(kx-ωt)+ i sin (kx-ωt)) p E ; != ! ! For non-relativistic particles k= p2 !k 2 E= " ! (k)= 2m 2m Has definite momentum and energy but location unknown ! X Wave Function of Different Kind of Free Particle : Wave Packet Sum of Plane Waves: $ ( x, 0) = +" & a (k )eikx dk #" $ ( x, t ) = +" & Combine many free waves to create a Localized wave packet (group) a (k )ei ( kx #! t ) dk #" Wave Packet initially localized in %X, %t undergoes dispersion The more you know now, The less you will know later Why ? Spreading is due to DISPERSION resulting from the fact that phase velocity of individual waves making up the packet depends on λ (k) Normalization Condition: Particle Must be Somewhere Example : ! ( x, 0) = Ce " x x0 , C & x 0 are constants This is a symmetric wavefunction with diminishing amplitude The Amplitude is maximum at x =0 # Probability is max too Normalization Condition: How to figure out C ? A real particle must be somewhere: Probability of finding particle is finite P(-$ + x + +$) = +$ , 2 ! ( x, 0) dx = -$ # 1 = 2C $ 2 ,e 0 "2 x x0 , -$ % x0 & dx = 2C ' ( = C 2 x0 )2* 2 # ! ( x, 0) = +$ 1 e x0 " x x0 C 2e "2 x x0 dx = 1 Where is the particle within a certain location x ± Δx Prob |Ψ(x,0)|2 Lets Freeze time (t=0) P(-x 0 # x # +x 0 ) = ? +x 0 + -x 0 x 2 ! ( x, 0) dx = +x 0 + C 2e "2 x x0 -x 0 $ x0 % "2 "2 $ % $ = 2C ) * '1 " e ( = '1 " e %( = 0.865 & 87% '2( 2 dx Where Do Wave Functions Come From ? • Are solutions of the time dependent Schrödinger Differential Equation (inspired by Wave Equation seen in 2C) ! 2 ! 2 " ( x, t ) !" ( x, t ) # + U ( x ) " ( x , t ) = i ! 2m !x 2 !t • Given a potential U(x) particle under certain force – F(x) = " !U ( x) !x Schrodinger had an interesting life Schrodinger Wave Equation Wavefunction " which is a sol. of the Sch. Equation embodies all modern physics experienced/learnt so far: E=hf, h p= , ! #x.#p ! " , #E.#t ! ", quantization etc Schrodinger Equation is a Dynamical Equation # # much like Newton's Equation F= m a $ " (x,0) $ Force (potential) $ " (x,t) Evolves the System as a function of space-time The Schrodinger Eq. propogates the system forward & backward in time: & d" ' " (x,% t) = " (x,0) ± ( %t ) * dt + t =0 Where does it come from ?? ..."First Principles"..no real derivation exists Time Independent Sch. Equation ! 2 " 2 #(x,t) "#(x,t) ! + U (x)#(x,t) = i! 2 2m "x "t Sometimes (depending on the character of the Potential U(x,t)) The Wave function is factorizable: can be broken up ( ) # x,t = $ (x) % (t) Example : Plane Wave #(x,t)=e i(kx-& t) = e i(kx)e-i(& t) In such cases, use separation of variables to get : -! 2 " 2$ (x) "% (t) % (t). 2 + U (x)$ (x)% (t) = i!$ (x) 2m "t "x Divide Throughout by #(x,t)=$ (x)% (t) -! 2 1 " 2$ (x) 1 "% (t) ' . 2 + U (x) = i! 2m $ (x) " x % (t) "t LHS is a function of x; RHS is fn of t x and t are independent variables, hence : ' RHS = LHS = Constant = E Factorization Condition For Wave Function Leads to: -! 2 # 2! ( x) + U ( x)! ( x) = E ! ( x) 2 2m # x #" (t ) i! = E" (t ) #t What is the Constant E ? How to Interpret it ? Back to a Free particle : # (x,t)= Aeikx e-i! t , " (x)= Aeikx U(x,t) = 0 Plug it into the Time Independent Schrodinger Equation (TISE) $ %! 2 d 2 ( Ae(ikx ) ) !2k 2 p 2 ( ikx ) + 0 = E Ae $E= = = (NR Energy) 2 2m dx 2m 2m Stationary states of the free particle: # (x,t)=" (x)e-i! t 2 $ # ( x, t ) = " ( x ) 2 Probability is static in time t, character of wave function depends on " ( x) A More Interesting Potential : Particle In a Box U(x) Write the Form of Potential: Infinite Wall U(x,t) = !; x " 0, x # L U(x,t) = 0 ; 0 < X < L • Classical Picture: •Particle dances back and forth •Constant speed, const KE •Average <P> = 0 •No restriction on energy value • E=K+U = K+0 •Particle can not exist outside box •Can’t get out because needs to borrow infinite energy to overcome potential of wall What happens when the joker is subatomic in size ?? Example of a Particle Inside a Box With Infinite Potential (a) Electron placed between 2 set of electrodes C & grids G experiences no force in the region between grids, which are held at Ground Potential However in the regions between each C & G is a repelling electric field whose strength depends on the magnitude of V (b) If V is small, then electron’s potential energy vs x has low sloping “walls” (c) If V is large, the “walls”become very high & steep becoming infinitely high for V→∞ (d) The straight infinite walls are an approximation of such a situation U=∞ U(x) U=∞ Ψ(x) for Particle Inside 1D Box with Infinite Potential Walls Inside the box, no force " U=0 or constant (same thing) " -! d ! ( x ) + 0 ! ( x) = E ! ( x) 2m dx 2 2 2 d 2! ( x) 2mE 2 2 " = # k ! ( x ) ; k = dx 2 !2 d 2! ( x) or + k 2! ( x) = 0 $ figure out what ! (x) solves this diff eq. 2 dx In General the solution is ! ( x) = A sinkx + B coskx (A,B are constants) Why can’t the particle exist Outside the box ? E Conservation ∞ ∞ Need to figure out values of A, B : How to do that ? Apply BOUNDARY Conditions on the Physical Wavefunction We said ! ( x) must be continuous everywhere So match the wavefunction just outside box to the wavefunction value just inside the box " At x = 0 " ! ( x = 0) = 0 & At x = L " ! ( x = L) = 0 % ! ( x = 0) = B = 0 (Continuity condition at x =0) & ! ( x = L) = 0 " A Sin kL = 0 (Continuity condition at x =L) " kL = n& " k = n& , n = 1, 2,3,...' L n 2& 2 ! 2 So what does this say about Energy E ? : E n = Quantized (not Continuous)! 2mL2 X=0 X=L Quantized Energy levels of Particle in a Box The Quantum Mechanics of Christina Aguilera! Christina at rest between two walls originally at infinity: Uncertainty in her location ΔX = ∞ . At rest means her momentum P=0 , ΔP=0 (Uncertainty principle) Slowly two walls move in from infinity on each side, now ΔX = L , so Δp ≠ 0 She is not at rest now, in fact her momentum P ≈ ± (h/2π L) L X Axis 0 Christina’s Momentum p On average, measure <p> = 0 but there are quite large fluctuations! Width of Distribution = !P ! !P = ( P 2 ) ave " ( Pave ) 2 ; !P " L Bottomline : Christina dances to the tune of Uncertainty Principle! What About the Wave Function Normalization ? The particle's Energy and Wavefunction are determined by a number n We will call n ! Quantum Number , just like in Bohr's Hydrogen atom What about the wave functions corresponding to each of these energy states? # n = A sin(kx) = A sin( n" x ) L for 0<x < L for x % 0, x % L =0 Normalized Condition : L L 0 0 1 = '# n *# n dx = A2 ' Sin 2 ( n" x ) L Use 2Sin 2$ = 1 & 2Cos 2$ A2 ( 2n" x ) 1= 1 & c os( ) + and since ' cos $ = sin$ * ' 2 0, L L A2 1= L 2 . A= So # n = 2 sin(kx) = L 2 L 2 n" x sin( ) L L ...What does this look like? Wave Functions : Shapes Depend on Quantum # n Wave Function Probability P(x): Where the particle likely to be Zero Prob Where in The World is Carmen San Diego? • We can only guess the probability of finding the particle somewhere in x – For n=1 (ground state) particle most likely at x = L/2 – For n=2 (first excited state) particle most likely at L/4, 3L/4 • Prob. Vanishes at x = L/2 & L – How does the particle get from just before x=L/2 to just after? » QUIT thinking this way, particles don’t have trajectories Classically, where is the particle most » Just probabilities of Likely to be : Equal prob of being being somewhere anywhere inside the Box NOT SO says Quantum Mechanics! How to Calculate the QM prob of Finding Particle in Some region in Space Consider n =1 state of the particle L 3L Ask : What is P ( # x # ) ? 4 4 3L 4 P= 1" L 4 2 1 2 dx = L 3L 4 !x $2% 1 1L sin L dx = ') L (* . 2 2 4 3L 4 2! x 1L (1 & cos L )dx 4 3L / 4 1 +L ,+ L 2! x , 1 1 P = - & . - sin = & . L / 2 0 / 2! L 0 L / 4 2 2! 2! 3L 2! L % $ . & sin . ( ' sin L 4 L 4* ) 1 1 P= & (&1 & 1) = 0.818 2 81.8% 2 2! Classically 2 50% (equal prob over half the box size) 2 Substantial difference between Classical & Quantum predictions When The Classical & Quantum Pictures Merge: n→∞ But one issue is irreconcilable: Quantum Mechanically the particle can not have E = 0 This is a consequence of the Uncertainty Principle The particle moves around with KE inversely proportional to the Length Of the 1D Box Finite Potential Barrier • There are no Infinite Potentials in the real world – Imagine the cost of as battery with infinite potential diff • Will cost infinite $ sum + not available at Radio Shack • Imagine a realistic potential : Large U compared to KE but not infinite Region I U Region II Region III E=KE X=0 X=L Classical Picture : A bound particle (no escape) in 0<x<L Quantum Mechanical Picture : Use ΔE.Δt ≤ h/2π Particle can leak out of the Box of finite potential P(|x|>L) ≠0 Finite Potential Well -! 2 d 2" ( x) + U" ( x) = E " ( x) 2 2m dx d 2" ( x) 2m $ = 2 (U # E )" ( x) 2 dx ! 2m(U-E) = ! 2 " ( x); ! = !2 $General Solutions : " ( x) = Ae +! x + Be #! x Require finiteness of " ( x) $ " ( x) = Ae +! x .....x<0 (region I) " ( x) = Ae #! x .....x>L (region III) Again, coefficients A & B come from matching conditions at the edge of the walls (x =0, L) But note that wave fn at " ( x) at (x =0, L) % 0 !! (why?) d" ( x) Further require Continuity of " ( x) and dx These lead to rather different wave functions Finite Potential Well: Particle can Burrow Outside Box Finite Potential Well: Particle can Burrow Outside Box Particle can be outside the box but only for a time Δt ≈ h/ ΔE ΔE = Energy particle needs to borrow to Get outside ΔE = U-E + KE The Cinderella act (of violating E Conservation cant last very long Particle must hurry back (cant be caught with its hand inside the cookie-jar) 1 ! Penetration Length ! = = " 2m(U-E) If U>>E # Tiny penetration If U $ % # ! $ 0 Finite Potential Well: Particle can Burrow Outside Box 1 ! Penetration Length ! = = " 2m(U-E) If U>>E # Tiny penetration If U $ % # ! $ 0 n! ! En = , n = 1, 2,3, 4... 2 2m( L + 2" ) When E=U then solutions blow up # Limits to number of bound states(E n < U ) 2 2 2 When E>U, particle is not bound and can get either reflected or transmitted across the potential "barrier" Simple Harmonic Oscillator: Quantum and Classical Spring with Force Const k m X=0 x Unstable a Stable U(x) b Stable Equilibrium: General Form : c Stable x 1 U(x) =U(a)+ k ( x ! a ) 2 2 1 Rescale " U ( x ) = k ( x ! a ) 2 2 Motion of a Classical Oscillator (ideal) Ball originally displaced from its equilibirium position, motion confined between x=0 & x=A 1 2 1 k 2 2 U(x)= kx = m # x ; # = = Ang . Freq Particle of mass m within a potential U(x) 2 2 m ! dU ( x ) 1 F(x)= E = kA2 " Changing A changes E dx 2 ! E can take any value & if A $ 0, E $ 0 dU ( x ) F(x=a) = = 0, Max. KE at x = 0, KE= 0 at x= ± A dx ! ! F(x=b) = 0 , F(x=c)=0 ...But... look at the Curvature: ! 2U ! 2U > 0 (stable), < 0 (unstable) 2 2 !x !x Quantum Picture: Harmonic Oscillator Find the Ground state Wave Function ! (x) 1 Find the Ground state Energy E when U(x)= m" 2 x 2 2 -! 2 # 2! ( x ) 1 Time Dependent Schrodinger Eqn: + m" 2 x 2! ( x ) = E ! ( x ) 2 2m # x 2 d 2! ( x ) 2m 1 $ = ( E % m" 2 x 2 )! ( x ) = 0 What ! (x) solves this? 2 2 dx ! 2 Two guesses about the simplest Wavefunction: 1. ! (x) should be symmetric about x 2. ! (x) & 0 as x & ' d! (x) + ! (x) should be continuous & = continuous dx 2 My guess: ! (x) = C0e %( x ; Need to find C0 & ( : What does this wavefunction & PDF look like? Quantum Picture: Harmonic Oscillator 2 P(x) = C 0e " (x) = C0e "2! x 2 C2 #! x 2 C0 x How to Get C0 & α ?? …Try plugging in the wave-function into the time-independent Schr. Eqn. Time Independent Sch. Eqn & The Harmonic Oscillator ! 2" (x) 2m 1 2 2 Master Equation is : = [ m # x $ E]" (x) 2 2 !x ! 2 d" (x) $% x2 $% x2 Since " (x) = C0 e , = C0 ($2% x)e , dx d 2" (x) d($2% x) $% x 2 2 $% x2 2 2 $% x2 = C0 e + C0 ($2% x) e = C0 [4% x $ 2% ]e 2 dx dx & C0 [ 4% x $ 2% ]e 2 2 $% x2 2m 1 2 2 $% x2 = 2 [ m# x $ E]C0 e ! 2 Match the coeff of x 2 and the Constant terms on LHS & RHS 2m 1 m# 2 2 & 4% = 2 m# or % = 2! ! 2 2m & the other match gives 2% = 2 E, substituing % & ! 1 E= !# 2 !!!!......(Planck's Oscillators) What about C0 ? We learn about that from the Normalization cond. SHO: Normalization Condition +# +# " m% x 2 2 ! 0 2 | ! (x) | dx = 1 = $ C e $ 0 "# "# +# Since dx $e " ax 2 dx = "# Identifying a= & a (dont memorize this) m% and using the identity above ! 1 4 ( m% + ' C0 = * & ! ) , Hence the Complete NORMALIZED wave function is : 1 4 ( m% + ! 0 (x) = * - e ) &! , " m% x 2 2! Ground State Wavefunction has energy E = h% /2 Planck's Oscillators were electrons tied by the "spring" of the mutually attractive Coulomb Force Quantum Oscillator In Pictures E = KE + U ( x) > 0 for n=0 Quantum Mechanical Prob for particle To live outside classical turning points Is finite ! U(x) C0 U -A +A -A +A Classically particle most likely to be at the turning point (velocity=0) Quantum Mechanically , particle most likely to be at x=x0 for n=0 Classical & Quantum Pictures of SHO compared • Limits of classical vibration : Turning Points (do on Board) • Quantum Probability for particle outside classical turning points P(|x|>A) =16% !! – Do it on the board (see Example problems in book) Excited States of The Quantum Oscillator m" x 2 2! m" x) e ; ! H n ( y) = Hermite Polynomials # ! n (x) = Cn H n ( with H 0 (y)=1 H1 (y)=2y H 2 (y)=4y2 # 2 H 3 (y)=8y3 # 12 y H n (y)=(-1) n e y 2 n # y2 d e dy n and 1 En = (n + )!" 2 Again n=0,1,2,3...$ Quantum # Excited States of The Quantum Oscillator Ground State Energy >0 always Measurement Expectation: Statistics Lesson • Ensemble & probable outcome of a single measurement or the average outcome of a large # of measurements " n < x >= n1 x1 + n2 x2 + n3 x3 + ....ni xi = n1 + n2 + n3 + ...ni $n x i i i =1 N = % xP( x )dx #" " % P( x )dx #" For a general Fn f(x) " n < f ( x ) >= $ ni f ( xi ) i =1 N = * ! % ( x ) f ( x )! ( x )dx #" Sharpness of A Distr: Scatter around average " % P( x )dx #" != != 2 (x " x ) % i N ( x 2 ) " ( x )2 ! = small # Sharp distr. Uncertainty $X = ! Particle in the Box, n=1, <x> & Δx ? " (x)= 2 %! & sin ' x ( L )L * $ 2 %! & sin ' x ( x L )L * <x>= 0 -$ 2 %! & sin ' x ( dx L )L * 2 %! & %! & = 0 x sin 2 ' x (dx , change variable # = ' x ( L0 )L * )L * L ! 2 1 2 2 + <x>= # sin # , use sin # = (1 - cos 2# ) 2 0 L! 0 2 ! ! 1 2L , + <x>= 2 . 0 # d# - 0 # cos2# d# 2 use 0 udv=uv-0 vdu 2! / 0 0 3 L %!2 & L + <x>= 2 ' ( = (same result as from graphing " 2 ( x)) ! ) 2 * 2 ! L2 L2 Similarly <x >= 0 x sin ( x)dx = - 2 L 3 2! 0 L 2 2 2 L2 L2 L2 and 4X= <x > - < x > = - 2= 0.18 L 3 2! 4 4X= 20% of L, Particle not sharply confined in Box 2 2 Expectation Values & Operators: More Formally • Observable: Any particle property that can be measured – X,P, KE, E or some combination of them,e,g: x2 – How to calculate the probable value of these quantities for a QM state ? • Operator: Associates an operator with each observable – Using these Operators, one calculates the average value of that Observable – The Operator acts on the Wavefunction (Operand) & extracts info about the Observable in a straightforward way gets Expectation value for that observable < Q >= +! $ #* ( x, t ) [Qˆ ] #* ( x, t )dx "! Q is the observable, [Qˆ ] is the operator & < Q > is the Expectation value Examples: [X] = x , [P]2 -! 2 % 2 [K] = = 2m 2m %x 2 ! d i dx % [E] = i! %t [P] = Operators Information Extractors ! d Momentum Operator i dx gives the value of average mometum in the following way: [p] or p̂ = +" <p> = # ! (x)[ p]! (x)dx = * Plug & play form -" +" $ ! ' d! * ! (x) # &% i )( dx dx -" Similarly : !2 d 2 [K] or K̂ = gives the value of average KE 2m dx 2 +" +" $ ! 2 d 2! (x) ' * * <K> = # ! (x)[K]! (x)dx = # ! (x) & * dx ) 2 % 2m dx ( -" -" Similarly +" <U> = # ! * (x)[U (x)]! (x)dx : plug in the U(x) fn for that case -" $ ! 2 d 2! (x) ' and <E> = # ! (x)[K + U (x)]! (x)dx = # ! (x) & * + U (x)) dx 2 2m dx % ( -" -" +" +" * * Hamiltonian Operator [H] = [K] +[U] The Energy Operator [E] = i! + informs you of the average energy +t [H] & [E] Operators • [H] is a function of x • [E] is a function of t …….they are really different operators • But they produce identical results when applied to any solution of the time-dependent Schrodinger Eq. • [H]Ψ(x,t) = [E] Ψ(x,t) ! !2 # 2 " ! #" & $ 2m #x 2 + U ( x, t ) ' % ( x, t ) = &i! #t ' % ( x, t ) ( ) ( ) • Think of S. Eq as an expression for Energy conservation for a Quantum system Example : Average Momentum of particle in box ! n (x) = 2 n" sin( x) L L +( & ! *n (x) = 2 n" sin( x) L L ( * #! d % # % ! p ! dx = ! ) $ & ) *$ i dx +&! dx '( '( L ! 2 n" n" n" < p >= ( x)cos( x)dx i L L )0 L L < p >= * Since ) sinax cosax dx = !# n" ,< p >= *sin 2 ( iL $ L 1 n" sin 2 ax ...here a = 2a L x= L % x + = 0 since Sin 2 (0) = Sin 2 (n" ) = 0 & x =0 We knew THAT before doing any math ! Quiz 1: What is the <p> for the Quantum Oscillator in its symmetric ground state Quiz 2: What is the <p> for the Quantum Oscillator in its asymmetric first excited state But what about the <KE> of the Particle in Box ? p2 < p >= 0 so what about expectation value of K= ? 2m < K >= 0 because < p >= 0; clearly not, since we showed E=KE ! 0 Why ? What gives ? n" ! Because p n = ± 2mEn = ± ; " ± " is the key! L AVERAGE p =0 , particle is moving back & forth p2 <p > 2 <KE> = < > ! 0 not 2m 2m ! Be careful when being "lazy" Quiz: what about <KE> of a quantum Oscillator? Does similar logic apply?? Schrodinger Eqn: Stationary State Form d 1 df (t ) Since [ln f (t ) ] = dt f (t ) dt "! (t ) 1 "! (t ) E iE In i! = E! (t ) , rewrite as = =# "t ! (t ) "t i! ! and integrate both sides w.r.t. time t =t 1 "! (t ) iE 1 d! (t ) iE &t=0 ! (t ) "t dt = &0 # ! dt % &0 ! (t ) dt dt = # ! t t iE $ ln ! (t ) # ln ! (0) = # t , now exponentiate both sides ! % ! (t ) = ! (0)e % ! (t ) = e # iE t ! # iE t ! ; ! (0) = constant= initial condition = 1 (e.g) & Thus ( (x,t)=' (x)e # iE t ! where E = energy of system Schrodinger Eqn: Stationary State Form + iE t ! " iE t ! iE iE t" t ! ! P ( x, t ) = # * # = ! * ( x ) e ! ( x ) e = ! * ( x)! ( x)e =| ! ( x) |2 In such cases, P(x,t) is INDEPENDENT of time. These are called "stationary" states because Prob is independent of time Examples : Particle in a box (why?) : Quantum Oscillator (why?) Total energy of the system depends on the spatial orientation of the system : charteristic of the potential situation !