Transcript Document

Time-Dependent Density-Functional Theory
Carsten A. Ullrich
University of Missouri-Columbia
Neepa T. Maitra
Hunter College & Graduate Center
CUNY
APS March Meeting 2010, Portland OR
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
1. Survey
Time-dependent Schrödinger equation



i  (r1 ,...,rN , t )  Tˆ  Vˆ (t )  Wˆ  (r1 ,...,rN , t )
t
kinetic energy
operator:
N
Tˆ   

j 1
2
2
j
2m
electron
interaction:
2
N
1
e
Wˆ  
2 j ,k r j  rk
j k
The TDSE describes the time evolution of a many-body state
starting from an initial state
 t ,
t0 , under the influence of an
Vˆ t   V r j , t .
N
external time-dependent potential
j 1
From now on, we’ll (mostly) use atomic units (e = m = h = 1).
1. Survey
Real-time electron dynamics: first scenario
Start from nonequilibrium initial state, evolve in static potential:
t=0
Charge-density oscillations in metallic
clusters or nanoparticles (plasmonics)
New J. Chem. 30, 1121 (2006)
Nature Mat. Vol. 2 No. 4 (2003)
t>0
1. Survey
Real-time electron dynamics: second scenario
Start from ground state, evolve in time-dependent driving field:
t=0
Nonlinear response and ionization of atoms
and molecules in strong laser fields
t>0
1. Survey
Coupled electron-nuclear dynamics
● Dissociation of molecules (laser or collision induced)
● Coulomb explosion of clusters
● Chemical reactions
High-energy proton hitting ethene
T. Burnus, M.A.L. Marques, E.K.U. Gross,
Phys. Rev. A 71, 010501(R) (2005)
Nuclear dynamics
treated classically
For a quantum treatment of nuclear dynamics within TDDFT (beyond the
scope of this tutorial), see O. Butriy et al., Phys. Rev. A 76, 052514 (2007).
Linear response
1. Survey
tickle the system
observe how the
system responds
at a later time
(r, t )
(r, t )
n1 (r, t )   dr dt   r, t , r, t V1 r, t 
density
response
density-density
response function
perturbation
Optical spectroscopy
1. Survey
● Uses weak CW laser as Probe
Photoabsorption cross section
● System Response has peaks at
electronic excitation energies
Na2
Green
fluorescent
protein
Na4
Theory
Energy (eV)
Vasiliev et al., PRB 65, 115416 (2002)
Marques et al., PRL 90, 258101 (2003)
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
Runge-Gross Theorem
2. Fundamentals
kinetic
external potential
For any system with Hamiltonian of form H = T + W + Vext ,
e-e interaction
Runge & Gross (1984) proved the 1-1 mapping for fixed T and W:
n(r t)
0
vext(r t)
For a given initial-state y0, the time-evolving one-body density n(r t) tells you
everything about the time-evolving interacting electronic system, exactly.

This follows from :
0, n(r,t)  unique vext(r,t)  H(t)  (t)  all observables
2. Fundamentals
Proof of the Runge-Gross Theorem (1/4)
Consider two systems of N interacting electrons, both starting in the same 0 ,
but evolving under different potentials vext(r,t) and vext’(r,t) respectively:
vext(t), (t)
o
vext’(t), ’(t)
Assume timeanalytic potentials:
RG prove that the resulting densities n(r,t) and n’(r,t) eventually must differ,
i.e.
same
2. Fundamentals
Proof of the Runge-Gross Theorem (2/4)
The first part of the proof shows that the current-densities must differ.
Consider Heisenberg e.o.m for the current-density in each system,
the part of H that
differs in the two
systems
;t )
At the initial time:
initial density
 if initially the 2 potentials differ, then j and j’ differ infinitesimally later ☺
2. Fundamentals
Proof of the Runge-Gross Theorem (3/4)
If vext(r,0) = v’ext(r,0), then look at later times by repeatedly using Heisenberg e.o.m :
…
*
As vext(r,t) – v’ext(r,t) = c(t), and assuming potentials are time-analytic at t=0,
there must be some k for which RHS = 0 
 proves j(r,t)
1-1
o
1st part of RG ☺
vext(r,t)
The second part of RG proves 1-1 between densities and potentials:
Take divergence of both sides of * and use the eqn of continuity,
…
2. Fundamentals
Proof of the Runge-Gross Theorem (4/4)
…
≡ u(r) is nonzero for some k, but
must taking the div here be nonzero?
Yes!
By reductio ad absurdum: assume
assume fall-off of n0 rapid enough
that surface-integral  0
Then
integrand 0, so if integral 0, then u  0
i.e.
 contradiction
same
1-1 mapping between time-dependent densities and potentials, for a
given initial state
2. Fundamentals
The TDKS system
n v for given 0, implies any observable is a functional of n and 0
-- So map interacting system to a non-interacting (Kohn-Sham) one, that
reproduces the same n(r,t).
All properties of the true system can be extracted from TDKS  “bigger-fastercheaper” calculations of spectra and dynamics
KS “electrons” evolve in the 1-body KS potential:
functional of the history of the density
and the initial states
-- memory-dependence (see more shortly!)
If begin in ground-state, then no initial-state dependence, since by HK,
0 = 0[n(0)] (eg. in linear response). Then
2. Fundamentals
Clarifications and Extensions
But how do we know a non-interacting system exists that reproduces a given
interacting evolution n(r,t) ?
 van Leeuwen (PRL, 1999)
for time-analytic potentials and densities
(& under mild restrictions of the choice of the KS initial state F0)
The KS potential is not the density-functional derivative of any action !
If it were, causality would be violated:
Vxc[n,0,F0](r,t) must be causal – i.e. cannot depend on n(r t’>t)
But if
then
But RHS must be symmetric in (t,t’)  symmetry-causality paradox.
van Leeuwen (PRL 1998): an action, and variational principle, may be defined,
using Keldysh contours in complex-time.
 Vignale (PRA 2008): usual real-time action is just fine IF include boundary terms
2. Fundamentals
Clarifications and Extensions
Restriction to time-analytic potentials means RG is technically not valid for
many potentials, eg adiabatic turn-on, although RG is assumed in practise.
van Leeuwen (Int. J. Mod. Phys. B. 2001) extended the RG proof in the linear
response regime to the wider class of Laplace-transformable potentials.
The first step of the RG proof showed a 1-1 mapping between currents and
potentials  TD current-density FT
In principle, must use TDCDFT (not TDDFT) for
-- response of periodic systems (solids) in uniform E-fields (see later…)
-- in presence of external magnetic fields (Ghosh & Dhara, PRA 1988)
In practice, approximate functionals of current are simpler where spatial nonlocal dependence is important
(Vignale & Kohn, 1996; Vignale, Ullrich & Conti 1997) … Stay tuned!
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
Time-dependent Kohn-Sham scheme (1)
3. TDKS
Consider an N-electron system, starting from a stationary state.
Solve a set of static KS equations to get a set of N ground-state orbitals:
 2
 (0)
 
 Vext r,t0   VH r   Vxc r  j r    j (j 0) r 
 2

The N static KS orbitals are taken as initial orbitals and will be propagated in time:

( 0)
j
r  j r, t0 ,
j  1,...,N
 2


i  j r, t    
 Vext r, t   VH r, t   Vxc r, t  j r, t 
t
 2

N
Time-dependent density:
nr, t     j r, t 
j 1
2
3. TDKS
Time-dependent Kohn-Sham scheme (2)
Only the N initially occupied orbitals are propagated. How can this be sufficient
to describe all possible excitation processes?? Here’s a simple argument:
Expand TDKS orbitals in complete basis of static KS orbitals,

 j r, t    a jk t 
r 
finite for
kN
( 0)
k
k 1
A time-dependent potential causes the TDKS orbitals to acquire admixtures of
initially unoccupied orbitals.
Adiabatic approximation
3. TDKS
nr, t 
VH r, t    d r 
r - r
depends on density at time t
3
Vxc nr, t 
(instantaneous, no memory)
is a functional of
nr, t , t   t
The time-dependent xc potential has a memory!
Adiabatic approximation:
adia
xc
V
nr, t   V n(t)r
gs
xc
(Take xc functional from static DFT and evaluate with time-dependent density)
ALDA:
2 hom
d
exc (n )
ALDA
LDA
Vxc (r, t )  Vxc nr, t  
dn 2
n  n ( r ,t )
Time-dependent selfconsistency (1)
3. TDKS
start with
selfconsistent
KS ground state
propagate
until here
t0
I. Propagate
T


old
2

i j   2  VKS t   j , t  t0 , T 
II. With the density
nt     j t 
2
calculate the new KS potential
j
new
KS
V
time
t   Vext t   VH nt   Vxc nt 
III. Selfconsistency is reached if
for all
t  t0 , T 
old
t   VKSnew t , t t0 , T 
VKS
Numerical time Propagation
3. TDKS
Propagate a time step
 t :  j r, t  t   e
Crank-Nicholson algorithm:
1
Problem:
i
2

e
iHˆ t
 iHˆ t
 j r, t 
1  iHˆ t 2

1  iHˆ t 2


tHˆ  j r, t  t   1  2i tHˆ  j r, t 
Hˆ
must be evaluated at the mid point
But we know the density only for times
t  t 2
t
Time-dependent selfconsistency (2)
3. TDKS
Predictor Step:
 (j1) t  t   Hˆ (1) t  t 
 j t 
nth Corrector Step:
 j t 
1
2
Hˆ t  t 2  
Hˆ t   Hˆ ( n ) t  t 

Selfconsistency is reached if
 (j n1) t  t   Hˆ ( n1) t  t 

nt 
remains unchanged for
t  t0 , T 
upon addition of another corrector step in the time propagation.
3. TDKS
Summary of TDKS scheme: 3 Steps
r,0
1
Prepare the initial state, usually the ground state, by
a static DFT calculation. This gives the initial orbitals:
2
Solve TDKS equations selfconsistently, using an approximate
time-dependent xc potential which matches the static one used
in step 1. This gives the TDKS orbitals:
r, t  n r, t
j 
3


Calculate the relevant observable(s) as a functional of
( 0)
j
 
nr, t 
3. TDKS
Example: two electrons on a 2D quantum strip
hard walls
periodic
boundaries
(travelling
waves)
initial-state density
exact
LDA
z
x
(standing waves)
Charge-density oscillations
Δ
● Initial state: constant electric field,
which is suddenly switched off
● After switch-off, free propagation of
the charge-density oscillations
L
C.A. Ullrich, J. Chem. Phys. 125, 234108 (2006)
Construction of the exact xc potential
3. TDKS
Step 1: solve full 2-electron Schrödinger equation
 12  22
1


 V z1 , t   V z2 , t  
 i   r1 , r2 , t   0

2
r1  r2
t 
 2
Step 2: calculate the exact time-dependent density
d
3
r2  r, r2 , t   n z , t   2   z ,t 
2
2
s1 , s 2
Step 3: find that TDKS system which reproduces the density
 1 d2

 V z ,t   VH z ,t   Vxc z ,t   i  z ,t   0

2
t 
 2 dz
3. TDKS
Construction of the exact xc potential
Ansatz:
nr, t 
 r, t  
expi r, t 
2
A
xc
V
Vxc r, t    V r, t   VH r, t 
1 2
1
2
  ln nr, t    ln nr, t 
4
8
1
2
  r, t    r, t 
2
dyn
xc
V
3. TDKS
2D quantum strip: charge-density oscillations
density
adiabatic Vxc
exact Vxc
● The TD xc potential can be constructed from a TD density
● Adiabatic approximations get most of the qualitative behavior right,
but there are clear indications of nonadiabatic (memory) effects
● Nonadiabatic xc effects can become important (see later)
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
4. Memory
Memory dependence
functional dependence on history, n(r
and on initial states
t’<t),
Maitra, Burke, Woodward (PRL 2002): Exact condition relating
initial-state dependence and history-dependence.
Almost all calculations ignore memory, and use an “adiabatic approximation” :
Just take xc functional from static DFT and evaluate on instantaneous density
vxc
But what about the exact functional?
4. Memory
Example of history dependence
Eg. Time-dependent Hooke’s atom –exactly solvable
2 electrons in parabolic well,
time-varying force constant
parametrizes
density
k(t) =0.25 – 0.1*cos(0.75 t)
Any adiabatic (or even
semi-local-in-time)
approximation would
incorrectly predict the
same vc at both times.
Hessler, Maitra, Burke, (J. Chem. Phys, 2002); see also other examples in the Literature handout
• Development of History-Dependent Functionals: Dobson, Bunner & Gross (1997),
Vignale, Ullrich, & Conti (1997), Kurzweil & Baer (2004), Tokatly (2005,2007)
4. Memory
RG:
Initial-state dependence
n(r t)
0
1-1
vext(r t)
But is there ISD? That is, if we start in different 0’s, can we get the same n(r t),
for all t, by evolving in different potentials? i.e.
Evolve 0 in v(t) 
n (r t)
t
? Evolve 0 in v (t)  same n ?
The answer is:
No! for one electron,
but,
Yes! for 2 or more
electrons
If no, then ISD redundant, i.e. the functional dependence on the density is enough.
4. Memory
Example of initial-state dependence
A non-interacting example:
Periodically driven HO
If we start in different 0’s, can
we get the same n(r t) by
evolving in different potentials?
Yes!
Re and Im parts
of 1st and 2nd
Floquet orbitals
Doubly-occupied
Floquet orbital
with same n
• Say this is the density of an interacting
system. Both top and middle are possible
KS systems.
 vxc different for each. Cannot be captured
by any adiabatic approximation
( Consequence for Floquet DFT: No 1-1 mapping between densities and timeperiodic potentials. )
Maitra & Burke, (PRA 2001)(2001, E); Chem. Phys. Lett. (2002).
Time-dependent optimized effective potential
4. Memory

t
N
0  i   dt  d 3r  Vxc (r, t )  u xcj (r, t )

j 1  

  k (r, t )k* (r, t ) j (r, t ) *j (r, t )  c.c.
k 1
where
exact exchange:
Axc i 
1
u xcj (r, t )  *
 j (r, t )  j (r, t )
u xj r, t   
N
1
3
d
r

*

 j r, t  k 1
 *j r, t k r, t k* r, t 
r  r
C.A.Ullrich, U.J. Gossmann, E.K.U. Gross, PRL 74, 872 (1995)
H.O. Wijewardane and C.A. Ullrich, PRL 100, 056404 (2008)
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
TDDFT in linear response
5. Linear Response
Poles at true
excitations
Poles at KS
excitations
adiabatic approx: no w-dep
Need (1) ground-state vS,0[n0](r), and its bare excitations
(2) XC kernel
Yields exact spectra in principle; in
practice, approxs needed in (1) and (2).
Petersilka, Gossmann, Gross, (PRL, 1996)
5. Linear Response
Matrix equations (a.k.a. Casida’s equations)
Quantum chemistry codes cast eqns into a matrix of coupled KS single
excitations (Casida 1996) : Diagonalize
q = (i  a)
 Excitation energies and oscillator strengths
Useful tools for analysis: “single-pole” and “small-matrix” approximations (SPA,SMA)
Zoom in on a single KS excitation, q = i a
Well-separated single excitations: SMA
When shift from bare KS small: SPA
5. Linear Response
How it works: atomic excitation energies
TDDFT linear response from
exact helium KS ground state:
LDA + ALDA lowest excitations
Exp. full matrix
SMA
SPA
Vasiliev, Ogut, Chelikowsky, PRL 82, 1919 (1999)
Compare different functional
approxs (ALDA, EXX), and
also with SPA. All quite
similar for He.
From Burke & Gross, (1998); Burke, Petersilka & Gross (2000)
Atomic excitations: Rydberg states
5. Linear Response
Generally, KS excitations themselves are good zero-order approximations to the
exact energies – except
when they are missing !
LDA/GGA KS potentials asymptotically decay exponentially (ground-state lectures)
 No -1/r tail  no Rydberg excitations.
Either paste a tail on (eg LB94, or some kind of hybrid…)
OR, use a clever trick to obtain their energies:
Quantum defect theory:
determined by
short-range
part of v
A. Wasserman & K. Burke, Phys. Rev. Lett. (2005);
5. Linear Response
A comparison of functionals
Study of various functionals over
a set of ~ 500 organic
compounds, 700 excited singlet
states
From: D. Jacquemin, V. Wathelet, E. A. Perpete, C. Adamo, J. Chem. Theory Comput. (2009).
5. Linear response
General trends
Energies typically to within about “0.4 eV”
Bonds to within about 1%
Dipoles good to about 5%
Vibrational frequencies good to 5%
Cost scales as N3, vs N5 for wavefunction methods of
comparable accuracy (eg CCSD, CASSCF)
Available now in many electronic structure codes
Unprecedented balance between accuracy and efficiency
TDDFT Sales Tag
5. Linear response
Examples
Can study big molecules with TDDFT !
-- Can study candidates for solar cells, eg. carotenoid-diaryl-porphyrin-C60
(632 valence
electrons! )
Photo-excitation of a light-harvesting supra-molecular triad: a TDDFT study,
N. Spallanzani, C. A. Rozzi, D. Varsano, T. Baruah, M. R. Pederson, F. Manghi, and
A. Rubio, J. Phys. Chem. (2009)
5. Linear response
Examples
Circular dichroism spectra of chiral fullerenes: D2C84
F. Furche and R. Ahlrichs, JACS 124, 3804 (2002).
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
Excitations in finite and extended systems
6. TDDFT in solids


0 nˆ r   j  j nˆ r 0
 r, r, w   lim 
 c.c.w  w 
 0
w  E j  E0  i
 j

j
The full many-body response function has poles at the exact excitation energies
Im w
Im w
finite
x
xx
x
x
extended
Re w
► Discrete single-particle excitations merge into a continuum
(branch cut in frequency plane)
► New types of collective excitations appear off the real axis
(finite lifetimes)
Re w
6. TDDFT in solids
Metals vs. insulators
plasmon
Excitation spectrum of simple metals:
● single particle-hole continuum
(incoherent)
● collective plasmon mode
Optical excitations
of insulators:
● interband transitions
● excitons (bound
electron-hole pairs)
6. TDDFT in solids
Excitations in bulk metals
Plasmon dispersion of Al
Quong and Eguiluz, PRL 70, 3955 (1993)
►RPA (i.e., Hartree) gives already
reasonably good agreement
►ALDA agrees very well with exp.
In general, (optical) excitation processes in (simple) metals are very well
described by TDDFT within ALDA.
Time-dependent Hartree already gives the dominant contribution, and
fxc typically gives some (minor) corrections.
This is also the case for 2DEGs in doped semiconductor heterostructures
(quantum wells, quantum dots).
6. TDDFT in solids
Elementary view of excitons
Excitons are bound electron-hole pairs created in optical excitations
of insulators.
Mott-Wannier exciton:
weakly bound, delocalized
over many lattice constants
Frenkel exciton:
tightly bound, localized on
a single (or a few) atoms
6. TDDFT in solids
Wannier equation and excitonic Rydberg series
 
e 
 
  (r )  E (r )
 2 mr  r 
2
2
r
2
Cu2O
R.J. Uihlein, D. Frohlich, and R. Kenklies,
PRB 23, 2731 (1981)
●  (r ) is exciton wave function
● derived from TDHF linearized
Semiconductor Bloch equation
● includes dielectric screening
GaAs
R.G. Ulbrich, Adv. Solid State Phys. 25,
299 (1985)
Simplified calculation of exciton binding energies
6. TDDFT in solids
from linearized TDDFT semiconductor Bloch equations (Tamm-Dancoff approx.):
 w
  Fkq (w) (w)  w  (w)
cv
q kq
cv
q
cv
k
q
2
Fkq (w )  2  d 3r  d 3r c*k (r)vk (r) f xc (r, r, w )v*q (r)cq (r)
 

● Finite atomic/molecular system:
single-pole approximation
involves two discrete levels
● “Single-pole approximation” for
excitons involves two entire bands
● Excitons are a collective phenomenon!
● TDDFT Wannier equation: nonlocal
e-h interaction (in real space)
V. Turkowski, A. Leonardo, C.A.Ullrich,
PRB 79, 233201 (2009)
6. TDDFT in solids
Optical absorption of insulators
Silicon
RPA and ALDA both bad!
►absorption edge red shifted
(electron self-interaction)
►first excitonic peak missing
(electron-hole interaction)
Why does the ALDA fail??
G. Onida, L. Reining, A. Rubio, RMP 74, 601 (2002)
S. Botti, A. Schindlmayr, R. Del Sole, L. Reining, Rep. Prog. Phys. 70, 357 (2007)
6. TDDFT in solids
Optical absorption of insulators: failure of ALDA
Optical absorption requires imaginary part of macroscopic dielectric function:
Im mac    limVG q  Im GG 
q 0
where
q  0 limit:
VG , G  0
   KS   KS V  f xc  , VG  
 0, G  0
~ q2
Long-range excluded,
so RPA is ineffective
2
Needs 1 q
component to
correct
KS

But ALDA is constant
for q  0 :
f xcALDA  lim f xchom q, w  0
q 0
6. TDDFT in solids
Long-range XC kernels for solids
● LRC (long-range correlation) kernel
(with fitting parameter α):
● TDOEP kernel (X-only):
f
OEP
x
f xcLRC q   
r, r  

q2
 f  r  r
k k
2
*
k
k
2 r  r nr nr
Simple real-space form: Petersilka, Gossmann, Gross, PRL 76, 1212 (1996)
TDOEP for extended systems: Kim and Görling, PRL 89, 096402 (2002)
● “Nanoquanta” kernel (L. Reining et al, PRL 88, 066404 (2002)
f xc 1,2  P01 1,3 G3,4 G5,3W 4,5G4,6 G6,5 P01 (6,2)
independent
quasiparticle
polarizability
screened
Coulomb
interaction
quasiparticle
Green’s function
6. TDDFT in solids
Optical absorption of insulators, again
Kim & Görling
Silicon
Reining et al.
F. Sottile et al., PRB 76, 161103 (2007)
6. TDDFT in solids
Optical absorption of molecular chains
Peierls-distorted H-chain
has optical gap and localized
excitons. ALDA fails.
Undistorted H-chain: no gap,
delocalized exciton.
ALDA works well
long-range f xc
works well in
all cases.
D. Varsano, A. Marini, and A. Rubio, PRL 101, 133002 (2008)
6. TDDFT in solids
Extended systems - summary
► TDDFT works well for metallic and quasi-metallic systems already
at the level of the ALDA. Successful applications for plasmon modes
in bulk metals and low-dimensional semiconductor heterostructures.
► TDDFT for insulators is a much more complicated story:
● ALDA works well for EELS (electron energy loss spectra), but
not for optical absorption spectra
● Excitonic binding due to attractive electron-hole interactions,
which require long-range contribution to fxc
● some long-range XC kernels have become available,
but the best ones are quite complicated.
● At present, the full (but expensive) Bethe-Salpeter equation gives
most accurate optical spectra in inorganic and organic materials
(extended or nanoscale), but TDDFT is catching up.
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
7. Where the usual approxs. fail
Ailments – and some Cures (I)
meaning, semi-local in space
and local in time
Local/semilocal approx inadequate.
Need Im fxc to open gap and 1/q2
Rydberg states
Polarizabilities of long-chain molecules
Optical response/gap of solids
Molecular Dissociation
Cure with orbital- dependent fnals
(exact-exchange/sic), or Nanoquanta
kernel or TD current-DFT
Haunted by static correlation in the
ground-state.
Long-range charge transfer
Adiabatic approx for fxc fails.
Conical intersections
Double excitations
Frequency-dependent kernel
derived for some of these cases
7. Where the usual approxs. fail
Ailments – and some Cures (II)
or, are questionable…
• Some strong-field dynamics calculations
Adiabatic approx fails -memory-dependence crucial
• Certain electronic quantum control problems
TD Static
correlation
• Momentum distributions (eg in ion-recoil experiments)
• Non-sequential double ionization
Cannot extract
observable
simply from KS
system
• Coupled correlated electron-ion dynamics
• Electronic transport
-- Need essential derivative discontinuity lacking in
approx
7. Where the usual approxs. fail
Double Excitations
Types of Excitations
Non-interacting systems eg. 4-electron atom
Eg. single excitations
Eg. double excitations
near-degenerate
Interacting systems: generally involve mixtures of (KS) SSD’s that may have
1,2,3…electrons in excited orbitals.
single-, double-, triple- excitations
7. Where the usual approxs. fail
Double Excitations
How do these different types of excitations appear in the TDDFT response
functions?
Now consider:
 – poles at true states that are mixtures of singles, doubles, and higher excitations
S -- poles only at single KS excitations, since one-body operator
can’t
connect Slater determinants differing by more than one orbital.
 has more poles than s
? How does fxc generate more poles to get states of multiple excitation character?
7. Where the usual approxs. fail
Double Excitations
Exactly Solve a Simple Model: one KS single (q) mixing with a nearby double (D)
Invert and insert into Dyson-like eqn for kernel dressed SPA (i.e. w-dependent):
Strong non-adiabaticity!
7. Where the usual approxs. fail
Double Excitations
General case: Diagonalize many-body H in KS subspace near the double ex of
interest, and require reduction to adiabatic TDDFT in the limit of weak coupling of
the single to the double 
usual adiabatic matrix element
So: (i) scan KS orbital energies to see
if a double lies near a single,
dynamical (non-adiabatic)
correction
(ii) apply this kernel just to that pair
(iii) apply usual ATDDFT to all other
excitations
Maitra, Zhang, Cave,& Burke JCP (2004);
Alternate derivations: Casida JCP (2005); Romaniello et al (JCP 2009); Gritsenko & Baerends
PCCP (2009)
7. Where the usual approxs. fail
Double Excitations
Example: Short-chain polyenes
Lowest-lying excitations notoriously difficult to calculate due to significant
double-excitation character.
Cave, Zhang, Maitra, Burke, CPL (2004)
• Note importance of accurate double-excitation description in coupled electron-ion
dynamics – propensity for curve-crossing
Levine, Ko, Quenneville, Martinez, Mol. Phys. (2006)
7. Where the usual approxs. fail
Long-Range Charge-Transfer Excitations
Example: Dual Fluorescence in DMABN in Polar Solvents
Rappoport & Furche,
JACS 126, 1277 (2004).
“normal”
“anomalous”
“Local” Excitation (LE)
Intramolecular Charge Transfer (ICT)
TDDFT resolved the long debate on ICT structure (neither “PICT” nor “TICT”),
and elucidated the mechanism of LE -- ICT reaction
Success in predicting ICT structure – How about CT energies ??
7. Where the usual approxs. fail
Long-Range Charge-Transfer Excitations
TDDFT typically severely underestimates long-range CT energies
Eg. Zincbacteriochlorin-Bacteriochlorin
complex
(light-harvesting in plants and purple
bacteria)
Dreuw & Head-Gordon, JACS 126 4007, (2004).
TDDFT predicts CT states energetically well below local fluorescing states.
Predicts CT quenching of the fluorescence.
! Not observed !
TDDFT error ~ 1.4eV
7. Where the usual approxs. fail
Long-Range Charge-Transfer Excitations
Why do the usual approximations in TDDFT fail for these excitations?
We know what the exact energy for charge transfer at long range should be:
exact
Why TDDFT typically severely underestimates this energy can be seen in SPA
-As,2
-I1
~0 overlap
i.e. get just the bare KS orbital energy difference: missing xc contribution to
acceptor’s electron affinity, Axc,2, and -1/R
(Also, usual g.s. approxs underestimate I)
Dreuw, Weisman, Head-Gordon, JCP (2003)
Tozer, JCP (2003)
7. Where the usual approxs. fail
Long-Range Charge-Transfer Excitations
Many approaches to try to fix TDDFT for CT:
Eg. Dreuw, Weisman, & Head-Gordon, JCP (2003) – use CIS curve but shifted
vertically to match SCF-DFT to account for correlation
Eg. Tawada, Tsuneda, S. Yanagisawa, T. Yanai, & K. Hirao, J. Chem. Phys.
(2004): “Range-separated” interaction in TDDFT matrix, with parameter m
Short-ranged,
use GGA
Long-ranged, use
HF, gives -1/R
Eg. Vydrov, Heyd, Krukau, & Scuseria (2006), 3 parameter range-separated,
SR/LR decomposition…
Eg. Zhao & Truhlar (2006) M06-HF – empirical functional with 35 parameters!!!
Ensures -1/R.
Eg. Stein, Kronik, and Baer, JACS 131, 2818 (2009) – range-separated hybrid, but
with non-empirically determined m
Eg. Heßelmann, Ipatov, Görling, PRA 80, 012507 (2009) – exact-exchange kernel
7. Where the usual approxs. fail
Long-Range Charge-Transfer Excitations
What are the properties of the unknown exact xc kernel that must be wellmodelled to get long-range CT energies correct ?
 Exponential dependence on the fragment separation R,
fxc ~ exp(aR)
 For transfer between open-shell species, need strong frequency-dependence.
step
Step in Vxc re-aligns the 2 atomic
HOMOs  near-degeneracy of
molecular HOMO & LUMO  static
correlation, crucial double excitations 
frequency-dependence!
“LiH”
(It’s a rather ugly kernel…)
Gritsenko & Baerends (PRA, 2004), Maitra (JCP, 2005), Tozer (JCP, 2003) Tawada
et al. (JCP, 2004)
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
8. TDCDFT
Situations not covered by the RG theorem
1
TDDFT does not apply for time-dependent magnetic fields or for
electromagnetic waves. These require vector potentials.
2
The original RG proof is for finite systems with potentials that
vanish at infinity (step 2). Extended/periodic systems can be tricky:
● TDDFT works for periodic systems if
the time-dependent potential is also
periodic in space.
● The RG theorem does not apply when
a homogeneous electric field (a linear
potential) acts on a periodic system.
N.T. Maitra, I. Souza, and K. Burke,
PRB 68, 045109 (2003)
ring geometry:
▲


Bt 
A(t), E(t)
▼
8. TDCDFT
V-representability of current densities
Continuity equation only involves longitudinal part of the current density:
nr, t 
   jL r, t 
t
If
jr, t   jL r, t   jT r, t 
then
jr, t   jL r, t   jT r, t 
[both have the same
comes from a potential
cannot come from
V r, t 
V r,t .
nr, t , and this would violate the RG theorem]
In general, time-dependent currents are not V-representable.
This makes sense, since j is vector (3 components), and V is scalar (1 component).
R. D’Agosta and G. Vignale, PRB 71, 245103 (2005)
TDCDFT
8. TDCDFT
generalization of RG theorem: Ghosh and Dhara, PRA 38, 1149 (1988)
G. Vignale, PRB 70, 201102 (2004)


N
1
1
1


2
1
ˆ
H (t )    p j  c Ar j , t   V r j , t   
 2 j  k r j  rk
j 1  2
N
The full current is uniquely
determined by the pair of
scalar and vector potentials
jr, t   jL r, t   jT r, t 
V,A 


1

2
ˆ
H KS t     p j  1c A KS r j , t   VKS r j , t 

j 1  2
N
uniquely determined up to gauge transformation
TDCDFT in the linear response regime
8. TDCDFT
j1 r, w    d r '  KS r, r, w A ext,1 r, w   A H ,1 r, w   A xc,1 r, w 
3

KS current-current response tensor: diamagnetic + paramagnetic part
 m r, r, w   n0 r  r  r m
where
fk  f j
1 
 
Pmkj r P jk r
2 j ,k  k   j  w  i
Pmkj  k* rm j r   j rmk* r
Effective vector potential
8. TDCDFT
Aext,1 r, w  :
external perturbation. Can be a
true vector potential, or a gauge
transformed scalar perturbation:
  j1 r, w 
A H ,1 r, w  
d r
2 
r  r
iw 

3
1
A ext,1 
Vext,1
iw
gauge transformed
Hartree potential

A xc,1 r, w    d r  f xc r, r, w j1 r, w 
3
ALDA:
A
ALDA
xc,1
r, w  

iw 2
the xc kernel is
now a tensor!
3
ALDA

r, r   j1 r, w 
d
r
f
xc

8. TDCDFT
Nonlocality in space and time
Visualize electron dynamics as the motion (and deformation)
of infinitesimal fluid elements:
r, t
r, t 
Nonlocality in time (memory) implies nonlocality in space!
Dobson, Bünner, and Gross, PRL 79, 1905 (1997)
I.V. Tokatly, PRB 71, 165104 and 165105 (2005), PRB 75, 125105 (2007)
8. TDCDFT
Ultranonlocality and the density
nx0 , t 
●
x0
An xc functional that depends only on the local density
(or its gradients) cannot see the motion of the entire slab.
A density functional needs to have a long range to see
the motion through the changes at the edges.
x
8. TDCDFT
TDCDFT beyond the ALDA: the VK functional
G. Vignale and W. Kohn, PRL 77, 2037 (1996)
G. Vignale, C.A. Ullrich, and S. Conti, PRL 79, 4878 (1997)
A xc,1 r, w   A
ALDA
xc,1

c
r, w  
   xc r, w 
iwn0 r 
xc viscoelastic stress tensor:
 xc, jk
2

 ~
~
  xc   j v1,k   k v1, j    v1 jk    xc  v1 jk
3


vr, w   jr, w  / n0 r 
velocity field
● automatically satisfies zero-force theorem/Newton’s 3rd law
● automatically satisfies the Harmonic Potential theorem
● is local in the current, but nonlocal in the density
● introduces dissipation/retardation effects
8. TDCDFT
XC viscosity coefficients
2
n
~xc n, w   
f xcT n, w 
iw
unif
d 2exc
n2  L
4 T
~
 xc n, w     f xc n, w   f xc n, w  
iw 
3
dn2



In contrast with the classical case, the xc viscosities have both real
and imaginary parts, describing dissipative and elastic behavior:
S xc w  shear modulus
~
 w    w  
iw
Bxcdyn w  dynamical
~
bulk modulus
 w    w  
iw
reflect the
stiffness of
Fermi surface
against deformations
8. TDCDFT
xc kernels of the homogeneous electron gas
Im f xcL
Re f xcL
Im f xcT
Re f xcT
GK: E.K.U. Gross and W. Kohn, PRL 55, 2850 (1985)
NCT: R. Nifosi, S. Conti, and M.P. Tosi, PRB 58, 12758 (1998)
QV: X. Qian and G. Vignale, PRB 65, 235121 (2002)
8. TDCDFT
Static limits of the xc kernels
2 unif
d
exc (n) 4 S xc (0)
L
f xc (0) 

2
dn
3 n2
S xc (0)
T
f xc (0) 
n2
The shear modulus of the electron liquid does not disappear for
(as long as the limit q0 is taken first). Physical reason:
w  0.
● Even very small frequencies <<EF are large compared
to relaxation rates from electron-electron collisions.
● The zero-frequency limit is taken such that local
equilibrium is not reached.
● The Fermi surface remains stiff against deformations.
8. TDCDFT
Applications of the VK functional
(A) In the (quasi)-static ω→0 limit:
● Polarizabilities of π-conjugated polymers
● Nanoscale transport
● Stopping power of slow ions in metals
These applications profit from the fact that VK does not
reduce to the ALDA in the static limit.
(B) To describe excitations at finite frequencies:
● atomic and molecular excitation energies
● plasmon excitations in doped semiconductor structures
● optical properties of bulk metals and insulators
Here the picture is less clear, but it seems that VK works
for metallic systems but can fail for systems with a gap.
8. TDCDFT
TDCDFT for conjugated polymers
ALDA overestimates
polarizabilities of long
molecular chains.
The long-range VK
functional produces
a counteracting field,
due to the finite shear
modulus at w  0.
M. van Faassen et al., PRL 88, 186401 (2002) and JCP 118, 1044 (2003)
8. TDCDFT
Stopping power of electron liquids
Nazarov, Pitarke, Takada, Vignale, and Chang, PRB 76, 205103 (2007)
(ALDA)
(VK)
(Winter et al.)
friction coefficient:
Q  Qsingle  Qxc
particle
Qxc    n0 r   vˆ n0 r  vˆ 
 Im f xc r, r, w 

d 3rd 3r 
w
w 0
► Stopping power measures friction experienced by a slow ion
moving in a metal due to interaction with conduction electrons
► ALDA underestimates friction (only single-particle excitations)
► TDCDFT gives better agreement with experiment: additional
contribution due to viscosity
8. TDCDFT
TDCDFT: discussion
► TDCDFT overcomes several formal limitations of TDDFT:
● allows treatment of electromagnetic waves, vector potentials,
uniform applied electric fields.
● works for all extended systems. One does not need the
condition that the current density vanishes at infinity.
► But TDCDFT is also practically useful in situations that could,
in principle, be fully described with TDDFT:
● Upgrading to the current density can be a more “natural” way
to describe dynamical systems.
● Helps to deal with the ultranonlocality problem of TDDFT
● Provides ways to construct nonadiabatic approximations
► The VK functional is a local xc vector potential beyond the ALDA.
● Works well for many metallic and quasi-metallic systems, but
has problems for systems with a gap.
● More work is needed to construct current-dependent xc functionals.
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
DFT and nanoscale transport
9. Transport
Koentopp, Chang, Burke, and Car (2008)
I
2


 dET E  f E   f E 
L
R
two-terminal Landauer formula

Transmission coefficient, usually obtained from
DFT-nonequilibrium Green’s function
Problems: ● standard xc functionals (LDA,GGA) inaccurate
● unoccupied levels not well reproduced in DFT
transmission peaks can come out wrong
conductances often much overestimated
need need better functionals (SIC, orbital-dep.)
and/or TDDFT
TDDFT and nanoscale transport: weak bias
9. Transport
 
 
  
3
j r ,w    d r'  0 r , r ' ,w Eeff r ' ,w 
Current response:
I w  0 
T0  F 



 d r' Eext w   EH r' ,w   Exc r' ,w 
3
XC piece of voltage drop: Current-TDDFT
Sai, Zwolak, Vignale, Di Ventra,
PRL 94, 186810 (2005)

4
 z n
 2 
dz
4
3e Ac
n
2
R
dyn
dynamical resistance: ~10% correction
9. Transport
TDDFT and nanoscale transport: finite bias
(A) Current-TDDFT and Master equation
Burke, Car & Gebauer, PRL 94, 146803 (2005)
● periodic boundary conditions
(ring geometry), electric field
induced by vector potential A(t)
● current as basic variable
● requires coupling to phonon
bath for steady current
(B) TDDFT and Non-equilibrium
Green’s functions
Stefanucci & Almbladh, PRB 69, 195318 (2004)
● localized system
● density as basic variable
● steady current via electronis
dephasing with continuum of
the leads
► (A) and (B) agree for weak bias and small dissipation
► some preliminary results are available – stay tuned!
Outline
1. A survey of time-dependent phenomena
C.U.
2. Fundamental theorems in TDDFT
N.M.
3. Time-dependent Kohn-Sham equation
C.U.
4. Memory dependence
N.M.
5. Linear response and excitation energies
N.M.
6. Optical processes in Materials
C.U.
7. Multiple and charge-transfer excitations
N.M.
8. Current-TDDFT
C.U.
9. Nanoscale transport
C.U.
10. Strong-field processes and control
N.M.
10. Strong-field processes
TDDFT for strong fields
In addition to an approximation for vxc[n;0,F0](r,t), also need an
approximation for the observables of interest, also with functional dependence
A[n;0,F0]
 Is the relevant KS quantity physical ?
Certainly measurements involving only density (eg dipole moment) can
be extracted directly from KS – no functional approximation needed for
the observable. But generally not the case.
We’ll take a look at:
High-harmonic generation (HHG)
Above-threshold ionization (ATI)
Non-sequential double ionization (NSDI)
Attosecond Quantum Control
Correlated electron-ion dynamics
10. Strong-field processes
High Harmonic Generation
HHG: get peaks at
odd multiples of
laser frequency
Eg. He
TDHF
correlation reduces
peak heights by ~ 2 or 3
L’Huillier (2002)
Measures dipole moment,
|d(w)|2 = ∫ n(r,t) r d3r
so directly available from TD KS
system
Erhard & Gross, (1996)
10. Strong-field processes
Above-threshold ionization
ATI: Measure
kinetic energy of
ejected electrons
Eg. Na-clusters
L’Huillier (2002)
30 Up
l= 1064 nm
I = 6 x 1012
W/cm2 pulse
length 25 fs
• TDDFT is the only computationally
feasible method that could compute ATI for
something as big as this!
• ATI measures kinetic energy of electrons
– not directly accessible from KS. Here,
approximate T by KS kinetic energy.
Nguyen, Bandrauk, and Ullrich, PRA
69, 063415 (2004).
•TDDFT yields plateaus much longer than
the 10 Up predicted by quasi-classical oneelectron models
10. Strong-field processes
Non-sequential double ionization
1
Exact c.f. TDHF
1
TDDFT
2
2
TDDFT c.f. TDHF
Lappas & van Leeuwen (1998),
Lein & Kummel (2005)
Knee forms due to a switchover from a sequential to a non-sequential
(correlated) process of double ionization.
Knee missed by all single-orbital theories eg TDHF
TDDFT can get it, but it’s difficult :
• Knee requires a derivative discontinuity, lacking in most approxs
• Need to express pair-density as purely a density functional – uncorrelated
expression gives wrong knee-height. (Wilken & Bauer (2006))
10. Strong-field processes
Non-sequential double ionization: Momentum
Ion-recoil p-distributions
computed from exact KS
orbitals are poor, eg.
Wilken and Bauer, PRA 76, 023409 (2007)
• Generally time-dependent KS momentum distributions don’t have anything
to do with the true p-distribution
( in principle the true p-dist is a functional of the KS system…but what
functional?! – “observable problem”)
A. Rajam, P. Hessler, C. Gaun, N. T. Maitra, J. Mol. Struct. (Theochem), TDDFT Special
Issue 914, 30 (2009);
10. Strong-field processes
,
Electronic quantum control
Is difficult: Consider pumping He from (1s2)  (1s2p)
Problem!! The KS state remains doubly-occupied throughout – cannot evolve into a
singly-excited KS state.
Simple model: evolve two electrons in a harmonic potential from ground-state
(KS doubly-occupied 0) to the first excited state (0,1) :
TDKS
• KS system achieves the exact target excited-state density, but with a doublyoccupied ground-state orbital !! The exact vxc(t) and observables are unnatural and
difficult to approximate.
• If instead, optimized final-state overlap – max possible with KS is ½ while true is 0.98
Maitra, Woodward, & Burke (2002), Werschnik, Gross & Burke (2007)
10. Strong-field processes
Coupled electron-ion dynamics
Classical nuclei coupled to quantum electrons, via Ehrenfest coupling, i.e.
Eg. Collisions of O atoms/ions with
graphite clusters
Freely-available TDDFT code for
strong and weak fields:
http://www.tddft.org
Castro, Appel, Rubio,
Lorenzen, Marques,
Oliveira, Rozzi,
Andrade, Yabana,
Bertsch
Isborn, Li. Tully, JCP 126, 134307 (2007)
10. Strong-field processes
Coupled electron-ion dynamics
Classical Ehrenfest method misses electron-nuclear correlation
(“branching” of trajectories)
How about Surface-Hopping a la Tully with TDDFT ?
Simplest: nuclei move on KS PES between hops. But, KS PES ≠ true PES,
and generally, may give wrong forces on the nuclei.
Should use TDDFT-corrected PES (eg calculate in linear response).
Many recent interesting applications.
Trajectory hopping probabilities cannot always be simply extracted –
e.g. they depend on the coefficients of the true  (not accessible in TDDFT),
and on non-adiabatic couplings (only ground-excited accessible in TDDFT)
Craig, Duncan, & Prezhdo PRL 2005, Tapavicza, Tavernelli, Rothlisberger, PRL 2007, Maitra,
JCP 2006, Tavernelli, Curchod, Rothlishberger, JCP 2009
To learn more…
Time-Dependent Density Functional Theory, edited by
M.A.L. Marques, C.A. Ullrich, F. Nogueira, A. Rubio,
K. Burke, and E.K.U. Gross, Springer Lecture Notes
in Physics, Vol. 706 (2006)
(see handouts for TDDFT literature list)
Codes with TDDFT
capabilities
NWChem
octopus
Gaussian
Yambo
QChem
ABINIT
GAMESS
Parsec
Turbomole
SIESTA
ADF
Acknowledgments
Collaborators:
• Giovanni Vignale (Missouri)
• Kieron Burke (UC Irvine)
• Ilya Tokatly (San Sebastian)
• Irene D’Amico (York/UK)
• Klaus Capelle (Sao Paulo/Brazil)
• Robert van Leeuwen (Jyväskylä/Finland)
• Meta van Faassen (Groningen)
• Adam Wasserman (Purdue)
• Hardy Gross (MPI Halle)
• Tchavdar Todorov (Queen’s, Belfast)
• Ali Abedi (MPI Halle)
Students/Postdocs:
• Peter Elliott
• Harshani Wijewardane
• Volodymyr Turkowski
• Aritz Leonardo
• Fedir Kyrychenko
• Ednilsom Orestes
• Daniel Vieira
• Yonghui Li
• David Tempel
• Arun Rajam
• Christian Gaun
• August Krueger
• Gabriella Mullady
• Allen Kamal
• Sharma Goldson
• Chris Canahui
• Izabela Raczckowsa