Time Dependent DFT
Runge-Gross Theorem
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:
n(r t) vext(r t)
For a given initial-state , 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
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
Examples
spectra of chiral fullerenes: D2C84
F. Furche and R. Ahlrichs, JACS 124, 3804 (2002). GS and EES geometries of diphenyl-
acetylene
240 optically allowed transitions were required
to simulate the spectrum
GS and EES dipole moments for the
DPP dye.
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:
A purely time-dependent function
vext(t), (t)
vext’(t), ’(t)
Assume Taylor-
expandability:
RG prove that the resulting densities n(r,t) and n’(r,t) eventually must differ,
i.e.
same
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
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’s 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 ☺
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
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 Taylor-expandable at
t=0, there must be some k for which RHS = 0
1-1
proves j(r,t) vext(r,t) 1st part of RG ☺
The second part of RG proves 1-1 between densities and potentials:
Take div. of both sides of * and use the eqn of continuity, …
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
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
Then assume fall-off of n0 rapid enough that
surface-integral 0
integrand 0, so if integral 0, then u 0 contradiction
i.e. sam
e
1-1 mapping between time-dependent densities and potentials, for a
given initial state
Adiabatic approximation
n r, t
VH r , t d r 3 depends on density at time t
r - r (instantaneous, no memory)
Vxc n r, t is a functional of n r, t , t t
The time-dependent xc potential has a memory!
Adiabatic approximation: V adia
xc n r, t V n(t ) r
gs
xc
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
TDKS scheme: 3 Steps
1
r ,0
Prepare the initial state, usually the ground state, by
a static DFT calculation. This gives the initial orbitals: (0)
j
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:
j r , t n r , t
3 Calculate the relevant observable(s) as a functional of n r , t
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
Time-dependent selfconsistency (1)
start with
propagate
selfconsistent
until here
KS ground state
t0 T
time
I. Propagate
2
old
i j 2 VKS t j , t t0 , T
n t j t
2
II. With the density calculate the new KS potential
j
V new
KS t Vext t VH n t Vxc n t for all t t0 , T
III. Selfconsistency is reached if V old
KS t V t ,
new
KS t t0 , T
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
Example: two electrons on a 2D quantum strip
initial-state density
hard walls
periodic exact
boundaries
(travelling LDA
waves)
z
x (standing waves)
Charge-density oscillations
● Initial state: constant electric field,
which is suddenly switched off
Δ ● After switch-off, free propagation of
the charge-density oscillations
C.A. Ullrich, J. Chem. Phys. 125, 234108 (2006)
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
Construction of the exact xc potential
Step 1: solve full 2-electron Schrödinger equation
2 2
1
1
V z1 ,t V z 2 ,t i r1 , r2 ,t 0
2
2 2 r1 r2 t
Step 2: calculate the exact time-dependent density
2
dr2 r , r2 ,t n z ,t 2 z ,t
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 dz
2
t
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
Construction of the exact xc potential
n r , t
Ansatz: r ,t exp i r , t
2
A
V
Vxc r ,t V r ,t VH r ,t
xc
1 2 1 2
ln n r ,t ln n r ,t
4 8
1 2
r ,t r ,t
2
dyn
V xc
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
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.
Adapted from C. Ullrich and N. Maitra. APS March Meeting 2008
TDDFT in linear response
Poles at true Poles at KS
excitations excitations
adiabatic approx: no -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)
Matrix equations (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
Transitions for
the Helium atom