YAMBO YPP Real-Time Post-Processing Documentation

Table of Contents

1. Overview
2. YPP RT Capabilities Summary
3. Occupations and Carrier Dynamics
4. Response Functions and Optical Properties
5. Nonlinear Optics
6. Transient Absorption Spectroscopy
7. Green's Functions Analysis
8. Database Manipulation
9. Field Analysis
10. Practical Examples
11. Input Variables Reference

Overview

YPP (Yambo Post-Processor) is the post-processing tool for YAMBO real-time simulations. It reads the output databases from yambo_rt calculations and performs various analyses including:

• Carrier dynamics: Occupations, lifetimes, density of states evolution
• Optical response: Linear and nonlinear susceptibilities, absorption spectra
• Transient spectroscopy: Time-resolved absorption, Kerr effect
• Green's functions: Two-time correlation functions, spectral functions
• Database manipulation: Manual carrier excitation, field analysis

YPP operates on the databases written by yambo_rt during the simulation, allowing you to extract physical observables without re-running expensive calculations.

YPP RT Capabilities Summary

Main RT Post-Processing Modes

Logical Flags (mod_YPP_real_time.F)

! Main modes
l_RealTime          ! General RT post-processing
l_RT_DBs            ! Database manipulation

! What to analyze
l_RT_occupations    ! Carrier occupations
l_RT_lifetimes      ! Carrier lifetimes
l_RT_density        ! Real-space density
l_RT_G_two_times    ! G^<(t,t')
l_RT_X              ! Response functions
l_NL_X              ! Nonlinear response
l_RT_abs            ! Transient absorption
l_RT_pol            ! Polarization analysis
l_RT_ARPES          ! Time-resolved ARPES
l_RT_fields         ! Field analysis

! How to plot
l_RT_bands          ! Band structure interpolation
l_RT_time           ! Time evolution
l_RT_energy         ! Energy-resolved
l_RT_dos            ! Density of states

Occupations and Carrier Dynamics

Overview

Analyze carrier occupations f_n(k,t) and their evolution in time, energy, or band structure.

Driver: RT_occupations_driver.F

Capabilities

1. Time Evolution (l_RT_time)

Plot occupation changes vs time for selected states.

Routine: RT_occ_time_plot.F

Outputs: - YPP-occ_K*_B*_S*.dat: Occupation vs time for each (k, band, spin) - Includes lifetimes if scattering is present

Key Features: - State selection via k-point circuits (BANDS_path) - Separate electron/hole contributions (SeparateEH) - Include equilibrium occupations (IncludeEQocc) - Group states (OCCgroup)

Input Example:

ypp -y
RealTime
RToccupations
RTtime
BANDS_path= "G K M G"
%QPkrange
  1 | 10 | 5 | 8 |
%

2. Energy-Resolved (l_RT_energy)

Plot occupations vs energy at different times, with Fermi function fitting.

Routine: RT_components_energy_plot.F

Outputs: - YPP-carriers_E.dat: Occupation vs energy - YPP-Ef_and_T.dat: Fitted Fermi level and temperature - Separate files for e-e, e-ph scattering contributions

Key Features: - Automatic Fermi function fitting: f(E) = 1/(1 + exp[(E-μ)/kT]) - Extracts effective temperature and chemical potential - Separate electron/hole analysis - Skip fitting with SkipFermiFIT

Input Example:

ypp -y
RealTime
RToccupations
RTenergy
%QPkrange
  1 | 10 | 1 | 10 |
%

3. Density of States (l_RT_dos)

Time-dependent DOS with optional interpolation.

Routine: RT_dos_time_plot.F

Outputs: - YPP-TD_dos.dat: DOS(E,t) - YPP-TD_INTERPOLATED_dos.dat: Interpolated DOS (if INTERP_grid set)

Key Features: - Gaussian/Lorentzian broadening (DOSESteps, DOSERange) - k-space interpolation for smooth DOS - Projected DOS on atoms (Proj_DOS) - Spin-resolved DOS (spinor calculations)

Input Example:

ypp -y
RealTime
RToccupations
RTdos
DOSESteps= 1000
DOSERange= -5 | 10 | eV
%INTERP_Grid
 10 | 10 | 10 |
%

4. Band Structure Interpolation (l_RT_bands)

Interpolate occupations along high-symmetry paths.

Routine: RT_occ_bands_interpolation.F

Outputs: - YPP-occ_BANDS_*: Occupation along band structure - Animated band structure evolution

Key Features: - Uses INTERPOLATION_driver for smooth bands - Follows time evolution of band occupations - Useful for visualizing carrier relaxation

Input Example:

ypp -y
RealTime
RToccupations
RTbands
BANDS_path= "G K M G"
BANDS_steps= 100

Lifetimes and Scattering Rates

When scattering is included in RT simulation, YPP extracts:

Lifetimes:

τ_n(k) = -ℏ / [2 Im Σ_n(k)]

Scattering rates:

Γ_n(k) = 1/τ_n(k)

Outputs: - Separate e-e, e-ph, e-photon contributions - Hole vs electron lifetimes - Energy-dependent scattering rates

Response Functions and Optical Properties

Overview

Compute linear and nonlinear optical response from current/polarization.

Driver: RT_X_response.F

Linear Response (l_RT_X)

Extract linear susceptibility χ^(1)(ω) from RT simulation.

Routine: RT_X_LRR_real_field

Theory:

χ^(1)(ω) = ∫ P(t) e^(iωt) dt / E(ω)

Supported Field Types: - DELTA/PULSE: Broadband excitation → full χ(ω) - SIN/SOFTSIN: Monochromatic → χ at specific ω - QSSIN: Quasi-static → SHG analysis

Outputs: - YPP-X_probe_order_1: Linear susceptibility - YPP-eps_along_*: Dielectric function ε(ω) = 1 + 4πχ(ω) - YPP-alpha_along_*: Absorption α(ω)

Input Example (Delta Pulse):

ypp -y
RealTime
RTX
Xorder= 1
% Probe_Freq
 0.0 | 0.0 | eV
%
TimeRange= 0 | 100 | fs
ETStpsRt= 1001

Input Example (Monochromatic):

ypp -y
RealTime
RTX
Xorder= 1
% Probe_Freq
 2.0 | 2.0 | eV    # Single frequency
%

Polarization Analysis (l_RT_pol)

Decompose polarization into contributions from different transitions.

Routine: RT_Polarization.F

Modes:

1. Slice mode (RT_pol_mode="slice"):
2. Reads G^<(t,t') from database
3. Reconstructs P(t) from density matrix

4. Useful for checking consistency

5. Transitions mode (RT_pol_mode="transitions"):

6. Decomposes P(t) by transition energy
7. Shows which transitions contribute to each frequency

Theory:

P(t) = -e Σ_{nm,k} d_{nm}(k) ρ_{nm}(k,t)

where d_{nm}(k) are dipole matrix elements.

Outputs: - YPP-TD_P_decomposition: P(ω) by transition energy - Shows resonance structure

Input Example:

ypp -y
RealTime
RTpol
RT_pol_mode= "transitions"
% EnRngeRt
 0.0 | 10.0 | eV
%

Time Configuration

All RT post-processing requires time range setup:

Variables:

TimeRange    ! Time window [t_start, t_end] (fs)
DampMode     ! Damping: "NONE", "LORENTZIAN", "GAUSSIAN"
DampFactor   ! Damping strength (eV)
ETStpsRt     ! Energy steps for Fourier transform
% EnRngeRt    ! Energy range for FT (eV)
 E_min | E_max | eV
%

Damping: - LORENTZIAN: P(t) → P(t) exp(-Γt) → Lorentzian broadening - GAUSSIAN: P(t) → P(t) exp(-Γ²t²) → Gaussian broadening - NONE: No damping (sharp features, may have artifacts)

Nonlinear Optics

Overview

Extract nonlinear susceptibilities χ^(n) from multi-frequency RT simulations.

Driver: NL_ypp_driver.F

Runlevel: ypp -n

Theory

For monochromatic field E(t) = E₀ cos(ωt):

Polarization expansion:

P(t) = χ^(1) E cos(ωt) + χ^(2) E² cos(2ωt) + χ^(3) E³ cos(3ωt) + ...

Harmonic generation: - χ^(1): Linear response at ω - χ^(2): Second harmonic at 2ω (SHG) - χ^(3): Third harmonic at 3ω (THG) - χ^(n): n-th harmonic at nω

Workflow

1. Run RT simulations at different probe frequencies: ```bash # Run 1: ω₁ yambo_rt -F input_w1.in

# Run 2: ω₂ yambo_rt -F input_w2.in

# ... Run N: ωₙ ```

1. Post-process with YPP: bash ypp -n -F ypp_nl.in

Capabilities

1. Frequency Scan (loop_on_frequencies)

Compute χ^(n)(ω) for multiple probe frequencies.

Input:

ypp -n
NonLinear
Xorder= 3              # Up to χ^(3)
% Probe_Freq
 1.0 | 3.0 | 0.1 | eV  # ω from 1 to 3 eV, step 0.1 eV
%
TimeRange= 50 | 100 | fs  # Use steady-state region

Outputs: - YPP-X_probe_order_0: DC component χ^(0) - YPP-X_probe_order_1: Linear χ^(1)(ω) - YPP-X_probe_order_2: SHG χ^(2)(2ω) - YPP-X_probe_order_3: THG χ^(3)(3ω)

Units: - χ^(1): Dimensionless - χ^(n) (n≥2): [cm/statV]^(n-1)

2. Angular Scan (loop_on_angles)

Compute χ^(n) for different field polarizations.

Input:

ypp -n
NonLinear
Xorder= 2
% Probe_Freq
 2.0 | 2.0 | eV
%
% Probe_Angles
 0 | 360 | 10 |    # Angles from 0° to 360°, step 10°
%

Outputs: - YPP-X_angle_order_*: χ^(n) vs polarization angle - Reveals anisotropy and tensor structure

3. Pump-Probe (N_pumps > 0)

Separate pump and probe contributions.

Setup: - Run 1: Pump + Probe - Run 2: Pump only (reference)

Input:

ypp -n
NonLinear
Xorder= 1
Pump_path= "pump_only"   # Path to pump-only run
Probe_path= "pump_probe" # Path to pump+probe run

Analysis:

P_probe(t) = P_pump+probe(t) - P_pump(t)

Outputs: - YPP-pump-*: Pump contribution - YPP-probe-*: Isolated probe response - YPP-polarization: Full pump+probe

Inversion Method

For monochromatic fields, YPP uses Fourier coefficient inversion:

Method: RT_coefficients_Inversion

Theory:

P(t) = Σ_n P_n cos(nωt + φ_n)

Fit P(t) to extract {P_n, φ_n}, then:

χ^(n) = P_n / E₀^n

Advantages: - Extracts all orders simultaneously - Robust to noise - Works for steady-state region

Requirements: - Simulation must reach steady state - Use TimeRange to select steady-state window - At least one full period: T = 2π/ω

Transient Absorption Spectroscopy

Overview

Compute time-resolved optical properties: absorption, Kerr effect, reflectivity.

Driver: RT_TRabs_driver.F → RT_transient_absorption

Runlevel: ypp -t (if compiled with RT support)

Theory

Transient dielectric function:

ε_ij(ω,t) = δ_ij + 4π Σ_trans R_i(trans,t) R_j*(trans,t) / [ω - E_trans(t) + iΓ]

Transitions: - cv: Conduction-valence (equilibrium) - cc: Conduction-conduction (excited carriers) - vv: Valence-valence (holes)

Observables: - Absorption: α(ω,t) = ω Im[ε(ω,t)] / c - Kerr angle: θ_K ∝ Im[ε_xy] / Re[ε_xx] - Reflectivity: R(ω,t) = |(√ε - 1)/(√ε + 1)|²

Capabilities

1. Transient Absorption (TRabsWHAT="abs")

Time-resolved absorption spectrum.

Input:

ypp -t
TransAbs
TRabsWHAT= "abs"
TRabsMODE= "cv+cc+vv"   # Include all transitions
TRabsDIP_plane= "all"   # All ε_ij components
% TRabsEnRange
 0.0 | 10.0 | eV
%
TRabsEnSteps= 1000
TRabsDamp= 0.1 | eV

Outputs: - YPP-eps_along_E_*: ε(ω,t) along field direction - YPP-alpha_along_E_*: α(ω,t) - Separate files for each time step

Modes: - cv: Equilibrium transitions only - cc+vv: Excited carrier contributions - cv+cc+vv: Full spectrum

2. Kerr Effect (TRabsWHAT="kerr")

Time-resolved magneto-optical Kerr effect.

Input:

ypp -t
TransAbs
TRabsWHAT= "kerr"
TRabsDIP_plane= "xy"    # Off-diagonal component

Outputs: - YPP-Kerr_*: Kerr angle vs ω and t - YPP-B_Hall_*: Hall conductivity

Theory:

θ_K + i η_K = -ε_xy / [ε_xx √(ε_xx)]

3. BSE Transient Absorption (l_BS)

Include excitonic effects in transient absorption.

Requirements: - BSE databases from yambo calculation - RT simulation with BSE kernel

Input:

ypp -t
TransAbs
TRabsMODE= "cv"         # Use BSE eigenstates
% TRabsExc
 1 | 10 |               # Excitons 1-10
%

Features: - Uses BSE eigenstates instead of single-particle transitions - Includes excitonic effects in ε(ω,t) - Can compute inter-exciton transitions (TRabsMODE="inter-exc")

4. Dipole Directions

Control which ε_ij components to compute.

Options:

TRabsDIP_plane= "none"  # ε_xx only (default)
TRabsDIP_plane= "xy"    # ε_xx, ε_xy, ε_yx, ε_yy
TRabsDIP_plane= "xz"    # ε_xx, ε_xz, ε_zx, ε_zz
TRabsDIP_plane= "all"   # All 9 components

Seed direction:

% TRabsDIP_dir
 1.0 | 0.0 | 0.0 |      # x-direction
%

YPP automatically generates orthogonal directions for y and z.

Gauge Considerations

Velocity gauge (default):

ε_ij(ω) = δ_ij - 4π Σ_trans v_i(trans) v_j*(trans) / ω²

Length gauge:

ε_ij(ω) = δ_ij + 4π Σ_trans r_i(trans) r_j*(trans) ω²

Use TRabsVelFullExpr to include 1/ω² term explicitly in velocity gauge.

Green's Functions Analysis

Overview

Analyze two-time lesser Green's function G^<(t,t').

Driver: RT_G_two_times_driver.F

Runlevel: ypp -y with RTGtwotimes

Theory

Lesser Green's function:

G^<_{nm}(k,t,t') = i f_{nm}(k,t,t')

where f_{nm}(k,t,t') is the two-time density matrix.

Physical meaning: - Diagonal (n=m): Occupation at time t - Off-diagonal (n≠m): Coherence between states - Time-diagonal (t=t'): Instantaneous density matrix - Time-off-diagonal (t≠t'): Memory effects, correlations

Capabilities

1. Build G^<(t,t')

Routine: RT_G_two_times_build.F

Reads G^< from RT databases and prepares for analysis.

Input:

ypp -y
RealTime
RTGtwotimes
%QPkrange
  1 | 10 | 5 | 8 |
%

2. Interpolate and Plot

Routine: RT_G_two_times_interp_and_plot.F

Interpolate G^< in k-space and plot.

Outputs: - YPP-G_lesser_K*_B*_t_tp.dat: G^<(t,t') for each (k,band) - 2D plots showing time-time correlations

Applications: - Dephasing analysis: Off-diagonal decay - Memory effects: How long system remembers past - Spectral functions: Fourier transform to energy

3. Spectral Function

From G^<(t,t'), compute spectral function:

Theory:

A(k,ω,t) = -2 Im ∫ G^<(k,t,t') e^(iω(t-t')) dt'

Physical meaning: - Time-resolved band structure - Shows how bands evolve during dynamics - Includes broadening from scattering

Input:

ypp -y
RealTime
RTGtwotimes
Rho_deph= 0.1 | eV     # Additional dephasing
% GrEnRange
 -5.0 | 10.0 | eV
%
GrEnSteps= 1000

Retarded Green's Function

For equilibrium or quasi-equilibrium analysis:

Variables:

% Ret_GF_bands
 1 | 10 |              ! Band range
%
Gr_E_step= 0.01 | eV   ! Energy resolution
GF_T_step= 0.1  | fs   ! Time resolution

Theory:

G^R(ω) = 1 / [ω - E_n(k) - Σ^R(ω)]

Database Manipulation

Overview

Manually create carrier excitations for RT simulations.

Driver: RT_DBs_carriers_setup.F

Runlevel: ypp -r

Purpose

Create initial non-equilibrium carrier distributions without running full RT simulation: - Test different excitation scenarios - Initialize pump-probe simulations - Study specific carrier configurations

Capabilities

1. Energy-Based Excitation (l_RTpump_energy)

Excite carriers in specific energy windows.

Input:

ypp -r e
RTDBs
RTpumpNel= 0.1         # Electrons per unit cell
% RTpumpEhEn
 -0.5 | 0.5 | eV       # Energy window (relative to VBM/CBM)
%
RTpumpEhWd= 0.2 | eV   # Energy width (Gaussian)
RTpumpDE= 2.0 | eV     # Energy difference (e-h pairs)

Method: RT_manual_excitation

Theory: - Select states in energy window around VBM (holes) and CBM (electrons) - Create e-h pairs with specific energy separation - Distribute carriers with Gaussian weight

Use cases: - Resonant excitation - Band-edge pumping - Specific transition excitation

2. Fermi Function Excitation (l_RTpump_Fermi)

Create thermal-like distributions.

Input:

ypp -r f
RTDBs
RTpumpNel= 0.1
% Eh_mu
 -0.2 | 0.2 | eV       # Chemical potential (holes | electrons)
%
% Eh_temp
 300 | 300 | K         # Temperature (holes | electrons)
%
h_mu_autotune_thr= 0.01  # Tolerance for Nel matching

Method: RT_Fermi_excitation

Theory:

f_h(E) = 1 - 1/(1 + exp[(E - μ_h)/k_B T_h])
f_e(E) = 1/(1 + exp[(E - μ_e)/k_B T_e])

Features: - Automatic μ tuning to match target Nel - Different T for electrons and holes - Realistic thermalized distributions

Use cases: - Hot carrier injection - Thermal excitation - Quasi-equilibrium initial states

3. K-Space Selection

Restrict excitation to specific k-regions.

Input:

ypp -r e
RTDBs
BANDS_path= "G K"      # Use high-symmetry points
RTpumpBZWd= 0.1 | iku  # k-space width around points
% RTpumpBZReg
 0.0 | 0.0 | 0.0 |     # Additional k-points
 0.5 | 0.5 | 0.0 |
%

Method: RT_select_k_space

Features: - Select k-points near high-symmetry points - Define multiple k-regions - Control k-space width

Use cases: - Valley-selective excitation - Momentum-resolved pumping - Study k-dependent relaxation

Output

Database: ndb.RT_carriers

Contains: - Modified occupation factors f_n(k,t=0) - Carrier density Nel, Nhole - Scattering components (initialized to zero)

Usage:

# 1. Create excitation with ypp
ypp -r e -F ypp_pump.in

# 2. Use in RT simulation
yambo_rt -F rt_input.in

The RT simulation will start from the manually created carrier distribution.

Field Analysis

Overview

Analyze and manipulate external fields.

Driver: RT_fields.F

Runlevel: ypp -y with RTfields

Capabilities

1. Field Fourier Transform

Convert field between time and frequency domains.

Routine: RT_Field_t_to_w

Input:

ypp -y
RealTime
RTfields
TimeRange= 0 | 100 | fs
TimeStep= 0.01 | fs

Outputs: - YPP-E_time_*: E(t) components - YPP-E_freq_*: E(ω) components - YPP-A_time_*: Vector potential A(t) - YPP-A_freq_*: A(ω)

Applications: - Verify field shape - Check frequency content - Analyze pulse spectrum

2. Field Chirping

Apply frequency chirp to field.

Input:

ypp -y
RealTime
RTfields
ypp_chirp= 0.1 | eV/fs  # Chirp rate

Routine: RT_1D_Fourier_Chirp

Theory:

E_chirped(ω) = E(ω - α·t)

where α is the chirp rate.

Applications: - Create chirped pulses - Frequency-swept excitation - Pump-probe with varying delay

3. Load Field from File

Read field from external file.

Input:

ypp -y
RealTime
RTfields
Field(1)= "FROM_W_FILE filename"

Format: Custom frequency-domain field file

Note: Currently under development (RT_Field_w_load)

Field Configuration

Time setup:

TimeRange= t_start | t_end | fs
TimeStep= dt | fs

Field definition (from RT simulation):

Field(1)= "DELTA"      ! Delta pulse
Field(1)= "PULSE"      ! Gaussian pulse
Field(1)= "SIN"        ! Monochromatic
Field(1)= "QSSIN"      ! Quasi-static

Practical Examples

Example 1: Linear Absorption from Delta Pulse

Goal: Compute ε(ω) and α(ω) from broadband excitation.

Step 1: RT simulation

# rt_input.in
negf
RT_Threads= 0
RTBands= 1 | 50 |
Integrator= "EULER"
RTstep= 0.01 | fs
NETime= 100 | fs
Field(1)= "DELTA"
Field_Freq(1)= 0.0 | eV
Field_Int(1)= 1.E-4 | kWLm2
Field_Dir(1)= 1.0 | 0.0 | 0.0 |

Step 2: YPP post-processing

# ypp_abs.in
RealTime
RTX
Xorder= 1
TimeRange= 5 | 100 | fs    # Skip initial transient
DampMode= "LORENTZIAN"
DampFactor= 0.05 | eV
% EnRngeRt
 0.0 | 10.0 | eV
%
ETStpsRt= 1000

Step 3: Run

yambo_rt -F rt_input.in
ypp -y -F ypp_abs.in

Outputs: - YPP-eps_along_E: ε(ω) - YPP-alpha_along_E: α(ω)

Example 2: Second Harmonic Generation

Goal: Compute χ^(2) for SHG.

Step 1: RT simulations at different frequencies

# Loop over frequencies
for omega in 1.0 1.5 2.0 2.5 3.0; do
  mkdir run_w${omega}
  cd run_w${omega}

  # Create input
  cat > rt_shg.in << EOF
negf
RTBands= 1 | 50 |
RTstep= 0.05 | fs
NETime= 200 | fs
Field(1)= "SIN"
Field_Freq(1)= ${omega} | eV
Field_Int(1)= 1.E6 | kWLm2
Field_Dir(1)= 1.0 | 0.0 | 0.0 |
EOF

  yambo_rt -F rt_shg.in
  cd ..
done

Step 2: YPP analysis

# ypp_shg.in
NonLinear
Xorder= 2
% Probe_Freq
 1.0 | 3.0 | 0.5 | eV
%
TimeRange= 100 | 200 | fs  # Steady state

Step 3: Run

ypp -n -F ypp_shg.in

Outputs: - YPP-X_probe_order_1: χ^(1)(ω) - YPP-X_probe_order_2: χ^(2)(2ω)

Example 3: Pump-Probe Transient Absorption

Goal: Time-resolved absorption after pump excitation.

Step 1: RT simulation with pump

# rt_pump.in
negf
RTBands= 1 | 50 |
RTstep= 0.05 | fs
NETime= 500 | fs

# Pump pulse
Field(1)= "PULSE"
Field_Freq(1)= 3.0 | eV
Field_Int(1)= 1.E8 | kWLm2
Field_FWHM(1)= 10 | fs
Field_Dir(1)= 1.0 | 0.0 | 0.0 |

# Include scattering
LifeInterpKIND= "FLAT"
LifeInterpSteps= 10 | fs

Step 2: YPP transient absorption

# ypp_trabs.in
TransAbs
TRabsWHAT= "abs"
TRabsMODE= "cv+cc+vv"
% TRabsEnRange
 0.0 | 8.0 | eV
%
TRabsEnSteps= 800
TRabsDamp= 0.1 | eV

Step 3: Run

yambo_rt -F rt_pump.in
ypp -t -F ypp_trabs.in

Outputs: - YPP-eps_along_E_T*: ε(ω) at each time T - YPP-alpha_along_E_T*: α(ω,T)

Analysis:

import numpy as np
import matplotlib.pyplot as plt

# Load data
times = []
alphas = []
for t in range(0, 500, 10):
    data = np.loadtxt(f'YPP-alpha_along_E_T{t}.dat')
    times.append(t)
    alphas.append(data[:,1])

# Plot 2D map
plt.contourf(data[:,0], times, alphas)
plt.xlabel('Energy (eV)')
plt.ylabel('Time (fs)')
plt.title('Transient Absorption α(ω,t)')

Example 4: Carrier Dynamics and Thermalization

Goal: Track carrier thermalization and extract effective temperature.

Step 1: RT simulation with scattering

# rt_therm.in
negf
RTBands= 1 | 50 |
RTstep= 0.05 | fs
NETime= 1000 | fs

# Initial excitation
Field(1)= "PULSE"
Field_Freq(1)= 4.0 | eV
Field_Int(1)= 1.E7 | kWLm2
Field_FWHM(1)= 20 | fs

# Scattering
LifeInterpKIND= "FLAT"
LifeInterpSteps= 10 | fs

Step 2: YPP carrier analysis

# ypp_carriers.in
RealTime
RToccupations
RTenergy
%QPkrange
  1 | 50 | 1 | 50 |
%

Step 3: Run

yambo_rt -F rt_therm.in
ypp -y -F ypp_carriers.in

Outputs: - YPP-carriers_E.dat: f(E,t) - YPP-Ef_and_T.dat: Fitted T(t) and μ(t)

Analysis:

import numpy as np
import matplotlib.pyplot as plt

# Load fitted parameters
data = np.loadtxt('YPP-Ef_and_T.dat')
time = data[:,0]
T_e = data[:,1]   # Electron temperature
T_h = data[:,2]   # Hole temperature
mu_e = data[:,3]  # Electron chemical potential
mu_h = data[:,4]  # Hole chemical potential

# Plot thermalization
plt.figure(figsize=(10,6))
plt.subplot(2,1,1)
plt.plot(time, T_e, label='Electrons')
plt.plot(time, T_h, label='Holes')
plt.ylabel('Temperature (K)')
plt.legend()

plt.subplot(2,1,2)
plt.plot(time, mu_e, label='Electrons')
plt.plot(time, mu_h, label='Holes')
plt.xlabel('Time (fs)')
plt.ylabel('Chemical Potential (eV)')
plt.legend()

Example 5: Manual Carrier Injection

Goal: Create specific carrier distribution for RT simulation.

Step 1: Create carrier database with YPP

# ypp_inject.in
RTDBs
RTpumpNel= 0.05        # 0.05 e/cell

# Energy-based excitation
% RTpumpEhEn
 -0.3 | 0.3 | eV       # ±0.3 eV around band edges
%
RTpumpEhWd= 0.15 | eV  # 0.15 eV width
RTpumpDE= 2.5 | eV     # 2.5 eV e-h separation

# K-space selection
BANDS_path= "G"        # Near Γ point
RTpumpBZWd= 0.2 | iku  # 0.2 iku radius

Step 2: Run YPP

ypp -r e -F ypp_inject.in

Step 3: Use in RT simulation

# rt_from_db.in
negf
RTBands= 1 | 50 |
RTstep= 0.05 | fs
NETime= 500 | fs

# No external field - just relaxation
# Carriers loaded from ndb.RT_carriers

LifeInterpKIND= "FLAT"
LifeInterpSteps= 10 | fs

Step 4: Run RT

yambo_rt -F rt_from_db.in

Analysis: Track how manually injected carriers relax.

Input Variables Reference

General RT Post-Processing

! Mode selection
RealTime            ! Enable RT post-processing
RToccupations       ! Analyze occupations
RTlifetimes         ! Extract lifetimes
RTdensity           ! Real-space density
RTGtwotimes         ! G^<(t,t')
RTX                 ! Response functions
RTpol               ! Polarization analysis
RTfields            ! Field analysis

! Time configuration
TimeRange= t1 | t2 | fs        ! Time window
TimeStep= dt | fs              ! Time step (for fields)
DampMode= "LORENTZIAN"         ! Damping type
DampFactor= Γ | eV             ! Damping strength

! Energy configuration
% EnRngeRt
 E_min | E_max | eV
%
ETStpsRt= N                    ! Energy steps

! State selection
%QPkrange
 ib_min | ib_max | ik_min | ik_max |
%
BANDS_path= "G K M G"          ! High-symmetry path

Occupations

! Plot types
RTtime              ! f(t)
RTenergy            ! f(E)
RTdos               ! DOS(E,t)
RTbands             ! f along bands

! Options
IncludeEQocc        ! Add equilibrium f
SeparateEH          ! Separate e/h
NoOcc               ! Skip occupation plots
OCCgroup            ! Group states
SkipFermiFIT        ! Skip Fermi fitting
CarrEnRnge          ! Energy selection

! DOS parameters
DOSESteps= N
% DOSERange
 E_min | E_max | eV
%
DOSEBroad= σ | eV

! Interpolation
%INTERP_Grid
 n1 | n2 | n3 |
%

Response Functions

! Linear response
RTX                 ! Enable
Xorder= 1           ! Order (1 for linear)
UseFFT              ! Use FFT (faster)

! Probe configuration
% Probe_Freq
 ω_min | ω_max | dω | eV
%
N_probes= N         ! Number of probes
Probe_path(1)= "path"  ! Path to probe run

! Pump-probe
N_pumps= 1
Pump_path= "path"   ! Path to pump run

! Polarization
RTpol
RT_pol_mode= "slice"       ! or "transitions"

Nonlinear Optics

! Nonlinear response
NonLinear           ! Enable NL mode
Xorder= N           ! Max order (0 to N)

! Frequency scan
% Probe_Freq
 ω_min | ω_max | dω | eV
%

! Angular scan
% Probe_Angles
 θ_min | θ_max | dθ |
%

! Options
PrtPwrSpec          ! Print power spectrum

Transient Absorption

! Transient absorption
TransAbs            ! Enable
TRabsWHAT= "abs"    ! "abs", "kerr", "trans", "refl"
TRabsMODE= "cv"     ! "cv", "cc", "vv", "cv+cc+vv", "inter-exc"

! Energy range
% TRabsEnRange
 E_min | E_max | eV
%
TRabsEnSteps= N
TRabsDamp= Γ | eV

! Dipole configuration
TRabsDIP_plane= "xy"  ! "none", "xy", "xz", "all"
% TRabsDIP_dir
 dx | dy | dz |
%

! BSE options
% TRabsExc
 i_min | i_max |    ! Exciton range
%
% TRabs_Exc_pol
 px | py | pz |     ! Exciton polarization
%

! Options
TRabsVelFullExpr    ! Full velocity gauge expression

Green's Functions

! G^<(t,t')
RTGtwotimes         ! Enable

! Retarded GF
% Ret_GF_bands
 ib_min | ib_max |
%
Gr_E_step= dE | eV
GF_T_step= dt | fs

! Dephasing
Rho_deph= Γ | eV

! Energy range
% GrEnRange
 E_min | E_max | eV
%
GrEnSteps= N

Database Manipulation

! RT databases
RTDBs               ! Enable DB mode

! Excitation type
! Use: ypp -r e  (energy-based)
!      ypp -r f  (Fermi function)

! Energy-based
RTpumpNel= Nel      ! Electrons per cell
% RTpumpEhEn
 E_h | E_e | eV     ! Energy (holes | electrons)
%
RTpumpEhWd= σ | eV  ! Energy width
RTpumpDE= ΔE | eV   ! e-h energy difference

! Fermi-based
% Eh_mu
 μ_h | μ_e | eV     ! Chemical potential
%
% Eh_temp
 T_h | T_e | K      ! Temperature
%
h_mu_autotune_thr= tol  ! μ tuning tolerance

! K-space selection
BANDS_path= "G K"
RTpumpBZWd= w | iku
% RTpumpBZReg
 k1x | k1y | k1z |
 k2x | k2y | k2z |
%

Field Analysis

! Fields
RTfields            ! Enable

! Chirping
ypp_chirp= α | eV/fs  ! Chirp rate

! Load from file
Field(1)= "FROM_W_FILE filename"

Output Control

! General
SkipOBS_IO          ! Skip observable I/O
FrMinDamp           ! Force minimum damping

! Projection
Proj_DOS            ! Project DOS on atoms
PROJECT_mode= "ATOM"  ! "ATOM", "LINE", "PLANE"

! Normalization
NormN               ! Normalize to Nel

Summary Table: YPP RT Capabilities

Tips and Best Practices

1. Time Range Selection

For Fourier transforms (RTX, NonLinear): - Skip initial transient: TimeRange= 10 | 100 | fs - Use steady-state region for monochromatic fields - Longer time → better frequency resolution: Δω ~ 1/T

For occupations (RToccupations): - Include full dynamics: TimeRange= 0 | T_max | fs - Sample at reasonable intervals

2. Damping

LORENTZIAN (recommended): - Physical broadening - Corresponds to finite lifetime - Use DampFactor ~ Γ_scatt

GAUSSIAN: - Smoother spectra - Better for noisy data - Use for inhomogeneous broadening

NONE: - Only if data is very clean - May show numerical artifacts

3. Energy Resolution

Balance resolution vs computational cost:

ETStpsRt= 1000      ! Good for most cases
ETStpsRt= 2000      ! High resolution
ETStpsRt= 500       ! Quick check

Energy range should cover all relevant features:

% EnRngeRt
 -2.0 | 12.0 | eV   ! Include below VBM and above CBM
%

4. State Selection

For occupations: - Include all active bands: QPkrange= 1 | N_bands | 1 | N_kpts | - Or select specific region: BANDS_path= "G K M G"

For transient absorption: - BSE: Use exciton range TRabsExc= 1 | 10 | - IP: Include enough bands for convergence

5. Nonlinear Analysis

Frequency scan: - Cover resonances: Probe_Freq= 0.5 | 5.0 | 0.1 | eV - Finer steps near resonances - Ensure steady state: TimeRange= T_transient | T_max | fs

Order selection: - Start with Xorder= 1 (linear) - Increase if higher harmonics visible - Xorder= 3 usually sufficient

6. Pump-Probe

Setup: 1. Run pump+probe: yambo_rt -F input_full.in 2. Run pump only: yambo_rt -F input_pump.in -J pump_only 3. Post-process: ypp -n with Pump_path= "pump_only"

Analysis: - Check pump alone first - Verify probe isolation - Look for pump-induced changes

7. Memory Management

Large systems: - Use SkipOBS_IO to reduce I/O - Select specific k-points/bands - Process time slices separately

Interpolation: - Use INTERP_Grid for smooth DOS - Balance accuracy vs cost - Check convergence

8. Troubleshooting

Problem: Noisy spectra - Solution: Increase DampFactor, use longer TimeRange

Problem: Missing features - Solution: Check EnRngeRt, increase ETStpsRt

Problem: Unphysical results - Solution: Verify RT simulation converged, check gauge consistency

Problem: Slow post-processing - Solution: Reduce ETStpsRt, use UseFFT, select fewer states

Code Structure Reference

Main Drivers

ypp/real_time/
├── RT_ypp_driver.F              # Main RT post-processing driver
├── NL_ypp_driver.F              # Nonlinear optics driver
├── RT_occupations_driver.F      # Occupations analysis driver
├── RT_TRabs_driver.F            # Transient absorption driver
└── RT_G_two_times_driver.F      # Green's function driver

Analysis Routines

ypp/real_time/
├── RT_occ_time_plot.F           # f(t) plots
├── RT_components_energy_plot.F  # f(E) and Fermi fitting
├── RT_dos_time_plot.F           # DOS(E,t)
├── RT_occ_bands_interpolation.F # Band structure
├── RT_X_response.F              # Linear response
├── RT_Polarization.F            # Polarization analysis
├── RT_X_effective.F             # Effective susceptibility
├── RT_transient_absorption.F    # Transient spectra
├── RT_G_two_times_build.F       # Build G^<(t,t')
└── RT_fields.F                  # Field analysis

Database Manipulation

ypp/real_time/
├── RT_DBs_carriers_setup.F      # Main DB setup
├── RT_manual_excitation.F       # Energy-based excitation
└── RT_Fermi_excitation.F        # Fermi function excitation

Utilities

ypp/real_time/
├── RT_1D_Fourier_setup.F        # Fourier transform setup
├── RT_coefficients_Fourier.F    # FT of observables
├── RT_damp_it.F                 # Apply damping
├── RT_OBSERVABLES_IO.F          # Read observables
├── RT_OBSERVABLES_damp_and_write.F  # Process and write
└── RT_time_configuration_setup.F    # Time grid setup

Module

ypp/YPPmodules/mod_YPP_real_time.F  # RT variables and types

References

YAMBO Documentation

• Main manual: https://www.yambo-code.eu/wiki/
• RT tutorial: https://www.yambo-code.eu/wiki/index.php/Real_time_approach_to_linear_response
• NL tutorial: https://www.yambo-code.eu/wiki/index.php/Nonlinear_optics

Key Papers

1. RT-TDDFT in YAMBO: Sangalli et al., J. Chem. Phys. 134, 034115 (2011)
2. Nonlinear optics: Attaccalite et al., Phys. Rev. B 88, 235113 (2013)
3. Transient absorption: Grüning & Attaccalite, Phys. Rev. B 89, 081102(R) (2014)
4. Pump-probe: Marini & Attaccalite, Phys. Rev. Lett. 120, 107401 (2018)

Related Tools

• yambo_rt: Real-time propagation
• yambo: Ground-state and BSE calculations
• ypp: General post-processing (bands, DOS, etc.)

Appendix: Quick Command Reference

Basic Usage

# Linear absorption
ypp -y -F ypp_abs.in

# Nonlinear optics
ypp -n -F ypp_nl.in

# Transient absorption
ypp -t -F ypp_trabs.in

# Manual carrier injection
ypp -r e -F ypp_pump.in  # Energy-based
ypp -r f -F ypp_pump.in  # Fermi function

Input File Templates

Linear absorption:

RealTime
RTX
Xorder= 1
TimeRange= 5 | 100 | fs
DampMode= "LORENTZIAN"
DampFactor= 0.05 | eV
% EnRngeRt
 0.0 | 10.0 | eV
%
ETStpsRt= 1000

Carrier dynamics:

RealTime
RToccupations
RTtime
RTenergy
%QPkrange
  1 | 50 | 1 | 50 |
%
BANDS_path= "G K M G"

Second harmonic:

NonLinear
Xorder= 2
% Probe_Freq
 1.0 | 3.0 | 0.2 | eV
%
TimeRange= 100 | 200 | fs

Transient absorption:

TransAbs
TRabsWHAT= "abs"
TRabsMODE= "cv+cc+vv"
% TRabsEnRange
 0.0 | 8.0 | eV
%
TRabsEnSteps= 800
TRabsDamp= 0.1 | eV
TRabsDIP_plane= "all"

Document Version: 1.0
Last Updated: 2025
YAMBO Version: 5.x
Author: Generated from YAMBO source code analysis

For questions and support, visit: https://www.yambo-code.eu