Predefined Protocols

Convenient wrappers are predefined to simplify the simulation of most important protocols.

IV Simulations

The simulation of an IV curve can be done by the IVProblem constructor

AnnA.IVProblem — Type

IVProblem(
    parm::Parameters,
    range::Union{AbstractArray,Tuple},
    rate;
    double_sweep = true,
    alg_control = AlgControl(dtmin = 1e-20,
        dt = 1e-6,
        reltol = 1e-4,
        abstol = 1e-12,
        tend = abs(sum(diff(range))) / abs(rate) * (1+double_sweep)
    ),  
)

Creates an IVProblem. The voltage parameter V defined in parm will be overwritten by the linear voltage sweep defined as V = t-> first(range) + rate*t. AlgControl.tend is forced to be abs(sum(diff(range))) / abs(rate) * (1+double_sweep), an overwrite can be done on the finalized object using Setfield: p = Setfield.setproperties(p::IVProblem, alg_control=AlgControl(...)).

Units

If no units provided for range and rate, V and V/s is assumed.

source

Here is a simple example on how to use

using AnnA
using Unitful, UnitfulRecipes
parm = Parameters(light = t -> 1.0,   
    vₙₕ= 10u"m/s" ,                 # electron surface recombination vel. at HTM
    vₚₕ= 0.01u"m/s" ,               # hole surface recombination vel. at HTM
    N=500,                          # grid size
    N₀=1e18u"cm^-3"                 # ionic concentration
)
prob = IVProblem(parm, [-0.5,1.7]u"V", 0.2u"V/s")
sol  = solve(prob)

The ProblemSolution object contains also grid and spatial information. All timesteps are stored in a DataFrame an can be acessed via the df field of sol. If we just want to plot the IV characteristics we can do:

using Plots
using UnitfulRecipes  # To interface Unitful with Plots
gr()
sol=sol.df
sol.j = sol.j .|>u"mA/cm^2" # scale to common units
plt = plot(sol.V[sol.fwd], sol.j[sol.fwd],label="Forward", ylims=(-25,40),xlims=(-0.5,1.3),legend=:topleft);
plot!(plt,sol.V[.!sol.fwd], sol.j[.!sol.fwd],label="Backward");
plt

Jsc vs. Voc Curve

The wrapper implements $j_{sc}(V_{oc})$ simulation by two consecutive simulation runs, where the illumination is increased exponentially over time. This reflects the aspect of a slow $V_{oc}$ built up under low illumination intensities.

Open Circuit Voltage Decay (OCVD)

To simulate a open circuit voltage decay a OCVDProblem is implemented:

prob_ocvd = OCVDProblem(
    parm,       # input parameter set
    50u"s",     # illumination time
    1e5u"s",    # time the decay will be simulated to
) 

sol = solve(prob_ocvd)  
plot(sol.t_decay,sol.V_decay,xscale=:log10,xlims=(1e-8,1e5))