Solar Cells

Solar Cell Domain

The solar domain provides evaluation models for thin-film photovoltaic absorber layers, with a focus on halide perovskites (ABX3), organic photovoltaics (OPV), and two-junction tandem cells.

Physics Model

• Power conversion efficiency (PCE) is calculated using the detailed balance (Shockley-Queisser) framework with corrections for non-radiative recombination losses. The bandgap is derived from composition using Vegard's law and bowing parameters from the perovskite literature.
• Open-circuit voltage (Voc) is estimated from the bandgap minus radiative and non-radiative voltage losses, where non-radiative losses depend on composition and film quality.
• Stability (T80) is modeled using Arrhenius-type degradation kinetics with composition-dependent activation energies. Cesium incorporation improves thermal stability; methylammonium fraction drives moisture-induced decomposition.

Default Parameters

| Parameter | Type | Bounds | Unit | Description | |—————-|———|————|———|——————-| | ma_fraction | continuous | [0.0, 1.0] | — | Methylammonium A-site fraction | | fa_fraction | continuous | [0.0, 1.0] | — | Formamidinium A-site fraction | | cs_fraction | continuous | [0.0, 0.20] | — | Cesium A-site fraction | | br_fraction | continuous | [0.0, 0.50] | — | Bromide X-site fraction | | film_thickness | continuous | [200, 800] | nm | Absorber layer thickness | | annealing_temp | integer | [80, 160] | C | Film annealing temperature | | antisolvent | categorical | [toluene, chlorobenzene, diethyl_ether] | — | Antisolvent quenching agent |

Default Objectives

| Objective | Direction | Unit | |—————-|—————-|———| | pce | maximize | % | | stability_hours | maximize | hours (T80 under 1-sun) |

Templates

materia init my-perovskite --template solar/perovskite
materia init my-tandem --template solar/tandem-cell
materia init my-opv --template solar/organic-pv

Key Trade-Offs

• Efficiency vs. stability: Novel compositions (triple-cation, mixed-halide) can reach PCEs above 24%, but long-term stability under illumination and humidity remains challenging. Higher MA content boosts initial PCE but degrades faster.
• Bandgap tuning: Bromide incorporation widens the bandgap (useful for tandem top cells) but introduces halide segregation under illumination, reducing operational stability.
• Film thickness: Thicker films absorb more light but increase recombination losses and series resistance.

Example: Triple-Cation Perovskite

name: triple-cation-perovskite
domain: solar

parameters:
  - name: ma_fraction
    type: continuous
    bounds: [0.0, 0.3]
  - name: fa_fraction
    type: continuous
    bounds: [0.5, 1.0]
  - name: cs_fraction
    type: continuous
    bounds: [0.05, 0.15]
  - name: br_fraction
    type: continuous
    bounds: [0.0, 0.3]
  - name: film_thickness
    type: continuous
    bounds: [300, 700]

objectives:
  - name: pce
    direction: maximize
    unit: "%"
  - name: stability_hours
    direction: maximize
    unit: hours

constraints:
  - expression: ma_fraction + fa_fraction + cs_fraction <= 1.0

optimizer:
  method: cma-es
  budget: 350
  seed: 42

Typical Results

Well-optimized campaigns discover compositions near the known high-performance region:

• High-PCE solutions: FA-rich compositions (FA > 0.8) with small Cs addition (0.05-0.10) and minimal Br, achieving PCE of 22-25%.
• High-stability solutions: Cs-rich, low-MA compositions with higher Br content, achieving T80 of 2000+ hours but lower PCE (17-20%).
• Balanced solutions: FA~0.75/MA~0.10/Cs~0.10 with Br~0.10, achieving PCE ~21% and T80 ~1000 hours.

Tandem Cell Mode

The tandem template optimizes both top and bottom cell compositions simultaneously, adding parameters for the wide-bandgap top cell and current-matching constraints:

constraints:
  - expression: br_fraction >= 0.15
    description: Wide bandgap required for tandem top cell (>1.7 eV)