Exploring the Potential of Ternary Blending for Two and Three-Junction RAINBOW Solar Cells

This study demonstrates that the scalable RAINBOW spectral-splitting architecture, which utilizes ternary-blended subcells to optimize bandgaps and minimize fabrication challenges, can boost organic photovoltaic efficiency from 12.9% in single-junction devices to 17.3% in three-junction configurations, confirming its viability for high-performance, manufacturable solar cells.

The Gist

Imagine you are trying to catch rainwater using a single, wide bucket. If the rain is light, you catch a little. If it's a heavy downpour, your bucket overflows, and you lose the excess water. This is exactly how traditional solar cells work today: they are made of one material that can only "catch" photons (light particles) of a specific energy. If the light is too weak, the material ignores it. If the light is too strong, the material catches it but wastes the extra energy as heat.

The researchers in this paper are trying to build a better "rain-catching system" using a clever new design called RAINBOW.

The Problem with the Old Way (Stacking)

Usually, to catch more types of light, scientists stack different solar cells on top of each other, like a sandwich. The top layer catches bright, high-energy light, and lets the rest pass through to the bottom layer. But this is tricky to build. It's like trying to stack delicate pancakes perfectly; if they don't line up just right, the whole thing collapses or stops working. Also, the layers have to match perfectly in how much electricity they produce, which is a headache for manufacturers.

The New Idea: The RAINBOW Approach

Instead of stacking the cells vertically, the researchers laid them out side-by-side, like tiles on a floor. They use a special optical mirror (the "optical element") to split the sunlight like a prism, sending different colors of light to different tiles.

• Blue light goes to the "Blue Tile."
• Green light goes to the "Green Tile."
• Red light goes to the "Red Tile."

Because they are side-by-side, they don't need to be perfectly stacked, and they don't need to match their electrical output exactly. This makes them much easier to manufacture using large, scalable tools like a squeegee (a process called blade coating).

The Missing Piece: The "Ternary" Blend

The team found that while the "Blue" and "Green" tiles worked well, the "Red Tile" (which catches the lowest energy, far-infrared light) was struggling. It was like a bucket with a hole in the bottom; it could catch the water, but it leaked a lot of energy.

To fix this, they didn't just use one material for the Red Tile. They created a Ternary Blend.
Think of a binary blend as a smoothie made of just two fruits (Donor and Acceptor). A ternary blend adds a third fruit.

• They took their struggling Red material and mixed in a third ingredient.
• This third ingredient acted like a "bridge" or a "helper." It helped the electricity flow better and stopped the energy leaks.
• Specifically, they mixed a material called COTIC-4F (the main catcher) with BTP-eC9 (the helper).

This new three-part mixture didn't just catch the same amount of light; it caught it more efficiently, turning more of that light into electricity.

The Results: A Better Catch

The team tested this idea in two ways:

1. Computer Simulations: They modeled what would happen if they combined these tiles. They found that a 2-junction system (Blue + Red) could reach 16.4% efficiency, and a 3-junction system (Blue + Green + Red) could hit 17.7%.
2. Real-World Tests: They actually built these side-by-side devices using their blade-coating method. The results were very close to the simulations:
   • 2-junction device: 15.9% efficiency.
   • 3-junction device: 17.3% efficiency.

This is a big jump from their single-material devices, which only reached about 12.9%.

The Future Outlook

The paper concludes that this "RAINBOW" design is a very promising, scalable way to make organic solar cells more efficient. However, they note one final hurdle: to push the efficiency even higher, they need to find materials that are really good at catching the very high-energy (wide bandgap) blue light. Currently, those materials aren't quite as good as the red and green ones yet.

In short: By laying solar cells side-by-side instead of stacking them, and by mixing a "third ingredient" into the red-light-catching material to fix its leaks, the team created a solar cell design that is easier to make and catches significantly more energy from the sun.

Technical Summary

Technical Summary: Exploring the Potential of Ternary Blending for Two and Three-Junction RAINBOW Solar Cells

Problem Statement
Organic photovoltaics (OPV) have recently surpassed 20% efficiency in single-junction devices, yet they remain fundamentally limited by the spectral mismatch between the incident solar spectrum and the bandgap of the photoactive material. While multijunction architectures (tandems) offer a pathway to higher efficiencies by reducing thermalization and absorption losses, vertically stacked configurations face significant fabrication hurdles, including current matching constraints, complex interconnect layers, and scalability issues. The "RAINBOW" architecture proposes a solution by placing subcells side-by-side and connecting them externally, utilizing optical elements to split the incident light laterally. However, the performance of RAINBOW devices is currently constrained by the availability of highly efficient materials across the full bandgap spectrum, particularly for narrow-bandgap (NBG, <1.3 eV) and wide-bandgap (WBG, >1.7 eV) regimes. Furthermore, existing NBG systems, such as PTB7-Th:COTIC-4F, often suffer from low open-circuit voltage ($V_{oc}$) and inefficient charge extraction, limiting their utility as red subcells.

Methodology
The study employs a multi-faceted approach combining material screening, ternary blend optimization, device simulation, and scalable fabrication:

1. Material Screening and Ternary Optimization: The authors evaluated seven binary donor-acceptor systems spanning bandgaps from 1.98 eV to 1.16 eV. Recognizing the limitations of the NBG binary PTB7-Th:COTIC-4F, they explored ternary blends by incorporating a third component (either a second acceptor or a fullerene derivative) to tune $V_{oc}$ and improve charge transport. Five specific ternary systems were tested, with a focus on the PTB7-Th:COTIC-4F:BTP-eC9 system.
2. Characterization: Devices were fabricated using high-throughput blade-coating with thickness gradients to determine optimal active layer thicknesses. Comprehensive characterization included current-voltage (JV) measurements under AM1.5G, External Quantum Efficiency (EQE) profiling, and synchrotron-based Grazing Incidence Wide Angle X-ray Scattering (GIWAXS) to analyze microstructure and crystallinity.
3. Simulation: Semi-experimental simulations were conducted to model 2-junction and 3-junction RAINBOW configurations. These models utilized experimental JV parameters and EQE curves to calculate partial currents and efficiencies for different spectral splitting wavelengths ($\lambda_{div}$).
4. Proof-of-Concept Fabrication: Monolithic RAINBOW devices were fabricated using scalable meniscus-guided blade coating, depositing up to six different active materials side-by-side on a single substrate. These devices were characterized under custom partial spectra generated by a programmable LED solar simulator to mimic the RAINBOW operating conditions.
5. Theoretical Analysis: A modified detailed balance limit was calculated to account for intrinsic non-radiative losses in organic semiconductors, specifically reorganization energy, to estimate the theoretical ceiling for this geometry.

Key Contributions and Results

• Ternary Blending for NBG Subcells: The study demonstrates that ternary blending is an effective strategy to overcome the limitations of binary NBG systems. Specifically, the PTB7-Th:COTIC-4F:BTP-eC9 ternary blend (optimized at a 1:0.6:0.9 ratio) significantly improved the performance of the red subcell. While the binary PTB7-Th:COTIC-4F achieved only 6.62% efficiency with a low $V_{oc}$ of 0.54 V, the ternary blend reached 8.71% efficiency with an increased $V_{oc}$ of 0.60 V. Crucially, the ternary blend maintained the extended NIR absorption (up to 1100 nm) while improving charge extraction efficiency, as evidenced by higher EQE values in the shared absorption region. GIWAXS analysis revealed that while the microstructure was dominated by COTIC-4F, the energy level alignment in the ternary system facilitated a cascading electron transfer from COTIC-4F to BTP-eC9, reducing recombination losses.
• Simulation of Multijunction Configurations: Simulations identified that the highest theoretical efficiencies for RAINBOW devices rely on the inclusion of these optimized ternary blends.
  • 2-Junction: The optimal configuration (PTQ10:FCC-Cl as the blue cell and the PTB7-Th:COTIC-4F:BTP-eC9 ternary as the red cell) achieved a simulated Power Conversion Efficiency (PCE) of 16.4%, a significant increase over the best single-junction cell (12.9%) and the binary-based RAINBOW (15.2%).
  • 3-Junction: Adding a green subcell (PM6:DTY6) to the optimal 2-junction stack yielded a simulated PCE of 17.7%.
• Experimental Validation: Proof-of-concept RAINBOW devices fabricated via blade coating validated the simulations.
  • The 2-junction device achieved an experimental PCE of 15.9% (compared to 15.1% for the binary-based equivalent).
  • The 3-junction device achieved an experimental PCE of 17.3% (compared to 17.0% for the binary-based equivalent).
  • These results confirm the viability of the side-by-side deposition method and the effectiveness of ternary blends in enhancing multijunction performance.
• Theoretical Limits: The authors propose a revised detailed balance limit for OPV that incorporates non-radiative losses. This model suggests the maximum theoretical efficiency for single-junction OPV is approximately 27% at a bandgap of 1.5 eV (lower than the standard 33% Shockley-Queisser limit). For multijunction RAINBOW devices, the model indicates that efficiencies can be pushed higher, provided that highly efficient WBG materials (>2 eV) with improved current and fill factors are developed.

Significance
The paper establishes the RAINBOW architecture as a viable, scalable alternative to vertically stacked tandems for high-efficiency OPVs. By demonstrating that ternary blending can effectively tune the $V_{oc}$ and charge transport properties of narrow-bandgap materials without sacrificing spectral coverage, the study provides a concrete pathway to overcome current bottlenecks in multijunction OPV design. The successful fabrication of 3-junction devices with efficiencies approaching 17.3% using scalable blade-coating methods highlights the potential for industrial application. Furthermore, the work underscores that future gains in this geometry are contingent upon the development of high-performance wide-bandgap materials, shifting the focus of material research toward the >2 eV regime to fully realize the potential of the RAINBOW concept.