Tandem Solar Cells

Introduction

A single band gap wastes low-energy photons by transmission and much of high-energy photon energy by thermalization. A tandem places absorbers with different gaps in the path of light: a higher-gap top cell uses energetic photons, and a lower-gap bottom cell captures light that passes through. This can exceed the ordinary single-junction efficiency benchmark, but only if the optical stack and electrical connections deliver more gain than they cost.

Core explanation

In a common tandem arrangement, the top cell has a wider band gap. It absorbs shorter-wavelength, higher-energy photons and transmits longer wavelengths. The bottom cell has a smaller gap and absorbs part of that transmitted light. Because each photon is absorbed closer to an appropriate gap than it might be in a single-gap absorber, total sub-gap and thermalization losses can be reduced. DOE's explanation of tandem cells describes how different layers use different spectral regions. The actual gain depends on top-cell transparency below its gap and bottom-cell collection.

A two-terminal (2T) tandem connects subcells electrically in series through an interconnection or recombination layer. The same current must pass through both subcells, so the weaker photocurrent limits the stack current under operating conditions. Their voltages add at a given current, less losses in the interconnection. Gap choice and thickness must therefore balance generated currents under the intended spectrum. NREL's tandem-coupling material emphasizes series current limitation and optical coupling between subcells.

A four-terminal (4T) tandem stacks cells optically but gives each its own electrical contacts and operating point. The subcells can work independently near their respective maximum-power points, reducing the strict series current-matching constraint. However, extra transparent contacts and wiring may add optical absorption, resistance, weight and module-integration complexity. NREL's tandem roadmap compares 2T and 4T designs and the distinct challenges of scaling them.

The interconnection between two 2T subcells must pass charge with low resistance and little optical loss. It also must remain chemically and mechanically stable during fabrication and operation. Processing the top cell can damage the bottom cell; a solvent, heat treatment or sputtered contact may degrade previously made layers. These constraints can be more limiting than an ideal calculation based on band gaps alone. Transparent contacts, texture and light management determine whether photons actually reach the lower cell.

Perovskite-on-silicon tandems are a prominent example because perovskite gap tuning can complement silicon's lower gap, and a thin top film can be placed over an established silicon platform. But perovskite stability under light, heat, moisture and bias still matters; the tandem's lifetime output can be limited by its less durable subcell. Other combinations use III–V semiconductors or thin-film absorbers, with different material cost and processing needs. NREL's tandem research roadmap discusses both efficiency and module manufacturing options.

Testing requires attention to spectral mismatch. A lamp could give the top cell too many above-gap photons and the bottom too few, altering the 2T current limit relative to real sunlight. Subcell-resolved spectral response and calibrated reference devices help determine each contribution. A single total current–voltage curve may hide one struggling subcell. Long-term operation should measure both subcells and the interconnection as they age, because degradation rates can differ.

Step-by-step reasoning

Choose high- and low-gap absorbers based on the intended spectrum. Estimate which photons the top absorbs and which it transmits to the bottom. For a 2T design, calculate each subcell's photocurrent and enforce series current matching; add voltages at the same current. For 4T, evaluate each maximum-power point separately, then subtract extra optical/electrical losses. Check interconnection compatibility, module patterning and durability under calibrated illumination.

Visual explanation

Draw incoming sunlight hitting a high-gap top cell first. Short-wavelength arrows stop there, while long-wavelength arrows pass to a low-gap bottom cell. Alongside, show two current–voltage curves; for 2T, a vertical line at the common current intersects both, and their voltages add. For 4T, each curve has its own maximum-power rectangle. A third schematic labels transparent contacts and interconnection loss that can reduce the theoretical spectral benefit.

Real-world analogy

Two filters sorting objects by size can use a mixed stream more effectively than one filter that accepts only a narrow class. If the filters share one conveyor belt, the slower stage sets throughput; if they have separate conveyors, each can run at its preferred speed but needs more equipment. This parallels 2T current matching and 4T independent outputs. The analogy ignores photon energies and semiconductor voltage, so it is not a quantitative model.

Real-world example

A perovskite/silicon 2T tandem initially has top-cell photocurrent 19 mA cm⁻² and bottom-cell photocurrent 17 mA cm⁻² under a calibrated spectrum. Series operation is limited near the bottom-cell current, so increasing top absorption alone may not raise stack power and could further starve the bottom cell. Thinning the top absorber or improving its transmission at longer wavelengths could balance currents. The design must also monitor voltage and recombination, since less top absorption can lower its output if changed indiscriminately.

Why?

Why can a tandem beat the familiar single-junction limit without violating physics? The limit assumes one absorbing junction with one band gap. A tandem has two thresholds and can harvest separate spectral regions with different voltage scales. It changes the model, so a higher possible efficiency is expected. It remains constrained by its own radiative, optical, resistive and recombination losses.

Common misconception

“Stacking any two good cells automatically adds their efficiencies.” The top cell changes the light available to the bottom, and extra contacts/interconnects add loss. Another misconception says two series subcells each contribute their own independent maximum current; series current is common and limited by the weaker subcell. A third assumes 4T has no coupling at all; it removes electrical current matching but the cells remain optically coupled.

Worked example

An illustrative 2T tandem has a top-cell operating voltage of 1.10 V and bottom-cell operating voltage of 0.65 V at the common current density 16 mA cm⁻². Ignoring interconnection loss, stack voltage is 1.75 V and output power density is 1.75 × 0.016 = 0.028 W cm⁻² = 28 mW cm⁻² . If the top could deliver 19 mA cm⁻² alone but the bottom only 16, the 2T stack cannot simply use 19 mA cm⁻² without changing conditions. Under incident power density 100 mW cm⁻², the illustrative stack efficiency is 28% before additional optical and electrical losses.

Quick check

1. In a 2T tandem with top and bottom photocurrent capabilities of 18 and 15 mA cm⁻², which subcell limits common current? Answer: The bottom subcell limits the series current near 15 mA cm⁻² under the simplified comparison.

Exam focus

Explain high-gap top and low-gap bottom spectral roles. State that 2T currents match and voltages add at the same current; distinguish 4T independent electrical operation with extra contacts. Include transparency, interconnect loss, spectrum and stability when claiming advantage. Do not add separate subcell efficiencies as if they received the same full sunlight independently.

Advanced insight

Subcells can couple through emitted luminescence as well as transmitted sunlight. A radiatively efficient top cell may emit photons that the bottom cell absorbs, complicating simple independent-current estimates. NREL's subcell-coupling measurements emphasize that tandem subcells are never fully independent optically. For modules, lateral resistive transport and patterned interconnections add another scale of optimization. An architecture that wins in a tiny research cell may lose its advantage after module-area and lifetime losses are counted.

Summary

Tandems reduce single-gap spectral losses by stacking absorbers with different gaps. Series 2T designs require current matching and add subcell voltages; 4T designs operate independently but need extra contacts and integration. Real benefit depends on optical transparency, interconnection quality, calibrated spectral tests and durable large-area manufacture.

Practice questions

1. Why is the top absorber normally chosen with a larger band gap? Answer: It uses high-energy photons while transmitting lower-energy photons that the lower-gap bottom cell can absorb.

2. In a 2T tandem, how do subcell current and voltage combine? Answer: The same current passes through both series subcells, while their voltages add at that current, less interconnection losses.

3. What advantage does a 4T design have over 2T, and what cost can accompany it? Answer: Each subcell can operate near its own maximum-power point without strict current matching, but extra transparent contacts and wiring can add optical and resistive losses.

4. Why can increasing top-cell absorption reduce tandem output? Answer: It can deprive the bottom cell of photons and lower the series-limiting bottom current.

5. Why does a high initial tandem efficiency not establish lifetime advantage? Answer: One subcell or its interconnection may degrade faster, so long-term power and module yield must be measured.