AI autonomous weapons: when the machine pulls the trigger and what it means

01The Photovoltaic Effect: Light Becomes Current

A solar cell, also known as a photovoltaic cell, is an electronic device that converts the energy of light directly into electricity by using the photovoltaic effect. It is a type of photoelectric cell, a device whose electrical characteristics vary when it is exposed to light. The effect was first observed in 1839 by Edmond Becquerel, who noticed that shining light on certain materials produced a small electric current. The explanation, however, had to wait nearly a century until the development of quantum mechanics revealed why photons could liberate electrons from their atoms.

The photovoltaic effect occurs when photons — particles of light — strike a semiconductor material and transfer their energy to electrons. If a photon carries enough energy, it can knock an electron loose from its atom, creating a free electron and a positively charged hole where the electron used to be. Left to themselves, the electron and hole would quickly recombine, releasing the energy as heat. The genius of the solar cell is in separating the electron and hole before they can recombine, driving the electron through an external circuit where it can do useful work before returning to fill the hole.

This separation is achieved through an internal electric field, created at the junction between two differently doped regions of semiconductor. The field sweeps electrons in one direction and holes in the other, producing a voltage and a current. This is the fundamental mechanism that converts the energy of sunlight directly into electricity, with no moving parts, no fuel, and no emissions.

02Semiconductor Physics: The P-N Junction

Solar cells are built from semiconductors, materials whose electrical conductivity sits between that of conductors and insulators. Silicon, the second most abundant element in the Earth's crust, is by far the dominant material. In its pure crystalline form, silicon holds its electrons tightly and conducts poorly. Doping — the deliberate introduction of impurity atoms — transforms it into a controllable electronic material.

N-type doping adds atoms with extra electrons (like phosphorus), creating a material rich in free electrons. P-type doping adds atoms with missing electrons (like boron), creating a material rich in holes. When n-type and p-type silicon are brought together, the excess electrons from the n-type side diffuse toward the p-type side, and the holes diffuse the other way. At the interface, they recombine, depleting the region of free charge carriers and leaving behind fixed ions that create an electric field. This depleted region, typically micrometers wide, is the p-n junction — the heart of every conventional solar cell.

The built-in electric field at the p-n junction is what makes the cell a cell rather than merely a piece of silicon. When a photon creates an electron-hole pair near the junction, the field pushes the electron toward the n-type side and the hole toward the p-type side. If an external wire connects the two sides, electrons flow from the n-type side through the circuit to the p-type side, generating a current. Each electron that traverses the external circuit carries a small packet of energy — the energy it received from the photon — and the sum of millions of billions of such events is the electrical output of a solar panel.

03Anatomy of a Silicon Solar Cell

A commercial silicon solar cell is a carefully engineered sandwich. The base is a thin wafer of p-type silicon, typically about 150 micrometers thick and doped with boron. On top of this sits a thin layer of n-type silicon, only a fraction of a micrometer thick, doped with phosphorus. The junction between them is where most of the photocurrent is generated. The front surface is textured into microscopic pyramids to reduce reflection, trapping more light by giving photons multiple chances to enter the silicon rather than bouncing off.

An anti-reflective coating, typically silicon nitride, further reduces reflection, cutting the fraction of light lost from about 30% to under 5%. The front surface is covered by a grid of metallic contacts — usually silver — that collect the current from the n-type layer while blocking as little sunlight as possible. The back surface is fully covered by a metallic contact, usually aluminum, that collects the holes. The grid design involves a trade-off: thicker and more numerous fingers reduce electrical resistance but shadow more of the active area, while thinner and fewer fingers capture more light but lose more energy to resistance.

Modern cells employ additional tricks to push efficiency higher. Passivation layers on the front and back surfaces reduce recombination at the surfaces, where the crystalline lattice terminates and dangling bonds trap carriers. Selective emitter structures concentrate doping near the contacts to reduce resistance while keeping it light elsewhere to preserve response to short-wavelength light. Rear-contact designs move all contacts to the back of the cell, eliminating front-side shadowing entirely, at the cost of more complex processing. These refinements have pushed commercial silicon cell efficiencies from around 12% in the 1980s to over 24% today.

04The Shockley-Queisser Limit

No single-junction solar cell can convert all of the sun's energy into electricity. The Shockley-Queisser limit, calculated in 1961, sets the maximum theoretical efficiency of a single-junction cell under standard sunlight at about 33.7% for a bandgap of 1.34 electron volts. The limit arises from two unavoidable losses. First, photons with energy below the bandgap cannot excite electrons at all and pass through the cell wasted. Second, photons with energy above the bandgap excite electrons but the excess energy is lost as heat — a process called thermalization. Together, these two effects throw away roughly half of the solar spectrum's energy before any electricity is generated.

Silicon's bandgap is 1.12 electron volts, close to but not exactly at the optimum. Its single-junction limit is about 32%, which is why even the best laboratory silicon cells max out around 26-27% — the remaining gap is filled by other losses including recombination, reflection, and contact resistance. For decades, this ceiling defined the expectations of the industry: silicon cells would get cheaper and incrementally better, but they would never be dramatically more efficient.

The path beyond the Shockley-Queisser limit is to use multiple junctions tuned to different parts of the solar spectrum. A multi-junction cell stacks layers with decreasing bandgaps: the top layer captures high-energy photons, the next captures medium-energy ones, and the bottom captures low-energy ones, each converting its slice of the spectrum efficiently. Theoretically, an infinite-stack cell could reach 86% efficiency. In practice, six-junction cells have reached over 47% under concentrated sunlight in the laboratory, though these devices are far too expensive for rooftop deployment.

05Panel Types: From Mono to Thin-Film

The solar market is dominated by three families of technology, each with its own trade-offs. Monocrystalline silicon cells are cut from single large crystals grown slowly and expensively by the Czochralski process. They offer the highest commercial efficiencies, typically 20-24%, and dominate the residential rooftop market where space is limited. The black, uniform appearance of mono cells has made them the aesthetic standard as well. Polycrystalline silicon cells are cast from melted silicon that solidifies into many small crystals, a cheaper and faster process that sacrifices some efficiency — typically 15-20% — due to recombination at grain boundaries.

Thin-film solar cells deposit a thin layer of photovoltaic material onto a substrate, using far less material than wafer-based silicon. The leading thin-film technology, cadmium telluride, has achieved commercial efficiencies of about 18-22% and is manufactured at scale by First Solar. Thin-film cells are lighter, flexible, and perform better in low light and high temperatures, but they have historically lagged in efficiency and face concerns about the toxicity and scarcity of some constituent materials. Copper indium gallium selenide (CIGS) is another thin-film technology with competitive efficiencies but complex manufacturing.

The newest entrant is the perovskite cell, named for the crystal structure of its absorber material rather than the specific chemistry. Perovskite solar cells have seen the fastest efficiency improvement in the history of photovoltaics, rising from 3.8% in 2009 to over 26% in the laboratory by 2024. They can be manufactured from solution at low temperatures, potentially at very low cost, and their bandgap is tunable, making them ideal for multi-junction stacks with silicon. The challenge is durability: perovskites are sensitive to moisture, heat, and ultraviolet light, and achieving the 25-year lifetimes expected of conventional panels remains the field's primary hurdle.

06The Cost Collapse and Global Deployment

The story of solar energy over the last two decades is not primarily a story of physics but of industrial scaling. In 2010, solar electricity cost roughly $0.38 per watt of installed capacity; by 2024, it had fallen below $0.10 per watt in many markets, a decline of more than 90%. This collapse has made solar the cheapest source of new electricity generation in most of the world, cheaper than coal, gas, or nuclear, without subsidy in favorable locations. The decline follows a learning rate of roughly 20-24% — every time cumulative production doubles, cost falls by about a fifth — a pattern known as Swanson's Law, the solar analog of Moore's Law.

The drivers of this cost decline are numerous. Economies of scale in manufacturing, particularly in China, which now produces over 80% of the world's solar panels, have driven down per-unit costs. Improvements in cell efficiency mean each panel generates more power for the same area and balance-of-system cost. Automation and process optimization have reduced labor and material waste. Standardization of modules and mounting systems has simplified installation. The result is that solar capacity has grown from about 40 gigawatts installed worldwide in 2010 to over 1,800 gigawatts by 2024 — a forty-fold increase in fourteen years.

This explosive growth has transformed the global energy landscape. Solar now accounts for the largest share of new electricity capacity additions each year, and in countries like Australia, Spain, and Germany, midday solar generation routinely covers a large fraction of demand. The challenge has shifted from generating cheap electricity to integrating intermittent solar into grid systems designed for dispatchable fossil fuels, driving investment in battery storage, grid flexibility, and demand management.

07The Future: Perovskites and Multi-Junction

The next frontier in solar technology is being defined by two complementary advances. Perovskite-on-silicon tandem cells, which layer a perovskite cell tuned to high-energy wavelengths on top of a conventional silicon cell, have already exceeded 33% efficiency in the laboratory, breaking the Shockley-Queisser limit for single-junction silicon. Commercial tandems are entering pilot production, and if the durability challenges can be solved, they could push rooftop panel efficiency from 22% to 28% or higher within a decade — a step change that would reduce the cost of the electricity produced per square meter even if the panels themselves cost slightly more.

All-perovskite tandems, stacking two perovskite layers with different bandgaps, offer a potentially cheaper route to high efficiency since both layers can be deposited from solution at low temperature. Laboratory efficiencies have passed 28%, and the theoretical limit is above 40%. The durability question remains the central obstacle: perovskite cells must survive decades of outdoor exposure to moisture, temperature cycling, and ultraviolet degradation, and while encapsulation strategies have improved rapidly, long-term field data is still being accumulated. Several companies, including Oxford PV and microquanta Semiconductor, are building pilot manufacturing lines, with commercial modules expected later in this decade.

Beyond tandems, the field is exploring transparent solar cells that could turn windows into generators, building-integrated photovoltaics that replace conventional building materials, and solar fuels that use sunlight to produce hydrogen or other chemicals rather than electricity. The unifying theme is that the photovoltaic effect — the simple principle that light can push electrons through a circuit — is far from exhausted as a basis for innovation. A century and a half after Becquerel first observed it, the conversion of sunlight into electricity continues to surprise, and the technology built on it continues to transform the world's energy system.