Narration script

Unit two. In unit one we said energy is always conserved. So here is a fair question: if no energy is ever lost, why can't a solar panel turn all of the sunlight into electricity? The answer is that not all energy is equally useful. Turning disordered energy, heat, into ordered energy, work or electricity, is limited by the second law of thermodynamics. That law sets the ceiling on every efficiency number you will meet in this field.

By the end you should be able to state the first and second laws in plain words, explain the difference between heat and work, calculate the Carnot efficiency, and explain why there are hard thermodynamic ceilings on solar conversion.

Every efficiency figure in solar is really a statement about where the unconverted energy goes. Twenty-two and a half percent for a module, thirty-three point seven percent for the Shockley-Queisser limit, twenty-nine point four percent for the intrinsic limit of silicon. In almost every case, the missing energy becomes heat. And heat matters twice, because a hot module also produces less voltage.

Start with two words. Work is ordered energy transfer: a force pushing in one direction, or a current driven around a circuit. Heat is disordered energy transfer, the random jiggling of atoms passing from a hot place to a cold one. The first law says the books must balance. For a module sitting in the sun, the absorbed sunlight equals the electrical power out plus the heat out.

Here are real numbers. One thousand watts per square metre falls on a module. Around ten percent is reflected or lost optically, which is an estimate, so roughly nine hundred watts per square metre is absorbed. The module delivers two hundred and twenty watts of electricity. So about six hundred and eighty watts per square metre must leave as heat. That is why a module in full sun runs twenty to thirty degrees above the air temperature. Most of the sunlight that hits a solar panel heats it.

Now the second law. In any real process, the total entropy of the universe increases. You can think of entropy as a measure of how spread out energy is, or more precisely, how many microscopic arrangements match what you observe. Heat flows from hot to cold, never the reverse on its own, because spreading increases entropy. The practical consequence is this: no device can take heat from one place and turn all of it into work. Some heat must always be thrown away to somewhere colder.

This diagram is the classic heat engine. Heat flows in from a hot source at the top. Work comes out to the side. And some waste heat must flow down to a cold sink. The second law forbids the bottom arrow from ever being zero.

In eighteen twenty-four, Sadi Carnot worked out the best any engine could ever do. The Carnot efficiency is one minus the cold temperature divided by the hot temperature, with both temperatures in kelvin. In words: the bigger the ratio between hot and cold, the more of the heat can become work.

This plot shows that limit as the hot side gets hotter, with the cold side fixed at three hundred kelvin, about room temperature. Boiling water can never do better than about twenty percent. A steam power plant near five hundred and fifty degrees Celsius is capped at about sixty-four percent, and real plants reach roughly forty to forty-five. Now look at the far right. A source as hot as the surface of the Sun, about fifty-seven hundred kelvin, allows around ninety-five percent. Sunlight is high-quality energy precisely because it comes from something so hot. The second, slightly lower curve is the Landsberg limit, which we will meet in a moment.

Is a solar cell a heat engine? Not mechanically. It does not first turn light into heat. It turns photons directly into separated electrical charges, which is the subject of unit eleven. But thermodynamics still applies, because sunlight is radiation from a hot body and the cell sits at room temperature. Three ceilings matter. The Landsberg limit, about ninety-three percent, is the most work you can ever extract from sunlight, accounting for the entropy carried by the light itself. A stack of infinitely many junctions could, in theory, reach about sixty-nine percent in ordinary sunlight. And a single junction, the kind in almost every panel sold today, is limited to about thirty-four percent. That big drop from ninety-three to thirty-four is not a thermodynamic accident. It comes from using one band gap to harvest a very broad spectrum, and it is the main reason researchers chase tandem cells.

Finally, always ask: efficiency of what, relative to what input? This table separates four numbers people often blur. Industrial cells, around twenty-five to twenty-six percent. Mass-production modules, about twenty-three and a half percent for TOPCon and about twenty-two for PERC, according to the I T R P V roadmap. The ideal Shockley-Queisser value for silicon's band gap, about thirty-three percent. And the realistic intrinsic limit for silicon, twenty-nine point four percent, once unavoidable Auger recombination is included. Module numbers are lower than cell numbers because the module area includes gaps, the frame, and losses in the glass and wiring.

Misconceptions to drop. Perfect materials would not give one hundred percent; thermodynamics forbids it. Waste heat is not just bad engineering; some of it is unavoidable. And entropy is not abstract: it is why a cell's voltage is always below its band gap, and why hot cells make less voltage.

In a factory, heat management is everywhere: glass transparency, temperature coefficients, hot-spot reliability. And the big furnaces, for diffusion, firing and crystal growth, are designed with exactly this energy-balance thinking. Interviewers also expect you to keep three ideas separate: the thermodynamic limit, the material limit, and what production actually achieves.

Try the review questions now, especially the Carnot calculation. Next unit, we leave heat behind and look at light itself, as a wave.