Second Law of Thermodynamics

Drop an ice cube into a glass of water and it melts, cooling the water. You have never seen a glass of water spontaneously freeze part of itself into an ice cube while the rest warms up, even though that would conserve energy just as well. The first law would allow it. The second law forbids it. Nature has a preferred direction.

The second law states that in any real process, total entropy of an isolated system increases. Heat flows spontaneously from hot to cold, never the reverse. No heat engine can convert thermal energy entirely into work; some heat must always be rejected to a cooler reservoir. The theoretical maximum efficiency for any heat engine is the Carnot limit: η = 1 - (T_cold / T_hot), with temperatures in kelvin.

This is why nuclear plants (steam at ~325°C) achieve only 33-34% efficiency while combined-cycle gas plants (combustion at ~1,400°C) reach 60%. Higher source temperatures raise the Carnot limit. No engineering, no matter how advanced, can exceed it.

Worked Example

A nuclear plant operates with steam at 325°C (598 K) and cooling water at 30°C (303 K).

• Calculate Carnot limit. η = 1 - (303/598) = 49.3%.
• The plant actually achieves 33% efficiency. The gap between 49.3% and 33% comes from real-world friction, turbulence, and heat leakage.

Solar panels convert sunlight to electricity at ~20% efficiency. Does this mean they are less efficient than a nuclear plant?

The comparison is misleading. Solar PV is not a heat engine. It converts photons via the photoelectric effect, governed by semiconductor physics (theoretical limit ~33% for single-junction silicon). Comparing PV efficiency to thermal plant efficiency compares two different physical limits. The second law's Carnot bound applies only to heat engines.

The first law is bookkeeping: every joule gets accounted for. The second law is the rule that decides which of those bookkeeping entries you're allowed to collect as work.

---