Silicon transistors are hitting a wall. The chips powering your phone—like Apple's A17 Pro—have transistors so small that electrons simply leak through the barriers meant to contain them, wasting energy and generating heat. Making them smaller doesn't help anymore. And building a factory to produce them costs over $20 billion.
So researchers are asking: what if we stopped trying to shrink silicon and started using something else entirely. Single molecules could work as electronic components, with electrons naturally flowing one direction more easily than another—like a tiny diode made from a few atoms.
This isn't new thinking. Molecular electronics has been theorized for decades. But only recently has the technology to actually test and measure these molecular-scale devices become reliable enough to move from lab curiosity to genuine engineering problem. A new review in Microsystems & Nanoengineering shows the field has crossed a threshold: this could work.
We're a new kind of news feed.
Regular news is designed to drain you. We're a non-profit built to restore you. Every story we publish is scored for impact, progress, and hope.
Start Your News DetoxThe math is striking
Theoretically, you could pack 10¹⁴ molecular devices into a square centimeter—roughly 1,000 times denser than today's silicon chips. The physics works differently at this scale. Instead of electrons flowing through continuous material, they tunnel across molecular junctions via quantum mechanics. Longer molecules carry less current; shorter ones carry more. It's controllable.
Even more elegant: in benzene-based molecules, electrons can take multiple paths through the ring structure. Depending on where you attach the electrodes, these paths either reinforce each other (constructive interference, high conductance) or cancel out (destructive interference, almost no current). One molecule, multiple logic states, all from quantum interference.
Building the impossible
The hard part isn't the theory. It's making it real. You need electrodes spaced less than 3 nanometers apart—smaller than most viruses. Researchers use methods like electromigration (running current until metal atoms migrate and create a gap) or liquid metal contacts that can touch molecular layers without destroying them.
Some systems deliberately break and reform contacts thousands of times, collecting data from each cycle to map how individual molecules conduct electricity. It's painstaking, but it works.
The obstacles are real
Organic molecules break down above 200 degrees Celsius, but standard chip manufacturing happens at 400 degrees. One solution: add the molecules only at the very end of fabrication, after the silicon has been processed. Another: use DNA origami—folded DNA structures that can guide molecules into precise positions like nanoscale construction scaffolding.
Heat management remains the biggest unsolved problem. Stacking molecular layers (using vertical channels like modern 3D chips) could increase density further, but also creates more heat to dissipate.
Early applications are already emerging. Molecular memristors—devices that remember their history—could enable neuromorphic computing that mimics how brains work. Molecular sensors can detect single chemical reactions, revealing details invisible to conventional instruments.
We're not there yet. But molecular electronics has moved from "interesting physics" to "engineering problem we're learning to solve." The next decade will show whether this becomes the successor to silicon, or remains a brilliant dead-end.
