This passage was adapted from an article published
Competition to make computer chips smaller and, consequently, faster and more efficient has driven a technological revolution, fueled economic growth, and rapidly made successive generations of computers obsolete. Yet at the current rate of progress this march toward miniaturization will hit a wall by about 2010—for many, an unthinkable prospect. The laws are investigating a different molecular pattern maker: peptides, amino acid chains that are shorter than proteins.
The project grew out of Belcher’s doctoral research on abalone. Her research group discovered in the mid-1990s that a specific peptide causes calcium carbonate to crystallize into the structure found only in the tough abalone shell. From that discovery, Belcher and Hu, Belcher’s postdoctoral adviser at the time, realized that if they resembling accelerated evolution, they developed additional related peptides from those that had the initially promising characteristics.
Hu says that in order to use such a method to assemble a set of circuit-building tools it would be necessary to identify many additional organic compounds that bind to circuit-component materials. The group is making progress on that quest. As they have expanded their targets to 20 more semiconductor materials, their glue. It will take that kind of finesse at the nanoscale to produce selfassembling circuits.
What this question is testing
Topic
The author is profiling a piece of chemistry research aimed at solving a problem the computer industry is about to run into.
Framework
Highlight Noteworthy. The author isn't fighting an opponent — they're explaining why this research matters and where it stands.
Main Point
Here's the simpler version: computer chips have been getting smaller every year, but that march is about to hit a physical wall — transistors can't shrink below 25 nanometers using current techniques. Living cells, though, build smaller structures all the time. So scientists are looking to biology for tools. Belcher and Hu are betting on peptides (short amino-acid chains): they've found peptides that can grab onto specific semiconductor crystals and even act like molecular glue. That's the kind of fine-grained tool you'd need to build circuits that assemble themselves at the nanoscale.
P1: A wall is coming, and biology might help
Chips have been getting smaller, faster, cheaper — but at the current rate, this hits a wall by about 2010. The laws of physics say current transistors can't go below 25 nanometers. Cells, however, build complex structures smaller than that all the time. So the question is whether we can harness those biological processes. Most researchers focus on DNA. Belcher and Hu are working with peptides instead.
P2: How the idea developed
Belcher had been studying abalone shells and found a peptide that controls how calcium carbonate crystallizes there. She and Hu reasoned: if we can find peptides that control crystal growth in semiconductor materials, we'd have a tool for building tiny electronics. No such peptide was known, so they took the bold approach of growing a billion random peptides and testing which ones grabbed onto silicon, gallium arsenide, or indium phosphide crystals. They found a handful, and then refined them by a process resembling evolution.
P3: Where the project is now
To make a real toolkit they'd need many more binding peptides. They're getting there — hundreds of them, across 20+ materials. They're also designing peptides that latch onto two different crystals at once, which acts like a tiny dab of glue. That kind of precision is what circuits that build themselves will need.
Reading along? Open the full official question in LawHub — we show a fragment here and keep the reasoning in our own words.