Science makes pies and avocados better

Easy as pie

Expert pie makers transform flour, water, and fat into light, buttery flakes of pastry. The process is simple but difficult to perfect.

Adding water to flour creates gluten, a network of wheat proteins that acts as a scaffold maintaining the pie’s structure. But too much gluten renders the crust tough. Matthew Hartings, an American University chemist, author of Chemistry in Your Kitchen, and member of C&EN’s advisory board, says making a good pie crust requires breaking links between gluten proteins with globs of fat. “It helps with the flakiness,” he tells Newscripts. “You’re making tiny little pockets of gluten.”

The fat must be globbed. If it spreads out evenly, the dough will be homogeneous and less flaky. That means bakers must be careful not to melt the butter or shortening with their warm hands.

Julie Pollock, a chemist at the University of Richmond, says vegetable shortening can be more forgiving than butter because it melts at a higher temperature. There’s a trade-off, though. “The butter adds more flavor,” she says.

Kimberly Jacoby Morris, a chemist and pie maker working as science, technology, engineering, and mathematics program coordinator at the US Air Force Office of Scientific Research, adds a tablespoon of sugar to her piecrust recipe to control gluten content. The sugar interacts with the water, pulling moisture away from the wheat to prevent gluten formation (Jacoby Morris notes that these opinions are her own and do not represent the air force’s official view on pie making).

Another option is replacing water with liquor. Since spirits like vodka are usually more than 30% ethanol there is less water in the same amount of liquid, leading to less gluten formation.

Saving avocados

Experienced avocado eaters can grab a fruit at the supermarket and, with a single squeeze, determine when it will be ripe.

This mysterious art is perfected over hundreds of bowls of guacamole, and supermarkets and avocado distributors have a hard time replicating the process at scale. They test some avocados by poking them or cutting them, but those methods are slow. The lack of information makes it hard to put the best fruit in front of customers, leading to waste. About one in five avocados in US supermarkets goes bad before it is sold, according to a 2015 US Department of Agriculture study (Agriculture, DOI: 10.3390/agriculture5030626).

The start-up Apeel Sciences hopes to reduce avocado waste by helping grocery stores and distributors estimate avocado ripeness. The company shines a light at avocados on a conveyor belt and measures what gets reflected. A machine learning model trained on thousands of avocados uses those data to estimate the fruit’s current firmness. The next step will be improving the model to predict when an avocado will be ready to eat.

“We can get a picture of every single avocado moving down the line,” Apeel’s vice president of technology, Chuck Frazier, tells Newscripts.

Frazier says the variability of avocados is what makes it hard to determine when a fruit will be ripe. The weather an avocado was grown in, its position on the tree, and the country it was harvested in all affect the ripening process. It would be nearly impossible to gather that type of granular information for each avocado, but Apeel’s machine learning model can make accurate guesses about ripeness without those details. Frazier says distributors can use the information to route the ripest avocados to local markets for immediate sale while sending firm ones farther away.

At a trade show in October, Apeel also demonstrated a device that consumers could use to scan individual avocados at the store. Until that device becomes widely available, consumers will still have to rely on the squeeze method.

Please send comments and suggestions to newscripts@acs.org.