The new epicurean trend has arrived: hydrocolloids:

Despite its imposing name, a hydrocolloid is a simple thing. A colloid is a suspension of particles within some substance. A hydrocolloid is a suspension of particles in water where the particles are molecules that bind to water and to one another. The particles slow the flow of the liquid or stop it entirely, solidifying into a gel.

Cornstarch used as a thickener is a hydrocolloid. So is plain flour. But the properties of hydrocolloids differ widely, depending on their molecular structure and affinity for water.

Today, Grant Achatz, chef of Alinea in Chicago, uses agar-agar, which is a hydrocolloid made from seaweed that is best known for growing bacteria in petri dishes, and gelatin, a more familiar hydrocolloid made from collagen in meat, to make transparent sheets that he drapes over hot foods. For a dish made of a confit of beef short ribs, he wanted to add a taste of beer so he draped a veil flavored with Guinness on top — “a thin, flavorful glaze that ensured the diner would get some beer flavor in every bit of the dish,” Mr. Achatz said. Plain gelatin would simply melt, and ruin the effect.

Personally, I associate agar agar with the sweet funk of incubating bacteria, not beer glazed short ribs. But that’s just me. A few years ago, I had the pleasure of spending some time in the kitchen of The Fat Duck, which helped pioneer much of this culinary chemistry. Although I’m a big of the restaurant – Heston Blumenthal deserves his three Michelin stars – I learned firsthand that many of these kitchen experiments end in failure:

I’m in the kitchen of the Fat Duck, a Michelin two-star 40 minutes west of London on a tranquil curve of the Thames. The chef, Heston Blumenthal, has a problem: His dessert tastes like horseshit. This is because Heston is trying to put Tryptophan and Valerian Acid, two proteins that induce sleepiness, into a little petit four. He wants you to relax as you’re paying the bill. The Tryptophan is easy to disguise in a dessert. It came from the chemical company in the form of a tasteless white powder that Heston dissolved in Vahlrona chocolate, thus creating the world’s first psychoactive chocolate souffle). The Valerian acid, however, came from the chemical company as an army green silt smelling of pond and manure – or, “tobacco and prune” as Heston says optimistically. So Heston mashes some prunes with fromage blanc, then adds a gram of the Valerian acid. He gives me a spoonful: “Horseshit,” I say, only more bitter. A line cook walks in and wonders what smells so bad. Heston, still a believer, adds prune liquor. Drunk horseshit. Clearly, this isn’t working.

Needless to say, Heston got the petit four right before sending it out to customers. The moral is that experimental food, like experimental science, requires lots and lots of experimentation.