Sourdough Science

Baking is applied microbiology. That may seem like an odd way to look at it, but it’s only a modest exaggeration. All yeast-leavened breads owe their shapes and textures to the actions of microbes. The yeast used to create bread can be commercially derived (baker’s yeast), or it can be cultivated from the environment around us in the form of a levain (sourdough starter). There are many reasons to use this popular preferment. Levains produce breads that have a depth of flavor that commercial yeast-based breads don’t and are more forgiving thanks to the longer fermentation time. Starting a levain takes time, though, and when you create a preferment using microorganisms from the environment, you must maintain the culture.

A variety of myths and legends surround sourdough starters, and many of them date far back in the long history of yeast and bread. Before it was possible to observe fermentation through a microscope, no one could have imagined—much less explained—how dough could leaven itself, as if by divine intervention. We’ve come a long way since then, and useful information about the science of levain and sourdough breads abounds today. And that’s important, because having a basic understanding of how the microbes in levain behave can make working with this preferment more straightforward.

Getting Cultured: Yeast and Lactic Acid Bacteria

A levain is a preferment used to make sourdough bread, composed of a mix of water and flour that is fermented by lactic acid bacteria (LAB) and wild yeast. By themselves, the raw ingredients that go into a sourdough are essentially flavorless. The sweet-and-sour flavors we love in these breads are by-products of the microbes’ mutually beneficial fight to survive and grow in a complex microscopic ecosystem. And the makeup of that ecosystem evolves over hours or days of fermentation.

Unlike commercial baker’s yeast, which are strains of yeast within the species Saccharomyces cerevisiae, the yeasts in levain are varied, including not only S. cerevisiae but also a mix of other species, such as S. exiguus, Hanensula anomala, and Candida tropicalis. This particular mix of yeasts makes each levain unique flavor-wise—and most importantly, gives the dough rise.

While many people think that their sourdough starter is made up primarily of wild yeast, it is far outnumbered by the lactic acid bacteria in the culture— LAB outnumber yeast cells in a mature sourdough starter by roughly 100 to one. In fact, a levain isn’t stable without the lactic acid bacteria that symbiotically live with the wild yeast.

Like yeast, many kinds of bacteria also engage in fermentation. Smaller than yeasts, most of these bacteria are members of the genus Lactobacillus, so named because the 200-odd species in this group produce lactic acid as they digest sugars. The fermentative power of an individual bacterium is far less than that of a yeast cell, which contains about 20 times the volume of a lactic acid bacterium such as Lactobacillus brevis. San Francisco–style sourdough bread, as well as many other sourdoughs from around the world, derives its characteristic tangy flavor from L. sanfranciscensis. Bacterial species from the genera Leuconostoc, Pediococcus, Enterococcus, Streptococcus, Weissella, and Lactococcus are also common in levain.

Yeasts and LAB coexist so well because each can grow alongside the other and tolerate, to a certain extent, the other’s defense mechanisms. Lactic acid bacteria, like yeasts, are greedy when it comes to resources. The two work together to poison their surroundings—the toxic cocktail they create is full of alcohol and acids that are made during fermentation. It’s a less than warm welcome for other microbes.

Lactic acid bacteria aren’t much inhibited by the ethanol that the yeasts give off. In fact, some strains of lactobacilli are more tolerant of ethanol than yeasts are. The LAB, meanwhile, secrete acids—notably, lactic acid and acetic acid—that lower the pH of the levain. (Scientists who have compared the pH of commercial yeast-based breads and sourdough breads have found that the pH of sourdoughs is much lower: 3.8 to 4.6 versus 5.3 to 5.8 typical of commercial yeast-bread breads.)

But the wild yeast species in levain are able to survive in the increasingly acidic mixture. Without each other, pure cultures of yeasts and LAB can be invaded by other microbes, and if left unchecked, both yeasts and LAB will produce more alcohol and acid than even they can tolerate.

When it comes to peaceful coexistence, it helps that sourdough yeasts and LAB like different foods. Yeasts are better able to make use of a wide range of sugars and starches. C. milleri and other yeasts are happiest eating glucose and fructose (and sucrose, which enzymes quickly break down into these two simpler sugars). L. sanfranciscensis and other LAB, in contrast, prefer maltose. Another display of teamwork is that yeast cells also produce amylase, an enzyme that splits the complex starches and polysaccharides in flour into sugars that are more digestible to the yeasts and their bacterial neighbors.

The Evolution of a Levain

When bakers create levain, they exploit one of the principal forces of evolution— natural selection—as they shape a microbial ecosystem into a tightly controlled tool for bread making. The process illuminates the remarkable ability of yeasts and LAB to adapt to specific environmental conditions.

The growth of yeast and bacteria depend on three key factors: availability of nutrients, acidity, and temperature. Because growth can happen exceptionally fast, species and strains that aren’t adapted to a specific diet (like flour) can quickly be overwhelmed and die out. This is precisely why the inoculants, such as raisin water, that some bakers use to jump-start their levain don’t make a difference. (We think flour, which is chock-full of microbes, and water work just fine.)

Additional factors, including hydration, also influence how a sourdough starter matures. Levain can vary in hydration. If you mix together equal parts water and flour, you’ll produce a levain that is fluid—that is, highly hydrated. We refer to this as a liquid levain (pictured on the right in the image below). If you add more flour to the mixture, say 120% flour to 100% water, the result will be stiff (left). In our experiments, we noticed perceptible differences in pH: the more liquid the starter, the more acidic it will be. (So if you like your sourdoughs good and sour, use a liquid levain.) Your culture can also be affected by contamination or invasion by dust particles, spores, and the like, which can introduce new microbes

Many bakers swear by their particular starter too. But from a microbiology standpoint, the makeup of a starter will be very different if the feeding schedule or temperature is inconsistent. If you aren’t careful, your special starter may be very different on day 1 than it is on day 20 (or even day 2). And different starters can create surprises, which isn’t a good thing if you’re trying to make consistent loaves.

A long-lived levain is almost certainly going to change in composition over time. Think of it like a city; a great city may be just as grand two centuries from now as it is today, but it will have different inhabitants—including some who are descended from the current residents and some who moved in later. A starter’s composition will stay the same only in a perfectly maintained sterile environment, more like a laboratory setting than a bakery. The community of microorganisms will fluctuate and adjust to whatever foods they are given and whatever living conditions they experience. If one strain finds the environment more welcoming than the others, it will quickly grow and crowd its neighbors.

But locking in a specific population of bacteria is not important. What matters is creating a hearty colony of yeasts and lactic acid bacteria that behaves predictably; in other words, as long as the levain is fed on the same schedule and kept at about the same temperature and hydration, it will ripen and mature as expected.

Why Does Baking Bread Smell So Good?

Here’s a fun thing to try: stand outside a bakery on an early summer morning, and watch how people react to the smell of baking bread wafting out the door as they walk by. Their heads turn, their noses lift, their eyes close . . . It’s only a matter of time until someone says, “Oh my God—that smells good!”

What is it about the aroma of bread in the oven that is so irresistible? Yes, for many people, the odors evoke powerful, pleasant memories of childhood. But even people who grew up on plastic-wrapped, essentially aroma-free Wonder Bread break into contented smiles when they enter a bakery while the ovens are going. The reason has as much to do with chemistry as it does with psychology.

We can get some clues as to where the aromas originate by considering wheat products that don’t smell quite as good. Wheat pasta, for example, has essentially no odor when boiled, and not much even when baked—that heartwarming aroma from a baked lasagna comes mainly from the sauce, cheese, and meat, not the noodles. Most unleavened crackers don’t do much for the nose, either.

But chemically leavened baked goods such as biscuits and muffins (made with baking soda and baking powder rather than yeast) can smell very tempting once they start to brown. The color change is a sure tip-off that Maillard reactions are happening. These reactions—in which sugars combine with amino acids to form tasty golden and umber complexes— throw off lots of volatile aromatic compounds that float through the kitchen air and into your nostrils.

Recipes for biscuits and muffins almost always call for added sugar of some kind: the lactose in buttermilk, the fructose in fruit, the dextrose in corn, or even crystals of sucrose sprinkled into the mix. Added sugars help kick-start Maillard reactions.

Another, even better way to generate pleasant aromatic compounds such as ethyl esters (ethyl acetate, hexanoate, and octanoate) is to leaven the flour with yeast. As a by-product of the microbes’ metabolic processes, the yeast cells produce chemicals that break down during baking into delicious-smelling aromatics. The longer the fermentation, the more pronounced the yeast flavors become since the microbes have more time to produce these compounds.

We have tried baking the same bread recipe with and without yeast, and the yeast bread develops a far more complex flavor profile. A big part of the difference is how much better yeast bread smells. The unleavened bread also doesn’t brown nearly as well. Thanks to yeast, your dough is stocked with amino acids that are an integral component of Maillard and other browning reactions.

So the next time you have a loaf in the oven and your kitchen smells like heaven, you have the tiny yeasts to thank.

Is Fresh Yeast Best?

Yeast—living, single-celled fungi—is one of the main reasons bread is so complex and special. These microbes behave like miniscule factories that specialize in the production of bubbles and booze by way of a process called fermentation. In addition to leavening dough, fermentation makes important contributions to the aroma, flavor, and texture of bread.

The yeast used to create bread can be commercially derived, or it can be cultivated from the environment around us in the form of a levain. Using a levain is considered to be the very definition of fermentation by some bakers who dismiss commercial yeast (also known as baker’s yeast) as not producing “real” fermentation. We reject that view; fermentation is fermentation, whether it involves levains or commercial yeast. One method is not more legitimate than the other. The fermentation method you choose depends heavily on your schedule, ability to plan ahead, and yeast preference.

Commercial Yeast

When it comes to commercial yeast, there’s an ongoing debate as to which type of yeast is best for baking bread: active dry, instant, or fresh. The main issue doesn’t seem to be about the “power” of the yeast but rather an unspoken stigma that persists for each kind. You might have heard that “fresh is best,” but in truth, yeast is yeast is yeast—Saccharomyces cerevisiae to be specific.

S. cerevisiae is a fermentation superstar—the species is used by bakers, brewers, and vinters, although the strains that they work with differ. Strains are often isolated, grown, and stored in tightly controlled conditions so that they are best adapted for particular situations, such as making a sourdough, a French bread, an ale, or a champagne. That means that you probably won’t get great results if you try to make bread using a strain developed for brewing beer or winemaking.

At this point you might still be wondering what type of commercial baker’s yeast you should use. If a baker uses the right techniques, there is no reason to use fresh yeast over instant yeast—in a lineup of baked loaves, you’d be hard-pressed to distinguish one from the other in terms of the yeast used. We like working with instant yeast rather than fresh yeast or active dry yeast. After you read our explanations of the differences between the three forms, you will be better equipped to make your own choices.

Fresh Yeast

Developed in the mid-19th century, fresh yeast is the oldest commercial form of yeast. It was originally sold as a cream of yeast mixed with a mash, which served as a growth medium. Fresh yeast is more commonly sold today in blocks of cake or compressed yeast that resemble crumbly, cream-colored modeling clay. Fresh yeast must be dissolved into a liquid but easily does so, dispersing efficiently throughout the dough, which is a plus.

Each gram of compressed yeast contains roughly six billion active yeast cells. Fresh yeast has the highest moisture content of any form of baker’s yeast, but also the shortest shelf life. Blocks require refrigeration and last for only 2–3 weeks after opening. Fresh yeast is highly perishable, a considerable drawback that can cause issues in bakeries as well as home kitchens. At the bakeshop, fresh yeast is likely to sit on the bakers’ worktable for hours while they mix many doughs. The warmth of the bakery will activate the yeast, and it will eventually die because it has nothing to eat. The home baker who buys a pound of fresh yeast must bake frequently to use it all up before it dies. The challenges that come with fresh yeast eventually sparked the next wave of yeast innovation: dried yeast.

Active Dry Yeast

Dried yeast was developed during the Second World War by Fleischmann Laboratories so that United States field infantrymen could bake fresh bread in their camps. The new active dry yeast was not as perishable as fresh yeast and therefore did not require refrigeration and had a longer shelf-life.

Dried yeast is an inert substance when you purchase it, but it becomes a living, thriving colony of microorganisms with the addition of some water and food. During the production process, water is removed from the yeast cells, reducing the moisture content from around 82% to 8% in the case of active dry yeast. The desiccation sends the cells into a state of dormancy. Particles of dormant yeast are coated with a protective layer of dead yeast cells to form tiny granules, which are then packaged for sale. Unlike fresh yeast, unopened packages of dormant, active dry yeast can be frozen for months.

Active dry yeast is more convenient than fresh yeast, but it still requires some additional work and comes with its own set of drawbacks. The dormant cells must be reactivated before use, which can be done by stirring the granules in lukewarm (40–43 °C / 104–109 °F) water. Active dry yeast is temperature sensitive—water that is too hot or too cold can damage or kill the cells, reducing the fermentation power of the yeast.

Around 25% of the yeast cells are killed during the production process, which means that active yeast has, ironically, the lowest amount of active yeast (by weight) of either fresh or dry varieties. Thus, more of it must be added to a recipe than other types of yeast. Dead yeast cells also leach a self-produced chemical called glutathione that relaxes dough. Small quantities of glutathione can be beneficial, depending on the dough, but it can quickly make dough become so relaxed that it’s difficult to handle. Active dry yeast is slower to ferment than both instant and fresh yeast. It needs to proof longer to achieve the same results as the other forms of commercial yeast; the time required will depend on the environment and amount of yeast in the dough. Still, smaller quantities of active dry yeast are often the only option available at supermarkets, which is likely why it’s still commonly used in home baking.

Instant Yeast

Instant yeast, also called quick yeast, was developed in the 1970s by French manufacturer Lesaffre. Like active dry yeast, instant yeast is sold as desiccated granules; it is even drier than active yeast, having a moisture content of just 5% or so. The difference is that instant yeast ferments faster, does not require activation, and is less sensitive to water temperature.

So why do we prefer instant yeast over all other forms of commercial yeast? For starters, instant yeast is truly instant—it does not need to be activated; although we prefer to bloom it, you can add it directly to your dough—and, as soon as it comes into contact with moisture, it will begin the fermentation process.

Instant yeast is made with a fast-acting strain of S. cerevisiae, and the noodle-shaped granules are finer than those of active dry yeast. The surface layer of dead cells is more porous than that of active dry yeast, which allows the granules to rehydrate more rapidly. During production, instant yeast is quick-dried, a process that produces significantly more living yeast cells. As a result its leavening power more closely resembles that of fresh yeast. Manufacturers add salts of fatty acids to the yeast to control rehydration and boost the yeast’s gassing power. The moisture content is lower, which increases the shelf life to 2 years in its vacuum pouch, or even longer when refrigerated. Once the package is opened and exposed to oxygen, instant yeast remains active for 1 year if it’s refrigerated after being opened—it’s the trade-off of the more porous surface. Compared with the active dry form, the instant variety produces more gas during fermentation.

Instant yeast is also available in a number of forms; the one you choose will depend on the type of dough you make. For example, enriched doughs with larger proportions of sugar require osmotolerant yeast; osmotolerant instant yeast requires less water than the instant yeast used in lean doughs. So instant yeast offers options you don’t have with active dry yeast, along with added convenience.

You’re likely to encounter an occasional bump in the road in the road when your local grocery store or purveyor only has one type of yeast on hand. Accidents happen and it’s all too easy to get the wrong package of yeast in a rush. Fortunately, this is one bread-making problem that’s easy to fix. By giving the yeast proper care and employing some basic math, you can use any form of yeast successfully. You’ll find our own conversion table on page 10 (volume 3) of Modernist Bread.

Visit to learn more about yeast and bread-making.






Bread Is Lighter Than Whipped Cream

The headline above is surprising but true, and you can test it yourself: put 1 L of whipped cream on the left pan of a balance scale and a 1 L brioche on the right. The scale will tip to the left. Whipped cream has a reputation for being light and airy, but it’s about twice as dense as brioche.

The demonstration is hard to believe because it violates our expectation that a foam should be lighter than a solid. But bread is also a foam—it is just a set foam. The brioche’s crust is solid enough, but the crumb inside is mostly air.

This simple experiment illustrates that the density of bread—that is, its mass divided by its volume—is less than that of almost any other kind of food. Ciabatta, baguette, brioche, sandwich bread, and other common yeast breads typically have a density of just 0.22–0.27 g/cm3. Whipped cream, by comparison, has a density of 0.49 g/cm3. A liter of whipped cream thus weighs twice as much as a brioche of equal volume.

Bread seems denser than it is in large part because its mass is not evenly distributed: a crunchy baguette crust, which resists cutting and chewing, is 50%–100% more dense than the crumb. The crust is about as dense as pinewood (and whipped cream), whereas the density of the crumb is more like that of cork.

But if the crust is as dense as whipped cream, why does crust feel heavier? The short answer is that the chemistry of these two foams differs. To bite through bread (a set foam), you have to tear apart strong chemical bonds among adjacent molecules. But to eat whipped cream (a colloidal foam), you merely have to push adjacent particles apart.

Intuitively, you might expect that airier breads, such as a baguette, are less dense than loaves that have a tighter crumb, such as pumpernickel and other rye breads. And, in fact, that’s true, as this chart shows.

As it turns out, brick-like rye breads are more dense than red pine—and less dense a kernel of wheat. Scientific insights like this are why we find bread endlessly fascinating and fun.

Gluten: How Does It Work?

Gluten has gotten a bad rap lately—it was practically a four-letter word when we started working on Modernist Bread—but in the world of bread, it’s your friend. As Jimmy Kimmel discovered, there’s a bit of confusion about what gluten is and what it does. Whether you avoid gluten or can’t get enough of it, we think it’s important to understand how it works.

Gluten is a protein found in wheat products. In bread making, it’s exceedingly important. Think of gluten as the miraculous net that holds bread together; it helps dough rise by trapping gas bubbles during fermentation and gives bread its unique texture. Although bread begins with many of the same ingredients as cookies, pastries, cakes, and even shortbreads, it has a completely different consistency. Gluten makes bread airy and satisfyingly chewy—it’s hard to imagine enjoying a chewy cake or a bread that crumbles like a cookie.

Gluten is formed when two of wheat’s native proteins, glutenin and gliadin, come into contact with water. That’s why it’s more accurate to talk about the gluten potential of a particular flour, rather than its gluten content. Either way you phrase it, the more gluten a flour can produce, the more able the dough is to hold gas bubbles, and those gas bubbles are what gives bread an open crumb.

Adding water to flour starts a chemical process that can eventually lead to gluten development. When we grind wheat flour, we destroy the structure of the seed (the cells and organelles), preventing germination. But a cascade of chemical reactions will still occur when the flour is hydrated because the materials that cause the reactions are still present. Gluten development occurs when we add water to flour and let the enzymes work as they were intended.

Gluten Development

From a baker’s perspective, gluten development begins during mixing. The basic point of mixing is to hydrate flour. Mixing matters not because it is necessary to develop gluten; you can develop gluten with minimal mixing (there really is no need to knead). Mixing is essential because it speeds up the hydration process and ensures that water is evenly dispersed throughout the flour.

When hydrated, the glutenin and gliadin proteins almost immediately bind and form gluten. The longer glutenin pieces link up with each other via disulfide bonds to form strong, stretchy units of molecules. These interlinked strands are among the largest protein molecules yet identified. More compact gliadin proteins allow the dough to flow like a fluid, whereas glutenins contribute strength.

Although hydration happens quickly, it takes time to form the chemical attachments that knit gluten proteins together into a strong network. Proteases (protein-snipping enzymes) begin cutting strands of gluten into smaller pieces that are able to make additional connections. Protease is found in very small amounts in wheat flour; an excess of it would cut gluten strands too much and have the opposite effect on the gluten network.

As mixing continues and the ingredients transform into dough, the chains of proteins become more numerous and elongated; they organize into a sort of webbing (the network can be seen in the image above, which was taken with a scanning electron microscope) that has both elasticity (the ability to stretch) and extensibility (the ability to hold a shape). Without this little protein tango, bread would be a very different thing: flatter, crumblier, denser, and less chewy.

The network of gluten will continue to develop, gradually becoming stronger and more complex, up until the dough is fully proofed. Enzymes have even more time to act while the dough rests and begins to ferment. Chains of gluten grow longer and stronger as more and more molecules stick together. During bulk fermentation, bakers periodically fold the resting dough to help align the gluten strands into an even, organized structure, which gives the dough the integrity it needs to expand as the carbon dioxide produced by the yeast and water vapor are introduced into the bubbles.

When the gluten network is strong enough, the dough can be shaped. Bakers check gluten development by performing the windowpane test, which involves stretching a portion of dough in your hands. A well-developed dough can be stretched so thin that it’s translucent. Gluten strands tighten and reorganize once again as the dough is divided and shaped. The tension created during shaping helps the dough expand at a steady rate, producing uniform loaves.

Most of the carbon dioxide production during fermentation happens in the final proofing stage. The largest volume increase comes during baking when the dough nearly doubles in volume in the oven. To expand during both processes, the dough must be strong enough to retain the gas that’s produced. Gluten makes the dough elastic enough that the bubble walls can expand like a little balloon without tearing up until the point where the bread overproofs. When carbon dioxide exerts more pressure than a proofed dough can withstand, the gluten structure weakens, releasing the gas and deflating the overproofed dough.

Other Factors

There are other factors that influence gluten development, such as the type of flour you use. Generally, bread bakers are shooting for an 11%–13% protein level, which will give good volume and texture to a loaf. Protein content varies among flours, and in most cases the higher the protein content, the more gluten the dough can typically form. This doesn’t mean that one flour is better than another; rather, different types of flour are better suited for different purposes. We use lower-gluten flour when we want to make cake, for example, because it won’t form a gluten network that can make a cake’s texture rubbery.

Some wheat varieties, including semolina and most ancient grains, don’t have good gluten-forming properties, which is why they are often blended with other wheat flours in bread recipes. Whole-grain wheat flours contain plenty of protein, as well as bran and germ, which chemically and physically affect the strength of dough. The thirsty particles prevent proteins from fully hydrating, excrete compounds that weaken gluten, and can create microscopic holes in the wall.

Other grains, including rice and corn, can’t form the gluten protein at all, although they do contain other proteins. Rye is a special case. It has some gluten, but not the kind that creates a network that makes for a light and airy bread. Rye becomes bread largely by means of pentosans. These polysaccharide molecules form a sticky gel when mixed with water. That gel—not gluten—is what gives breads with a high percentage of rye their structure.

The quantity of water present also plays into the gluten-forming process. Adding too little water won’t work; the flour must be sufficiently hydrated to activate the proteins that form gluten. Too much water also causes problems, resulting in more of a batter than a dough, in which a gluten network will form but never produce a cohesive mass.

Salt provides more than flavor—it strengthens gluten bonding. Although the gluten proteins naturally repel one another, the chloride ions in salt help them overcome that repulsion and stick together. You can see this change happen within dough when you add salt later in the mixing process: as the salt mixes in and dissolves, the tacky dough firms up.

Fats, such as butter and oils, slow down the gluten-forming process by coating the protein strands, which is one reason enriched doughs such as brioche call for longer mixing times. The coating acts like a barrier that prevents gluten proteins from sticking to one another, stunting the growth of long chains. It’s because of these clipped strands of gluten that we can intricately shape enriched doughs, such as challah. With a small addition of solid fat (1%–3%), lean dough becomes stretchier (allowing it to rise higher) and easier to handle. Fat-enriched recipes, like brioche, can call for large amounts of fat. Fat in these quantities hinder gluten formation and lead to a soft, tender crumb that is more like that of a cake.

Certain inclusions can have the same weakening effect. Any inclusion that contains lots of gluten-killing enzymes, for example, is generally tough on dough. That includes raw papaya (rich in papain) and pineapple (high in bromelain). A workaround is to cook these ingredients first; high heat destroys the enzymes.

Time serves as a general tool for controlling gluten development; the longer the flour and water spend together during the hydration process, the more numerous the gluten bonds will be, while a longer mixing time will speed up hydration by forcing the water into the flour. Time also allows enzymes to assist in gluten development, and most notably extensibility.

Mixing methods also matter. Hand-mixing techniques won’t hydrate the dough—and develop the gluten—as fast as machines. Using an electric mixer can make many breads feasible that would otherwise be difficult to mix by hand, like challah.

The next time you make bread, keep these factors in mind. If you want a taste of Modernist Bread, give our Chocolate and Cherry Sourdough, Portuguese Sweet Bread, and Pork Cheek Hum Bao recipes a try.

Soup and a Side of General Relativity

As a chef, Nathan’s passion for creating new dishes is twofold—he creates dishes and the dishes they’re served on. When thinking about a new bowl to create for a 2014 dinner, he found inspiration from a source not often associated with food: the general theory of relativity. The vessel, which is designed to hold two vibrantly colored soups, has a unique shape. Until recently, the bowl might have appeared to be an intricate spiral, but now, the inspiration for its shape—gravitational waves generated by two colliding black holes—has been receiving international attention.

Modernist Cuisine Gravity Bowl

Last week, news broke that on September 14, 2015, the LIGO Scientific Collaboration, a network of scientists from 15 countries, detected gravitational waves for the first time ever. Naturally, the Modernist Cuisine team was excited by this monumental discovery. In honor of Nathan’s cosmic muse, it seemed like a fitting time to share the story of his “gravitational waves” bowl.

In 1915, Albert Einstein came up with the general theory of relativity—in short, it theorizes that gravity is due to the warping of space–time. It was a radical idea for the era, but in 1916 he proposed yet another radical idea with the prediction of gravitational waves, ripples in the curvature of space–time that propagate from their source at the speed of light. He expanded his theory by positing that gravity is made of waves, just as radio signals are made of waves of electromagnetic radiation.

Photo credit: R. Hurt/Caltech-JPL
Photo credit: R. Hurt/Caltech-JPL

About 1.3 billion years ago, a system of two black holes orbited each other and emitted gravitational waves along the way. They eventually coalesced into a supermassive, spinning black hole, and for decades scientists have looked for evidence of its gravitational waves, but to no avail. Amazingly, the LIGO team figured out a way to not only detect gravitational waves but to also prove it was two black holes that produced them—a momentous discovery that helps confirm Einstein’s theory.

Well before Nathan began working on Modernist Cuisine, he spent his days researching general relativity and quantum theories of gravitation. Although his career took him down another pathway, his interest in those subjects remained and would continue to influence his work in new ways. The book Gravitation, for example, influenced the design and approach of Modernist Cuisine. Coauthored by Kip Thorne, one of the founders of the LIGO project, it’s a landmark in the study of gravity.

It’s no wonder Nathan found inspiration in the cosmos yet again. When he thought about colliding black holes in the spring of 2014, he didn’t just see gravitational waves—he also saw an otherworldly bowl that would make two soups spiral around one another.

Modernist Cuisine Gravity Bowl Aluminum 3-D model

To design the bowl, Nathan used Wolfram’s Mathematica to create a mathematical model that mimicked the gravitational waves of orbiting black holes just before they merge. Next, the Modernist Cuisine team turned Nathan’s Mathematica surface into a 3-D model using the machine shop’s five-axis CNC mill to carve the prototype out of solid aluminum.

Modernist Cuisine Gravity Bowl Plaster Mold

From there, the aluminum 3-D model was used to make a negative mold out of plaster.

Modernist Cuisine Gravity Bowl Mold

Local Seattle potter Wally Bivins then made porcelain bowls from the plaster mold by using a technique called “hump molding.” The clay is rolled into a sheet and tamped down over the mold so that the top surface of the plaster mold becomes the soup-side of the bowl. The soft clay bowl is fired in a kiln and glazed, which transforms it into a bright white porcelain dish.

Modernist Cuisine Gravity Plate

Even before the announcement last week, we liked the story behind the gravitational wave bowl because it illustrated that the source of culinary creativity can come from anywhere—even outer space. Before, we were able to tell guests about a bowl inspired by Einstein’s theories. Now we look forward to telling Cooking Lab visitors that we serve soup from what will surely be one of the most important scientific discoveries of our time.

So, remember: if you’re ever at The Cooking Lab, those aren’t noodles in your soup—they’re ripples in the fabric of space–time.

Dinosaur Bone Broth

Bone broth is in. Technically it has been for a really, really long time. The resurgence of bone broth inspired us to create a Modernist Cuisine spin on the trend.

When people try to describe the Cooking Lab, and the building it’s housed in, you hear a lot of comparisons to a certain fictional chocolate factory. The analogy is fair, though we’ve yet to replicate Wonka’s three-course dinner gum. Truthfully, one of the best things about coming into the kitchen is that you can expect the unexpected: new breads, experiments, lasers, even dinosaurs.

Nathan loves food and cooking, but he also really loves dinosaurs. It’s not uncommon to come across fossilized bones at the lab. He’s contributed to paleontology literature and led expeditions in the Montana Badlands—his T. rex count is 12. Sometimes we get to examine some of the fossils that are brought in, but he’s never merged both of these interests. Until now.

There’s a lot of interesting work going on in the craft beer world. Geneticists, paleontologists, archaeologists, microbiologists, and master brewers have been teaming up to extract yeast from archaeological sites and from fossils to reconstruct old recipes and create new brews. We took a little inspiration from these efforts and applied the ideas to cuisine.

Dem Bones

We were able to obtain some fossils from some well-studied dinosaur species. These bones were superfluous so we decided to put them to a creative use The fossils turned out to be fragments of bone from the tail of a Triceratops, which was recovered from one of Nathan’s more recent trips to the Hell Creek Formation, located outside of Jordan, Montana. This formation is known for the incredible diversity of bones discovered there. Most date to the Cretaceous Period, which began 145.5 million years ago. It was the final portion of the Mesozoic Era, and the longest, lasting 79 million years. The Cretaceous Period ended with the Cretaceous-Paleogene (K-Pg) extinction event, 65.5 million years ago.

Blog_Dinosaur Broth_08700_M=C

During this period, the area had a subtropical climate that supported a varied population of plants, mammals, and dinosaurs. One of the most recovered animals from the formation has been Triceratops horridus, a species that typically grew to 9 m/30 ft long and 3 m/10 ft tall and weighed between four and six tons. Fossils from the massive herbivore can be quite large, but bones from the tip of the tail can fit in the palm of your hand.

Nom Nom Dinosaur

There are no blueprints or rules to working with dinosaur fossils in the kitchen—it’s uncharted territory. You can touch a fossil to the tip of your tongue to determine if it’s legit (real dinosaur bones will slightly stick to your tongue, thanks to their porous structure), but there are no books to consult for cooking techniques or recipes.

The fossilization process also places some restrictions on how you can utilize prehistoric bones. Here’s a quick review. After an animal dies, soft tissues like organs and bone marrow begin to decay, leaving spaces where the tissue was. In a process called permineralization, the animal is covered in sediment from ash, silt, and runoff. The sediment protects the bones from decaying, and, eventually, minerals from the sediment fill the spaces left in the bones and replace the calcium phosphate to form a cast. Fossils from the Hell Creek Formation typically contain iron oxide and coal as well as the minerals quartz, feldspar, mica, and pyrite, all of which comprise the mudstone and sandstone found there.

With no soft tissue, marrow-based dishes were out of the question. Instead, we started to think about how we use one of the dinosaur descendants: chicken. The idea of Dino Broth was a quick revelation from there. Instead of pulling out the flavors of soft tissue to flavor our liquid, we would extract the minerals for an earthy broth.

Blog_Dinosaur Broth_08765-Edit

Instead of simmering the fossils for days, head chef Migoya adapted our Pressure-Cooked White Chicken Stock recipe from Modernist Cuisine at Home. With only a few fossils to work with, we were concerned that a two-day simmer would compromise the bones and also fail to draw out all of the subtle aromatics of the minerals. By pressure-cooking the bones, we dramatically reduced our cook time, accelerated the extraction of flavors, and prevented aromatics from escaping into the air. A little salt, pepper, and MSG were added to the broth—just enough to enhance, but not alter, the mineral flavor. We all sampled it and thought it was quite good; the broth was comforting, complex, and earthy—the epitome of how terroir creates unique dimensions of flavor. It’s the next level of bone broth.

Big in Austin

We discovered that there’s one more rule to constructing a broth from fossils: it needs an epic debut. So off it went, packed away with the rest of the prep for a dinner at Qui. Admittedly, we did a bit of strategizing as we looked over the list of attendees. Who would take us seriously? Who would actually taste this stuff? Andrew Zimmern, that’s who.

IMG_0409 (2)

We were fortunate to share a breakfast with the Bizarre Foods host during SXSW. Over coffee, eggs, and monkey bread, the conversation finally turned to dinosaurs. We revealed our broth and then eagerly fed him a sample, which we served steaming out of a makeshift shot glass that we made by hollowing out a raptor tooth. We’re not sure if it’s possible to surprise Andrew Zimmern with any food, but we’d like to think we might have done just that.

His first response was a hearty, satisfied laugh—we were hopeful. After one more swig, he responded, “It tastes like chicken, but it has a riverbed, river-stone vibe. There’s a citrus quality that’s really nice and appealing.” After passing the raptor tooth on to the rest of the party, he joked, “In two years, this is going to be on the [TGI] Fridays’ menu.” But then he got quite earnest. “That is unbelievable. I haven’t felt this alive since I was locked away in the Alcatraz vault and Geraldo set me free.” We thought the debut was a success.

Right after breakfast, we made our way to La Barbecue, where we met up with our friend Kerry Diamond, who joined us on our barbecue crawl before her interview with Nathan. Kerry has visited the lab twice; she attended our 35-course dinner last June. It’s safe to say that at this point, she’s also learned to expect the unexpected from us. The line at La Barbecue is really long. Waiting offered the perfect opportunity to gather another valued opinion from someone who knows and loves food.


Another laugh at our raptor claw and another swig. “I spent all night researching Nathan and all of his interests, including dinosaurs, and now he’s feeding me bone broth that was made from fossils he found.” After another taste, she added, “It’s familiar like homemade chicken broth, but really distinctive at the same time. Is this gluten-free?” It’s definitely local, but we’re not sure if dinosaur is seasonal.

We’re still experimenting and refining the broth. We’re interested to see if fossils from different dinosaurs will differ in flavor profiles or if the excavation location is what matters. Next, the culinary team will be testing broth made from Apatosaurus fossils that Nathan unearthed during a dig in Colorado. Our Jurassic Broth could end up on one of our dinner menus when we have access to extra fossils; however, as Andrew Zimmern pointed out, this project illustrates the most remarkable thing about food. It’s an experience; when you’re curious and experiment, even with the simplest ingredients, you can create incredible moments for people.

Blog_Dinosaur Broth_08823

Tastes Like Chicken

As Andrew Zimmern noted, our broth did indeed taste like chicken—for a good reason. That’s exactly what it was. This post is really our Modernist spin on April Fools’ Day. Thank you to chef Zimmern and Kerry Diamond for their help in our hijinks. As for real dinosaur bone broth, we can dream. For now, we hope you have a happy April Fools’ Day.

Tastes like chicken

The Delicious Science of Guinness

Guinness isn’t just tasty — the company has a long history of technical and scientific innovation.

Guinness draft beer is famous for both its taste and its velvety, foamy head. The creamy foam of dry stouts is notably different from the bitter, more carbonated foam of other beers because of the addition of nitrogen. In fact, kegs that dispense stouts are pressurized with nitrogen, which has a low solubility in liquids and works to displace carbon dioxide (CO2), imparting a unique head with a pleasant mouthfeel.

The bubbles of other beers, as in lagers, form as dissolved CO2 comes out of solution slowly. But this doesn’t translate well to canned beers. So, in the late 1980s, Guinness developed an answer: a special can pressurized not just with 2 but also nitrogen. Cans — and, more recently, bottles — of Guinness contain a floating plastic container called a widget, which releases additional nitrogen when the container is opened. This sophisticated combination has been very successful in mimicking draft Guinness.

In 2006, Guinness introduced another option: the Surger, an ultrasonic device that sits under a pint glass and sends out a pulse of ultrasound to create cavitation, which drives bubbles out of solution.

We cut open a Guinness Bottle to examine how the widget works its magic.

Guiness-Layout-VOL4 CH16_240 2

– Adapted from Modernist Cuisine: The Art and Science of Cooking

Beer-can chicken: popular classic based on science

BY W. WAYT GIBBS Associated Press

You may not find too many restaurant chefs plopping their poultry on cans of PBR, but all those tailgaters and beachside grillers are on to something.

There are solid scientific reasons that chicken really does roast better in a more upright, lifelike pose than when it is flat on its soggy back. And by adding a couple of extra prep steps to the technique and taking your care with the temperature, you can get the best of both worlds: succulent, juicy meat and crispy, golden brown skin. On top of all that, you get to drink the beer! The chicken doesn’t actually need it.

Beer-can chicken recipes are everywhere on the Internet, but most of them don’t address the two biggest challenges of roasting poultry. The first is to avoid overcooking the meat. Nothing is more disappointing at a Labor Day cookout than to bite into a beautiful-looking chicken breast only to end up with a mouthful of woody fiber that seems to suck the saliva right out of your glands. The solution to this first challenge is simple: take your time, measure the temperature correctly and frequently, and choose the right target for the core temperature (as measured at the deepest, densest part of the thigh). When you cook the bird slowly, the heat has more time to kill any nasty bacteria living in the food, so you don’t have to cook the heck the out of thing. The federal government recommends bringing the meat to 165 F for at least 15 seconds. But guidelines issued by the USDA’s Food Safety and Inspection Service show that 35 minutes at 140 F achieves the same degree of pasteurization, even in the fattiest chicken. The recipe below calls for several hours in the oven and a core temperature of 145 F to 150 F, which will meet those guidelines as long as you slow-cook the bird at a low temperature. But be sure you use a reliable, oven-safe thermometer and place it properly as directed in the recipe. The tip shouldn’t be touching or near any bone. The second challenge that most beer-can chicken recipes fail to overcome is crisping the skin. Here, liquid is the enemy, and adding additional liquid in the form of a can full of beer is the wrong approach. So empty the can first _ the specifics of that will be left as an exercise for the reader_ and use the empty can merely as a way to prop up the bird and to block airflow in its interior so that the meat doesn’t dry out.

Also, give the skin some breathing room by running your (carefully washed) fingers underneath it before roasting. As the subdermal fat melts away, it will trickle downward; a few well-placed punctures provide exits without compromising the balloon-like ability of the skin to puff outward under steam pressure. Held apart from the juicy meat, the loose skin will dry as it browns, especially during a final short blast of high heat in a hot oven. Done right, each slice of tender meat will be capped with a strip of wonderfully flavored skin, which will be at its crispiest when it emerges from the oven. So have your table ready, and don’t be slow with the carving knife. But do take a moment to remove the can before you tuck in.

Modernist Cuisine Beer Can Chicken
Photo credit: Ryan Matthew Smith/ Modernist Cuisine, LLC

Start to finish: 3 1/2 to 4 1/2 hours (30 minutes active) Servings: 4 1 medium roaster chicken 12-ounce can of cold beer (any variety you like to drink) Set an oven rack in the lowest position in the oven. Remove the upper racks. Heat the oven to 175 F, or as low as your oven will allow if its controls do not go this low.

Wash your hands well with soap. Remove the neck and bag of giblets, if included, from inside the chicken. Slide your fingertips underneath the skin at the neck opening and gently work the skin away from the meat. Use care to avoid tearing the skin as you pull it loose from the body; continue as far as you can reach on both the front and the back. Turn the chicken over, and repeat from the cavity opening at the base of the bird, making sure to loosen the skin on the drumsticks so that it is attached only at the wings and the ends of the legs. Use a knife to pierce the skin at the foot end of each leg and at the tail end of the front and back. These small incisions will allow the cooking juices to drain away so that they don’t soak into the skin. Pour the contents of the beer can into a glass, and enjoy it at your leisure. Push the empty can into the tail end of the bird far enough that the chicken can stand upright as it rests on the can. If the neck was included with the chicken, use it like a stopper to close up the opening at the top of the bird. Otherwise you can use a bulldog clip to pinch the skin closed so that steam inflates the loose skin like a balloon and holds it away from the damp meat as the chicken roasts. Set a baking sheet in the oven. Insert the probe of an oven-safe thermometer into the deepest part of the chicken’s thigh. Stand the chicken upright (on the can) on the baking sheet and roast until the core temperature reaches 145 F if you want the white meat to be juicy and tender; for more succulent dark meat, continue roasting to a core temperature of 150 F. A medium-size roaster will need 3 to 4 hours.

After the first 30 minutes of roasting, check the effective baking temperature by inserting a digital thermometer through the skin to a depth of 3/8 inch. The temperature there should be within 5 F of the target core temperature (either 145 F or 150 F). If it is too high, open the oven door for several minutes; if too cool, increase the oven setting slightly. Repeat this check of the near-surface temperature every half hour or so. When the core temperature hits the target, take the chicken out and let it rest, uncovered, for 20 minutes. Meanwhile, increase the oven temperature to its hottest baking setting. Don’t use the broiler, but do select a convection baking mode if your oven has one. Return the bird to the hot oven, turn on the light, and watch it carefully as it browns. The goal is crisp, golden brown skin. The skin will start to brown quickly, and browning will accelerate once it starts. So keep your eye on it. Once the chicken is browned, remove the can, carve it, and serve immediately, while the skin is still crispy.

Making your grill (or broiler) shine this summer

Associated Press

Compared to other basic cooking techniques, grilling is hard: the temperatures are high, timing is crucial and slight differences in the thickness or wetness of the food can dramatically affect how quickly it cooks.

Bad design choices by equipment makers—kettle-shaped grills with black interiors, for example—make it harder still. But if you’re willing to do some simple arithmetic or break out a roll of foil, you can reduce the guesswork and get better performance from your grill. Similar tricks work for broiling; after all, a broiler is basically just an inverted grill.

Every grill has a sweet spot where the heat is even. You know you’re cooking in the sweet spot when all of the food browns at about the same pace. In most situations, the bigger the sweet spot, the better. One notable exception is when you need to reserve part of the grill for cooking some ingredients more slowly or keeping previously cooked food warm.

If you find yourself continually swapping food from the center of your grill with pieces at the periphery, that’s a sure sign that your sweet spot is too small.

You can get an intuitive feel for where the edge of the sweet spot lies by looking at the heat from the food’s point of view. I mean that literally: imagine you are a hotdog lying facedown on the grill. If the coals or the gas flames don’t fill your entire field of view, then you aren’t receiving as much radiant heat as your fellow wiener who is dead-center over the heat source. If the falloff in the intensity of the heat is greater than about 10 percent, you’re outside the sweet spot.

You can use the table below to estimate the size of the sweet spot on your own grill. The 26-inch-wide gas grill on my deck has four burners with heat-dispersing caps that span about 23 inches. The food sits only three inches above the burner caps, so when all four burners are going, the sweet spot includes the middle 16 inches of the grill. But if I use only the two central burners, which are 10 inches from edge to edge, the sweet spot shrinks to a measly 5.4 inches, too small to cook two chicken breasts side by side. I can use this to my advantage, however, if I have a big piece of food that is thick in the middle and thinner at the ends, such as a long salmon fillet. By laying the fish crosswise over the two burners, I can cook the fat belly until it is done without terribly overcooking the slimmer head and tail of the fillet.

Sweet spots are narrowest on small grills, such as little braziers, kettles, hibachis, and the fixed grilling boxes at a public parks. If the sweet spot on your grill is too confining for all the food you have to cook, you can enlarge it in several ways.

If the grill height is adjustable, lower it. Bringing the food a couple inches closer to the heat can easily expand the sweet spot by 2 to 3 inches. The effect on the intensity of the heat is less than you might expect: typically no more than about a 15 percent increase.

If your grill is boxy in shape, line the sides with foil, shiny side out. Your goal is to create a hall of mirrors in which the heat rays bounce off the foil until they hit the food. A hotdog at the edge of the grill then sees not only those coals that are in its line of sight, but also reflections of the coals in the foil-lined side of the grill.

The foil trick unfortunately doesn’t work well on kettle grills because their rounded shape tends to bounce the radiant heat back toward the center instead of out to the edges. But if you can find a piece of shiny sheet metal about 4 inches wide and 56 inches long, you can bend the metal into a reflective circular ring and build the coal bed inside of it. All food within the circumference of the ring should then cook pretty evenly.

Jury-rigging a grill in this way wouldn’t be necessary if grills came shiny on the inside and we could keep them that way. But, presumably because nobody likes to clean the guts of a grill, the interiors of most grills are painted black, the worst possible color for a large sweet spot. A black metal surface doesn’t reflect many infrared heat rays; instead it soaks them up, gets really hot, then re-emits the heat in random directions.

Someday, some clever inventor will come up with a self-cleaning grill that has a mirror finish inside, and the sweet-spot problem will simply vanish.



For grills, measure the width of the coals or gas burners (including any burner caps that disperse the heat). Then measure the distance from the top of the coals or burners to the upper surface of the grill grate. Find the appropriate row in the table to estimate the size of the sweet spot, centered over the heat source. This table assumes a nonreflective grill.

To calculate the sweet spot of an electric broiler — which is the ideal vertical distance between the top of the food and the broiler element — measure the distance between the rods of the heating element. Multiply that measurement by 0.44, then add 0.2 inches to the product. For example, if the rods are 2.4 inches apart, the sweet spot is 1.25 inches from the element to the top of the food.

Width of heat source (inches) Height of the food above the heat source (inches) Width of grill sweet spot (inches)
14 3 8.1
14 4 7.7
14 5 7
16 3 9.9
16 4 9
16 5 8.3
20 3 13.2
20 4 12
20 5 11.20
23 3 16.1
23 4 15
23 5 13.3
29 3 21.8
29 4 19.7
29 5 18.9


Photo credit: Ryan Matthew Smith / Modernist Cuisine, LLC