How to Make Cemita Rolls, The Ultimate Sandwich Bun

20150422-cemita-rolls-recipe-01.jpg
Sweet-and-savory homemade cemita buns are easy to make and perfect for sandwiches. Photographs: J. Kenji Lopez-Alt

A cemita roll and a pile of cemita toppings are lying, breathless, in bed. One of them lights up a cigarette and says, "Well, I guess that answers that question."

But which one came first, the bun or the toppings? I ask only because the cemita is one of those rare sandwiches where toppings and bun seem perfectly matched, almost as if one were designed to go with the other. These days, the Mexican sandwiches hailing from Puebla can be found all over Mexico and anywhere in the U.S. with a sizable Mexican population (though as Daniel points out, cemitas in the U.S. tend to be constructed a bit differently from their Mexican counterparts).

What makes the buns and toppings such perfect partners in crime? Well, if you know anything about tortas or cemita sandwiches, it's that they're stacked tall with toppings that are either soft or extremely moist, like avocado, shredded cheese, refried beans, or chipotle chilies. That means that the right structure is of utmost importance when designing a bun for them.

A cemita bun ought to have a rich, golden, egg-enriched crumb. It should be tender enough that fillings don't get crushed or squeezed out the back of the sandwich as you bite through, but substantial enough that it can hold up to layers of moist toppings without crumbling or disintegrating. It should be soft in the center but have a crisp upper and lower crust to offer some textural contrast to the sandwich and it's flavor should be both sweet and savory to complement the salty, meaty fillings inside. A good cemita roll should bear more than a passing resemblance to a great sesame seed hamburger bun.

I mean, just look at this:

cemita-queens-assembly-collage.jpg
Photograph: Vicky Wasik

For a sandwich that massive, that fully-stacked, only the freshest bun will do, and what better way to get a fresh bun than to make it yourself?

In my quest for the ultimate cemita roll, I started by scouring the internet and testing out a half dozen recipes. Turns out that most recipes are quite similar. You make an enriched dough with flour, sugar, salt, yeast, milk, eggs, and some form of fat. The dough is mixed, proofed, shaped, re-proofed, brushed with a wash of some kind (either an egg wash or a dairy wash), sprinkled with sesame seeds, then baked. Pretty straightforward, but that doesn't mean I didn't need to do some tweaking to get things just right.

Fat

First up, I had to decide which liquid and which fat to use in my dough. On the liquid front, I tested milk, water, buttermilk, and evaporated milk in various ratios, along with some powdered milk for good measure. As for fats? I tested everything from vegetable oil to lard to butter.

When it came to liquids, the choice was clear. Milk offered the ideal amount of tenderness and flavor, not to mention that it's inexpensive and almost always on-hand. The choice of fat was more difficult. Different fats yield major differences in texture: The vegetable oil-based breads were very delicate and almost cake-like, while the lard- and butter-based breads were tender, while still maintaining some structure. Why was that?

20150422-cemita-rolls-recipe-02.jpg

See, in order for flour to form gluten—the protein network that lends structure to bread—it needs two things: moisture and friction. First the flour proteins need to get wet. Then they need to be rubbed directly against each other either through kneading or a prolonged rest (a.k.a. the no-knead method). Fat inhibits the formation of gluten by coating flour proteins and preventing them from rubbing up on each other too closely. Sort of like a really bad wingman on a singles' night out.

Fats that are liquid at lower temperatures—like, say, vegetable oil—are much more effective at this job, since they coat the proteins virtually as soon as they're added to the dough. Solid fats like butter or lard will still get in the way, but they're far less mobile in the dough as it's being kneaded. Lesson learned: for enriched, fatty breads, the more fat you add, the more tender the dough, and the more solid the fat at room temperature, the less cakey your final bread will be. For my money, the version made with milk and butter had the best flavor and texture.

But there was another question burning in my brain: I'm adding milk and butter. That's really just a mix of water and milk proteins with a very high concentration of butterfat. What else has those three exact things in a very similar ratio as its main constituents?

That's right: heavy cream.

Heavy cream is, in fact, so similar to a mixture of milk and butter that with a powerful enough homogenizer, you can actually make heavy cream by emulsifying butter into milk. So why not just use cream instead of both milk and butter in my bread?

20150422-cemita-rolls-recipe-04.jpg

I knew that more liquid fats would make for overly tender breads, but I had a hunch that fats locked into an emulsion would have an even harder time coating flour particles and preventing the formation of gluten. That hunch turned out to be right. Not only does a cemita roll made with cream in place of milk and butter have no problem forming a good, tender-yet-robust gluten network, it also makes for a far more foolproof recipe with fewer ingredients.

I sometimes find even brioche—a butter-enriched bread—to be too cake-like. I may well start using heavy cream in my brioche recipes as well, for that rich, buttery flavor coupled with a moister, stretchier crumb.

Kneading

Next, I nailed down the basic ratios of the bread to come up with a dough that's quite sticky but still easily workable. The moister the dough, the more the buns will puff and rise as they bake and the lighter and more tender the crumb will be. The ratio I landed on was eight ounces of heavy cream to 12.5 ounce of flour, two eggs, a teaspoon of instant yeast, three tablespoons of sugar, and one and a half teaspoons of kosher salt. But what's the best way to combine them?

I'm a typically huge proponent of the no-knead method. It involves combining all the ingredients in a bowl, covering it, and letting it sit undisturbed. As the yeast slowly works its way through the dough and produces bubbles of carbon dioxide gas, those bubbles in turn stretch and knead the dough very, very slowly on a microscopic level. With a normal lean dough made of flour, water, and yeast, this process can take overnight at room temperature.

An enriched dough with eggs and fat, on the other hand, has to be kept refrigerated until baking, which can extend that no-knead time up to 48 hours. It works,* but that's a long time to wait for sandwich perfection. I decided to stick to regular kneading. For small batches of moist dough, I prefer using the food processor over a stand mixer. It makes very rapid work of the process, producing a great, stretchy gluten network in moments with very little chance for the dough to oxidize or develop the off-flavors that can occur in over-mixed doughs.

*and if you're not squeamish, I can also confirm that letting the dough sit out at dangerous room temperature levels for 12 hours also works.

20150422-cemita-rolls-recipe-03.jpg

The rest of the process was a snap. If you're a regular Serious Eats reader, you'd know that the very best way to measure ingredients for baking is to use an accurate digital scale, adding your ingredients one at a time and taring the scale in between additions to confirm amounts.

I start by adding 12 1/2 ounces of flour (that's anywhere from two to three cups, depending on how tightly they're packed. See why you shouldn't use a volume measure for flour?), followed by a couple of whole eggs.

20150422-cemita-rolls-recipe-05.jpg

To that, I add my sugar, yeast, salt, and heavy cream, and then process it until the dough comes together into a very sticky ball. It should only take about 45 seconds or so until the dough sticks to itself enough that it'll begin riding around the blade.

20150422-cemita-rolls-recipe-06.jpg

Next, I transfer the dough to a large bowl for its initial proofing phase.

20150422-cemita-rolls-recipe-07.jpg

During this phase, yeast will start to digest carbohydrates in the flour, producing carbon dioxide gas to leaven the dough along with other flavorful by-products. At the same time, natural enzymes in the flour will begin to break down proteins into smaller pieces, which in turn aids with browning and flavor development later on.

20150422-cemita-rolls-recipe-08.jpg

Once the dough has risen to about one and a half times its original size (this takes about four hours; it's such a rich, delicate dough that it never really doubles in size the way a highly elastic lean dough will), I sprinkle it with flour and turn it out onto a work surface.

20150422-cemita-rolls-recipe-09.jpg

I reshape the dough into a ball just so that I can more easily divide it with my bench scraper.

20150422-cemita-rolls-recipe-10.jpg

Using a bench scraper, I split the dough evenly in half. When dividing dough, it's always a good idea to do the major divisions first, instead of trying to estimate how big each individual piece should be from the very start. For instance, if I want 16 pieces of dough in the end, I'll divide it in half first, then divide each half in half to make quarters, then each quarter in half to make eighths, and finally each eighth in half to make 16 even pieces.

A good bench scraper is invaluable for this step, not just for dividing, but for reigning in any loose flour or sticky dough.

20150422-cemita-rolls-recipe-11.jpg

In this case, I'm going for sixths, so I divide the dough in half, and then divide each half into thirds.

20150422-cemita-rolls-recipe-12.jpg

To form balls, I start by rolling each chunk of dough around on a relatively clean section of my work board. The idea is that I want it to stick just a little so that it rolls around instead of sliding.

20150422-cemita-rolls-recipe-13.jpg

Once it's formed a ball, I pick it up and stretch the top downwards, pinching it together at the bottom in order to form a very smooth, tight, top surface.

20150422-cemita-rolls-recipe-14.jpg

Th dough balls go onto a parchment-lined rimmed baking sheet for their final rise, which takes about an hour. This second rise is to add back any air bubbles that were deflated in the process of dividing and shaping the balls for baking. We want the dough to be airy and relaxed enough that it'll poof up dramatically as it bakes in the oven.

20150422-cemita-rolls-recipe-15.jpg

To prevent the top surface from drying out, I cover them with plastic wrap that I weigh down with a clean kitchen towel.

20150422-cemita-rolls-recipe-16.jpg

Just before baking, the dough balls get a brush of egg wash. I tried various other liquids like milk and heavy cream, but an egg wash produces not only the best color, but also makes it easiest to get sesame seeds to adhere.

20150422-cemita-rolls-recipe-18.jpg

For several rounds, I'd been baking these cemita buns with just the classic sesame seed top, but I always felt like they could just just a bit more flavor. A sprinkle of coarse sea salt was the ticket, giving a crunchy burst of salt to contrast their sweetness as you bite through them.

20150422-cemita-rolls-recipe-19.jpg

They take about 15 minutes in a hot oven to bake through, and while you may be tempted to grab them and start eating immediately, their texture improves significantly as they cool a bit, especially if you then split them and toast them in butter before making a sandwich.

20150422-cemita-rolls-recipe-21.jpg

And there she is. Isn't she a beaut? Tender and rich, but substantial with lots of buttery flavor (despite having exactly zero butter in the ingredients) and a crumb that can stand up to the biggest pile of toppings (or the juiciest burger, if that's your preference). These are gonna become my sandwich rolls of choice for the summer. Give 'em a shot and I'm pretty sure they'll become yours too.

The only real question remaining is, which of your guests is going to come first after they hear about the hand-held feast you're putting on?