The Food Lab: The Science of Pie Dough
It's time for another round of The Food Lab. Got a suggestion for an upcoming topic? Email Kenji here, and he'll do his best to answer your queries in a future post. Become a fan of The Food Lab on Facebook for play-by-plays on future kitchen tests and recipe experiments.
If there's one thing that instills fear into the hearts and minds of American cooks, it's pie crust. I know. At one time, I was one of those people. Pie crusts were the Mumm-ra to my Lion-O, and it was all because they were a mystery to me. What makes them flaky? What makes them tender? And most importantly, how come mine used to come out like pliant pieces of leather instead of buttery and delicious?
What I'm after: The kind of crust that's substantial enough that it doesn't sog-out from a juicy filling but tender enough that it flakes in your mouth into buttery shards. A crust with substance, but not chew. A crust that divides along deep faults into many distinct layers separated by tiny air spaces and that cracks when bent. A crust that is never leathery or pliant, but not so tender or crisp that it crumbles instead of flakes. And of course, it should have a deep butteriness coupled with a balanced sweet and salty flavor.
Easier said that done, right? For many people, making pie crust is a crap shoot. Sometimes it comes out perfectly flaky, other times tough. Sometimes you need just a couple tablespoons of water, sometimes a full 1/2 cup. What gives?
Turns out that the science of pie crust is really not all that complex, and once you get a grasp of what's really going on in between those flaky layers, then making a perfect crust becomes a matter of smarts, not luck.
Life of Pie
On paper, a classic American pie crust is a mind-bogglingly simple recipe. Combine flour with a bit of salt and sugar, cut in some butter and/or shortening, then add just enough cold water to get it to come together into a disk. Roll it out, and bake. That's it.
The underlying difficulty in the technique comes during the first stage of cutting the butter into the flour. It's simply impossible to accurately cut butter into flour to the same degree on a consistent basis. Don't cut it in quite enough, and you need to add extra water to absorb the excess dry flour, resulting in the over-formation of gluten, and a tough, leathery crust. On the other hand, cut it just bit too far, and you end up with way too little dry flour. Rather than having well-structured flaky layers, you end up with a crust that crumbles into sandy pieces.
This is the reason why you sometimes need to add a couple tablespoons of water and other times up to twice as much to get the dough to come together—it's got nothing to do with the relative humidity of the air as many books will have you believe. Indeed, in the short time that it takes to make a pie crust, flour will absorb approximately 0.1 percent of its weight in water, even in the most humid of environments. That's a small enough amount to effectively be zero.
So how does the simple action of cutting butter into flour result in layers and layers of flaky pastry? It's all got to do with the balanced interaction of gluten and fat.
Old school pastry books will tell you that when you cut butter or some other solid fat (like shortening or lard) into flour, what's happening is that you are encasing pockets of flour inside a shell of fat. Add water, and the flour is moistened, whereupon gluten—the network of proteins that lend structure to baked goods—is formed. When you subsequently roll this dough out, these pockets of fat stretch and stretch, eventually forming sheet of fat that separate sheets of gluten-enforced flour. Then, as the pastry bakes, the fatty layers melt, allowing the floury layers to separate from each other, solidify, and form the layers you see in a great pie crust.
It makes sense. Sort of. Unfortunately, it's not a particularly accurate picture of what's going on. For starters, how could the action of cutting a solid fat into a relatively fluid mass of flour possibly cause it to coat pockets of flour in distinct bubbles? And even more importantly, if the fat is really coating these pockets of dry flour, then how would they get moist when you add water to the mix? Wouldn't the fat prevent any water from reaching the flour?
How Pie Crust Really Works
Let me digress for a moment. A few years ago, I developed a pie dough recipe while working at Cook's Illustrated (you can find that recipe here). The one trick that got majorly hyped up about it was the inclusion of vodka in place of some of the water in the recipe. It was a pretty neat trick, if I do say so myself, and it solved one of the major problems people have with pie crust.*
* Unfortunately, due to some legally binding document I signed, ironically, as the creator of the recipe but not the owner, I am now the only person in the universe who is not allowed to write about it. Never mind that. You can read about the science of it over at The Kitchn. Gotta love lawyers.
But there's a good chunk of that article that seemed to have gotten glossed over by pretty much everyone, and I believe that it's a far cooler part than the headline-friendly vodka trick, and it has to do with the basic structure of dough.
You see, it turns out that when it comes to pie dough, our existing model has it wrong. In fact, it's not the fat that's coating pockets of dry flour. It's the reverse. It's the flour that's coating pockets of pure fat. With this model, things make much more sense. You can easily and intuitively see how fat gets coated with flour (think about dropping a pat of butter into a pile of flour, but on a much smaller scale), and with this model, when you add water, you are indeed moistening dry flour so that it can form sheets of gluten.
But there's a third element at play here that's been ignored: the fat/flour paste that forms at the interface between the pure fat and the dry flour. With this paste, what you're getting is essentially single particles of flour that are completely coated in fat. Since they can't absorb any water, they end up behaving in much the same way as pure fat.
To summarize, here's what we've got as we're forming a pie dough: Dry flour, completely uncoated by fat, that absorbs water when you add it, forming gluten that then gets stretched out into wide layers. Pure pockets of fat that will flatten out into long, wide, thin sheets as you roll out your dough, separating the layers of gluten-enforced flour from each other so that rather than forming a solid, leathery mass, they separate and gently puff as they bake. A flour/fat paste that functions much in the same way as pure fat does. As it bakes, the fat melts and a tiny amount of individually separated flour bits will deposit themselves and become incorporated into the gluten-enforced layers.
And if this flour/fat paste functions similarly to pure fat, doesn't that imply that we can completely replace the pure fat with this paste? Indeed, we can.
Don't believe me? Take a look.
This beautiful looking pie crust was made with the overprocessed dough you see in the image directly preceding this one. (You know, the one that is pretty much a solid paste of flour and butter?) But there's got to be some trick going on. Simply bake that dough above, and you end up with a short, sandy, crumbly cookie-like crust. Not the beautiful flaky layers you see here.
In order to get this crust here, all I had to do was add some extra plain dry flour to the fat/flour paste I'd formed, pulse it a couple times in the food processor just to distribute it evenly, then proceed just like an normal pie crust: I added water, chilled it, rolled it, and baked it.
Any baker would have taken a look at the finished dough and declare it a failure without even bothering to bake it. After all, how could a crust come out flaky if there aren't even visible bits of fat left in it? By classical pie crust definitions, it can't. Yet we've just proven here that it can indeed, and when physical evidence indicates that an existing model is wrong, it's time to modify that model.
Why Do I Care?
So it's all very interesting, but the question here is, why should you care? How does this make forming pie crust any easier?
In two ways. First off, it completely removes the variability of a traditional pie crust recipe. By measuring out a given amount of flour and fat and combining them together until they form a near homogeneous paste and then adding the remaining dry flour to that paste, you are very strictly defining exactly how much flour is used for gluten formation and how much ends up coated with fat. No more trying to visually judge whether your fat is properly cut. No more adding ice water a drop at a time until a dough is formed. The dough comes out the same, every single time.
The other key advantage is that your dough becomes much more pliable. The fat/flour paste formed at the beginning is much softer and more malleable than pure butter is, which means that your final dough rolls out smoothly and easily with little to no risk of cracking like a traditional pie crust, even without the vodka.
How's that for pie flinging?
On Butter, Shortening, and Lard
There's still a number of questions that come up in regards to pie crust. Which fat makes the best crusts is prime stomping grounds for eternal debate between bakers, and it comes down to a battle between texture and flavor.
- Pros: Excellent flavor, forms distinct, large flaky layers.
- Cons: Difficult to work with. Butter melts at a relatively low temperature (below body temperature) and has a very narrow workable range. It's also got a relatively high water content (about 15 to 17 percent), which can cause excess gluten to form and turn your crust leathery if you aren't careful.
- Pros: Very easy to work with, produces crusts that are extremely tender.
- Cons: It has no flavor at all other than grease. It's also soft over a wide range of temperatures, greatly increasing your chance of overworking your dough and turning it crumbly instead of flaky.
- Pros: The best for working with—it has a wide workable temperature range and is not nearly as soft as shortening within that range. Creates very tender, flaky crusts.
- Cons: Unless you slaughter your own pigs and render your own leaf lard, it's extremely hard to find good lard. The stuff sold in supermarkets has a very piggy aroma, which makes for very piggy-tasting crusts. Most of the time, that's not good.
Because of these various characteristics, most recipes call for some combination of butter and shortening. Because shortening is so much softer than butter at room temperature, it's much more likely to form the fat/flour paste while the butter remains in discrete chunks. But here's the good news: with the method I've outlined above, you can cut the amount of shortening down to virtually nothing. Depending on how tender you want your crust to be, anywhere below a 4:1 ratio of butter to shortening will work. To be honest, most of the time I'll make pure butter crusts, simply because shortening is just not something I keep lying around the house.
Tips To Perfect Pie Crust
Finally, let me offer you a few tips to make the most out of your crust, no matter what recipe you decide to use.
1. Weigh Your Flour
Get yourself a scale, then do yourself a favor and throw out your dry measuring cups. Honestly. Do it. You will not regret it. Weighing dry ingredients by volume is simply not accurate. Depending on how tightly packed it is, the weight of a cup of all-purpose flour can vary by as much as 50 percent. 50 percent!!!. No such problems with a scale. Five ounces of flour (the equivalent of one cup) is five ounces of flour, no matter how tightly it's packed.
2. Use a Food Processor
If there's one reason to own a food processor, it's to make pie crust. Nothing is as efficient or as consistent at cutting fat into flour. A stand mixer will do the job reasonably well; a pastry cutter will, too, though it requires much more work. You can even get a decent crust using just your fingers. But if you've got the processor, use it.
3. Keep Everything Cold
Just like with grinding meat, the key to great pie crust is to make sure that your fat doesn't melt too much. If your home is too hot, make sure that you re-chill your dough as you work with it. In the summer with my oven on, the apartment pushes 80°F. Under these conditions, I'll put my dough back in the fridge for ten minutes immediately after incorporating the fat, for at least 2 hours after forming the dough into a disk and wrapping it, and for another 10 minutes after draping it in the pie plate before trimming and fluting the edges.
The temperature of your water has less of an effect than people lead you to believe. A couple tablespoons of warm water won't significantly chill a pound or two of butter and flour. Still, it doesn't hurt to use cold water or even ice water.
4. Use a Spatula to Incorporate Water
Your processed flour and butter are sitting there in the bowl of your food processor and you're tempted to save yourself a bit of cleanup by just adding the water directly in there, right? Don't do it! Sure, it'll work out ok, but you won't form nearly the same level of flakiness as you do if you incorporate with a spatula. Here's what a crust made 100 percent in the food processor looks like:
Compare that to the crust I made above with the spatula:
See the difference? By using the spatula to fold the water into the crust, you give flaky layer formation a head-start even before you pull out the rolling pin.
5. Use a Tapered Rolling Pin
While the ball-bearing-based heavy-duty cylindrical rolling pin might be better for whacking husbands over the head in cartoons or fitting into Norman Rockwell paintings, the slender, slightly tapered French-style rolling pin offers far more control, is easier to clean and store, and is cheaper to boot.
And that's basically all I know about pie crust. Or at least as much as I'm willing to write before my book comes out. I gotta save something to keep my publishers happy, right?
For a full step-by-step walkthrough including how to mix, roll, transfer, trim, and flute a pie crust, click through the slideshow above.