Not just any soufflé
“Molecular gastronomy” is a term that gets used a lot to describe a certain type of cooking, often with little care for what the phrase might actually mean. In particular, given that this concept’s most high profile advocate in Britain is the rarely away from these pages Heston Blumenthal, it is often used (both in a derogatory way and by those who see their own cooking in this fashion) as a synonym for wacky, unusual or spectacular recipes or food presentation. However, this is not really understanding what the concept of “molecular gastronomy”, the science of culinary phenomena, actually involves. Hervé This is the French chemist who first coined the term and for his followers “molecular gastronomy” means using the scientific method of questioning and testing to inquire into whether traditionally held beliefs and culinary methods really do produce the results that they are supposed to.
Working for the Institut National de la Recherche Agronomique (National Institute for Agronomic Research) This puts old wives’ kitchen tales to the test to understand better how to make better food. This was the subject of his excellent first book to be translated into English, Molecular Gastronomy: Exploring the Science of Flavour. Through both this book and his experiments at INRA, This is probably most famous for his research into the correct temperature to boil an egg, making his work rather more akin to the famous Delia Smith basic advice on how best to boil an egg than he is to somebody preparing a bacon and egg ice cream. Given all of that, it should be obvious that the element that makes Heston Blumenthal a follower of molecular gastronomy is not so much his making a liver paté that looks like an orange, but his pricking holes in potatoes to see if it makes better chips.
I’ve recently come into possession of This’ more recently translated book, Kitchen Mysteries: Revealing the Science of Cooking (a title which is sort of shared by this blog’s subtitle, so you know it’s going to be good), so I decided to see whether This’ research and advice could improve the results in my own kitchen. Obviously whatever I made had to be something fairly tricky to get consistently good results with, otherwise This’ advice might have made no discernible difference. Given his past background in egg preparation the chapter on soufflé stood out.
“How to succeed at soufflé every time?” asks This, “We will obtain good results most reliably if we truly understand what a soufflé is and how its constituent parts react”. Essentially a soufflé consists of a foam of egg whites combined with some form of preparation, in a savoury soufflé this is a béchamel sauce. I decided to make a simple cheese soufflé out of parmesan and gruyeres so I wouldn’t be distracted by adding a complex recipe to an attempt to follow This’ advice.
The most important element of preparing a roux, the basic mix of butter thickened by flour that forms the basis of a béchamel sauce, is for it to be cooked slowly on a relatively low heat, according to This. He points out that starch is useful in thickening sauces as, when it is heated, the energy of the water molecules inside it is enough to weaken the links between the amylose and amylopectin molecules. This establishes hydrogen bonds between starch molecules and water molecules, creating swollen starch granules that This describes as “microscopically enormous”. This is enough to make molecular motion difficult and, thus, a liquid viscous. However, This says: “amylose molecules have only weak thickening powder and a floury taste”. Cooking the flour in butter for a longer time breaks down these amylose molecules into smaller sugars, reducing that floury taste. At 93˚C a roux reaches its maximum viscosity. Allowing it to boil will cause it to reliquefy as this breaks the swollen starch molecules into smaller fragments and increases flow.
So, trying to avoid letting it boil, I slowly melted my butter and mixed the flour in. After making the roux, I added the cheese and some seasoning and spices. The next step is to add egg yolks, having separated the eggs so that the whites could be used to make a foam. In order not to cook the yolks while mixing, This advises to allow the roux to cool before adding them. Adding the egg yolks two by two like animals entering an arc is, apparently, the popular way to proceed with this and that provided a perfect opportunity for This to put a piece of traditional handed down advice to the test. So, he describes, every Sunday he would serve up a cheese soufflé for his friends and each time he would add the egg yolks in a different way. “I obtained the best result when I added the yolks two by two,” he explains, “The cooks seemed right, but the mystery remained (I know now that the two by two trick is no use; I probably just finally learned how to make this soufflé”! With this advice in mind, I felt quite confident adding the egg yolks to my sauce all at once (especially given I was only making two small soufflés, so I wasn’t using more than two yolks in total anyway).
Now came the most important part of all, the egg whites. It is in their preparation, the whisking, that the soufflé rises or falls (both figuratively and literally) so, naturally, This has plenty of advice and investigation to impart on this subject. The essential element here is the introduction of air. The way this works particularly with egg whites is that the ovocumin and conalbumin, the proteins in the egg white, have parts that bond with water and parts that repel it. These create bubbles as the proteins gather around the interface between water and air, spreading air bubbles through the water.
We all know that we shouldn’t let any of the yolk get into the whites, but This can tell us why. It’s because the fats in the yolk bond with the hydrophobic parts of the white, meaning that those have no opportunity to bond with air. This is why you can combine the whites with something that contains the yolks once the whites have been properly beaten, as there is no place for those undesirable bonds to form. Incidentally, this is also the reason why This insists that you can’t use a plastic bowl for whipping egg whites (fats stick to plastic, so fat molecules that remain on the surface of the bowl can have the same effect as a bit of yolk getting into the mix).
The whisking process is what turns larger bubbles in the egg whites into smaller and smaller ones. The smaller the bubbles, the more effectively surface tension becomes a stronger force acting on them than gravity. Thus, better whisking makes a stabler egg white mix. So, I guess the question is: how do you know when you’ve whisked the whites sufficiently?
This tells us that, for those keen to measure, an egg of 3.5 centilitres produces a well whipped white of 15 (so it should more than quadruple in size). More practically, he suggests that removing your whisk and turning it upside down should show the foam attached to it “taking the shape of a pointed tuft, like a clown’s wig”. Finally, he suggests that a sufficiently whisked egg white is strong enough to support the weight of a whole egg in its shell. Being sure to use a ceramic bowl and not a plastic one, I whisked my egg whites up until I could try and see whether there was any truth in this latter advice. Sure enough, once I’d whipped my whites into something suitably strong it really did support the weight of an egg.
Finally, then, came the point where the béchamel and the whipped egg white needed to mix. The difficulty here is in not undoing all the good work of whipping the egg whites, making sure to mix them with the much more viscous sauce without breaking up the foam. To successfully mix the two, This advises, you have to pour the lighter egg whites over the heavier sauce before cutting through both with a spatula and bringing the heavier sauce over the top of the lighter part, turning the bowl to make sure everything is well mixed together. If the mixture appears foamy when you pour it into your buttered ramekins then that seems to be a good sign.
But, how long should a soufflé cook for and at what temperature? And why do you have to make sure that the oven door stays closed? To make a soufflé rise properly it needs to bake at a temperature where the proteins can coagulate before the bubbles explode and the foam collapses, however too high a temperature causes that same coagulation to prevent the interior from rising at all. According to This, anywhere between 150˚C and 200˚C, depending if you want something with a moist centre and dark crust or something more uniform, is an appropriate temperature. A large soufflé can take about half an hour to bake, a smaller one more like 15 minutes. So, I put my two little cheese soufflés in the oven at 180˚C for 15 minutes to see how well that worked.
It is indeed important to keep the oven door closed and the reason why, This tells us, is that before the egg proteins have coagulated to form something rigid enough to support the soufflé, it is the air bubbles in it and the air in the oven that support it. Opening the oven causes a sudden drop in temperature in the oven that causes the air bubbles in the soufflé to contract and the soufflé to fall.
On that subject, This also points out that, scientifically, it doesn’t make sense to say that the soufflé rises because the air bubbles expand in the heat as these can only expand by about 30% and a soufflé may double or treble in size. It is rather because the water evaporates and the vapour enlarges the bubbles, all kept in place as the egg coagulates. To avoid water evaporating from the top of the soufflé preparation it is important that it is heated from below and thus This advises it should be placed on the bottom shelf of the oven.
So, putting all this advice into action, was I able to make a soufflé that worked perfectly? Well, it rose well, but even as I was taking this photo it began to sink. Otherwise, though, I was quite happy with the taste and texture. What final comment, then, does This have to make on the possibility of soufflé sinking and how to avoid it? Well, not a great deal actually. He used his scientific method to assess the value of adding something acidic like vinegar to the egg white mix and placing the soufflé in a warm water bath before cooking and found there was very little discernible difference in the results.
At the end of the day, then, soufflé success remains something that science can only help us so much with. I’ve learnt a lot to enable me to understand how a soufflé works, but, I suspect, it will take a similar weekly experiment to This’ own for me to get the practice to perfect it.
Professor Plum in the Dining Room: A few months ago I had a go at soufflé, using a recipe from the BBC Goodfood website. Slightly different combination of cheeses, but broadly the same result (apart from the one I managed to drop before cooking!). I can never shake the sneaking suspicion that soufflés are easier than people let on, that if you get the egg whites just right everything after that should just come together. Either that, or restaurants cheat and actually make cheesey cakes disguised as souffles.
Of course, if you don’t make them hang around while you take photos, they come to the table much higher!