SCIENCE OF THE KITCHEN
by Jack Lang
- WHY COOK
- WHAT HAPPENS WHEN WE COOK MEAT
- An Experiment
- The Structure of Meat
- The Effect of Heat
- Conductivity of Meat
- To Cook A Joint
- Wet Heat
- Making Stock
- COOKING WITHOUT HEAT
- SUMMARY TABLE OF TEMPERATURES
The Science of The Kitchen module of the eGCI will be formed of a number of courses to be published over the coming semesters and will give a brief overview of some of the science behind food preparation. Areas to be covered include: the effects of heat on muscle, protein, carbohydrates and fat; sources and transmission of heat; browning; fermentation and preservation.
Understanding the processes behind the transformations that food undergoes as it cooks may give insights in how to cook, and in the hands of inspired chefs lead to new dishes, or better ways of cooking old ones. Some have dignified this with the title "Molecular Gastronomy."
This is very much a high-level view, and much of the detail is glossed over or left to the references. No specific technical knowledge or mathematics is assumed. Although the science is, by its nature universally true, the material here is aimed at the home cook and restaurant chef as well as the cookery (culinary) student rather than at industrial scale processes.
Why cook at all? Why not just eat food raw, as some advocate? One reason is that it tastes good. Few can resist the smell of fresh toast or frying bacon in the morning. It tastes and smells good because that is our body's way of saying it is good to eat, suitable for our digestive systems and (probably) won't poison us. In ancient times, getting this right was important for survival.
Our digestive systems have evolved to only deal with certain foods. We don't chew the cud like cows and other ruminants and can't usefully digest cellulose, so eating grass is not particularly nutritious. Ancient humans appear to mostly have eaten fruit and seeds with the occasional bounty of whatever else they could find or catch such as eggs, but basically we are originally carrion eaters.
This leads to the second reason to cook; it makes food more digestible, allows us to eat a wider range of food, and releases nutrition locked up in the raw material that we otherwise could not digest. Our success as a species owes a lot to being able to utilize a wide range of foods because we discovered cooking.
The third reason is food safety. Cooking processes kill or denature bugs, some poisons and other nasties that come with the raw food. Although less important today (most of the food we buy in supermarkets in the civilized world is safe to eat raw), bugs are everywhere in our environment. There are still reports of salmonella endemic in chickens, and occasional reports of trichinosis, tapeworms and other parasites in pig meat.
Reflecting the increased safety of our food supplies, today's cooking is much lighter than that of our forebears. A rare steak is, as we shall see, essentially uncooked in the centre, and certainly not cooked enough to kill spore-forming bacteria such as those that cause botulism or certain parasites.
We start by looking at cooking meat. You may choose not to eat meat, but it has always been an important component of the human diet and certain nutrients are only naturally available from it.
Meat (and the edible portion of fish) is mostly muscle. A joint of meat is a section of a large muscle or muscles, together with associated fat, bones, and other structures such as nerves and veins.
You can do this at home as well, although it's a terrible thing to do to a decent bit of steak. You can always add the leftovers to the stock pot. You do have a stock pot, don't you?
Take a nice piece of steak. This is filet, but any will do.
Divide it in half. We will use one half in a moment. Divide the other half into eight or nine cubes, about 1cm/˝ inch per side.
We are going to cook each of these cubes at a different temperature, in 5C/10F steps from 45C/110F to 75C/170F. The easiest way to do this is in pan of water, since we can control the temperature more easily and get more even heating. Use a remote reading digital thermometer. A digital thermometer is the one gadget above all that will do the most to transform your cooking.
Heat the water to the desired temperature. Put in the meat. Leave it until the temperature of the meat has stabilized at the desired temperature.
Here is a cube of meat in a pan of water.
Here are the results:
|C||F||FEEL||JUICE WHEN CUT?||COLOR||COMMENTS|
|45||110||Soft||No||Dark red/blue||Still raw|
|50||120||Soft but firmer||No||Dark red/blue||Blue|
|65||150||Firmer||Yes||Some red juice||Medium|
|70||160||Firm||No||Mostly Grey||Well done|
What is going on here? How is a juicy steak transformed into a rubbery tough grey lump?
Muscle is a wonderful piece of natural engineering. It is both strong and flexible at the same time. It consists of bundles of active fibers held together and anchored with a natural glue, mostly collagen, and lubricated with pads of fat that also act as natural dampers.
The fibrous nature gives meat it texture. Meat is usually carved across the fibers, chopping them into short lengths, since this makes the meat easier to chew and digest. (Fish muscle has much shorter lengths of fiber, divided by flexible sheets, and so is more delicate and cooks at a lower temperature (45C/130F)).
Heat basically screws up the elegant structure, bursts the cell walls, and disrupts the delicate chemistry. The mobile and flexible strands of protein shrink and tangle, squeezing out the lubricating (and tasty) fluid. As the temperature increases these tangles get tighter and firmer. Your steak gets smaller and tougher. The lubricating and flavorful juices separate from the tough dry meat. These juices in the old days were called the "osmazone" which the famous eighteenth century gourmet, Brillat Savarin described as "The soul of the meat."
The color changes because the red pigment, mostly myoglobin, is turned to its grey form. Myoglobin is the muscle's equivalent of hemoglobin, the oxygen carrying molecule in the blood. Myoglobin carries the oxygen to power the muscle's complex chemical engine. Muscles that work more often, like leg muscles in chicken, tend to have more myoglobin, and so are darker. Fish have two sorts of muscle: the large white (or pink if the fish easts shellfish) muscle, which is the massive engine that used only occasionally to attack or get out of trouble, and the small dark muscles used most of the time to swim slowly around.
Like hemoglobin in blood myoglobin can exist in several forms with different colors.
The central atom that binds to the oxygen is iron so the color of the molecule follows the color of iron compounds. The oxygenated form, oxymyoglobin, is bright red. The de-oxygenated form is dark purple/blue. The oxidized form, metmyglobin is brown, like rust. The changes between these forms are reversible. Some chemicals bind more tightly to the iron and so cannot be reversed.
The most common is nitric oxide, to give the pink form nitro-myoglobin, which is the color of cured meats such as ham or bacon. The saltpetre used in curing (and in making gunpowder) is potassium nitrate, KN03. Other chemicals, such as sulfur, can bind irreversibly to myoglobin to create other colors, such as green, but these are not usual in cooking, except as indications of contamination or bacterial action. A more complete list is given in the table.
|Fe++||Ferrous (covalent)|| || |
| ||:H2O||Purple||Reduced myoglobin|
| ||:NO||Cured pink||Nitro-myoglobin|
|Fe+++||Ferric (ionic)|| || |
Because we associate freshness with bright red meat, the meat packaging trade uses all sorts of tricks, such as feeding the animals with vitamin E, and using oxygen permeable films to keep the meat looking red. Properly hung meat should be the dark purple/brown of reduced myoglobin, since the biological processes in maturing will have used up the oxygen.
Heat breaks up (denatures) the complex myoglobin molecule to create the greyish denatured hemochrome. This occurs at a slightly higher temperature than that at which the proteins denature, roughly 65C/150F, so a grey steak is an overdone steak.
Meat conducts heat very poorly. We can use the myoglobin color change as a rough thermometer to watch what happens as we cook a steak on a hot pan.
You will see as it cooks that it shrinks the piece of meat gets smaller and tougher. Even after 20 minutes, the heat has only penetrated an inch or so. The heat also melts the fat interspersed with the muscle which runs into the pan, along with the meat juices and is lost.
Steak after 25 minutes…burnt on one side, but still raw on the other.
Solid meat is not a very good conductor of heat. The actual conductivity is quite complicated. For example, it is different along the grain or across it. It also changes with temperature; frozen meat conducts less well than unfrozen meat. Meat conducts better as the fat melts, and as the proteins denature, and as the water moves more freely. The length of time a piece of meat takes to cook (or freeze) is governed by the maximum depth, rather than the weight. A rough rule of thumb is about 20 minutes for each inch of depth. Thin pieces cook faster than thick joints; meat for quick stir frying needs to be cut wafer thin.
If the object is to raise the internal temperature of the meat to 60C/140F, then why put it in a 200C/400F oven, overcook the outside, undercook the inside and hope they even out during the resting period? They won't. The heat changes are irreversible. All that happens is the inside cooks a bit more from the retained heat and some of the juices squeezed out from the middle migrate a bit to the outside.
A much better idea is to put the meat in a much lower oven, say 65C/150F, for a long enough time for the entire joint to come up to temperature&emdash;about 5 hours. This also has the beneficial effect of letting the collagen denature to gelatin, as we shall see later. The meat will be uniformly tender, juicy and delicious. People will marvel. There is no danger of overcooking, so timing is not critical.
This is well above the temperature where harmful bacteria can thrive, despite what the "Food Police" say, who just look at absolute temperature. Recent work has shown the length of time at any particular temperature is important, and lower temperatures for longer times are also safe. Unfortunately this has not yet been incorporated into the regulations, so meat at public functions is often still served overcooked by law. If you need to hold meat for extended time (more than two hours) between cooking and serving you are advised to hold it at at least 60C/140F.
The oven needs to be at about 65C/150F, slightly above the desired temperature. This is because of the second law of thermodynamics: heat flows from a hotter body to a cooler body. As Flanders and Swan remind us, this means we are all going to cool down, and so there will be no more heat and perfect peace…
The second law says that the flow of heat into the meat will be proportional to the temperature difference between the outside and the inside of the meat. For meat, the thermal conductivity, k, is about k = 0.08 + 0.0052w, where w is the water content, and the units are watts per square meter per degree. Mathematically:
- Q= k*A*(?T)/x
- Where Q is the rate of heat transfer in watts (Joules per sec.)
- K is the thermal conductivity as above
- A the surface area in square meters
- X the depth in meters
- DT the temperature difference casuding the heat flow
However, the heat flowing in will increase the temperature of the inside. One Joule raises 1cc of water by 1 degree centigrade. Since the inside is hotter, the temperature difference is less and so less heat will flow in. Mathematically this means that the temperature will rise in an inverse exponential, asymptotically to the temperature of the oven. The graph shows an example
For a typical small joint, say 2 inches thick we can model the temperature rise at the center for various temperature ovens:
You can see from the graph the inside is cooked rare (60C/140F)
- After about 45 minutes in a hot 200C/ 400F oven, leave it another 15 minutes and it will be 75C/170C, gray, dry and tough. Worse, when the inside is cooked, the outside half inch will be at 100C/210F or very overdone.
- After about 110 minutes at 100C/210F, the outside will be at 83C/180F. At this temperature the center will be overdone after another hour, so timing is less critical.
- After 5 hours at 65C/150F, the outside will be at 64C/ 148F, a little more cooked, but not significantly so. Leaving it there another hour or even several will have little effect – the curve is almost flat at that point.
Authorities differ on the thermal conductivity of bone. McGee states that bone conducts more than the meat, Wolke says it less then the meat. The measurements I have been able to find also vary widely, with values for k from half to double those of the meat. In the context of roasting or freezing a joint, I'd say that bone, with its cellular structure, would conduct less well than meat. If it conducts poorly, it may also explain why the meat next to the bone is meant to be better. If the joint is roasted conventionally the poor conduction of say the bones of a rib joint will tend to protect the meat next to it from the heat, overcooking and drying.
To cook a joint, first brown the outside either in a hot pan, with a blowtorch or for 10 minutes in a scorching hot oven. Maillard reactions are the chemical processes involved in browning and developing the typical flavours of roast meat and will be covered in detail in a later course. This browning is entirely for taste. It has been shown, contrary to popular belief, that it does nothing to seal in the juices. If you don't overcook it, you won't get any significant juices – they will still be in the meat adding their flavor and juiciness, not squeezed out when the proteins abused from overcooking curl and contract.
You can alternatively brown the meat after it has been cooked, and for some varieties such as pork or the skin of birds, this crisps the outside. However, for beef or lamb, I prefer to do it beforehand. It avoids any danger of overcooking and I can do it at leisure rather than at panic time.
Leg of lamb cooked at 65C/150F for 7 hours Gigot a sept heure. Internal temperature 60C/ F.
Beef done the same way. (Prime Scottish Rib, boned)
Brown on all sides. This is for flavor– it does NOT seal the meat. Note thermometer probe sticking out of the side. The browned onions are for the benefit of the gravy.
After 6 hours in a 65C/150F oven. Final temperature is 55C/130 F – rare.
Since the meat has lost very little juice, there are almost no pan deposits, or pink juice when you cut the meat.
Perfect Roast Beef for Sunday lunch. Good all the way to the edge.
Succulent, and moist.
If some of your family demands it well done, either cook off their pieces separately in a hot pan, or cook a separate piece for them to a higher temperature, say 70C/160F. It will be tougher and drier, but to each their own.
Stewing, boiling, or braising uses wet heat, and another mechanism comes into play. The glue that holds the muscle together dissolves slowly. This glue is made up of different proteins, primarily collagen. Collagen consists of three strands of molecules wound around each other. The older the animal, the greater the amount of collagen. Likewise, the more active the muscle, the greater the amount of collagen.
With long, slow cooking, those strands will unwind and turn to soft, succulent gelatin, providing the juiciness to tough cuts of stewing beef like oxtail and shin This however is a comparatively slow process. If you cook it long enough for all of the collagen to turn to gelatin, and hot enough for the contraction of the meat to squeeze out the liquid, you have just the cooked meat fibers. If they have been overcooked, you are left with irretrievably dry and stringy meat. Even if it is swimming in liquid, you can't get that juiciness back into the fibers that the curled up proteins have squeezed out.
A cube boiled for 30 minutes at 75C. The collagen is beginning to dissolve.
Collagen starts to turn into gelatin and dissolve at around 60C/140F. This process (and also the fat melting) takes energy. Experienced BBQ cooks know that during the long slow smoking of brisket there is a "temperature stall" at around 72C/165F, where the internal temperature, instead of continuing to climb, stays steady for a long time before increasing again. That is the period the collagen is converting to gelatin. Once the temperature starts to climb again the conversion is complete, and the meat is tender. Any more cooking tends to dry the meat without improving tenderness.
Heston Blumenthal says that softening the collagen also improves even normally tender cuts of meat, such as the roast beef above. He suggests holding the temperature of the beef for up to 10 hours at 55C/130F (longer will start to generate 'off' flavors) to make beef that is "unbelievably tender."
Wet cooking is appropriate for the tougher (but flavorful) pieces of meat that have a lot of connective tissue. The long, slow moist cooking melts the toughness into smooth unctuousness. The tougher pieces of meat can stand the long cooking and will become tender as the collagen dissolves. Even so, although they are fairly tolerant, overheat them and they too will fall apart into dry, tough shreds. These joints have much more flavour than the softer joints
For stock making, we want to extract the maximum goodness from the meat. We want the contracting proteins to squeeze out the interstitial liquid into the stock and the collagen to turn to gelatin and dissolve. The gelatin dissolved in the liquid gives the stock body and mouth feel, so stocks need to be cooked hot&emdash;boiled. However, vigorous boiling will emulsify the fat into the stock, which creates a cloudy stock. Besides, fast boiling is only a few degrees above a gentle simmer. Thus the traditional advice to make good stock is to do so at a gentle simmer for a long time.
An alternative approach is to use a pressure cooker. A pressure cooker allows the temperature of the stock to be raised before it boils, and the pressure reduces the turbulence. It also seals in all the flavor. So for the ultimate stock, cook in a pressure cooker, at maximum pressure for an hour or so, adjusting the heat input so that steam barely escapes from the valve. Use tough, cheap cuts of meat but leave the bones out. Be sure to let the pressure come down on its own after removing from the heat (as opposed to any of the quicker methods such as holding the closed pot under cold tap water, or releasing the pressure valve)otherwise the contents will boil aggressively when the lid is opened, making the stock cloudy.
Collagen is also the main component of tendons and is also the glue that holds bones together, so, given long enough wet heat, they start to dissolve as well. Cook a stock for too long, and the calcium from the bones also begins to dissolve, giving a "bone taint."
Thus for cooking meat (and other muscle):
- Light dry cook to 55C-65C (130F-150F) (45C/110F for fish) to just set the proteins
- Long slow cook to dissolve the collagen,
- Long hot wet cook to make stock.
An example recipe indicating long, slow wet cooking is Steak and Kidney Pudding. Pudding, not pie. Pie is just a stew with a pastry lid. In Steak and Kidney Pudding, the meat is sealed in a suet crust and boiled for six or more hours to melting, tasty, meaty loveliness. To quote Dr Marigold (one of Charles Dickens' more obscure characters), describing his pudding; "A beefsteak-pudding, with two kidneys, a dozen oysters and a couple of mushrooms thrown in. It's a pudding to put a man in a good humour with everything, except the two bottom buttons of his waistcoat."
DR MARIGOLD'S PUDDING
Serves 12. Good, cheap eating.
For the pastry:
8oz/250g shredded beef suet
1 tsp black treacle/molasses (gives the pastry a golden color and taste)
Cold water to mix
Line a large (2pt) pudding basin ( or heat-resistant bowl). Reserve 1/3rd of the pastry for a lid.
2.5lbs/1Kg stewing beef cubed. Skirt steak is good.
1lb/500g chopped beef or veal kidney
A dozen oysters, or a can or two of smoked oysters
A couple of large Mushrooms cut up, or even better, dried morels
2 onions, chopped and softened
2 Tbs flour
Salt and pepper;
1 Tbs Worcesteshire sauce
Easy on the seasonings. Since everything is sealed in, the flavors intensify.
You can, I suppose, omit the kidneys and the oysters, but it will not be as rich. You can use anchovies instead of oysters, but watch the salt level.
Mix well and pack into the lined basin/bowl. Fill with a little stock or water, but there won't be much room for liquid.
Cover with the reserved pastry
Tie a piece of greasproof paper over, leaving a fold for expansion. Hint: easiest if you secure it with a large elastic band before tying with string. Don't forget to leave a loop of string over the top, tied on both sides as a handle to help get it out of the hot pan after cooking.
Put in a pan of water. Add a cut lemon to the water to protect the pan.
Simmer (or rather not quite simmer) for 6 to 12 hours.
Once simmering, it may be easier to put the whole pan in a low oven (90C/200F) for most of the cooking. Check the water level occasionally and if needed top it up to stop it boiling dry.
Turn out into a deep dish, as there will be lots of gravy. Unfortunately I did not manage to snap this step before the hungry guests got at it. Serve with brussel sprouts, and mashed potatoes.
The meat is deeply flavored and spoon-tender. The kidneys and oysters have combined to form the most wonderful sauce.
There are other methods to make meat tender enough to eat besides cooking it. The most obvious way is mechanical: mince it up into small pieces, or slice it very very thin across the grain so that the muscle fibers are chopped up and can be easily chewed. Thinly sliced raw meat examples are carpaccio and delicacies like Parma ham.
The extreme example is Steak Tartar, said to be named after the practice of the fierce Mongolian and Tartar warriors who softened their steaks before eating them raw by putting them under their saddles before riding off on raiding or war parties. These days you don't need a horse to prepare steak tartare. The steak is chopped fine (but not into a puree). Its interesting to note that tartar sauce was originally a sauce for steak tartare.
A less extreme example is hamburger (and sausages) where the fine mincing makes otherwise tough cuts tender enough to eat with only short cooking times. Of course, flame browning the outside but otherwise not cooking your hamburger over 55C/140F, but holding it at this temperature for several hours, will immensely improve it.
Protein is also degraded by acid. Ceviche (fish "cooked" in lemon or lime juice) is an example. Meat proteins can also be digested by various enzymes: papain, from unripe papayas, Bromelin from pineapples, ficin from figs. The effect has been known for thousands of years. Primitive tribes wrapped their meat in papaya leaves. These enzymes are destroyed by heating above 70C/180F so have to be from fresh fruit and are not present in pastaurised or tinnned juice. You can buy them as "tenderizing salt" where the extracted enzyme is mixed with salt. In use they must be injected or otherwise put into the meat. Just soaking tends to only tenderize the surface.
It's easy to try: soak a steak in some fresh pineapple juice. Pierce the steak with a fork in lots of places to let the juice get inside.
Here is the steak and the pineapple
I pureed about an inch of pineapple, and soaked half the steak in the puree
After about three hours. The half on the left was not soaked:
The enzyme in the fresh pineapple puree has eaten the steak. The steak fell apart as I tried to lift it, like tearing damp blotting paper.
Personally I don't like the effect; I think it turns the meat pasty and mushy, with odd hard bits of tendons that have not been softened. Apparently pineapple softens lentils as well. Pineapple jelly is difficult to make with fresh pineapple and gelatin as the bromelin denatures the gelatin. If you want to make pineapple jelly you must use agar, or cook the pineapple to destroy the enzyme.
Enzymatic tenderizing also occurs when meat is hung. The enzymes (and the lactic acid) naturally present in the meat slowly break down the long muscle fibers, making the meat more tender. The process is called proteolysis, which means breaking down (fragmentation) of the protein strands. The enzymes which carry out the process are thus called proteoses, meaning that they fragment the protein strands.
Many complex processes including initial relaxation of rigor mortis happen while the meat is aged, but the primary two are the enzymatic breakdown and dehydration. Freshly killed beef is not very nice, and doesn't even taste much of beef, but rather metallic. Aging increases the beefy and "gamey" flavors as a byproduct of the natural breakdown of the meat, friendly microbial and bacterial action and Maillard reaction processes on the surface of the meat. The dehydration (beef is 70% water, and loses some 20% during aging) concentrates the flavor.
Traditionally beef was dry-aged. The entire carcass was hung in a refrigerated room just above freezing at a temperature of between 32-34F/0C-2C, 80-85% relative humidity, and with an air velocity or 0.5-2.5m/s (about 3 miles an hour) for between 21 and 28 days. The low temperature discourages growth of spoilage organisms and the high relative humidity stops the meat from drying out too much. There is typically a 15-20% weight loss. The air movement stops water condensing on the meat. A dark crust forms on the outside of the meat which must be cut off and discarded. The wastage, the loss in weight and the time involved means that dry aged beef is a more expensive product.
You can reproduce these conditions at home by keeping a large piece of meat in the refrigerator, uncovered, on a wire rack with a drip tray underneath it with water in it. However, a professional meat storage room is a better place to do this and your friendly local butcher may be persuaded (at your risk, and if you pay in advance) to hang your beef for you for that extra week or two.
When aging beef there is always the danger of spoilage, which increases the longer the beef is aged. Discard it if there is any hint of an unwholesome smell, or sliminess.
The change in the meat is most noticeable in the first 14 days, and then the effect decreases, with little change after 21 days and none in tenderness after 28. Some advocate hanging for much longer, such as 90 days. Others feel that in this length of time the meat generates "off" flavors, and there is a significant risk of spoilage.
In order to overcome the problem of spoilage and length of time taken, the meat industry has developed various other aging techniques. These include accelerated aging where the beef holding temperature is higher, but bathed in ultraviolet light to retard the microbial growth. Anther technique, now used almost universally, is wet aging; the beef is cut into joints, sealed wet in aseptic cryovac bags and held at 32-34F/0-2C. Although the enzymatic processes are the same, there is no dehydration and consequent weight loss, nor Maillard conversion of the surface, so that the taste is different. However, there is much less risk of spoilage with wet aging so less waste and easier handling.
Brining is soaking the meat in brine, typically 1 cup each of salt and sugar to a gallon or water before cooking. It can immensely improve bland meats like turkey or pork.
It works because the fluid inside a meat cell is saltier than the outside, so by osmosis additional water is sucked in, along with some of the sugar and any flavoring you have added. Of course, if you overcook the meat and squeeze all the water out, your good work will be undone.
Penetration of the meat happens quite slowly, typically 1cm/hour. An advantage is that when submersed in the brine or marinade, which is usually quite acidic, nasty bugs are kept away from the surface, and the meat can be stored for a long time in a conventional fridge.
The illustration shows the diffusion of a marinade into a meatball, imaged using MRI scanning, courtesy of the University of Cambridge.
- Harold McGee:
- Harold McGee:
- Jeffrey Steingarten:
- Jeffrey Steingarten:
- Robert L. Wolke:
- Shirley Corriher:
- Peter Barham:
Course Editors: Carolyn Tillie & Andy Lynes
HTML Formatting: Andy Lynes (updated by Chris Hennes, 10/11/2010)