L'Atelier Cumbria

Wednesday, August 30, 2006

an experiment with onions, acid formation

another experiment with onions.

When onions are cut, they become pungent very quickly. This is because when the cells of the onions are damaged, sulphur containing compounds are acted upon by enzymes (chemical agents) which turn rather innocuous substances into the pungent onion smell we all know. These pungent smells get into the air and eventually some can get into ours eyes, with the well known effect. These substances are acids, and can produce a bitter taste; in many modern recipes, the author usually suggests adding sugar to a dish, to counter this bitterness. If, however, the onions are fried quickly after being cut, the pungent chemicals are not produced in the same quantity, and can be driven off by the heat.
My experiment was to determine how quickly these pungent chemicals were produced. I decided the preliminary experiment should monitor pH and conductivity of the diced onion in pure water, over time. The pH would give an indication of the acidity being produced, and the conductivity would indicate the presence of the acids produced by the onion.

I cut up half a medium sized onion (25g) into 3mm dice and immediately placed them in a beaker which contained 50ml deionised water. A pH probe and conductivity probe were immersed in the liquid, and monitored every two minutes.
The results showed that conductivity increased rapidly for the first 3 minutes, then became relatively constant after 10 minutes, and for the next 20 minutes (which was monitored). The pH was the same, decreasing in the same way.
This experiment showed that the reactions taking place started very quickly, and by 2 minutes were 50% complete. At the end of 15 minutes, I analysed the onion water using a technique called ion chromatography, which separated the acid ions into separate entities.

The graph shows pH and conductivity against time, though the conductivity values shown have been divided by 100 to get them on scale.

What does all this mean for the cook?
I think it means that if we want to cook chopped onions, then one has to put them in hot oil as quickly as possible, if we boil them we should expect bitterness. I believe 3mm dice is about right for dishes which need the onions to be cooked in oil until golden, and would strongly argue against grating onions, which would 'damage' the onion cells too much, producing even more acid.

an experiment with frying onions

Some of you might know I'm a practicing scientist, but I apply my science to food related topics too. I ran an experiment recently, the results of which you may find interesting.

I wanted to see how water is driven off from frying onions, and how much water is left in the oil after frying.
I cut a medium sized ordinary onion (120g) into 3mm dice, and threw it into 100g of hot (120°C) rape seed oil contained in a beaker which was placed on a hotplate (halogen type) which was arbitrarily set so that the onion cooked without browning. Into the oil was placed a thermocouple to monitor the temperature, and I had a balance (scales) to determine weight loss.
I ran the experiment for a total of 70 minutes, noting temperature of the oil and the total weight of the beaker, with contents, every two minutes.

I thought the results were interesting. The temperature settled to 100°C very quickly, as expected, this because it cannot rise above this temperature when free water is present. This situation lasted for about 20 minutes, the temperature fluctuating a few degrees in this time. From then on, the temperature began to rise, a little at first, and increasingly as time passed, until, at the termination of the experiment, it was 141°C. The loss of weight (mass of water) was remarkably constant, at 1.63g/min (2 x sd =0.67), for the first 60 mins, when the rate slowed (as most of the water had been driven off).

The results showed me that for a given temperature of oil, the rate of water loss was constant. Total amount of water loss equated to 87.2% which I think agrees well with literature values (I need to check). The onions did not brown for the first hour, except for a few bits which were above the oil level (nearly all the diced onion was covered with oil, not something we would normally do when cooking golden onions).I determined the water content of the oil at the end of the experiment, and was amazed to find it was only 0.017%.This suggests that onions can be fried so that all the water is boiled off and only oil remains; this could be critical if the next step in the recipe was adding woody aromatic spices (cardamom, cassia, cloves, mace) when the flavour compounds will be extracted. If water were present, less extraction would take place, I think.

Next experiments in this series are to do the same with garlic, ginger and chilli, and to check on the extraction efficiency of the chilli colour.

where's the flavour? about Kow

Octanol water partition coefficient

What on earth is an 'octanol water partition coefficient' and what possible relevance could it have to cooking?
Well to answer the second part first, it is very relevant, it can be enormously useful when considering flavour distribution in our food

It can be explained as follows: for the moment, instead of octanol (an alcohol with eight carbon atoms) think vegetable oil.
Take a glass bottle, say 200mm high and 50mm diameter (the size is actually irrelevant!) and add a little water to it, say 50mm depth. Now add the same volume of a vegetable oil. This oil will float on top of the water. Now add a teaspoon of turmeric powder, the yellow coloured powder used in Asian cooking, then cap and shake the bottle for a while. Let it settle, so that you end up with two layers of liquid, and a sediment of turmeric powder. Look at the two layers of liquid. What do you see?

When I did this experiment, I was quite surprised to see the oil layer was a golden yellow colour, and the water only lightly coloured, mostly from suspended turmeric powder.
What had happened was that the yellow colour in the turmeric, a substance called curcumin, was dissolved out of the turmeric powder, and into the water and oil, except that most of it ended up in the oil, and very little in the water, which was evident from the depth of colour in each.
In fact, what had been performed was a very useful chemistry experiment, (although chemists use octanol instead of vegetable oil, as veg oil is quite a variable product, and octanol is a well characterized pure chemical). The curcumin is said to have partitioned between the oil and the water, that is, quite simply, part of the curcumin is in the oil, and part in the water. This is quite usual with most organic chemicals.
If I had the equipment to measure the amount of curcumin in each liquid, I could calculate the amount in each, and give a value for the amount in the oil and the amount in the water. This would be very similar to a value using octanol and water, and the value is known as the 'octanol water partition coefficient'.

[stop press! I did measure the optical density of the two layers, after diluting the oil phase 1000 times with acetone, and filtering the water phase through a 0.2µm filter, at a wavelength of 420nm. My result was a veg oil/water partition coefficient (Kvow ?) of 2000 exactly, which agrees very well with the literature value for Kow of 1950 !]

How does this help the cook/chef?

With an understanding of the above and a little chemistry, it is possible to estimate (or even look up) values for the octanol water partition coefficient. Most organic chemicals will have values between 100 and 1,000,000, this value is given the abbreviation Kow. The numbers (the coefficients) can be quite large, so it is common to use the log of the number, and it's usual to look up the log Kow of a substance.

So how is knowing the Kow or log Kow of a substance going to help with cooking?

Well, the larger the number (Kow), the more the substance will partition into the oil. A substance with a Kow of say 100 will be 100 times more likely be concentrated in the oil than in the water, as long as the oil and water are in contact with each other, so the substance can move freely (partition) between the two liquids.

If we consider making a dish which has oil and water in it, we could estimate where the flavours would partition into, that is, estimate how the flavour would be distributed!or did you think the flavours were evenly distributed throughout the food? Nothing could be further from the truth.

In reality, as you might expect, the picture is more complicated, flavours not only partition between the oil and the water, but also between meat, bone, air, vegetables and possibly even the cooking vessel itself.
(See 'fugacity' below, for a fuller explanation)

So, the Kow is important in cooking, it can give us an idea of where flavours are likely to be. Remember the cooks saying, "the flavour is in the fat!" Now you know why! Most flavours are organic substances that have a relatively high Kow, so these substances will partition into the fat/oil, and also onto meat surfaces, especially if associated with fat. The Kow value also explains why it is better to fry spices in oil rather than add them to a dish when there is water present. The hot oil/fat will extract the flavours, the water will be much less efficient at doing so. It is common to fry whole spices in oil/fat when making Indian dishes, for example, such spices as cassia, cardamom, cloves and mace all contain flavour compounds with Kow's of 1,000 or more, so these flavours will end up in the oil and fatty meat, even though there may be a lot of water-based gravy. In many Indian dishes, it is suggested that the gravy be cooked until the oil just sepates from the water, at that stage an emulsion can form and the oil and water mix to some extent, thus distributing the flavours more evenly.
So to get the maximum flavour from your ingredients, think oil/fat! and think where your flavours might end up!!

Fugacity

.....a culinary twist on an environmental concept

Fugacity is a word you may not know, though it comes from the Latin word fugit, meaning to leave or move on, as in tempus fugit....time flies.

Here, it will be used as a way of determining the distribution of flavour compounds in a cooked dish, for example, where spice flavours are likely to be concentrated. The theory is quite well proven and accepted in environmental work, where it is used to determine the distribution of one or more (polluting) chemicals throughout the region/s of interest. I am applying the same rationale to one pot cooking.

In environmental work, one first needs to define something which is called the unit world. The unit world (that to be considered) may include soil, water, sediment, air, vegetation and other biota, such as fish. The theory predicts, after suitable data manipulation, how a pollutant is likely to be distributed between the various parts of the defined 'unit world'. It is assumed that the pollutant is in equilibrium with its environment, that is, there has been time to exchange from, and to, each and every part of the unit world.

In culinary terms, the unit world is defined by the dish/es that are cooking. It is necessary to define the parts of the 'unit world', and estimate the volume of each part.
So, if we consider cooking a one pot dish such as a curry, the unit world may comprise: oil/fat, water, vegetables, meat, bones, the air above the ingredients, and maybe one or two other things which may be discussed later. We can estimate the volume of all the ingredients and the cooking vessel, and we could consider whether the pot needs a lid!
Next, we must consider which flavours we are interested in, and how much there is going to be of each. Then we need to find out a few things about each flavour compound, such as the molecular weight, solubility in water, vapour pressure and the octanol water partition coefficient. Have I lost you yet?
The molecular mass is a way of expressing the size (and therefore mass) of the flavour molecule, all flavour chemicals are very small molecules (<300 Dalton) compared to some other biochemicals such as proteins, starches, DNA and the like.

The solubility in water is self explanatary, but it will be temperature dependent. The vapour pressure is the pressure of the vapour phase of the chemical above its surface. It is also temperature dependent, and when a chemical's vapour pressure is equal to atmospheric pressure at the point on the earth where the chemical is, it is said to have reached its boiling point. The octanol water partition coefficient has been discussed above, but it is less affected by temperature.

So, we have a pot containing our ingredients, say a joint of meat on the bone, some fat/oil, some water, a little vegetable puree, and we must not forget the air above the cooking pot, and whether the pot will have a lid on (effectively producing a closed system) or lid off, meaning the flavour chemicals are free to disperse into the oven or the whole kitchen space! After cooking for an hour, we may assume that all phases (parts of our unit world) are in equilibrium, which means the flavours have had time to partition into every part of the dish, and the air around it.

So how do we work out where the flavour chemicals are? Well, it works the same as the octanol water partition coefficient (Kow) experiment above except, not only do we have a partition between the oil and the water (the Kow) but also between everything else as well. Relatively simple equations enable us to work out how much of each flavour compound there is in each part of the dish (the unit world). If we know how much flavour compound we put in and the volumes of our unit world, we can calculate where it will end up and the proportions of each. And the answers may surprise you.

Here is an example.

Firstly, we must define our unit world.
I will start with a one pot dish (a 10 litre pot with a lid on), containing water, oil, meat and vegetable puree. So the unit world, with appropriate volumes, is as follows:

pot (air space): 9 litres
water: 200ml
oil: 100ml
meat: 500ml
veg. 200ml

Next we must add some flavourings, say spices, which contain certain chemicals, say:

cloves containing eugenol
cardamom containing cineol
cassia containing cinnamaldehyde
pepper containing limonene, linalool, alpha phellandrene and 3-carene, among others.

We will assume we put in the same amount of each spice chemical (never true in practice, but see later)

Then we need to find the relevant physico-chemical data we require, put it all into a spreadsheet and carry out the calculations. We assume that ever part of the unit world (in our pot) is in contact with every other part, and that the dish is cooked for long enough for everything to come to a steady state (equilibrium). We also have to guess at the level of fat in the meat and the vegetable puree (I guessed 10% for the meat, and 0.2% for the vegetables).

to be continued......