mmmmmm …

January 12, 2011

There has been a lot of discussion lately in the chemistry community about making chemistry more approachable and visible to the general public. Paul ponders this issue at the end of part 1 of his 2010 Chemmy Awards post in his ruminations on whether or not chemistry needs a “hero”. In this vein, the fine crew at Nature Chemistry took it upon themselves to conduct a quick poll to decide who is the greatest chemist of all time. This post and poll generated lots of chatter and strong opinions, which included Paul expanding on his top-5 chemists. Icanhasscience has some really fun ideas (including possibly utilizing Lady Gaga and, dare I say it… Justin Bieber) on what we can do to more effectively communicate chemistry. One problem with this, which Chemjobber and Th’Gaussling point out, is that chemists live in the details. We need all of our discussions to be factually correct and to be expressed with the proper conventions. Let me assure you, dear reader, that this gets tedious even for the chemist.

One approach that I think may be useful, and one that I am currently using in a general education chemistry course, is to discuss the chemistry that the students already unknowingly do.

Let’s do a quick little exercise. Which of the following two chemical reactions would you rather do?

This one:

To see a larger version click here. Notes, this reaction involves mixing the two reactants and washing the resulting white/crystalline product with dilute acid to purify it.

Or, this one:

To see a larger version click here. Notes: For a satisfactory product, two competing reactions must be controlled. First, the heat and acid catalyzed decomposition into smaller molecules must proceed in a manner that produces a sufficient number of these molecules. Caution must be exercised as over-heating will cause all of these molecules resulting from decomposition to leave your final product. Second, the chemist must ensure that the competing process of oligomerization is sufficiently carried out to produce colloids of known diameter and coloration. Warning, overheating these colloids will cause them to burn and turn into carbon.

I suppose this choice could also be communicated in the following fashion:

Would you rather 1) synthesize 2-acetoxybenzoic acid or 2) make caramel?

When put this way, the non-chemist would very quickly choose to make caramel, even though the process of caramelization is incredibly complex and aggravatingly (at times) difficult to adequately control. (I’m sure many of you recognize reaction 1. It is the synthesis of aspirin – which our friend Katherine Haxton can tell you all about – and it is ubiquitous in high school/general chemistry/organic chemistry laboratories.) But, back to the caramelization. There aren’t as many people who are more intimidated by caramelization than by the synthesis of 2-acetoxybenzoic acid. Cooking can be used as a very nice platform in which to talk about some really interesting chemistry. And caramelization is just as nice a jumping-off-pint as any other.

So, let’s talk a little bit more about caramelization. When you are making caramels or caramel sauce, there are really three things that you are trying to achieve. 1) The proper aromas. 2) An appealing brown color. 3) An acceptable texture – not too sticky/easy on the tooth. Unfortunately these three processes are often at odds.

In order to do a better job of caramelization, why don’t we try to better understand some of the individual steps first.

Decomposition of Sucrose

Before any of the “interesting” reactions occur, sucrose (table sugar, a disaccharide – one molecule containing two monosaccharides) has to break down into fructose and glucose. (In the figure, fructose is the pentagon-shaped portion on the left and glucose is the hexagonal shaped portion on the right). This decomposition occurs at roughly 180^oC (356^oF). You cannot caramelize sugar on a double boiler, which won’t get over 100^oC. Even, though you’d like to have better control over the temperature in your pan, you really need a lot of heat to make this process work.

Decomposition of Fructose and Glucose into Aroma Molecules

The first noticeable change in the caramelization is the appearance of aromas coming from the sugar. I can’t smell sugar. I really don’t know if anyone can. It’s not volatile (i.e. it doesn’t appear in vapor form at pressures and temperatures that we’re used to on Earth). But, when fructose and glucose start to break down into smaller, more volatile compounds, your brain can very easily detect that there are new molecules present. Some important molecules that are produced during caramelization are the furans (have a nutty aroma), diacetyl (smells like butter), maltol (toasty), and ethyl acetate (fruity). From my reading, it appears that the break-down of fructose and glucose into these molecules is acid catalyzed. (This means that having acids present will enable the reaction to occur at lowertemperatures and proceed at a faster rate.) I know that warm solutions of sugar are more acidic than cold solutions, but I have no idea if molten sugars are acidic in and of themselves. Because of the role of acidity, it seems as though a cook would develop more aromas by initiating the caramelization process in water. (If this is truly an acid catalyzed, sucrose may start decomposing into aroma molecules at the lower temperature of boiling water).

Oligomerization of Fructose and Glucose

This is the stickiest part of the process. In the oligomerization reactions, the brown coloration and portions of the texture are developed. Obviously, these are crucial to a good caramel. While the overall reactions occurring here are still unknown to chemists (there has been some research on this process going on since the late 1700′s), some recent research has put several parts of sugar oligomerization into clearer view. First, the individual sugars dimerize (two sugars come together to form one molecule) into a new form that contains two rings attached by a third central ring (see compound a in the figure). In the case of fructose, this structure is called a di-D-fructose Dianhydride. From this point, the chemistry gets a little hand-wavy. The difructose dianhydride molecules can further react on three different pathways. On the first, one molecule loses 12 water molecules from its structure to form a compound called caramelan (C₁₂H₁₂O₉). Caramelan aggregates to form small, brown particles that are 460 nanometers (0.46 micrometers, 0.000018 inches) in size. A second type of molecule that the difructose dianhydrides can form is called caramelen (C₃₆H₁₈O₂₄). Caramelen aggregates to form small brown particles that are 950 nanometers in size. Finally, the difructose dianhydrides can also form caramelin (C₂₄H₂₆O₁₃) from the combination of two difructose dianhydrides and the elimination of 27 water molecules. Caramelin forms aggregates that are 4333 nm in size and darker in color.

Chemists really still don’t have a good grasp on what these molecules are. We know that they have similar structures to sugars, i.e. that they are still in ring form. (For our chemist friends, the IR spectra are similar.) Chemists have also figured out that there are free-radicals in the system. (Free radicals are molecules that don’t quite have enough atoms to fill all of the bonding requirements.) This is part of what makes caramels sticky.

(If any chemists want to try their hand at guessing what the structure of caramelan, caramelen, and caramelin are just send me a chemdraw to matt A-T sciencegeist DOT net. The reference that has some spectral analysis of these chemicals is: Piotr Tomasik “The Thermal Decomposition of Carbohydrates Part 1″ in Advances in Carbohydrate Chemistry and Biochemistry volume 47 1989. p203)

My suggestions for making a good caramel sauce

Add 1 cup (240 g) of sugar and 1/2 cup of water to a pan. (The water will increase the acidity and should help to develop aroma).
Heat over high heat with stirring until the water has evaporated.
Once the water has evaporated, the sugar should melt and start to brown.
Quickly turn down the burner (this will stop the caramelization process).
With continuous stirring add 1cup (240 g) of heavy cream.
Note – you really need to pour the cream slowly and stir really quickly. Failure to do so will cause the caramel to cool down too rapidly giving you crunchy caramel crystals in the middle of your beautiful caramel sauce.

So, there you go. Enjoy. Go home and do some chemistry tonight. And then pour your chemistry all over some vanilla ice cream!

Cheers

-mrh

References:

Books:
On Food and Cooking by Harold McGee. Scribner.
Food Chemistry 2nd edition by Belitz and Grosch. Springer.

Journal Articles:
Piotr Tomasik “The Thermal Decomposition of Carbohydrates Part 1″ in Advances in Carbohydrate Chemistry and Biochemistry volume 47 1989. p203
Suarez-Pereira et al. “Di-D-fructose Dianhydride-Enriched Products by Acid Ion-Exchange Resin-Promoted Carmelization of D-Fructose: Chemical Analyses” in the Journal of Agriculture and Food Chemistry volume 58, 2010. p1777.