I recently stumbled over “Norwegian egg coffee”. At first I thought it was a joke, but it turned out that this is indeed an “egg coffee” – coffee prepared with an egg! I have never heard about it here in Norway, but the fact that it’s popular among Americans of Scandinavian origin in the Midwest suggests that it could be something immigrants brought with them from Norway (feel free to fill me out on the historic origins of this!). I mentioned egg coffee to my mom, and although she had never heard of it before, she did mention that skin or swim bladders from fish were used when boiling coffee to help clearify it. In fact the Norwegian name for this – klareskinn – literally means “clearing skin”. The English name is isinglass (thank’s Rob!). Could it be that the fish skin originally used was replaced by eggs, perhaps due to a limited availability of fish in the Midwest? After all, both are good protein sources.


Egg coffee is amber colored and you can clearly see some precipitate from the egg-coffe mixture. If serving the coffee warm it seems to be difficult to totally avoid the precipitate unless you filter the finished coffee through a cheese cloth or filter paper. The coffee in this picture has not been filtered yet.

When looking into the chemistry behind this it isn’t as strange as it may sound. Fish skin as well as eggs contain proteins. The addition of proteins while preparing the coffee serves two purposes: 1) it helps the coffee grounds to flocculate, allowing them to sink faster to the bottom of the pot (this effect is probably more pronounced when using eggs) and 2) the proteins bind irreversibly to astringent and bitter tasting polyphenols in coffee to form insoluble complexes that will precipitate. The end result is a clearer coffee with a pleasant and mild taste. The bitterness is only barely noticeable, but the coffee still has enough “body” so it doesn’t feel too thin!

Norwegian egg coffee
80 g coarsly ground coffee (rouhgly 200 mL)
1 egg
100 mL cold water
2.5 L boiling water
250 mL cold water

Mix coffee with an egg and 100 mL cold water to a thick paste. Add this mixture to the boiling water, stir carefully and leave to boil for 2-3 min. Remove pot from stove and add the remaining cold water. Let the grounds settle for a couple of minutes, skim off any floating particles, filter through a fine meshed sieve, a cheese cloth or filter paper and serve.

The first time I made this I stirred quite a bit to break up the big lumps. But I was curious whether stirring had any influence on the amount of fine particles, so I repeated the whole process with as little stirring as possible. The lumps of ground coffee where significantly larger, but I couldn’t really see a difference on the prepared coffee. There was however a small difference when looking at the glasses from below (see picture below). My conclusion so far is that there is not a big difference, and that it’s OK to stir a little at the start to break up the biggest lumps. This will also allow a more complete extraction of the ground coffee.

Difference between much (left) and little (right) stirring as the coffee boils as seen from the precipitate at the bottom of a glass of egg coffee.

The addition of the cold water helps formation and settling of the precipitate. Home brewers talk about a “cold break” when they cool wort rapidly in order to precipitate proteins which have been extracted from the malt. And while we’re talking about beer chill haze also comes to my mind. This is the cloudiness that occurs upon cooling beer, and again it’s caused by precipitation of protein-polyphenol complexes. The effect of adding only 10% cold water to the still hot egg coffee is of course limited, and won’t really be enough to give a “cold break”. But since egg coffee has a pleasant taste even when cold, I have decided to cool a whole pot of egg coffee before filtering it. I may post more on how this turns out later, but initial tasting suggests that it’s going to be a very nice iced coffee!

The interesting thing about the protein-polyphenol complexes is that we also encounter them when drinking wine (a quick reminder here that polyphenols is a group of compounds which includes tannins). There’s a nice experiment you can do to illustrate this which has been published on Khymos previously. When we drink red wine, the tannins react with proteins in our saliva to form water insoluble protein-tannin complexes. A precipitate is formed and as a result, the lubricating properties of the saliva are lost and our tongue feels rough and dry. In other words, we experience the astringency of the red wine. To ilustrate this, try the following (I was first introduced to this experiment at the 2004 International workshop of molecular gastronomy in Erice):

Take a sip of a dry red wine, preferably rich in tannin. Keep the wine in your mouth for 10-20 seconds without swallowing. Spit it into an empty glass and watch how a precipitate forms (this might take a minute or two). Notice how the color changes from red to light red or even pink (see picture below). Rinse your mouth by chewing a piece of bread and drink some water. Take a small sip of the wine that you just spat out (if you dare!). Since the tannins of this wine have already reacted with your saliva, it is as if they were removed from the wine, leaving a fad and flat wine without much taste at all.

Top: red wine. Bottom: formation of precipitate in red wine mixed with saliva.

The saliva flow rate and the concentration of proteins varies from person to person (the latter with a factor of 20). Furthermore the flow rate and protein concentration also varies throughout the day and is also influenced by what you are eating/drinking and even by the smell of food. As a consequence, a person with a high saliva flow rate and/or a high concentration of proteins is more likely to approve of a red wine rich in tannins than someone with a low saliva flow and a lower protein concentration. Knowing this, you should not be surprised that wine preferences can be very individual.

Last year, while visiting family in Germany, I decided to pick some walnuts to bring home to Norway. They were not ripe, which was good, because I was planning to make nocino, a walnut liqueur. You can easily find a number of recipes by googling and there is also a nocino-thread over at eGullet.

What fascinated me the first time a saw nocino mentioned in a book about liqueurs was the nearly black color. Many recipes comment that after steeping, the liquid looks more like used motor oil than something edible. The color is really amazing and I also observed that most recipes recommended the use of gloves as the stains from the unripe walnuts would not easily come off. The juice from the walnuts is a light yellow green color to start with, but when exposed to air it quickly turns dark brown. Color chemistry is always fascinating and I couldn’t resist the temptation to investigate this further.

This picture was taken only 3 days after I steeped the walnuts in ethanol.

I did a little research and found a number of scientific papers about nocino. Some of them focus on the high content of phenolic compounds and possible health effects. Personally I can’t really see how nocino will ever become a significant source of phenolics in anyones diet. Even the article entitled “Influence of processing variables on some characteristics of nocino liqueur” is mainly concerned with how the process variables affect the antioxidant activity. There is no mention of taste or aroma (and this is perhaps why I would not classify these papers as molecular gastronomy), but the authors conlcude that walnuts picked earlier (more unripe) combined with a long steeping time (3 months) give a higher content of phenolics. They correlate the phenolic content to a measure of firmness, but for those making nocino at home an easy way to test the walnuts is to pierce them with a knitting pin – the walnut should be soft all the way through. As a side note I should mention that the walnuts I used were picked in August which is way too late – the inner walnut shell was stone hard, but I decided to give it a try anyway (more on how it all worked out in a comming follow-up post). The closest I came to some input regarding aroma was an article were 12 different phenolic compounds were analyzed in walnut extracts made with 40, 60 and 96% ethanol. Although the total phenolic content was highest when using 96% ethanol, they found that the concentration of some phenolics (protocatechuic, sinapic and p-coumaric acid) increased with the concentration of the ethanol used for extraction, whereas other phenolics (gallic, chlorogenic, vanillic and syringic acid, (+)-catechin, juglone) were best extracted with 40% ethanol. Polyphenolic compounds are normally bitter or astringent. The low molecular weight compounds are typically more bitter. With increasing molecular weight bitterness decreases whereas astringency generally increases. The solubility in water decreases with higher molecular weight. As I mentioned in a previous post on ethanol extractions this is the reason why 30-60% ethanol is most commonly used for infusions and extractions.

Even though the contents of the jar seems to be black, it’s actually still a bright green color if you shine enough light through it. The reason for this is that the jar was kept well closed. Once the nocino is filtered and thereby exposed to plenty of air the color will change to dark orange brown.

Despite all the papers on nocino I still didn’t know more about the colors of nocino, and what I was really searching for was an account of the color changes I observed. The incredible staining of skin and the change from bright green to dark brown and nearly black upon contact with air. Back in 1856 Reischauer and Vogel [1] studied and isolated a compound they named juglone after the latin name of walnut (juglans). In 1885 the structure was elucidated by Bernthsen and Semper who two years later also synthesized the compound [2]. Later Mylius [3] isolated a precursor to juglone, α-hydrojuglone, which upon oxidation yields juglone. And in 1950 Daglish [4] showed that what had become known as “apparent vitamin C” was in fact a hydrojuglone glucoside, a precursor to α-hydrojuglone. This is a nice illustration of how the most stable molecules are discovered first, before one realizes that there are more reactive precursors.

It’s also interesting to note that the structure of juglone is closely related to lawsone, a compound found in the henna plant (Lawsonia inermis) which is used for skin coloring. The hydroxyl group on the quinoid double bond causes lawsone to easily undergo oxidative dimerization (which I belive leads to compounds with a darker color – this would explain why henna color typically needs some time to develop). In the henna plant lawsone exists as a glucoside, and I would suspect that it is in the form of hydrolawsone glucoside which looses glucose and is oxidized as is the case with hydrojuglone.

Read on: Nocino – walnut liqueur (part II) – includes recipes!

References:
[1] Vogel, Reischauer Jahresberichte 1856, 693.
[2] Bernthsen, Semper Berichte der Deutschen Chemischen Gesellschaft, 1885, 18, 203. Bernthsen, Semper Berichte der Deutschen Chemischen Gesellschaft, 1887, 20, 934.
[3] Mylius, F. Berichte der Deutschen Chemischen Gesellschaft 1885, 17, 2411. Mylius, F. Berichte der Deutschen Chemischen Gesellschaft 1885, 18, 463.
[4] Daglish C. Biochem J. 1950, 47, 452. Daglish C. Biochem J. 1950, 47, 458. Daglish C. Biochem J. 1950, 47, 462.

(Picture by IreneM entiteld “coffee with a “drop” of milk” from DPchallenge – OK, it’s not tea, but I just love this picture!)

By measuring the blood vessel’s ability to expand (and thereby reduce the blood pressure) the researchers found that this ability was improved by tea, but the effect was completely blunted if milk was added to the tea. It was found that the caseins were responsible for the observed inhibition, probably by formation of complexes with catechins. It is believed that catechins (polyphenolic compounds, belong to the group of flavonoids, structure of epicatechin shown below) trigger the release of other active substances that are responsible for the expansion of blood vessels (also known as vasodilation).

The results of this study are not limited to tea, because catechins are found in many other foods, including citrus fruits, wine and chocolate.