Wednesday, August 10, 2011

Candi syrup secrets, and how to make your own

There is a lot of misinformation on the internet about candi syrup, what it is, and how it's made. After making 100 or so batches of candi syrup, and hundreds of hours of work and research, I have some insight on this.

First, let's start with what it is. "Candi syrup" or "Kandijsiroop" or "sirop de candi" is dark, richly flavored syrup made from sugar utilizing Maillard reactions and caramelization. In Belgium, tariffs and subsidies make beet sugar much cheaper than cane sugar, so the "authentic" Belgian syrups use beet sugar. There is nothing particularly good about beet sugar. It's just what they use because it's cheap and convenient. The sugar refining process removes any impurities, leaving only the pure sucrose. Beet sucrose is chemically identical to sugar cane sucrose.

Maillard reactions are a complicated set of interactions between a nitrogen source and reducing sugars. Examples of reducing sugars are glucose and fructose. Sucrose is not a reducing sugar, therefore it is impossible for Maillard reactions to occur with sucrose. It is impossible to have Maillard reactions with pure beet or cane sugar in its natural state. To allow the Maillard reactions to occur, one must invert the sugar from sucrose to glucose and fructose.

Sugar can be inverted by heat or acid, or both. The rate of chemical reactions increases with temperature. While it may take several days for acid to invert sucrose at room temperature, it can happen in 30min if boiling. Once the sugar has been inverted, the sugars have become reducing sugars, which can form Maillard products. Any remaining acidity must be neutralized with a base of some kind.

All the manufacturers of candi syrup that I'm aware of claim their product contains only sugar, and is made by repeated heating and cooling alone. This is unlikely to be true based upon the chemistry involved.

The misinformation over candi syrup is caused by two factors. The first is that recipes cannot be patented. The only way to keep your secret recipe from being stolen is to keep it, well, a secret. The other factor is FDA regulations on food labeling. "Process aids" do not have to be included on ingredient labels. "Process aids" are defined as "Substances that are added to a food for their technical or functional effect in the processing but are present in the finished food at insignificant levels and do not have any technical or functional effect in that food." Candi syrup manufacturers are not required to disclose which acids and bases they use in producing their syrups, and to disclose them would mean anyone could reverse engineer their product.

The FDA has guidelines on what products can be used in the production of caramel. Acceptable food-grade carbohydrates:
dextrose (glucose)
malt syrup
starch hydrolysates

Acceptable acids:
acetic acid
citric acid
phosphoric acid
sulfuric acid
sulfurous acid
(In the UK hydrochloric acid is also used)

Acceptable bases:
Ammonium hydroxide
Calcium hydroxide
Potassium hydroxide
Sodium hydroxide

Acceptable salts include any combination of the following:
Ammonium, sodium, or potassium carbonate, bicarbonate, phosphate, sulfate and sulfite.

Lyle's Golden Syrup is a type of invert syrup made with hydrochloric acid and sodium hydroxide. The acid leaves chloride ions, and the base leaves sodium ions. Together they form sodium chloride, or common salt, in the final product.

Knowing accepted ingredients and a little bit about chemistry, we can make an educated guess into the true nature of candi syrup. I have no doubt the product begins as sucrose. Some acid and heat is added until the sucrose inverts. Any residual acid needs to be neutralized. Ammonium hydroxide (NH4OH) would both neutralize any residual acidity and provide a source of nitrogen for the Maillard reaction. If an alternative base is used, some other sort of nitrogen would be needed. Phosphates increase the rate of Maillard reactions, so an ammonium-phosphate salt would have the dual purpose of providing nitrogen and phosphate.

With this information, it is not difficult to devise a substitute for candi syrup. Here is a recipe I've come up with:

3oz Dextrose (corn sugar)
13oz Sucrose (table sugar)
1c H2O
1/2tsp diammonium-phosphate
1/2tsp potassium bicarbonate
1/2tsp lemon juice

1) Bring water and sugars to a boil
2) add 1/2tsp lemon juice
3) simmer for 30min between boiling and boiling+10*F (212*F-222*F at sea level)
Add water as necessary to keep temperature in correct range
4)Add 1/2tsp DAP
Add 1/2tsp KHCO3
Stir well to dissolve

5) Heat to 320*
Add 1/2c H2O
Stir well (be careful as sugar will sputter)

6) Heat to 310*
Add 1c H2O
Stir well
Pour in mason jar or other heat-tempered glass jar, or allow to cool and store in plastic

The reason I include a portion of dextrose (glucose) is that I've found dextrose will darken at lower temperatures than fructose, so having more dextrose will result in a roastier, darker syrup, while more fructose will result in a fruitier, lighter-colored syrup.

Another tweak you can make is heating the syrup to lower temperatures (you can get some nice vanilla flavors around 260*). You can also experiment with the number of heating/cooling cycles.

I encourage lots of experimentation to find what works for you. Maillard reactions are responsible for some amazing flavors, but also foul, acrid flavors. Finding the proper balance between the good and bad flavors is the hardest part of making your own candi syrup.


  1. have you done any fermentability tests on these?

  2. I've done a little bit of testing. Once I get a recipe nailed down exactly how I want it to taste, I'll do more complete trials. In a 2L trial, 60% DME and 40% syrup, my syrup in the recipe here got down to 1.011, while the D2 got down to 1.007. In a full batch of beer, where D2 got down to 1.010, my syrup got down to 1.012, accounting for as many variables as I could. The interesting thing was that the D2 beer actually tasted thicker and slicker, even though it had a lower FG. Mine tasted drier and more complex, despite the higher FG. The results were encouraging enough to keep at it.

  3. Also, depending on the makeup of your water, you may need to play around with the amount of acid and base necessary to get the pH where you want. My water is high in bicarbonates (400ppm or so) so with softer water you will probably need less acid and more base. I recently found pure citric acid in the canning section of the grocery store, so in the future I'll be using that instead of lemon juice.

  4. very interesting...I'd be interested to see attenuation data for just the syrup isolated. have you done taste comparison + pH check for syrups +/- of the potassium bicarb? i've noticed a slight 'metallic' or bright flavor to mine that is not in the D2, which seems much 'simpler' but also smoother, softer. I haven't yet tried adding Kbicarb to mine yet...

  5. The pH for the D2 is pretty high. I measure the syrups at the same dilution I use to measure points per pound per gallon. The syrup I made is 4.3, D2 is 6.3, and the D-180 syrup from that new company is 5.3. A lot of the bright/acrid flavors in your syrup will be mellowed by base addition.

    If I had access to a really strong base that would be ideal, some like potash or lye. The strongest base I've used has been slaked lime, which works reasonably well, flavor-wise, but I got a pretty good chemical burn from that, so wear gloves and goggles if you go that route.

    Just like roasted grains become more acidic the darker the roast, the same will be true for your syrup. The darkness of the commercial syrups tells me either a)they're adding some amount of caramel colouring, or b) they're adding a lot of base to get the pH that high.

    The problem with the potassium bicarbonate is that in large amounts you'll get a kind of minerally/chalky flavor. I tried calcium bicarbonate first, and the taste threshold was really low. Potassium bicarbonate is more flavor neutral, but it works best as a buffer to keep the pH from dropping too low, not to raise the pH one it's already too low. For that, the lime works better.

    And, if you want to try something really crazy, just use any protein source you can think of as a nitrogen source for the syrup. If you make cheese at all, when your "break" the milk with acid, you'll get curds and whey. The whey is full of protein and lactose, and some lipids. You can get some interesting flavors that way.

  6. huh, the pH on the spec sheet for D2 is 4.5-5.5.

    so maybe at the flavor thresholds, the stronger bases are really the way to go...but my limits for experimentation are reached when I start considering options the janatorial supplier and soapmaking shelves!

  7. I've never checked the pH of the plain syrup. I don't want to get my meter all gunked up. I've only measured it diluted.

    I agree with not wanting to use stronger bases. I don't think it's really necessary, but it would probably work a little bit better. Some of the minerally flavors you can pick up in the syrup don't carry over to the finished beer, so using the KHCO3 or picking lime isn't a problem, in practice.

  8. I'll let you know when I find time to take another stab at it...have you tried contacting dark candi or the new company to see if they'll divulge anything?

    my guess is that they really would not be interested in discussing any trade secrets.

  9. The Candisyrup guy told me they have a new patented process to make their "authentic" syrup, but they also have a secret process, and they only use pure beet and date sugar. It's all marketing BS.

    If it's a patent, it's not a secret, and he wouldn't give me the patent number. He also wouldn't say anything about the process or ingredients, which is why I started looking into labeling rules regarding process aids, and industrial caramel manufacture techniques.

    The laws of chemistry tell me it's impossible that they're telling the truth.

  10. Of course...we need to send a mole in to their operation, find out what 'ingredients' they are ordering!

    Agreed re: 'only sugar + heat + time' ...its just not possible. keep plugging away...I'll do the same.

    cheers, JC

  11. The Euro process for sugar beet refining is called "carbonatation." There are a few methods, but all involve increasing the alkalinity of the raw beet slurry coming in to precipitate and break down proteins. So I do believe that the imported stuff truly is "a byproduct of sugar beet refining." The beets provide the protein, and the carbonatation method provides the alkalinity. If you're starting with the refined product, you'd have to work backwards, and provide your own alkalinity protein.

  12. If I get really, really bored this winter, I might try buying a bunch of beets and making my own sugar from scratch.

  13. Nateo, check this out...I think Ryan has done the experiments that speak to the increased pH. Now, it just seems we need to figure out viable amino acid source beyond DAP/yeast nutrient.

  14. Cool, thanks JC for the link. Glad to see others working on this!

  15. Look-up how maple syrup is made. There are amino acids in the maple sap when it is harvested from the tree. When the sap is boiled it not only concentrates the sugars, but the amino acids create maillard reactions which give the syrup its color and flavor. Three are probably some amino acids that you can purchase that is derived from natural sources. Dates, are actually a good source of amino acids. I thought about grinding up some body building tablets to see if that works. Must not contain milk or meat amini acids or flavor will be off.

  16. Richard - I'll check out maple syrup. I've tried using various isolated amino acids (for body building) but the flavors aren't right. I did make one syrup that tasted exactly like almond extract that way. I'd have to check my notes to remember which amino acid it was. Date sugar is usually just pulverized dates, so I'll check that out. I've used date sugar before, but only on a couple batches.

  17. Has anyone tried just sucrose, a bit of acid, and date sugar yet? I have used sucrose + DAP a few times with great results but I would like to try sucrose and date sugar next but will have to come up with some proportions to start with.

  18. I am in Canada where food-grade slaked lime (ie. pickling lime) is hard to come by due to some home picklers adding too much to a batch of pickings, over adjusting the pickle brine ph from acidic to neutral, and giving themselves a nasty case of botchalism.

    Has anybody tried this with epsom salts (ie. magnesium sulfate) in place of the lime to raise the alkalinity of the sugar solution???

  19. Nate, I've been using your Candi Syrup recipe for some time now. But my batch in the keg, I burnt the syrup. I still used it, and I *hope* it will mellow in a few months. Am I wasting my time, or will it come good?

  20. Matthew - I'd caution against using epsom salts. Magnesium tastes pretty bad. You'd probably have better luck baking baking soda for an hour or so. That's an old pretzel-makers trick:

    Mike - It's hard to say, but it's really hard to guess the beer's flavor from the syrup's flavor. I've had some good-tasting syrups makes blah beer, and some bad-tasting syrups make good beer.

  21. Nate, the recipe you posted here, is this the best out of 100 attempts? What flavour difference will there be between NaHCO3 and KHCO3? The KHCO3 is very difficult to get here in The Netherlands, while NaHCO3 is more common.

  22. Just tried 50/50 sucrose and lme. Stuff starts putting acrid notes/smoke off at 270F (where I stopped), and remains incredibly frothy throughout the process, I presume from protein content. Flavor isn't horrid, tastes much like chocolate malt, coffee notes, maaaaybe dark fruit/cherry. Nothing like the sucrose DAP with all its wonderful vanilla almond notes. Very dark though. Before it started getting too dark, it smelled a little meaty/savory. re-adding water leaves a protein foam on top, like wort. Next experiment will be all sucrose with DAP and Gelatin as an amino source.

    1. Hi Elijah, have you tried the DAP and gelatin combination yet?

  23. Hi Nate, Thanks for doing all this work. I have a ton of info on the production of candi syrup in Belgium. In Belgium, they can use HCL and they do. They also use NAOH. Candi syrup is a blend with about 4% being caramelized syrup and the rest invert. I totally agree with you about the US producer and their secret process. The Belgian producers maintain the highest quality food certification possible. The US producer has nothing. That really worries me. I hope I am helping. - Brian


Note: Only a member of this blog may post a comment.