DIY Belgian Candy Syrup 2: Experiments

[Miniseries part one, two]

In part one, I went in to some of the science behind making syrup from sugar.  Here, I’ll go into some of the experiments I tried before coming up with something I felt worthy of risking five gallons of brew over.

The general procedure is to heat sugar, water, and a source of amino acids.  As the water evaporates, the sugar solution’s boiling point will rise.  While heating, keep in mind that as the temperature rises and water evaporates, there is less thermal mass in the pan, so the temperature will rise more quickly (ie if it took 10 minutes to go from 240-250F/ 115.5-121C, it might only take a minute or two to go from 280-290F/ 138-143C).  The ending boiling temperature, and thus the ending water content, will determine the state of the sugar when it is cooled; 240F will result in a syrup.  Simply heat to the desired temperature for chemical reactions, cool with the addition of water below 240F/ 115.5C, then bring back to 240F/ 115.5C before cooling.  Be extremely careful when adding water- the solution is well above the boiling point of pure water and can instantly vaporize and spit the sticky hot syrup well out of the pan.  To store the syrup, I like to use some small mason jars; you’ll need to cool the syrup first or the glass will shatter from thermal shock (or I suppose you could heat the glass to the temperature of the syrup if you wanted to be difficult).

Home made Belgian candi syrup

My first attempt was a bit of a stab in the dark.  I used enough tap water to dissolve 1 lb/ 0.45 kg of sugar and 3 tbsp/ 44.3 mL of DAP (diammonium phosphate).  The DAP is meant to provide a source of amino acids, necessary for the Maillard reactions as covered in part one; you’ll notice an ammonia smell as the DAP decomposes when it heats.  I quickly found out that my Walmart meat thermometer was useless at the required temperatures, and was essentially running blind.  You’ll want a very accurate thermometer that can read into the hundreds of degrees (a good candy or laboratory thermomter), as a few degrees will change the end product consistency drastically.  The result was a black, acrid burnt syrup which mellowed after a few weeks, but not enough to be usable.

I attributed the acrid flavor to fructose which scorches at a lower temperature than other sugars.  I decided to add pickling lime to make the solution more basic, because sucrose will invert into glucose and fructose faster in a more acidic environment.  Additionally, the solution will acidify as it caramelizes and undergoes Maillard reactions.  But, as I research it further now, DAP is known to decrease the charring temperature, so this might explain a lot of things too.  Perhaps the DAP is lowering the scorch temperature of the sugars and causing them to burn.  I don’t know enough about the specific chemistry taking place with the DAP to offer more than a guess.

In the next batch I added molasses and used Wyeast yeast nutrient in addition to the DAP to increase the complexity of the amino acids present for reaction.  I also started using distilled water to cut down on experimental variables; distilled water is of a predictable pH and has no other dissolved solids.  Just keep in mind that once distilled water is exposed to the atmosphere, it will slowly drop a bit below a pH of 7 by dissolving CO2 into solution, forming carbonic acid (the same stuff that gives your tongue the prickly sensation when drinking anything carbonated).  I only went to 268F/ 131C with this batch, but only because the temperature started falling.  As it turns out, the sugar had crystallized on the bottom of the pan, forming an insulating layer.  Lesson learned: stir the sugar more or less constantly (if there are any candy makers in the audience, they might disagree- the stirring can cause course sugar crystals to form in some instances by forming seed crystals, but this is not really a concern once the sucrose begins to invert).  This batch (1 lb/ 0.45 kg sugar, 1/2 tsp/ 2.5 mL lime, 1 tsp/ 4.9 mL nutrient, 1 tsp/ 4.9 mL molasses) turned out light brown and melt-in-your-mouth buttery, and completely solidified as it cooled- not the most fun to clean up.

I used the same general process and took it to 300F/ 149C (1 lb/ 0.45 kg sugar, 1/2 tsp/ 2.5 mL lime, 1/2 tsp/ 2.5 mL nutrient, 1/2 tsp/ 2.5 mL DAP, 1 tsp/ 4.9 mL molasses) and got a syrup with molasses-like flavors and raisins, with some burnt sugar up front.  This was a syrup I felt good enough to brew with, but I wanted to push it further.  I also noticed some grainy clumps in the syrup, which I think was the lime or more likely something in the nutrient not dissolving, but I was never able to be sure.  This didn’t concern me for brewing, as I knew it would just be dissolved or fall out of solution at some point.

I next tried multiple heating cycles, reasoning that I could induce more flavor without increasing the temperature and charring.  For this batch (1 lb/ 0.45 kg sugar, 1/2 tsp/ 2.5 mL lime, 1/2 tsp/ 2.5 mL nutrient, 1 tsp/ 4.9 mL DAP, 1 tsp/ 4.9 mL molasses), I used 3 heating cycles, each to 290F/ 143.5C, cooling to below 240F/ 115.5C each time with the addition of water.  Unfortunately burnt and tart flavors overpowered a delightfully fruity background.  During cooking, there was a divine cherry-rum aroma which I was never able to reproduce in the flavor.  Another odd thing I noticed was that upon each reheating, there was less of a tendency for the sugar to crystallize.  I attribute this to the further inversion of the sugar.  Glucose and fructose will disrupt the crystalline structure of sucrose, slowing crystallization.

My next batch (1 lb/ 0.45 kg sugar, 1 tsp/ 4.9 mL lime, 1/2 tsp/ 2.5 mL nutrient, 1 tsp/ 4.9 mL DAP, 1 tsp/ 4.9 mL molasses), I heated to 280F/ 138C four times.  The result was a flavorful but slightly tart syrup (indicating acid), with no burnt flavor, but otherwise completely uninteresting.

By this point, I got tired of trying to avoid charring.  As I was attributing it to the fructose from the inverting sucrose, I decided to use some glucose sugar I had for priming bottles.  I immediately noticed a change in consistency and a lack of crystallization, as well as how much more rapidly the syrup darkened (even before 240F/ 115.5C), presumably a result of how much more reactive this simpler sugar is.  For this batch (1 lb/ 0.45 kg glucose, 1 tsp/ 4.9 mL lime, 1/2 tsp/ 2.5 mL nutrient, 1 tsp/ 4.9 mL DAP, 1 tsp/ 4.9 mL molasses), I heated to 290F/ 143C, then to 270F/ 132C four more times.  Oddly, this batch also had a burnt flavor, even though it was never heated anywhere near the decomposition point of glucose.  I suspected it was some product of the reactions, but I now think it may have been the DAP causing scorching.

I went back to basics with my next batch (1 lb/ 0.45 kg sucrose, 1 tsp/ 4.9 mL lime, and 1.5 tsp/ 2.5 mL DAP).  I also went with a relatively low temperature cycling between 240F/ 115.5C and 260F/ 126.5C fifteen times.  This batch I would call a success, with a definite raisin and plum flavor with no burnt character, though it seems as if it could go through additional cycles.

My final batch (2 lb/ 0.91 kg sucrose, 1.25 tsp/ 6.2 mL lime, 3 tsp/ 14.8 mL DAP) was based off of the previous batch, and was meant to be what I used in my beer, hence the scaling up.  I had read that lime should not be scaled up directly with the rest of the recipe to avoid a mineral off flavor, but given the tart burnt flavor, I suspect I needed more.

Since this batch didn’t turn out, I used two previous batches in my beer (the third and seventh).  It turned out quite well, with a flavor profile very close to Chimay thanks to the White Labs WLP575 yeast blend, though it was quite a bit more thin due to the low mash temperature of 140F/ 60C.  My one complaint is a slight dry acridness on the back of the tongue about 30 seconds after swallowing which I think is from the syrup.  But, that’s being very persnickety- no one else has noticed this off flavor.

This work was based on some great experiments done over at Ryan Brews and Nate O.’s Brew Log.  And another source, from another engineer.  [2013.11.21 UPDATE: Another great recent experiment over at Sui Generis: Part One and Two.  He beat me to trying DME as the sugar source and it seems to have produced good results.] [2014.11.14 UPDATE: Found another source of info from back in 2009 on the homebrewtalk forums.]

[Miniseries part one, two]

– Dennis,
Life, Fermented

Advertisement