Polydextrose

[NulloModo]

Yes, Mannitol and Malitol are both pretty nasty as sugar alcohols go. I have had good luck with Xylitol which seems to have a very natural sugary taste, and less of the laxative effect that some of the others.

[JSkilling]

I made a simple brownie recipe today and used the Litesse sample I received. Here is the starting recipe:

4 ounces unsweetened chocolate (I use Scharffenberger)

1 stick butter, unsalted

1 cup sugar

1/2 t salt

1 t vanilla

2 eggs

1/2 cup all-purpose flour

Using Scott's calculations, I subbed the sugar for:

1 cup Litesse

2 T Xylitol

2 T Erythritol

1/2 t Splenda syrup

The batter was clearly not sweet enough but looked very much like my normal batter though it was a bit thicker. I added 3 T Whey Low and adjusted to about the right amount of sweetness. I don't like chocolate without the hit of sucrose and this probably added about 1 T with the added sweetness of the fructose.

The brownies had a nice gooey texture so the Litesse was very helpful in getting the sugar to hold the moisture instead of the flour, but the outside crusted to a very chewy consistency. Not to the point of being inedible but not what I'd serve. The application specs say to use Litesse at about 20% for baked goods and this experiment showed me what happens when the usage is higher than it should be. I'd cut this back to 1/4 cup the next time to see if I can get a good dense texture with a bit of a shattering on top and not so much chew.

I've a note into the applications development folks at Danisco to see how they are using it. I've already been communicating with their application development manager for frozen goods and she's been very helpful. That ratio is for only 10% usage and .20% of stabilizer. Since I was looking to replace atomized glucose and invert sugars in about the same percentage, this might just give me the mouthfeel I want in sorbet. Raspberry is the next experiment this weekend.

[scott123]

Josette, I stand by my math :) That was a cup of sugar equivalent. In the context of such an intense/bitter chocolate as scharffenberger, 1 C. of sugar would have been not nearly enough. Have you done 4 oz. sharffenberger to 1 cup sugar brownies before? My brownies are 4:1.5 and that's a very mild, alkalized Nestle baking chocolate. With the high cocoa content of Scharffenberger, I'm surprised an extra 3 T. of Whey Low was enough.

I, personally, adore that chewy crusty exterior, but I can see how someone might find it texturally obtuse. To prevent it from happening, I'd suggest three things:

1. Bake them at a lower temp - if you went with 350, try 300.

2. Bake them on a lower shelf.

3. Reduce the PDX to 1/2 C.

1/4 C. PDX translates into about 7% of the total ingredients (by weight). If you're going to cut back for less chew, I'd recommend trying 1/2 cup (14%). 1/2 C. and a lower shelf/lower temp should give you exactly what you're looking for.

P.S. Another way around the crusty top is to cover them when you take them out of the oven. The steam will soften them right up.

Edited January 29, 2005 by scott123 (log)

[JSkilling]

Yes, one cup of sugar is generally sweet enough for the Scharffenberger and the Cocoa Barry unsweetened, both. The original recipe calls for 1 1/4 cup but that's too sweet. I have many times mixed these two chocolates to success.

I just reread the specs and they say to use from 5% - 20% in baked goods. This is a case for the lower value to get what I want in texture. I had already planned to lower the temp on the next round. I'm going to use the least amount on this trial and then adjust towards the middle to see which percentage works out.

Then I'll need to do the same with Inulin and try that.

[nathanm]

I have a couple questions for the quite impressive sugar expertise on this thread. I am the one that posted the savory sorbet thread mentioned above, and that one interest. In that context I want the texture attributes of sugar without the sweet taste.

I am also interested in low-carb ice cream and sorbet, where I DO want the sweet taste. Obviously, to be truly low-carb, the main flavoring ingredient would have to be low-carb too. Raspberry would work, honey or caramel would not. Also, if you are really concerned about it, the lactose in milk might be an issue, but probably not for reasonable serving sizes.

1. Any issues with freezing PDX? Presumably it will lower the freezing point in a manner similar to other sugars, but I'd like to ask.

2. From what is said in the thread, it sounds like PDX is best incorporated into warm or hot liquid.

3. More generally, what would you recommend for low-carb ice cream or sorbet ? Will the blend of PDX, Xylitol, Erythritol, Stevia, Splenda discussed in the thread work or is it mainly for hot baked items?

4. Any particular flavors that work well with low carb ice cream or sorbet? I would guess that strong flavors, or those complimentary to the aftertaste from the sweeteners would be best...

[scott123]

Nathan, polydextrose makes phenomenal ice cream.

The chemistry of low carb ice cream gets a little complex, though. I've been reading everything I can on it, and still don't fully comprehend it. Low carb ice cream needs more than just polydextrose/artificial sweeteners to work. You'll need stabilizers/emulsifiers/water immobilizers to obtain the right mouthfeel/scoopability.

1. Polydextrose, like sugar, is a freezing point depressor. It's high molecular weight causes it to have less anti-freeze qualities than sugar, though (I read somewhere that it's half). Scoopability tends to be a major concern with low carb ice cream. Erythritol is a good freezing point depressor, so that will help. Glycerin, in small amounts, is another freezing point depressor that's popular in low carb formulas. And, if you're making this for adults, a zero carb alcohol like brandy, rum, vodka, gin, whiskey will lower the freezing point as well.

2. For a non baked application, dissolving the PDX in a warm liquid is essential. Actually, hot liquid. The heat necessary to denature the egg proteins for a creme anglais might not be sufficient. It would be better to heat the PDX with the cream until it dissolves, let it cool, then start adding the eggs/egg yolks.

3. Cold impairs the ability to taste, so an ice cream base has to be very sweet to be the correct perceived sweetness when frozen. Because of the intensity of sweetness required, combining multiple sweeteners isn't just a good idea, it's mandatory. The list you mention is excellent, although xylitol, with it's potential laxative effect should be used sparingly, if at all. I'd also add a tiny amount of ace k (5% of the sweetening equivalent).

Because you're dealing with a low carb product, fat is not an issue. Some milk might be good, but the majority of your formula should be cream (not UHT!). There's also low carb milks available.

Emulsifiers/Stabilizers are a whole other topic. I'm running out the door at the moment and will add more about those later.

Btw, I have a working low carb recipe that seems to produce good results that I'd be happy to share with you. If you're like me you'd probably want to learn how to fish rather than be provided with a fish, though :)

[nathanm]

Thanks very much for the info.

I use a Pacojet and make ice cream to order so scoopability is not as much an issue. I use stabilizers in some conventional ice creams already.

Sure, I would like the recipe if you don't mind. It is great to learn principles, but starting somewhere can be a big help!

[Sebastian]

I've had some very good erythritol ice creams. use of pdx or inulin wasn't necessarily required in that application, as the cooling effect of erythritol, well... it's ice cream 8-) plus, too much pdx or inulin in a high water application tends to get gummy. you'll never get a hard ice cream with erythritol, as molecularly it's very small, which means it'll depress the FP quite a bit. when working with HIS, i tend to prefer sucralose, but it can be difficult (and expensive) to get nowdays. there's also a fair amount of synergy that can be realized by using sucralose in conjunction with ace-k, but personally i don't care for ace-k straight. Neotame is extraordaarily sweet, and seems to have good flavor, but because it's so sweet, it can be hard to formulate with - however there are companies out there that are diluting it just for that reason. There's also a company out there (the name escapes me at the moment) who's extracting neohesperadin dihydrochalcone from orange peels (NHDC - forget trying to say that 5x real fast, i'd give you a dollar if you could do it 1 time real slow!), and labelling it as a natural flavor (very label friendly) - it's got a decent flavor, but use is restricted by law (in the us at least), to 10 ppm, and is really meant to be used as a flavor enhancer, not an sweetener, although it does have tremendous sweetening capabilities. In europe, you can use as much as you want of it for sweetening. The company, however, is saying it's not NHDC and that you can use as much of it as you want, which concerns me a bit as I've analyzed it already and know that it is, and frankly have issues with a company behaving like that.

[JSkilling]

Yes, one cup of sugar is generally sweet enough for the Scharffenberger and the Cocoa Barry unsweetened, both.  The original recipe calls for 1 1/4 cup but that's too sweet.  I have many times mixed these two chocolates to success. 

I just reread the specs and they say to use from 5% - 20% in baked goods.  This is a case for the lower value to get what I want in texture.  I had already planned to lower the temp on the next round.  I'm going to use the least amount on this trial and then adjust towards the middle to see which percentage works out.

Then  I'll need to do the same with Inulin and try that.

←

I ended up starting at 1/4 cup and added another 1/4 as the batter came together since it needed the PDX to emulsify the fat back in. I liked them at 1/2 cup PDX and still got a nice, dense fudgy brownie. I'm going to try around 3/4 cup and see how that comes out. Even at 1 cup in the mixture, the results on day two were much better for the chew in the crust. It has softened enough that it was better than day one, but I'll try adjusting down between 1 cup and 3/4 cup and see what I think. I've also switched to my pure liquid sucralose from McNeil - about 3 drops from a liquid dropper seemed to do it with my other sweeteners. That Splenda bite still comes shining through in the batter but seems to bake out as there is no distinct taste even as the product ages.

[JSkilling]

I've had some very good erythritol ice creams.  use of pdx or inulin wasn't necessarily required in that application, as the cooling effect of erythritol, well... it's ice cream 8-)  plus, too much pdx or inulin in a high water application tends to get gummy.  you'll never get a hard ice cream with erythritol, as molecularly it's very small, which means it'll depress the FP quite a bit.  when working with HIS, i tend to prefer sucralose, but it can be difficult (and expensive) to get nowdays.  there's also a fair amount of synergy that can be realized by using sucralose in conjunction with ace-k, but personally i don't care for ace-k straight.  Neotame is extraordaarily sweet, and seems to have good flavor, but because it's so sweet, it can be hard to formulate with - however there are companies out there that are diluting it just for that reason.  There's also a company out there (the name escapes me at the moment) who's extracting neohesperadin dihydrochalcone from orange peels (NHDC - forget trying to say that 5x real fast, i'd give you a dollar if you could do it 1 time real slow!), and labelling it as a natural flavor (very label friendly) - it's got a decent flavor, but use is restricted by law (in the us at least), to 10 ppm, and is really meant to be used as a flavor enhancer, not an sweetener, although it does have tremendous sweetening capabilities.  In europe, you can use as much as you want of it for sweetening.  The company, however, is saying it's not NHDC and that you can use as much of it as you want, which concerns me a bit as I've analyzed it already and know that it is, and frankly have issues with a company behaving like that.

←

I'd have to pay you the $ because I can't get past the first syllables

I used PDX in a raspberry sorbet over the weekend and didn't find it to be gummy in the product. I did learn that I really have to keep the usage down low if I'm going to have fruit at 50%. I started with a base that measured out at 16 Baume and added about 1/4 PDX. I ended up at 27 and added water to get to 25 and decided to just spin it and see what I got. Very fluffy. And it's never going to freeze solid. The taste was fine and I've actually eaten some sorbet that was this fluffy in consistency, though it was made from all sugar and so sweet that it was tooth decaying immediately. Next experiment will take the Baume back down to 17-18 and adjust the water and fruit with the PDX to accomodate. Litesse has recommended 30% fruit, 69% water, 10% Litesse and some sweeteners % that can't remember off hand.

[redfox]

I am also very intrigued by tagatose -- if there's anyone around who can get their hands on some pure tagatose from Arla Foods, I'd sure be interested in buying some.

[scott123]

Thanks very much for the info. 

I use a Pacojet and make ice cream to order so scoopability is not as much an issue.  I use stabilizers in some conventional ice creams already.

Sure, I would like the recipe if you don't mind.  It is great to learn principles, but starting somewhere can be a big help!

←

A Pacojet, awesome! Gotta love it!

Here's the working recipe:

Frozen Custard Base:

3C. heavy/whipping cream (not UHT!)

1C. low carb milk (carb countdown whole)

.5 C. Erythritol

.75 C. Polydextrose

12 drops liquid Splenda (1/2 C. sweetening equivalent)

1 tsp glycerin

.5 tsp liquid lecithin

.5 tsp xanthan gum

4 egg yolks, beaten

Sprinkle xanthan gum into cream/milk while whisking vigorously. Heat everything but the eggs in a double boiler until PDX/E are dissolved. Let cool momentarily. Slowly whisk in the eggs and continue whisking until the mixture coats the back of a spoon. Strain.

Chill in a sealed container. Makes 2 medium-sized batches of ice cream in a 1qt. ice cream maker.

-----------------

Notes:

This recipe was formulated for people with a small selection of sweeteners. It would hugely benefit from the addition of some ace k (no more than 6% sweetening equivalent), with a liquid splenda adjustment for synergy.

Sebastian is correct about polydextrose, when used in excess, making a gummier ice cream. I would call it 'chewier' as well. I've kept the erythritol relatively low in this because of my recrystallization paranoia. You might want to play around with increasing the erythritol/decreasing the PDX and see where that gets you.

This is an unflavored base. Vanilla is an obvious addition, as are low carb fruits, peanut butter, chocolate, you name it.

Edited February 5, 2005 by scott123 (log)

[scott123]

I've had some very good erythritol ice creams.

←

What kind of percentage of E are we talking about here? To achieve a sugary mouthfeel from just E and no other bulking agents, that must jack up the quantity, does it not? 1:1 replacement for sugar, texturally? No recrystallization issues?

If I can up my E without it going sand/grainy in the final product, my scoopability issues would be gone forever and it would be less chewy as well. Tell me it's true! ;)

[scott123]

Josette, I'm happy the PDX is working out for you in your brownies. Would you post your working sorbet recipe? A raspberry sorbet sounds gooood.

[Sebastian]

I've had some very good erythritol ice creams.

←

What kind of percentage of E are we talking about here? To achieve a sugary mouthfeel from just E and no other bulking agents, that must jack up the quantity, does it not? 1:1 replacement for sugar, texturally? No recrystallization issues?

If I can up my E without it going sand/grainy in the final product, my scoopability issues would be gone forever and it would be less chewy as well. Tell me it's true! ;)

←

Lemme talk to one of my collegues who's got more experience here than i - i can eat it with the best of them, but haven't formulated a single etoh ice cream yet 8-)

[nathanm]

I just finished some experiments with low carb ice cream.

I made three ice creams.

High Carb

-------------

This is conventional high carb, adapted from the book Los Secretos de Helado by Angelo Corvitto

Milk 330 grams

Cream 50 grams

Non fat dry milk 30 grams

Egg yolks 60 grams (3 yolks)

Dextrose 90 grams

Sucrose 30 grams

Invert sugar 12 grams

Stabilizer 3 grams

Vanilla bean 1

Low Carb

------------

This is basically the high carb ice cream, with the sucrose sweetening equivalent of the sweetner mix recommended by Scott123

Low carb milk 330 grams

Cream 50 grams

Non fat dry milk 30 grams

Egg yolks 60 grams (3 yolks)

Polydextrose 110 grams

Erythritol 16 grams

Splenda 6 grams (0.2 cup sweetening equivalent)

Xylitol 16 grams

Stabilizer 3 grams

Vanilla bean 1

==============

Note that the amounts are scaled to fit in a Pacojet beaker, which is about 600 grams total. In both cases I mixed in a Thermomix blender and heated to 85C / 185F. There was no problem with the PDX dissolving at that temp, but it was reluctant to dissolve at room temperature. Then I blended in the egg and brought pack to 85/185, then froze. All of these were frozen hard, then spun in a Pacojet.

I didn't have liquid splenda - I used solid, so there is some malodextrin in there too, but that is not the end of the world. I'll order some liquid splenda shortly.

The high carb ice cream was excellent - not much of a surprise there.

The amazing thing (to me anyway, this was my first attempt) is that the low version was nearly identical.

The texture was identical - no problem there at all. There was no issue at all with being chewy.

In the aftertaste was a bit "blunter" than the high carb version - i.e. there was less of a finish, and the ice cream tasted less sweet at the end.

However, that isn't a criticism it simply tasted like a recipe with slighly less sugar - but only at the end. The inital taste was just as sweet however.

All in all this was a VERY successful test. Thanks to everybody on the thread. Frankly I had thought that low carb sweets would have to taste pretty bad - based mainly on the various low-carb bars which are not that great. However I was really won over. THANKS FOR THE SUGGESTIONS!

Texture wise it is pretty clear that polydextrose is not going to be ideal for non-sweet savory ice cream - I tried an experiment in that direction too and the texture was too chewy / mealy - just as discussed in the thread. I clearly need a small molecule additive that will depress freezing point a bit. The Erythritol and Xylitol in the sweet versions really made a big difference in that regard.

So, I need to find a low molecular weight non-sweet additive with a neutral flavor. Any suggestions?

[nathanm]

One thing I forgot - any suggestions on good internet sources for liquid splenda / sucralose, isomalt, inulin ?

[v. gautam]

Would anyone know if levo-sucrose is produced commercially? Any other information about food uses of levo-sucrose?

TIA

g

[scott123]

Nathan, I'm happy that your low carb ice cream experiment turned out so successfully.

Was the low carb version just a little bit sweeter?

Assuming the invert sugar you used was 1.3 times as sweet as sugar and the dextrose .75 times, your high carb recipe had about 112 g sugar equivalent or half a cup.

From what I can tell, your low carb version had about 1/2 C. sugar equivalent pre synergy, but after synergy was taken into account, it came out to .6 to .75 of a cup.

Which brand of low carb milk did you use? If you used carb countdown which contains ace k, the ace k would drive the synergy to that higher .75 value.

It would be helpful for me to know, as I'm attempting to track the effects of synergy for different combinations. So far it's mostly guessing.

If you're looking to bring the sugar/carbs down a bit, you might want to see if the recipe works without the nonfat dry milk - there's a few lactose carbs there.

Salt, being an electrolyte, is a much more powerful freezing point depressor than non electrolytes like erythritol, xylitol, and polydextrose. Can you increase the salt a bit in your savory ice cream?

The most reliable/cheapest retail source for liquid splenda is sweetzfree. Because of the splenda shortage, the owner has placed limitations on how much a customer can order per month.

I know a person in Thailand that can get their hands on the powder, but I can't say I know them all that well.

[nathanm]

Sorry for the delay.

The reference book that I used - Los Secretos del Helado, calculates sugar equivalents and uses a 1.3X for invert sugar, but 0.70 for dextrose rather than 0.75. So that works out to be 108 grams of sucrose equivalent. I scaled the low-carb sweeteners accordingly.

I used Carb Countdown.

The subjective aspect is that it did NOT taste sweeter than the sugar version. This is somewhat personal I realize.

I also made another batch of ice cream based on the recipe that Scott123 provided - more on that later...

[misstenacity]

I also made another batch of ice cream based on the recipe that Scott123 provided - more on that later...

←

BUMP - I just finished reading this whole thread and am curious to hear how your later ice cream experiments turned out.... Thanks!

Andrea

http://tenacity.net

[nathanm]

I also made another batch of ice cream based on the recipe that Scott123 provided - more on that later...

←

BUMP - I just finished reading this whole thread and am curious to hear how your later ice cream experiments turned out.... Thanks!

Andrea

http://tenacity.net

←

Sorry, I got busy and did not post all of my experimental results. My experiments have been very successful. As others have posted, one key issue is using a blend of several sweetening agents.

Another is to have a very good base recipe - I used a vanilla here, but with real beans and a proper creme anglais base. The stronger the flavor, the more likely it is that you will overcome subtle differences in taste between good old sucrose and the blend of sweeteners.

Adjusting the sweetness is important too - if the ice cream is too sweet you are more likely to have a subtle flavor difference come across.

Finally, texture is important too. That seems to be the main contribution of polydextrose, which by itself is not that sweet. I use a pacojet which has a different set of issues than a conventional churn based ice cream freezer.

[abooja]

Resuscitating an old thread...

I recently baked a red velvet cake with PDX, and was wondering if anyone had any suggestions as to what went wrong.

Some background: After reading this thread around a year ago, I managed to make some pretty darn good low-sugar ice creams for my husband. I use scott123's ratio of 1 cup of PDX, 4 tablespoons erythritol, 4 tablepoons xylitol, and around 20 drops of liquid Splenda for every cup of sugar in a recipe.

I realized that my use of sugar substitutes is limited to ice cream, and that I should give low-sugar baking another shot. The only other time I tried, I baked low-sugar, gluten free brownies that were okay, but not to-die-for. I have no way of knowing if it was due to the sugar substitutes, the gluten free flours, or some evil symbiosis of both.

So, I went ahead and did it again. I baked a gluten free, low-sugar red velvet cake, based on Jaymes' excellent recipe, the original of which I've baked and loved. It was red and moist, rose well, and was topped with full sugar cream cheese icing, so that might have helped convince my husband that it was good, but I thought it tasted...burnt. I had to bake it a bit longer than the recipe states, since I made it in a 9" x 13" sheet pan, but it didn't actually burn. It just tasted that way.

I think it was the food coloring. Last time, I used two full tablespoons of powdered red food coloring, which was probably overkill, but I never figured out how to convert liquid to powder measurements. Nevertheless, the cake did not taste bitter or burnt, just delicious. This time, I used less -- one slightly heaping tablespoon. My gluten free flours (brown rice, potato, and tapioca) were not rancid, and I used the same PDX ratio as above, just doubling the amount. I'm guessing that the sweetness of the cake was not enough to overpower the bitterness of the food coloring. What do you think?

By the way, I sifted the PDX, erythritol, and xylitol in with the other dry ingredients, and it incorporated beautifully.