The name of the game here is to be sure to try it on your chocolate before buying.  Everyone's pushing online calibrations (ie it reads your sample, and adds it to it's baseline calibration file with the argument that it makes everyone's stronger).  Because of the high matrix sensitivity of NIR, "Jimmys" calibration curve doesn't mean anything really to "Johnnys" actual chocolate - as Jimmy has different process, raw material sourcing, and formulation than Johnny does.

Their argument is sort of a 'wisdom of crowds' argument applied to calibration.  It just doesn't work that well i'm afraid.  if you've got a very similar type of chocolate and process it works fantastically well (in fact i've built networks of matrixed NIR nodes all around the world to do just that) - but as a general 'catch all' for anything chocolate - not so much.

Spectral Engines has a new NIR instrument: http://www.spectralengines.com/applications/food-composition that has some interesting (at least to me) aspects.

Unity Scientific also has an NIR instrument - but importantly it has databases of characterized samples that could serve as baseline measurements. I think the machine is under $30k.

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@DiscoverChoc

Remember that NIR is VERY matrix dependent (and temperature dependent).  Be sure you have a calibration curve for each type of chocolate that you're plan to run (ie, a generic 'milk chocolate' calibration won't be very accurate for all milk chocolates, and certainly not for dark or white chocolates).  Run all your samples at a fixed temperature (including the plate the samples will sit on - probably easiest to get a small warming oven and set it to 50C and just keep everything inside of it).

good luck!

Hi Sebastian,

Thank you for your opinion.

I will need to do about about 3 on average and a maximum of 7 test per day, over 2 shifts so I may start with gravimetric and if this takes too much time buy the instruments. Based on your point about impact of metal content on results NIR looks like a better option.

Peter 

Sorry - one more thing that's relevant - how many/how often do you expect to do this testing? If it's once a day, i'd say go gravimetric.  You'll need gravimetric to validate the calibration curves of the instruments anyway.  If you need to run 5 samples every hour - than you'd  need a small army of tech to prepare the samples, and an instrument's your way to go.

Lots of pros and cons here.  A few highlights to consider:

NMR and NIR will be be expensive instruments to acquire ($50-90k?).

NMR can be heavily influenced by metals content - so your process becomes very important.  IE if you have a ball/attritor mill, the media will abrade off small pieces of metal and will impact your results.  

NIR is heavily matrix and temperature dependent - you'll need to have a robust calibration curve for accuracy.  Actually that's true with NMR as well.

If you have the funds to spend, and the expertise to create and maintain a calibration curve - i prefer the NIR, as you can calibrate it to predict other things as well.  I've never used Oxford instruments specifically, but that they indicate in their literature that you only need 2 calibration points would concern the hell out of me.

A 3rd option is gravimetric extraction.  Grinding your sample up, extracting the lipids with a solved, evaporating the solvent, and measuring what's left behind.  Far less expensive up front costs, but you are now dealing with a solvent, and the consistency of the tech doing the prep work is important.

Hi Sebastian,

I understand that by NIR you mean Near Infrared Transmission Spectroscopy. 

I need to set up a lab which will be able to do reasonably accurate (no more accurate than +/- 0.5%) measurements of cocoa butter content in cocoa mass.

We use blends of in house processed beans in our recipes and I need to adjust mixer contents to variation of cocoa butter in the beans to improve chocolate consistency. I have just started to look around for solutions and uncle Google shows this:

https://www.oxford-instruments.com/OxfordInstruments/media/industrial-analysis/magnetic-resonance-pdfs/Determination-of-Total-Fat-Content-in-Chocolate-and-other-Cocoa-Derivatives.pdf

Would you have an opinion on which method would work better?

Peter

Sure - put it into a calibrated NIR and push the start button.  

I assume you mean an inexpensive, easily accessible way for the home user?  No. Perhaps make friends with a local university and ask them to make it a class project to calibrate their equipment to your product, in exchange for some free product...

I wonder if anyone knows of an easy way to measure cocoa butter content in cocoa liquor, or in chocolate

For what its worth, you can often find 'specification sheets' for callebaut products online. Just search for "product name specification".

The spec sheet will often include both the % cocoa mass and the % non-fat cocoa solids.

You can use these spec sheets to double check your calculation technique for 'extracting' the relevant info from ordinary nutrition information.

Finally, several of the Callebaut dark chocolates are made with cocoa mass, defatted cocoa powder, sugar, vanilla, lecithin. These are not high end single origin chocolates, rather middle of the road mass market ingredients. These chocolates are very, very thick when melted and have relatively poor texture, but are quite nice for doing things like make ganache where the extra flavor hit per unit fat is a benefit.

-Jon

Also, in your example you have "1. 56g total wt". But the nutritional information is based on a serving size. For Dagoba bars the whole bar is usually 56g but the serving size may only be half of that. Was your example using Serving Size from the nutritional info or the total size of the bar?

Drew,

Thanks for the reply but I think you missed a crucial point that wasn't made clearly enough in my OP. You can only do this with dark chocolate that has no other inclusions such as milk powder, fruit, nibs, or nuts added.

The only 37% bars I have ever seen are milk chocolate bars, so they have milk powder and other ingredients added, including some that have fat in them. You CANNOT use this method with any bars that have inclusions. Does that make sense?

A better explanation is below:

---

Its actually pretty easy to figure out the percentage of cocoa butter in a chocolate bar. Note, though, that you can only do this with dark chocolate that has no other inclusions such as milk powder, fruit, nibs, or nuts added. You also have to have a package with the nutritional information that includes the fat content (in grams) because the fat is from cocoa butter. Follow these steps:

1) Note the serving size (in g)

2) Calculate the amount of cacao per serving by using the cacao percentage. Keep in mind that the total serving size = cacao + sugar + (optional other ingredients, 1-2%).

The amount of cacao = serving size X cacao percentage (as a decimal number).

Ex- a 40 g serving X .75 = 30 g cacao.

3) Note the Total Fat (in g)

4) Divide Total Fat by Cacao Amount in step 2. This gives you a decimal. Then convert this decimal number to a percentage (i.e. 0.52 is 52%) and then you have the cocoa butter percentage.

An example, Fresco Dominican Republic 223, 72%:

1) Serving size = 45 g

2) Cacao percentage = .72

3) Weight of cacao = 45 x .72 = 32.4

4) Fat content = 16.2 g

5) 16.2/32.4 = .50, so 50% cocoa butter

ChocoFiles-

I think that you are incorrect not in the mathematics but in the assumptions. Let me follow your example with a lower % bar

Dagoba 37%:

1. 56g total wt

2. 20g fat wt / 21g cacao wt

3. 36% fat

4. 37%-36% = 1% or 97% butter

I can show you many more examples like this. In fact some 30% bars have a negative cacao %.

I am a chemical engineer and I did all of this before I posted my first thread-- which included my assumptions on what cacao% means. I also attached a rough spreadsheet which calculated weight percentages of each constituent. I believe that the OP incorrectly assumes that %cacao is without fat. I made a bold assumption on the fat% and when entered into my spreadsheet shows a wide variation of dry cacao and a very low variation on fat% which makes sense to me. It shows that the texture is well agreed upon and the % cacao is inversely related to sugar based on personal taste preference.

So, I used the method above and calculated the cocoa butter percentages for more than 300 bars... then I discovered that method was flawed, so I threw away all of that data. I had neglected the fact that the serving size was the total weight of cacao plus sugar plus any extra ingredients (vanilla and/or soy lecithin)

Here is the correct method:

1) Note the serving size (in g)

2) Calculate the amount of cacao per serving by using the cacao percentage. Keep in mind that the total serving size = cacao + sugar + (optional other ingredients, 1-2%).

The amount of cacao = serving size X cacao percentage (as a decimal number).

Ex- a 40 g serving X .75 = 30 g cacao.

3) Note the Total Fat (in g)

4) Divide Total Fat by Cacao Amount in step 2. This gives you a decimal. Then convert this decimal number to a percentage (i.e. 0.52 is 52%) and then you have the cocoa butter percentage.

An example, Fresco Dominican Republic 223, 72%:

1) Serving size = 45 g

2) Cacao percentage = .72

3) Weight of cacao = 45 x .72 = 32.4

4) Fat content = 16.2 g

5) 16.2/32.4 = .50, so 50% cocoa butter

I wouldn't be surprised if makers add cocoa butter solely to appeal to a certain clientele, namely those who favor texture over flavor andto ease the transition into higher percentages for those folks who are just tooreticent of going that route. chocolate can have a fantastic texture withoutoverloading the fat content with cocoa butter or vegetable oils or other"texture-enhancers." Particle size achieved through the grinding andrefining has much to do with the finished texture. A rough estimate of 50% iseven easier to calculate. I forget which beans have more fat but I think it isforastero from Ghana.At least I remember that the butter content in those beans is supposed to beharder, which makes them more suitable for milk.

"So why do think there's generally higher fat content in higher percentage bars? By adding extra cocoa butter the makers are decreasing their profit margin. It doesn't seem like it's for taste because IMO extra cocoa butter weakens the taste."

IMO, whether the cocoa butter is deodorized or not, I think each manufacturer uses whichever to achieve the flavor they wanted.

It may be true that cacao powder is more flavorful than cocoa butter, but I think chocolate is not just about "chocolatey" flavour burst.

The notes, the mouthfeel, and the aroma might be enhanced or muffled by the ingredient they use.

Too much cocoa powder might overpower the taste details and notes of flavor. That may be one of the reasons why they decided to use cocoa butter instead of powder.

CMIIW.

"I wasn't arguing the amount or intensity of the flavor of a cocoa butter, just that cocoa butter has flavor."

Forgive me for misunderstanding, Clay, because I thought that was precisely what you meant when you said here that "If the cocoa butter is pressed from the same beans that the chocolate is made from and the butter is not deodorized then the flavor isn't diluted."

Langdon:

I wasn't arguing the amount or intensity of the flavor of a cocoa butter, just that cocoa butter has flavor. I agree that the affect adding a deodorized cocoa butter has on the flavor of a chocolate depends on how "flavorful" the chocolate was to begin with and how much cocoa butter gets added.

You are right, few origin chocolate bars contain undeodorized cocoa butter made from the same beans as the liquor. The only origin chocolate maker in the US that I know of that has a butter press is Askinosie.

FWIW, it's my opinion that in order to be truly called a "single" origin chocolate it's necessary to use butter from beans of the same origin (deodorized or no), not butter from an unknown source. If the chocolate maker is making a single-plantation bar, then the butter needs to be from beans from the same plantation, not the same region or country, in order to be a "true" "single-origin."

:: Clay

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@DiscoverChoc

"I was pointing out that cocoa butter does have flavor."

Even so, Clay, x amount of cocoa butter has much less flavour than the same amount of cocoa liquor. So it's fair to say, as Lowe did, that extra cocoa butter weakens the flavour.

In some bars, where only a small amount of cocoa butter is added, the weakening effect might be negligible, but in other bars (like Choklat 70%, for example, which contains 30% added cocoa butter) the effect is undoubtedly significant, whether the cocoa butter is deodorized or not.

Also, much chocolate is made with deodorized cocoa butter anyway - e.g. milk and white chocolates.

It also seems safe to assume that origin bars would often contain deodorized cocoa butter, too - because no serious chocolate maker would want the flavour of their "Chuao" (or whatever) beans competing with flavourful cocoa butter from a completely different origin (very few artisan manufacturers press their own cocoa butter).

Langdon:

Technically true, Minifie et al and the USDA refer to the edible fat component. However,I am not aware of any commercially available cocoa butter that is not "contaminated" (in the sense that you use the word) by at least some amount of volatile aromatics. I was making the point with references to Lowe's impression that adding cocoa butter muted the flavor of chocolate - I was pointing out that cocoa butter does have flavor. I may not have chosen the clearest way to express that thought.

:: Clay

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@DiscoverChoc

Glad you found my reply useful.

I call the part of the cacao nib that isn't fat "non-fat cocoa solids", or NFCS for short.

Green & Black's White was a mistake that shouldn't have been in there. That was done early in my data collection process and has now been removed, as will any other White bars. White can't be included because of the other added fats. Thanks for pointing that out. Please let me know of any other discrepancies or errors.

From Langdon "what exactly does "Percent" indicate? And how about "Cacao fat"?"

Percent = the percentage of all cacao that is printed on the label

Fat = from the nutrition facts. Calculated from Fat grams / serving size and changed from a decimal to a percentage. e.g.- 14 g fat in a 40 g serving = o.35 or 35% fat

I'll try to explain somewhere where these numbers come from. Thanks for the advice.

Langdon,

Thanks so much for that long and very informative reply.

What do you call the part of the cacao seed that isn't fat?

More to reply to, but I'll start with that.

Lowe

updated by @chocofiles: 09/08/15 00:33:53

Lowe, in the linked thread about migraines, Clay said that "Cocoa butter has more than fat in it in the same way that butter has more than fat in it".

This is incorrect. Minifie, Beckett, and the USFDA all agree that the term "cocoa butter" means the _edible fat_ obtained from cocoa beans.

It might be difficult (or even practically impossible) to remove 100% of the non-fat compounds from cocoa butter, but that just means that your cocoa butter is "contaminated" with other compounds. It doesn't mean that cocoa butter isn't pure fat.

By contrast, "butter" in the more common sense of the word refers to a dairy product made from cream. The butter in my fridge is only 81% fat (most of the rest is water). Dairy butter also naturally contains a small percentage of protein and carbohydrates.

Also, Lowe, I've taken a look at the file on your website labelled "Cocoa Fat in Chocolate", and I have a couple of questions and suggestions.

First of all, it would be helpful for the columns to include clear unit measurements.

For example, what exactly does "Percent" indicate? And how about "Cacao fat"?

The lack of clear explanation of these terms makes the data confusing. For example, I don't know what you mean when you state that Green & Black's white chocolate is:

Percent: 30
Cacao fat: 47
Non-fat solids: 0

If the bar is 30% cocoa solids, and 100% of the cocoa solids is cocoa butter, then how did you arrive at the 47% figure? The only way that figure makes sense to me is if you've included milk fat in the "Cacao fat" column.

Also, you asked: "So why do think there's generally higher fat content in higher percentage bars? By adding extra cocoa butter the makers are decreasing their profit margin."

1. You can't simply assume that "higher fat" means "added cocoa butter" (for example, many bars also contain milk fat, and/or other vegetable fats, e.g. Green & Black's white chocolate, as mentioned above).

2. A bar with a higher percentage of cocoa solids would naturally contain a higher percentage of cocoa butter, if expressed as a _percentage of the total bar_ (this is because pure cocoa liquor is always roughly 55% cocoa butter - hence, the more cocoa liquor you use, the more cocoa butter comes along for the ride).

So, an 80% bar with no added cocoa butter would contain about 44% cocoa butter, while a 60% bar with no added cocoa butter would contain only about 33% cocoa butter.

3. Cocoa butter isn't necessarily a more expensive ingredient than cocoa liquor. For example, if you produce your own cocoa liquor on a relatively small scale from whole beans, you can probably purchase cocoa butter from a large manufacturer for a fraction of the price that it costs you to produce your own cocoa liquor.

4. A lot of people (women especially, I've noticed) prefer the taste and mouthfeel of chocolate with a generous amount of added cocoa butter - so it sells very well in certain demographics.

Is there anything other than fat in cocoa butter? Or is cocoa butter 100% fat?

Here's information I got from The Nibble : Cocoa butter is the natural vegetable fat present in the cacao bean... It solidifies into a yellowish-white fat, solid at room temperature." Is this completely accurate?

I'm moving some info over from another thread because I think this is better home for it:

Lowe:

Clay's reply:

By Lowe:

JCandy et al,

I've attached a file with info I've collected on the percent of cocoa butter and cocoa solids for some of the bars I've reviewed. I also give my rating of my enjoyment of these bars. In general, the bars with a higher fat content were lower on my enjoyment scale.

(If you're curious, "Class Rating" refers to a system I devised to compare similar type chocolates. Some of my "Classes" are Darks: <59%. 60s, 70s, 80s, 90s, 100%, White, Milk, Mint, Spices, Spicy Hot, Nuts, Nibs, Orange, Other... I did this as a way to compare bars of a similar group such as "White". This allows me to know which bars I liked best in a certain group even if I don't particularly like that kind of chocolate. "White" is a good example because I don't particularly like White chocolate but there are some bars I gave a White class rating of 10 because if I have a whim for a White bar then that's one I enjoyed the most. I hope this makes sense.)

You can also find more of my review files at Choco Files .

Well, I'm on a quest to find chocolate with high cocoa solids & low cocoa butter, then! I've seen that even among chocolates of the same "cocoa" percentages, the fat content can vary. So that would imply that "cocoa content" can refer to either solids or butter. My high-solids search continues! The proportion of cocoa butter in a chocolate is what determines it's fluidity when melted. This is important if the chocolate is to be used for dipping or molding, both of which require very fluid chocolate. It also effects the mouth-feel of the chocolate; low-fat chocolates tend to coat the mouth with a clingy residue. For a chocolate to go nice and fluid when it melts, it will have to have a cocoa butter percentage >38% (most fine covertures will be in the low 40's). the goal is to produce one of the edgier 85% cocoa chocolates, then the sugar would, of course , be only 15%. Here, a 40% cocoa butter content would result in a cocoa solids percentage of 45%, WAY too much. So the manufacturer will have no choice but to lower the solids and increase the butter, and he'll end up with something like 48% cocoa butter, 37% cocoa solids and 15% sugar. It will be very fluid and very expensive.

I was wondering how Domori does it?

Cocoa butter is a favor carrier, on its own it is tasteless.

Cocoa butter is essential as an ingredient as far as encapsulating the non-soluble particles, with insufficient fat, chocolate will not melt smoothly in the mouth. The ideal amount of fat is thus dictated by the particle size, smaller sizes 15-18 microns will require more butter to encapsulate due to a larger surface area. Even smaller sizes don't make sense as our tongues cannot identify particles beyond this range, and the chocolate will feel muddy.

Cocoa butter is priced between premium beans and commodity-grade beans. So for premium chocolates, quantity of cocoa butter will tend to be at least sufficient because it reduces cost to use cocoa butter. For commodity chocolates, cocoa butter may be insufficient if the producer wants to cut costs.

Left this out: we use a dry-type bacterial incubator to approximate the temperatures in larger natural ferments... we incubate with different temperature curves but typically between 35 - 50 C.

For small-scale ferments we've been using acrylic and polycarbonate cylinders with a perforated drainage plate and sweatings reservoir. We drop a disk of the same material into the cylinder to cap the fresh seed/pulp. I've experimented with inoculating using both yeasts and Acetobacter pasteurianus . With very small fermentations we've found that with no inoculation at all, it's a total crapshoot with regard to yeast/bacteria/mold competition... with the molds winning most of the time and ruining the entire mass. In contrast, too much yeast inoculant appears to completely inhibit the acetic phase, much to the detriment of the finished product. One of the signs of a complete ferment including Acetobacter spp. (besides the vinegar reek) is a rapid and easily observable reddish darkening of the pulp and seed coat.The harvest season in Hawaii is starting very soon, so there will be a lot more research regarding post-harvest handling and fermentation in Hawaii this winter.

Dan -Any photos of the equipment (vessels and incubator) you're using for the microferments? Also - are you "seeding" the piles with specific yeasts?:: Clay


--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@DiscoverChoc

The thought of single tree bars is mildly mind-numbing. How does one come across something so esoteric? Is this the ultimate frontier? Do the trees start getting names?In all seriousness, how different are the trees? I wonder whether they are different like children can be different. The appearances are rather distinct - are the flavors? This is very exciting...

Hey edward, i was just wondering how much nitrous oxide will you need to aerate a 1/3 bowl of chocolate? I think I would mix a layer of aerated choc with something else (biscuits/caramel..), for diversity's sake besides, you are absolutely right about the weight.

im guessing this formula only applies to dark chocolate since milk has fat content as well

Any fine bar out there now is going to have a fair bit of fat in it, regardless of whether they add separate cocoa butter. Often, it seems to me, extra cocoa butter seems to be for the smoothing of the texture (at best) or the dilution of unpleasant flavors (at the more "worst" end of the spectrum).Fortunately, chocolate can have a fantastic texture without overloading the fat content with cocoa butter or vegetable oils or other "texture-enhancers." Particle size achieved through the grinding and refining has much to do with the finished texture, and Domori has worked out how to have their chocolates feel like pure silk without refining them to death. Cluizel tends to be more on the VERY refined side, and I think some of the innate flavors suffer accordingly.

Whatever is right or wrong with Lindt - their "Excellence" bars do have real vanilla - and a fair bit of it.

If the density of the non-cocoa-butter solids and and cocoa butter is the same, that would mean there is about 10% (of the total weight of the bar) added cocoa butter (based on the normal just-over-50%-cocoa-butter percentage of the average bean). Since the densities of the cocoa butter and the remainder of cocoa solids are bound to be different, it seems worthwhile to compare that to a bar with no added cocoa butter like one from Domori or Patric.*pause to run to the kitchen**further pause to put the girls down*As it turns out, those bars don't have much by way of Nutrition Facts - guess where my priorities haven't been. Anyone know of a bar of chocolate with no added cocoa butter that has published Nutrition Facts? Also, with the number involved, I am guessing we won't get very accurate percentages.

Dan, were these trees of the same genetic stock?

Are we not forgetting the natual fat in the liquor? Or should it be considered insignificant?