Proper Decarb Time?

As I said in my last post, I checked that chart to see how it compared with two well-known, accurate decarbs, 240 F for 40 min, and 250 F for 30 min.
It came close enough to those that I decided to call that last column "hours."

There appear to be huge discrepancies between values listed in the paper, and the two most popular decarbs.
For example, if 230 F really does require 1.83 hours, there is no way 240 F for 0.67 hours can be correct.
In particular, the 250 F for 30 min (or something very close) is presented in one of the most popular decarb graphs ever, after scientific study.
This will require some study.

 

Yeah, I agree about the inconsistencies in the literature. I believe that one of the charts that circulates the most was based on an article published in 1990. Chemistry definitely isn't my field so I can't say which study was better or more accurate, and since the findings directly conflict I'm not sure what to make of it. The reason I posted though, is to clear up inaccurate information that you posted. I'm not trying to give you a hard time here, but your calculations are essentially meaningless and it seems inappropriate to post what is simply conjecture based on mixing commonly used methods with the results of a scientific study that really don't go together. I'd hate for people to ruin their medicine based on inaccurate information that you decided was close enough. If you wanna say 240F for 40 minutes or 250F for 30 minutes etc. because those are commonly accepted effective methods, then fine, but I think you should delete the list you posted. If you disagree that's fair, but at least the information is here for people to read.

 

One last post on the topic. I did a little more research and found a brief review article that looked at different studies measuring decarboxylation of THCA to THC. The 1990 study that shows 230F takes approx. an hour to decarb, but the 2 other studies cited with data on that temp found it takes much longer. If you look at the combined trend line it is very close to the 110 minutes found in the previous study I posted. I'm not saying this makes that study correct, just more data.



https://static.t-cdn.net/5f7bf9bee3...-25-Decarboxylation-of-THCA-to-active-THC.pdf

 

Sorry to post again about this, but you can't interpolate the data like you did. The reaction rate (seen in the chart I posted earlier) depends on the temperature and you cannot average them to find the solution.

 

Yes you can interpolate those data.
Interpolation is never 100% accurate, but in this case it's probably accurate to within few degrees.

Until proved otherwise, I stand by my statement that those reaction rate values in the last column can also be thought of as hours of time.
This is in line with those two well known decarbs of 250 F for 30 min, and 240 F for 40 min.
But those times should be viewed as approximate.

I also own Ardent Lift, whose ability to stay in business depends on accurate decarb.
That device runs for 1:45, including warm up and cool down.
According to Ardent, Lift's temp never exceeds 218 F.

Something is wrong here, but I haven't looked over that paper enough to figure out what.
It is certain that the decarb time you presented of 230 F for 1.83 hours is incompatible with what most people are doing. In this forum that's 240 F for 0.67 hours. The "reaction rates" column gives 1.0. If we adopt the approximation that those reaction rates can be viewed as hours, then 230 F for 1.0 hours is very much in line with 0.67hr @ 240 F and 0.5 hr @ 250 F.

 

It's really simple.
Ask someone who actually creates high potency meds day after day and needs them to function in the real world.
In my search to get the absolute best from my medicine I've tested every possible decarb and processing time.




Hundreds of grams of Kief decarbed 4 times a year.
8 years making the exact same oil.
Well over 30,000 capsules YES I typed Thirty Thousand.
If a better decarb existed I'd have found it as I've tested it ALL. Hotter, Colder, Longer, Shorter. Been down this road several dozen times.


First round
160 micron dry ice Kief.
230-240 F past 60 minutes doesn't change a damn thing. No improvement in stone at all. As you reach 2 hours degrade sets in and it gets progressively more sleep inducing the longer you decarb.

Second round

Third Round

And
240 F for 40 minutes for all of it.
Empirical testing is a thing.

<-- will put my work up against anything you can create.

BNW

 

This just isn't right. You can't take the formula they found and use it to make calculations when you are saying that the times the formula was based on are wrong. Without having some kind of confirmatory evidence, which you do not have, the values you provided are meaningless outside of the 250F for around half an hour that has been consistently verified. I want to emphasize that I am not trying to advocate for any particular time or temperature. I am not trying to say that what they found was correct; however, your determination to treat the multiplier column on time is based solely on your opinion, and there is no evidence to support that determination. Unless people have the resources to test the results, most people are probably best off using the standard 250F for 30 mins or 240F for 40 minutes that you and most people reference.

 

I completely agree with you, and this is exactly my point. 240F for 40 minutes is rock solid and well tested. Its important to realize that many of the time values in that chart were derived from reaction rates determined by regression models based on the temps and times that they did test. 230F was the lowest temperature that they actually tested. If you're reading this because you're interested in a long slow decarb, please do not use the numbers in that chart blindly with large amounts of medicine without testing.

 

You CAN interpolate anything you decide to, but that does not mean it is appropriate to do so. You say that, "it's probably accurate to within a few degrees," but 5 minutes (12.5%) is a large error from your interpolated estimate (240F for 40 vs 45 minutes).

 

Apparently you do not understand what interpolation is.
When there are two different values for something, but a value between them is missing, interpolation is appropriate.
Interpolation fails if the two original values are far enough apart that there isn't a definite linear relationship between them.
All I did was to interpolate the 240 F value that lies between 230 and 248 F.

Here are the mechanics of that interpolation, using the final column as hours:
230 F = 60 min
240 = ?
248 F = 30 min (approx)

The difference between 230 and 248 is 18 degrees.
The difference between 30 and 60 min is 30 min.
240 F is 10/18 of the way between 230 and 248.

The time will also be about 10/18 of of the way between 30 min and 60 min.
10/18 * 30 min = 17 min
So the interpolated time for 240 F is 60 - 17 = 43 min, which comes close to the trial and error method of 240 F for 40 min.

Until proved otherwise we should assume that 230 F for 1.83 hours is incorrect because it is so far astray from other accepted experimental values.

 

Interpolation may be used to calculate values that you do not have data for; however, the method of interpolation that you used is inappropriate for this data.

Herein lies the problem with your interpolation. Your interpolation method is assuming that the lines for different temperatures have the same slope, representing rate of change. In this case, it was found that there is a linear relationship between time and decarb, with the rate of change or slope depending on the temperature. What this means is that you can't interpolate the way you did, because there isn't a straight linear association of the slopes between differing temperatures. So while your interpolation method would yield fairly accurate results using data appropriately fitted to a regression line, for example, they are not appropriate or accurate with this data. The formula that the values in that chart were derived from is based on the kinetic energy present at differing temperatures as you can see below:

I agree. So logically it follows that if you consider their findings to be incorrect or dubious, you should also consider the chart based off of their results to be highly questionable as well. Please remove your original answer that includes times and temperatures that are not supported by evidence.

 

You are absolutely wrong about interpolation, and seem not to understand it.
Interpolation led to this result, assuming that the final column values are approximate times.
230 F = 60 min
240 F= 43 min
248 F = 30 min (approx)

240 F is SOMEWHERE between 230 F and 248 F.
Interpolation is simply a method of figuring out where exactly.
You might even agree that the temp 239 F, which is exactly midway, between 230 and 248, would have a time of 45 min, the exact midpoint (average) of those values.
That's the easiest kind of interpolation.

You did point out info that shows that the last column times that I said contained hours, should be viewed as approximate.
I now believe that there is a systematic error that makes those times usable as hours, rather than multiples of some other value , because they are in line with accepted values.

 

I'm trying to be nice and respectful about this, but you are making it hard. I took several advanced statistics courses in grad school, and I have a solid grasp on the interpolation issues that we are talking about. As a scientist whose field is not chemistry I may not know a lot about chemistry, but I know a lot about the scientific method, statistics, and the evaluation of evidence in order to draw conclusions.

Interpolation as you described only works with data that follows the same or similar linear trends as I mentioned. Interpolation is a way of *estimating* an inferred value that was unmeasured based on existing data. If you do not use the proper methods of interpolating, your estimate will not be accurate. A good example of this is shown in the graph from the review article that I posted. The lines aren't linear, because higher temperatures can lead to exponentially faster decarb. If we apply your interpolation method to the patent results shown in yellow on the graph (the only one we can do this for and know how accurate it is because it was measured 3 times) you will see what I mean. Ignoring whether or not the values seen are the optimal decarb time/temp combinations, using your methods will demonstrate the large amount of error in your estimate:

The actual observed data:

Data point 1: (140C/284F, 30 minutes)
Data point 2: (120C/248F, 120 minutes)
Data point 3: (103C/217F, 240 minutes)

284F - 217F = 67F
240min - 30min = 210min

248F is 31/67 (46.27%) of the way between 217F and 284F
210min * 0.4627 = approx 97

So the interpolated estimate is 127 vs 120 minutes. That might not seem like a huge discrepancy at these temperatures, it's only 5%, but it will be much larger at the extreme values, imagine what happens when you apply this method to the lower end of the temperature range in the list you posted. If you combine the error in your predictions with the fact that most people are using devices that do not maintain a consistent temperature, the use some of the low and slow combinations following the values on your list could very likely result in severe under or over-decarbing. I'm not really sure what you're gaining by leaving information publicly posted as a reference when you don't even have confidence in the findings.

Edit: see the wikipedia entry for polynomial interpolation for example:

Polynomial interpolation - Wikipedia