GCMS and HPLC -- any relationship?

[Photon]

Paying for both GCMS and HPLC is very costly.
170 + 120 = 290.

I was wondering if there is a way to estimate HPLC mg/ml, via GCMS.
These are the GCMS that can be found on this forum, credit goes to @MuscularMD for doing most of them.

		A	B	C	D = A - C	E = D / A
		HPLC MG/ML	GCMS %	B * 1000	Variance	Variance %
PRIMAL	TC	204	19.98%	199.8	4	2%
DEVON	TC	255	26.36%	263.6	(9)	-3%
WWB	TE	273	28.85%	288.5	(16)	-6%
AXLE	TC	302	34.36%	343.6	(42)	-14%
COMPOUNDED RX	TC	194	23.88%	238.8	(45)	-23%
Arail	TNE	47	3.57%	35.7	11	24%
SSA	BU	535	62.71%	627.1	(92)	-17%
Scholar	TRA	108	12.43%	124.3	(16)	-15%

Looking at PRIMAL, DEVON and WWB, you can see that the variances are pretty small.
However the variances between the others are much wider.

Anyone got any inside into why this could be the case?

 

[readalot]

Paying for both GCMS and HPLC is very costly.
170 + 120 = 290.

I was wondering if there is a way to estimate HPLC mg/ml, via GCMS.
These are the GCMS that can be found on this forum, credit goes to @MuscularMD for doing most of them.

		A	B	C	D = A - C	E = D / A
		HPLC MG/ML	GCMS %	B * 1000	Variance	Variance %
PRIMAL	TC	204	19.98%	199.8	4	2%
DEVON	TC	255	26.36%	263.6	(9)	-3%
WWB	TE	273	28.85%	288.5	(16)	-6%
AXLE	TC	302	34.36%	343.6	(42)	-14%
COMPOUNDED RX	TC	194	23.88%	238.8	(45)	-23%
Arail	TNE	47	3.57%	35.7	11	24%
SSA	BU	535	62.71%	627.1	(92)	-17%
Scholar	TRA	108	12.43%	124.3	(16)	-15%

Looking at PRIMAL, DEVON and WWB, you can see that the variances are pretty small.
However the variances between the others are much wider.

Anyone got any inside into why this could be the case?

Jano dropped price for combo to 240 usd after I asked. 50 bucks off.

Will look at the rest when I can. Short answer is no.

Edit:

HPLC gives absolute number.

GCMS will only give you relative info the way it is currently set up. The %API of total is a function of volatility and ionization amount of the other stuff that comes with it.

 

[Photon]

Jano dropped price for combo to 240 usd after I asked. 50 bucks off.

Will look at the rest when I can. Short answer is no.

Edit:

HPLC gives absolute number.

GCMS will only give you relative info the way it is currently set up. The %API of total is a function of volatility and ionization amount of the other stuff that comes with it.

Yes i agree, but there shouldn't be huge variances in the "other stuff" if all one is using is MCT BB BA Raws (+EO for China oils).

Unless ofc the raw purity is crap and there's alot of extra stuff.
Similarly, doesn't this also mean that the lesser the variance between GCMS% and mg/ml, the cleaner the brew?

The combo is missing the comments(?), which is an issue if there are false positives.

 

[readalot]

Yes i agree, but there shouldn't be huge variances in the "other stuff" if all one is using is MCT BB BA Raws (+EO for China oils).

Unless ofc the raw purity is crap and there's alot of extra stuff.

The combo is missing the comments(?), which is an issue if there are false positives.

Are they all MCT? Could build a linear model if you knew all the other compositions.

I like where you are going with this.

 

[Photon]

Are they all MCT? Could build a linear model if you knew all the other compositions.

I like where you are going with this.

I guess if we separate by oil then we need more data for any sort of modeling
Table can't fit, i had to use an image..

 

[janoshik]

If we could get reliable readings from GCMS, we'd do that.

The injection process and ionizations vary too much and you all are spoiled by our average RSD already )

 

[readalot]

You can view full size by clicking on each thumbnail. Fantastic work compiling these data. No added adjustable parameters. Just straight linear model between GCMS and HPLC.

Actually I am shocked, but that is why you plot the data and run the regression....

My added values in blue.


Linear Regression Plot



ANOVA results


Line fit plot


We can actually predict the HPLC result within 8% "on average" (RMSE) based on the 8 data points.

Great job @Photon. You could almost just do the GCMS and get the HPLC mg/ml for free.

Incorporating more linear model parameters (oil type, etc) would most likely improve fit.

@janoshik, have you done this analysis previously??

 

[readalot]

If we could get reliable readings from GCMS, we'd do that.

The injection process and ionizations vary too much and you all are spoiled by our average RSD already )

The fit is actually incredible to be honest. I am surprised.

 

[Photon]

The fit is actually incredible to be honest. I am surprised.

We can try applying the model based off the Tren experiment.
Hard to find any GCMS data here.

 

[readalot]

We can try applying the model based off the Tren experiment.
Hard to find any GCMS data here.

yes that would be good test. I am surprised. The R2 and fit is incredible really.

Regression equation (line of best fit) on the regression plot.

 

[readalot]

We can try applying the model based off the Tren experiment.
Hard to find any GCMS data here.

if all those data you shared are tabled correctly, this is pretty wild.

 

[readalot]

Any feedback? Comments? Lol.

 

[TallandSmall]

No feedback other than I’d been wondering about this too. I’d almost be willing to trust the vendor’s HPLC for dosing if there is a blind GCMS that comes back “clean,” whatever that means. Based on your analysis, we may not be able to tell the mg exactly, but we might be able to tell if a product advertised at 200mg/ml is underdosed at only 100 mg/ml. If the GCMS suggests it’s most likely in the ballpark, can we trust the vendor’s HPLC?

 

[Photon]

Any feedback? Comments? Lol.

The goodness of fit is very good.
If you exclude the non MCT oils, the variance is <5%, but the sample size drops to 5 so the data isn't really good. It's been years since i touched this..lmao

 

[readalot]

The goodness of fit is very good.
If you exclude the non MCT oils, the variance is <5%, but the sample size drops to 5 so the data isn't really good. It's been years since i touched this..lmao

View attachment 343249

Is isn't good. It is fantastic.

 

[readalot]

No feedback other than I’d been wondering about this too. I’d almost be willing to trust the vendor’s HPLC for dosing if there is a blind GCMS that comes back “clean,” whatever that means. Based on your analysis, we may not be able to tell the mg exactly, but we might be able to tell if a product advertised at 200mg/ml is underdosed at only 100 mg/ml. If the GCMS suggests it’s most likely in the ballpark, can we trust the vendor’s HPLC?

Incredible fit. That's where we are headed it appears. Thanks for all your data and support!!

 

[readalot]

We can try applying the model based off the Tren experiment.
Hard to find any GCMS data here.

I gotta go back and add QSC data. I think there are some pairs (?).

 

[readalot]

3 more pairs

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 8, 2025

We have seen the 3 different GC MS results performed on the 3 samples of Testosterone Decanoate, Nandrolone Decanoate, and Boldenone Undecylenate, it is entirely reasonable and fair to request a GCMS test from any customer who complains of a certain impurity, non-standard and non-common, for any request for correction or compensation, especially if there is any accusation of foreign contamination, which is something we see quite often from whining raws newbies, the harm reduction community should request systematically a GCMS from these kind of members, the same way we request an HPLC...

• Qingdao Sigma Chemicals

• 
• 

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 8, 2025

Deca 200mg/ml GCMS analysis of impurities: Janoshik Analytical

GCMS enhanced testing attached.

Nandrolone decanoate is famous for being a low purity raw material.
We reached +96% on several batchs, but in the majority of cases, it's about 92%-94%.
GCMS is an interesting way to check the impurities.

No harmful impurities detected. The GCMS compounds detected are the same found in the Testosterone Decanoate GCMS analysis performed on our TD 250mg/ml sample.

• Qingdao Sigma Chemicals

• 
• 

Post in thread 'Enhanced Testing Database'

Jan 8, 2025

• Qingdao Sigma Chemicals

• 

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 7, 2025

Test Decanoate finished oils results.

An extra GCMS Enhanced testing was added due to suspicions on the quality of raws, tested 90.16% by Deadbeef:

The finished oil was brewed earlier, before his test, assuming the purity is 99%.

The HPLC finished results ended up accurately dosed.



GCMS testing shows no contamination.

The lines 2 and 3 are to be ignore, because it's from Janoshik previous GCMS test of another sample.
17.beta.-Acetoxyandrosta-4,6-dien-3-one is a common...

• Qingdao Sigma Chemicals

• 
• 

 

[Photon]

3 more pairs

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 8, 2025

We have seen the 3 different GC MS results performed on the 3 samples of Testosterone Decanoate, Nandrolone Decanoate, and Boldenone Undecylenate, it is entirely reasonable and fair to request a GCMS test from any customer who complains of a certain impurity, non-standard and non-common, for any request for correction or compensation, especially if there is any accusation of foreign contamination, which is something we see quite often from whining raws newbies, the harm reduction community should request systematically a GCMS from these kind of members, the same way we request an HPLC...

• Qingdao Sigma Chemicals

• 
• 

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 8, 2025

Deca 200mg/ml GCMS analysis of impurities: Janoshik Analytical

GCMS enhanced testing attached.

Nandrolone decanoate is famous for being a low purity raw material.
We reached +96% on several batchs, but in the majority of cases, it's about 92%-94%.
GCMS is an interesting way to check the impurities.

No harmful impurities detected. The GCMS compounds detected are the same found in the Testosterone Decanoate GCMS analysis performed on our TD 250mg/ml sample.

• Qingdao Sigma Chemicals

• 
• 

Post in thread 'Enhanced Testing Database'

Jan 8, 2025

• Qingdao Sigma Chemicals

• 

Post in thread 'Qingdao Sigma Chemical Co., Ltd (International, US, EU, Canada and Australia domestic'

Jan 7, 2025

Test Decanoate finished oils results.

An extra GCMS Enhanced testing was added due to suspicions on the quality of raws, tested 90.16% by Deadbeef:

The finished oil was brewed earlier, before his test, assuming the purity is 99%.

The HPLC finished results ended up accurately dosed.



GCMS testing shows no contamination.

The lines 2 and 3 are to be ignore, because it's from Janoshik previous GCMS test of another sample.
17.beta.-Acetoxyandrosta-4,6-dien-3-one is a common...

• Qingdao Sigma Chemicals

• 
• 

I added the data in.
If i run the linear regression using only MCT oils (yellow), then use it to estimate for everything, it is pretty accurate for MCT and MCT related (+EO) oils. I'd be curious to see the Tren MCT GCMS data and to see if it works on it.

y=707.4012413x+67.85563911

 

[readalot]

I added the data in.
If i run the linear regression using only MCT oils (yellow), then use it to estimate for everything, it is pretty accurate for MCT and MCT related (+EO) oils. I'd be curious to see the Tren MCT GCMS data and to see if it works on it.

y=707.4012413x+67.85563911

View attachment 343293

You just broke MESO! Lol!!