Sciencemadness Discussion Board

MSDS are inaccurate in 30-100% of cases (study)

Fyndium - 26-10-2020 at 10:29

https://pubmed.ncbi.nlm.nih.gov/18651574/
Quote:

Of the 280 unique articles retrieved, 24 fit the review criteria. Eligible articles included a range of methodologies: laboratory analyses, site audits, surveys and qualitative inquiry. Articles were grouped into three main topic categories: accuracy and completeness, awareness and use, and comprehensibility. Accuracy and completeness were found to be relatively poor, with the majority of studies presenting evidence that the MSDSs under review did not contain information on all the chemicals present, including those known to be serious sensitizers or carcinogens. Poor presentation and complex language were consistently associated with low comprehensibility among workers. Awareness and use of MSDSs was suboptimal in workplaces where these factors were studied.


When assessing sources for materials, MSDS can only be used as a reference for so far. I have faced multiple cases where MSDS were off and they contained mystery compounds and response during purification or processing was unexpected. Paint thinner that was supposed to contain toluene, did not, and meanwhile the other thinner consisted basically of toluene and minor fraction of acetone, while it was supposed to not contain any of either.

Then there are these MSDS that go as far as stating "0-50%" or "50-100%" and I have once seen "0-100%" content. Usually it just contains this one reagent in this instance, but technically it gives leeway for zero quality guarantee and they can use pretty much anything that is a) a matter in solid, physical form b) might be white, black or something between. In food industry, this is seen written in a peanut bag in form "might contain peanuts".

For example, I just processed a sample of fertilizer that, according to MSDS, contains about 50% of KNO3 and few % of urea phosphate, but when dissolved, an insoluble fraction of about 20% of total mass remains at 100C which was decanted off, and precipitate occurs immediately after heating is stopped; when 10% more water was added, the solution remained clear, and crystallized out. I'm yet to determine if the desired reagent content might actually be higher, or does it co-precipitate something else too.

Ubya - 26-10-2020 at 10:45

Yeah i found about this the hard way. a weekend project turned to be a 2 years one, i just wanted to distill some DCM out of paint thinner, the composition on the label was just a big fat lie.

I found a guy on an italian forum that worked in a paint thinner facility years ego, he told me that they just mix whatever they can find on the market for cheap, they just care that the end product behaves as a thinner. They can't print a new label for each batch, so they just make one that maybe has most of the compounds inside the product.
I tried distilling a can of paint thinner that had "only" toluene and DCM, what i found through days of distillations, and different testing methods is that there is toluene and DCM, but also at least acetone, methanol, and i suppose naphta as the boiling point of the thinner at the end of the distillation got near 180°C.

I really can't stand the blatant lies written on those products, but without official testing i can't even argue with those companies

valeg96 - 26-10-2020 at 12:09

Distilling any paint thinner or solvent that is not clearly labeled as "Acetone" "Toluene" "Tetrachloroethylene" "Xylene" and so on is an utter waste of time. I once advised a very stubborn friend not to do it. The result? Basically a continuous fraction from 40°C to 120°C. If you are lucky you get a low boiling fraction with all the ketones, alcohols and esters and a high boiling fraction with all the alkenes and arenes.

For example, what is listed as "C9 hydrocarbons, aromatic" (918-668-5) isopropylbenzene, mesitylene, ethyltoluene, propylbenzene and methyl-phenylethylene in random proportions. Even if a paint thinner was just C9 hydrocarbons, aromatic, there's no way you can fractionate all of those at home without tears and blood. Add some alkanes and you're done for. And the same goes for lighter fractions.

MSDSs are just a way to legally classify your product in a hazard category. They rarely reflect the actual composition, just serve to state that there is some toluene, some butyl acetate, some MEK, so you get to know the hazards and safety measures involved.

Thinners are basically the cheapest petrochemical fraction available, diluted with what ester or ketone is best available at the time. And the same goes with any "mixture" that is not expressely labeled.

Also, if a MSDS was accurate, there couldn't be much industrial secret: think about specialty cleaning products or specially formulated industrial mixes, such as electroplating solutions.

Fyndium - 26-10-2020 at 12:26

I suppose almost any composition can be determined via analytical methods, including but not limited to NMR. So there should not be that much of a trade secret.

Sometimes I also wonder if some products are intentionally adulterated or composed so that their believed misuse would be made hard or too expensive. (Apart from denatured alcohol and colored fuel oil)

Ubya - 26-10-2020 at 13:17

Quote: Originally posted by Fyndium  

Sometimes I also wonder if some products are intentionally adulterated or composed so that their believed misuse would be made hard or too expensive.


well i wanted to use paint thinner as a reagent, can't use it as there's so much random crap it is impossible to get a pure product without days of esotic purifications.

I think some psychoactive seeds are sprayed with toxic compounds (i've read they used mercury compounds in the past), not sure if to protect the seed or to warn a potential abuser

valeg96 - 26-10-2020 at 14:30

Quote: Originally posted by Fyndium  
I suppose almost any composition can be determined via analytical methods, including but not limited to NMR. So there should not be that much of a trade secret.


They can, but they also can't. For example, a water-based floor polish I can't name is made, according to the MSDS, of "nothing dangerous to the user on the environment in sufficient amounts to warrant a declaration, according to directive (UE) 1272/2008 (CLP)". That's a good way to keep it secret. As a matter of fact I know that it contains mostly water, polyethoxylated tridecylphosphate, isoalkanes C9-C12, and fillers. Assuming there is only the monoethoxy surfactant (as if!), there still are about what, 30 possible branched alkanes, and what, a bunch of buffers and fillers? Good luck finding a company that invested in analytical chemistry to sort that out, especially at a quantitative level. And this is something very simple! Try working out what a concrete additive contains: there could be dozens of molecules in minute amounts.

Edit: and let's not forget the actual formulation process, which can produce very different products depending on the conditions, order of addition, mixing, and so on.

Edit: I'm now very tempted to make a proton NMR of a synthetic solvent mixture to see the kind of crap that shows up. Unfortunately I don't have any, though.

[Edited on 26-10-2020 by valeg96]

[Edited on 27-10-2020 by valeg96]

Fyndium - 26-10-2020 at 22:39

Of course no one would do that for fun, but if someone sees a profit margin from industrial espionage, sure. I actually believe analysis is very widespread in this matter, and also for controlling product quality, verifying authenticity and source, and also as an investigative tool.
Quote: Originally posted by Ubya  
well i wanted to use paint thinner as a reagent, can't use it as there's so much random crap it is impossible to get a pure product without days of esotic purifications.

I think some psychoactive seeds are sprayed with toxic compounds (i've read they used mercury compounds in the past), not sure if to protect the seed or to warn a potential abuser


Likely what's said above, paint thinner being a general term for anything that "thins", aka basically anything that has solvent properties.

Poisoning drugs nowadays? It has generally been a trend to use non-toxic, but non-potable denaturants, for example methanol is not usually used in europe for ethanol denaturation, because someone will drink it anyways, either by accident or intentionally, and poison themselves, and treating such patients can quickly cost hundreds of thousands, not including possible permanent disability.

valeg96 - 27-10-2020 at 01:46

Quote: Originally posted by Fyndium  
Of course no one would do that for fun, but if someone sees a profit margin from industrial espionage, sure. I actually believe analysis is very widespread in this matter, and also for controlling product quality, verifying authenticity and source, and also as an investigative tool.


Yes, it's widespread, but I assume very few do it seriously. Small local companies that survive on formulation chemistry can't afford that much expertise, and would rather just develop their mixtures. And in some fields, like steelmaking, the key factor is the secrecy on timing and processes, and much less on the actual composition, which can be determined fairly easily. I remember once visiting a top-notch steelmaking plant (we're talking about steel for NASA, not pig iron for manholes) and the only person that knows the production steps for a specific steel is a supervisor that uses a recipe booklet that must then be returned to the company safe after every run and never left unattended.

I believe seeds were doused with Hg and As salt because of their insecticide and anti-mold properties; this was very common in the 1800s where As and Hg were used virtually everywhere.

Denaturants used to be purposefully toxic (see back in the 1920s when gov't scientists were rushed to find the most toxic and unremovable denaturants possible to help the prohibitionism) but now they are just really bad-tasting substances like denatonium benzoate, which is kinda hard to remove from rubbing alcohol even by distillation, and the occasional ketone. Ironically, methanol is harder to come by in Europe than ethanol.

macckone - 27-10-2020 at 19:32

MSDS are only required to list 'hazardous' ingredients.
For some definition of hazardous.
They can also exclude 'trade secret' ingredients as long as the hazard is listed.
Ie. may cause cancer, flammable, toxic, etc.

If you are having to scrounge chemicals, you have no choice but to distill things like gasoline to get solvents and reactants. Toluene for example is hard to come by now in colorado but available in neighboring wyoming. However 93 octane gasoline has a good percentage but it is hard to separate. While xylene with minimal adulterants is readily available and can be partially oxidized and decarboxylated much easier. Or heated with benzene and anhydrous aluminum chloride, then distilled to get toluene. The yield of both methods sucks but it is better than distilling gasoline. Of course I can drive across the border and get three gallons in less time than it takes to make the aluminum chloride necessary. For some reason the place I get it always stocks three gallons. I suspect they order six and someone always buys three as it is used as an octane booster.

Fyndium - 8-11-2020 at 12:33

Another fancy MSDS surprise.

Thinner that clearly states everywhere, including on the main etiquette that it contains 80% of toluene and 20% of acetone, showed variable results.

First, a water and brine wash reduced the volume about 20%, which was very promising. After fractional distillation though, about half the volume distilled before 110C, and the rest right on point. I have faced this same ratio in other thinner products as well, so I suspect they all originate from the same source.

I'm not quite sure what it contains, because it starts to distill at around 55C and slowly creeps up to 110c without any clearly noticeable plateaus. To be honest, I did not watch it all the time, just hung around until the thermometer alarmed at 110 and I changed the receiver. For the price, 50% recovery is not bad, but the mislabeling just annoys. I can, though, use the lower fraction as an actual general purpose rinsing solvent for all kind of crap.

As an extraction solvent, toluene shouldn't be that bad if it has other non-polar compounds with it, but as a reagent it'd be a different matter.

monolithic - 8-11-2020 at 18:12

Quote: Originally posted by valeg96  
Quote: Originally posted by Fyndium  
I suppose almost any composition can be determined via analytical methods, including but not limited to NMR. So there should not be that much of a trade secret.


They can, but they also can't. For example, a water-based floor polish I can't name is made, according to the MSDS, of "nothing dangerous to the user on the environment in sufficient amounts to warrant a declaration, according to directive (UE) 1272/2008 (CLP)". That's a good way to keep it secret. As a matter of fact I know that it contains mostly water, polyethoxylated tridecylphosphate, isoalkanes C9-C12, and fillers. Assuming there is only the monoethoxy surfactant (as if!), there still are about what, 30 possible branched alkanes, and what, a bunch of buffers and fillers? Good luck finding a company that invested in analytical chemistry to sort that out, especially at a quantitative level. And this is something very simple! Try working out what a concrete additive contains: there could be dozens of molecules in minute amounts.

Edit: and let's not forget the actual formulation process, which can produce very different products depending on the conditions, order of addition, mixing, and so on.

Edit: I'm now very tempted to make a proton NMR of a synthetic solvent mixture to see the kind of crap that shows up. Unfortunately I don't have any, though.

[Edited on 26-10-2020 by valeg96]

[Edited on 27-10-2020 by valeg96]


They've analyzed simple solvents but you might find this enjoyable. http://ilrc.ucf.edu/sample_detail.php?sample_id=2