I am reposting the final plot graph for the 87 octane gasoline, I should have saved it in PNG form here it is again:

This is very interesting, but I am so new to this and I have a question or two. Why are you adding ethanol to the solution? Doesn't the gasoline have an ethanol content? Can this be used to determine the presence or absence of 2 stroke oil in a 95 octane gasoline sample?

Hey estabean, great question, well I use ethanol because it can dissolve both polar and non-polar compounds, what this means is, like oil, gasoline is a hydrocarbon (aromatic hydrocarbons to be exact.) which is non-polar (like in an electrical charge, positive and negative, although a little more complex than that.) if I just sample that gasoline "as is," I'm not going to be able to decifer any information from the peak because all the info is hidden inside, so I have to elute, or separate those components that I want to see.

Yes gasoline can contain ethanol but its usually 10% by volume, so it's concentration is very trivial. Yes if you wanted to see if there was 2 stroke oil contamination in a sample of gasoline there are steps you can follow in the data sampling, because you can read the signature peaks and wavelength data of the oil sample that I did.

Hm, Dave, this is interesting, but I have a couple questions:

each sample @ 3ml was dissolved in 5ml of regent, both samples allowed to separate naturally through gravity. The top opaque layer was removed using a disposable pipette, samples then transfered to cuvettes (1.5ml capacity each.)

Can you post a photo of what this separation process looks like? Why does it separate -- due to the reagent?

What is the NaOH reagent for, and where do you obtain it? I'm sorry, your posts are just too fast, and I can't keep up, so I wonder if you could try to link to a page explaining some of this, so that people can keep up with your work.

Hey Jeff, I'll start posting more references, sorry, ok, the reason for this particular regent (solvent combination,) is to separate the aromatic hydrocarbons from the oil sample, if you don't do this, the data will be completely convoluted together and hard to discern upon processing.

The reason for the baking soda (which I use lab grade quality,) is for this type of data collection on this sample I was looking for the benzene rings and naphthol's. Which I should have pointed out out were at 422nm ( 2-naphthol) and 481nm ( naphthacene- 1-naphthol.) Also, ethanol is a non-polar substance, excellent for dissolving hydrocarbons, but ethanol has many polar properties which make it ideal in situations where you may want to dissolve and separate both polar and non-polar compounds, it can do both!

Reference's for compound identification solvents used and emission wavelength data can be found in Lange's hand book of chemistry sixteenth edition Section 3-Spectroscopy, Table 3.32

Point number 2 for diluting a sample, is to eliminate the "inner filter effect" caused by the fluoresced intensity to be re-absorbed back into the cuvette, not good. I know chemisrty pretty good, so it helps, since really that's what spectroscopy is, fingerprinting molecules.

Yes, I can post pic's of what the oil separation looks like, no problem!

Also, ethanol is a non-polar substance, excellent for dissolving hydrocarbons, but ethanol has many polar properties which make it ideal in situations where you may want to dissolve and separate both polar and non-polar compounds, it can do both!

This is such an interesting and great point. I think this could be its own entire research note -- maybe the reference you cited could fit too?

We struggled to find time to test alcohol solvents and we're very surprised to see isopropyl work. This insight of yours explains why!

Hey jeff! Bingo, yes, Isopropyl will work in some situations, although it is a polar compound, but like ethanol it does exhibit non-polar properties. I have found research on the internet from some big oil companies testing Isopropyl for analyzing their products.

Here is a set I did with the proligght on some oil samples you might find interesting, https://spectralworkbench.org/sets/3239

Thanks -- helpful. Also, though, can you elaborate a bit on this?

separate the aromatic hydrocarbons from the oil sample,

How does baking soda separate them out? Can you cite some further reading I can do to catch up on this technique? Thanks again and sorry for the deluge of questions!

No jeff, ask as many questions as you like, there should be a lot more discussion and interaction between all of us here. I'm not using NaOH3, which is sodium bi-carbonate (baking soda,) for the oil samples, I am using 0.1M of NaOH, which is sodium hydroxide ( a very strong base, hence the high Ph.) these are called "buffer" solutions.

The sodium hydroxide was mixed with the ethanol for the oil samples, the baking soda for the fluorescein.

ok, aromatic hydrocarbons, which are organic compounds, have unique chemical properties, you can either cause a reaction by heating and what is called fractional distillation (difficult and needs expensive equipment,) or by a catalyst, hence a chemical reaction. Using different chemical solvents, you can separate the single and double carbon and hydrogen bonds of the molecules, it's like a recipe, here are a couple of good links explaining a lot of this process if your interested:

https://www2.chemistry.msu.edu/faculty/reusch/virttxtjml/react3.htm https://www.boundless.com/chemistry/textbooks/boundless-chemistry-textbook/organic-chemistry-23/aromatic-hydrocarbons-165/properties-of-aromatic-compounds-635-3608/

I find this very exciting, yet it is too difficult for someone like me without the extensive knowledge in chemistry you and others here are showing! Would it be possible to help me design a workflow that would allow me to say whether there is 2 stroke oil present in a gasoline sample or not? What would happen if I just separated (centrifuged) a sample and scanned it. Would I see the different spikes or would the information, as you say, be convoluted and impossible to see?

Hey @eatabean, you could centrifuge it yes, but the easiest way would be to do an absorbance spectral analysis, where you can get a snap shot of all the absorption peaks, because gasoline will fluoresce the same as oil because of the phospher in it.

Hi, @dhaffnersr, are you saying htere are absorption peaks in the visible range (400-750nm) for 2-stroke oil and/or gasoline? I wasn't aware of that. I'm not convinced that the two will fluoresce the same; our #oil-testing-kit work (https://publiclab.org/wiki/oil-testing-kit) was based on the red-shift in the fluorescence spectrum of heavier petroleum products.

@warren , Yes jeff, oil under a UV laser pointer will fluoresce a "whitish blue" color, gasoline will fluoresce more "white" because it contains more phosphor, you can easily test this theory out on any gas or oil sample, it doesn't even have to penetrate the sample, the color will present itself on the surface as either whitish-blue or white.

Gasoline, even though it is refined, is still a petroleum product (an aromatic hydro carbon) with specific Benzene rings.

@warren , Jeff, Please re-visit my research study on the crude oil sample I did:

http://publiclab.org/notes/dhaffnersr/05-18-2016/crude-oil-analysis-sweet-crude-laramine-county-carpenter-wyoming-usa