[ssfrstlstnm]

The usual way to do what you want is to use a GC. The analyst will extract the soil with some solvent, and run that sample on the GC. He then compares the GC response for the sample that was extracted from the soil to the response from a standard (known concentration) sample to estimate the concentration in the original soil sample. For gasoline hydrocarbons, this is never an exact number because the "standard" that the analyst has will not be exactly the same as the gasoline contamination in the soil (different origin, partially degraded by soil bacteria, partially evaporated from soil, etc.)

The least expensive way to do it is with your nose (a very sensitive detector indeed). Smell the soil.

If the contamination is high. You may be able to send a sample of the contaminated soil along with some clean soil from a nearby location to a soil testing lab for organic carbon analysis. This test is performed frequently for agricultural soils and is not too expensive ( surely less than $50). I'm not sure how the lab will feel about the gasoline contamination as they usually only do agricultural soils. You could also try to do this on your own if you have a really hot oven. The soil is first heated to 110C for 24 hours to remove all of the water, then weighed. Then it is heated to a very high temperature to burn off all of the carbon and weighed again. This will give you percent organic carbon. If the gasoline contamination was large, you should see a difference between the clean and contaminated soils.

[Borek]

The soil is first heated to 110C for 24 hours to remove all of the water

...and you have just lost all volatile HCs from gasoline. Bad idea.

[Joeay]

Perhaps if you will ask for "total hydrocarbons" it will be much easier - it will be enough to add all HC concentrations to get some number of more or less well defined meaning.

That would be even better actually. I am assuming you were talking about the heating the soil method, ssince you were speaking of different volatilities.

Idea 1:
I suppose this could be done with a heating element, and a top sealed box. Since the vapours are lighter than the air, they would go up, being trapped in the top sealed box. Any extra oxygen would be forced down, where there would be an opening. Now, how do I measure the total hydrocarbons which were evapourated?

If the amount of oxygen would need to be known, I would be able to have a pilot light, heating a thermocouple in which would result in creating a very little voltage. This could control a solenoid, which would either let oxygen in, or close.

Idea 2:
This one is much more basic. A heater element, the soil, and a pressure sensor would be placed in a completely sealed box. The pressure of the trapped air would be taken before the soil is heated, then compared after. This would have to be done a few times and tested at a real lab to create our own scale. (ex. 10 psi of air, 15 psi after heated, 15-10 = 5psi difference, send to lab, find out that 5 psi is equal to so much.) I suppose a pressure release valve could also be included for safety reasons, though I do not imagine that much extra pressure would be created.

Would these ideas work? Or am I just dreaming?

Thanks for the replies so far guys. I really appreciate it

[Borek]

Would these ideas work? Or am I just dreaming?

I am afraid they won't.

Don't expect vapours separation in small box, I doubt in separation even if the box will be two meters high.

Second method sounds more reasonable, but the pressure will build up not only because of gasoline, but becasue of water too. Thus you will measure some combination of gasoline, water and probably even humus decomposition. Hardly precise.

[Joeay]

Well, you said it would be a lot simpler if I would detect all hydrocarbons. Which method were you speaking of when you said this? Or is there a way to improve on my second idea?

Anything to do with he electrical would not be a problem for me as my carreer is being an electrician, perhaps eventually leading to electrical engineer.

I'm sorry if I'm being a bother, I've been searching and reading on the net for some time now, and this seems to be the only place I could get answers for exacly what I need.

[Borek]

Well, you said it would be a lot simpler if I would detect all hydrocarbons. Which method were you speaking of when you said this?

Perhaps there are methods of detecting sum of HC that I am not aware off. Simplest way seems to be GC or GCMS plus addition of individual concentrations/amounts.

[mike]

What do you want to determine exactly? Are you looking to analyse the soil where there is an expected gasoline contamination? or are you looking to find an oil field?

[Joeay]

What do you want to determine exactly? Are you looking to analyse the soil where there is an expected gasoline contamination? or are you looking to find an oil field?

I want to analyse the soil where there is a gasoline contamination, however hydrocarbons in general would probably be even better.

[mike]

"Detection of Hydrocarbon fuel spills using distributed fibre optic sensor"

MacLean et al. Sensors and Actuators A 109 (2003) 60-67

[Joeay]

WOW! That's impressive! Would of never thought of using an OTDR for this! I'm VERY impressed! Only downside is the price, however it is possible to take a long length of tests instead of individually! I'm impressed if it works as easily as it explains in the article.

I think everyone should google that article. Definately a good read!

I'm still for a look out for a cheapter method. If anyone has information please post it.

[ssfrstlstnm]

...and you have just lost all volatile HCs from gasoline. Bad idea.

Yeah, didn't think about that. We were doing oil and diesel fuel contamination where volatilization wasn't so much of a problem.

Maybe you could extract the soil with something like pentane. Then let pentane evaporate and weigh the residue. If you compare the residue to that of a clean soil, you should get some idea of contamination. Pentane is a weak solvent and should extract much from the soil except for the gasoline hydrocarbons. Of course evaporation of the gasoline is still something of an issue, but maybe you could evaporate at low temp and catch it before too much gasoline evaporates. Still not as quantitative as GC, but it's cheap!

[ssfrstlstnm]

You could also artificially contaminate some soil and try the pentane extraction to see how much gas you recover in the residue. Probably not the most environmentally friendly way, but like I said it's cheap

[Borek]

"Detection of Hydrocarbon fuel spills using distributed fibre optic sensor"

MacLean et al. Sensors and Actuators A 109 (2003) 60-67

This one?

http://cmp.eee.strath.ac.uk/detection%20hydrocarbon%20fuel%20spills.pdf

[Borek]

Pentane is a weak solvent and should extract much from the soil except for the gasoline hydrocarbons.

Quite contrary - I would rather expect pentane to extract most of the gasoline.

[ssfrstlstnm]

I would rather expect pentane to extract most of the gasoline

That was my point. Pentane will only extract the gasoline from the soil and not much else.