[bw58]

Hi all,
I hope this is the right place to ask this question. I'm a process engineer so I'm not 100% familiar with this chemistry.
In a forward osmosis membrane process we recently found unexpected loss of performance due to damage or fouling (not sure yet).
We use treated wastewater (final effluent) as our feedwater for the process.
To find out what was causing the damage or fouling we sent a sample off for GC-MS analysis. Their automated library search of chemicals identified a relatively strong presence of dodecane but no other alkanes.
Has anyone heard of dodecane being present in wastewater? There is no evidence of any other alkanes. I looked at the treatment process but this site only uses silicone oil and silica as antifoam, and some basic chemicals such as sodium hypochlorite.
Treatment site shows no alkanes being discharged with final effluent.
My questions:
1. How could a chemical such as dodecane end up in wastewater with no other alkanes detected? OR:
2. What kind of chemical could look like dodecane in a GC-MS analysis?

I'm out of thoughts here... The only additional information I have is that when we boiled and condensed a sample, we got an unknown liquid dissolved in the distillate that smelled acrid and had a conductivity far higher than distilled water should have. We also did GC-MS on this sample and found only dodecane and . heptane?
Sample fouling is not possible - multiple samples, multiple GC-MS units.
Oil leak is unlikely - would show way more than alkanes than just dodecane.
What mystery chemical are we dealing with that has a high electrical conductivity and looks like an alkane?
Thanks in advance for any thoughts!

[rolnor]

It can come from hydrolysis of silicone oil maybe?

[bw58]

Thank you for the suggestion rolnor. I'm not sure if that would only show dodecane and heptane as a result though. You'd think there would be many more different chain lengths ending up in the sample?

[rolnor]

Its very strange. If it was petrol it should be a mixture. I checked silicone oil, it does not explain dodecane.

[bw58]

Agreed, it's really a mystery. The problem is we can't solve it until we know what we're looking at... Any wild ideas are welcome...

[rjb]

Something about the identity of your compound (dodecane) has been bothering me for days and I can't quite put my finger on why...

Whilst I can't give you an answer to your problem, there are a couple of points that could have relevance if only to cross them off your list.

Dodecane was (possibly still is?) used to help determine FID response in GC systems and my first thought is that somebody had either contaminated your sample or was possibly using it as some kind of internal standard? Given what you've later described, this doesn't seem too likely although but I have never heard dodecane described as a having an acrid smell.

However, I understand that dodecane is also used as a model oil (crude oil surrogate) in wastewater research (something I know virtually nothing about), so could there be something going on upstream so to speak that you are not aware of?

Just a couple of thoughts.

[bw58]

Thanks for your input rbj!
I'll check with our lab to see if they could have any contamination, you never know. If they work with these chemicals they may just have made a mistake.

I've contacted the operators and process engineers but they all said there's no way we have an oil leak without it showing up on their regular QC tests. I wasn't aware of the use as model oil/diesel so it's something to check with the lab. Thanks for this thought, I'll also double check with them which chemicals they add in their sludge treatment!
B

[MOTOBALL]

You have had one half of your GC/MS investigation done; that gave a "fingerprint" (extensive fragmentation) type spectrum , i.e., an Electron ionization mass spectrum.

Identification of the unknown as dodecane is just the best match of the spectrum Vs. the spectrum of authentic dodecane in the database.

As such, and based on your other info, it is likely NOT to be dodecane.

The second half of your GC/MS investigation is to repeat the analysis, on the same sample, using a chemical ionization (CI) source.

This will / should give you info regarding the Mol. Wt. of the unknown---also pointers as to the presence of Halogens (F, Cl, Br, I) and Sulphur , Oxygen & odd or even number of N atoms if it contains N.

The Molecular ion region will be shown as [M+H]+, and losses of H2O, HX (X = F, Cl, Br , I) etc may occur.

Finally have a C,H,N,O combustion analysis done.

Regards,
Motoball

[rolnor]

If it is a error in the GC-MS interpretation I think the operator is inkompetent, its really bad if they dont exclude technical error. Elemental analysis would be impossible, you need a pure sample for this, you can not run this on a water solution.

[bw58]

Thanks for the explanation Motoball. I agree now that it's unlikely to be dodecane. I wasn't aware of a GCMS with chemical ionization. I'll propose this to the lab.

Our GCMS analyis was done by a PhD student who doesn't have extensive experience, so I'll make sure to bring this up.

We did do a DMC / dimethyl carbonate liquid-liquid extraction to remove any non-volatiles from the mess of the wastewater sample. Will also pass on the info on the C,H,N,O combustion analysis.