Chromatographic analysis of fuel oils in fish and shell fish

[Don]

Hi there, could some one please help guide me in the development of a method for
"Chromatographic analysis of fuel oils in fish and shell fish"

As iam in college and must develop a method?

[foodchem]

A very simple procedure that I found to work well is to extract fish tissue using pentane, concentrate extract using K-D evaporator, then use GC to analyze extract with FID detector.

[Don]

Propose suitable methods for the pre-treatment and quantative analysis of fuel oil contamination of fish and shell fish (ug/g).

1. Outline analytical problem and potential difficulties.
2. Propose a sampling protocol.
3. Suggested essential sample pre-treatment and cleanup
4. Select the most cost-effecient procedure that will deliver an analytical result appropriate to the sensetivity requirements.

Therefore I must justify any decisions I make.

Thanks

[tom jupille]

1. Outline analytical problem and potential difficulties.

Okay, so what have you come up with for this first stage (we can provide critique and guidance, but we're not going to do your homework for you!)?

Some possible questions:
- what detection limits will you need?
- what fish (species? big? little?)
- what part of the fish (whole thing?, edible?, offal?)

[Don]

I appreciate your help and accept taht this must be my own effort in developing methods. However due to my lack of experience inthis field I just need some help to guide me in a logical path.

In the case of
what part of the fish?
The whole thing I presume. Although what are the implications of using only some parts of the fish to any possible methods.

The species was not specified. However I think it best to consider oily and non-oily fish. That is if they can be categorised in theese groups?
I was not provided with a detection limit, only given a (ug/g) term. So any range of detectors and methods that are able to achieve this are acceptable.

Also it was suggested that I research papers on the subject. However I have found it difficult sifting through large volumes of unrelated data using www.sciencedirect.com.
Would you be able to suggest any other sites or am I meerly searching in the wrong way?


Thanks again
Don
Ireland

[tom jupille]

Here's where some telephone calls might come in handy. You might try:

The Western Regional Fisheries Board, Weir Lodge, Earls Island, Galway, Rep. of Ireland.
Tel: +353 (0)91 563110 • Fax: +353 (0)91 563111 • e-Mail:info@wrfb.ie Web: www.wrfb.ie

Be persistant!

Now, as to the reason for the question about species and organs to be analyzed: those relate to parts 2 and 3 of your project.

Some species tend to concentrate certain pollutants (I would assume that "oily" species would tend to lock up more petroleum pollution than non-oily species). That suggests limiting your project (at least in the first stage) to only one species (removes a source of variability). If you can find a good "marker" species (one that concentrates a lot of petroleum) it will make your analytical job easier.

In a similar vein, I would expect that fuel oil would distribute among different organs. You might want to select organs having the hightest level.

The general procedure for the analysis was outlined very nicely by foodchem: drop the sample in a blender with pentane, then concentrate and inject the result (yuck!).

[Don]

Thanks I recieved alot of information from that web site.

However I am still left wondering what type of sample pre-treatment and cleanup I should use?
Would this envolve the use of SPE and if so how?

Any suggestion?

Don

[tom jupille]

Go back and look at the post from foodchem. Extraction with something like pentane or hexane should be fairly selective for your target (neutral, very hydrophobic) molecules.

[Don]

Thanks for the advice so far. As my knowledge begins to expand I fear my questions also do? ups
Excuse any stupid questions?

From what I have read of journals and been advised on this site Iam thinking of going along the line of liquid extraction using something like pentane. Does this involve SPE? and if so are there any other requirements that should be taken into account?
I think it best to use HPLC (would it not be better to use GC because of its greater sensetivity or is this not a factor because MS may be used) for seperation followed by MS. However is this not a very laboryious and expensive process? which may be avoided?

How is the oil content quantised within the matrix, is this done by spiking also is it required to set up a number of standards? How would I determine the concentration range of my standards? Is it required to perform a "blank" with a fish that has no oil contamination? Would this blank be useful to account for the oils present in some fish?

Finally are GC-MS and HPLC-MS the only methods of providing the sensetivity required i.e. i micro g/g?

ta
Don

[tom jupille]

Now you're ready to start working on a PhD thesis!

Your questions are all good ones; hopefully, some more experienced environmental analysis people will chime in.

Presumably, what you want to be able to do is to monitor the level of fuel oil contamination or exposure in the fish. That means that you need to find a compound (or set of compounds) that are "markers" (essentially a fingerprint) of exposure to fuel oil. The likely candidates are the heavier alkanes or aromatic hydrocarbons. As far as I know, there is no metabolic pathway in fish that will either create them or get rid of them (note, however, that this is way outside my area of expertise!). They are quite thermally stable and reasonably volatile, which makes them an ideal candidate for GC analysis.

I've spent most of my career doing LC, but generally speaking, if you have a compound that is volatile and stable at any temperature under about 300 degrees C, GC is preferable.

SPE (solid phase extraction) is essentially a low-resolution form of LC (carried out in small cartridges) to do a preliminary separation of complex mixtures. In your case, there's a good chance that it won't be necessary; a simple liquid-solid extraction with pentane or hexane, followed by filtration and partial concentration by evaporation could do the job.

General procedure would be to start with "clean" fish

[Don]

Would GC provide the sensetivity that I require i.e. micro gram/ gram.

Also what version of standards should I use i.e. internal standards, standard addition or external addition and why?



Don

[tom jupille]

Given that you can evaporate down your extract, you should be able to get virtually any sensitivity you need/want (defined in concentration terms).

Because the proposed sample prep involves evaporation of the extract, you will definitely need an internal standard (sometimes also called a "surrogate") to compensate for variations in the concentration process.

Doing quantitation by standard additions may be required if you can't find a suitable blank (uncontaminated fish).

[Don]

In order to determine PAHs in fish I will probally need to perform many analyses on a large number of samples using an automated technique. In order to quantatite the mehtod which type of standardisation would be more useful?
Standard addition
Internal standard
or any other method? any I choose a particular method they what are the advantages and disadvantages of that chosen method as well as the not chosen method.
Basically Iam tryiny to decide what method of standardisation to choose and to justify it?

By the way thanks so far for the help!!!!!!!!!

[tom jupille]

Actually, I'm tempted to move this thread offf the "Student Projects" board and onto Gas Chromatography; you're now getting to a quite professional level of detail and questions.

But, to answer the actual question, internal standard and standard addition are not mutually exclusive.

In "external standardization" (which is what you are doing unless otherwise specified", you construct a calibration plot relating mass (or concentration) of analyte to the area (or height) of the peaks. This is done by running calibrators at known concentrations. The catch is that the samples have to be run under exactly the same conditions as the calibrators. "Exactly" includes sample workup, injection volume, etc., in addition to the chromatography conditions per se.

----------------------------
In your situation, the potential exists for significant error in the sample workup (assuming you will be evaporating your extract in order to concentrate it).

Internal standardization adds an additional compound (called, appropriately enough, the "internal standard") at a known level to the sample *before* you run the error-prone step. When you do the chromatography, you measure the area or height of both the internal standard and the analyte peaks. The calibration plot then relates peak area *ratio* to mass *ratio*. Computationally, all the rest is the same as in external standardization. The advantage is that any errors in sample workup or injection volume will presumably affect both the internal standard and the analyte peak in the same proportion, and so will cancel when the ratios are calculated.

If the dominant source of error in your analysis is sample workup or injection volume, then the use of internal standards will usually improve precision. On the other hand, if the dominant source of error is integration or "signal to noise ratio", the use of internal standards will make the precision worse.

---------------------------------------------
Standard additions come into play when you can't find a clean matrix blank. In essence, the calibrators are added to the actual sample at two or more different levels. The plot of peak area vs concentration added is then extrapolated back to zero area. The X-intercept is the negative of the amount of analyte in the sample.

Note that you can use internal standards with this approach as well.

The advantage is that you don't need a clean matrix blank. The disadvantage is that it requires more runs per sample.

[HW Mueller]

This effect of lowering precision due to the use of an internal standard can easilly be circumvented by using both the external and internal standard method. All you have to do is add an internal standard, but run your samples the way you should when running the external st. method (one example: Inject a standard containing all the components including int. st., then inject the same volume of your unknown with known amount of int. st.).
Note that with manual injection it is difficult to do the external standard method with high precision.