Strategies to Identify and Measure Organics

SteveL

New member
Randy:

Enjoyed your Reefkeeping article on organic copounds in our reef tanks. Couldn't agree with you more that it's the big chemical unknown that all our tanks are exposed to be we're all essentially blind to what is going on.

I believe there are several strategies to investigating how and what organics are turned over in our aquaria but it's going to need some pretty heavy analytcal instrumentation horsepower. We spoke briefly at MACNA about using liquid chromatography-mass spectrometry (my own profession) but nuclear magnetic resonance (NMR) could probably also play an important role.

I spoke with Bob Minard (co designer and curator of the Penn State reef tank with Sanjay Joshi) along a similar theme. Bob is a mass spec guy also. I don't know if there's any other LC/MS or NMR guys reading these boards in the US or elsewhere but here's a bone for someone to pick up on in terms of a strategy.

Consistent with your article, we're undoubtedly facing a complex mixture of organics in our tanks. The fact that most are closed systems differentiates them from the ocean (and the fact we have no where near the biodiversity - so much so that we can hardly even use that term) means that we probably have very unnatural accumulations of organics unless our skimmers and mechanical media are stripping everying out ... which I doubt. So if we're going to set about identifying and quantifying various organics we need a means of separating the "soup". State of the art here for the polar organics we can expect in an aqueous media (i.e. our saltwater) is liquid chromatography (LC). This is a technique where a sample is separated into ideally all it's individual components such that they elute from the end of a column (think narrow tube). On the end of this column we have a detector and I'm proposing that this is a mass spectrometer (MS) for our interests and needs of identification and quantitation.

MS identifies a given compound according to its molecular weight and how the structure "fragments" in the instrument. By combining LC and MS we are able to look at the complex mixtures which we can expect in our aquaria.

If we were to establish a standard LC/MS protocol (which would include an extraction of a saltwater sample prior to the LC/MS) we could start to do some comparisons of say:

1. A simple control system vrs one with a skimmer
2. A control vrs use of carbon
3. A control vrs a system with fighting corals
etc etc

By doing a compare and contrast an LC/MS experiment could focus on those compounds that differ between say a skimmed tank versus an unskimmed tank. In the initial stages just getting an "LC/MS fingerprint" would be interesting to see if we can detect the differences when we attempt to remove organics.

Woud like to hear your thoughts, Bob Minards or any other organic analytical chemists opinions on this one. I'll also attempt to tackle LC/MS questions if this thread generates interest.

In the meantime I will be trying to do some "weekend work" over the winter to investigate these ideas.

SteveL
 
Hi Steve,

Just curious (sorry, engineer intruding in the chemistry) as to what advantages an MS detector for the HPLC would have vs. say an electrochemical, FTIR (water interference?), or other detector may have. (Again, an engineer with just about enough A-Chem knowledge to be dangerous).

Anyway, I would be interested in helping out with yours (or anybody's) weekend work, if possible, even if it's only to the extent of providing samples.
 
Steve:

I will pass this on to Bob Minard. He does not frequent the reef boards.

sanjay.
 
Skearse:

There are a number of detectors suitable for coupling to LC other than the mass spectrometer (MS). A UV detector is the most common means of obtaining a signal from a component eluting off the end of a LC column. However, a MS offers one tremendous advantage in that it provides structural information on the compounds being separated by the LC. In comparison a UV detector gives a response based on a compounds ability to absorb UV light of a particular wavelength. Unfortunately this is not sufficient to differentiate between organics while MS will. If we had a simple mixture (and we knew what was in there), chromatography may be enough to separate all the components and a UV detector would suffice but when we're trying to identify unknown components in a complex mixture structural information is key. The only other detector that can rival the MS is an LC/NMR instrument but even that has some disadvantages compared to LC/MS for this application (sensitivity, cost and expertise to run one) making it a pratical no-no.

Sanjay:

Thanks for passing on to Bob. Could you PM me his e-mail as I could chat to him about possibilities. He gave me it at MACNA but in fine fashion I lost it. Still hoping to catch up with you guys on the 24th also.

SteveL
 
With so little info available, I'd think a good place to start is a simple LC with UV detection to see if we are talking about a total smear, or discrete peaks. I just don't know what to expect.

Then if we have any discrete peaks, it would be niced to run them through the MS to see what we get.
 
Now that I think about it, refractive index detection may be better than UV for a first survey as some of the organics may have very little UV absorbance.
 
Randy:

I agree, a refractive index/light scattering detector would be a good universal detector for this application (better than UV) and could serve as a first pass.

There is three additional components of the experiment that are going to be very influential upon the resulting chromatogram (technical term for the plot of time versus detector response which we get from a HPLC instrument):

1. The sample preparation
2. The HPLC column used
3. The mobile phase we use to conduct the chromatography separation

The sample prep will need to remove as much of those inorganic salts that for these purposes we're not interested in. If we don't do this I suspect we'll be lucky to get anything useful out. I have my reasons for this but it's going to get pretty technical quickly. If anyone wants to question this I'll explain.

A simple hexane shake (equal volume of saltwater sample with the non-mixable hexane - like shaking oil and water) will allow a broad base of organics to leave the saltwater sample and go into the hexane. If we then blowdown to dryness and reconstitute in the HPLC mobile phase we should be ready to go. I need to do a bit of a literature search (or if anyone reading this knows better) to check if this is appropriate - I'm used to dealing with blood plasma and urine samples not environmental samples.

I'm proposing that we go with a reverse-phase column. All this really means is that it's a column used to separate compounds such that the water-liking (hydrophylic) compounds come off first and the water-hating (hydrophobic) compounds come off last. Typically this is going to mean that the small organic compounds will come off early and the large ones will come off later. This is a very broad generalization but will help us get a rough handle on things.

A simple mobile phase gradient elution going from 5% methanol in water to 95% methanol in water over 10 minutes is a starting point. The mobile phase is what we pump through the column to induce the separation to occur.

Just doing this experiment on saltwater from the tank compared to a sample of protein skimmer skimate may be very enlightening. You know I've got to get a mass spec on this eventually.

SteveL
 
Hi Steve,

The link below is for the EPA's testing methods for solid & hazardous waste analysis (also covers aqueous samples such as groundwater). A large number of the methods are for GC; however, methods 8321 and 8325 are for solvent extractable non-volatile organics using reverse method HPLC/TS/MS or UV detection. Methylene chloride is used as the extraction for at least one of the methods.

http://www.epa.gov/epaoswer/hazwaste/test/main.htm

HTH!
 
I'd be afraid that hexane will leave too many organics behind. Something like methylene chloride might get more of the organics. We'd still need to understand that the extraction might only get a portion of the total organics.
 
Randy Holmes-Farley said:
I'd be afraid that hexane will leave too many organics behind. Something like methylene chloride might get more of the organics. We'd still need to understand that the extraction might only get a portion of the total organics.

I had seen that the pH influences "what" is extracted when heavily polluted (far far more than normal) was extracted with IIRC cyclohexane.

My feeling is that organic pollutants difficult to remove by skimming and activated carbon are (somewhat) polar in nature and have perhaps not too high molecular weight (say less than 1000 a.m.u.)

UV spectroscopy of the water gives some information (200 - 220 nm) together with the full UV-VIS spectrum.

I'll see if I can find back spectra of NSW, "clean" tanks and polluted tanks.

All not much for characterization of the structure of the pollutants but might be of some use. :)
 
Just before turning the computer off and going to sleep I remebered something. Let me write it down here as a note before I forget it.

I had done extractions and looked at changes in UV-VIS spectra.
 
Good, looks like we're getting some more qualified organic analytical experience chipping in here.

Skearse: Thanks for the EPA link to standard methods. I'll start looking through this info. With this and Randy's suggestion it looks like we may be going the way of a methylene chloride extraction. Another which occured to me today would be ethyl acetate which I believe will also pull out more of the polar organics than hexane. Both hexane and EtOAc will take up some water (probably around 5%) which may pull over more junk than we want.

Randy: Yes we're not going to get everything but I figure the liq/liq extraction has wider coverage (and is cheaper) than going the solid-phase extraction approach. We should still keep this in mind though for the future just in case we may want to be more selective and look at particular families of compounds (am I an optimist or what !).

Habib: UV-Vis may give us some insight - same way as LC/UV. I just can't help thinking that we're going to need some structural information if we're really going to start understanding what we're dealing with. I'm making plans to have an LC/MS available for some weekend work. Doing the initial LC development off-line with a UV detector is still worthwhile though.

Keep these ideas rolling please. I'm away for this weekend but will pick up again on Monday.

SteveL
 
I've got a good book called:

Biogeochemistry of Marine Dissolved Organic Matter by Dennis A. Hansell, Craig A. Carlson (Hardcover - June 2002)

I've not looked at it with an eye for analytical methods, but I'll look through it this weekend.
 
Skearse and Randy:

The EPA methodolgy Skearse directed us to is useful. The mass spec methodology is as dated as precambrian corals but there's good info on liquid/liquid extraction protocols using methylene chloride and there's a place to start with HPLC gradient conditions. Considering the complexity of what we'll be looking at, I'm leaning towards the standard HPLC gradient used for LC/MS of carbamate pesticides in waste water samples (Method 8321A).

The standard MeCl2 extraction is on a scale appropriate for pulling out trace pesticide levels (with relatively insensitive MS detection c.f. what we have today) so I may elect to go with something more simple. I'm trying to avoid using gallons of solvent for extraction and since we'll be dealing with the major organics first there's not a whole lot of point worrying about whether we're extracting >80% of something at the ppb level. A worthwhile addition though may be to spike in an internal standard before extraction and used this as an indicator of recovery. I'm thinking if I spiked in caffiene before extraction and then compared the response of caffiene post extraction to a neat solution of caffiene (all of known concentration of course) we'll have some sort of handle on recovery .... well recovery of caffiene that is. I'm really not sure how much use this is other than making me comfortable that we can detect an organic that we know was in the sample to start with at a known concentration.

Randy: if you have any pearls of wisdom from your biogeochemistry book, I'd like to hear them.

-SteveL
 
There is a chapter titled "Chemical Composition and Reactivity" in this book that has a lot of interesting information in it. None of it gives details to actual procedures, but there are many references to the original papers that presumably do. It details what chemicals have been quantified, and how. Moslty, the quantitation are for simple things like amino acids or monomeric carbohydrates.

Unfortunately, one section starts with this sentence:

"In general, there is a paucity of data on the molecular composition of DOM in the ocean. This limitation results in large part from the lack of suitable chromatographic methods with sufficient sensitivity for the characterization of DOM in seawater"

Still, if one wants to pursue chromatographic methods, this looks like a good place to start, and I'd suggest finding a copy at a library. :)
 
Hi guys,
I just got back into the hobby after a lay-off with the birth of twins. Randy mentioned this thread to me and I thought I could help out. Where are you guys with this idea?

My backgroud is a Ph.D. in Organic synthetic chemistry, I work at a large pharmaeutical company in cardivascular research and have access to almost every type of equipment you can think of. I also, teach at Villanova (tough braek in NCAA) in the evenings, Org Chemistry. They have a strong researc group also.

I look forward to hearing form you.

john
 
JWUllrich:

Well I'm still pulling this experiment together. Can't believe winter is essentially over when I thought I'd get this moving by now. Something to do with running a coral farm, a contract bioanalytical lab and taking January off on vacation chewed through my time.

Anyway would be delighted to run through my thoughts and if we can get any additional help on the instrumental analytical chemistry I think this would all be very interesting. Are you at liberty to say which big pharma company you work for .... I could probably guess to one of three ... it begins with a G a M or a W. Am I close?

SteveL
 
SteveL,
Good Guess, Begins with a W, Wyeth.. I am located at the discovery site in Collegeville, PA. Lets chat about what we want to do. depending on the complexity I may bring my analytical staff on board.

john
 
John:

Yes I know several of your colleagues. We work with your GXP drug development people - just had your QA people here last week.

I'll get back to you soon with some more details of what I have in mind. I'll probably take it off-line and use your PM until we've got something of substance to inform other RC people. Got to run tonight but I'll PM you soon.

SteveL
 
Back
Top