Hardware and method for Sulfur analysis

[rikkert]

Hello all,

I'm trying to get an old GC started for Sulfur analysis. I don't know what hardware is recommended and what methods arevused for sulfur analysis.

The GC I'm talking about is a Chrompack CP9001. It has a split/splittles injector and a cappillary column installed (Wall coated, fused silica with a length of 10 meter and 0,53 mm bore), the detector is an FID.

My questions in a row:

- I know that for sulfur analysis a PFPD detector is reccomended, but a FID will do fine right?

- A long column is reccomended, so is it better if I install the same column with 25 meters of length and 0,32 mm bore?

- Can anyone give me an example of a method for the detection of sulfur compounds?

If you have any other tips for me regarding the sulfur analysis; please tell me

Thanks in advance...

[Peter Apps]

What do you want to find out about the samples ? - the total content of sulfur or the amount of each sulphur compound ?

What kind of samples do you hope to analyse ?

What sulphur compounds do you expect, and at what concentrations approximately ?

Have you searched the literature ?

Peter

[rb6banjo]

Sulfur-containing molecules can be problematic. They can be active (like alcohols) and adsorb to surfaces. Getting them through your system - especially if you're looking for small concentrations - can be difficult. They also generally don't smell nice so they're difficult to handle.

You can use a flame like with any other combustible analyte. The use of a S-specific detector is nice because everything else pretty much vanishes and you don't have to be so concerned about the separation. If you're looking for trace levels of them, the S-specific detector is very desirable (I prefer SCD's to FPD).

There are countless ways to attack this. I'd start with an app note from one of the column suppliers (Restek, Supelco, Agilent, etc.) that has what you're searching for and modify it from there.

Good luck.

[rikkert]

What do you want to find out about the samples ? - the total content of sulfur or the amount of each sulphur compound ?
The total content of sulfur will do (we look for a way to make our oil sulfur-free)

What kind of samples do you hope to analyse ?
As said in the answer above: we analyse oild samples (maybe even crude oil if that is possible)

What sulphur compounds do you expect, and at what concentrations approximately ?
We expect thiophene (50-80% of the sulfur compounds in oil is tiophene), also dipropylsulfide and dibutylsulfide. At concentrations of approximately 1%. But as I said, we look for a way to get the sulphur out of our oil, so it is best if we can also see the sulphur concentrations of 0,01% and lower maybe (just to make sure there is no sulphur left in the oil, or negligible low concentrations)

Have you searched the literature ?
Yes, but I wasn't really able to figure out if any of the methods I found would work for our setup.

[rikkert]

Sulfur-containing molecules can be problematic. They can be active (like alcohols) and adsorb to surfaces. Getting them through your system - especially if you're looking for small concentrations - can be difficult. They also generally don't smell nice so they're difficult to handle.

You can use a flame like with any other combustible analyte. The use of a S-specific detector is nice because everything else pretty much vanishes and you don't have to be so concerned about the separation. If you're looking for trace levels of them, the S-specific detector is very desirable (I prefer SCD's to FPD).

There are countless ways to attack this. I'd start with an app note from one of the column suppliers (Restek, Supelco, Agilent, etc.) that has what you're searching for and modify it from there.

Good luck.

Thanks, we are having contact with Agilent at the moment, a employee there is trying to find out what is best for our desired application. But it is more like a second opinion here, because Agilent tells us to buy a PFPD and some other stuff from their assortment, but I read that an FID would also do just fine (with the rest of our setup)

[Peter Apps]

There are standard methods for sulfur in fuels and oils - why not use one of them ?

Peter

[rikkert]

There are standard methods for sulfur in fuels and oils - why not use one of them ?

Peter

where do I find these?

[Peter Apps]

There are standard methods for sulfur in fuels and oils - why not use one of them ?

Peter

where do I find these?

Probably the American Society for Testing and Materials (ASTM) would be the place to start. Seek and ye shall find - Google is your friend. The top several hits are about total sulfur by X-ray fluorescence rather than by GC, so maybe you need to rethink your approach.

Peter

[rikkert]

There are standard methods for sulfur in fuels and oils - why not use one of them ?

Peter

where do I find these?

Probably the American Society for Testing and Materials (ASTM) would be the place to start. Seek and ye shall find - Google is your friend. The top several hits are about total sulfur by X-ray fluorescence rather than by GC, so maybe you need to rethink your approach.

Peter

Thanks a lot, I let you know when I find something

[Andy F]

There are standard methods for sulfur in fuels and oils - why not use one of them ?

Peter

where do I find these?

Probably the American Society for Testing and Materials (ASTM) would be the place to start. Seek and ye shall find - Google is your friend. The top several hits are about total sulfur by X-ray fluorescence rather than by GC, so maybe you need to rethink your approach.

Peter

Many, many years ago I used to do total sulphur in oils by XRF. The method was IP 336 and I think LOQ was around the 0.01 or 0.02%. Really quick and easy method once the calibration is set up, which has been mentioned is a bit smelly.

[Steve Reimer]

" I know that for sulfur analysis a PFPD detector is reccomended, but a FID will do fine right?"

" we analyse oild samples (maybe even crude oil if that is possible)"

Have you looked at a chromatogram of crude oil using an FID? How would you see 0.01% thiophene in the envelop of unresolved hydrocarbons?
You need a detector that doesn't respond to the hydrocarbon background. That is why Agilent is recommending a sulfur specific detector.

[muGC]

You say you want total sulfur:
ASTM D5623 (Sulfur Compounds in Light Petroleum Liquids by GC and Sulfur Selective Detection) clearly states in the scope (Section 1.2) that GC is not the preferred method for determination of total sulfur. The preferred method is ASTM D2622 - Sulfur in Petroleum Products by Wavelength Dispersive X-Ray Fluorescence Spectrometry. If you want to know specific concentrations of individual sulfur compounds, then you use D5623 or a modified version of it for heavier petroleum samples.

You say you want to see sulfur in oils, even crude oil, by FID:
Oils contain many more hydrocarbon compounds than sulfur compounds, which have a higher response factor on the FID than sulfur compounds. You will have difficulty separating the sulfurs from the hydrocarbons, especially on your 10m x .53mm column. If each of your sulfur compounds really is 1 wt%, then maybe you will be able to see them but I am skeptical. 0.1 wt% is an even more difficult task on the FID. You need a sulfur selective detector to ignore the matrix of hydrocarbons and only see sulfur compounds.

I'm surprised Agilent is recommending the PFPD, especially considering they sell an SCD and a standard FPD. I recommend the SCD of these 3 technologies, I have worked with all 3 on petroleum samples.

[CE Instruments]

For total Sulphur GC is problematic as you cannot be certain you have found every peak and the detectors are not always linear to Sulphur content so you need additional calculations. The ASTM methods are either XRF or Combustion
A list of all methods below
D 129 Bomb combustion and precipitation For products with >0.1m% sulfur
D 1266 Lamp combustion and wet chemistry Labor intensive; not widely used
D 1552 High temperature combustion; iodate titration or IR detection For non-volatile products
D 2622 Wavelength dispersive X-ray fluorescence Expensive instrument but widely used
Sample – standard matrix must match
D 2784 Oxyhydrogen burner or lamp combustion and wet chemistry For LPG only
D 3120 Oxidative pyrolysis and microcoulometry Poor precision; not widely used
D 3246 Oxidative pyrolysis and microcoulometry For LPG only
D 4045 Hydrogenolysis and rateometric colorimetry Not widely used
D 4294 Energy dispersive X-ray fluorescence Precision poor at low sulfur levels but widely used
D 4927 Wavelength dispersive X-ray fluorescence Used for additives and lube oils
D 4951 Inductively coupled plasma atomic emission spectrometry Used for additives and lube oils
D 5185 Inductively coupled plasma atomic emission spectrometry Widely used for lube oils, and also additives
D 5453 High temperature combustion UV-fluorescence Good precision; widely used
D 6334 Wavelength dispersive X-ray fluorescence Sample – standard matrix must match
D 6428 Oxidative combustion electrochemistry Not widely used
D 6443 Wavelength dispersive X-ray fluorescence For lube oils and additives
D 6445 Energy dispersive X-ray fluorescence Similar to D 4294
D 6667 High temperature combustion UV-fluorescence For LPG only
D 6920 Oxidative combustion electrochemistry Similar to D 6428

List taken from an ASTM publication 2004

Don't waste your money buying a non approved method, you should be able to get a second hand Combustion TS system (D5453) for a similar price. If in the UK visit my home web site

[Peter Apps]

Plenty of wheels there that don't need to be reinvented !!

Peter

[oiltest]

Producer and exporter of Trace Sulfur Analyzer in China.
High quality and Competitive price.
For more information, please contact us through www.dgaequipment.com.