Liquid vs. gas sampling valves for LPG analysis

[thohry]

Hi all,

In analyzing LPG (ASTM D 2163 for instance) I see there are two methods of sampling:
1. Vaporizing cylinder and gas sampling valve, in which an exact amount of liquid LPG is vaporized and then the gas is piped to the gas sampling valve for analyzing.
2. GC liquid sampling valve: the liquid sample is led directly to the valve and the valve switches, the sample vaporizes in loop and analyzed.

My question is: which method is more accurate and preferred ?

Thanks for any input.

[larkl]

You're usually better off testing the sample in the state that it's captured. So, if the material is sampled and delivered to the lab as a liquid, you'd use a liquid valve.

[Bruce Hamilton]

In general, the analysis should be defined in the request. It's usual for suppliers of bulk LPG to use the liquid phase, whereas a consumer may specify vapour phase analysis. A problem with performing the volatization and injecting the gas is that you have to ensure all the sample is fully volatized and representative, whereas direct injection of all the liquid avoids that problem, however other issues ( such as discrimination ) may arise.

The choice is more defined by the customer, rather than one method being "better".

[thohry]

Thanks you both for replying.
The sample is always in liquid phase (under pressure) so I think the choice of liquid sampling valve is better.

[rrtymko]

Have you investigated Transcendent Enterprises which manufactures a Heated Pressurized Liquid Sampling valve (HPLIS)?
www.transcendent .ca

[rb6banjo]

I have found that making injections where the injection loop is pressurized - even a little - leads to problems with precision. We analyze liquid CO2 for impurities and we have chosen to fill a loop with liquid, then expand that liquid into previously evacuated "expansion chamber". Then, we can feed the rotary valve located in the GC oven to make the injection for analysis. I have a small rotometer connected to the outlet of the flow path for the sample so that I can tell that there is flow from the canister containing the expanded liquid sample and I also know when the pressure drops to atmospheric so I can actuate the valve.

[muGC]

Thohry,
I work with this ASTM method frequently. Section 6.4.2 of ASTM D2163 recommends that the liquid sampling valve be the mode of injection. I have never seen anyone vaporizing a sample to perform D2163. larkl is right, you should always inject a sample in the same state that it was capture. Inject LPG as an LPG, inject gas as a gas. You can perform the vaporized method from Section 6.4.3, but as Bruce said you must ensure that you are injecting a representative sample. You have more certainty that you are injecting a representative sample when injecting as an LPG.

rrtymko,
The HPLIS by Transcendent is not appropriate for ASTM D2163. If you are injecting as an LPG, you must not be heating the sample otherwise there will be a phase transition before the sample enters the column and you will be injecting both liquid and gas but in unknown proportions. The HPLIS from Transcendent is not for LPG samples, it is for "difficult to chromatograph samples such as broad boiling range analytes in light condensates, viscous or highly polar samples" (from their website). The proposed ASTM method with the HPLIS from Transcendent is for live crude oils and condensates, not LPG.

rb6banjo,
If Thohry is following ASTM D2163, then they likely cannot operate outside the parameters of the method. The precision issue you see with liquid CO2 injections is not a problem that occurs with LPG if the method and instrument are set up properly. Pressurized liquid injections work fine.

[rrtymko]

HPLIS is not required for LPG analysis. PLIS alone is more than adequate to vapourize the sample providing the concentration of C1 is below 1% for single phase injection.
With PLIS there is no pulse heating applied to the sample as there is adequate thermal energy derived from the Injection port to gasify the sample.
PLIS has been successfully used in LPG analysis with superior results to Rotary valve injection as the transfer zone is extremely short (less than 80 uL) with no interruption of carrier gas flow.

[muGC]

HPLIS is not required for LPG analysis. PLIS alone is more than adequate to vapourize the sample providing the concentration of C1 is below 1% for single phase injection.
With PLIS there is no pulse heating applied to the sample as there is adequate thermal energy derived from the Injection port to gasify the sample.
PLIS has been successfully used in LPG analysis with superior results to Rotary valve injection as the transfer zone is extremely short (less than 80 uL) with no interruption of carrier gas flow.

Agreed, I use Agilent's version of the PLIS (they call it the HPLI). It works great and provides superior results to the traditional LSV configuration. It's not a carrier gas flow issue though, it is the pressure drop between the column head and the LSV. This pressure drop allows a phase transition from liquid to gas and cause band broadening. Comparisons of PLIS and traditional LSV injections show the PLIS with sharper peaks due to a nearly 100% liquid injection. However, if you purchase a GC configured for ASTM D2163 from Agilent, PAC, etc., it will just be configured with a traditional LSV and GSV.

Traditional LSV configuration and vapor sampling are still acceptable.

[dhanu07]

Hi Friends!

i'm new to this forum. here i read your forum discussions that contains many useful infromations, then i'm also share an informations about this topic.

Several methods use the Gas chromatography technique for determining the composition of liquefied petroleum gas (LPG) after a sample expansion. The LPG vaporization is often source of errors and bias. Obtain a representative vapor sample from a pressurized liquefied gas is challenging and difficult to handling. Many procedures are used in laboratories to quantitatively expand LPG samples from a liquid phase to a homogeneous gas phase prior to analysis. Most of them are off-line methods, not safe for operators and environment, user-sensitive, causing lack of reproducibility and repeatability on analysis results. SRA Instruments developed an automated and user-independent LPG Vaporizer to handle under control the critical step of LPG phase change.

[Legadulab]

Hi all,

about this LPG Vaporizer device, you can get more details from this link:

http://www.sra-instruments.com/en/index ... =myvap.htm

Regards