Does the Amount of Wool in Prepacked Liners Matter?-Part II: Results

In Part I of this series, I posed the question of the impact of the amount of wool on liner performance.  As a reminder, this study specifically examines Topaz liners that are prepacked with wool and then deactivated. With your own hand-packed liners these results will not apply.

The first criteria I wanted to examine was inertness, as it’s generally the largest concern when using wool.  Since wool has a high surface area, it can be difficult to thoroughly deactivate, leading to the theory that the more wool, the more activity.  While this is likely true if you’re packing your own liners, the results for the Topaz liners in this study, where the wool was deactivated in place, did not show this trend.

Average response factors for active acids and bases injected splitless at 0.5 ng on different liners can be found in Figure 1:

Figure 1: Average relative response factors for several active acids and bases for each group of liners, organized by weight-length of wool plug. Non-reactive compounds are included for comparison. Error bars represent 1 standard deviation.

As you can see from the above results, differences in relative responses for most of these active compounds were small, even with liners that were overpacked (10 mg and 15 mg of wool).  On the other hand, for 2,4-dinitrophenol, an often-troublesome acidic compound, the liners that were packed with 15 mg of wool showed a significant decrease in response.  Interestingly, for two of the difficult to analyze bases, dicyclohexylamine and benzidine, response factors slightly increased as wool amount increased.  Perhaps the wool has a slightly basic character.

The above data is presented using a response factor, n-tridecane as an internal standard.  This allows us to examine differences caused by activity by normalizing to the internal standard.  But what if we remove this normalized data and see the general effect on vaporization? Figure 2, below, illustrates raw peak area for the acids and bases for the different liner groups.

Figure 2: Average peak area counts for several active acids and bases for each group of liners, organized by weight-length of wool plug. Non-reactive compounds are included for comparison. Error bars represent 1 standard deviation.

Notice that the overall trend is that peak areas tend to increase as wool amount increases, though the differences are generally pretty small.  This could be due to the enhanced vaporization from the higher surface area of the wool.  2,4-dinitrophenol is an exception, as you can see the large loss in area when overpacked with 15 mg of wool.

Besides response factors and raw area response, we can also examine tailing factors of active compounds.  Figure 3 shows a comparison of tailing factors for select active compounds.  There were not any significant differences, even with the liners that are very overpacked with wool.  I found this interesting for 2,4-DNP, since I did observe a reduction in response on the 15 mg packed liners, yet the tailing did not increase.  This tells me that 2,4-DNP is likely experiencing irreversible adsorption within the wool, rather than reversible, which might be exhibited by increased tailing.

Figure 3: Tailing factors for select active compounds for each group of liners, organized by weight-length of wool plug.

In addition to inertness, I wanted to examine retention times of volatile compounds, which could be affected by flow differences or increased interactions with the wool.

The most volatile compound in this study was cis-1,2-dichloroethene.  As everything else in the system was left unchanged, any differences in retention should be due to the differences in wool packing.  The average observed retention times for each group of liners is shown in Figure 4 below.

Figure 4: Retention times of cis-1,2-dichloroethene for each group of liners, organized by weight-length of wool plug.

The liners with more wool had more retention of cis-1,2-dichloroethene.  For the liners packed with 15 mg of wool, this volatile compound eluted around 4 seconds later than the liners that were packed within specifications.  While differences are seen when packed far beyond the specifications, liners at the upper and lower specs for wool did not show significant variation in retention time.

If we look at the most volatile compound from the active acids and bases mix, 4-picoline, we can see a similar trend (Figure 5).  The liners with 15 mg of wool showed a 6 second increase in retention for 4-picoline.

Figure 5: Retention times of 4-picoline for each group of liners, organized by weight-length of wool plug.

One final observation was with regards to liner “bleed”.  Generally, upon installing a new liner, you may notice some peaks that are related to excess deactivant “bleeding” off of the liner, especially during the first injection or two.  In general, I have observed very little bleed coming off of Topaz liners.  I did observe, however, that the liners that were packed at 2 or 3 times the weight specification for wool showed increased bleed on an initial solvent blank (Figure 6).

Figure 6: Liner bleed comparison between 3 mg, 10 mg, and 15 mg wool plug on first injection after installation. The boxes highlight differences, with the 3 mg liners showing no significant bleed peaks.


The above data demonstrates that for Topaz liners packed within the specifications, there are no major differences in performance based on the amount or density of the wool.  When packing at 2-3 times the maximum weight specification, there were some differences observed: 2,4-DNP response was negatively affected, retention time shifts occurred, and there was increased liner bleed.  I believe the lack of differences observed in many cases is owed to the deactivation process itself, with the wool being deactivated in place using vapor deposition.

I don’t want to sound like a broken record, but I’ll again point out that results may look completely different if you are packing your own liners.  This is due to breaking fibers and exposing active sites during the process.  If anything, this data should help to further convince you to try prepacked liners with the wool deactivated in place, since in addition to superior inertness, you will get better reproducibility.

To see more information about the experimental set-up, visit Part I:

Does the Amount of Wool in Prepacked Liners Matter?-Part I: Experimental Setup

Tags: ,

8 Responses to “Does the Amount of Wool in Prepacked Liners Matter?-Part II: Results”

  1. Dear Lynx,

    Very nice work! Have you considered testing different types of septa as well as different inlet septum purge settings?

    With kind regards,
    Lars Kürstein, Copenhagen

  2. Hi Lars,

    Thank you for reading! I like your suggestion on looking at septa bleed based on septa type and different flow rates. I haven’t seen anything out there showing this kind of data. I will definitely consider this as a future study.


  3. Dear Linx,

    I’ve been thinking about using a 1/8 inch. PTFE tubing instead of 1/8 inch. copper tubing to connect the split vent socket and the split vent filter on an Agilent S/SL injection port. PTFE tubing are transparent. I think I will thus be able to quickly spot particles and deposits visually through the tubing close to the split vent socket.

    Would you think a PTFE tubing could be used? Or will hydrogen or helium diffuse through the PTFE material, disrupting proper split vent gas flows and pressures in the S/SL injection port? I don`t think high temperatures at the S/SL port would be able to degrade the PTFE tubing close to the split vent socket.

    This could also be a future S/SL study? :-)

    Best regards –

  4. Hi Lars,

    Thank you for reading and commenting! Having a way to visually see the dirt build up on a split vent line (like you could with a liner) would be awesome. This is an interesting idea. I am concerned, however, that diffusivity of small carrier gas molecules such as helium and hydrogen could be an issue. Perhaps this would be less of an issue with nitrogen.

    In addition, I think the high temperatures at the interface of the connection to the inlet could lead to off-gassing from the PTFE tubing, which could potentially interfere with the analysis. I would also be concerned that the high temperatures actually could degrade the tubing over time, affecting the seal. This could become a safety issue if you are using hydrogen as a carrier gas.

    Kind regards,

  5. Yuki says:

    Hi Linx,

    Thanks for the nice blog.

    I have a few questions.
    1. Do you think the results would be similar between silica wool and glass wool?

    2.In Fig. 2., average peak areas increased with the increasing amount of wool, but do you think this increase will be more pronounced if injection solvent is either water or methanol(i.e. high evaporation energy).

    3. Fig. 6. “liner bleed” is acquired with FID, but any MS data to find out what those bleeds are??

  6. Hi Yuki,
    Thank you for reading and commenting. To address your questions:
    1. I think the results could be different for glass wool, since it typically has significantly more active sites compared to fused silica (quartz) wool. As I mentioned, I also think hand packing wool that has already been deactivated will also lead to increasingly poor performance as wool amount increases, due to breaking of fibers, exposing active sites.
    2. Honestly, I am not sure about this. I suppose it is possible. This may be an interesting experiment to try in the future. Thanks for the idea!
    3. I did not check by MS, but the bleed is most likely various siloxanes from the deactivant.

  7. James Ball says:

    Hello Linx,

    Great study.

    As for the quartz vs glass wool one thing to consider is that the glass wool is more flexible and usually doesn’t break as easily at the quartz wool does so there may be an advantage to that.

    When I was using packed liners I would pack then silanize them myself for just the reason you mention if I was using them for a full semivolatile list with active analytes. For something like DRO it really didn’t matter.

    Has anyone done a comparison between wool packed liners and the cyclo liners for efficiency and inertness?

    Best Regards


  8. Hi James,

    Apologies for the late reply. You are correct about quartz being a lot more fragile than borosilicate wool. This is why you’ll generally find that when manufacturers sell deactivated wool for the customers to pack their own liners, it is most commonly glass wool, rather than quartz.

    I have actually not done a direct comparison between wool packed and cyclo liners for inertness. That is a good idea for a study in the future, as we tend to assume some benefits in inertness without having wool, so it would be nice to have the direct comparison data. I have done studies comparing molecular weight discrimination and reproducibility between these liners and found that for splitless injections, the cyclo liners perform very similar to single taper liners with wool. For split injections, I observed poor performance for the cyclo liners. Fortunately, the use of wool is generally more forgiving for split injections, due to the decreased residence time in the liner.

    These studies can be found here if you’re interested:

    Thanks for reading and commenting!

Leave a Reply

Restek Domestic Customer Service



Your Full Name

Your Email

Company Name


Spam Block (Please leave this blank)

all fields required

Thank you

Your message has been sent. We will be in touch shortly.

Message not sent

Sorry, your message could not be sent at this time. Please try again later, or contact Restek or your local Restek representative via phone.