Quick & Painless Email Subscription

Use the “Subscribe” link at right. It just takes a few seconds!

How Dirty Are You? Part 4…Manual Syringe Rinsing…The Answer

So how many times do you need to rinse a syringe?

Well, I can tell you the answer for the experiment we did. Let me first refresh your memory on specifically what we did.

Sequence of manual syringe rinsing:

1. 20 µL draw of a 500 ppm hydrocarbon standard and expel it
2. 20 µL draw of acetonitrile – eject into sample vial – test via GC-FID
3. Repeat step 2 many times

Each “rinse” was tested via GC-FID.

How many 20 µL acetonitrile rinses were needed to show no signal (FID)?

The answer is 4.

 

I want to thank everyone that commented when I posed the question. Most people were right on track with about 4 rinses. It is also funny that people tended to do “extra” rinsing regardless of what they thought was required. I still do this with about 6 to 10 rinses depending on the situation.

I wanted to share some helpful hints from people who commented:

  • - Use more than one rinse solvent especially in situations were analytes have a wide range of polarity
  • - Use more than one rinse vial
  • - Use the same solvent for rinsing as is used for the sample and standard solutions

Here is a chromatogram of a 50 ppm hydrocarbon standard followed by chromatograms of consecutive rinses. The 50 ppm chromatogram can be used as an intensity reference because the y-axis is the same for all chromatograms.

50ppm

rinse1v2rinse2v2rinse3v3rinsetable

Peak Capacity in Capillary GC

Peak Capacity in capillary gas chromatography is different than sample loading capacity, which is something I’ve posted on multiple times recently in ChromaBLOGraphy.  Peak capacity is simply the number of theoretical peaks that can “fit” inside a chromatogram under some definition of how much they should be separated (e.g. baseline resolved or some other criterion).  While peak capacity is something we like to maximize in GC, it always comes at the cost of speed of analysis.

I did some experiments to define peak capacities on a 30m x 0.25mm x 0.25µm Rxi-5ms using hydrogen carrier gas under efficiency-optimized flow (EOF), speed-optimized flow (SOF), and optimal heating rate conditions, and then filled in spaces between and outside-of those starting points.  I analyzed SV Calibration Mix #5, a polycyclic aromatic hydrocarbon (PAH) standard, with split injection – GC-FID.  This standard includes 16 PAHs across a wide volatility/elution range from naphthalene to benzo[ghi]perylene.  Split injection via a Precision split liner with wool minimizes injection band widths, which is critical to estimating peak capacity based on the column conditions.

As you can see from the graph below, peak capacity plateaus in the range of 1.3 to 2.5 mL/min hydrogen carrier gas under the imposed criterion of using an optimal heating rate (OHR) of 10°C divided by the holdup time in min.   Importantly, using SOF with an OHR drops the analysis time substantially without a huge loss in peak capacity.

The Performance Measurements table shows that resolution between PAH isomers benzo[b]fluoranthene and benzo[k]fluoranthene holds up well for EOF and SOF conditions.  Not surprisingly, as faster column flow and heating conditions are used, signal-to-noise is better for analyzed components.

In summary, if you are looking to maximize the number of peaks you can put in a chromatogram, while simultaneously paying attention to speed of analysis and detectability, give EOF and SOF and OHR a try.  These are great concepts as method development starting points.  Use the EZGC™ Method Translator and Flow Calculator to help with holdup time and other considerations.

Finally, don’t forget stationary phase selectivity, which is the most important parameter for separating specific components.  But we’ll get back to that in a later post…

Theory of Fast Capillary Gas Chromatography – Part 3: Column Performance vs. Gas Flow Rate
Leonid M. Blumberg
Journal of High Resolution Chromatography  – 1999, 22, (7) 403-413

Optimal Heating Rate in Gas Chromatography
L.M. Blumberg and M.S. Klee
Journal of Microcolumn Separations – 2000, 12 (9), 508-514

Plate Height Formula Widely Accepted in GC is Not Correct
Leonid M. Blumberg
Journal of Chromatography A – 2011, 1218, 8722-8723

Temperature-Programmed Gas Chromatography
Leonid M. Blumberg
Wiley-VCH Verlag & Co. – 2010

Peak Capacity Graph 2

Peak Capacity Table

Alternate GC Carrier Gas: Helium to Nitrogen, 30m x 0.25mm x 0.25µm Column to 20m x 0.15mm x 0.15µm Column

I sent my earlier post, “Switching from Helium to Nitrogen Carrier Gas for GC by Switching from a 30m x 0.25mm x 0.25µm Column to a 20m x 0.15mm x 0.15µm Column” to Roy Lautamo at Restek West for review, as he has a critical eye for particulars.  He wanted to see finer detail in the elution area starting around 1400 sec in the first figure and took a sideways swipe at me by calling what I showed “chromahistograms”.  Indeed!  He also asked about peak widths, so I post some here for comparison, and I even included elution temperatures.  The zoomed in chromatograms in that 1400+ sec region look very similar to me.

Compare He N2 1

1400 Compare

Another cup of PAH tea please!

It seemed not too long ago that Julie Kowalski and I were analyzing tea, a lot of tea, and only tea.  To this day we cringe at the thought of tea, but hopefully our pain can be your gain. We analyzed pesticides, natural products, and polycyclic aromatic hydrocarbons (PAHs) in tea.   Julie and Amanda Rigdon first started analyzing PAHs in Yerba Mate tea and found quite high levels of PAHs due to the roasting process.  This spurred more research into other teas that also have a roasting pretreatment.  Longjing tea, also known as Dragon Well tea, has a very specific hand roasting process that involves pan frying the leaves just after harvest.

Amanda and Julie put in a lot of work developing the sample preparation method for analyzing PAHs in tea.  The procedure is below in Figures 1 and 2.  After the QuEChERS extraction and silica SPE cleanup the extracts were analyzed by GCxGC-TOFMS.  We used a 60 m x 0.25 mm x 0.10 µm Rxi-PAH column in the first dimension and a 1 m x 0.25 mm x 0.10 µm Rxi-1HT in the second dimension.  Since many PAHs are isobaric, chromatographic separation is necessary.  With a GCxGC setup, it is important to preserve the first dimension separation by matching the modulation period for 3-4 slices across the peak.  We chose a 1.5 sec modulation period in order to maintain the separation of closely eluting chrysene and triphenylene and also the benzo fluranthenes (Figures 3 and 4).

We did find ppb levels of PAHs in the Dragon Well tea that we analyzed, including those PAHs in the European Food Safety Authority (EFSA) priority 4 list (Table I). Surprisingly, I still enjoy drinking tea, even though I really don’t want to analyze that very complex matrix again!

Figure 1: A modified QuEChERS extraction was used for the analysis of PAHs in tea. We also used this extraction method for PAHs in mussels.

Figure 1: A modified QuEChERS extraction was used for the analysis of PAHs in tea. We also used this extraction method for PAHs in mussels.

Figure 2:  The silica SPE cleanup procedure removed the large amount of chlorophyll in the Dragonwell tea extract.  Chlorophyll is nonvolatile and can quickly degrade the GC inlet performance.

Figure 2: The silica SPE cleanup procedure removed the large amount of chlorophyll in the Dragon Well tea extract. Chlorophyll is nonvolatile and can quickly degrade the GC inlet performance.

Figure 3: Coupling the Rxi-PAH with a fast modulation time maintains separation of isobaric PAHs in the first dimension.

Figure 3: Coupling the Rxi-PAH with a fast modulation time maintains separation of isobaric PAHs in the first dimension.

Figure 4:  Separation of closely eluting benzo fluoranthenes are baseline separated on the Rxi-PAH column.

Figure 4: Benzo[b]fluoranthene, benzo[k]fluoranthene and benzo[j]fluoranthene are all baseline separated in the first dimension using the Rxi-PAH column.  This separation is maintained in the GCxGC system using a fast (1.5 sec) modulation period. 

Table I: Incurred PAHs found in the Dragonwell Tea.  The red dot signifies the EFSA PAH4.

Table I: Incurred PAHs found in the Dragon Well tea. The red dot signifies the EFSA PAH4.

Switching from Helium to Nitrogen Carrier Gas for GC by Switching from a 30m x 0.25mm x 0.25µm Column to a 20m x 0.15mm x 0.15µm Column

Our own Jaap de Zeeuw is a master chromatographer and he sometimes does his best work while sleeping.  Recently he was sharing with me an idea that woke him up in Singapore at 3 AM about switching from helium carrier gas for GC, which is still under threat from a supply shortage, to nitrogen, an abundant gas.   This “dream” Jaap had was to use nitrogen and get the same efficient separations as helium delivers while using exactly the same oven program rate to get the same analysis time.  I’d call that a fantasy, until Jaap demonstrated with the EZGC Method Translator how it could be done.  This got me excited enough that I gave up my weekend to do the necessary lab experiments using GC-FID on fragrances, including the one in the example below.

Essentially, operating a 30m x 0.25mm x 0.25µm Stabilwax at helium efficiency-optimized flow and optimal heating rate and a 20m x 0.15mm x 0.15µm Stabilwax at just below nitrogen speed-optimized flow (matching the holdup times with the Method Translator) allows you to use the same oven program rate and deliver the same looking chromatograms in very close to the same time!

Nothing comes for free, so sample loading capacity is about 5x less for the 0.15mm versus the 0.25mm column, but the concept works and I will continue to explore its implementation in my lab.

Keep on dreamin’, Jaap!

30 He to 20 N2 MTFC


Compare He N2 1

 


Compare He N2 2

 

How Dirty Are You? Part 4…Manual Syringe Rinsing…The Question

reanimator

I don’t even want to guess how many times you would need to rinse this syringe before removing all traces of reanimation “reagent”

This How Dirty Are You? blog should really be credited with inspiring the entire project. I was discussing the number of times I rinse my syringes with a colleague, Michelle Misselwitz, a few years ago. We were working in the same hood and I noticed that she rinsed her syringe a lot…I do too. I was even reminded of my rather obsessive rinsing at Thanksgiving this year. I was washing a few dirty dishes by hand and my mother mentioned that I was rinsing them a lot…and that my grandmother would have been upset because I was wasting so much water…but “What about the soap residue?” I said.

Over the years, I have been told different things about the number of times a syringe should be rinsed to prevent carryover/contamination…and I have been asked the same question many times. I was told early in my lab career that 7 was the magic number…so just to be sure, I got in the habit of using 10 rinses.

What We Did

There are a lot of variables to consider…sample matrix, compound concentration, sample solvent, rinse solvent(s), rinse volume etc. But to keep things simple, we used a hydrocarbon standard at a high level and rinsed with the same solvent as the sample solution, acetonitrile. We also defined the volume of a “rinse” as 20 µL. We started the experiment with a 20 µL draw of a 500 ppm hydrocarbon standard. The hydrocarbon standard was expelled and then consecutive rinses were performed. Each rinse was tested by GC-FID.

Sequence of manual syringe rinsing:

1. 20 µL draw of a 500 ppm hydrocarbon standard and expel it
2. 20 µL draw of acetonitrile – eject into sample vial – test via GC-FID
3. Repeat step 2 many times

Question

Post or Email me with your answers and look for the data in my next blog.

1. How many 20 µL acetonitrile rinses were needed to show no signal (FID)?

 (Hint: The answer is between 1 and 15…I didn’t say it was a good hint.)

 

 

 (photo credit: http://philosophicalasides.blogspot.com)

Does a different oven temperature programming rate change overload for PAHs on a GC column?

Jaap proposed that it might make a difference, so I wanted to test that possibility and add it to the following series on sample loading capacity, or “loadability” as Jaap calls it, for gas chromatography.

Sample loading capacity for PAHs on a 30m x 0.25mm x 0.25µm Rxi-5ms GC column

Does a thicker film GC column increase sample loading capacity for PAHs?  30m x 0.25mm x 0.50µm Rxi-5ms

Sample loading capacity for PAHs, part 3:  30m x 0.32mm x 0.25µm Rxi-5ms

Finally, some increased sample loading capacity for PAHs! Part 4: 30m x 0.32mm x 0.50µm Rxi-5ms

As previously, I used an Rxi-5ms GC column with efficiency-optimized hydrogen carrier gas flow and optimal heating rate (in this case, 10.1°C/min) and test conditions employing 0.5x and 1.5x that heating rate, 5.1 and 15.2°C/min, respectively.

The resulting chromatograms are shown below and to my eye there are no obvious practical differences in peak shapes that would suggest overload has been changed by any of the GC oven heating rates.  The chromatograms look different as regards separation though because the PAHs are eluting at different temperatures.  I’ve listed the elution temperatures for the last PAH on the chromatograms.

As always, the EZGC Method Translator and Flow Calculator was used to guide this work.

Overload Temp

 

Lab Hack: Quickly Reducing GC Inlet Pressure on an Agilent GC Prior to Changing the Septum and/or Inlet Liner

Have you ever been changing your Agilent GC inlet liner (you do change that routinely, right?), particularly a single taper with wool type, and pulled it out to see the wool at the top of the liner (Figure 1) when it should be at the bottom of the liner (Figure 2), the proper placement where injected sample can hit it and be transferred to the GC column?  How does this occur?  Sudden depressurization of the GC inlet can cause the wool to pop to the top of the liner.  It probably doesn’t matter if this happens when you’re changing the liner, since you’re going to throw the old liner away, but it can also happen when you’re changing the GC inlet septum if you haven’t properly depressurized the inlet, and then unknowingly you’re left with the liner wool at the top of the inlet, where it’s not effective for its purposes of keeping the sample from hitting the bottom seal (where compounds like DDT can degrade), heat transfer, and preventing nonvolatile “dirt” from collecting on the head of the more expensive GC column.

In the following sets of figures, I’ll show you how I depressurize my Agilent 6890 GC inlet to prevent the inlet liner wool from jumping up.  As always, it is much better to completely cool all heated zones before inlet maintenance procedures.

Go to the Agilent GC Keyboard, and press “Col 1” (Figure 3), depending on the Column and Inlet in use.  You’ll see the Display showing similarly to Figure 4.  If necessary, use the “Up” or “Down” symbol key on the Keyboard to move the Display arrow to “Pressure” (it is on “Flow” in the Figure 4 example).   When the Display arrow is on “Pressure”, use “0.1 Enter” (or even “0 Enter”) from the Keyboard (Figure 5), which should cause the Pressure to drop down relatively quickly, but not so quickly that it causes the inlet liner wool to move (the pressure does not drop all the way to zero, typically, Figure 6).  [Note:  On newer Agilent GCs you may hear the “rattlesnake” as the pressure drops.]  Use the “Up” symbol key to move the Display back to “Pressure” and then press “Off” on the Keyboard (Figure 7).  Again, the “Pressure” may not completely zero out, but the GC inlet is now ready for a septum and/or liner change.  Why don’t we just use “Off” immediately instead of “0.1”.  When you use “Off” the pressure bleeds off very, very slowly through the GC column and if you open the inlet too soon the liner wool can move in the liner.

Don’t forget to restore pressure/flow before you heat the GC inlet and column up, and it’s always good to leak check after inlet maintenance.

Fig 1x


Fig 2x




Fig 3


Fig 4


 


Fig 5xx


Fig 6



Fig 7


Fig 8

Need help finding the correct ferrule to install your GC column? Part 1: Agilent GC ferrules

You’re ready to install a column into your GC and realize you do not have ferrules to do the job. But which ferrule do you choose?  The parameters to consider when choosing the correct ferrule are instrument make and model, nut size and type, ferrule material, and column ID size. A visit to our ferrules home page will provide a general overview of variety of ferrules offered by Restek.  By navigating the page, you can narrow the selection by choosing these parameters. This is the first in a series of posts that will hopefully make this selection process easier. We begin with Agilent systems because they have more choices available.

Capillary Column Ferrules

For this section, I will assume the GC is already configured to install capillary columns with 1/16 inch fittings. Restek offers two options on nuts and ferrules for Agilent capillary column installation. There are compact ferrules and standard ferrules, each with their own style of nut. The compact ferrule and nut are considered the Agilent-style. The broad taper on the compact ferrules was designed to match the taper on the bottom of the inlet seal. Standard ferrule and nut can also be used effectively on an Agilent system.

Here are the visual differences between the standard and compact ferrules:

Standard                                                                         Compact

standferrule                                                             Compferrules          

Capillary Ferrules                                   Compact Ferrules (Fit ≤ 0.28mm ID Fused Silica Columns)

                                                                Compact Ferrules (Fit 0.32mm ID Fused Silica Columns)

                                                                Compact Ferrules (Fit 0.45/0.53mm ID Fused Silica Columns)

 

Here are the catalog numbers of the corresponding applicable nuts:

FTnut                    capnut                 hotswapFT2nut

 

Finger-Tight Capillary Column Nuts Capillary Column Nuts Hot Swap Capillary Column Nuts Looped Finger-Tight Capillary Nuts
Brass Stainless Steel
Standard Ferrules 21041 21879 20883 22347 21312
Compact Ferrules 21040 21878 21884 22348 21311

 

This is a cross-cut look at the insides of the standard capillary nut and shows the length of the ferrule and the depth of the nut hole. This helps to show how necessary it is to match the ferrule with the proper nut; otherwise there will not be an adequate seal. If you are unsure which of these nuts you have on hand to install your columns, take note the Agilent style nut has a notch on the hex-nut portion, the standard does not.

 

comp                                     stand

 

“Compact”  Agilent-style Ferrule and nut                       Standard  ferrules and nut

 

Once you’ve matched the nut type to the correct ferrule type, now we need to match the capillary column ID to the ferrule ID. The table on sizes below gives an overview of how to match fused silica and MXT capillaries.

 20141121105433156

The next step in determining the correct ferrule is the material. The chart below explains the differences between the material choices. For installation into the inlet and most detectors, either graphite or the softer 60/40 vespel graphite should work fine. For GC/MS, it is recommended to use the standard, long ferrule vespel/graphite installed with an MSD source nut.  This material is a good fit for MS because it will not fragment or allow oxygen to permeate into the system under vacuum applications.

Agilent has suggested that 100% vespel ferrules be only used for isothermal analysis.

MSD Source Nut

20643_ph_co_msd

20141118133453588-21

A note on the Agilent TCD, it has an 1/8 inch compression fitting port. In order to use our Agilent style or standard design nuts and ferrules you would need to have the Agilent supplied adaptor fitting. Restek does not have that fitting. Alternatively, Agilent does supply a two piece ferrule to install capillary column with a 1/8 inch compression nut. Restek does not sell the two piece ferrule and we do not recommend using a one piece graphite reducing ferrule.

 

Packed Column Ferrules

Typical Agilent packed column inlets are sized to accommodate ¼ inch OD packed columns; verify the fitting size of your inlet and detector. Agilent systems generally use standard compression style nuts and ferrules. If you doing on column injections, installing a 3/16 or 1/8 inch OD column will require inlet adaptors to center the column inside the ¼ inch injection port. If your instrument is equipped with an adaptor with a glass insert, on column injection will not be necessary, therefore a nut and ferrule would be all that is required to connect the column. Again you will need to verify the size nut and ferrule required. If you are installing a packed column into a capillary column inlet and detector we suggest a pig-tail configuration.

Notes on Agilent detectors- verify the size of the fitting required to connect the column. Agilent TCD’s for example are 1/8 inch compression fittings. If you want to install a 3/16 packed column, make sure your detector is adapted to accommodate ¼ inch packed column so you can use a reducing ferrule and ¼ inch nut.

Packed Column Inlet Adaptor

inletadaptorkit_ph_co_pak

In summary, the key points to remember are:

  • Match the ferrule with the proper nut
  • Know the ID of your column to match with the correct ferrule size.
  • Know system requirements for the ferrule material.
  • For packed columns: Verify the fitting sizes of you inlets and detectors.

 

Look for future posts on nuts & ferrules for other instrument manufacturers. Thanks for reading.

I would like to thank my Colleague, Erin Herrold for her assistance and insight with this blog post.

Residual Solvents in Cannabis…and Terpenes with Simon & Garfunkel

In my last blog post, I wrote about our ongoing method development for residual solvents in cannabis. We’ve been really busy since then, and now we have a complete application note published on this subject. Because we can’t legally get our hands on real cannabis concentrates here in Pennsylvania, the application note proves our concept with some supporting data from our collaborators at CAL Green Solutions. You can check out the full application note here.

By now, a lot of people in the industry know that headspace-gas chromatography (HS-GC) is the ideal analytical technique for analysis of residual solvents in cannabis concentrates, but wouldn’t it be nice to be able to perform another pertinent cannabis analysis on our fancy headspace instrument? It turns out that we can. The beauty of HS-GC is its ability to separate volatile analytes from difficult sample matrices with very little sample preparation. While terpenes are less volatile than residual solvents, all cannabis terpenes (with the exception of phytol) are volatile enough to analyze using HS-GC. With that in mind, we developed a preliminary method for HS-GC analysis of terpenes in cannabis.

In order to keep things as simple as possible, we developed our terpenes method on the Rxi-624Sil MS column, which is the same column that is recommended for residual solvent analysis. Luckily, Jack Cochran had already established that the column does a really good job at resolving terpenes. This means that you can run two different analyses on the same instrumental setup, getting more use out of your instrument investment – hooray! Here’s what the chromatography looks like for a comprehensive set of terpenes:

 

 

Beautiful Terpene Chromatogram

The Rxi-624Sil MS nicely separates a comprehensive set of terpenes.

 

To test out the method sans cannabis, we decided to determine terpene profiles for some of the following common herbs: parsley, sage, rosemary, and thyme (now the title makes sense!):

 

Terpene profiles of common herbs using the Rxi-624Sil MS

Terpene profiles of common herbs using the Rxi-624Sil MS

 

As a disclaimer, this data should be considered semi-quantitative, and the method is preliminary, but our overall profiles matched up well with what we could find about terpene profiles of these herbs. We also profiled hops, which are a close cousin to cannabis. This work can be found in this application note which describes the analytical procedure and our results in detail.

I will be discussing this and other work at The Emerald Conference in San Francisco on January 23rd. The conference agenda includes some good speakers from the cannabis industry, and it looks like it will be a valuable conference for anyone who can attend.