An introduction to the benefits of using split injection when performing semivolatiles analysis by 8270D – the instrument checkout mix

This blog is part of a series; the previous installments can be found here and here.

We have mentioned several times that the reduced residence time resulting from the fast sample transfer that occurs during a split injection reduces in-inlet degradation and adsorptive loss. Proof of this can be seen in the first run of the 8270 analysis batch – the tune, inlet inertness, and column performance verification check.

Section 7.6 of 8270D indicates that a 50 ng/µL solution of decafluorotriphenylphosphine (DFTPP), 4,4’-dichlorodiphenyltrichloroethane (p,p’-DDT), pentachlorophenol and benzidine should be used. Specifically, the total amount for each compound should be 50 ng or less; the concentration can vary.

  • DFTPP is used to verify that the detector is properly tuned (more on this subject here).
  • Degradation of DDT is not to exceed 20% – by the criteria listed in EPA Method 8081. Thankfully, Linx has already covered this topic in detail here.
  • Benzidine and Pentachlorophenol:
    • are to be present at their “normal” responses. This is vaguely worded – perhaps the intent was ± 20% like the CCV evaluation.
    • should not exceed a tailing factor of 2 as calculated in Figure 1 of EPA Method 8270D (also blog Figure 1). NOTE: AB>BC is a symptom of fronting (i.e., column overload), indicating a need for less mass on column to properly perform the tailing factor check.
tailing factor

Figure 1 – 8270D Tailing Factor Calculation

Figure 2 is an example of the system suitability check mix (cat# 31615) analyzed at 50 ng/µL (in dichloromethane) by a 1 µL splitless injection. Pentachlorophenol and benzidine tailing factors are 0.75 and 0.53 respectively, indicating column overload. At 50 ng on column, overload is not a surprise, but it could be problematic as the system gets dirty. The tailing factor is supposed to be a helpful metric to indicate when maintenance should be done before client samples are analyzed. Serious overload significantly biases the tailing factor measurement and can result in a value of less than 2.0, even when the inlet is dirty and maintenance should be performed.

blog 3 graphics 1

Figure 2 – System suitability check mix (cat# 31615) analyzed at 50 ng/µL in dichloromethane by a 1 µL splitless injection.

Something else to consider: injecting enough material to overload the column not only negates the effectiveness of the tailing factor, but also the DDT degradation check and the pentachlorophenol and benzidine response check. It is likely the active sites in the inlet are also being overloaded, so you are not getting an adequate indication of inlet activity, either.

Figure 3 shows the same system suitability check mix analyzed by a 1 µL 10:1 split injection; the system configuration is the same as that used in Figure 2, except the single taper inlet liner with wool has been replaced with a split precision liner. Pentachlorophenol and benzidine now have reasonable tailing factors. The 0.94 for pentachlorophenol could indicate a slight overload, but this value of less than 1 is more likely due to the scan rate and peak apex assignment (read more here).

blog 3 graphics 2

Figure 3 – System suitability check mix (cat# 31615) analyzed at 50 ng/µL in dichloromethane by a 1 µL injection split 10:1.

Finally, some evidence to support the initial claim –the fast sample transfer that occurs during a split injection reduces in-inlet degradation and adsorptive loss. We see improved tailing factors and the DDT breakdown number is significantly better. The improved breakdown number is not due to reduced sensitivity. As summarized in Table 1, the TIC area for DDT was actually higher under split because the gain factor was increased to maintain sensitivity at the low end of the calibration (another future blog topic).

breakdown table 1

Table 1 – DDT Breakdown comparison for the splitless and split injection examples.

3 Responses to “An introduction to the benefits of using split injection when performing semivolatiles analysis by 8270D – the instrument checkout mix”

  1. Steve Pruskin says:

    Hi Chris, Would you please tell us what column you’re using since that figures into it being overloaded. Also what instrument since that relates to the sensitivity i.e. what you can see after splitting. Thanks.

  2. I used an Rxi-5MS in the 30 m x 0.25 mm x 0.25 µm format. The full instrument conditions for split and splitless can be found here: http://blog.restek.com/?p=19801

  3. Steve Pruskin says:

    That explains it. I’m with a .32mm column and a much older mass spec.

Leave a Reply