ADV06-15_1-412

32 / 413

| 1-800-356-1688 or 1-814-353-1300 | Feedback? E-mail 6
By Chris Rattray, Jack Cochran, and Chris English
• Perform large volume splitless injection with an unmodified Agilent-style split/splitless GC inlet.
• Reliably detect 1,4-dioxane down to 5.0 ppt in drinking water.
• Improve quantitative accuracy by introducing more analyte to the detector.
Global concern over the carcinogenic potential of 1,4-dioxane, along
with its identification as a Group 2B compound by the World Health
Organization’s International Agency for Research on Cancer (IARC),
has led to increased regulatory interest in this compound. For exam-
ple, as part of Unregulated Contaminant Monitoring Rule 3 (UCMR3),
the U.S. EPA is requiring increased monitoring of 1,4-dioxane in drink-
ing water and has revised the 1x10-6 cancer risk assessment level*
down to 0.35 µg/L. As a result, the proposed minimum reporting
level (MRL) for 1,4-dioxane as part of UCMR3 is 0.07 µg/L [1].
Concurrent solvent recondensation–large volume splitless injection
(CSR-LVSI), a technique described by Magni and Porzano [2,3], can
be advantageous when trying to analyze trace-level contaminants
in clean matrices like drinking water. Since more target compound is
introduced onto the analytical column, detectability is improved; how-
ever, a specialized injection port, such as a PTV, is generally required
for LVSI [4]. Building on work by chemists at Thermo Scientific, our lab
has been exploring the use of CSR-LVSI with a completely unmodified
Agilent-style inlet. We use a fast autosampler injection with liquid
sample band formation in a liner containing glass wool, a retention
gap press-fitted to the analytical column, and a starting GC oven
temperature below the boiling point of the solvent (see next page for
instrument setup and analytical conditions). Previously, we have suc-
cessfully analyzed a wide variety of compounds, including PAHs, BFRs,
organochlorine pesticides, and semivolatiles, using this technique (see
and enter “LVSI” in search). Here we assess its potential
to lower detection limits for 1,4-dioxane in drinking water.
Evaluating CSR-LVSI With a Standard Splitless Inlet
To determine if CSR-LVSI with an unmodified split/splitless inlet was
compatible with the volatile compounds in this application, linear-
ity and interferences were assessed. Calibration curves at levels well
below typical minimum detection limits displayed excellent correla-
Table I: Calibration curve (0.5–50 pg/µL).
Level Prepared
Standard (pg/µL)
10 µL Injection
On-Column Amount (pg)
Equivalent Concentration
in 500 mL Samples (µg/L)
1 0.50 5.0 0.010
2 1.0 10 0.020
3 5.0 50 0.10
4 10 100 0.20
5 50 500 1.0
tions across a wide range (R2 = 0.9998 for 1 to 1,000 pg/µL [10 to
10,000 pg on column] and R2 = 0.9996 for 0.5 to 50 pg/µL [5 to 500
pg on column]). Calibration levels and equivalent concentrations are
shown in Table I for the lowest curve, which was used to quantify
recoveries from extracted drinking water samples.
While results for injected standards were quite promising, this
analysis is very sensitive to interference from co-extracted material
because the SIM ions are at a relatively low mass to charge ratio.
Although CSR-LVSI introduces more matrix onto the column than a
typical injection, no interferences for 1,4-dioxane were observed. As
shown in the analysis of a fortified drinking water extract in Figure 1,
1,4-dioxane is chromatographically separated from any interferences.
Using CSR-LVSI to Lower Detection Limits
Having established that CSR-LVSI with an unmodified GC inlet is an
appropriate technique, we wanted to assess its potential for lowering
detection limits. The 10 µL CSR-LVSI in Figure 1 (approximately 5 pg on-
column) produced a signal-to-noise ratio of 16 for the quantitation ion
(m/z 88), which is above the threshold of 10. In contrast, when 1 µL of
the same extract was injected, the resulting peak is barely distinguish-
able from the noise and the confirmation ion cannot be seen (Figure 2).
Ultimately, the improved signal-to-noise ratios obtained using CSR-LVSI
resulted in recoveries of 1,4-dioxane and surrogate 1,4-dioxane-d8 that
were within the expected range (Table II) and that matched published
method development data very well [4].
Lowering Detection Limits for 1,4-Dioxane in DrinkingWater
Using Large Volume Injection in an Unmodified Split/Splitless GC Inlet
*A 1x10-6 cancer risk assessment level corresponds to the lifetime probability of one individual
in an exposed population of one million developing cancer.