Evaluation of the Ultra Inert Liner

Deactivation for Active Compounds

Analysis by GC

Technical Overview

Authors Abstract
Limian Zhao, Allan D. Broske, Endrin and DDT breakdown and active semivolatiles tests were used for the Ultra Inert
Agilent Technologies, Inc. liner deactivation performance evaluation. The results indicate that the Ultra Inert
2850 Centerville Road deactivated liners provide superior inertness for analysis of active compounds.
Wilmington, DE 19808
USA Introduction
David Mao, Allen Vickers
The analysis of active compounds by gas chromatography (GC) is challenging in areas
Agilent Technologies, Inc. such as pesticides, food, environmental and drug analysis. In order to achieve reliable
91 Blue Ravine Road and solid results for these active compounds, it is critical to minimize the interaction
Folsom, CA 95630 of active analytes along the GC flow path starting with the injector, to the column and
finally to the detector. The inlet liner plays a key role in influencing the inertness of
the entire flow path. Active sites on the liner can selectivity latch onto active analytes,
leading to the degradation or adsorption of these analytes, resulting in poor linearity
for calibration curves and loss of sensitivity. Therefore, it is critical to deactivate the
inlet liner completely to make it fully inert.
Many pesticides are labile, and intentionally designed to decompose easily so that
accumulation in the environment is minimized. Endrin and DDT are two well-known
compounds that can degrade excessively during gas chromatographic analysis if the
inertness of GC flow path is not well controlled, Endrin decomposes to Endrin
Aldehyde (EA) and Endrin Ketone (EK), and DDT degrades to DDE and DDD. Most
breakdown reactions occur on hot inlet surfaces. The degradation increases when
liner deactivation degrades with continuous use, and fragments of septa and non-
volatile residue from dirty samples slowly accumulate in the inlet. These create addi-
tional active surfaces and cause the breakdown reactions of Endrin and DDT to
increase. Therefore, Endrin and DDT breakdown is a very good probe to evaluate not
only the efficiency of liner deactivation, but also the stability of liner deactivation
treatment over the time of multiple injections. The calculation of Endrin and DDT
breakdown is listed below.
(Peak areaEA + Peak areaEK)
% Endrin breakdown = × 100
(Peak areaEA + Peak areaEK + Peak areaEndrin)

(Peak areaDDE + Peak areaDDD)

% DDT breakdown = × 100
(Peak areaDDE + Peak areaDDD + Peak areaDDT)

USEPA Method 8270 has been used widely to determine the concentration of semi-
volatile organic compounds in the environment. This method has been used to ana-
lyze a mix of acids, bases and neutrals that must be measured concurrently. This test
is a challenge for GC instrument due to the interaction of the analytes with GC flow
path, where the inlet liner can be a significant contributor for the system activities.
The active sites on liner surface can cause the unwanted adsorption of these com-
pounds, and lead to the loss of system responses. The most active compounds, such
as the nitrophenols, showed lower response factors (RFs) at the low concentrations,
and this causes poor linearity of the calibration curve and low sensitivity of the ana-
lytes. Among those active analytes, 2,4-dinitrophenol, usually showed RFs below EPA
method requirements and failed the run. A test mix used for evaluating the new liner
deactivation, including potentially troublesome compounds: N-nitrosodimethylamine,
aniline, 2,4-dinitrophenol, 4-nitrophenol, 4,6-dinitro-2-methylphenol,
4-aminobiphenyl, pentachlorophenol, benzidine, 3,3-dichlorobenzidine, benzo[b]fluo-
ranthene, benzo[k]fluoranthene. This test mix was used previously for the USEPA
8270 method improvements on the Agilent GC/MSD system. [1] A calibration curve
from 2 – 80 µg/mL was used for linearity evaluation.

2
Experimental
Chemicals and Reagents
Endrin and DDT stock standards were purchased from AccuStandard (New Haven, CT,
USA). 8270 customized standard and internal standards were obtained from Ultra
Scientific (North Kingstown, RI, USA). Ultra Resi-analyzed grade isooctane and meth-
ylene chloride were from J. T. Baker (Phillipsburg, NJ, USA).

Solutions and Standards

The 100/200 µg/mL Endrin and DDT stock solution was diluted 2000 times with
isooctane to 50/100 ng/mL Endrin and DDT testing solution. This testing solution
was stored at 4 °C in the refrigerator and used for no more than three days. Isooctane
was also used for solvent blank injection and syringe rinse
solvent.

The 8270 custom standard was purchased as 2000 µg/mL mixture in methylene chlo-
ride. A series calibration standards were prepared at 2, 5, 20, 40 and 80 µg/mL by
appropriate dilution with methylene chloride. The 8270 semivolatile internal standard
(IS) mixture at 4000 µg/mL in methylene chloride was spiked into standards with
appropriate volume to generate constant 40 µg/mL IS concentration.

Instrumentation
The Endrin and DDT breakdown test was done on an Agilent GC equipped with an
Agilent 7683B autosampler and a µECD. Semivolatile test was done on an Agilent GC
equipped with a 7683B autosampler and FID. Table 1 and 2 list the instrumental condi-
tions used on each test. Table 3 lists flow path
consumable supplies used in these experiments.

3
Table 1. Instrumental Conditions for Agilent GC/µECD System Used for Endrin and DDT Test

Autosampler Agilent 7683B, 10 µL syringe (p/n 5181-3354),

1 µL injection volume.
Preinj solvent A (Isooctane) washes: 4 (for sample run), 0 (for solvent run)
Sample pumps: 3 (for sample run), 1 (for solvent run)
Postinj solvent B (Isooctane) washes: 4 (for sample run), 1 (for solvent run)

Carrier gas Helium at 0.9 mL/min (31cm/s), constant flow

Inlet Splitless mode; 250 °C, 30 mL/min purge flow at
0.75 min
Analytical column Agilent HP-5MS UI column, 15 m × 0.25 mm,
0.25 µm, (p/n 19091S-431UI)
Oven profile For sample run: 120 °C (1 min), 30 °C/min to
220 °C (0 min), 8 °C/min to 280 °C (1 min).
For solvent run: 250 °C (5 min)
Detector µECD, 280 °C, Constant column + makeup flow with combined flow of
60 mL/min

Table 2. Instrumental Conditions for Agilent GC/FID System Used for EPA 8270 Semivolatile Active
Compounds Test
Autosampler Agilent 7683B, 5 µL syringe (p/n 5181-5246),
1 µL injection volume.
Preinj solvent A (Methylene chloride) washes: 1
Sample pumps: 3
Postinj solvent B (Methylene chloride) washes: 3
Carrier gas Helium at 3 mL/min constant flow
Inlet Splitless mode; 250 °C, 30 mL/min purge flow at
0.75 min
Analytical column Agilent Ultra 2 column, 25 m × 0.32 mm, 0.52 µm,
(p/n 19091B-112)
Oven profile For sample run: 40 °C (1 min), 15 °C/min to
310 °C (0 min)
Detector FID, 250 °C, H2/Air/Makeup N2: 40/450/45 mL/min

Table 3. Flow Path Supplies

Vials Amber screw cap (Agilent p/n 5182-0716)

Vial caps Blue screw cap (Agilent p/n 5182-0717)
Vial inserts 150 µL glass w/ polymer feet (p/n 5183-2088)
Septum Advanced Green Non-Stick 11 mm (p/n 5183-4759)
Ferrules 0.4 mm id, 85/15 Vespel/graphite (p/n 5181-3323)
0.5 mm id, 85/15 Vespel/graphite (p/n 5062-3514)
O-rings Non-stick liner O-ring (p/n 5188-5365)
Non-stick Flip-Top Liner O-ring (p/n 5188-5366)
Inlet liners Agilent UI single taper splitless liner (p/n 5190-2292)
Agilent deactivated single taper splitless liner
(p/n 5181-3316)
Agilent inert single taper splitless liner (p/n 5181-3316i)
Restek Siltek deactivated gooseneck splitless liner

4
Results and Discussion
The purpose of these tests was to evaluate the Agilent Ultra Inert liner deactivation
and compare it with other available liner deactivations. However, there are other fac-
tors possibly contributing to the unexpected degradation or adsorption of active com-
pounds, including the inlet gold seal, column, detector and so forth. Therefore, it is
critical to minimize other factors’ impact on the flow path activity, and also keep the
testing conditions consistent for accurate comparison.

We purposely used FID and µECD as detectors to eliminate any activity contributed
from the mass spectrometer. We also selected the Ultra Inert column to minimize the
column activity. Different liner configurations have different effects on the liner activi-
ties. Although, by using direct connect liner, the inlet gold seal activity contribution
can be eliminated, it does increase the operation difficulties, and has limited applica-
tions. As splitless injection has been used almost universally, the splitless injection
mode and single taper splitless liner configuration were used for this evaluation and
parallel comparison. A new gold seal was used for each liner test. It was recommend-
ed that a Cool On Column (COC) should be used to test the column performance [1]
because the on-column injection eliminates any inlet activity. The COC test was not
performed in the study. Firstly, splitless injection is much more practical in applica-
tions. Furthermore, the activity comparison can be made as long as all of tests were
performed at same conditions.

Endrin and DDT Breakdown Test

As described above, Endrin and DDT breakdown test was used for liner deactivation
evaluation. A 50/100 ppb Endrin and DDT solution made in isooctane was used for
multiple injections test. Before a new liner was installed, the inlet was inspected and
carefully cleaned if necessary. A new gold seal and washer were installed and a new
liner O-ring was used. An isooctane blank was normally injected as the very first run
on the tested liner. A standard sample was run after to collect the initial data for the
Endrin and DDT breakdown. Nine solvent injections were run after the standard run.
This cycle was repeated nine more times until 100 injections were made. One more
standard was then run, and this was the data for the 101st injection. A fast solvent
run method was used for solvent run to save the time. The Endrin and DDT standard
runs data were collected; and these data were used to generate the breakdown pro-
file. In addition to the Ultra Inert single taper deactivated liner, the Agilent proprietary
deactivated liners (p/n 5181-3316) and Restek Siltek deactivated liners with equiva-
lent liner configuration were tested for the parallel comparison. Multiple Ultra Inert
liners (n=16), Agilent Proprietary deactivated single taper liners (n=8) and Restek
Siltek Deactivated gooseneck liners (n=8) were tested. Data were collected for com-
parison.

Figure 1 shows the sample chromatograms of Endrin and DDT standards run on an
Agilent Ultra Inert deactivated single taper liner. The first injection delivered 1.2%
Endrin breakdown and 2.5% DDT breakdown; and the 101st injection had 12.2%
Endrin breakdown and 3.0% DDT breakdown. This data indicated that the Endrin and
DDT breakdown were well controlled by the liner inertness provided by the Ultra Inert
liner deactivation process.

5
 Hz % Breakdown 5 Peak identification:
Endrin DDT 1. DDE*
4000 1st injection: 1.2 2.5
5 2. Endrin
101 st injection 12.2 3.0 3. DDD*
2 2 4 Endrinaldehyde*
3000 5. DDT
6. Endrinketone*
2000 *Breakdown products

1000
1 3 4 6

1 3 6
0
5 6 7 8 9 min

Figure 1. Endrin and DDT breakdown test chromatograms by Agilent Ultra Inert single taper splitless
liner (p/n5190-2292)

Figure 2 shows the comparison of Endrin breakdown results over 100 injections for
Agilent Ultra Inert deactivated single taper splitless liners, Agilent Proprietary deacti-
vated single taper splitless liners, and Restek Siltek deactivated gooseneck splitless
liners. The DDT breakdown is well under control and below 10% for all of tested liners
over 100 injections. The results indicate that while all chemistries had similar initial
decomposition, the Agilent Proprietary and new Ultra Inert deactivations demonstrat-
ed greater stability throughout the test sequence.

30

25
% breakdown for Endrin

20

15

10
Agilent UI Liners (n=16)
5 Agilent s tp Liners (n=8)
Restek Siltek gooseneck Liners (n=8)
0
0 50 100
Injections

Figure 2. Endrin breakdown profile over 100 injections for Agilent Ultra Inert single taper splitless liner
(p/n 5190-2292) (blue), Agilent Proprietary single taper splitless liner (p/n 5181-3316) (red),
and Restek Siltek gooseneck splitless liner (green).

6
Semivolatile Active Compounds Test
In addition to Endrin and DDT breakdown test, an alternative and more sensitive test,
EPA 8270 semivolatile active compounds analysis, was conducted. A previously estab-
lished test mix [1] was used, which includes four phenols, several bases, and several
neutral components. These compounds were selected to not only represent 8270
active analytes, but also to be resolved easily and unambiguously detected by GC/FID.
Figure 3 shows a sample chromatogram of the test mix with 20 ng on column. The
EPA 8270 method does not specify a calibration range, yet traditionally a dynamic
range of 20 to 160 ng on column has been widely used in USEPA Contract Lab
Program (CLP). However, with the increased sensitivity of newer GC/MS systems,
users are moving toward lower and more challenging detection limits. A calibration
range of 2 to 80 ng on column was chosen for this analysis. The 2 ng on column quan-
titation limit gave the satisfied response and peak shape on FID with S/N ratio over 20
for 2,4-DNP, which has the lowest response. In order to evaluate the linearity of
GC/FID system over the calibration curve range, the Response Factor (RF) at each cali-
bration level was calculated as below and the overall RF values across the curve were
used to calculate the relative standard deviation (RSD).
Peak AreaAnalyte × ConcentrationInternal Standard
RF =
Peak AreaInternal Standard × ConcentrationAnalyte

According to USEPA 8270 method requirement [2], the RSD for each target analyte
should be less than 20%. Table 4 shows the average RF value across the calibration
range and RSD for each testing compound. All of compounds were below the EPA
requirement of less than 20% RSD over the 2-80 ng on column range.
3,3-Dichlorobenzidine data is not available due to its complete co-elution with peak of
IS 5, Chrysene-d12 (Figure 3).

Table 4. Average RFs and RSD Results for Tested 8270 Semivolatile Active Compounds Using Agilent
Ultra Inert Deactivated Single Taper Liners (n=6) (p/n 5190-2292)

Average RF over
Compounds 2 – 80 ng on column RSD (%)
N-Nitrosodimethylamine 0.563 3.1
Aniline 1.513 1.0
2,4-Dinitrophenol 0.288 15.6
4-Nitrophenol 0.457 3.9
4,6-Dinitro-2-methylphenol 0.382 8.5
4-Aminobiphenyl 0.913 1.1
Pentachlorophenol 0.299 6.1
Benzidine 0.619 4.8
Benzo(b)fluoranthene 0.938 3.3
Benzo(k)fluoranthene 0.897 7.1

• 3,3-Dichlorobenzidine data is not available due to complete co-elution

with IS 5.

7
 IS 2
9000000 IS 3 9, IS 5
IS 4
8000000
7000000
6000000 IS 6
5000000 IS 1
4000000 2 6 11
3000000 8 10
3 4 5 7
2000000 1
1000000

4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00

1 N-Nitrosodimethylamine 7 Pentachlorophenol IS 1 Dichlorobenzene-d4
2 Aniline 8 Benzidine IS 2 Naphthalene-d8
3 2,4-Dinitrophenol 9 3,3-Dichlorobenzidine IS 3 Acenaphthene-d10
4 4-Nitrophenol 10 Benzo(b)fluoranthene IS 4 Phenanthrene-d10
5 4,6-Dinitro-2-methylphenol 11 Benzo(k)fluoranthene IS 5 Chrysene-d12
6 4-Aminobiphenyl IS 6 Perylene-d12

Figure 3. A sample chromatogram of 8270 short test mix with peaks identification by Agilent Ultra Inert
single taper splitless liner (p/n 5190-2292).

Six critical compounds were selected for parallel comparison with other equivalent
liners. They are 2, 4-dinitrophenol (2,4-DNP), 4-nitrophenol (4-NP), 4,6-dinitro-2-
methylphenol (4,6-DN-2-MP), 4-aminobiphenyl (4-ABP), pentachlorophenol (PCP) and
benzidine. For the parallel comparison, 5 liner deactivation chemistries but equivalent
liner configurations were tested with repeats of n ≥ 4 for each type of liner. Table 5
shows the comparison results for five deactivations. Since 2,4-dinitrophenol
(2,4-DNP) is the most active compound, as shown with larger RSD, its’ comparison
data is shown in detail in Figure 4. The comparison data indicate that, in this EPA
8270 semivolatile active compounds application, the new Ultra Inert deactivation is
equivalent to or slightly better than the current Agilent Proprietary, Restek Siltek, and
Restek Sky liner deactivations and they all behave better than the SGE deactivation. It
should be notified that although they all give similar RSD values across the calibra-
tion range, the Ultra Inert deactivation generated higher responses for 2,4-DNP
(mean RF of 0.288) than other liners and this number is close to the COC data

Table 5. Average Response Factors and RSD Results Comparison of Five Liner Deactivations of Single Taper Splitless Liner (or Equivalent Configuration)
2,4-DNP 4-NP 4,6-DN-2-MP 4-ABP PCP Benzidine
Ave. RF RSD Ave. RF RSD Ave. RF RSD Ave. RF RSD Ave. RF RSD Ave. RF RSD
Ultra Inert Deact.
0.288 15.6 0.457 3.9 0.361 6.3 0.935 7.5 0.299 6.1 0.617 3.3
Liners (n = 6)

Restek Siltek Deact.

0.245 16.5 0.449 3.6 0.352 7.2 0.934 7.2 0.298 11.9 0.615 2.7
Liners (n = 4)

Restek Sky
0.262 18.2 0.433 5.4 0.352 9.1 0.881 2.4 0.292 5.7 0.495 12.9
Liners (n = 4)

Agilent Proprietary
0.262 15.2 0.475 4.2 0.379 4.9 0.938 7.7 0.325 16.4 0.631 3.1
Deact. Liners (n = 4)

SGE Deact. Liners

0.252 22.7 0.459 6.6 0.369 8.3 0.921 7.2 0.320 15.0 0.604 5.9
(n = 4)

8
 obtained previously.[1] This allows lower detection limits, as was demonstrated with
0.5 ppm of 2,4-DNP test done on GC-MS. As shown in Figure 5, Ultra Inert deactiva-
tion liners had higher response for 2,4-DNP at the concentration of 0.5 ppm. It should
also be mentioned that the average RFs value of benzidine on Restek Sky liners are
about 20% lower than other liner deactivations, and the RSD is 2-3 times higher than
other liners. This indicates that Restek’s Sky liners are less suitable for basic active
compounds like benzidine than other liner deactivations.

2,4-DNP RRF for single taper SL liner w/o wool

1.600
(Ave. RFs, %RSD)
0.288, 15.6
1.400
0.262, 15.2 0.262, 18.2
0.221
0.245, 14.5 0.249, 22.7
1.200 0.202 0.194
0.188
0.205
0.267
1.000 0.233
0.252
0.194
0.207 2

0.800 0.308 5
0.263 0.276 0.257
0.244 20

0.600
40

0.308 80
0.284 0.295 0.293
0.271
0.400

0.200
0.336 0.308 0.298 0.311 0.312

0.000
Agilent UI Agilent proprietary Restek Siltek Restek Sky SGE
deactivated deactivated deactivated deactivated deactivated

Figure 4. Parallel comparison of splitless single taper liners for Agilent Ultra Inert liners (p/n 5190-2292), Restek Siltek Deact. liners, Restek Sky
Deact. liners, Agilent proprietary deact. liners (p/n 5181-3316), and SGE deact. liners using 2,4-Dinitrophenol as a probe.

9
Figure 5. 0.5 ppm 2, 4-Dinitrophenol test by GC/MS SIM mode. Agilent Ultra Inert single taper splitless liners (P/N 5190-2292) give out higher responses for 2,
4-DNP than Restek Siltek liners at lower concentration.

Conclusion For More Information

Agilent Ultra Inert deactivated liners were evaluated and com- For more information on our products and services, visit our
pared with equivalent liners including the Agilent Proprietary Web site at www.agilent.com/chem.
deactivated liners, Restek Siltek and Sky deactivated liners,
using Endrin and DDT breakdown test and EPA 8270 semi-
volatile active analytes test as two probes. In the Endrin and
DDT breakdown test, the Ultra Inert deactivated liners had
<20% Endrin and DDT breakdown over 100 injections at 50 ppb
standard level. In 8270 semi-volatile test, the Ultra Inert deacti-
vated liners generated less than 20% RSD for RF values over
the calibration range from 2-80 ng on column. For the most
active compound, 2,4-Dinitrophenol, the mean RSD over 2-80 ng
on column was 15.6% with replicates of six. The Ultra Inert
mean RFs value of 0.288 for 2,4-DNP was higher than those
obtained by other liners. The results of liner deactivation evalu-
ation and comparison demonstrate that, given the equivalent
liner configuration, the new Ultra Inert liner deactivation is well
suited for high sensitivity analysis of active analytes. www.agilent.com/chem
Agilent shall not be liable for errors contained herein or for incidental or consequential
References damages in connection with the furnishing, performance, or use of this material.

1. M. Szelewski, B. Wilson, P. Perkins, “Improvements in the Information, descriptions, and specifications in this
publication are subject to change without notice.
Agilent 6890/5973 GC/MSD system for Use with USEPA
Method 8270.” Agilent Technologies publication © Agilent Technologies, Inc., 2011
5988-3072EN. Printed in the USA
March 9, 2011
2. USEPA 8270D method, 5990-7380EN
http://www.epa.gov/osw/hazard/testmethods/sw846/pd
fs/8270d.pdf