Energetic chemicals used in explosives are generally toxic and remain a major class of compounds monitored by the EPA and other environmental agencies. TNT (2,4,6-trinitrotoluene), RDX (hexahydro-1,3,5-trinitro-1,3,5 triazine) and other explosives are consistently listed in the top 100 Superfund environmental toxins, especially at former military bases[1].
EPA method 8330 is an HPLC method introduced in 1990 for quantitating explosive compounds from environmental samples. The method describes a cleanup protocol of soil and water samples followed by HPLC analysis of 14 major explosives and related compounds. Because several of these compounds are so closely related that they differ by only minor aromatic substitutions, resolution of these compounds can be difficult on a fully porous C18 HPLC column, making accurately quantitating these compounds difficult. In the EPA 8330 method, this resolution issue was addressed by specifying a secondary column using a different selectivity to confirm the quantitation for several of these closely eluting compounds[2].
HPLC column technology has improved dramatically since the 1990s with improvement in silica bonding chemistries as well as silica morphology. Such improvements have delivered better resolution of closely eluting compounds, better analyte sensitivity, and shorter analysis times.
Core-shell media is the latest technological development where a narrow shell of porous silica media is grafted on to a non-porous silica core; the resulting particle morphology minimizes the analyte diffusion path in and out of a porous particle. This optimized diffusion path results in HPLC peak efficiencies significantly higher than what would be expected for a fully porous particle of the same size[3]. In comparison studies with Phenomenex’s Kinetex 2.6-µm columns, efficiencies for the core-shell media are generally observed to be on par or better than what are observed for most sub-2-µm media. Unlike fully porous sub-2-µm particles that require specialized UHPLC systems to handle the higher backpressures that such columns generate, core-shell materials operate at backpressures amenable to conventional HPLC systems.
Efforts were undertaken to see if the performance improvements theorized by core-shell technology are realized when performing demanding environmental applications where analytes of interest are closely related.
The resolution equation R= √N/4 * (ά-1/ά) * (1+k/k) indicates that any improvement in efficiency should increase resolution; however, that is provided that selectivity and capacity remain unchanged when using core-shell media.
Efforts were also undertaken to see if method times could be reduced while maintaining selectivity. Finally, alternate selectivities were investigated to determine if a confirmation column was still deemed necessary for the analysis of EPA 8330 with core-shell columns.
Materials and methods
Figure 1. Standard 5 µm C18 250 x 4.6 mm using gradient from 55 to 65% B in 9 min at a flow rate of 1.2 mL/min and detection at 254 nm. Components: 1. HMX, 2. RDX, 3. 1,2,5-Trinitrobenzene, 4. 1,3-Dinitrobenzene, 5. Tetryl, 6. Nitrobenzene, 7. 2,4,6-Trinitrotoluene, 8. 2-Amino-4,6-Dinitrotoluene, 9. 4-Amino-2,6-Dinitrotoluene, 10. 2,6-Dinitrotoluene, 11. 2,4-Dinitrotoluene, 12. 2-Nitrotoluene, 13. 4-Nitrotoluene, 14. 3-Nitrotoluene. Note the poor resolution for peaks 7-9 and the run time around 12 min. Click to Enlarge. |
EPA 8330 explosive standards were purchased from Sigma Chemical, St. Louis, Mo., and solvents were obtained from EMD Chemical, San Diego, Calif. An SPE manifold and strata-X-L 500-mg/6-mL SPE tubes were obtained from Phenomenex Inc., Torrance, Calif. A fully porous 5-µm C18 column (250 x 4.6 mm), Kinetex 2.6-µm C18 columns (150 x 4.6 mm and 100 x 4.6 mm) and a Kinetex 2.6-µm PFP column were manufactured at Phenomenex.
SPE was used to concentrate a low-level explosive mixture from dilute spiked tap water samples. In short, strata-XL tubes were conditioned with 10 mL of acetonitrile and then equilibrated with 30 mL of DI water. The dilute sample was loaded at 5 to 10 mL/min, washed with 5 mL of DI water, and eluted with 5 mL of 85:15 acetonitrile/water. Aliquots of sample were then diluted with DI water 3:1 and injected on HPLC.
HPLC analysis was performed using an Agilent 1200SL HPLC system equipped with an in-line degasser, autosampler, column heater and multi-wavelength detector; Chemstation software version A.08.03 was used for instrument control and data analysis. Water was used as mobile phase A, and methanol was used as mobile phase B. A gradient from 55% to 65% over 6 to 9 min was used for a majority of the separations at a flow rate of 1.2 mL/min. For accelerated methods, the gradient time was shortened to 5 min; in some cases flow rate was increased to 1.4 mL/min to utilize the increased optimal flow rate of core-shell media, as well as reduce run time. Analyte elution was monitored at 254 nm.
Results and discussion
Figure 2. Kinetex 2.6 µm C18 150 x 4.6 mm using gradient from 55 to 65% B in 6 min at a flow rate of 1.2 mL/min and detection at 254 nm. Components are labeled as in Figure 1. Note the significant increase in peak height and resolution versus the standard 5-µm C18 column, as well as the 50% reduction in run time. Click to Enlarge. |
SPE-based concentration from highly dilute tap water spiked samples was utilized to better approximate working with groundwater samples. Previously performed studies (data not shown) demonstrate roughly quantitative recovery of major energetic analytes using SPE cleanup/concentration. A methanol/water extraction could be adapted in conjunction with this method for soil samples to avoid the lengthy “salting out” methods specified in the sample preparation section of the EPA 8330 method[2].
To get a good benchmark on any performance gains that are recognized with core-shell media, a separation on a fully porous 5-µm C18 column (250 x 4.6 mm) was performed using the EPA 8330 standard mixture. The chromatogram is shown in Figure 1. It is readily apparent that several components are not baseline resolved, most notably TNT and related compounds. This lack of resolution between compounds typically found together is one of the reasons that a secondary column using a polar selective phase is recommended in the method to confirm quantitation of such analytes.
If improvement from next-generation C18 separation of explosives provided adequate resolution so that all analytes could be resolved, then a confirmatory column would not be necessary.
An example of the explosive mixture run on a Kinetex 2.6-µm C18 core-shell column (150 x 4.6 mm) is shown in Figure 2. Note that all of the analytes are better resolved than the 5-µm C18 column even though a shorter column was used (150 x 4.6 mm vs. 250 x 4.6 mm), and faster gradient was used (6 min vs. 9 min).
Upon closer inspection one can also note the near doubling of the peak height using the core-shell column for all the components versus the 5-µ fully porous column; such improvement in peak heights suggests significantly higher sensitivity using the core-shell media with subsequently lower LOD and LOQ for the Kinetex 2.6-µm C18 column.
Figure 3. Accelerated gradient using Kinetex 2.6 µm C18 100 x 4.6 mm. Flow rate was increased to 1.4 mL/min, and a shorter column was used to reduce run times. The gradient was also shortened to 5 min. Note the further reduction in run time to slightly >3 min while resolution maintained for most compounds. Click to Enlarge. |
Using the shorter column and gradient resulted in a +50% reduction in run time while exhibiting better resolution than the fully porous media. The selectivity of such media appears to be on par or better than the fully porous media for this application, and resolution is such that even aromatic isomers are nearly baseline resolved; this improved resolution may make the use of a confirmatory column unnecessary.
While using a 150- x 4.6-mm core-shell column realized a 50% improvement in throughput, more studies were performed to see if additional steps could further reduce run times while maintaining adequate resolution of similar compounds. Based on the van Deetmer equation, smaller particle diameter and core-shell media should demonstrate improved efficiency at higher flow rates than a 5-µm media; thus separations were performed with a shorter core-shell column (Kinetex 2.6 µm C18 100 x 4.6 mm) using a higher flow rate (1.4 mL/min) and shorter gradient (55% to 65% B in 5 min). The results of this faster analysis are shown in Figure 3. All the components are nearly baseline resolved in <3 min with the exception of the nitrotoluene isomers, which are only partially resolved. The faster analysis clearly shows the advantage of high-efficiency core-shell media over fully porous 5-µm media, with a 4-fold decrease in run time.
The EPA 8330 method specifies the use of an alternate selectivity column to the C18 column as a confirmatory column to better identify and quantitate compounds that elute at similar retention times on the C18 column. While a cyano column is specified in the method, many groups use different polar functionalized phases in place of a cyano column because of reproducibility issues often observed with CN columns. In these studies, the selectivity of a core-shell PFP phase (Kinetex 2.6 µm PFP 100 x 4.6 mm) was compared to the C18 core-shell to see if such media provided different selectivities for compounds from the EPA 8330A mixture. The results are shown in Figure 4. Both the retention time and elution order of the tested compounds are different for the two phases, indicating that the core-shell PFP is a useful confirmatory column for EPA 8330.
Conclusions
Figure 4. Different selectivity using Kinetex 2.6 µm 100 x 4.6 C18 versus PFP media. Flow rate was 1.2 mL/min. and gradient was from 55 to 65% in 6 min. 1. 1,3,5-Trinitrobenzene, 2. HMX, 3. Nitrobenzene, 4. 1,3-Dinitrobenzene, 5. 2,4,6-Trinitrotoluene, 6. RDX, 7. 2,4-Dinitrotoluene. The Kinetex 2.6-µm C18 column is shown in black, and the Kinetex 2.6-µm PFP column is shown in red. Note the changes in both elution time and order for the PFP phase. Results indicate that the Kinetex PFP provides alternate selectivity needed for an EPA 8330 confirmatory column. Click to Enlarge. |
The EPA 8330 environmental method identifies and quantitates closely related nitroaromatics and nitroamines by HPLC. As many of these compounds are geometric isomers, EPA 8330 is also a good test of HPLC column selectivity and suggests the use of a confirmatory column to verify the identity and quantitation of the analyzed compounds.
The recent introduction of core-shell media, which provide dramatically higher efficiency separations at HPLC backpressures, offers the possibility of switching to a high-performance solution for existing HPLC systems so that a confirmatory column may not be necessary (versus sub-2µ columns that require a new UHPLC system).
While a direct replacement of a core-shell column using existing conditions realizes improved resolution of energetic analytes, as well as reduced run times, increased efficiency offers the prospect of shortening column length and increasing flow rates.
Finally, alternate selectivity using a core-shell PFP was demonstrated for cases where a confirmatory column was desired with the same efficiency benefits achieved with the C18 column. These results demonstrate the utility and versatility of using core-shell media for environmental application like EPA 8330.
Fore more information, contact Michael McGinley at michaelm@phenomenex.com
References
1. 2007 Comprehensive Environmental Response Compensation and Liability Act (CERCLA) priority list of hazardous substances. www.epa.gov/superfund
2. 1994. EPA Method 8330: Nitroaromatics and nitramines by high performance liquid chromatography. Revision 0. www.caslab.com/EPA-Methods/PDF/8330.pdf
3. Experimental and specification data available on Kinetex Web site: www.phenomenex.com/Phen/EM/ws63990808/technology.html