Modern Sample Preparation Methods for POPs

Persistent organic pollutants (POPs) pose an on-going threat to human health, but are often trapped within environmental samples, thereby making analysis challenging. Bethany Degg of The Column spoke to Lourdes Ramos from the Department of Instrumental Analysis and Environmental Chemistry Institute of Organic Chemistry of the CSIC (Madrid, Spain) about her innovative research on new sample preparation methods for POPs.

Photo Credit: Christine Schneider/Getty Images

Q. Why are persistent organic pollutants (POPs) a concern? When did scientists recognize the threat of exposure to human health?
A: Persistent organic pollutants (POPs) are mainly halogenated organic chemicals that are toxic; highly resistant to any type of chemical, biological, and photolytic degradation; and bioaccumulative. In addition, they can be transported long distances from the emission point by wind and water. These characteristics give POPs, also classed as PBTs (persistent, bioaccumulative, and toxic), the potential to adversely affect human health and the environment around the world - as recognized by the international agreement signed in May 2001 known as the Stockholm Convention.

POP classes include intentionally produced chemicals currently or once used in agriculture, disease control, manufacturing, or industrial processes, and by-products of industrial and combustion processes. Under the Stockholm Convention, countries agreed to reduce or eliminate the production, use, and release of legacy chemicals and to conduct a constant revision process, which has led to the gradual incorporation of other POPs of global concern to the list over the years.

Q. In your view, what are the main challenges associated with the analysis of POPs in environmental samples?
A: The biggest challenge from an analytical point of view is the complexity of the environmental matrices in which POPs are sorbed or entrapped, combined with the very low levels of detection required. This is particularly true in the case of (semi-)solid matrices and for those chemicals and by‑products that have already been subjected to specific regulations and controls, which are typically found at ultratrace levels in environmental samples. This combination of low concentration levels and matrix complexity has led to the use of labourious multi-step sample treatment protocols followed by highly sophisticated instruments for final instrumental determination. The aim is to ultimately ensure the quantitative extraction of target compounds and subsequently exhaustive purification and fractionation of co-extracted matrix components.

Generally speaking, the lower the concentration of the analyte to be determined, the more demanding the sample treatment. However, advances achieved over the years in the field of analytical instrumentation - in particular of mass spectrometry (MS) - allow highly sensitive and accurate analyses to be performed with instruments that are available in most academic and commercial laboratories today.

Unfortunately, the potential offered by advanced instrumentation has not always been used to simplify the previous sample preparation protocol.

Q. Your research interests include the development of new miniaturized sample preparation methods for determining trace organic pollutants with chromatographic techniques. What led you to begin this work and why is it important to develop miniaturized sample preparation methods?A: In a typical gas chromatography (GC)‑based analysis of POPs, the purified extract is usually diluted to a final volume of ca. 50–100 µL, of which only 1–2 µL are injected in the splitless mode in the analytical instrument. In practice, this means that only a small fraction of the initial amount of sample is used for final instrumental determination. Consequently, a simple reduction of the solvent used to dissolve this final extract to an acceptable volume of 10–20 µL could promote an in‑line reduction of the initial sample amount without affecting the detectability of the investigated analytes. Nowadays, using any of the large volume injection (LVI) techniques available can contribute to further reducing the initial sample amount while maintaining the total amount of analyte injected in the GC system. Today, it is therefore possible to significantly reduce the amount of sample use for POP determination in environmental samples without affecting the quality of the analysis.

In addition to being the best alternative when dealing with the analysis of size‑limited samples, miniaturization has many other advantages including faster sample preparation times, minimal sample manipulation, reduced reagents consumption, and reduced waste generation.

Q. You recently developed a miniaturized pressurized-liquid extraction (PLE) with in-cell purification method for the simultaneous extraction of endogenous PCBs in feedstuffs. Can you tell us how you developed this method?
A: This method can be considered a logical continuation of my previous research in the field of miniaturized sample treatment of biological tissues and fat-containing foodstuffs for the analysis of legacy PCBs. These former studies were based on the dispersion of freeze-dried tissue on the surface of an appropriated sorbent. As well as providing a sorbent-like mixture that can easily be packed in a solid-phase extraction (SPE) cartridge, this process contributed to the preliminary purification of the matrix extract. Packing of this homogeneous matrix solid-phase dispersion (MSPD) mixture on top of an appropriate co-sorbent (in this case, silica modified with sulphuric acid, SiO2‑HSO4) allowed complete sample treatment to be performed, in this instance, quantitative extraction of the target compounds plus fat removal in a single step with minimum reagents and time consumption. Despite its simplicity, this sample treatment procedure resulted in ready for analysis extracts and, when combined with LVI–GC–ITD (ion trap detection) (MS–MS), the methodology provided results similar to those obtained using the conventional large-scale procedure involving GC–high-resolution mass spectrometry (HRMS) in use in our laboratory, even for the less abundant dioxin‑like PCB congeners.

Q. What were the challenges you faced and how did you overcome them? What are the advantages of this approach compared to other methods?
A: The analysis of more sorptive analytes, like polybrominated diphenyl ethers (PBDEs), in a more sorptive material, like feedstuffs, made it necessary to adopt a different analytical strategy. Direct application of the previous method would have resulted in an undesirable increase of the final extraction solvent volume and of the analytical time. The use of one of the modern enhanced solvent extraction techniques, and pressurized-liquid extraction (PLE) in particular, was the most evident and advantageous alternative. However, commercially available PLE instruments were not really adapted for the treatment of small size samples under the principles of the minimum reagent consumption and waste generation we were interested in.

We consequently decided to develop our own miniaturized-PLE system. This homemade instrument had technical and operational specifications similar to commercial systems but was ready to hold tailored extraction cells that would fit the investigated sample. The latter provided an additional flexibility when setting up methods that were impossible to attain with the commercial instruments. In practice, that meant that we were ready to develop a new miniaturized PLE-based methodology for the simultaneous and exhaustive extraction of PCBs and PBDEs from highly sorptive matrices, such as feedstuffs.

The final optimized method was based on the MSPD of the water-normalized feedstuff with a mixture of anhydrous Na2SO4 and SiO2-HSO4. The homogenous and dried MSPD mixture was then packed in the extraction cell on top of SiO2-HSO4, which acted as co-sorbent and allowed the in-cell purification of the extracts. When combined with either GC–ITD(MS–MS) or GC coupled to negative chemical ionization (NCI)–MS, the proposed method allowed the accurate determination of the endogenous PCBs and PBDEs, respectively, at the levels typically found in commercial (that is, non‑contaminated according to the legislation then in force) feeds using only 250 mg of sample, 8 mL of organic solvent, and 3.5 g of sorbent. Complete sample preparation was done in less than 45 min, which sharply contrasted with the several hours of work required by more conventional large-scale (in this instance, off-line) approaches for these types of determinations.

Q. In another study, you developed an approach using ultrasound‑assisted extraction followed by disposable pipette purification for the determination of PCBs in small biological tissue samples. Why did you choose ultrasound-assisted extraction over other methods?
A: The determination of trace organic compounds in samples of very small size (less than 100 mg) represents an analytical challenge but, in my view, also an opportunity to explore the feasibility of new analytical approaches and configurations. With such a small sample size, the challenge starts before the analysis to ensure the representativeness of the sub-sample to be analyzed. In precedent application studies, freeze-drying of the investigated heterogeneous sample followed by its grinding, thorough homogenization and pre-treatment of a representative (2–5-times larger than required) amount of sample, contributed to minimize this problem. This approach was also followed in the present study. However, it should be noted that the idea of analyzing such small samples came from another research study carried out in our group.1 In that case, samples were actually that size (in fact, they were below 10 mg), but they were individual samples, thus eliminating the problem of its representativeness.

Manual sample handling can also become a problem when dealing with the analysis of very small-size samples. Apart from the (possible) difficulty associated with manipulation and quantitative transfer through the sample preparation procedure, the risk of contamination increases at this level, at which residual background levels can easily become apparent in analytical and laboratory blanks and ruin the analysis. At such a level, rigorous QC procedures should be adopted to prevent (and detect) any possible sample contamination through storage and processing. The use of appropriate (clean) laboratory material, recovery surrogates, and control samples becomes not only advisable but, in most instances, mandatory. In this context, the lower the manipulation of the sample, the better.

For this reason, we decided to develop a sample treatment that could be performed entirely in the Eppendorf in which such a type of sample could be collected and stored until analysis. With this idea in mind, treatment with a 2-mm ultrasonic tip probe appeared as the best alternative to ensure an exhaustive extraction of the target microcontaminants from the investigated biological tissues. After only 20 pulses of 2 s with the ultrasonic tip, the matrix was completely desegregated and PCBs were quantitatively extracted in the 150 µL of n-hexane used as extractant. The supernatant was then separated by centrifugation for 2 min and slowly aspirated with a micropipette into a 5-mL polypropylene tip modified to contain the clean-up sorbent. After 10 s of contact, the purified extract was ejected into a chromatographic vial or, in the case of very fatty samples, into a new (clean) Eppendorf to perform a second clean-up step. In all cases, ready-for-analysis extracts were obtained and, as demonstrated by the analysis of appropriate reference materials, when combined with GC–ITD(MS–MS), the proposed method allowed accurate determination of most of the investigated PCBs - even for such a small amount as 50 mg. In addition, and because of the simplicity of the operations carried out during sample treatment, the proposed method exhibited potential for (at least partial) automation.

Q. Have you been involved in other projects involving POPs analysis?
A: My PhD, completed in the mid-1990s, dealt with the development of analytical methods for the determination of PCBs and PCDD/Fs in fatty foodstuffs, a type of analysis considered a challenge at that time. I then undertook a postdoctoral stay in Amsterdam, at Vrije University, under the supervision of Prof. Dr. Udo A.Th. Brinkman. My goal was to learn about the systems and procedures used for miniaturized SPE of liquid samples and to subsequently explore the possibility of developing (as much as possible) equivalent set-ups and systems for the treatment of semi-solid and solid matrices. Whilst there, I had the opportunity to start working in that line of novel research in the late-1990s and developed some miniaturized methods for the analysis of pesticides and PAHs in non-fatty matrices, such as fruits, soils, and sediments. Back in Madrid in 2000, I continued with that type of investigation and I rapidly expanded to the analysis of POPs and other trace organic compounds in fat-containing samples, from serum to biological tissues and foodstuffs among others. Today, the development of novel miniaturized sample preparation procedures for the analysis of trace microcontaminants and components in (semi-)solid samples can be considered a well established and active research line in my department.

Q. Is sample preparation still the main bottleneck in environmental analysis?
A: Sample preparation is still recognized as the bottleneck of many analytical procedures. The efforts carried out during the last two to three decades have resulted in the development of a number of novel analytical approaches and techniques that have contributed to solve some of the most pressing shortcomings of conventional (large‑scale) sample treatment procedures, namely long analytical times, large consumption of sample and reagents, high risk of extract contamination because of their continuous manual manipulation, and generation of large amounts of waste. Today, on-line coupling (with or without automation) is a recognized and accepted approach in many application areas dealing with the analysis of gases or volatile compounds and with the treatment of liquid samples. Attempts to develop equivalent procedures for the treatment of (semi-)solid samples have been much more limited, probably because of the difficulty of the initial extraction step, for which large-scale (off-line) approaches are mainly used.

Results reported on the development of hyphenated systems for gaseous and liquid samples have demonstrated that miniaturization of the techniques and approaches are probably a key aspect when attempting (at least partial) integration of the different analytical steps. Some of the modern simplified, faster, cheaper, and greener techniques designed for the treatment of solid matrices (in particular MSPD and PLE) have demonstrated their potential in this field through a number of illustrative examples.2,3 However, to achieve a level of development similar to that shown at present for other solvent- and sorbent-based techniques used in coupled systems, more work is still required from both academia and companies, who should support and promote the development of appropriate analytical instrumentation.

References

1. J. Sanz-Landaluze, M. Pena-Abaurrea, R. Muñoz‑Olivas, C. Cámara, and L. Ramos. Enviro. Sci.Technol.49, 1860–1869 (2015).
2. J. Escobar-Arnanz, L. Ramos. TRAC Trends Anal. Chem. (2015) http://dx.doi.org/10.1016/j.trac.2015.02.023
3. J.L. Tadeo, C. Sanchez-Brunete, B. Albero, A. Garcia-Valcarcel, and R.A. Pérez, Central European Journal of Chemistry 10, 480–520 (2012).

Lourdes Ramos currently holds the position of Senior Scientific Researcher of the Spanish Scientific Research Council (CSIC, Madrid), at the Department of Instrumental Analysis and Environmental Chemistry of the Institute of Organic Chemistry. Her research interests include the development of new miniaturized sample preparation methods for the fast determination of trace organic pollutants in environmental and food samples and the evaluation of new chromatographic techniques, especially GCxGC-based approaches, for unravelling the composition of complex mixtures of organic microcontaminants. Member of the editorial board of various journals and invited editor of several special issues, she has published over 80 peer-reviewed scientific papers, 12 book chapters, and has edited a multi-authored book on comprehensive two-dimensional
gas chromatography.

E-mail: l.ramos@iqog.csic.es
Website: http://www.iqog.csic.es/iqog/es/node/18689

This article was first featured in LCGC's digital magazine, The Column. Click here to view the full issue>>