LIBS and XRF: Complementary Analysis Techniques for Pharmaceutical Development

The atomic spectroscopy techniques of laser-induced breakdown spectroscopy (LIBS) and X-ray fluorescence spectroscopy have different strengths. Lydia Breckenridge, a senior research investigator at Bristol-Myers Squibb, uses both techniques in her work in pharmaceutical development. Here, she shares some of the advantages and challenges of using these techniques, and how the greatest benefits are sometimes derived by focusing on their complementarity, and using them in combination.

You use both X-ray fluorescence spectroscopy and laser-induced breakdown spectroscopy (LIBS) for elemental analysis in your work in pharmaceutical development (1). What are your main applications for LIBS, and what are the main advantages and limitations of the technique for your work? What are they for XRF?

In the atomic spectroscopy lab at Bristol-Myers Squibb, we have been exploring the use of LIBS for a variety of different applications for over a decade. We have applied its use to the characterization of tablet coatings and excipient homogeneity in formulated tablets, for the identification of counterfeit drugs, to the assessment of tablet hardness and compaction parameters (2), as well as for various catalyst remediation studies. However, we have found that the greatest niche for this technique in pharmaceutical development lies in its utility in foreign matter and quality investigations that require rapid answers to the fundamental question, “What is this?” LIBS has proven itself ideal for such investigations because of its minimal-to-no sample preparation requirements, microdestructive nature on limited solid samples, and provision of rapid, qualitative identification of elemental components. Reliance on the technique for typically qualitative analyses avoids the inherent limitations of poor precision and requirement of matrix-matched standards that are associated with quantitative LIBS.

XRF has also emerged as another solid-analysis technique in the analytical toolbox of pharmaceutical development. In our lab, my colleague Sharla Wood has had great success implementing the technique as a tool for process chemists to rapidly assess accurate, quantitative levels of residual catalysts (such as palladium) in solids or liquids (3). The advantages of XRF for this application include its portability, broad working range, and simplicity of use (new users need only undergo a brief initial training). Although limitations to the technique exist-namely, higher detection limits relative to conventional plasma-based techniques, matrix dependency, and relegation to elements of masses greater than sodium-with optimization, it has proved a suitable technique for fit-for-purpose applications.

How are the two techniques complementary in your work?

Both techniques share two significant similarities: They can be applied to solids with practically no sample preparation, and they can provide elemental fingerprints of the sample in question.  However, the major difference is that XRF can more easily provide relative, accurate quantitative data while LIBS can provide more broadband qualitative data on a much shorter timescale. As such, we have found they can be complementary techniques if LIBS is used as an initial, rapid screening tool before XRF analysis. For instance, upon receipt of an unknown foreign matter, LIBS can be directly applied to the sample and almost instantaneously provide information regarding composition, whether it is inorganic or organic, and, if the former, what metals may be present. With this knowledge, XRF can then be used (possibly on the same sample!) to both confirm the LIBS results and further provide relative concentrations of the elemental composition. Combined, the techniques can provide a more complete, focused picture of the identity (and potentially the source) of the foreign matter.

In a recent presentation on LIBS and XRF (1), you described a case study of pipe rouging in a vial washer system in which you used LIBS and XRF for pharmaceutical and solid analysis. What conclusions did you draw about the two techniques from this study?

Pipe rouging occurs when stainless steel passivation fails, resulting in oxidation of the outermost steel layer and subsequently the formation of iron oxide (rust). It often manifests as a red, particulate deposition either on the piping or accumulated further downstream of the oxidation event. Typically, the identification of iron is all that is required to confirm rouging. To this end, for this case study, the red residue was sampled using cotton swabs, which were subsequently analyzed by both LIBS and XRF with no additional sample preparation. Both techniques were more than capable of identifying iron as the primary component in the residue; however, LIBS provided the data more rapidly than XRF (less than two min versus approximately 20 min). Despite the speed of analysis, some spectral interpretation of the LIBS data may be required by a trained analyst, which may present a challenge if the technique is to be deployed in the field. Data processing on the XRF requires no interpretation or analyst expertise (following method validation).

Your presentation stated that the combination of LIBS and XRF is a good choice for quality investigations, and for raw and vendor material characterization. Are there other areas where these combined techniques might be useful?

LIBS and XRF can be utilized as complementary techniques for any application than can benefit from the rapid, nondestructive, qualitative screen provided by LIBS followed by a more comprehensive, element-specific quantitative analysis by XRF. Such application possibilities are endless but in pharmaceutical development could include solvent and reagent purity assessment, catalyst development support, contamination characterization, drug product formulation support, excipient quality control, and packaging integrity assessment. In some instances, one technique may be preferential to the other; however, in combination they can serve as mutually confirmatory techniques that provide a greater body of data with minimal resource expenditure relative to other more-involved confirmatory techniques such as inductively coupled plasma–atomic emission spectroscopy (ICP-AES) or ICP-mass spectrometry (MS).

Can you describe some of the challenges involved in combining LIBS and XRF in pharmaceutical analysis?

When applying both techniques to the analysis of a single sample it is important to ensure that the integrity and availability of the sample is maintained. To this end, while LIBS may be the technique of choice for first pass, preliminary analysis, it remains a microdestructive assay, and, in the case of powder analysis, will require that the sample is fixed (often to tape) before interrogation by the laser. Such sample manipulation and consumption may preclude subsequent analysis by XRF. Additionally, both techniques may suffer from interferences from other metallic elements in the sample or the matrix, which may limit the sensitivity of either technique for specific elements of interest. Finally, as with any solid-analysis analytical technique, the matrix matching of standards for accurate quantitative analysis continues to be a challenge. Both LIBS and XRF are highly matrix-dependent techniques and creating appropriate standards for either is a significant challenge in pharmaceutical development where sample of a specific matrix is often very limited or, in some cases, unknown or unavailable. 

What are your next steps in exploring or implementing the use of LIBS and XRF in pharmaceutical development?

We are continuing to devote significant effort to expanding the quantitative capabilities of XRF as a tool that can be used by our process chemist colleagues during synthetic optimization. This includes optimizing methods that can be applied to multiple matrices and for a variety of different elements simultaneously. Implementing the technique for this purpose has resulted in significant resource savings over more-traditional plasma-based techniques. LIBS continues to be the first-choice technique for rapid assessment of incoming foreign matter investigations and serves as a preliminary screen for determining which is the more confirmatory technique to employ. We plan to explore the use of custom blend standards in a fit-for-purpose matrix that can be applied for quantitative analysis comparison between both techniques.

References

1.     L. Breckenridge and S. Wood, “LIBS and XRF: Complementary Solid State Analysis Techniques in the Pharmaceutical Lab,” presented at SciX, Reno, Nevada (2017).

2.     A. Narang, L. Breckenridge, H. Guo, J. Wang, A. Wolf, D. Desai, S. Varia, and S. Badaway, J. Pharm. Sci.106, 200–207 (2017).

N. Lewen, M. Soumeillant, J. Qiu, J. Selekman, S. Wood, and K. Zhu, Org. Process Res. Dev. 19(12), 2039–2044 (2015).