Almost all sectors of analytical chemistry are finding applications for ambient ionization mass spectrometry (AI–MS) because of its ease of use, speed of analysis, and sensitivity. Although emphasis has been placed on developing new hardware that can help analyze unique samples across various applications, there has not been much innovation in the functionality of software tools and mass spectral libraries to support applications like non-targeted searching. In this article, we discuss new algorithms and libraries that have enabled non-targeted analysis of small molecules using AI–MS, as well as some of the key considerations and outstanding questions in the field.

From forensic science (1) to healthcare (2) to food analysis (3), ambient ionization mass spectrometry (AI–MS) continues to find new application areas (4). The appeal of AI–MS—which is listed in almost every article on the topic—includes minimal to no sample preparation, rapid analysis times, the ability to alter ionization chemistry on the fly, and desirable sensitivity. However, there is a one (major) drawback to AI–MS, which is that the resulting mass spectra can be difficult to interpret. Given the lack of chromatography, an AI–MS spectrum consists of ions generated from all compounds in the sample overlaid on top of one another.

For some applications, like olive oil authentication (5) or diabetes detection (6), it may be logical to consider an AI–MS mass spectrum as a fingerprint of the sample. In this case, a barrage of statistical tools (for example, principal component analysis [PCA], linear discriminant analysis [LDA], and machine learning) can be applied to the collections of spectra to group which samples are similar, identify the potential source of a sample, or answer another classification-related question. Other applications, like forensic seized drug analysis, require the analyst to understand the precise chemical makeup of a sample. In these applications, treating an AI–MS mass spectrum as a fingerprint and using classification tools is inadequate.

So how does one accomplish non-targeted analysis with AI–MS? Given that most mass spectral search tools are designed for chromatography approaches where the spectra of pure compounds are compared, we need to use new approaches. Recognizing that most AI–MS sources are soft ionization techniques, and therefore typically produce intact molecular ions (that is, protonated or deprotonated molecules), one approach is to identify individual peaks in the spectrum using simple lookup tables of compound 

 values. Although this approach can be fruitful, lookup tables can provide inconclusive results when several compounds share the same 

 (for example, isomeric or isobaric compounds). This limitation is exacerbated when using unit resolution instrumentation.

For example, take the AI–MS spectrum of a seized drug sample shown in Figure 1. The main peak at nominal 

 238 could be attributed to ketamine, a common drug of abuse, 4’-chloro-α-pyrrolidinopropiophenone, an isomeric synthetic cathinone, or both. As a result, we need a different approach if we want to determine whether one or both compounds are truly in the sample.

FIGURE 1: An example of (a) a low energy is-CID mass spectrum and (b) a mid energy is-CID mass spectrum of a seized drug sample containing ketamine. The table at the bottom provides the reverse match factor (RevMF) scores for ketamine and 4’-chloro-α-pyrrolidinopropiophenone (PPP) obtained by analyzing the low and mid is-CID spectra with the ILSA. Note the significant drop in score for 4’-chloro-α-PPP when the mid is-CID spectrum is included in the search. Scores range from 0 to 1 with 1 being a perfect match.

There are currently two approaches that improve the specificity of the lookup table approach: tandem mass spectrometry (MS/MS) with collision induced dissociation (CID), or in-source CID (is-CID). In MS/ MS, we can specify that the 

 238 ion should be isolated, fragmented, and a product ion scan collected. These product ions can then be used to help elucidate the structure of the ion. Of course, specifying an ion to be isolated is not ideal when performing a true non-targeted analysis, especially in AI–MS when your sample is only ionized for a few seconds. With is-CID, one or more “fragment spectra” are generated at multiple is-CID energies without any precursor ion isolation. The issue here is that product ions in fragmentation spectra cannot be directly associated with their precursors. If we rethink the data analysis approach, these high energy is-CID fragmentation spectra can actually be used as a rich source of information in non-targeted analysis.

The inverted library search algorithm (ILSA) is a new data analysis approach developed at the National Institute of Standards and Technology (NIST) to use multiple is-CID mass spectra and unlock non-targeted screening with AI–MS (7–9). As the name implies, the ILSA inverts the traditional mass spectral library search process. Instead of asking the traditional question of “Does my questioned spectrum match this known spectrum?”, we instead ask “Does this known spectrum match my questioned spectrum?” Given that AI–MS spectra are mixture spectra, there is an important additional piece to the ILSA question. Specifically, we ask, “Does the known spectra match a sub-pattern within my questioned spectrum?” The sub-pattern piece is critical because we must assume our questioned spectrum is not pure. Although we can do this on just one is-CID spectrum (a low energy is-CID spectrum that produces intact protonated ions), the value is increased when we combine multiple is-CID spectra, enabling us to not only match intact ions but also match fragment ions that are produced at higher energies. By implementing the ILSA alongside a similarity metric of choice, we can obtain scores that can be used for deciding the identity of a compound in the mixture. Consider our example in Figure 1. When utilizing the ILSA along with the reverse match factor and three is-CID spectra, we can see the scores from ketamine are much higher than those of 4’-chloro-α-pyrrolidinopropiophenone, allowing us to have more confidence that the sample contains ketamine.

To truly make use of any library search algorithm, one needs a high-quality mass spectral library. The ILSA currently uses the NIST Direct Analysis In Real Time MS (DART–MS) Foren- sics Database (10,11), a regularly updated resource that contains multiple is-CID mass spectra of over 1100 compounds of interest to the forensic community. The library was built and evaluated with the assistance of a combination of automated and interactive software tools that could be used to produce libraries of measurements collected with other AI–MS platforms as well.

There is still one major question that needs to be addressed before we can be fully certain about best practices for non-targeted analysis using AI–MS and the ILSA, and that has to do with the effectiveness of centrally collected spectral libraries. Given the rapidly changing landscape of AI sources and MS platforms, are the data collected using a particular source on a specific MS platform useful to a chemist using a different configuration? The corollary to that question is equally important—even using the same configuration of instruments, is there other laboratory-to-laboratory variability (12,13) that impacts the effectiveness of a centrally collected AI–MS spectral library? In traditional chromatographic techniques where ionization occurs under vacuum, spectral reproducibility across instruments and laboratories is quite high and so libraries (for example, the NIST series of electron ionization [EI]–MS libraries) are reliable resources. Some preliminary studies in our laboratory focused on the platform-to-platform utility of a central library are promising, but much more work needs to be done. Unfortunately, there is currently limited literature on the laboratory effects of AI–MS measurements (14–17). Answering these questions will help us determine the costs and benefits of centrally created libraries like the NIST DART- MS Forensics Database compared to using custom libraries created under the specific conditions of the laboratory.

Although our recent work has made non-targeted analysis with AI–MS practically possible, the final piece of the puzzle is the identification of novel compounds. The ILSA, like most mass spectral library search algorithms, is limited to identifying known compounds that are contained within the search library. For chromatographic techniques producing mass spectra of pure compounds, NIST has a few approaches (for example, MS Interpreter, hybrid search, fentanyl classifier) to assist in novel compound classification. These methods are not naturally amenable to is-CID mass spectra of mixtures but variations of them might be applicable with further work. Developing additional tools and resources for novel compound identification, using AI–MS or other mass spectral platforms, is a worthwhile pursuit.

Non-targeted screening using AI–MS and is-CID mass spectra could potentially provide critical, and previously unthinkable, capabilities to nearly all sectors of analytical chemistry in the future. Accomplishing this objective would require the development of algorithms and data analysis procedures that improve capabilities even more. Near-instantaneous non-targeted compound identification may seem far-fetched, but the research is moving in that direction and the road to success seems clear.

Certain commercial products are identified to adequately specify the procedure. This does neither imply endorsement or recommendation by NIST nor does it imply that such products are necessarily the best available for the purpose.

These opinions, recommendations, findings, and conclusions do not necessarily reflect the views or policies of NIST or the U.S. Government.

https://doi.org/10.1016/j.forc.2020.100294

(2) H. Su, M.-Z. Huang, J. Shiea, and C.-W. Lee, 

(3) A. Arrizabalaga-Larrañaga, J.F. Ayala-Cabrera, R. Seró, J.F. Santos, and E. Moyano, in 

, C.M. Galanakis, Ed. (Academic Press, Cambridge, 2021), pp. 271–312.

https://doi.org/10.1016/j.trac.2020.116046

https://doi.org/10.1007/s11306-020-1637-8

(8) A.S. Moorthy, S.S. Tennyson, and E. Sisco, 

(9) E. Sisco, A.S. Moorthy, S.S. Tennyson, and R. Corzo, 

(10) E. Sisco and A.S. Moorthy, NIST DART-MS Forensics Database (is-CID), (2020). 

(11) E. Sisco, A.S. Moorthy, and L.M. Watt, 

(19), 9603–9611 (2014). https://doi. org/10.1021/ac502075t

(14) G.A. Newsome, L.K. Ackerman, and K.J. Johnson, 

(15) S.R. Kumbhani, L.M. Wingen, V. Perraud, and B.J. Finlayson-Pitts, 

https://doi.org/10.1007/s13361-018-2105-9

(17) G.A. Newsome, L.K. Ackerman, and K.J. Johnson, 

https://doi.org/10.1007/s13361-015-1259-y

 is with the National Institute of Standards and Technology, in Gaithersburg, MD. Direct correspondence to: 

Articles

Non-targeted analysis of small molecules with ambient ionization mass spectrometry (AI-MS) is enabled by new mass spectral libraries and algorithms, such as the inverted library search algorithm (ILSA) developed at NIST.

Algorithms and Databases: Unlocking Non-Targeted Screening of Small Molecules with Ambient Ionization Mass Spectrometry

After three “lost” years, the 50th International Symposium on High Performance Liquid Phase Separations and Related Techniques, or HPLC 2022, returned. The symposium was held at the Town and Country Hotel and Conference Center in sunny San Diego, California, USA, from 18Ð23 June 2022. The HPLC symposium continues to be the premier event bringing together leading scientists in the field of liquid chromatography (LC) and related techniques. The conference was chaired by Frantisek Svec of Charles University and, especially under the uncertain times leading up to the event, was very successful and welcomed by the attendees.

After three “lost” years, the 50th International Symposium on High Performance Liquid Phase Separations and Related Techniques, or HPLC 2022, returned. The symposium was held at the Town and Country Hotel and Conference Center in sunny San Diego, California, USA, from 18–23 June 2022. The HPLC symposium continues to be the premier event bringing together leading scientists in the field of liquid chromatography (LC) and related techniques. The conference was chaired by Frantisek Svec of Charles University and, especially under the uncertain times leading up to the event, was very successful and welcomed by the attendees.

In this instalment of “Column Watch”, we report on the observed highlights and trends from the HPLC 2022 conference. In a similar fashion to the previous HPLC symposium review articles (1–3), many colleagues in attendance at the symposium were asked for their insights regarding the most interesting topics they observed at the event. Figure 1 attempts to capture the major topics of interest as determined by attendees in an infographic gleaned from these responses, and what follows is a synopsis of highlighted topics along with some personal views.

The topic of column technologies immediately took centre stage with the opening plenary lecture by Gert Desmet of Vrije Universiteit Brussel called, “Technological (R)evolutions in Column Making: Quo Vadis Chromatography?” Desmet provided a historical review of column development and highlighted the concept of Moore’s Law and its pertinence to developments in column technology. The “law” predicts that there will be a certain steady rate of gain in a given area of development until something revolutionary happens to change that rate of gain. Desmet went on to demonstrate how measurements of performance, such as column efficiency, have followed this law pretty well since the 1980s (see Figure 2), and asked the question, “how do we break out of it?” The advantages, disadvantages, and potential for some emerging technologies intended to create highly ordered structures such as three-dimensional (3D)-printing, micro‑machining, and micro-groove columns were discussed and contrasted. It was interesting to learn that the 3D-printing technology has developed enough to create the microstructures necessary for the creation of chromatographic devices, but the time to process is still extraordinarily long and materials amenable to printing are very limited.

In a plenary lecture entitled “3D-Printed Microfluidics for Solid-Phase Extraction and Biomolecule Separation,” Adam Woolley (Brigham Young University) discussed modern tools in 3D printing to create microfluidic devices for solid‑phase extraction (SPE) and biomolecule separations. By utilizing such a device, Woolley’s group developed a reliable assay for monitoring protein biomarkers for pre‑term birth. In the future, such a small device could be amenable to rapid detection in the clinic in real time; thus, it demonstrates an area of promise for 3D printing. In addition to the Woolley lecture, there was an excellent short course put on by Woolley and Greg Nordin (Brigham Young University) as well as a focus session within the programme featuring noted talks by Nordin and Rosanne Guijt (Deakin University). Although the 3D-printed liquid chromatography (LC) column may be a long way away, 3D printing is certainly gaining ground in the area of microfluidics and integrated fluidic devices.

In addition to 3D printing, there were several other prominent column and system technologies that stood out. These include an increase in the presence and number of portable LC systems, continued advances in microscale separations, and the current trend towards bio-inert or metal-passivated columns and systems.

Prominent at the symposium was the concept of miniaturized and portable LC instrumentation. Several talks were given, including “A Modular Approach to a Portable Analytical Separation System for Pharmaceutical Analysis” by Llewellyn Coates; “Taking the Chromatograph Out of the Lab: High and Low Pressure Field DeployableInstruments for Environmental Applications” by Brett Paull; “Evolving Capabilities of Compact, Portable Capillary LC” by Milton Lee; “Analysis of Small Molecules in Complex Mixtures using Compact LC–UV-MS Instrumentation” by Samuel Foster; and “Implementation of Compact Capillary LC into Analytical Workflows” by James Grinias. The latter presenter noted in his talk that portability affords “laboratory-to-sample” rather than “sample-to-laboratory”, which could be utilized in improving point-of-care and many environmental analyses. The reduced solvent and power consumption also leads to a lower overall cost as well as moving towards a more “green analytical chemistry” approach. Grinias demonstrated several applications where the portable LC systems were fit for purpose.

The upward trend in the development of portable instruments was also apparent on the exhibition floor with Axcend, Trajan, and CEM (Lucidity) all promoting their latest high performance LC (HPLC) systems.

Robert Kennedy of the University of Michigan presented “Capillary LC at 35,000 psi and Beyond for Complex Mixture Analysis”, a talk centred on using elevated pressure, combined with capillary LC–mass spectrometry (MS), to analyze complex mixtures. Kennedy’s group employed longer columns (20–100 cm) in length running at pressures of 35,000 psi to separate lipid and metabolite mixtures. With 1-m long columns, he demonstrated a peak capacity of ~1000 in 3-h analysis times, suitable for complex mixtures. When employing a novel 1.1 µm particle in this format, analysis times were reduced to 1/3 of those achieved by a more conventional 1.7 µm particle in a 15-cm long column at 15,000 psi. Taking this aspect one step further, Kennedy used two-dimensional (2D-) LC (hydrophilic‑interaction chromatography [HILIC] and reversed‑phase liquid chromatography) to analyze the lipid complement of plasma to yield a peak capacity of 1900 in a MS-compatible format (see Figure 3).

He also reported packing HILIC particles into 75 cm columns at 35,000 psi to yield 213,000 plate separations.

In his talk entitled “Nanoscale Separations to Enable Bottom-up, Top‑down and Native Proteomic Profiling of ng- and Sub-ng-level Samples”, Alexander Ivanov (Northeastern University) employed nano-flow methodologies with narrow-bore monolith columns to achieve sensitive results for analyzing sample-limited protein mixtures in a proteomics workflow. The use of low flow solvent delivery system (12 nL/min) was key to achieving the sensitivity needed in very low abundant protein biomarkers in the clinical specimen.

The need for small sample analyses (for example, in single cells) is proving the need for very narrow columns. Examples were also shown by Ryan Kelly of Brigham Young University in his talk, “Miniaturized LC Separations for Single-Cell Proteomics”. The need for reduced dilution and high resolution is, and will be, key for small samples. The technique also appears to be moving closer to clinical applications as exemplified later in this article under the “Highlighted Areas of Application” section.

 The topic of bio-inert or metal-passivated systems and columns was prevalent at HPLC 2022. As noted in the 

 2022 review of new LC columns and accessories (4), column and system hardware with various metal passivation strategies have been on the rise recently. David McCalley (University of the West of England, Bristol) kicked off the discussion with a highly cited, Monday morning keynote lecture entitled, “Managing Deleterious Solute Interactions with Metals in Hydrophilic Interaction Chromatography”. McCalley noted that interactions of analytes such as nucleotides and catecholamines with metal surfaces can occur, and there are several ways one may mitigate these undesirable interactions: (1) use columns/systems constructed from metals other than stainless steel; (2) use PEEK- or PTFE-coated systems; (3) use systems and columns deactivated by vapour‑deposition of some “inert” material; or (4) use complexing agents in the mobile phase. Although the focus of the work was geared towards the favourable impact of vapour-deposition deactivated columns and systems, the additional use of high pH as well as low-level complexing agents in the mobile phase provided the best overall separation of the adenosine phosphate probes employed.

In addition to several posters, many vendors were demonstrating their various approaches to bio-inert systems and columns. ThermoFisher displayed their Vanquish systems that replace stainless steel with titanium or MP35N components. Shimadzu showcased Nexera XS Inert, their bio-inert system that utilizes PEEK‑lined stainless steel to offer a metal-free flow path. Waters delivered numerous talks and presented quite a few posters on their Premier systems that utilize chemical vapour deposition of hybrid silica to block metal interactions. Agilent also had their suite of biosystems, which were either metal-free (PEEK) or utilize alternative metal alloys labelled as “iron-free”. Silcotek was also present and showcased their chemical vapour deposition solutions and how their product performance is on par with PEEK when it comes to keeping LC flow paths metal-free.

As noted later in the “Highlighted Areas of Application” section, HILIC was also prominent in various aspects of practical research, and HILIC is also often employed in multidimensional chromatography.

Multidimensional chromatography was a large part of the show, with two dedicated sessions as well as many lectures utilizing various multidimensional techniques throughout the conference. Peilin Yang (Dow Chemical) presented an excellent lecture entitled, “Multidimensional Separations with Simultaneous Evaporative Light Scattering Detection (ELSD) and MS Detections for Complex Industrial Polymers”, where she showed examples of how comprehensive 2D‑LC can be a powerful tool for polymer analysis. Yang focused her examples on industrial applications such as terpolymer composition heterogeneity, tri-block polymeric impurities, and nonionic copolymer surfactants analysis. Combinations of size-exclusion chromatography (SEC) and reversed‑phase LC, as well as HILIC and reversed-phase LC, were highlighted along with the intelligent use of various detectors based on information needs. Unfortunately, little detail on any additional multidimensional talks and topics were available at the time of writing. However, it is interesting to note that attendee thoughts on the adoption of multidimensional chromatography in the industry varies widely—some indicate a clear adoption whereas others feel this is still solidly in the hands of academics.

Several separation modes were considered dominant in their presence this year by conference attendees. Notably supercritical fluid chromatography (SFC), HILIC, and multidimensional separations seemed to take centre stage.

A number of colleagues expressed surprise at the number of presentations focusing on SFC, and many were perplexed by the lack of adoption of SFC in general practice. Some also expressed the sentiment that perhaps SFC is finally breaking through towards acceptance. As for the conference, there were presentations on SFC fundamentals, practical tips and tricks, and, as noted in the forthcoming “Highlighted Areas of Application” section, important real‑world applications in cancer diagnosis.

In his talk entitled, “Measurement and Modelling of Longitudinal Diffusion in Supercritical Fluid Chromatography”, Ken Broeckhoven of Vrije Universiteit Brussel measured both molecular diffusion and longitudinal diffusion (that is, the 

-term) in SFC, which allowed the authors to model particle diffusion and surface diffusion, possibly for the first time, in SFC. In another notable talk, Chun-Kai Chang of the University of Minnesota used molecular simulations of the mobile phase composition stationary phase pores to better understand retention mechanisms in SFC in the talk entitled, “Exploration of the Retention Mechanism in Supercritical Fluid Chromatography by Molecular Simulations”. Although they only showed results for C18 surfaces, this work might shed some light on the rather poorly understood retention mechanisms in SFC.

Attila Felinger of the University of Pecs discussed the adoption of methanol on SFC stationary phases in his talk entitled, “Solvent Interactions in Supercritical Fluid Chromatography”. Lucie Novakova of Charles University (“Problems of Stationary Phases in SFC”) followed this up by providing highly practical evidence on how the history of using a column in SFC can strongly affect separation. Columns exposed to additives and organic modifiers may be chemically changed so as to perform differently than the original or a new column. Novakova suggested dedicating columns to certain methods because of this, but also noted that the change is not necessarily permanent. The group is running a large amount of experiments to test a wide range of columns and conditions to find adequate procedures to better understand these findings.

SFC was also noted for its potential as a more “green” technique compared to HPLC. Stefan Lamotte (from BASF) provided a keynote lecture entitled, “The Good, the Bad, and the Ugly - Towards the Separation of Polar Analytes in Supercritical Fluid Chromatography”, where he provided comparisons of SFC with HPLC and HILIC as well as examples of important SFC separations. Lamotte noted that SFC has several real advantages over HPLC including speed of analysis, but noted solubility issues in SFC may need to be overcome. In a post-conference communication, Lamotte added that the chemical industry is clearly being asked to minimize waste and develop sustainable methodologies, and his efforts are clearly moving in the direction of SFC.

HILIC was a hot topic of discussion. The topic garnered its own session, and there were several additional HILIC-based talks throughout the conference. As mentioned previously, David McCalley’s talk on the impact of metal passivation in HILIC was a noted highlight. Brady Anderson (University of Michigan) presented a well‑received lecture on the power of HILIC at pressures up to 50,000 psi. Thomas Walter and Jinchuan Yang from Waters Corporation gave excellent talks on characterizing and using a zwitterionic HILIC phase, respectively. Joe Pesek from San Jose State University discussed method development for hydrophilic compounds using aqueous normal phase (ANP) chromatography and contrasted this to HILIC. Although the modes employ similar mobile phase compositions, ANP uses silica hydride-based stationary phases as opposed to more common silica-based chemistries (see Figure 4). Pesek noted that this silica hydride surface exhibits less than one-half a monolayer of water, whereas silica-based columns can have between 4–10 monolayers under similar mobile phase conditions. It is this difference that Pesek surmises results in differences in retention mechanisms (dominant adsorption in ANP compared to dominant partition in HILIC) and other attributes such as fast equilibration. Separation examples are then provided for several important classes of polar analytes.

Pressure-Enhanced–Liquid Chromatography (PE–LC): 

Many colleagues noted Szabolcs Fekete’s talk on how pressure can be used as an additional and unique parameter to tune the selectivity of large molecules in LC as an interesting highlight of the conference. Fekete introduced pressure as a potentially important parameter for selectivity tuning, especially for large molecules, as pressure may impact analyte tertiary structure. Equilibria associated with separation processes such as ionization, adsorption, partition, and complexation are also pressure-dependent. Fekete also noted that pressure may also impact the stationary phase structure and thus the interactions available for retention and selectivity. A system utilizing a pressure-regulating fluid interface was described and utilized to test the retention and selectivity dependence of a set of large and small molecule probes. PE–LC was shown to provide unique separations and be applicable to large molecules. PE–LC also showed efficacy in several modes of chromatography including reversed-phase LC, ion-exchange, hydrophobic interaction chromatography (HIC), and ion-pair reversed-phase LC.

Machine Learning and Molecular Dynamics Simulations:

Many attendees noted the increased interest and prevalence of machine learning and molecular dynamics as tools for improving our understanding of chromatographic principles. Two talks on molecular dynamic simulations and model reconstruction of porous media to predict chromatographic separations were given by Ulrich Tallarek from Philipps-Universität Marburg and Fabrice Gritti from Waters Corporation. Gritti contrasted the molecular dynamics approach to more traditional modelling methods such as empirical modelling and rate theory measurements (both with limitations). Although the molecular modelling requires a deep understanding of physical chemistry method and techniques, it may lead to a vast expansion of our knowledge of retention and mass transfer mechanisms as well as improve the robustness of LC modelling.

Machine learning approaches were also noted as an emerging trend. Torgny Fornstedt from Karlstad University gave a talk on ion-pair chromatography of oligonucleotides, discussing the importance of machine learning to interpret and act on the abundance of data collected for these complex analytes. In the following lecture, Imad Haidar Ahmad from Merck & Co. discussed the multifactorial modelling of data for multidimensional chromatography in an attempt to facilitate method development in both one- and two-dimensional separations.

A major theme throughout the conference centred on green analytical chemistry and some of the major topics discussed previously, such as SFC, microscale separations, and even modelling, can be related to this theme. As with other areas of chemistry and life in general, analytical chemists will certainly be called upon to get creative and generate less waste, use less power, and establish overall sustainable methods of analyses moving forwards.

The application of modern separation tools in life sciences was once again prominent at the symposium. Whereas recent HPLC meetings have focused largely on proteomics and related therapeutics (that is, monoclonal antibodies [mAbs] and antibody–drug conjugates [ADCs]), HPLC 2022 also had a strong focus on oligonucleotides, lipids, and the emerging general topic of personalized medicine.

Oligonucleotides, RNA, DNA, and Nucleic Acids:

The most recognized applied topic for the conference was centred on the separation of oligonucleotides, including various ribonucleic acid (RNA) and deoxyribonucleic acid (DNA) classes. A simple search of the conference technical programme using the term “oligonucleotide” revealed the topic was highlighted not only in many lectures and posters but also in several vendor seminars and short courses. The focus on oligonucleotides likely stems from the location of the conference (San Diego is a recognized hotbed for oligonucleotide research), recent success of oligonucleotide‑based therapeutics, and, of course, the elevated general community recognition of oligonucleotides because of Covid-19.

In a session dedicated to nucleic acids and oligonucleotides, Kelly Zhang, Claus Rentel, and Alexandre Goyon from Genentech impressed many attendees with their recent efforts in analyzing oligonucleotides using clustered regularly interspaced short palindromic repeats (CRISPR). In her lecture entitled, “Characterization of Guide RNA for CRISPR Gene Editing”, Zhang described methods for characterizing guide RNA (gRNA) for CRISPR gene editing systems and therapeutics. Zhang described ion-pair reversed-phase LC (IP-reversed-phase LC) strategies and noted the best results were achieved by using large pore stationary phases, triethylamine (TEA)- and hexafluoroisopropanol (HFIP)-supplemented mobile phases, and temperatures less than 40 °C to resolve variants. For fine discrimination of gRNA that only differ by one nucleotide, Zhang employed the use of HILIC–MS. Overall, Zhang’s platform of methods includes IP-reversed-phase LC, SEC, and HILIC–MS in analyzing these biotherapeutics.

Rentel discussed the analysis challenges and methods for synthetic oligonucleotides by LC–MS in the paper “Analytical Techniques for Oligonucleotide Therapeutics”. Three different modes of chromatography were employed in characterizing oligonucleotides: ion-pairing, reversed-phase chromatography (IP-RPC) using various IP reagents, weak anion exchange (WAX) chromatography, and metal ion complexation chromatography. The IP-RPC was used to resolve positional isomers of nucleic acids to detect truncated or modified oligos. The WAX chromatography was employed to resolve diastereoisomers and deamidation by-products during formulation. Alternatively, the metal ion complexation chromatography methods were also used to examine diastereoisomers.

In the lecture entitled “Development of Innovative Analytical Approaches for the Separation of Large RNA Oligonucleotide Impurities and Their Sequencing”, Goyon discussed a 2D-LC–MS platform to analyze these impurities. The first-dimension column was packed with various nuclease enzymes to digest the large oligonucleotides into smaller fragments that would be more amenable to ionization and interpretation in the MS instrument. Goyon shared with the audience that this technique is successful in allowing for the detection of very minute modifications in the oligonucleotide product, including even single oxidation events on one nucleotide. The second-dimension column was a HILIC phase because it provided better resolution than several IP-RPC columns and methods.

In a second session dedicated to nucleic acids and oligonucleotides, Gregory Jones of Alnylam Pharmaceuticals discussed ways to automate impurity characterization of oligos utilizing multidimensional chromatography in his talk, “RNAi Therapeutic Impurity Characterization: Workflow Automation using Multidimensional Chromatography”. Also noted as a highlight by several attendees was Ken Cook’s presentation, “Optimization of mRNA Sequence Analysis by UHPLC/HRMS”.

Separation of oligonucleotides is complicated and, as expressed by many of the presenters, requires a number of separation tools. IP-RPC and HILIC appear to be the main modes of separation, sometimes in combination (multidimensional approaches). It was also evident that column and system metal-passivation (bio-inert) is highly desirable for this class of analytes.

A highly touted lecture from Michal Holčapek, University of Pardubice, highlighted the use of ultrahigh-performance supercritical fluid chromatography coupled with MS (UHSFC–MS) for high-throughput quantitation of lipids in biological samples. Holčapek described the SFC approach and compared it to HILIC and reversed‑phase LC separation modes (also coupled with MS). Although each mode has its advantages, the UHSFC–MS system offered excellent selectivity and sensitivity (among other attributes) at comparatively short analysis times (see Figure 5). Holčapek went on to describe how the methodology was impressively applied to the early diagnosis of pancreatic and other cancer types. In a post-conference communication, Holčapek noted that “We have already established the spin-off company LipiDiCa (abbreviation of Lipidomic Diagnosis of Cancer), hired [our] first employees, are installing instruments in the laboratories now, and we are working on the preparation of clinical validation, which is a crucial step in our ambitious effort.” The reader is referred to two recent key papers that the presentation was derived from in the literature (5,6).

Lipid research was also well represented by both Robert Kennedy (“Capillary LC at 35,000 psi and Beyond for Complex Mixture Analysis”) and Erin Baker (“Combining Multidimensional Measurements with Standards and Isotopologue Workflows to Detect, Identify, and Validate Molecules in Omic Studies”). Analysis of lipids continues to a big opportunity for separation science, as they are challenging analytes to identify with MS alone. Isomers, underdeveloped databases, and a huge range of sizes and polarities make them a gold mine for separations research.

The emerging topic of personalized medicine was very apparent at the meeting. In her plenary lecture in the opening session, titled “Next Phase of Precision Medicine: Application of LC–MS”, Jennifer Van Eyk (Cedars-Sinai Medical Center) discussed the utilization of LC–MS in personalized medicine.

Perhaps the most personally memorable lecture of the conference was given by Steven Ray Wilson (University of Oslo) entitled “Organoids, Organ-on-a-chip, and HPLC”. After describing the advantages and disadvantages of current biomedical research models for disease modelling, Wilson introduced the ideas of using organoids, or laboratory-grown organ models derived from human pluripotent stem cells (Figure 6). The research effort aims to couple organoids with LC as a means for improving, among others, precision medicine and drug development. After describing initial success studying the role of oxysterols in non-alcoholic fatty liver disease using a basic “offline” setup, Wilson went on to describe a more elaborate liver “organ‑on‑a‑chip” for drug metabolism studies that utilized electromembrane extraction (EME) to sample solutions in contact with the operative organoid. Lastly, Wilson described packing “columns” with organoids, representing a new area of research where devices could be developed for research purposes (that is, drug metabolism studies) or as bioreactors. For the latter, the potential of a “pancreas-on-a-chip” was assessed by the online coupling of a disulfide bond cleavage reactor prior to MS analysis for measuring excreted insulin. The disulfide bond cleavage was used to fully characterize the insulin released from the organoids when using a nano-LC-based setup.

The HPLC 2022 symposium was a welcome return to face-to-face communication that was well-organized and lively. The meeting once again engaged researchers interested in analytical science from around the globe. The conference was full of new and expanding column and system technologies, offered new insights into many different separation modes, and exhibited important application to our craft in real‑world situations. The meeting also revealed some clear emerging topics such as a focus on oligonucleotide separations, the need for metal passivation of instruments and columns, and the increased focus on green analytical technologies. That said, it was mentioned by many attendees that there were no monumental or revolutionary developments noted at the conference, so it appears the science is still trending in accordance with Moore’s Law. Be it from 3D-printing advancements, developments in microscale separations, or perhaps revelations from molecular dynamics studies, the scientific community lies in anticipation for that next revolutionary discovery.

Coverage of such a large symposium is impossible without a great amount of assistance. The authors would like to acknowledge the invaluable input and discussion from Robert Kennedy, Joseph Pesek, David McCalley, Steven Ray Haakon Wilson, Ken Broeckhoven, Mariosimone Zoccali, Michal Holčapek, Alan McKeown, Michael Lämmerhofer, Kevin Schug, Stephen Groskreutz, Andrea Gargano, Martina Catani, Alberto Cavazzini, Fabrice Gritti, Szabolcs Fekete, James Grinias, Michael Dong, Nathaly Reyes, Diego Lopez, Jesse Bischof, Jason Anspach, Mark Schure, and Gert Desmet for providing notes, text, insights, figures, and enlightening discussions.

D.S. Bell, C.E. Muraco, and C. Flannery, 

D. Wolrab, R. Jirásko, E. Cífková, et al., 

D. Wolrab, R. Jirásko, O. Peterka, et al., 

 is the Biomolecule Workflows Manager at MilliporeSigma, the life science business of Merck KGaA, Darmstadt, Germany.



is a Research Fellow in Research and Development at Restek.

This instalment of “Column Watch” presents many of the highlighted topics and trends observed at the symposium.

Highlights from the 50th International Symposium on High Performance Liquid Phase Separations and Related Techniques

As I write this, I’m also grading the final exam from my “Chromatography and Separations” course. We spent approximately the first 75% of the course covering column chromatography, capillary electrophoresis (CE), and field-flow fractionation (FFF). The remainder of the course focused on chromatographic sample preparation. The final exam was based on this last section of the course. Some questions make a good review of the fundamentals of various extraction techniques. The basis for these questions may have been discussed in previous iterations of this column, in analytical chemistry textbooks, or in books on chemical separations. I generally give the final exam in my graduate courses as a take-home exam, so you are free to use whatever resources you have available to you. Here’s your opportunity to test yourself, as the answers are provided.

Question 1: Consider a system comprised of water and a neutral solute. An aliquot of this mixture is used to partially fill a septum-capped vial and solid-phase microextraction (SPME) is used to extract the solute from the headspace vapour. Using Gibbs’ Phase Rule, how many degrees of freedom describe this extraction system? Considering these degrees of freedom, what are the important parameters in developing this extraction?

, describes systems at thermodynamic equilibrium, where 

 is the number of degrees of freedom describing the system; 

 is the number of components in the system; and 

 is the number of phases. Because the phase rule applies to “pressure/volume/temperature [pvT]” systems, the constant 2 accounts for the pressure and temperature of the system. For example, if we have a single solute that may partition between an aqueous and an organic phase, there are three components (the aqueous and organic phases and the solute) and two phases (the aqueous and organic), so 

. Thus, if the temperature and pressure are constant, the concentration of solute in one phase dictates the concentration in the other phase because the system is at equilibrium. In other words, concentration is the third degree of freedom, which provides the basis for the concept of distribution ratios. In a slightly more complex, illustrative example, consider that the solute in the previous example is an amine. Now we have four components (the aqueous and organic phases, the deprotonated amine and its protonated counterpart) with the two phases, or

. As in the previous example, temperature, pressure, and concentration are degrees of freedom that describe the system. The fourth degree of freedom is the pH (relative to the pKb of the amine), which determines whether the amine is protonated or deprotonated.

Getting back to our question, in this case, we have four components (water, solute, headspace vapour, and SPME sorbent) and three phases (water, headspace vapour, and solid-phase microextraction [SPME] sorbent), or 

. In addition to the temperature and pressure, because all three phases are in equilibrium with each other, [solute]

, the concentration of solute in one phase dictates the solute concentration in all phases. Thus, headspace SPME can be considered a quantitative technique provided the SPME fibre is exposed to the headspace for a sufficient period to allow equilibria to be attained.

Question 2: Refer to the graph in Figure 1 for a given pair of extractions. For each of the two data sets depicted, state what property limits the extraction from producing quantitative yields? How might you adjust the extraction conditions to produce a more quantitative yield in each case?

The kinetics of most extractions generally follows the curve shown in Figure 1. In these plots, the cumulative amount extracted is shown as a function extraction progress, which may be, for example, solvent volume or time in a flow‑based extraction like pressurized liquid extraction (PLE) or supercritical fluid extraction (SFE), individual separative funnel extractions, or cycles in a Soxhlet apparatus. The kinetics show a rapid extraction period because of analyte solubility in the extraction solvent under the prescribed conditions, followed by a slowing of the curve, where the extraction kinetics become dependent on the diffusion of the solvent (with or without dissolved analyte) through the sample phase, asymptotic to the maximum amount extracted. Figure 1 demonstrates two different extraction examples.

In the upper blue curve, the initial slope is quite large, demonstrating high solubility of the analyte in the extracting solvent. Approximately 55% of the analyte is extracted in one cycle and a combined two-thirds following the second cycle. However, the total extraction never exceeds 70%. We can conclude that the rate-limiting step for this extraction is the diffusion of the extracting solvent into the pores of the sample particles (these diffusion-limited situations are much less prevalent when extracting liquid or gaseous samples) to solvate the analyte trapped inside the sample. There are two primary means of overcoming the diffusion limitation. The first is to shorten the diffusion pathlength (that is, increase the sample surface area) by grinding the sample. Extractions of solids are generally faster as sample particle size decreases. For example, consider making a pot of coffee with whole beans as opposed to grinding the beans! The one thing to watch out for in this situation is to minimize grinding the sample too fine (say the consistency of flour or talcum powder) to avoid the particles packing together and impeding solvent flow. The second means to address this situation is to enhance the diffusivity of the solvent. One may choose a less viscous solvent, but solubility is not the issue, so we would likely stay with the original solvent used. Supercritical fluids are one type of potential extracting solvent exhibiting fast diffusion rates. The easiest way to improve solvent diffusivity is to heat the system. As the temperature increases, the diffusion is faster, which is the principle behind techniques such as PLE or microwave-assisted extraction (MAE). PLE and MAE allow for higher temperatures than the atmospheric boiling point to be used. One caveat is that we must avoid thermal decomposition of the sample or analyte. Another less often used and more difficult way to increase solvent diffusivity is to add a secondary chemical, such as a surfactant or a gas like carbon dioxide (perhaps in the supercritical region), which lowers the surface tension and enhances diffusion.

The example shown in the lower, red curve in Figure 1 is limited by the analyte solubility in the extracting solvent. Only approximately 20% of the analyte is extracting during the first cycle before the curve flattens, though it still rises at a visible rate. This extraction is solubility limited. Again, increasing temperature generally results in increased analyte solubility. A different solvent, or a solvent additive, is most likely needed.

Plotting the extraction kinetics during method development, especially when extracting solid samples, often helps guide the extraction process: Should I switch solvents, heat, or grind the sample? However, the analyst should keep in mind two common situations where the shape of the kinetic plot may differ from that shown in Figure 1. The first common situation is where the analyte may decompose (because of light, heat, or oxidation) during the course of the extraction. In the second common situation, the analyte may be tightly adsorbed onto the sample matrix or may even be encapsulated by the matrix. These situations may require more complex method development.

Question 3: Over the past 20–25 years, there have been serious concerns over the environmental consequences(for example, flammability, toxicity, and waste generation) of the use of organic solvents in chemical extractions. Discuss how modern extraction methods reduce or eliminate the use of organic solvents from the perspectives of (a) extracting liquid samples and (b) extracting solid samples.

Since the 1990s, the chemistry community has seen the rise of green chemistry, a thought process on the practice of our discipline considering the health and environmental consequences of our products and processes. At approximately the same time, there has been the concurrent development of the next generation of extraction techniques with improved performance. Techniques like solid-phase extraction (SPE), SPME, hollow-fibre membranes, SFE, PLE, and MAE are among the techniques that possess significant performance advantages compared to their traditional counterparts and happen to have green advantages as well. These approaches are preferred because methods cannot possess green advantages only; they must also be cheaper, faster, and more quantitative (equal performance will not cut it) than the methods they are replacing.

Extracting liquid samples has been greatly influenced by the use of adsorbents, such as chromatographic stationary phases. This led to the development of SPE, followed by SPME, stir-bar sorptive extraction (SBSE), immobilized liquid extraction (ILE), and others. The key feature of each of these techniques is that the dissolved (typically aqueous) sample is added to the sorbent phase, with a variety of configurations that define the techniques, and small amounts of organic solvent are used in the washing steps, the desorption steps, or both. The same organic solvents may be used in conventional extractions or different solvents may be used.

As these methods—particularly SPME and SBSE—were developed, thermal desorption of the analyte from the sorbent led to the complete elimination of organic solvent use. The various formats of these sorbent-based extractions became the basis for other techniques such as hollow-fibre membranes and support-assisted liquid–liquid extraction (SALLE), which also dramatically reduced organic solvent use. Single-drop microextraction (SDME) can be envisioned as a derivative of SPME where microlitres of immiscible organic solvent may be considered as the sorbent phase.

Meanwhile, the extraction of solid samples was advanced by the development of instrumental approaches that utilize energy to impact the extraction. Generally, the energy is manifested in the form of heat to improve solubility, diffusion, and desorption energy, as discussed in Question 2. These techniques include PLE, MAE, and ultrasound-assisted extraction (UAE). The advantage of these techniques is a reduction of solvent use from hundreds of millilitres in Soxhlet extraction, for example, to tens of millilitres. The techniques also happen to be faster (from up to a day to 15–30 min per extraction) and more reproducible.

In addition to the newer techniques described above, another approach applied to both liquid and solid samples is the use of alternative solvents with green advantages. Beginning in the mid-1980s, SFE saw increasing use. Supercritical carbon dioxide has solubility and selectivity advantages for certain compounds and favourable diffusion. The instrumental configuration of SFE also inspired the subsequent development of PLE and MAE. Natural products like D-limonene, α-pinene, ethyl lactate, deep eutectic solvents, and ionic liquids are all finding use in analytical extractions. Coacervative extractions, such as cloud-point extraction (CPE), use surfactants to encapsulate solutes in the form of miscelles to isolate and separate the solutes from the liquid solution.

Question 4: Choose one of the more “advanced” methods (that is, automated Soxlet, UAE, PLE, SFE, or MAE) for extracting solid samples. Compare this method with the traditional Soxhlet procedure in terms of method operation, time, solvent use, cost, safety, and other performance variables.

Table 1 summarizes the attributes of each of the subject extraction techniques. Each of the techniques is much quicker than Soxhlet extraction, which takes up to 6–24 h. SFE, PLE, and MAE use the least amount of solvent, even though automated Soxhlet and UAE are also reduced-solvent use techniques. The automation characteristics of SFE, PLE, and MAE result in less labour. The instrumental methods suffer from a high acquisition cost, but that is amortized to a lower operating cost per sample. Although each technique is safe if operated properly (Soxhlet extraction should be performed in a fume hood), the instrumental methods have built-in features to avoid exposure to solvent vapours.

of 0.616. A cup (237 mL) of coffee contains 95mg of caffeine. If you want to extract caffeine from a cup of coffee using LLE with a separatory funnel and ethyl acetate as the solvent, how many extractions and how much solvent will be used to extract 92 mg of caffeine, if 237 mL each of coffee and ethyl acetate are used in each extraction?

The octanol–water partition coefficient, or K

, can be used to estimate the amount of solvent used in a LLE. This value is tabulated for many common compounds of biochemical or environmental interest in several references. The K

 is the equilibrium constant showing the distribution of an analyte (A) between a nonpolar solvent (

-octanol approximates all nonpolar organics. One thing I try teaching students is to estimate an answer using what is known. With a K

 of approximately 0.5, one-third of the solute should partition into the organic solvent with each extraction, or seven steps should extract approximately 95% of the caffeine in this case.

 is the concentration of analyte remaining in the sample following 

 is the initial concentration of analyte in the sample, and 

 are the volumes of aqueous sample and organic extracting solvent, tells of the efficiency of an extraction. Using our ballpark estimate of seven extractions: [A]

 = (237 mL coffee)/([237 mL ethyl acetate × 0.616] + 237 mL coffee)

 × 95 mg caffeine = 3.3 mg caffeine remaining. This will result in a whopping 1659 mL of ethyl acetate to extract the cup (237 mL) of coffee.

Therefore, seven extractions will result in a cup of coffee with 91.7 mg caffeine remaining. From an analytical perspective, selection of a solvent with a greater K

 for caffeine will provide a more efficient extraction. From a coffee drinker perspective, start with decaffeinated beans!

 Both benzoic acid and ethyl benzoate (structures shown below) are soluble in organic solvents. How can you use extraction to separate them from a solution?

Knowledge of the acid-base properties of benzoic acid is key to solving this problem, similar to the amine example in Question 1. The p

 of benzoic acid is 4.20, meaning that it will be in the deprotonated, or anionic, form at neutral pH or at any pH greater than 1.5 pH units higher than the p

 or pH 5.60. Hence it should be water soluble. The ethyl ester is not impacted by pH and is essentially insoluble in water. Thus, a LLE using neutral water and a water‑immiscible organic solvent should achieve the desired separation. As seen in Question 5, selection of the organic solvent should be based on the K

I hope that this month’s column provides an understanding of miscellaneous extraction and sample preparation concepts. You are challenged to recall your knowledge, or learn it for the first time, using the format of exam questions. Because our course is taught in alternate years, I’m likely to teach the course no more than one more time before retirement. Therefore, other professors are welcome to reuse or modify my questions, but it is also likely that these questions will be available to your students on various internet sites. Future instalments of this column may present other exam questions I’ve developed over the years.

 is a Department Head and Associate Professor at South Dakota State University, USA. His research interests include green chemistry, alternative solvents, sample preparation, high-resolution chromatography, and bioprocessing in supercritical fluids. He earned his Ph.D. in 1990 at Brigham Young University under the direction of Milton L. Lee. Raynie is a member of 

’s editorial advisory board. Direct correspondence about this column via e-mail to amatheson@mjhlifesciences.com

As I write this, I’m also grading the final exam from my Chromatography and Separations course. Here’s your opportunity to test yourself; answers are provided.

Testing the Fundamentals – An Examination of Your Extraction Knowledge

In recent years, there has been an effort to categorize how emerging contaminants are identified in water samples by liquid chromatography–mass spectrometry (LC–MS). This process is often referred to as a workflow, and it has been the focus of many talks and conference sessions around the world. Many scientists have looked to unify and generalize the workflow, but do we really need a generic workflow to identify a non-target in an environmental sample? Would the scientific method approach (applied to each individual problem to solve) be enough and reliable?

One would not want to write about a topic without defining the key terms that are used frequently in the environmental community. When focusing on non-targeted screening for environmental contaminants in mass spectrometry (MS), we often use terms such as 

There is no doubt about what a target analyte constitutes. It is one that the user is looking for, for which a standard is available (usually in the laboratory), and the properties of which are already known, such as its chromatographic retention time, tandem mass spectrometry (MS/MS) spectra, and chemical structure.

The ambiguous terminology comes when one refers to 

. These compounds might be analytes that one is not looking for, but for which standards are available somewhere (they might even be in the laboratory freezer, but one has not made the standards or analyzed them!). They could also be compounds that one is not aware of (such as new pesticides coming onto the market or new pharmaceuticals being prescribed). Once a non-targeted compound is discovered, it becomes a target. Each individual user will have a different approach for this particular process. Mine is usually to analyze the new identified standard (if available), obtain the accurate mass, record the retention time, perform an MS/MS analysis, and understand the fragmentation of the analyte. It is then that it becomes a target.

However, my problem starts when someone talks about 

. What are those? Some may claim those are analytes that are in a database but for which the user does not have the standard. As a result, the suspect compound has never been analyzed through the methodology; therefore, nothing is known about the compound. But wait, aren’t we getting ahead of ourselves? Databases are part of a workflow scheme, and we have not gotten to it yet. Suspect is the worst ambiguous term to use. Do you really have to start the process of suspecting which compounds will be in your sample? How often do you do that? Again, it depends on your approach, what type of sample you have, and what the end goal is for it. For now, I will leave this term up in the air. We have yet to talk about 

 is next. This term gives me the most trouble. What really is an unknown? How do we definite it? It might be unknown to you, but not to me, or vice versa. Wouldn’t that be the same as non-target, a compound someone knows is out there in the environment, but for which the particular user does not own the standard and has never been analyzed using the method? Once the standard for that compound is analyzed, it becomes a 

 would be more fitting for a new natural product, such as one discovered in plants, which has never been reported before. It might also be some analyte for which the standard does not exist commercially yet, some new drug that has just been developed, or some pesticide that has just been manufactured and is ready to be released. The environmental community has used this term 

 to describe compounds that are not in databases, and compounds that the user had no idea were out there. To me, that is more of an educated guess. Once you have gathered enough information and done your research, you should know what that particular unknown (to you) is!

 is relative to the observer. One might be an experienced analyst and realize some of those non-targets, suspects, or unknowns are in fact targets and very well-known compounds that just happen to be in one’s memory. I myself have memorized hundreds of 

 masses for pesticides and pharmaceuticals and can quickly make an identification. That comes with experience, and everybody’s learning curve is dif- ferent. So, when defining terms, I think one should be aware of experience, knowledge, and the methodology used by the analyst. A beginner analyst might need another set of terms, and maybe a different approach.

This might be a simplistic view, but it has proven to work time after time. To me, there are no suspect compounds or even unknowns; a compound is either a target (you have the standard in your laboratory) or a non-target (you might not even know that the compound exists). Here, I give some examples and refer to some of the work our group has published successfully when referring to non-target analyses.

Instrumentation for Non-Targeted Analysis: High-Resolution
Mass Spectrometry (HRMS)

There is no doubt that the most commonly used and accepted methodology for non-targeted analysis is high-resolution accurate mass spectrometry. It has been approximately 25 years since the development of the first benchtop liquid chromatography (LC)–accurate mass MS instruments using time-of-flight (TOF) or high-resolution MS (HRMS). The instrumentation has evolved tremendously, making these instruments very sensitive, with high resolving power capable of distinguishing between very close isobaric compounds (which have the same nominal mass but different accurate masses). The environmental community has finally accepted these methodologies as unique and capable of identifying an analyte in a sample. Back in the early 2000s, there was still a debate in food and veterinary chemistry of what constituted a positive finding, depending on a point system if using low-resolution mass spectrometry or HRMS (1).

Moreover, there have been attempts to use tandem MS instruments for non-targeted analyses, and those are valid as long as one follows the golden rules of MS and identification. In fact, ion traps were used at the beginning to aid in the identification of non-targeted analytes because of their fragmentation powers and structural elucidation capabilities. Those instruments were often underestimated as to what could be identified in a sample following a few simple approaches.

This overview focuses on using accurate mass MS for the identification of non-targeted compounds.

 is often used to refer to the process that leads to an identification of an analyte in a sample. To me, the identification process is simple, maybe because I learned to do it before any workflow software was available. But new generations of analysts have lost sight of the “golden rules” of MS, and other processes that are intuitive and follow pattern recognition. But I would still ask, do we really need a workflow?

As an author, reviewer, and editor, I have become aware of the increase in papers relating to workflows for non-targeted analyses. There has been an effort to unify this process to make it easier for the analyst to identify as many compounds as possible in a given sample. One important aspect of a workflow is the software. Nowadays, every accurate mass MS instrument manufacturer has its own software package that aids the user in the identification process. However, some of these software programs are not easy to use, have tedious learning curves, and are quite expensive. On top of that, software does not guarantee that you will correctly identify everything that is in a sample. False positives and negatives are the most common drawbacks encountered. In the next subsections, I discuss what I think needs to be considered when identifying non-targets in samples.

Sometimes, the most important process in environmental analysis is forgotten: sample preparation. Every sample, every project, and every problem is different. Sample preparation is an important step in environmental analysis, and it will determine which compounds you identify or don’t identify at the end. For example, fire-related compounds that are organic acids need a pH adjustment before the sample is processed through solid-phase extraction (SPE). If that step is not done, then no identification of polycarboxylic acids will be possible (2). Some workflows described in the literature do not take into account the important step of sample preparation.

There are myriad instruments and software packages, so where does one start? How about not having a generic workflow? How about developing a unique workflow each time a sample arrives in your laboratory? After giving careful consideration to sample preparation, one has to decide the acquisition method one wants to use. That alone will also determine what we will be able to identify. There are quite a few different acquisition methods depending on instrumentation and manufacturer type. Mainly, there are two methods that I can generalize when using quadrupole HRMS: data-independent acquisition, and data-dependent acquisition. In my view, data-independent acquisition (using a high in-source voltage) has been proven very effective in identifying compounds and their metabolites as a first step (3). Of course, data-dependent acquisition (MS/MS triggered) is also a highly effective approach for unequivocal identification using MS/MS spectral mass, libraries, and chemical structural elucidation.

Depending on which workflow one follows, at some point there is a step that involves formula generation from a measured accurate mass. Again, it is the user (not the software!) that has to set the elements that are included in the formula generation. A visual inspection of the mass spectra data will reveal if a sulfur atom or halogen has to be included. For pesticide analysis, phosphorus and fluorine are also elements to consider. These are all user-dependent parameters; the analyst has to make the choice and guide the software, not the other way around. Thus, the analyst should be well trained in accurate mass MS analysis. In one study, for example, halogen signals proved to be a key to the identification of a new anti-depressant in water samples (4).

Then, there are the databases. There are hundreds of them, and accurate mass MS data for each individual compound and their fragments can be acquired easily nowadays. There are homemade and commercial databases. Again, the database that is used determines what you find at the end. There have been attempts to have accurate mass spectral libraries as well, but each instrument is different and will give different ions and abundances for each fragment ion, depending on collision energy. Therefore, one has to decide if that is the best approach for a given sample. The advantage of using generic accurate mass databases is that this parameter of accurate mass for the protonated or depronated molecule is universal; it will not depend on the instrument or acquisition method used (5).

Therefore, the approach one follows is highly dependent on the sample type, sample preparation, acquisition method used, software (or lack of) used to extract data, and finally, user interpretation. Can that be unified into one general workflow? I don’t think so. The analyst still has to do the final interpretation. The instrument will give you accurate data (assuming it is well calibrated and performing correctly), and the software will guide you and try to interpret, but only to a point. The final decision is based on the knowledge, data, and experience of the analyst. That is what will prevail at the end.

I mentioned above some of the golden rules in MS that have been forgotten or seldom considered. I will comment on a few of them here.

It might seem obvious, but accurate mass has to be calculated correctly. Not too long ago, there were still some software programs that did not take into account the mass of an electron when calculating accurate mass (6).

A second rule is the nitrogen rule, which is an easy one to take into account if we only consider even-electron ions for a formula generation. However, there are software programs that will not account for this (7).

Mass defect has proven to be very useful for the identification of homologue series (such as surfactants) by measuring differences in alkyl chain lengths (8).

Obvious halogen and sulfur isotopic signals (abundance and intensity) are also highly useful in the structural elucidation and identification of compounds.

There is the potential for adduct formation with sodium, ammonium, and potassium. The identification of sucralose is an example of considering the formation of a strong sodium adduct in positive ion mode (9).

Labile analytes fragment easily and do not show a protonated or deprotonated ion; instead, one of the fragments is observed in electrospray. Examples of this are diphenhydramine and bupropion.

Finally, we cannot forget fragmentation patterns and fragmentation trees, some of which have also been included in advanced software programs. Again, the user has to guide the software to make an unequivocal and ultimate identification.

HRMS instruments are capable of unequivocally identifying many targeted and non-targeted analytes. The approach to follow will always depend on the type of sample, the analyst’s knowledge, the end goal to achieve, and the tools available (software packages). Generic workflows are not necessarily needed because rapidly changing variables and parameters can be used, depending on the need of each individual problem and user. Trying to unify all these complex pathways in a single workflow or identification system will never work. In the end, it is the analyst who is in charge of the data obtained from the instrument. The instrument produces the data, and the software makes an interpretation, but the analyst has to guide the whole process for successful analysis.

(1) A.A.M. Stolker, E. Dijkman, W. Niesing, and E.A. Hogendoorn, 

(2) I. Ferrer, E.M. Thurman, J.A. Zweigenbaum, S.F. Murphy, J.P. Webster, and F.L. Rosario-Ortiz, 

(5) E.M. Thurman, I. Ferrer, O. Malato, and A.R. Fernández-Alba, 

(7) E.M. Thurman, I. Ferrer, O.J. Pozo, J.V. Sancho, and F. Hernandez, 

(9) I. Ferrer, J.A. Zweigenbaum, and E.M. Thurman, 

 is with the University of Colorado, in Boulder, Colorado. Direct correspondence to: 

Are generic workflows really needed in environmental analysis? In the end, it is the analyst— not instrumentation or software—that is in charge of what data are obtained, after thoughtfully considering what sample preparation and acquisition method to use.

What’s Your Workflow? Non-Targeted Analysis of Water Samples

Analytical requirements in the biopharmaceutical, pharmaceutical, and food industries, among several others, are more demanding than ever. Chromatographic techniques are excellent tools to acquire detailed information on a vast number of molecules and sample types. The present challenge in research and development (R&D) as well as in quality control (QC) laboratories is to collect as much sample information as possible. However, even with the current one-dimensional (1D) analytical portfolio, it is not possible to fully ensure that all the relevant information from a sample is captured. This article illustrates the power of an online two-dimensional liquid chromatographic (2D-LC) setup to unravel the complexity of biopharmaceutical and pharmaceutical samples. This technology tremendously increases the resolving power in all areas where LC is applied and drastically reduces the risk of missing an unknown in the samples.

Chromatographic instrumentation and column technology are continuously evolving and the state-of-the-art equipment used regularly today is quite impressive. Ultrahigh-pressure liquid chromatography (UHPLC) has enabled laboratories to boost the chromatographic resolution, and hence the number of compounds that can be separated in an analytical run. Even with high-end systems and well-developed one-dimensional LC (1D-LC) methods, the probability of missing critical information on the sample composition, for example, because of chromatographically overlapping peaks, is very real. Figuring out ways to minimize this risk must be continuously explored.

The hyphenation of chromatography with mass spectrometry (MS) generally improves analytical sensitivity and selectivity significantly. Using MS can help the user to detect compounds that otherwise would be overlooked as a result of a lack in signal intensity or interference from other compounds present in the sample. LC–MS techniques are a big leap forwards, but even these instruments are not always able to provide the answers, and LC–MS methods have more restrictions in terms of the chemicals (such as the mobile‑phase buffers and additives) that can be used to execute the analyses compared to other more traditional detectors. Furthermore, high-end MS instrumentation is expensive to acquire and operate, requiring substantial user expertise.

All of the above leads to the conclusion that even with the high performance of UHPLC and MS equipment, the probability of missing critical information about the sample persists. Missing relevant compounds in the sample and not meeting the given analytical or regulatory requirements can have serious economic consequences (delays), damage the reputation of the department or company, and above all, affect safety and efficacy. The latter is obviously the reason for the considerable regulatory burden associated with drug development.

An emerging tool to substantially reduce the above risk is two-dimensional LC (2D-LC). In 2D-LC, two different chromatographic separation mechanisms are combined and peaks that are eluted from the first column are further separated on a second column that ideally has an orthogonal separation behaviour. As a result, separation power can to a great extent be boosted, complementary information can readily be obtained (such as information related to molecular properties like size, charge, hydrophobicity, and affinity), and separations can be rendered MS-compatible, allowing the analyst to obtain a detailed understanding of the sample under investigation. Peaks can be transferred from one dimension to the other, either in an online or an offline manner. Historically, offline transfer has been the method of choice, but with robust, state-of-the-art, online 2D-LC instrumentation and the elegant software solutions made commercially available in recent years, this technology has found its way to mainstream laboratories. Such advanced systems offer the further benefit of automation and reduced sample manipulation, thus preventing solute loss and degradation. In addition, automation inherently adds more reliable tracking of the entire workflow, in contrast to manual or external tracking applied in offline approaches, thereby improving traceability in regulated environments.

This short contribution discusses the potential pitfalls associated with 1D‑LC, and the 2D-LC approaches that can address those shortcomings.

This discussion is briefly illustrated with some relevant applications related to pharmaceutical and biopharmaceutical analysis (1–17). However, the reader should take note that the same principles apply in other disciplines such as food, environmental, toxicological, chemical, and petrochemical analysis, to name a few.

Coelution Resulting from Lack of Selectivity

One of the most recurring analytical concerns in the pharmaceutical industry relates to the lack of selectivity of methods, causing peaks to be coeluted, or a failure to prove that the results demonstrate adequate selectivity. Hence, it is crucial to incorporate as much selectivity as possible into any chromatographic method. This goal of ensuring sufficient selectivity is often supported by using forced degradation or even process samples during method development. However, it is impossible to ensure full selectivity for all potential and real impurities and degradants in the drug substance or drug product using a single LC method. Therefore, evaluating other methods with alternative selectivity is a common and good practice to assess peak coelution and prove peak purity.

One obvious example is the investigation of chirality for a given active pharmaceutical ingredient (API) and its impurities separated in a purity LC method, typically in reversed-phase LC mode. Because the purity method is achiral, it would not supply information about the presence of enantiomeric impurities in a 1D-LC setup. The sample or collected fractions would need to be analyzed with another method and alternative (chiral) stationary phases. By using an online 2D-LC instrument that combines a reversed-phase LC and a chiral LC column, data on both the achiral and chiral impurities can be obtained from a single injection in a fully automated manner.

Another typical challenge is related to being able to demonstrate the absence or presence of impurities due to degradation or the synthesis reactions that might be coeluted with the main peak. The main peak is often overloaded and too broad to reach the 0.05% reporting limit and can mask the presence of low-abundance compounds. Selective detectors, such as MS, can sometimes reveal these impurities based on differential 

; however, the high API and matrix content loaded onto the LC–MS system can suppress the ionization and detection of coeluted impurities. In addition, coeluted isomers would often not be differentiated. Therefore, it is advised to investigate such samples with chromatographic methods with alternative selectivity. An example of reversed-phase LC analysis of a small‑molecule API (metoclopramide) is shown in Figure 1. The large peak of the API observed in the UV chromatogram hides any impurities that may be eluted in this region. MS data on this run only show the presence of the API. On the other hand, data from comprehensive 2D-LC (LC×LC), in which the entire first dimension is sent in discrete fractions to the second dimension with alternative selectivity, reveal the presence of an unknown impurity hidden underneath the main peak.

It is crucial to detect these impurities and degradation products early in the method development process to streamline final method development. It is because of this that multiple 1D-LC methods are frequently developed and applied—to cover as many impurities as possible. However, such an approach is time and resource consuming, and peak tracking is not an easy task. Bundling alternative selectivities in a single 2D-LC method may be the better and more productive approach.

A multitude of examples can be found in the literature that apply to small molecules, oligonucleotides, peptides, proteins, monoclonal antibodies (mAbs), antibody–drug conjugates (ADCs), and more, illustrating the lack of selectivity of 1D-LC and how 2D-LC sheds light on peak and drug purity and drug quality.

Insufficient Resolving Power for a Given Sample Complexity

The above discussion focuses on potential chromatographic issues encountered in relatively simple yet challenging and difficult-to-resolve mixtures. The biopharmaceutical industry is evidently also confronted with many samples of high complexity such as natural products, fermentation samples, reaction mixtures, digests, antibodies, mRNA, and so on.

It is obvious that when sample complexity increases, the risk of losing or not detecting valuable information rises.

Peptide mapping is an appreciated methodology for the in-depth characterization of therapeutic proteins, such as mAbs and ADCs. The analysis is part of an extensive portfolio of chromatographic methods that, on their own, already demonstrate the need for complementary selectivities. Interestingly, there is a tendency to combine several of these methods in multidimensional and parallel setups, a principle known as 

 (12). Peptide mapping demands the best in terms of chromatographic separation because the complexity of the sample is substantially increased following the generation of peptides. In digests of larger therapeutic proteins such as mAbs, it is not uncommon to encounter hundreds of peptides with different physicochemical properties that are present in a wide concentration range. Despite the feasibility of generating peak capacities up to 1000 in UHPLC, the random distribution of peaks requires peak capacities in excess of 10,000 to resolve 98% of a tryptic digest containing 100 peptides. As a result, peptide mapping using 1D-LC does not provide the complete picture. In contrast, 2D-LC offers a substantially higher peak capacity as long as the two dimensions are orthogonal and separation obtained in the first dimension is maintained upon transfer to the second dimension (13).

Regulations do not only require full characterization of the API; guidelines concerning the evaluation of excipients are also becoming more demanding. Some excipients, such as surfactants, are highly heterogeneous and their analysis as a raw material or in a formulation represents an additional challenge.

In addition, composition can change over time and stability needs to be evaluated with the best chromatographic performance. Polysorbates (Tween), for example, are commonly used surfactants in pharmaceutical and biopharmaceutical formulations and exhibit high complexity resulting from the synthesis process and the inherent distribution already present in the raw materials used. If the distribution within each group were relatively narrow (degree of ethoxylation, fatty acid), a well‑developed reversed-phase LC method could potentially suffice to profile the polysorbate sample. Generally, however, natural mixtures of fatty acids are used in synthesis and the ethoxylation degree is relatively broad, making it very difficult to clearly separate all groups from each other. MS can be used to increase selectivity to a certain extent, but care should be taken for components with the same elemental composition and molecular weight. Consequently, 1D-LC will not be able to resolve all compounds within those samples. Nevertheless, impressive profiles and complexities have been revealed using 2D-LC (Figure 2) (14).

MS is nowadays an indispensable tool in analytical laboratories, both in research and development (R&D) and quality control (QC) environments. However, the direct hyphenation of LC methods to MS is not always an option because of mobile-phase incompatibility. For legacy and compendial LC methods, the main reason is the use of involatile mobile‑phase additives such as phosphate and ion‑pairing reagents that are incompatible with MS. In addition, chromatographic modes, such as ion exchange (IEC), hydrophobic interaction (HIC), and size‑exclusion chromatography (SEC), are used to highlight charge, hydrophobic, and size variants of protein biopharmaceuticals, and are ideally run with inorganic or nonvolatile mobile phases, which are also incompatible with MS.

In the absence of MS data, the probability of missing crucial sample information is real. A solution could be to perform an LC–MS experiment using similar yet MS-compatible conditions, by, for example, replacing phosphoric acid with formic acid. In certain cases, success will lurk around the corner. However, more frequently, selectivity would have changed too drastically, and, in a worst-case scenario, separation would be completely destroyed. Another approach to retrieve an MS readout from an MS-incompatible run is to collect the peaks of interest offline using a fraction collector (or manually in tubes) and desalt or concentrate them before MS analysis. Desalting or concentration is ideally carried out during the re-injection of the collected fraction onto an MS-compatible LC method, typically reversed-phase LC-based, in which salts are eluted before the compounds of interest.

The above process can be fully automated using (multiple) heart-cutting 2D-LC in which one or a couple of first‑dimension peaks or fractions are collected in loops installed on a valve followed by sequential transfer to a second chromatographic dimension that takes care of the desalting, concentration, or both. The second dimension optionally adds further separating power or selectivity, which is further beneficial to discriminate in-source fragments from real signals. The existing method, and therefore the previously acquired chromatographic profile, remains unchanged and peaks or regions of interest observed in the first-dimension profile can be identified. Additional advantages of online 2D‑LC–MS are full traceability and the limited sample manipulation, which prevents sample loss and degradation. It is not uncommon for low-abundance peaks to remain undetectable when they are collected in an offline manner and yet to be fully retrieved using the online 2D-LC approach.

Some of the common applications of this approach are MS characterization of protein variants observed in the SEC, IEC, or HIC chromatograms of mAbs or the identification of unknown or new impurities detected with legacy and validated methods that are not compatible with MS. Figure 3 shows the analysis of the peptide glucagon using an MS-incompatible mobile phase in accordance with the monograph in 

This article explores potential pitfalls associated with 1D‑LC and how 2D‑LC can overcome these obstacles.