Exposing Contamination: Advances in Food and Environment Analysis

Gas Chromatography Meets Non-Targeted Screening

Bioanalysis of Small-Molecule Drugs and Metabolites in Physiological Samples by LC–MS, Part 1

Where Has My Efficiency Gone? 

Facilitating Gas Chromatography Separation of Complex Mixtures

The Role of Microextraction Techniques 

Enhancing Performance of Analytical Methods

Ion Mobility–Mass Spectrometry (IM–MS)

Interest in three-dimensional (3D) printing technology is expanding rapidly. What impact can it have in mass spectrometry? Robert Winkler, of the Center for Research and Advanced Studies Irapuato, in Guanajuato, Mexico, is exploring this question. We recently spoke with Prof. Winkler about this work.

What makes 3D-printing technology valuable in ion mobility spectrometry and mass spectrometry?

Before we started using 3D printing, we spent quite a lot of time building our prototypes. On a daily basis, new exciting ideas for improving analytical instruments are published, and we would like to test some of them. For example, ambient ionization, mass spectrometry imaging, ion manipulation, and portability are very active research areas in ion mobility spectrometry and mass spectrometry. But often, the reported devices are difficult to reproduce and are commercially not available. Using 3D printing, we can quickly build, modify, test, and share prototypes. This approach also reduces our dependence on large industrial providers, speeds up innovation, and lowers costs.

In a recent article (1), you highlight many recent developments in 3D-printing, including fused deposition modeling (FDM). How can FDM help advance ion mobility spectrometry?

Fused deposition modeling (FDM) is a widespread technology, and simple FDM printers are on sale for a few hundred US dollars. Thus, this technology is affordable for higher education and research institutions “on a budget.” Although the electronics setup is not trivial, cheap ion mobility spectrometry (IMS) devices could be built and distributed as do-it-yourself kits using FDM. IMS has less analytical power than mass spectrometry, but it is very sensitive for detecting known compounds. Thus, IMS is commonly used for chemical screening in airports.

Having IMS devices that are easy to build with FDM could support the further development of the technology and its applications, such as environmental monitoring. In the review, we presented different IMS devices and components that were built using FDM. The production of ten unibody drift tubes with a high global accuracy showed the feasibility of 3D-printed IMS analyzers (2).

Three-dimensional printing has been shown to allow for the fast and cost-effective production of microreactors for bioassays. What effect has this development had?

The development of bioassays can be tedious because multiple conditions must be tested. Producing cartridges or microreactor systems with conventional methods is only feasible for extensive studies or routine use. With 3D-printing, small series of microreactors can be manufactured and enable reproducible experiments. Later, the design of the 3D-printed gadgets can be adopted for mass production.

In your article, you discuss how 3D printing enables the "peer production" of scientific devices. Can you expand on what "peer production" means and how that connects to the results 3D-printing has achieved so far in ion mobility and mass spectrometry?

“Peer production” is the community-driven development and manufacture of goods. Its participants have diverse motivations—not only monetary—for sharing their knowledge. The concept of peer production applies mainly to non-rival goods such as information (3). Wikipedia and Linux are examples of successful peer production, with thousands of volunteer contributors.

We develop analytical devices with public funding and are employed in public institutions. Thus, we should make our results accessible to the people and companies that support our work. If another research group or company adopts our designs, my salary is not affected (“non-rival”), and maybe the users cite us, which helps us justify new project funding. If the situation is the other way around, we also profit from using published files of 3D-printable devices and related literature.

Several groups share their 3D-printing files for IMS and MS parts, either as supplemental material to their papers or on personal request. For example, the files for building or modifying our 3D low-temperature plasma (LTP) probe are freely available 

 (4). Anyway, providing the necessary files for reproducing 3D-printed prototypes is good practice for publications in analytical chemistry and should be further encouraged by the journals.

What are the major challenges that remain with developing 3D-printing technology for use in mass spectrometry? Have some of these challenges been resolved?

Additive manufacturing has various limitations. Some shapes and geometries, for example, overhangs, are difficult to print. Also, the surface quality and spatial resolution of parts may be insufficient for specific applications. Also, not all materials are available for 3D printing. However, there's already a wide variety of polymer filaments with diverse properties, such as flexible or conductive.

Three-dimensional printing is a powerful production method, but we have to design printable parts. In some cases, a professional service can print the final component. But often, the combination of 3D printing with conventional manufacturing methods, and using off-the-shelf parts where possible, leads [subject of the verb is 

There are also innovative solutions to overcome particular problems, such as the metal coating of 3D-printed parts for making them conductive (5).

What are your next steps in researching 3D-printing technology, and in what direction do you see the application of 3D-printing heading in the future?

Combining different methods with a low analytical resolution often gives more information for identifying and quantifying compounds than using a single high-resolution method. Thus, we develop functional modules that fit together, resulting in analytical systems for given applications. For example, our 3D-printed ion source and our Open LabBot (built with 3D-printer technology) enable mass spectrometry imaging and high-throughput

 (6). Now, we are working on a modular miniature mass spectrometer for the real-time analysis of volatile organic compounds (VOCs) and a platform for metabolic phenotyping. Three-dimensional printing plays a central role in the development of our prototypes. Besides, we are creating software to process such multi-dimensional data.

The possibility of drawing objects in free software and producing them with good quality at low cost motivates students and scientists to bring their ideas to practice. In this way, analytical chemists are not only consumers anymore but also empowered to create their own devices and share them with the community.

(1) H. Guillen-Alonso, I. Rosas-Roman, and R. Winkler, 

(2) B.C. Hauck, B.R. Ruprecht, P.C. Riley, and L.D. Strauch, 

The Wealth of Networks: How Social Production Transforms Markets and Freedom 

(New Haven and London, Yale University Press, 2006).

(4) S. Martinez-Jarquin, A. Moreno-Pedraza, H. Guillen-Alonso, and R. Winkler, 

(6) I. Rosas-Roman, C. Ovando-Vazquez, A. Moreno-Pedraza, H. Guillen-Alonso, and R. Winkler, 

 is the PI of the Laboratory for Biochemical and Instrumental Analysis in the Department of Biochemistry and Biotechnology at the Center for Research and Advanced Studies (CINVESTAV) Irapuato, in Guanajuato, Mexico. Direct correspondence to: 

Articles

Three-Dimensional Printing in Ion Mobility and Mass Spectrometry

Since the detection of N-nitrosodimethylamine (NDMA) in a batch of valsartan in 2018, at levels exceeding ICH acceptable intake limits for mutagenic impurities, the analysis of nitrosamines has become an intense focus point for the pharmaceutical industry. The identification and low-level determination of nitrosamines in potentially affected materials is challenging and requires the application of highly sensitive analytical techniques. This article reviews the chronological development of the story and the regulatory landscape that has evolved. It will then discuss the development of analytical methods for the determination of a series of nitrosamines referenced by regulatory authorities, demonstrating separation of these compounds from the active pharmaceutical ingredient (API) and looking at how mass spectrometry (MS) can be applied to ensure that the required detection limits can be reached.

In the summer of 2018, the pharmaceutical landscape for the manufacture of small molecules changed with the discovery of a small mutagenic compound in a batch of pharmaceuticals manufactured in China. Zhejiang Huahai Pharmaceuticals produce valsartan, which is a prescription-only drug used to treat high blood pressure and heart failure. It is a selective angiotensin II receptor blocker (ARB) that dilates blood vessels and so consequently reduces blood pressure. It is often given to patients directly after a heart attack. In performing the routine analysis, quality control (QC) chemists identified the presence of N-nitrosodimethylamine (NDMA) and went on to report an average level of 66.5 parts per million in affected batches (1), which is high enough to have a detrimental impact on patient safety. On 5 July 2018, the European Medical Agency gave notice on their website that it was reviewing medicines containing valsartan from Zhejiang Huahai following detection of an impurity. It was also noted that national authorities across the EU had started to recall medicines containing valsartan from Zhejiang Huahai. Since 2018, NDMA and other forms of nitrosamines have been observed in a range of different pharmaceuticals. This article will look at the chronological development of the story and how nitrosamines have caused the pharmaceutical industry to re-evaluate their manufacturing procedures. It will look at the synthetic pathways that can cause the generation of a range of nitrosamines and the regulatory landscape that has evolved as a consequence of the initial findings. The article will then discuss the development of a series of applications that will allow for the determination of a series of nitrosamines referenced by regulatory authorities, initially showing how to perform a separation of these compounds from the active pharmaceutical ingredient (API) and then looking at how mass spectrometry (MS) can be applied to the analysis to ensure that the required detection limits can be reached.

The role of the quality control (QC) department is to check the quality of a final product and, in the case of the pharmaceutical industry, there are two key aspects that the QC department will monitor:

The amount of active ingredient to ensure that the drug is efficacious,

The amount and type of impurities to ensure that the drug is safe to use.

These analyses can be subdivided into a range of activities, with separation science being a fundamental component of the process.

Impurities that are separated from the major component above a certain level as defined by the regulatory bodies must be identified and then quantified in subsequent analyses of the final product. Figure 1 outlines the decision tree that is used by pharmaceutical organizations
to determine if further investigation is required to determine the impact (2).


This decision process is associated with a new drug application (NDA), although for the testing of a generic drug, where the process is defined as an abbreviated new drug application (ANDA), the process would be different in accordance with Section 505(j) of the Federal Food, Drug, and Cosmetic Act (FD&C Act) (2), in which case the applicant is looking to demonstrate the drug is equivalent to a reference listed drug (RLD).

The regulations stipulate the detection levels that are required for various types of impurities that are identified through the drug development process. For example, for drug substances with a maximum daily dose ≤ 2 g/day, any impurity detected that is nominally above 0.1% requires structural identification. Any impurities exceeding the 0.15% qualification threshold require further assessment to determine biological safety (3). The next course of action will be very dependent on the outcome of this and will determine if the drug product is suitable for release.

In the summer of 2018, it was separation science (with mass spectrometry [MS] identification and quantification) that isolated a small mutagenic compound, a nitrosamine, in a batch of valsartan. Mutagens are chemicals that damage the genetic information within a cell, resulting in mutations that may lead to cancer. The damage to the cell is caused by interactions with the DNA sequence and the DNA structure. The alteration to the DNA may also result in permanent heritable changes to the somatic cells of the organism or germ cells that can then be passed on to future generations. It is clearly important that mutagenic compounds can be detected and their production avoided wherever possible. The bacterial reverse mutation assay, also referred to as the 

 (4), is often used to test for gene mutation and it has been used to identify a range of chemical entities that have been reported to be mutagenic (5), including nitrosamines. The World Health Organization has used the principles outlined in the ICH M7(R1) (6) guideline to determine an acceptable daily intake. This information is used to determine the allowable daily intake of mutagenic impurities for a specific medication, as will be detailed later in this article.

Nitrosamines are a class of compounds that have a chemical structure containing a nitroso group bonded to an amine, as shown in Figure 2. The toxic properties of nitrosamines were first reported by Barnes and Magee, who reported that N-nitrosodimethylamine (NDMA) produced liver tumours in rats. Subsequent studies, which involved the testing of over 300 nitrosamines, showed that nearly 90% of these compounds were carcinogenic to a wide variety of animals (7). They have been identified in a variety of sources, including environmental sources, drinking water, and processed food products. The potential for formation in pharmaceutical drug products, such as aminophenazone, in the presence of nitrosating agents was established in the 1970s. In the literature, several other active pharmaceutical ingredients (APIs) have also been reported to contain NDMA, including aminopyrimidine, amitriptyline, chloramphenicol, oxytetracycline, promazine, propoxyphene, chlorpromazine, diphenhydramine, doxylamine, trimipramine, tetracycline, erythromycin, imipramine, and methapyrilene (8,9). In 2018, the detection of high levels of NDMA in batches of valsartan drug product prompted renewed focus on this class of impurity.

Nitrosamines can be formed by the reaction of a secondary amine with nitrite (Figure 3). Therefore, it is suspected that NDMA in valsartan may come from the ZnCl2-catalyzed disproportionation of dimethylformamide (DMF) to dimethylamine and carbon monoxide (CO). Dimethylamine then goes on to react with sodium nitrite and the resulting product is NDMA. It has also been proposed that carryover of nitrites or amines from subsequent steps may also result in the formation. Notably, contamination from external sources has been highlighted as a potential source of nitrosamines (1), such as the use of recycled solvents, for example DMF—which is quenched with sodium nitrite to destroy residual azide as part of the recovery process—is thought to be a potentially significant source of NDMA. Recycling of materials, catalysts, and solvents is deemed to be a major factor in the formation of nitrosamines, along with impurities that could react further downstream if equipment is not adequately cleaned. Similar reaction schemes have been proposed for the formation of other nitrosamines that have been identified during impurity profiling. An additional theoretical source of nitrosamines that should be considered is the formation of higher molecular weight API-like nitrosamines within a drug product. In the case of valsartan, two valsartan-specific nitrosamines were confirmed by a single API manufacturer to the European Medicines Agency (EMA), albeit at acceptably low levels (8).

Regulatory authorities first became aware of the presence of nitrosamines, specifically NDMA, in pharmaceutical product in July 2018. It was first detected in an angiotensin II receptor blocker, valsartan, and since then other nitrosamines have been detected in other sartan drug products that contain a tetrazole ring, along with other API/medicinal products, including ranitidine and pioglitazone. The FDA and EMA have highlighted several nitrosamines that could be generated during the production process as potentially existing within drug products. These are highlighted in Table 1, with the required acceptable intake limits as specified by the regulatory authorities (8,10,11).

The nitrosamines specifically highlighted as of being of concern are: N-nitrosodimethylamine (NDMA), N-nitrosodiethylamine (NDEA), N-nitrosoethylisopropylamine (NEIPA), N-nitroso-diisopropylamine (NDIPA), N -nitroso-N-methyl-4-aminobutyric acid (NMBA), 1-nitroso-4-methyl piperazine (MeNP), N-Nitrosodibutylamine (NDBA), and N-nitrosomethylphenylamine (NMPA). It should be noted though that both regulatory authorities do stipulate that any carcinogenic impurity should be assessed in accordance with the ICH M7(R1) guidelines, and indeed the EMA documentation refers to 16 different nitrosamines that are routinely monitored under the United States Environmental Protection Agency (EPA) guidelines (8,12). Separation and detection of these impurities within solvents, intermediates, drug substance, and drug product has therefore become of high importance within the pharmaceutical industry.

The regulatory landscape has evolved very quickly since the first observation of NDMA in valsartan, with the FDA releasing documentation related to controlling nitrosamine impurities in human drugs in September 2020, which was recently updated in February 2021 (10). There has also been legislation released by the EMA that defines the allowable daily intakes. This information can be used to determine the allowable concentrations in a given drug product (8). Current acceptable intake limits required by the pharmaceutical industry are given in Table 2. The acceptable daily intake of nitrosamines is related to the maximum daily dose (MDD) of the drug substance, therefore different concentrations of nitrosamines are acceptable for different drugs.

These limits are applicable only if a drug product contains a single nitrosamine. If more than one of the nitrosamine impurities identified in Table 1 is detected and the total quantity of nitrosamine impurities exceeds 26.5 ng/day (the acceptable intake [AI] for the most potent nitrosamines) based on the MDD (10), the manufacturer should contact the agency for evaluation. For drug products with an MDD of less than 880 mg/day, a recommended limit for total nitrosamines of 0.03 ppm is no more than 26.5 ng/day and is considered acceptable. For drug products with an MDD above 880 mg/day, the limit for total nitrosamines should be adjusted so as not to exceed the recommended limit of 26.5 ng/day. The wording here is taken directly from the February 2021 guidance to industry document on “Control of Nitrosamine Impurities in Human Drugs” (10). In Europe, the total risk level for all N-nitrosamines detected must not exceed 1 in 100,000 lifetime risk. An alternative approach where the sum of all detected N-nitrosamines is below the limit of the most potent nitrosamine may also be used (8,11). This clearly demonstrates the complexity of developing an analytical method for the analysis of nitrosamines in an API, since there is a degree of interpretation required to define the assay dynamic range and this varies depending on the nature of the nitrosamine impurities being analyzed.

It is very evident from the above discussion that since the initial identification of NDMA in valsartan that there has been a lot of activity both in terms of identification of nitrosamine impurities and in terms of the required legislation to address the potential toxicity implications. Figure 4 gives an overview of the main activities associated with this unfolding story from the initial discovery in June 2018, through the triggering of article 31, which is when the EMA initiated non-pharmacovigilance referral procedures, to the more recent guidance updates by the FDA on the analysis of nitrosamines and the more recent observations of nitrosamines in other APIs.

Liquid chromatography (LC)–UV experimental work was performed on an Agilent 1290 binary UHPLC system, with data acquisition and analysis performed using Agilent OpenLab CDS and ChemStation software (Agilent Technologies). The LC–MS application was run on a Sciex QTRAP 6500+ LC–MS/MS system equipped with an ExionLC AD UHPLC system (Sciex). The MS was operated in positive ionization mode using the atmospheric pressure chemical ionization (APCI) probe on the IonDrive Turbo V source. Data acquisition was managed by Sciex Analyst software, whilst data was processed using Sciex OS. The LC separations were performed on Avantor ACE UltraCore C18 superficially porous columns in 100 × 3.0 mm and 100 × 2.1 mm dimensions and a 3.5‑µm particle size (VWR). Solvents and reagents were purchased from VWR. NDMA, NDEA, NPIDA, NDPA, NDBA, NMPA, NDMA-d6, and valsartan standards were purchased from VWR, NMBA and NEIPA from Enamine Ltd, and NMBA-d3, NDEA-d10, and NDBA-d18 from LGC Ltd.

For the LC–MS application, a standard solution containing 200 ng/mL of NDMA, NMBA, NEIPA, NDIPA, NDPA, NMPA, and NDBA was prepared, along with a 132 ng/mL solution of NDEA. The internal standard solution contained 10 µg/mL NDMA-d6 and NMBA-d3 and 1 µg/mL NDEA-d10 and NDBA-d18. L1–L7 calibration solutions were prepared as outlined in the proposed 

 method (13). The spiked valsartan sample was prepared by spiking 80 mg of drug substance with 6 µL and 9 µL of the respective standard solutions and 12 µL of internal standard solution, followed by 1173 µL 1% formic acid (aq.). The sample was vortexed at 2500 rpm for 20 min and then centrifuged at 6000 rpm for 20 min. It was then filtered using a Whatman Mini‑UniPrep 0.45 µm PVDF syringeless filter.

The low-level detection of nitrosamines including NDMA is highly pertinent due to potential trace-level formation in the manufacture and processing of various consumer products, including drinking water, food and beverages, tobacco‑related products, rubber products, pharmaceuticals, and cosmetics (8,14,15,16). The potential for the formation of nitrosamines in such products, for example, the reaction of nitrite with nitrosatable amines in meat and fish at high temperatures, has caused major concern. The chromatographic separation and quantitation of nitrosamines is therefore well established in a variety of matrices using both gas chromatography (GC) or LC. GC is ideally suited to the analysis of more volatile nitrosamines and has been utilized extensively. Several detection options have been explored for GC analysis of nitrosamines. Nitrogen chemiluminescence detection (CLD), also termed 

), has been used to provide a highly sensitive detection method for N-nitrosamines (9,17); however, potential for cross reactivity exists from organic and inorganic nitrites, nitrates, and N-nitramines (8). The use of MS detection has become increasingly established, providing highly sensitive, analyte-specific detection capabilities.

Liquid chromatography is also commonly used and can overcome some reported issues found with GC analysis, for example, allowing determination of thermally unstable nitrosamines (14) and involatile nitrosamines such as NMBA, and avoiding potential thermal degradation of ranitidine drug substances under GC conditions to yield NDMA (18). LC also enables the simultaneous determination of nitrosamines alongside drug product in a single run. The separation of nitrosamines can be achieved by reversed-phase LC, although given the hydrophilic, polar nature of some nitrosamines, retention of some of the compounds can be challenging. Various stationary phases have been utilized to obtain satisfactory retention, with C18 or pentafluorophenyl (PFP) phases often specified. The proposed 

 general chapter on nitrosamine impurities, currently under review (April 2021), provides monograph methods for the use of both these LC phases (13). The FDA additionally specifies the use of a C18 phase with an integrated phenyl ring for the determination of NDMA in ranitidine (19,20).

To demonstrate the applicability of LC for the separation of nitrosamine impurities in pharmaceutical preparations, a set of eight nitrosamines were spiked into a valsartan secondary standard and separated on a 100 × 3.0 mm solid-core C18 column (Figure 5). The set of nitrosamine impurities selected for this example include the majority of the nitrosamines identified as being of concern by the 

, EMA, and FDA (8,10,13). The sample analytes possess log P values ranging from -0.17 to 2.60, therefore gradient elution was required to provide suitable retention of all analytes. A simple gradient ranging from 4 to 95% B (2.8% to 66.5% [

] acetonitrile) was found to achieve baseline separation of all sample components. An initial 1 min isocratic hold was employed to aid retention of NDMA. It is worth noting that the peaks for asymmetric NMBA and NEIPA are observed as doublet peaks of the syn- and anti-conformers due to hindered rotation around the N-N bond (21). From assessment of the analyte physicochemical properties, mobile phases buffered with potassium phosphate at pH 2.7 and 6.5 were assessed during method development. It was found that an acidic pH was required to improve the retention and peak shape of the valsartan API peak and to also achieve retention of NMBA, with the NMBA peak eluting on the column void time at intermediate pH. In this example, UV detection was employed at a wavelength of 254 nm.

Whilst the example shown in Figure 5 clearly demonstrates the applicability of a C18 phase for the separation of these nitrosamine impurities, the low detection limits specified by regulatory authorities typically precludes the use of UV detection. Nitrosamine analysis therefore requires the use of highly sensitive and specific detection modes, such as MS, although the low molecular weight of nitrosamines and their relatively low hydrophobicity can make developing a sensitive and robust method somewhat challenging. The use of LC–MS/MS has become increasingly common, with nitrosamine detection, quantitation, and confirmation achieved through monitoring multiple analyte specific MRM transitions. Alternatively, LC–high resolution MS (LC–HRMS) has also been employed to provide highly specific detection and quantitation.

As discussed above, an acidic mobile phase provides good retention and selectivity for nitrosamines, although the nonvolatile phosphate buffer used in Figure 5 should be replaced with an MS-compatible alternative. Formic acid is a convenient additive that is widely used in the mobile phase for nitrosamine analysis by LC–MS. Ngongang et al. reported an increase in MS signal intensity for five out of nine nitrosamines tested over non‑buffered mobile phase, when 0.05 or 0.1% (

) formic acid was added to the mobile phase, with 0.1% proving optimal for lower response nitrosamines (14). Higher concentrations of formic acid were found to compromise signal intensity. Electrospray ionization (ESI) has been used for LC–MS analysis of nitrosamines, however, it has been reported that ESI may be impacted by ion suppression due to matrix effects. Therefore, positive mode atmospheric pressure chemical ionization (APCI) is preferential for nitrosamine analysis, providing much improved sensitivity (8,22).

To demonstrate the use of the same stationary phase in Figure 5 for the high sensitivity LC–MS/MS quantitation of nitrosamines in drug substances, an LC–MS/MS method for the determination of six nitrosamines (NDMA, NDEA, NMBA, NEIPA, NDIPA, NDBA) in selected sartans—specified in the proposed 

 general chapter on the analysis of nitrosamine impurities—was used as a starting point (13). The monograph method specified the use of a 150 × 3.0 mm column packed with 2.7-µm particles bonded with a C18 phase. A 12.2 min gradient profile was used, which would give an approximate cycle time of 18–24 min (assuming 5–10 column volumes of mobile phase for post‑gradient re-equilibration). Given the high resolution obtained using the LC method in Figure 5, it was decided to develop an alternative gradient method targeting a >50% reduction in overall cycle time. In addition to the six nitrosamines specified in the 

 method, NMPA (identified as a nitrosamine of concern by EMA and FDA) and NDPA (isobaric to NDIPA) were also included. The method should provide full chromatographic separation of all analytes, in particular the isobaric pair NDPA and NDIPA. Resolution of the syn- and anti‑conformers for NMBA and NEIPA was not deemed to be necessary, as these would be integrated as a single peak. As per the 

 methodology, four deuterated internal standards were included for quantitation: NDMA-d6, NMBA-d3, NDEA-d10, and NDBA-d18 (NEIPA, NDIPA, NDPA, and NMPA are quantitated against NDEA-d10). As shown in Table 2, analytical methods for nitrosamine determination in drug substances require low ppb detection limits to ensure applicability for high daily dose drug substances.

The individual transitions specified in the 

 method for quantitation and confirmation were assessed by direct infusion of reference standard solutions of each compound and were either confirmed or replaced by higher performing transitions. The individual transitions were optimized by infusion, as detailed in Table 3, whilst MS source parameters were optimized by flow injection analysis, targeting maximization of the responses for NDMA and NMBA (Table 4).

Once optimization was complete, the final LC separation was refined on a 100 × 2.1 mm solid-core C18 column. An initial isocratic hold of 0.2 min was used to maximize retention of NDMA and NMBA, whilst the remaining analytes were eluted on a 4.0 min gradient (full LC conditions are detailed in Figure 6). Seven-point calibration curves were constructed by injection of standard solutions of nitrosamines according to the 

 method (Table 5 and Figure 7). Excellent linearity was observed across the calibration range for all nitrosamine standards, with all compounds giving r2 > 0.99, in compliance with method system suitability criteria. The accuracy and precision for the method was very good, as detailed in Table 6, demonstrating reproducibility across the calibrated range.

To further assess accuracy and precision, six injections of a QC sample containing the six nitrosamines specified in the 

 method were performed (summarized in Table 7). The QC standard was prepared at a concentration of 20 ng/mL, which would correspond to 0.3 ppm with respect to an 80 mg sample of drug substance (AI limit for NDMA and NMBA in valsartan). Determined accuracies were within 15% of the expected range, with %CV values substantially lower than 15%, demonstrating excellent reproducibility.

Finally, the method was applied to the analysis of valsartan drug substance spiked with nitrosamine standard solutions at a level of 1 ng/mL (equating to 15 ppb of drug substance, according to the sample preparation procedure). A simple extraction of nitrosamines with 1% aqueous formic acid was used following reference 11, which successfully extracted the spiked nitrosamines from valsartan drug substance. For the analysis of other drug substances, such as other sartans, ranitidine, or metformin, alternative extraction methods may be required to achieve suitable recovery whilst preventing solubilization of the API (18,23,24).

The analysis of finished drug product (that is, drug substance and excipients) presents further analytical challenges, particularly for drug products with a high MDD. The potential for interference from drug substance or excipients and the low detection limits required means that in some cases sample cleanup and concentration approaches, such as solid-phase extraction (SPE), may need to be employed to mitigate the impact of the matrix (18,25). Additionally, the potential interference from other low‑molecular‑weight trace impurities, such as residual DMF, has also been noted (26).

Figure 6 shows an example MRM chromatogram. Chromatographic resolution was obtained for all analytes and the isobaric analytes NDIPA and NDPA were clearly separated. It is worth noting that the syn- and anti-conformers of NEIPA were partially resolved but were integrated as a single peak for quantitation. According to Table 2, the AI limits for NDEA in a high daily dose drug substance correspond to 33 ppb, therefore the method clearly shows excellent performance in the low ppb ranges required by regulatory authorities. LOD and LOQ values were estimated based on the signal-to-noise ratio determined from individual MRM chromatograms obtained for the L1 calibration levels and are summarized in Table 8. These values clearly demonstrate the low ppb sensitivity that can be achieved using the tested method.

Since the initial discovery of NDMA in valsartan in 2018, the detection and quantitation of nitrosamine impurities in drug substances has become a major concern. Risk-based assessment of the nitrosamine contamination risks has been widely implemented to identify and mitigate potential sources of nitrosamine contamination in pharmaceutical products. Where risks are present, analytical determination of nitrosamines is essential to ensure the ongoing safety of drug products. Due to the low acceptable daily intake limits established for nitrosamines, which must be considered against the maximum API daily dose, chromatographic analysis necessarily involves the application of highly sensitive and selective mass spectrometric detection capable of quantitation at the low ppb level.

This article has discussed the potential origins of nitrosamine contamination in drug substance, the regulatory guidance governing the allowable limits of nitrosamines, and discussed the use of LC–MS/MS for low-level nitrosamine quantitation. A proposed 

 method for nitrosamines in sartans was selected as a starting point to develop a rapid method for nitrosamine analysis. The final LC–MS/MS method reduced the overall cycle time by 50%. The excellent separation and peak shape provided by the LC column and the separation conditions employed allow this LC–MS/MS approach to readily provide the sensitivity to quantify nitrosamines at the low levels required by regulatory authorities. The final method was demonstrated to show excellent accuracy and precision across the calibration concentration range assessed.

Lessons learnt from presence of N-nitrosamine impurities in sartan medicines

 (June 2020, online). Available: https://www.ema.europa.eu/en/documents/report/lessons-learnt-presence-n-nitrosamine-impurities-sartan-medicines_en.pdf (Accessed June 2021).

Guidance for Industry ANDAs: Impurities in Drug Substances 

(FDA, Rockville, Maryland, USA, 1999, online). Available: https://www.fda.gov/files/drugs/published/ANDA%27s--Impurities-in-Drug-Substances.pdf

ICH Topic Q 3 A (R2), Impurities in new Drug Substances

 (October 2006, online). Available: https://www.ema.europa.eu/en/documents/scientific-guideline/ich-q-3-r2-impurities-new-drug-substances-step-5_en.pdf (Accessed June 2021).

B.N. Ames, W.E. Durston, E. Yamasaki, et al., 

J. McCann, E. Choi, E. Yamasaki, et al., 

M7(R1) Assessment and Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals to Limit Potential Carcinogenic Risk 

(online) Available: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/m7r1-assessment-and-control-dna-reactive-mutagenic-impurities-pharmaceuticals-limit-potential.

Nitrosamine impurities in human medicinal products, Procedure number: EMEA/H/A-5(3)/1490

 (25 June 2020, online). Available: https://www.ema.europa.eu/en/documents/referral/nitrosamines-emea-h-a53-1490-assessment-report_en.pdf (Accessed March 2021).

Control of nitrosamine impurities in human drugs

 (FDA, Rockville, Maryland, USA, February 2021, online) Available: https://www.fda.gov/media/141720/download

Questions and answers for marketing authorisation holders/applicants on the CHMP Opinion for the Article 5(3) of Regulation (EC) No 726/2004 referral on nitrosamine impurities in human medicinal products

 (23 April 2021, online). Available: https://www.ema.europa.eu/en/documents/referral/nitrosamines-emea-h-a53-1490-questions-answers-marketing-authorisation-holders/applicants-chmp-opinion-article-53-regulation-ec-no-726/2004-referral-nitrosamine-impurities-human-medicinal-products_en.pdf (Accessed June 2021).

United States Environmental Protection Agency https://comptox.epa.gov/dashboard

United States Pharmacopeia Proposed General Chapter <1469> “Nitrosamine Impurities” (United States Pharmacopeial Convention, Rockville, Maryland, USA, September 2020 ,online). (Accessed 2021 February).

S.-J. Park, M.-J. Jeong, S.-R. Park, et al., 

Liquid Chromatography-High Resolution Mass Spectrometry (LC–HRMS) Method for the Determination of NDMA in Ranitidine Drug Substance and Drug Product

 (FDA, Rockville, Maryland, USA, September 2019, online). Available: https://www.fda.gov/media/130801/download (Accessed March 2021).

Liquid Chromatography-Tandem Mass Spectrometry (LC–MS/MS) Method for the Determination of NDMA in Ranitidine Drug Substance and Solid Dosage Drug Produc

t (FDA, Rockville, Maryland, USA, October 2019, online). Available: https://www.fda.gov/media/131868/download (Accessed March 2021).

T.W. Iwaoka, T. Hansen, S.T. Hsieh, et al., 

O. Scherf-Clavel, M. Kinzig, A. Besa, et al., 

F. Sörgel, M. Kinzig, M. Abdel-Tawab, et al., 

K.M. Shaik, B. Sarmah, G.S. Wadekar, et al., 

J. Yang, T. Andres Marzan, W. Ye, et al., 

 is a senior research scientist at Avantor.



 is an R&D manager at Avantor, heading a team of specialist scientists in developing next-generation stationary phases for HPLC.

This article discusses the development of a series of applications that will allow for the determination of a number of nitrosamines that have been identified by the FDA as genotoxins to monitor, initially showing how to perform a separation of these compounds from the API and then looking at how MS can be applied to the analysis to ensure that the required detection limits can be reached. 

Low-Level Determination of Mutagenic Nitrosamine Impurities in Drug Substances by LC–MS/MS

 spoke to Hans-Gerd Janssen from the Unilever Foods Innovation Centre in The Netherlands, about the advantages of a novel, rapid method to analyze antioxidants that inhibit oxidation, and he explains that the maturity of gas chromatography (GC) technology is actually an advantage to the analyst.

Q. You recently developed a fast gas chromatographyÐmass spectrometry (GCÐMS) method for efficacy assessment of natural antioxidants (1). Why is this type of analysis important?

As part of Unilever’s commitment to help people transition towards healthier diets, we are reducing the levels of unhealthy saturated fats in our products by replacing them with more healthy unsaturated oils. These unsaturated lipids are more prone to lipid oxidation, resulting in a poor smell and a rapid loss of product quality.
To avoid this, antioxidants are needed. Up till now synthetic chemicals were used to this end, but we now try to use “nature’s ingenuity”. Nature has not only provided us with unsaturated fatty acids, but also with natural antioxidants! Many spices and herbs, for example, have antioxidant properties. It is just that you need to find which spices and herbs work and for that you need to test them. But because there are many potential natural antioxidant sources you need to test a lot of them. Hence the method needs to be fast.

Q. What is novel about the method you developed and what benefits does it offer the analyst?

 In literature you will find many methods for testing antioxidant efficacy. Unfortunately, most of these methods are very far from reality. They test radical scavenging activity, metal chelation capacity, or oxygen scavenging. We wanted a test where we are much closer to reality. We wanted to monitor a lipid over its shelf life and monitor the products that are causing the rancid smell of oxidized lipids. Hexanal is one of these species. We have therefore developed an assay where we oxidize lipids under realistic, yet accelerated, conditions and can monitor the effects of antioxidants by following a key product of the oxidation reaction. The main novelty for me is the much more realistic nature of our assay. Moreover, the method is very fast. One measurement takes less than 2 min. The method is also safe for the analyst. There are no toxic chemicals involved. Just sunflower oil, salt, and air.

Q. Were there any particular difficulties you had to overcome when developing the method?

There were numerous difficulties. First, there are many potential food ingredients in nature that have an antioxidant capacity. So, we had to narrow the area in which we would search. Spices and herbs were a logical focus area because we wanted to stabilize dry soups and sauces. But even with this focus there were still many possibilities to test and we also knew there could be rather strong synergy effects. We clearly needed a fast method. We also wanted to be realistic, meaning that we wanted to study the actual oxidation reaction, not some type of partial aspect of the complex process of oxidation. So, we needed to accelerate the ageing in a realistic way. We studied many routes for acceleration and asked our sensory experts to make sure that the off-smell in the accelerated ageing was the same as in real ageing. We had already fixed hexanal as the read-out for the assay, but we needed to find a fast method to measure it rapidly at very low levels. GC–MS with selected ion monitoring provided the solution here. A final problem was caused by the heterogeneity of many spices and herbs. We needed to take at least 50 to 200 mg to be representative. But this meant that a rather large volume of oil was needed.
In the end our assay was done in 250-mL jars. There were many of them, some 200, so we needed a large climate cabinet for ageing.

The main findings we have patented. We found mixtures of three to five common spices or herbs in specific ratios that exert very strong antioxidant effects. From an analytical perspective I would say the main finding was the high speed and reliability of headspace GC–MS measurements of hexanal. By now we are probably close to 10,000 injections without any major instrumental problems. The rapid hexanal method is now used not only to find new antioxidants, but also to study shelf life stability of our new range of plant-based meat alternatives. We are also thinking of using fast GC–MS on short columns in other applications, such as the dynamic release of flavours from ice cream or the formation of Maillard reaction flavours in cooking.

Q. Are there any specific method development strategies that can turn a “standard” GC method into a “fast GC–MS” method in food analysis?

We are very much used to GC runs that take half an hour and basically take that for granted. Admittedly, with fully automated systems this means you can still do some 50 analyses each day. Such a speed you can probably not meet with the sample preparation, data interpretation, and all the paperwork. I think very often the actual GC run is not the rate-determining step. But if you are interested in a shorter run time the use of a column with a slightly reduced inner diameter is a very simple option. We used a 150 micrometre column. Such a column is roughly twice as fast as the standard 320‑µm column. Reducing the column length is also a simple way to reduce time, at least if you have over-resolution. Finally, GC–MS allows you to separate compounds that are not chromatographically resolved simply by exploiting mass spectral differences. This isalso a good method for reducing the analysis time.

Q. GC is often regarded as a mature technology. Are there any technology developments in GC that offer the analyst “new horizons” in food analysis or any areas of GC that you find particularly innovative?

GC is absolutely a mature technology. But that should not be interpreted as the technique is dull, no longer important, and not relevant anymore. GC continues to be a very important workhorse in the analytical laboratory. GC being mature also means it is reliable, it has proven its value, and, just as for people, we should think of “lifelong learning and improvement” rather than allow people or techniques to dwindle into oblivion. Field-portable instruments that can be used by less‑experienced users, stable retention times that can be repeated in other laboratories, and elimination of the need for calibration are just a few fields where there is still a lot of work to be done. Recent developments that I think are important include the better software we now have for comparisons of chromatograms, hyphenated systems for complex analyses such as mineral oil saturated hydrocarbons/mineral oil aromatic hydrocarbons (MOSH/MOAH) analysis, and high-resolution mass spectrometry. We should also not forget improvements in the sample preparation area, such as SPME Arrow and thin-film microextraction.

H.-G. Janssen, C. Gah, H. Steenbergen, E. Rosing, and M. Spraul, 

 is Science Leader Chromatography and Mass Spectrometry at the Unilever Foods Innovation Centre, in Wageningen, The Netherlands. He is also a part‑time professor at Wageningen University on the special chair of recognition‑based analytical chemistry. His research interests include the development of chromatographic systems, methods, and theories for food analysis and other applications. He has a Ph.D. degree in analytical chemistry from Eindhoven University (The Netherlands) and has over 35 years of experience in chromatography research.

LCGC Europe spoke to Hans-Gerd Janssen from the Unilever Foods Innovation Centre in The Netherlands, about the advantages of a novel, rapid method to analyze antioxidants that inhibit oxidation, and he explains that the maturity of GC technology is actually an advantage to the analyst.

Fast GC–MS for Analyzing Antioxidants and Industrial Food Analysis Applications

ChromTalks provided a new approach to a scientific symposium: Get the speakers to talk about mistakes and hard lessons learned that ultimately led to some of their successes. There were four days of presentations, each including industrial and academic speakers and vendor presentations. The four days encompassed sample preparation, liquid chromatography (LC), mass spectrometry (MS), and gas chromatography (GC).

The full presentations are available online at https://view.ceros.com/mjh-life-sciences/lcgc-ca-symposium/p/1 or point your browser to www.chromatographyonline.com and click the link for ChromTalks or search “ChromTalks”.

The sample preparation sessions focused largely on techniques that are commonly used with GC, especially QuEChERS (Quick, Easy, Cheap, Effective, Rugged, and Safe), presented by Dr. Steven Lehotay, of the United States Department of Agriculture, one of the inventors of the QuEChERS approach, and solid-phase microextraction (SPME), presented by its inventor, Professor Janusz Pawliszyn of the University of Waterloo (Ontario, Canada).

Mass spectrometry was discussed by Koen Sandra of the Research Institute for Chromatography in Belgium, Kevin Schug of the University of Texas-Arlington (USA), and Susan Richardson of the University of South Carolina (USA).
All three speakers spoke of successes, failures, and strange stories as they worked to figure out two instrumental techniques at once: chromatography and mass spectrometry.

The gas chromatography session was led off by Dr. John Hinshaw, the former editor of this column for nearly 25 years. John’s chromatography career spans over 40 years with stints at Varian, PerkinElmer, and currently with Serveron. John spoke of experiences and incidents over his career that many scientists can relate to. In a broad sense, John shared the following key points for experimenting, optimizing, and troubleshooting: focus on safety first, not results; be observant and ask the right questions (What? When? Who?); get to root causes (How? Why? Test!); and make appropriate changes (desired, possible, and practical). 

Lesson #1: In practical troubleshooting or optimizing, the result may be cost-effective, time-bound, or high performanceÑyou get two out of three, and, perhaps most importantly, only change one thing at a time.

John discussed the early computers and electronics that provide the foundations for today’s modern instruments and data systems, discussed in some recent “GC Connections” instalments (1). These included starting on a plug-board manual computer (have you ever heard of one of these?), time shared BASIC (Beginners All-purpose Symbolic Instruction Code) programming in the late 1960s, and working with punch card computing. The original version of the famous DryLab chromatography optimization software was written in BASIC. If a punch card had one error (or even a typo), the program would fail. Additionally, other adventures were discussed, including building a pirate AM radio station that was subsequently shut down.

 Lesson #2: Understand how your electronics and data systems work, and beware of federal and governmental regulations.

John started working for Varian as fused-silica capillary columns were becoming available. They used static coating in a fish tank (complete with fish!) to apply the stationary phase to the inside of the capillary tubing. The big challenge with making early columns was drawing out the fused silica with traditional glass coating machines that did not work well, so they changed to purchasing the fused silica. Early gas chromatographs, especially when used with capillary GC, had many interesting engineering challenges. John related a troubleshooting problem that the baseline on a flame ionization detector (FID) would elevate at seemingly random intervals. They eventually solved the problem: Opening the laboratory door caused pressure changes in the laboratory that caused flow changes in the detector, so they re-engineered the detector to eliminate the flow problem. I encountered a similar problem some years later, from accidentally using a contaminated
air supply. 

Lesson #3: Troubleshooting is a perfect example of John’s first admonitions: Ask the right questions, get to the root causes, and make practical changes, one at a time.

Most of John’s career was spent at PerkinElmer as an applications manager, a principal scientist in instrument and applications development, and later as a GC engineering manager. He described development of a quadrupole GC–MS system with software development in California, product management in Connecticut, development of the quadrupoles in Scotland, and the vacuum system in Germany. Multidisciplinary teams at multiple sites for the development of new products were somewhat unusual then, but are the norm today. 

Lesson #4: Get as much multidisciplinary training and experience as you can.

Dr. Jaap de Zeeuw of Restek has been making and working with capillary columns for over 40 years, and he shared perspectives and stories from his work as a long-time column maker. He spoke about practical experiences with capillary columns, how they have evolved, and how to extend their application and lifetime. He provided true inside stories about how the chemistry behind stationary-phase coating and bonding has evolved significantly in the four decades since invention of the fused-silica capillary columns commonly in use today.

Jaap discussed how we came to use fused-silica tubing, how fused-silica columns are drawn and made, and why they have become so popular. Prior to the introduction of fused silica in 1979, stainless steel and glass were used, and, later, aluminium. Glass is not user‑friendly and is fragile, although it is very stable. Aluminium seems promising, but coating is difficult, special connectors are needed, and aluminium is not robust for aggressive temperature programming. Deactivated stainless steel can be used, but is difficult to work with, and the thermal mass is higher. Fused silica solves most of these problems and can be run up to about 360 °C due to the polyimide coating on the outside of the column. Fused-silica columns can break if scratched by accident. 

Lesson #5: Long‑term operation of fused-silica columns at high temperatures can cause the polyimide coating to degrade and reduce column lifetime. Handle columns with care.

In the past, most stationary phases were made from industrial polymers and were very “dirty”. Today, major column manufacturers have in‑house polymer laboratories that produce very “clean” materials. This is one reason that the temperature stability of capillary columns has improved greatly over the years. Today, there are at least a dozen polysiloxane‑based stationary phases in use. Over time, however, even the newest stationary phases will degrade with temperature, so stabilizing groups and cross‑linking or cross-bonding are used to make the polymer chains more rigid. 

Lesson #6: Stabilized and cross-linked columns have better temperature stability, but they may not have the same selectivity as their non-stabilized equivalents. They are different polymers.

Coating columns is a “special art”. There are many techniques that can be used to coat a column, but the most important idea is that the coating must be homogeneous. Any distortions in the coating thickness will cause a loss of separation efficiency. Deactivation and phase stability are among the most important developments in stationary phases over the last 30 years. Columns must be inert, with deactivation of the fused silica in a manner which prevents formation of reactive silanol groups (is this sounding familiar, liquid chromatography folks?) with organic reagents, or by covering the surface with a polymer. However, deactivation can alter the retention and selectivity of the stationary phase. 

Lesson #7: Deactivation chemistry for the column must lead to minimal retention.

Yes, you still need to condition columns before use. Columns should be heated to the maximum temperature for at least 12 h. This is also a test to ensure that the column deactivation can withstand the upper temperature limit of the column. Exposing the column in the inlet can also cause degradation. This heated portion of the column will be exposed to a small amount of air every time a sample is injected. This air eventually results in column degradation. In addition, the liner will hold high-boiling contaminants that can eventually evaporate into the column and cause degradation. Derivatization reagents are often highly reactive, and they can also attack the column. It is possible for an initially inert column to lose performance quickly. 

Lesson #8: Keep the system and samples clean, condition the column, trim the column on a regular basis, and use the lowest possible temperature.

Jaap also shared insights about column bleed: 

Bleed products can deactivate transfer lines, couplings, and ion sources. He described a simple experiment that showed this idea dramatically. High‑bleed columns can often show more inertness than low-bleed columns. He also described “priming” techniques where mixtures are pre-injected onto a column prior to analytical runs to “protect” the column from the analytes, or the analytes from the columns. In my own experience, we used to do this for an anabolic steroid method. 

Lesson #10: “Analyte protectants” can be used to prepare the column for highly reactive analytes such as pesticides (2).

Finally, I closed the session as “tail-end Charlie” for the academic and industrial speakers, and felt like a kid in this room with just over 30 years’ experience as a gas chromatographer. About 25 years ago, as a much younger chromatographer, I predicted that mass spectrometry would become the only major detector for GC, because it can be both universal and selective, and was becoming easier to use. Also, gas chromatographers love to use GC–MS with capillary GC due to simple interfacing: The capillary column is inserted directly into the ionization source.

Based on my (incorrect) prediction from 25 years ago, my title in the ChromTalks was “Should MS be the only detector for GC?” A simple look at the table of contents of a recently published comprehensive book on GC, or of my own basic text, should tell us the answer (3,4). The Poole text has five separate chapters on detectors and mine has two. New developments in detectors, such as vacuum ultraviolet (VUV) and catalytic reactors for the FID, are strong and ongoing. 

Lesson # 11: Classical GC detectors, including mass spectrometry, are alive and well.

Using examples and lessons from my own experiences, I discussed how a scientific career includes many (often overlapping) career arcs, and many people. When starting a project, I, too, have often asked the questions that John presented at the beginning of the symposium: What? When? Who? I was a very early adopter of SPME with GC–MS, but looked to work on projects that had not already been taken, so my students and I worked on SPME of drugs. Figuring out how to derivatize drugs extracted into an SPME fibre took over a year of failures, and destroyed fibres (with a 1995 list price of over $100 each) (5). It is a good thing that chromatographers are a community, and the vendor was willing to provide them for free. We were ultimately successful, and there have been hundreds of related papers by many authors since. 

Lesson #12: Every experiment, even the ones that “don’t work”, teaches us a lesson, and brings us closer to solving the problem.

The first ChromTalks online symposium in May 2021 was highly successful and informative. The biggest challenge we face when searching the literature and reading references is that, most often, the best and final results are presented, but not the inside stories, missteps, mistakes, and difficulties that led to the successful work. My own first paper on SPME of drugs was a short communication that was the result of about a year of failures and almost giving up. John concluded his presentation by reminding us to “never stop learning and innovating”. Jaap reminded us that we can only take advantage of the high inertness and excellent separation chemistry of today’s capillary columns if we also pay close attention to sample preparation and overall system cleanliness. My own closing remarks were a reminder to remember that chromatographers, whether just beginning or with years of experience, are a community of scientists solving difficult problems together.

Finally, I offer a “shout out” to the four sponsors of the GC session: relative newcomers Lucidity, the makers of miniature GC systems; Activated Research, who provide a catalytic reactor upgrade for FIDs and a low-cost comprehensive two‑dimensional gas chromatography (GC×GC) solution; long-time vendors LECO, the makers of GC×GC and GC×GC–MS systems as part of a broader product range; and PerkinElmer, one of the original players in GC and home to Marcel Golay, inventor of the capillary column. All four of these vendors also provided excellent presentations with useful and practical advice. I hope that we will be able to come together again soon as teachers, users, vendors, and colleagues, in-person, to share our successes—and our mistakes and failures.

Additional information on ChromTalks, including access to all presentations on-demand, is available by scanning this QR code.

M. Anastassiades, K. Maštovská, and S.J. Lehotay, 

 (Elsevier, Amsterdam, The Netherlands, 2nd ed., 2021), pp. 343–442.

H.M. McNair, J.M. Miller, and N.H. Snow, 

 (John Wiley & Sons, New York, New York, USA, 3rd ed., 2019), pp. 113–176.

 is the Founding Endowed Professor in the Department of Chemistry and Biochemistry at Seton Hall University, and an Adjunct Professor of Medical Science. During his 30 years as a chromatographer, he has published more than 70 refereed articles and book chapters and has given more than 200 presentations and short courses. He is interested in the fundamentals and applications of separation science, especially gas chromatography, sampling, and sample preparation for chemical analysis.His research group is very active, with ongoing projects using GC, GC–MS, two-dimensional GC, and extraction methods including headspace, liquid–liquid extraction, and solid-phase microextraction. Direct correspondence to: amatheson@mjhlifesciences.com

The May 2021 ChromTalks, presented by LCGC and CHROMacademy, brought together 12 world-renowned chromatography experts discussing some of the most important lessons and experiences that shaped their careers. This month, we look back at ChromTalks and share the lessons learnt from these great speakers, with topics including sample preparation, GC, and GC–MS. 

Hard-Won Lessons in GC

Dispersion (broadening, or spreading) of analyte zones (peaks) outside of chromatography columns can seriously erode the resolution that good columns provide. In this instalment, we focus on the contributions of dispersion in connecting tubing and detectors to the total level of extracolumn dispersion in a liquid chromatography (LC) system.

In the instalments of “LC Troubleshooting” from the last two months, we reviewed the basic concepts of extracolumn dispersion (ECD) (1), mentioned how the level of ECD associated with a particular instrument can impact the apparent quality of a separation (for example, as measured by resolution), and discussed dispersion associated with the injection step in some detail (2).

In this month’s instalment, we continue our discussion of details associated with the contributions of specific system components. Figure 1 illustrates the contributions of different system components to the total observed variance, starting with the injection step and ending with the detection step. In principle, the total observed peak variance is simply the sum of the variances contributed by each system component. This is only rigorously correct if the dispersion in each element of the system is independent of the others (3), but under most conditions, equation 1 is accurate enough to guide method development and system optimization (4). In this instalment of “LC Troubleshooting”, we continue by discussing dispersion in connecting tubing and detectors.

It is a well known fact that increasing the length and inner diameter of connecting capillaries in a chromatographic instrument increases dispersion. This dispersion is the result of the parabolic flow profile that is established in an open capillary under laminar flow conditions. Under these conditions, the fluid in the centre of the tube moves at twice the average velocity, whereas the fluid near the wall is stagnant (that is, stuck to the wall). Estimating the contribution of this dispersion that occurs in connecting capillaries to the total extracolumn band broadening in a liquid chromatography (LC) system is not straightforward. The two limiting conditions for which analytical solutions exist are as follows: 1) dispersion in a short, straight capillary at high-flow rates (with negligible radial equilibration between velocity streams); and 2) dispersion in very long capillaries at low-flow rates (that is, where full radial equilibration between velocity streams occurs by diffusion). In the first case, the dispersion contribution of the capillary is given by the Atwood-Golay equation below (5):

 are the length and diameter of the capillary, respectively. In the second case, the dispersion is given by the Taylor-Aris equation, where 

 is the flow rate through the capillary and 

 is diffusion coefficient of the analyte in the mobile phase (6).

Most conditions used for practical LC methods lie between the extreme conditions where equations 2 and 3 are valid, and thus require that we consider a partial radial equilibration of the analyte across the tubing diameter (that is, across different mobile phase velocity streams). For this purpose, we can use either a Giddings-style coupling of the previous expressions (equation 4) (7), or an exponential model based on numerical simulations (equation 5) (8):

. In practice, however, equations 4 and 5 both seem to overestimate the degree of dispersion due to these capillaries in comparison to experimental observations, especially at high flow rates (7). One possible explanation for this observation comes from the fact that these equations are only valid for a straight tube, which is seldom used in real instruments where the capillaries are coiled and bent to fit in different instrument components, such as the injection valve, mobile phase pre‑heaters, the column, and the detector inlet. As a result of this coiling, centripetal forces acting on the liquid cause secondary radial flow effects in the liquid. Beyond a critical flow rate, these radial flow effects are pronounced enough to enhance radial mixing and thus the equilibration of the analyte across the diameter of the tube. This in turn relaxes the axial dispersion that normally results from the parabolic flow profile. Nevertheless, equations 2–5 make it clear that using shorter and narrower capillaries reduces contribution of these tubes to the total extracolumn dispersion. This is especially true when using gradient elution, where the volumes of peaks exiting the column are typically much smaller than they are when using isocratic elution. Figure 2 shows that even for a relatively short (140 mm) piece of tubing between the column outlet and the detector inlet, the diameter of this tube can have a dramatic effect on peak width observed at the detector (3). When considering the use of narrow capillaries, however, one must be cautious about the pressure required to push the mobile phase through these tubes at the desired flow rate (Δ

), because this increases along with the inverse fourth power of the column diameter as shown in equation 6 below (where η is the dynamic viscosity of the mobile phase).

For example, changing from a commonly used 120 µm i.d. stainless steel capillary to a 75 µm i.d. capillary comes at the cost of a 6.5-fold higher pressure drop.

Because of the widespread use of UV absorbance detectors in LC, most investigations into the contributions of detectors to extracolumn dispersion have been focused on UV detectors, although most fundamental aspects can also be applied to other types of detectors with flow‑through detection cells. In a first, rough approximation, these flow cells behave as open‑tubular flow paths that are either circular or rectangular in cross‑section. As a result, the equations that govern dispersion in circular tubes can be used to estimate their contributions to extracolumn dispersion. However, the entire fluidic path in a detector module is more than the detection cell itself, and often involves additional in- and outlet tubing, sharp turns, and changes in the cross‑sectional area or shape of channels. Nevertheless, in the literature, the volumetric variance of dispersion attributed to the detector has often been related to the geometrical detection cell volume (

) using an empirical relationship similar to that of the injection volume contribution (2).

The proportionality between cell volume and dispersion is not surprising because equations 2–5 show that a longer or broader flow path results in more dispersion. Thus, there is a tradeoff between dispersion and sensitivity during flow cell design because longer flow cells result in higher sensitivity (9), and broader cell design reduces signal noise. This is illustrated in Figure 3, which shows how narrower peaks are obtained using the flow cell with the smaller volume; however, this comes at the cost of reduced sensitivity (that is, lower peak height) and a much higher baseline noise level. Innovations in the fluidic and optical characteristics of these cells over the past decades have enabled improvements in sensitivity (higher) and noise (lower), while maintaining or decreasing flow cell volume.

Recent studies have shown that equation 7 provides poor estimates of the actual contributions of flow-through cells to the total extracolumn dispersion (10,11). As was discussed for the injection step (2), if a rectangular plug of analyte would move through the detector flow cell in a perfect plug flow without mixing, 

 would be 12. On the other hand, if the flow cell would behave as a “perfect mixer”, 

 would be 1. However, if there are zones in the flow cell that are poorly swept because of flaws in the cell design, then values of 

 below 1 can be observed (10). In addition, another consequence of the similarities between dispersion in a flow-through cell and dispersion in a capillary is that dispersion in the flow cell is expected to be flow rate dependent. Using a custom detector that allowed continuous variation of the cell pathlength, Dasgupta and others also showed that for short cell path lengths, the in- and outlet connections to the illuminated part of the flow cell make up the majority of the measured dispersion (10). To a first approximation, the dispersion at higher flow rates, typically above 0.5 mL/min, from modern low volume flow cells can be estimated using equation 7 with 

= 0.5–0.8 (11). Given the fact that the dispersion of a flow cell is significantly affected by its internal design, some instrument vendors no longer report the geometrical cell volume, and instead report the expected contribution of the flow cell to peak widths (11).

In discussions on dispersion associated with detectors it is important to note that settings associated with the electronic components of the detector (for example, sampling frequency, and detector rise time or time constant) can also affect peak width and shape. Although strictly speaking these do not contribute to extracolumn dispersion in the same way as the other factors we have discussed (for example, detector cell volume), the effects of these settings can influence the observed peak variance and be confused with other contributions. Readers interested in learning more about this topic are referred to prior “LC Troubleshooting” articles (12), and some recent journal articles on the topic (13,14).

Finally, even though mass spectrometric (MS) detection is being used in more and more laboratories, there are far fewer studies of the contributions of MS detectors to extracolumn dispersion than there are for UV detectors (15,16). In the studies that have been done, however, it was found that when optimized MS settings are used, the ionization source and MS detector itself had little impact on observed peak widths. In fact, it was found that the tubing connecting the LC instrument to the mass spectrometer was the most critical contributor to the extracolumn dispersion for these hyphenated systems.

In this instalment of “LC Troubleshooting”, we have continued our discussion of details associated with the contributions of specific LC system components to extracolumn dispersion, this time focusing on connecting tubing and detectors. In the next instalment in this series, we discuss how the impact of extracolumn dispersion can be different under isocratic and gradient elution conditions, and discuss the impact of postcolumn flow splitting on the total dispersion observed at the detector.

, 115619 (2019). https://doi.org/10.1016/j.trac.2019.115619

K. Vanderlinden, K. Broeckhoven, Y. Vanderheyden, and G. Desmet, 

, 73–82 (2016). https://doi.org/10.1016/j.chroma.2016.03.016

, 353–370 (1979). https://doi.org/10.1016/S0021-9673(00)95261-0

, 67–77 (1956). https://doi.org/10.1098/rspa.1956.0065

K.J. Fountain, U.D. Neue, E.S. Grumbach, and D.M. Diehl, 

, 5979–5988 (2009). https://doi.org/10.1016/j.chroma.2009.06.044

, 1325–1337 (2009). https://doi.org/10.1016/j.chroma.2008.12.065

P.K. Dasgupta, C.P. Shelor, A.F. Kadjo, and K.G. Kraiczek, 

, 2063–2069 (2018). https://doi.org/10.1021/acs.analchem.7b04248

In recent articles, the authors reviewed the basic concepts of extracolumn dispersion and how this phenomenon can impact the quality of an LC separation. We now specifically discuss the effects of dispersion that can occur due to tubing and detectors.