Technology Forum: Food & Beverage Analysis

LCGC asked a group of experts about the techniques and technologies they chose to use in food analysis, and how they made those choices. Joining us for this forum are Lauryn Bailey PhD, global marketing manager for food & environmental markets, AB Sciex; Gareth Roberts, international business development manager, ALMSCO International; Gerry Broski, food safety marketing director, Thermo Fisher Scientific, Michael Pickering, president, Pickering Solutions; Hayley Crowe, mass spec detection product specialist, PerkinElmer; and Sky Countryman, manager, PhenoLogix & Applied Technology.

Food analysis is often plagued with extensive sample preparation challenges, leading to slow method development and long analysis times with low sample numbers being able to be conducted on a daily basis. What are the most promising technologies to answer these challenges? Is it routine sample techniques with standard formulas such as quick easy cheap effective rugged safe (QuEChERS), or is it the direction of fast extraction dilution flow-injection mass spectrometry (FED-FIMS), where sample handling is minimal and chromatography is eliminated?

LB: Both QuEChERS and FED-FIMS are promising technologies that answer some of the challenges that plague food analysis, but neither is a “catch-all” solution. QuEChERS works well for fruit and vegetable matrices and covers a large number of pesticide compounds but results are often highly dependent on the matrix type and could require matrix-matched calibration for accurate quantification, adding a significant amount of time and labour to the analysis. Additionally, QuEChERS results can be highly variable or even unacceptable for highly pigmented or complex matrices (such as those that are botanical or herbal).

FED-FIMS is a great tool for fast, high-throughput analysis but is limited in the number of compounds that can be analysed in a single run, thus making screens with a large number of compounds difficult or even not possible.

The most promising technologies for food testing combine the best features from both of the above approaches and couple those with high sensitivity mass spectrometers. The use of either FED or QuEChERS for a fast sample extraction (then followed by a large dilution) allows quick sample preparation and limited matrix interferences, and when combined with a highly sensitive mass spectrometer, allows the detection of many compounds while still meeting the detection limits required by most regulatory agencies and food producers.

GR: Food analysis covers such challenging substrates and a range of analytes that you have got to be really careful when selecting your preparation technique. Reducing the number of sample preparation steps is clearly desirable but to avoid selective losses of target analytes, I think it is safest to stick with simple but relatively gentle methods, and let the instrumentation take the strain. This is especially important when detecting trace-level components in complex matrices, where preconcentration methods (such as thermal desorption) and chromatography are indispensable. Moreover, when combined with powerful MS technologies such as triple-quadrupole or time-of-flight (TOF), it is now becoming possible to achieve analyte identification, ultra-low detection limits and quantitative accuracy — all in a single run.

GB: QuEChERS has certainly eased the burden of sample preparation, however like any technique it does have limitations. As technology improves, direct sample analysis using atmospheric pressure techniques such as direct analysis in real time (DART), desorption electrospray ionization (DESI), and solids analysis probes are gaining traction as alternatives that virtually eliminate sample prep, but they also have their limitations. Sample prep accounts for 60% of the analysis workflow and any time savings in sample prep will be tangible. However, given the vast number of matrices and analytes it will be difficult to have a single technology or technique, which will be sufficient to address all needs. Several papers are available on the subject of DART coupled to high resolution MS. Automated sample preparation methods have been developed using turbulent flow chromatography, which uses size exclusion combined with solid-phase extraction (SPE) to minimize or eliminate manual sample prep. The technology comes out of the clinical market where high sample volumes are the norm and cost effectiveness is a driver for sample prep efficiency.

MP: Sample preparation is the largest contributor to analysis variation. Current sample preparation techniques such QuEChERS being portrayed as a standard for all samples are more marketing hype than reality. For example, with fatty matrices gel permeation chromatography (GPC) continues to be the standard. FED-FIMS is an interesting technique that has its place with less complex matrices.

Developing nondestructive analysis methods such as laser spectroscopy to measure headspace in packaged foods for freshness eliminates sample preparation and promises to be the future of food testing.

Meanwhile, sample preparation automation technologies are making significant advancements to increase throughput and improve reproducibility.

HC: The complexity of food matrices requires that sample preparation be considered an integral part of any analysis. Methods such as QuEChERS are addressing some of the complexity. As part of an analysis portfolio, we must have QuEChERS. However, it does not address the limitations on method development or long analysis time.

We believe that the direct sample analysis approach, such as FED-FIMS is the direction for food analysis. However, the best solution is to eliminate or minimize sample preparation altogether. To accomplish this, some radical, innovative and transformative approach is needed.

We believe that direct sample analysis for food-related analysis is the optimum approach as it will not only save cost and time, but will also allow better insight into the raw sample without comprising the sample integrity.

SC: There is no question that improved techniques for real-time field testing are necessary, but there will always be a need for laboratory-based confirmation testing using chromatography. Historic chromatographic methods required more extensive clean up due to the nonspecific nature of the detection mode (UV, electron capture detection, flame photometric detection etc). Techniques such as gas chromatography–tandem mass spectrometry (GC–MS–MS) and liquid chromatography (LC)–MS–MS combined with high efficiency chromatographic separations provide much higher specificity and sensitivity, which reduce the need for as extensive sample clean-up. General sample preparation techniques like QuEChERS can be good for screening, but they are not amenable to automation. SPE on the other hand can be easily automated. The combination of SPE and LC–MS–MS or GC–MS–MS is the most likely strategy for high-throughput testing available today.

What special considerations do you place on sample preparation when dealing with trace quantification from different food matrices? Are there some overriding factors to consider?

GB: Given the fact that the purpose of sample prep is to separate the matrix from the analyte of interest, there are many considerations that must be taken into account depending on the matrix and what type of analysis technique you are using. Considerations include the various steps of the process including extraction and dilution, derivatization and standard addition, heating and cooling and reconcentration. Matrix considerations around extracting include fat content, reactivity and binding effects. A proven technology gaining popularity in the food industry is accelerated solventless extraction (ASE). This technology significantly reduces time, labour and high solvent usage required by traditional methods such as Soxhlet or sonication.

MP: The most challenging consideration for any analysis is dealing with interfering matrix components. Special care must be given to trace analytes so as not to lose the recovery of the targets when removing interfering non-targets, compromising sensitivity. Additional improvements in sensitivity can be achieved by derivatizing the analytes post-column to alter the selectivity of detection.

HC: Maintaining the sample integrity and the process used for sample prep is a significant factor. The issue then is not in detection, but in ensuring that the trace compounds of interest are separated and transmitted to the detection and measurement system.

SC: In trace analysis achieving the highest absolute recovery is critical so it is best to target fewer analytes and optimize the extraction and clean up to meet your detection limits. Good sample clean-up is also critical and the more specific the technique (such as SPE) the better. But it is also important to know how low is realistic? Our lab has had several cases where it was nearly impossible to find a "blank" matrix. In the case of chloramphenicol, even certified organic shrimp had high-ppb-level contamination. In another case, the stability of the analyte was a problem. Certain pesticides are pH labile; when analysing citrus fruits or some types of lettuce you may not be able to achieve low detection limits simply due to the natural pH of the sample.

GR: When detecting and quantifying trace analytes (for example, persistent organic pollutants) in food samples, there are three main factors to consider: (a) sample preparation should be applicable to a wide variety of sample types without major changes; (b) it should not compromise the quality of the results; (c) it should not be overly complicated — it is better to have a relatively simple sample prep procedure in conjunction with a robust analytical technique. If looking more generally at volatile organic compound and semi-volatile organic compound (VOC/SVOC) profiling, options such as direct thermal desorption, use of microchambers and sorptive extraction offer rapid, cost-effective sampling at the same time as analytical excellence.

LB: Sample preparation is often the most important element of food testing procedures, particularly given that each food sample varies significantly from the next. Special consideration always needs to be given for different food matrices — for example, botanicals, herbs, and spices may require a more specific sample preparation approach to remove matrix components that could interfere with the analytical analysis. Additionally, different labs may have different needs regarding the list of compounds being analysed and the limits of quantification needed for those compounds in the specified matrix, influencing the way the samples are prepared and analysed. Often for food, every analysis could be completely independent due to these variables.

What needs are currently unmet in food and beverage analysis?

HC: Challenges in food and beverage analysis include making analysis faster, easier and providing analytical methods to give food companies the ability to be responsive in a dynamic analytical environment. As new and unknown contaminants emerge and as detection levels become more challenging, the analytical systems employed by food companies must be able to match these needs.

An emerging need and opportunity for greater efficiencies is in the area of information management. Food is a global product and the information to manage and ensure the safety and quality of the end product comes from all over the globe. Effective information management systems are yet to be realized in the food industries.

GR: It is not so much a case of finding solutions to needs that are unmet (after all, analysts will always find a way round a problem), but about improving the existing solutions. Examples include: the need to detect ever lower levels of analytes, driven by regulatory and market pressure; the desirability of reducing the cost and complexity of sample preparation; ensuring that methods are amenable to a broad range of both liquid and solid products; and being able to add new chemicals of concern to existing methods without having to start from scratch.

LB: The biggest unmet need in the food and beverage analysis industry is the availability of “official, validated methods” (preferably that cover large screens of compounds and apply to a variety of matrices), which have been verified in collaborative studies and are widely accepted by regulatory agencies, accreditation agencies and food suppliers. The process of validating methods in the laboratory is often long, tedious and laborious, and results can still be called into question by regulators and accreditors given that the method may only be “single lab validated.” The ability to adapt an official reference method, which has been verified in a collaborative study of several labs, enables laboratories in the industry to get methods running in their lab quickly and have results more readily accepted by regulators, accreditors and food suppliers.

Agencies and organizations, such as AOAC and others, which have the reputation of producing verified official methods, are often slow to get these methods validated and published to the industry, and methods often do not achieve the “more compounds, more matrices, faster analysis” requirements of the testing labs in the industry. This is a major need for the industry and will be particularly valuable to growing laboratories.

SC: There is a real need in the food and beverage industry for updated testing methods that utilize better analytical technologies. Much of the testing is being done in the food producing countries such as China and they need reliable tests that will prevent food related problems.

Melamine is a perfect example of an outdated method that resulted in needless deaths when better analytical techniques like LC–MS–MS or GC–MS existed and could have identified the faulty nitrogen reading before the food reached consumers. Since this outbreak, we have had many of our scientists travel to China to work directly with the government to implement new testing protocols to better identify contaminants in food. We have a long way to go but we are committed to working with global food agencies to help develop and implement these new analytical procedures.

GB: Food and beverage analysis is a unique and dynamic market and needs are constantly changing. Variables include the perishability of food, the staggering number of food matrices, an ever-expanding list of contaminants including new threats brought on by deliberate or economic adulteration and the immediate threat of microbiological organisms. In most simple terms, we need rapid testing capabilities that are easy to use. Given the amount of testing required, the tests should be multi-analyte and have to be cost effective. Non-targeted analysis techniques are becoming important due to fraudulent practices reported. Powerful but user-friendly software that integrates to high-level data from different systems is typically requested by users. Analysis results have to be unambiguous and accurate with minimal or no false positives and false negatives. The areas where research is being conducted to improve existing methods is allergen detection, rapid microbiological analysis and non-targeted food screening approaches.

MP: The biggest concern for the food and beverage industry is the additional analytical technologies and methods needed to address the high numbers of newly registered and soon-to-be registered chemical compounds.

Additionally, as natural products and herbal supplements move into an FDA regulated environment, that will further strain the industry’s existing analytical capacity.