Application of Novel Balance Systems: Next Step Towards Laboratory Automatization?

Jef Halbardier

The Column

The Column, The Column-07-16-2018, Volume 14, Issue 7
Pages: 29–35

The key differences between a classical balance and currently available automated systems from the good weighing practice (GWP) perspective and within the scope of ISO9001:2015 quality standard are discussed. The systems under review are: (i) manual analytical balances; (ii) semi‑automatic systems; (iii) fully automatic systems; and (iv) integrated systems that use on-line coupling of the analytical balance with the instrument used for analytical measurements. The parameters defined in GWP guidelines, such as accuracy, uncertainty, minimum weight, and risks (including out‑of‑specification results), will be evaluated for all of these systems.

The key differences between a classical balance and currently available automated systems from the good weighing practice (GWP) perspective and within the scope of ISO9001:2015 quality standard are discussed. The systems under review are: (i) manual analytical balances; (ii) semi‑automatic systems; (iii) fully automatic systems; and (iv) integrated systems that use on-line coupling of the analytical balance with the instrument used for analytical measurements. The parameters defined in GWP guidelines, such as accuracy, uncertainty, minimum weight, and risks (including out‑of‑specification results), will be evaluated for all of these systems. Additionally, data integrity, costs, and time demands associated with respective approaches will be discussed. Finally, the weighing process efficiency will be reviewed in three diverse example applications including gravimetric methods for pharmaceutical analysis, semi- or fully automatic balance in preparation of reference standard solutions in pesticide residue analysis, and the use of a fully integrated system in a high-throughput good manufacturing practices (GMP) release laboratory.

Despite substantial advances in mechanization and automatization of repetitive, labour-intensive, or hazardous tasks in the laboratory environment, the work of the analytical chemist remained largely manual for most of the 20th century. Indeed, the process of analytical laboratory automatization, particularly in sample preparation and instrumental analysis, began in the 1980s (1). The use of automated analytical balance systems remains a tool that is sporadically implemented in various fields of analytical chemistry. When an analytical result depends on the weigh precision, a question arises: Why don’t analytical laboratories invest in versatile automatization systems to improve analytical results to avoid errors incurred during manual weighing?

The Automated System in Analytical Chemistry

In the 15th century, the first development of a “robot-like” system was reported (2,3). Then, with the Industrial Revolution, system automatization and mechanization developments led to humans being replaced for repetitive, labour-intensive, or hazardous tasks. However, the analytical chemistry laboratory remained manual for a large part of the 20th century. Despite automatization being recognized as reducing errors and improving accuracy and reproducibility, only tasks such as pipetting, centrifugation, mixing, and autosampling were automated early. Currently, a variety of automated systems (autosampler, micropipette) find their place in different analytical laboratories (1), but the automatic analytical balance remains a tool that is used sporadically.

An automated instrument can be either “off-line” or “on-line”. For the standalone or “off-line” instrumentation, human manipulation is still required for moving the samples from the preparation system, for example, dilution, transfer, solid‑phase extraction (SPE), to the analytical measurement system. In the integrated instrumentation or “on-line”, the samples carry on automatically, without human intervention, from the preparation system to the analytical measurement system (1).

The benefits of automatization in the analytical laboratory are not only a reduction in manual labour and the risks involved in hazardous tasks, but also an improvement in data integrity, downscaling, improvement in accuracy, speeding up of analysis processes, a reduction in expenditure on costly chemicals, and a reduction in sample contamination and human error (1). Specifically, balance automatization and gravimetric methods reduce error risk from using a volumetric flask (calibration, filling to the line, contamination, cost, and mixing), the labels, and the calculated final concentration (4,5).

 

Weight Accuracy and Good Weighing Practice

As a reliable analytical result depends on the weight precision, authorities and accreditation bodies encourage quality control to be set up. Generally, duplicate preparation on a calibration system and solution comparisons are required. For example, ISO9001:2015 enforces a process approach for risk management and quality. In parallel, good weighing practice (GWP) guidance in the form of a science-based global standard for efficient life cycle management of weighing systems was introduced in 2013 (6). This guide includes the routine testing of equipment that will not be discussed in this article. However, the measurement uncertainty and minimum weight concept will be detailed below.

True and Precise

The International Organization for Standardization (ISO) defined accuracy as true and precise. The trueness refers to the closeness of agreement between the arithmetic mean of many test results and the true or accepted reference value. The precision refers to the closeness of agreement between test results. This accuracy is the main factor of measurement uncertainty, meaning that a normal distribution around the measure weight can be observed. This distribution widens with the small masses measurement. The minimum weight is defined by the minimum weight that a system can measure within an acceptable uncertainty. In addition, the manufacturer adds a safety factor to this minimum weight to take into consideration changing environmental conditions that affect the performance of the instrument, such as vibrations, drafts, wear and tear, or temperature changes. As mentioned, the system drifts over time and needs to be checked and recalibrated in a defined time scale (6).

Accuracy, uncertainty, minimum weight, and risks (including out-of-specification [OOS]) will be evaluated throughout the different automated systems.

 

The Different Weighing Systems

This article will focus on analytical balances, micro balances, and ultra-micro balances, with readability from 0.1 mg for the analytical balance to 0.1 µg for the ultra‑micro balance. Semi-automatic, automatic, or integrated systems are upgrades from the manual system. Therefore, the instrument description will be limited to the option to improve weight accuracy and the automatization option. But first, the definition and advantage of gravimetric methods versus volumetric will be explained.

Gravimetric Method Versus Volumetric Method: The definition of a gravimetric method is weighing not only the solid but also the solvent to enable a specific concentration to be prepared accurately and precisely, whilst a volumetric method uses an accurate volume container. Generally, a volumetric method is performed in a volumetric flask, however, to reduce contamination risk, an exact volume can be dispensed by a positive displacement pipette. Indeed, a volumetric flask is a potential source of contamination, interferents, or noise (from electrolytes for a more selective detector) (7).

Why are gravimetric methods more accurate than volumetric? Volumetric flasks are not recalibrated in-house, and there are also error risks associated when using a volumetric flask. These are numerous: (i) the weighing boat: weighing by difference (transfer of the contents, weigh the remaining), or transfer all the contents with a liquid; (ii) the volumetric flask: the working temperature and the meniscus reading.

On the other hand, gravimetric methods reduce contamination risks as well as error risks, giving only a known uncertainty on both the compound and the solvent weight (7).

The Manual Balance: Apart from the accuracy of the different systems (understanding the accuracy improvement from analytical to micro or ultra-micro balances), different options can be added to improve manual weighing, for example, isolation of environmental variation, the plate, and the anti-static charge.

Analytical balances are sensitive to environmental factors including air flow, bench vibration, or movement. To decrease this effect, the balance can be isolated on a specific table; a graphite plate can also isolate the balance from the bench environment. The weighing plate can be isolated by suspension to decrease static elements such as powder between the plate, and the measurement system or wall around the plate (a double wall for micro balance, or a wider one for ultra-micro balance) can decrease the air flow effect.

Plate suspension has already been discussed, however, a grid plate (with a centre mark for a large plate) or smaller plate can be used. The first one can be used only for a suspended plate on analytical balance whilst the size reduction is used for more accurate micro- and ultra-micro balances.

Generally, modern balances are electronic and sensitive to electrostatic charge. These charges can come from the weighing product itself, the weighing container, or the manipulator. There are different options to reduce it, for example, using an anti-static gate, either off or on the system.

The Semi-Automatic System: The system is qualified as semi-automatic because the final container of the preparation has to be loaded manually between each weighing. There are three different systems: one for liquid dispensing, one for solid dispensing, and a combination of liquid and solid dispensers. One of the typical applications for the combined system is standard preparation.

The combined system includes a manually adjustable dispenser at head height (the height is manually commutable) to weigh directly into the destination container, a suspension plate with a reduced size compared to a manual balance, optional anti-static gate, and double wall for micro‑automatic balance. The accuracy is improved by reducing the manual operation and thus reducing electrostatic interferences. The accuracy is improved by a factor of three compared to manual weighing. Another advantage is that the final concentration is recorded directly into the electronic laboratory notebook, which is easily transferable to a laboratory information management system (LIMS) and printable on a label.

The Fully Automatic System: This system has the same technical parts as the semi‑automatic but with an autosampler. Sample weighing or small-scale production are two possible applications for this system.

On-Line System: A manual balance is used within a robotic system and thus without human manual handling. It offers the same accuracy as a semi-automatic balance. A typical application example is difficult weighing samples (that is, sticky samples).

Balance Comparison: A comparison between the accuracy of manual balances and automatic balances and a comparison between volumetric methods on manual balances (traditional across the sector and laboratories) and gravimetric methods on automatic balances are summarized in Table 1 (7,8,9).

A precision of 0.57% is observed for manual balances whilst a precision of around 0.2% is observed for an automatic system, including the integrated system (5). The trueness (recovery) is three times closer to the true value for automatic systems than manual. Better accuracy of the automatic system led to a smaller minimum weight. If for a manual system the minimum height is X, for the same system under automatic conditions the minimum weight will be around 0.67 X. A smaller minimum weight allows the method to be scaled down and made more cost‑effective in term of solvent, waste, and costly reference compounds (7,8,9).

 

Applications

Pharmaceutical Sector: Expensive reference materials, active pharmaceutical ingredients (APIs), or compendium standards in parallel with the huge amounts of volumetric flasks used, unknown peak identification (contamination from volumetric flasks), and a request for accurate (precise and accurate) data make automatic balances a useful tool for an analytical chemistry laboratory performing pharmaceutical release, method development, or small-scale research and development.

The example presented here focuses on data released in parallel with method uncertainty (as defined in ORA-LAB. 5.4.6) (10).

First, FDA<1251> (11) defines the use of automatic balances, meaning that automatic can be used in line with regulations. Data integrity is a regulation hot topic, whilst an automatic balance integrated easily into all the systems can be configured to communicate directly to a LIMS system. In addition, printing accurate labels can guarantee a better traceability. Therefore, the use of an automatic balance is recommended by FDA<1251> (11).

A further reason why FDA<1251> recommends automatic balances is related to method uncertainty (11). Indeed, one of the tightest release criterion on a drug is ± 2% of the certified value. With a precision of approximately 0.5%, a liquid chromatography–ultraviolet (LC–UV) analytical measurement is often performed. However, the weighing methods will have an impact on the overall uncertainty. Table 2 presents a compound analyzed by LC–UV (injector precision of 0.21%) with a gravimetric preparation on an automatic system. A 0.2% precision is achieved with this system without any volume or weight outside the method procedure criteria (7). Compared to the 0.6% relative standard deviation (RSD) on the weight, a tolerance of 0.4% on the volume of a volumetric flask, and the 0.5% of the analytical measurement system-cumulated up to 1.5% for a manual volumetric method-the automatic system provided a tolerance of approximately ten times smaller than the criterion limit whilst the manual system uncertainty represented 75% of the tolerance limit. Using an automatic balance improved the closeness between the value measure and the true value ensuring patient safety. This is also illustrated in the example from Pfizer (Figure 1) (7).

The other important aspect is the scale‑down process and the lower minimum weight to support expensive or rare API research and development.

Standard Preparation for Pesticides Analysis: Multiresidue analysis methods like pesticide residue methods or QuEChERS (quick, easy, cheap, effective, rugged, and safe) can include more than 500 different reference materials to weigh. On a manual instrument, all of these compounds have to be weighed in duplicate (generally a laboratory goes for a commercial solution as a second duplicate), the concentration calculated, and then transferred to a labelled vial. The process is time-consuming and can occur every year for a high-throughput laboratory.

The concentration of working solutions is in the ηg/mL scale, whilst stock solution preparations are in the mg/mL scale or commercial solutions in the µg/mL scale. What are the advantages of using an automatic balance? First, an automatic system is fully compliant under ISO 17025 (normal for pesticides analysis). Second, it allows better data integrity in terms of concentration and solution labels, but also accelerates the process with the possibility to store the reference materials inside the weighing head of the freezer. Third, it allows mixes to be prepared at lower concentrations. Finally, a weighing scale allows more accurate solutions to be prepared. For example, a 25‑mg reference material can be weighed three times accurately (minimum weight of 7 mg) in a unique stock solution of 1 mg/mL, 21 mL stock solutions can be prepared, and 2.1 L of solutions at 10 µg/mL (commercial mix solution currently available). A higher volume and a longer use of the reference standard is therefore observed compared to manual weighing.

With a full automatic system that includes an autosampler for 30 vials, multiple stocks or mix reference solutions can be completed and dispatched easily. These solutions have their weight and concentration directly recorded and associated to each vial filled in the automatic system.

High-Throughput GMP Laboratory: This example focuses on a high throughput GMP release. Indeed, clinical laboratories and other sectors can find benefits to an integrated system (7).

In addition to high accuracy, the integrated system can reduce sample handling by the operator and can run in parallel to the analytical measurement system, reducing degradation overtime and overall sample preparation time. This last benefit gives the analyst more time to interpret the data, which can have an improved traceability over the sample preparation.

 

Conclusion

Versatile weighing automatic systems exist and their applications look promising in terms of accuracy, time savings, cost, and data integrity. Instrument cost remains acceptable compared to the potential source of revenue and the day‑to-day cost reduction (solvent used and waste reduction included). However, a full evaluation of the investment cost and revenue must be calculated on a case-by-case basis.

References

  1. G. Rakesh D. Prabhu, and P.L. Urban, Trends in Analytical Chemistry88, 41–52 (2017).
  2. M.E. Rosheim, Leonardo’s Lost Robots (Springer, Heidelberg, Germany, 2006).
  3. M.E. Moran, J. Endourol.20, 986–990 (2006).
  4. White paper Mettler Toledo (ISO 9001:2015 and weighing: Managing risk and quality).
  5. White paper Mettler Toledo (Gravimetric sample preparation: Reducing sample size and OOS error).
  6. White paper Mettler Toledo GWP® (The standard: Science based weighing).
  7. White paper Mettler Toledo (Gravimetric sample preparation: the alternative to volumetric flask) (2015).
  8. F. Foster, J.R. Stuff, and E. Pfannkoch, GSPE 6 (Automation of sample preparation steps using a robotic X-Y-Z coordinate autosampler with software control ORALaboratory procedure) Anatune App Note (2008).
  9.  Gerstel Solutions Worldwide 15, 16–18.
  10. US Food and Drug Administration, Estimation of Uncertainty of Measurement ORA-LAB. 5.4.6
  11. General chapter <1251> “Weighing on an Analytical Balance,” in United States Pharmacopeia 41–National Formulary 36 (United States Pharmacopeial Convention, Rockville, Maryland, USA, 2018).

Jef Halbardier graduated in 2007, with an M.Sc. in bioengineering in chemistry and biochemistry from the Faculty of Science Agronomy Gembloux in Belgium. He has held different roles including research scientist for the National Reference Laboratory (pesticide residues analysis, polychlorinated biphenyls [PCBs], polycyclic aromatic hydrocarbons [PAHs], hormones, and veterinary drugs), study director at Baxter, and research chemist at Covance. He is the co-author of
several publications including one AOAC method awarded as best method of 2016. Before joining Reading Scientific Service Limited, he was the owner of his own consulting company and worked as a mass spectrometry technical specialist in various areas.

E-mail:jeanfrancois.halbardier@rssl.comWebsite:www.rssl.com