Aaron Hineman

 <232> and <233> for the determination of metals in pharmaceutical products, the need has arisen for analysis of a wide variety of sample types, ranging from raw materials to final medications. Depending on how the medication is administered (oral, inhalation, or parenteral) and its daily dose, the maximum permissible exposure of the various elements changes. Although inductively coupled plasma–mass spectrometry (ICP-MS) is generally thought of as the preferred analytical method because of its ability to measure extremely low levels, ICP-optical emission spectroscopy (OES) can easily meet the requirements for many medications with low daily doses. With its advantages (compared to ICP-MS) in matrix tolerance, robustness, simplicity, and cost, ICP-OES is the technique of choice for the analysis of medications with low daily doses. This article will focus on sample preparation considerations and the analysis of a variety of medications with ICP-OES to meet the 

 <232> and <233> for the analysis of elemental impurities in finished drug products (1,2), it is important for manufacturers to monitor the metal content of the final products to be in compliance. The maximum permitted daily exposure (PDE) of different target elements is primarily defined in 

 <232>, and is based on the route of administration of the medications (1).

 <233> defines the analytical requirements, which are all based on the J values for each element. The J values are a function of the PDE, the maximum daily dose of the medication, and the dilution factor used in sample preparation. The four analytical criteria specified in the quantitative procedures for 

 <233> are accuracy, repeatability, ruggedness, and system suitability.

	Pharmaceutical tablets are complex mixtures that can contain a wide variety of excipients. Because of the variation and complexity, closed-vessel microwave digestion is the most efficient technique to get the tablets into solution without losing volatile elements, such as mercury (Hg). However, when tablets contain silica (SiO

) as excipients, digestion becomes more complex as hydrofluoric acid (HF) must be used. The presence of HF also requires the use of non-glass sample introduction components, unless the HF is dealt with during the digestion process.

	Inductively coupled plasma–optical emission spectroscopy (ICP-OES) can easily meet the requirements for many medications with daily doses of 1 g or less. With its advantages (compared to ICP-mass spectrometry [MS]) in matrix tolerance, robustness, simplicity, and cost, ICP-OES is the technique of choice for the analysis of medications with low daily doses.

	This article focuses on sample preparation considerations, and the analysis of a variety of medications with ICP-OES to meet the 

	A variety of medicines were purchased locally, as summarized in Table I. These samples represent common, over-the-counter medications that are popular among consumers, and include both tablets and nasal sprays. In addition, the acid reducers contained titanium dioxide, while the sleep aid and motion sickness medications had silica dioxide as an inactive ingredients. As a result, these medications required special treatment during sample preparation.

	Because of the variety of sample types and analytes, closed-vessel microwave digestion with a Titan MPS microwave preparation system was used for sample preparation. The high temperature and pressures easily broke down the medications while using a minimum amount of acid (1.5–3 mL HNO

 + 0.5–1.5 mL HCl, depending on the analytes), while the sealed vessels facilitated the recovery of volatile elements, such as mercury and selenium. Table II shows the microwave digestion program used. For each digestion, one tablet was used. For tablets containing SiO

, hydrofluoric acid was required, followed by a HF-complexation step (3 mL of saturated boric acid solution), which allowed a standard glass sample introduction system to be used for all medications.

	All analyses were performed on a PerkinElmer Avio ICP-OES system, using standard sample introduction components and plasma conditions. All measurements were made against external calibration curves prepared in a mixture of nitric and hydrochloric acids to match the acid composition of the samples after digestion. Yttrium (Y) and scandium (Sc) were added to all blanks, standards, and samples as internal standards.

	The digestion procedure was designed to minimize acid use and time. While most elements require nitric acid to be stable in solution, the platinum group elements (PGEs: Au, Ir, Pd, Pt, Rh, Ru) are only stable in hydrochloric acid. However, silver (Ag) precipitates in the presence of low HCl concentration, but remains stable at large excess HCl levels. Therefore, when measuring Group 2B elements, the challenge was to find appropriate HNO

 and HCl concentrations to maintain the stability of all elements in solution, which turned out to be 3% HNO

	Sample preparation time was further reduced by incorporating a cooling step in the microwave program. This step allowed the vessels to be handled 15 min after the digestion was complete, as opposed to 1–2 h if the cooling step were not used.

	Although a single tablet of each medication is required to meet the USP requirements, the digestion methodology was tested with multiple tablets of each medication in a single digestion vessel. It was found that the methodology could completely digest two of the SiO

-containing tablets (total mass = 0.86 g) and five of the allergy tablets (total mass = 0.5 g). Higher levels were not evaluated.

	One of the concerns with the analysis is that some analytes can interfere with other analytes. Consequently, it is possible to obtain results that meet the criteria of 

 <233> while actually measuring an interference on an analyte instead of the analyte itself. To address this situation, all analytes were run individually as single-element solutions at their 1.5J concentrations, and their spectra compared. Where interferences were observed, alternate interference-free wavelengths were chosen when available. For elements where they were not, the effects of the interferences were removed using the system's multicomponent spectral fitting (MSF) feature. Although interelement corrections (IECs) could also be used, MSF was found to be simpler to implement and more robust.

	The concentrations of all elements in each tablet analyzed were less than 0.3J, signifying that these medications do not pose a health risk.

 <233> defines four validation criteria that must be met when performing an analysis: suitability, accuracy, repeatability, and ruggedness. Each medication that was analyzed was subjected to these validation criteria.

	To meet the system suitability criterion, a 1.5J standard must be run at the beginning and end of each batch run, and the difference between the results must be less than 20%. Figure 1 demonstrates that each type of medication easily meets this criteria, with the largest variance being less than 6%.


	Figure 1: System suitability plot showing the drift over a batch analysis from 1.5J standards measured at the beginning and end of batch analyses for allergy tablets and TiO

	With the suitability of the system established, the accuracy of the methodology was evaluated. The accuracy criteria requires that spikes of 0.5J, 1J, and 1.5J be added to the microwave vessels before digestion and carried through the complete sample preparation scheme to determine if analytes are lost during sample preparation. The results are shown in Figure 2, indicating that all spike recoveries are less than 10% for all elements in all medications, significantly less than the method acceptance criteria of 70–150%.

	With the accuracy proven, the repeatability of the methodology was assessed. According to 

 <233>, the repeatability is evaluated through the preparation and analysis of six individual samples of the same medicine spiked at the J values. The RSD of the analysis of these six samples must be less than 20%. As shown in Figure 3, the RSDs are all less than 6%, easily surpassing the USP criterion.


	Figure 2: Accuracy plot showing 0.5J, 1J, and 1.5J predigestion spikes in allergy (green), TiO

-containing (pink) tablets. The black lines define the USP <233> limits.

	With the repeatability of the methodology demonstrated, the ruggedness was evaluated by analyzing the tablets in Figure 3 on a second day. According to USP <233>, the RSDs of the 12 measurements over two days must be less than 25%, which is easily met, as shown in Figure 4.


	Figure 3: Repeatability plot showing RSDs of six predigestion spikes of individual tablets containing TiO

	This article has demonstrated the ability of ICP-OES, in combination with closed-vessel microwave digestion, to meet the requirements of 

 <232> and <233> for allergy tablets and medicines containing TiO

 as excipients. With the inclusion of volatile elements and difficult matrices, high-pressure, high-temperature microwave digestion in sealed containers simplifies and speeds up sample preparation while preventing the loss of volatile elements. Although ICP-MS can also be used for these analyses, ICP-OES has the advantages of lower cost, higher matrix robustness, and ease of use for medicines with a daily dose of less than 1 g per day.


	Figure 4: Ruggedness plot showing RSDs of 12 measurements over 2 days of allergy medications and tablets containing TiO

, spiked at the J level for all analytes.

	(1) General Chapter <232> Elemental Impurities – Limits: Second Supplement of 

	(2) General Chapter <233> Elemental Impurities – Procedures: Second Supplement of 

 is a Product Line Leader at PerkinElmer in Seattle, Washington. Direct correspondence to: 

 is a Senior Application Specialist at PerkinElmer in Shelton, Connecticut. Direct correspondence to: 

Articles

With its advantages-compared to ICP-MS-in matrix tolerance, robustness, simplicity, and cost, ICP-OES may be the technique of choice for the analysis of medications with low daily doses. For such analyses, the use of high-pressure, high-temperature microwave digestion in sealed containers simplifies sample preparation and prevents the loss of volatile elements.

Meeting the USP  and  Guidelines with ICP-OES

This study focuses on the determination of a suite of Class 1 and 2A elemental impurities (Pb, As, Cd, Hg, Co, Ni, and V) in various over-the-counter antacid tablets by inductively coupled plasma–mass spectrometry (ICP-MS), as defined in the 

International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use 

. The analysis is performed according to the recommended sample preparation procedure and validation protocols laid out in 

 <233>, by optimizing both the microwave digestion procedure and the analytical methodology for these types of sample matrices that contain extremely high levels of calcium, magnesium, and other mineral components.

International Conference on Harmonization

) (2) have aligned the list and levels of 24 elemental impurities in pharmaceutical materials, which will be implemented on January 1, 2018. The elements of toxicological concern have been categorized into three groups based on their permitted daily exposure (PDE) levels, which are defined in 

 <232> (3). To show compliance, manufacturers will need to measure the elemental contaminants by following the analytical methodology described in 

 <233> (4). To supplement these new chapters, the United States Food and Drug Administration (FDA) also issued draft guidance recommendations for the pharmaceutical industry on elemental impurities in drug products that represent its current thinking on the topic from a regulatory perspective (5).

	Elemental Impurities Analytical Methodology

 <233> recommends the use of either inductively coupled plasma–mass spectrometry (ICP-MS), inductively coupled plasma–optical emission spectroscopy (ICP-OES), or any other suitable technique that meets the necessary validation protocols, for the determination of elemental impurities in finished drug products. The analytical capabilities of ICP-MS make it the most suitable technique for performing determinations of the Class 1 elements at these low levels on a routine basis-especially for drug products with a large daily dosage (>10 g/day). Among this category of medications and supplements, antacids present a unique set of analytical challenges for ICP-MS because of their extremely high calcium or magnesium content. High levels of total dissolved solids (TDS) can cause ionization suppression in the plasma as well as instrumental drift and increased maintenance because of deposition within the instrument. However, as shown in this work, these challenges can be overcome with optimized sample preparation and instrumental analytical conditions. A list of Class 1 and 2A impurities and their PDEs defined in 

	The eight antacids chosen for the evaluation represent a typical cross section of antacids available over the counter and are shown in Table II.

 <233> also makes suggestions on a closed-vessel or microwave digestion procedure to get solid samples into solution. A typical digestion for antacid samples uses nitric and hydrochloric acids, but because silica (SiO

) is present in some of the samples, a source of fluoride is required to dissolve the silica-typically hydrofluoric acid or tetraethyl fluoroboric acid.  

	In this work, all samples were digested using a microwave digestion system (Titan MPS System, PerkinElmer Inc.) with standard 75-mL TFM (modified polytetrafluoroethylene) vessels. Approximately 3–5 g of material were crushed and homogenized. Then 0.30 ± 0.01 g of each sample was added to a digestion cup and placed in a digestion vessel, followed by 5 mL of nitric acid (70%), 1 mL of hydrochloric acid (35%), 1 mL of hydrogen peroxide (30%), and 0.5 mL of hydrofluoric acid (49%) when needed. The vessels were allowed to sit uncapped for 10 min to allow for any prereactions to occur safely before being capped and digested following the program in Table III. To evaluate the effect of the sample preparation on analyte recovery, spikes were added to the microwave vessel before the addition of acid.

	When the digestion was complete, all samples were diluted with deionized water to a final volume of 50 mL, resulting in a total dilution factor of 167x with a reagent matrix of 10% HNO

, 2% HCl, and 1% HF. Calibration standards were prepared in this same matrix. To stabilize mercury, 200 ppb gold (Au) was added to each sample, standard, and blank.

	For this validation, all analyses were performed on a NexION 2000 P ICP-MS system (PerkinElmer Inc.) (6) with a SMARTintro high-throughput and high-matrix sample introduction module using standard operating conditions. This configuration offers enhanced speed of analysis through the use of an in-line flow-switching valve and provides high matrix tolerance through the use of the all matrix system (AMS), which uses aerosol dilution for enhanced matrix tolerance. Although only vanadium (V), cobalt (Co), nickel (Ni), and arsenic (As) are affected by spectral interferences originating from the sample matrix and sample preparation procedure, the ICP-MS system was operated in collision mode for all analytes and samples, providing a simple method for the analysis. The instrument conditions are shown in Table IV, and the elements and masses appear in Table V.

 <233> outlines the requirement to calibrate using a matrix-matched blank and calibration standards using concentration values of 0.5J and 1.5J. The 

 is defined as the permitted daily exposure limit for the element of interest, appropriately diluted to the working range of the instrument, after the sample preparation process is completed. In the case of antacids, large daily dosages are typical. For the eight samples in this study, the largest maximum dosage was 30 g of the antacid per day. Therefore, to test all samples using one set of calibration standards, a 30-g/day dosage was used in calculating the calibration range (0.5J and 1.5J). The elements and standard concentrations were calculated using PerkinElmer’s J Value Calculator software, as shown in Figure 1. 

 <233> also contains the following requirements for method validation:

		Accuracy: The matrix and materials under investigation must be spiked with target elements at concentrations that are 50%, 100%, and 150% of the maximum PDE. Mean spike recoveries for each target element must be within 70–150% of the actual concentrations. 

		Repeatability: Six independent samples of the material under investigation must be spiked at 100% of the target limits defined and analyzed. The measured percent relative standard deviation (%RSD) must not be more than 20% for each target element. 

		Ruggedness: The repeatability measurement testing procedure should be carried out by analyzing the six repeatability test solutions either on different days, either with a different instrument or by a different analyst. The %RSD of the 12 replicates must be less than 25% for each target element. 

		System suitability: Standard 1 (1.5J) must be measured before and at the end of analyzing a batch of samples. The difference between the two results must be less than 20% for each target element. 

	All quantitative sample data were less than the 0.5J standard and, as a result, were less than the target limits for the elemental impurities. Therefore, for the purposes of this study, we chose to report the method validation results for the sample that produced the greatest matrix suppression of the internal standard signal. For this reason, one of the calcium- and magnesium-based over-the-counter tablet formulations was selected because it represented the most difficult analytical challenge out of all the samples examined.

	The accuracy data of this antacid sample is exemplified in Table VI, which shows that the predigestion spike recovery test in the sample matrix passes at all three spike levels (50%, 100%, and 150% of the target limits) with the mean spike recoveries for each target element well within the 70–150% acceptance criteria.

	Six independently prepared samples of the Ca–Mg antacid were digested and then spiked at 100% of the target limit and analyzed. As shown in Table VII, the %RSDs for all target elements were within 2%, which is well under the 20% acceptance limit.

	The six samples used for the repeatability study shown in Table VII were analyzed on two different days. The RSDs for these 12 measurements are all <2.5% (as shown in Table VIII), well below the method requirement of 25%.

	To accept the sample validation data, the instrument drift must be determined by measuring Standard 1 (1.5J) at the beginning and end of the analyses. The difference between the two results for each target element was less than 6%, which is below the acceptance limit of 20%, as demonstrated in Table IX.

	This study has shown that the methodology easily handles these types of high-matrix antacid tablets. The Class 1 and 2A elemental impurity levels were well below the daily dosage PDE limits for 

 in the samples analyzed. In addition, to confirm the method validation, the most difficult antacid sample matrix was chosen and the method used for this work easily passed the acceptance criteria for the testing protocols described in 

		The United States Pharmacopeial Convention online: http://www.usp.org/usp-nf/key-issues/elemental-impurities. 

		International Conference on Harmonization, 

ICH Q3D Step 4 - Guideline for Elemental Impurities

http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q3D/Q3D_Step_4.pdf

		General Chapter <232> “Elemental Impurities in Pharmaceutical Materials - Limits” in second supplement to 

United States Pharmacopeia 39-National Formulary 34

 (United States Pharmacopeial Convention, Rockville, Maryland, 2016); updates published in Pharmacopeial Forum 42(2). 

		General Chapter <233> “Elemental Impurities in Pharmaceutical Materials - Procedures” in second supplement to 

United States Pharmacopeia 38-National Formulary 33

 (United States Pharmacopeial Convention, Rockville, Maryland, 2015).

Guidance for Industry: Elemental Impurities in Drug Products

https://www.fda.gov/downloads/Drugs/Guidances/UCM509432.pdf

		“Thirty-Minute Guide to ICP-MS,” Perkin-Elmer Technical Note, 2017, 

http://www.perkinelmer.com/lab-solutions/resources/docs/TCH-30-Minute-Guide-to-ICP-MS-006355G_01.pdf

 is with PerkinElmer Inc., in Shelton, Connecticut. Direct correspondence to: 

Antacids present a unique set of analytical challenges for ICP-MS. These challenges can be overcome with optimized sample preparation and instrumental analytical conditions.

Determination of Elemental Impurities in Antacids by ICP-MS According to the Validation Protocols Defined in USP Chapters  and  and ICH Q3D Step 4 Guidelines