Introduction

Reference materials serve a critical function in analytical testing by establishing a standardized basis for measurement or value assignment. ISO Guide 30 defines a reference material as a “[m]aterial or substance one or more of whose property values are sufficiently homogeneous and well established to be used for the calibration of an apparatus, the assessment of a measurement method, or for assigning values to materials” [1]. A Certified Reference Material (CRM), according to ISO Guide 34, is a “reference material characterized by a metrologically valid procedure for one or more specified properties, accompanied by a certificate that provides the value of the specified property, its associated uncertainty, and a statement of metrological traceability” [2]. For manufacture of active pharmaceutical ingredients, the US Food and Drug Administration (FDA) defines a primary reference standard as a “substance that has been shown by an extensive set of analytical tests to be authentic material that should be of high purity” [3].

With the FDA’s release of the Draft Guidance on New Dietary Ingredient (NDI) Notifications in July 2011, compliance with dietary supplement cGMPs (current good manufacturing practices) is predicated on the use of CRMs in analytical testing to support identity, purity, strength, quality, and composition requirements [4, 5]. Increased enforcement activities by the FDA from inspections and warning letters underscore the importance of compliance not only in dietary supplement manufacturing but also in quality control testing, labeling, packaging, distribution, and marketing [4, 6]. Since these regulations impart dietary supplement manufacturers with the responsibility of validating product label claims [6], use of an accurate and quantitative reference material is vital to ensure accuracy of results, safety and reliability of the product, and, ultimately, consumer health.

Dietary supplement and food manufacturers utilize reference standards or CRMs in their testing methods for a variety of applications, from confirming accuracy of label claims and identifying potential nutritive value to validating methods, characterizing materials, and identifying adulterants. Manufacturers and testing laboratories, for example, have developed methods using reference standards to confirm concentration levels for vitamins B [7], D [8], and K [9] in nutritional supplements and infant formula. Researchers have used reference materials to analyze biomarkers in plants and herbs as well as to identify their potential pharmacological effect, nutritional value, or antioxidant activity [10–12]. Government laboratories have developed dietary supplement CRMs to establish traceability to a primary reference standard, to characterize quality, or to validate testing methods [13–15].

Quantitation of complex natural products such as phytochemicals and vitamins presents special analytical challenges. Phytochemical reference materials often consist of powders or extracts derived from the plant. They may be enriched in the active ingredient, but may not be highly purified, and may contain high levels of residual solvent or materials left behind after extraction that are not detected by HPLC. The handling and solution preparation of vitamins including A [16, 17], C [18, 19], and B [19, 20] offer significant difficulties due to their sensitivity to air, light, and solution pH. Hygroscopicity of natural products, such as stevia [21], also presents an additional challenge during material handling to minimize potential water absorption that would lead to inaccurate preparations of calibrators and controls.

The case for a solution-based CRM

Solution-based CRMs in a single-use, ampouled format offer a viable alternative to powder or matrix-based reference standards. Analytical chemists often use powdered reference materials by dissolving in solution and preparing volumetrically for use at the bench. Solutions are stored with stoppers for short intervals with no protection against evaporative changes or exposure to air and/or light. The solutions must be discarded and remade frequently, resulting in material waste and disposal cost. Unlike bench-top solutions derived from powders or extracts, the ampouled format prevents changes in concentration from storage over time due to evaporation and eliminates the potential for cross-contamination from repeated use. The inert atmosphere of the sealed amber ampoule promotes multi-year stability by eliminating degradation in solution due to air or light exposure. Increased solution stability minimizes waste and allows for the preparation of larger batch sizes with concomitant larger weighings in the gravimetric preparation, which in turn, provides greater accuracy of the solution concentration [22]. Using the same lot of solution over an extended period of time provides greater long-term consistency of analytical results.

Solution-based CRMs are used as a critical part of analytical testing methods in several industries. Certified ampouled solution reference standards and CRMs are widely used as calibrators and controls in the forensic/toxicology, clinical/diagnostic, and environmental industries [23, 24]. In the forensic and clinical setting, test results are used to support forensic investigations, therapeutic monitoring, and clinical decisions [24]. Pharmaceutical companies and contract research organizations have used solution-based reference standards and CRMs in many applications including research and development, method development and validation, toxicology/safety assessment, and pharmacokinetic/drug metabolism (PKDM) studies. Over the past few years, the pharmaceutical industry has introduced the use of solution-based reference standards for active pharmaceutical ingredient (API) release testing of small-molecule organic and protein-based pharmaceuticals citing significant improvements in efficiency of labor and materials; reduction of variability in the reference from use of a single lot over a longer period of time; and elimination of handling hygroscopic, air-sensitive, or toxic powders [22, 25].

Importance of proper neat material characterization

Dietary supplement and food testing laboratories encounter multiple challenges in natural products testing of botanicals such as Ginkgo biloba, ginseng, green tea, and vitamins. These natural products present significant handling issues due to inherent problems with hygroscopicity and/or sensitivity to air and light. In addition, neat powders are available with varying degrees of certification. It is important to understand how the neat material was certified and what handling controls should be implemented in routine use. Preparation of the reference solution from the neat powder may require reanalysis of water content before use (for hygroscopic compounds) or necessitate use of an inert atmosphere glove box to prevent moisture absorption or degradation.

An ampouled solution-based phytochemical or vitamin CRM eliminates the need for these controls in the testing laboratory. The first step in preparation of the solution CRM is proper characterization of the neat powder and proper handling techniques.

Characterization of the neat reference material comprises assignment of a mass balance purity factor or weighing adjustment. This value should include corrections for chromatographic impurities and other residuals such as solvent, water, or impurities that do not chromatograph by HPLC. The neat material purity factor is important, because the analyte mass used in preparation of a solution reference standard from a neat reference material should be corrected for all impurities, considering the calculated uncertainties from the analytical testing, to ensure accuracy of the solution concentration [24]. This correction factor is especially important for natural products to account for residual solvents and other impurities resulting from extraction or isolation of the botanical using an organic solvent. For example, phytochemical biomarkers from Ginkgo biloba and green tea used in solution-based CRMs were found to contain significant levels of residual solvent, water, or inorganic impurities upon characterization of the neat material (Table 1).

Table 1 Neat material characterization of Ginkgo biloba and green tea phytochemicals
Full size table

Pre-formulation studies: defining diluent and storage specifications

The development of phytochemical and vitamin solution-based CRMs should include proper selection of diluent for solubility and solution stability as well as material handling controls to ensure accuracy of the solution concentration. Diluent selection should also consider compatibility with method diluent in the laboratory’s analytical technique. Pre-formulation studies involve an initial identification of potential diluents compatible with analytical methods used for the compound as well as the analytes’ solubility and stability in those diluents at various storage conditions. In developing solution-based CRMs of riboflavin (vitamin B2) and retinol (vitamin A), pre-formulation studies were essential to select the proper diluent and storage conditions.

Riboflavin

Riboflavin is known to exhibit moderate solubility in water and high solubility, but low stability, in alkaline solutions [26]. Aqueous solutions were evaluated as a potential method-compatible diluent based on literature information and expected use of HPLC as a primary analytical method in vitamin testing [26]. In addition to water alone, additives including formic acid, acetic acid, citric acid, and urea along with dissolution aids such as sonication or heating were examined for potential improvements in solubility during solution preparation (Table 2). Acidic additives did not improve the solubility of riboflavin in water. Mild sonication and heating, on the other hand, were found to aid dissolution of riboflavin up to a concentration of approximately 100 μg/mL. Further studies indicated that concentrations of 10–50 μg/mL are optimal for maintaining solubility and homogeneity.

Table 2 Riboflavin solubility study in several diluents
Full size table

Accelerated stability of the riboflavin solutions in water at 100 μg/mL and 1 % urea in water at 50 μg/mL was evaluated by liquid chromatography/mass spectrometry (LC/MS) over 4 weeks at the following storage temperatures: refrigerator, room temperature, and 40 °C (all storage temperatures and their respective ranges cited throughout this article are defined in USP <797>) [27]. After 4 weeks in water, no change was observed in chromatographic purity of the neat material at all three storage conditions. For the samples stored in 1 % urea, whereas no purity changes were noted after 4 weeks at refrigerator and room temperature conditions, a decrease in purity to 78.5 % was observed after 4 weeks at 40 °C. These results demonstrate the importance of accelerated stability studies at the pre-formulation stage for evaluating potential instability of the phytochemical or vitamin in the chosen diluents. Accelerated stability also establishes suitable conditions for transportation.

Vitamin A

Use of additives in formulation design may be appropriate not only for enhancing solubility of the analyte, but also for promoting analyte stability. Stabilizers are required in retinol formulations containing organic solvents because of retinol’s sensitivity to light and ease of oxidation in air [26, 28]. On the basis of this information, stability of retinol in ethanol was evaluated in accelerated stability studies with and without using BHT (butylated hydroxytoluene), a common food additive and antioxidant stabilizer [28].

Solutions of retinol were prepared under inert conditions at a concentration of 100 μg/mL in ethanol in the presence and absence of BHT. Solutions were maintained at storage conditions of freezer, refrigerator, room temperature, and 40 °C and examined by HPLC [27]. In the absence of BHT, decreases in chromatographic purity after 2 weeks were observed at all four temperature conditions, ranging from 88.9 % at freezer to 70.7 % at 40 °C (Table 3). Improved stability of retinol in ethanol, especially at elevated temperature conditions of room temperature and 40 °C, was observed with the addition of BHT. At 40 °C retinol chromatographic purity decreased just 1.1 % in 4 weeks compared to the initial time point value of 98.7 %, whereas a decrease of only 0.5 % was observed after 4 weeks at room temperature (Table 4).

Table 3 Accelerated stability study of retinol (100 μg/mL) in ethanol without BHT stabilizer
Full size table
Table 4 Accelerated stability study of retinol (100 μg/mL) in ethanol with 0.1 % (w/v) BHT stabilizer
Full size table

Ginsenoside solutions

Pre-formulation studies of multi-analyte solution mixes offer an added level of complexity from potential changes in solubility and stability due to analyte interaction in solution. In addition, the analytical method used to evaluate such mixes must be capable of resolving impurities and detecting such changes. The solubility and stability of a mixture of eight ginsenosides (Rb1, Rb2, Rc, Rd, Re, Rf, Rg1, and Rg2) were assessed at a concentration of 100 μg/mL for each ginsenoside in three diluent systems—acetonitrile/water (80:20), water/methanol (60:40), and water/methanol (60:40) with 5 % 1 M HCl. At 100 μg/mL per analyte, the ginsenoside mix showed limited solubility in the acetonitrile/water diluent and complete solubility in water/methanol (60:40) and water/methanol (60:40) with 5 % 1 M HCl.

Accelerated studies compared the stability of the mix in the neutral versus acidic water/methanol diluents. Over a time period of 4 weeks, the chromatographic purity by HPLC (sum of % peak area) for the eight ginsenosides in the presence and absence of the acid was evaluated at five storage conditions: sub-freezer, freezer, refrigerator, room temperature, and 40 °C [27]. Degradation was observed during preparation of the acidic samples as evidenced by the lower initial purity value of 88.9 % compared to 97.3 % for samples in neutral water/methanol. At storage conditions of refrigerator and higher, rapid degradation after 1 week was observed in the acidic samples with chromatographic purity values ranging from 71.9 % at refrigerator to 9.01 % at 40 °C. Samples in the acidic diluent stored at sub-freezer conditions showed no degradation during the 4-week study (Table 5).

Table 5 Accelerated stability study of ginseng ginsenosides (Rb1, Rb2, Rc, Rd, Re, Rf, Rg1, and Rg2) in water/methanol (60:40) and water/methanol (60:40) with 5 % 1 M HCl
Full size table

In contrast, the samples without acid demonstrated improved stability at refrigerator, freezer, and sub-freezer conditions with decreases in purity of less than 1.5 % observed at each storage condition after 4 weeks. Compared to the 1-week results for the acidic diluent, the samples in neutral water/methanol showed significantly higher chromatographic purity values of 96.2 % and 96.6 % at room temperature and 40 °C, respectively (Table 5). As with the riboflavin example, these results further illustrate the importance of diluent evaluation and accelerated studies not only for single-component solutions of vitamins and phytochemicals, but for multi-component solution mixes as well.

The riboflavin and ginseng ginsenoside examples demonstrate that evaluation of solution pH is critical for proper diluent selection and long-term solution stability. The retinol example, on the other hand, illustrates the importance of minimizing exposure to air during solution preparation and packaging. The reactivity of phytochemicals and vitamins necessitates the use of environmental controls such as handling in a glove box and packaging in flame-sealed, amber ampoules. These process controls ensure that an inert environment is maintained during raw material handling, solution preparation, and packaging to protect the integrity of the compounds.

Establishing long-term shelf life of solution-based CRMs

Shelf life (expiration) of solution-based CRMs in a sealed ampoule format is established through long-term stability studies. Purity and concentration of the reference solution are evaluated at regular intervals up to 5 years for assignment of expiration dates. Long-term studies have established multi-year stability in solution for a diverse group of compounds ranging from opiates, synthetic opioids, and antipsychotics to natural products. Shelf life of 5 years or greater has been achieved for solutions of fentanyl, codeine, or haloperidol in methanol and 6-acetylmorphine in acetonitrile sealed under argon in amber ampoules [29]. Stability studies of anthocyanin solutions in methanol for cyanidin-3-glucoside, petunidin-3-glucoside, and peonidin-3-galactoside demonstrated a shelf life of 4 years with a 2.2 % or less decrease observed in solution purity [29]. Solutions of rebaudioside A, the major steviol glycoside in stevia sweeteners, had a minimum shelf life of 4 years in sealed amber ampoules. Although long-term stability studies in real time offer the most accurate means for determination of shelf life, potential stability issues prior to completion of the long-term studies must be evaluated as well.

Accelerated stability studies are a useful initial tool to estimate shelf life of a solution-based CRM before long-term stability data is available. The accelerated study monitors degradation of the compound in solution at multiple storage conditions over a time period of up to 1 month. Temperature conditions include freezer storage as a control along with four storage temperatures of refrigerator, ambient, 40 °C, and 60 °C [27]. Degradation is monitored as a percent decrease in peak area from the HPLC analysis and is reported in degradation rate curves for the four storage temperatures. Assuming first-order degradation kinetics, an Arrhenius plot [30], determined from degradation at the different storage temperatures, can be used to predict the degradation rate of the solution-based CRM at the recommended storage temperature. These studies make the assumption that rate of degradation is linear at all temperatures studied; however, non-linear degradation at higher temperatures could bias the results.

For solutions of thiamine HCl (vitamin B1) and a mixture of caffeine and seven green tea catechins, degradation rate curves and Arrhenius plots were developed using the decrease in HPLC peak areas at storage temperatures of refrigerator, ambient, 40 °C, and 60 °C (Figs. 1, 2, 3, 4, and 5). Green tea catechins in the mix consisted of (−)-catechin 3-gallate, (−)-epicatechin, (−)-epicatechin 3-gallate, (−)-epigallocatechin 3-gallate, (−)-gallocatechin, (−)-gallocatechin 3-gallate, and (+)-catechin. The model predictions from the Arrhenius plots for thiamine HCl and (−)-epicatechin estimate a shelf life of 4 years for each solution-based CRM in freezer storage. These examples illustrate the importance of accelerated stability studies not only for estimating shelf life of a solution-based CRM but also for understanding the interaction of analytes in a mixture as it pertains to overall solution stability. A combination of accelerated predictive stability and real-time long-term stability can be used effectively to develop solution-based CRMs.

Fig. 1
figure 1

Degradation rate curves of (−)-epicatechin at four storage conditions. Results at 60 °C indicate that (−)-epicatechin is unstable at elevated temperatures and are not necessarily indicative of long-term stability. The % decrease in peak area for the compound increases at elevated temperatures, indicating a higher percentage of degradation

Full size image
Fig. 2
figure 2

Arrhenius plot of (−)-epicatechin at four storage conditions

Full size image
Fig. 3
figure 3

Predictive stability study showing chromatograms of green tea catechin mix (caffeine, (−)-catechin 3-gallate, (−)-epicatechin, (−)-epicatechin 3-gallate, (−)-epigallocatechin 3-gallate, (−)-gallocatechin, (−)-gallocatechin 3-gallate, and (+)-catechin) after 4 weeks at each storage condition

Full size image
Fig. 4
figure 4

Degradation rate curves of thiamine HCl at four storage conditions. The % decrease in peak area for the compound increases at elevated temperatures, indicating a higher percentage of degradation

Full size image
Fig. 5
figure 5

Arrhenius plot of thiamine HCl at four storage conditions

Full size image

Concluding remarks

Phytochemicals and vitamins offer unique challenges as CRMs for analytical testing methods. These challenges—from sensitivity to air, light, solution pH, and temperature to extracts with high levels of residual solvent and impurities that do not chromatograph—increase the difficulty in providing accurate, consistent, and stable reference standards. Solution-based CRMs in a single-use, ampouled format offer a viable alternative to powder or matrix-based reference standards when appropriate parameters are chosen in the design, development, preparation, packaging, and storage of the solution standard. With defined specifications as the result of full neat material characterization, pre-formulation studies, and evaluation of long-term stability, solution-based CRMs can provide the level of accuracy and consistency required for the changing regulatory requirements in the dietary supplement and food industries.