Analysis of Organic Acids using a Mixed-Mode LC Column and an ACQUITY QDa Mass Detector

Standard Preparation

Individual OAs were dissolved in MilliQ water at 1.0% (w/v) to make stock solutions. Fumaric acid was dissolved in absolute ethanol at 1.0% (w/v). A solution containing all OA, each at 600 μg/mL (record 3 significant figures) was prepared by mixing individual OA stock solutions together and diluting with MilliQ water. Lower concentration solutions were prepared by serial dilution with water from this OA mix stock solution.

Sample Preparation

Samples of fruit juices, including apple, grape, and pomegranate, were purchased from local stores. These juices were diluted with MilliQ water at a 1 to 10 ratio (1 mL juice mixed with 9 mL water) and filtered with a 0.45 μm glass microfiber membrane filter (GMF). The filtered juice solutions were further diluted with water as needed.

Optimization of Chromatography

Mobile phase composition 

The Atlantis PREMIER BEH C₁₈ AX Column is a mixed-mode column. Its stationary phase contains not only C₁₈ groups, but also tertiary alkylamine moieties, which become positively charged below approximately pH 8.⁽³⁾ It was found that the mobile phase composition (organic solvent content) had a significant impact on the retention of OA. Figure 2 shows capacity factors (k’) of OA in isocratic elution with different mobile phase compositions. In water with 50 mM ammonium formate and 0.9% formic acid, the organic acids had k’ values from 0.4 to 2.5. When the organic content (mobile phase B, acetonitrile with 0.9% formic acid) was increased to about 50%, these acids exhibited the least retention on the column and all k’ values were less than 0.5. When the organic content was further increased to high organic content, the OA had high k’ values, again. The elution order of OA was also altered in different mobile phase compositions. 

Mobile phase buffer concentration

The buffer concentration is another major factor in method optimization for OA analysis. As shown in Figure 3, it affects both the retention and the selectivity of organic acids. When the concentration of ammonium formate was decreased from 50 to 0 mM, the k’ of citric acid and malonic acid were significantly increased, while other OA had moderate increases in k’. This significant increase in k’ for citric and malonic acids is related to their low pK_a values compared to other OA’s in Fig. 3. The elution order of OA was also altered when the buffer concentration was changed. 

pH of mobile phase

The pH of the mobile phase is also an important factor in method development. The pH of the mobile phase affects the acid dissociation equilibrium of OA, which affects their retention and peak shape. pH 2.9 was used in this study to obtain good peak shape. 

Analysis of organic acids

A gradient elution program has been developed and used for the OA analysis (see the Experimental section). It is essentially a binary solvent gradient with an additional segment of isocratic elution with water containing no buffer (low ionic strength). With the isocratic elution with water containing no buffer followed by a gradient elution, we obtained greater retention for the early eluting OA and good resolution for the important quinic acid, tartaric acid, and malic acid.

Chromatograms of 15 organic acids were collected in 11 Single Ion Recording (SIR) channels (Figure 4). Most molecular ions [M-H]^- of these OA have different mass to charge ratios (m/z). They were detected in separate SIR channels, which eliminated the possible interference from each other. The RT, limit of quantification (LOQ), and calibration results of 15 organic acids are in Table 2. The LOQ of organic acids in solvent solution were in the range of 0.2 to 6 ppm (μg/mL in solution). 

The accuracy of quantitation of organic acids was evaluated in a spiking study. Four important organic acids, quinic acid, tartaric acid, malic acid, and citric acid, were spiked at two levels, 0.1 mg/mL and 4 mg/mL, in apple, grape, and pomegranate juices. The recoveries, following AOAC guidelines ,⁽⁴⁾ were calculated and shown in Table 3. The majority of the recoveries were within the 80-120% range. External calibration was used in the spiking study. However, it is recommended to use internal standards if more accurate results are needed. Using internal standard is a common practice to improve the accuracy and precision in analyses, especially when mass spectrometry is involved. 

The main benefits of this method for OA analysis can be highlighted in pomegranate fruit juice analysis. Figure 5 shows an overlay of pomegranate juice chromatograms with an insert of a chromatogram overlay at enlarged scale. A tartaric acid content as low as 0.007 mg/mL in pomegranate juice was determined (see Table 3). Please note the flat baseline in the tartaric acid SIR chromatogram, and how many peaks are present near the tartaric acid peak (RT 1.5 min) in Figure 5. For non-selective detectors, it would be very difficult to quantify the tartaric acid at such a low level with many peaks eluting so closely to it. Tartaric acid was used as a marker for adulterated pomegranate juice. The level of tartaric acid in adulterated pomegranate juice was found to be 0.07 mg/mL or higher.⁽⁵⁾ Sensitive detection of OA, such as tartaric acid for pomegranate juice in this case, is necessary for fruit juice authenticity testing.

The analysis of OA can be a challenging task given the difficulty of separating the large number of naturally occurring OA that could be present in samples. Waters UPLC/QDa System with an Atlantis PREMIER BEH C₁₈ AX Column offers an excellent approach for OA analysis. The key advantages of this OA analysis approach include:

• Greater retention and improved resolution
• Highly sensitive detection that is beneficial in certain applications
• Highly selective detection that makes the analysis less prone to interference from co-eluting compounds in the sample matrix
• Fast analysis with run time less than 8 minutes

1. Monica Gonzalez and Venerando Gonzalez. Organic Acids in Food Analysis by HPLC; 3^rd ed. Leo M.L. Nollet, Fidel Toldra., Eds; CRC Press Taylor & Francis Group, Boca Raton, FL, 2013, pp 443.
2. Hyoung S. Lee, HPLC Method for Separation and Determination of Non-Volatile Organic Acids in Orange Juice, J. Agric. Food Chem. 1993, 41, 1991–1993.
3. Thomas H. Walter, Bonnie A. Alden, Cheryl Boissel, Jessica Field, Nicole Lawrence, Donna Osterman, and Amit Patel, A New Mixed-Mode Reversed-Phase/Anion-Exchange Stationary Phase Based on Hybrid Particles, Waters Application Note, Waters Corporation, 720006742en, 2020.
4. AOAC Guidelines for Single-Laboratory Validation of Chemical Methods for Dietary Supplements and Botanicals (2012) Official Methods of Analysis, 19^th Ed., Appendix K, AOAC INTERNATIONAL, Rockville, MD, USA. http://www.eoma.aoac.org/app_k.pdf.
5. Stefan Ehling and Shannon Cole, Analysis of Organic Acids in Fruit Juices by LC-MS: An Enhanced Tool for Authenticity Testing, J. Agric. Food Chem. 2011, 59, 2229–2234.