A simple, rapid, and robust ultrahigh-pressure liquid chromatography (UHPLC) method for the simultaneous determination of
polyphenols, methylxanthines, sweeteners, and flavouring substances as well as some common preservatives has been developed
using an automated method scouting and method optimization workflow. The most suitable mobile phase and stationary phase combination
was identified in an overnight scouting run. These conditions were used to create a two-dimensional model using computer simulation
software. Temperature and gradient time were varied to establish the best separation conditions. This approach resulted in
a 1.5 min gradient method that could screen 17 compounds of interest.
Chocolate is one of the most popular indulgences available. As well as the well-known negative effects of any high-fat, high-sugar
food product, there is also evidence of health benefits related to chocolate consumption. These positive effects, such as
lowered blood pressure, antioxidant properties, reduction in the risk of cardiovascular disorders, and an increase in cognitive
abilities, are all attributed to the flavonoid content, which is only relevant in dark chocolate products (≥ 40% cocoa) (Flavonoids
can also be found in milk chocolate but in a very low concentration).
PHOTO CREDIT: MALERAPASO/GETTY IMAGES
However, a high cocoa and a high flavonoid content comes with an even higher amount of less desirable alkaloids, namely the
stimulants caffeine and theobromine. In addition to the natural products found in cocoa beans, a large number of permitted
additives can be added to chocolate products to alter the taste, texture, or expiry date, such as artificial or natural sweeteners,
and flavouring compounds or preservatives. To monitor the composition and therefore the quality of cocoa in the food product,
an analytical method needs to enable individual quantification of any of these potential ingredients.
This article describes the development and optimization of a rapid ultrahigh-pressure liquid chromatography (UHPLC) screening
test for the separation and quantification of polyphenols, methylxanthines, sweeteners, and flavouring substances as well
as some common preservatives for the quality control of dark chocolate products.
Equipment and Chromatographic Methods: For UHPLC method scouting, a Shimadzu Nexera X2 Method Scouting System was used, consisting of two quaternary solvent pumps,
autosampler, and column oven, including a six-column switching valve. The system was also equipped with a high resolution
UV photo diode array (PDA) detector and an LCMS-8040 triple quadrupole mass spectrogmeter via an electrospray ionization (ESI)
Method scouting was performed in an overnight sequence using 5 min gradient runs at 40 °C. Five combinations of stationary
phase and mobile phase were selected. The columns comprised a variety of bonded phases that exploit multiple mechanisms of
separation and therefore provide alternative selectivity to common C18 phases. The aqueous and organic mobile phase combinations
contained the same buffer salt/acid (AA/BA; AB/BB; and AC/BC).
After method scouting a column with a polar embedded group for enhanced polar selectivity was found to offer the best separation
from the columns selected with all analytes of interest at least partly separated using a mobile phase consisting of A: 10
mM ammonium formate in H2O (pH 2.8) and B: 10 mM ammonium formate (pH 2.8) in MeCN/H2O (90:10 v/v). These conditions were used to create a two-dimensional DryLab model (Molnár Institute) using 2 min and 6 min
gradient runs at 25 °C and 50 °C as input data. These experiments resulted in a colour-coded resolution map for simple identification
of the optimum separation conditions (Figure 1).
Figure 1: Colour-coded resolution map for UHPLC method development.
The software predicted an optimum separation with a minimum resolution of the critical peak pair of 1.6 in a gradient run
from 5% to 85% B in 1.5 min at a flow rate of 1.2 mL/min at 42 °C.
To obtain the best possible resolution the column was packed with 2-µm particles instead of the 3-µm particle size tested
initially. The selectivity of the column was identical and the pressure increase was not higher than predicted by the software
so the change in particle size did not cause any problem for the separation. A comparison of the predicted chromatogram with
an actual sample run is displayed in Figure 2.
Figure 2: Comparison of the predicted and actual chromatogram of the UHPLC analysis of alkaloids and permitted additives in
dark chocolate products.
Sample Preparation: The chocolate sample was defatted and the compounds extracted into a sample solvent suitable for high performance liquid chromatography
A 2 g sample of dark chocolate (≥ 63% cocoa) was spiked with approximately 100 µg of standard substances before being defatted
twice with 15 mL of n-hexane for 10 min in an ultrasonic bath at 30 °C. The mixture was then centrifuged for 10 min at 6000
rpm. The supernatant with the dissolved fat was discarded and the precipitate was air-dried and reconstituted in a mixture
of acetone, water, and acetic acid (70:29.8:0.2 v/v/v) and sonicated for another 15 min at 30 °C. The sample was then filtered
and heated to 40 °C to evaporate the organic solvent. The remaining aqueous fraction was filtered again and used for UHPLC
Figure 3: UHPLC analysis of the extracted chocolate sample in a 1.5 min gradient with PDA detection. Column: ACE Excel 2 C18-Amide,
100 × 2.1 mm, mobile phase: 10 mM HCOONH4, pH 2.8 in A: H2O and B: MeCN/H2O (90:10 v/v), 1.2 mL/min flow, 42 °C.