Determination of Caffeine Content in Beverages by Liquid Chromatography

Caffeine is a natural methylxanthine alkaloid widely present in raw materials such as coffee beans, tea leaves, and cacao beans. It serves as a core functional additive and characteristic flavor compound in commercially available beverages, including carbonated drinks, energy drinks, tea-based beverages, and coffee drinks. Moderate intake of caffeine can exert physiological effects such as refreshing the mind and alleviating fatigue. However, excessive and long-term consumption may readily induce adverse reactions, including palpitations, insomnia, and nervous hyperexcitability. Moreover, susceptible populations—such as pregnant women, children, and individuals with cardiovascular diseases—have even lower tolerance thresholds. Consequently, food safety standards in various countries have established clear regulatory limits on caffeine content in beverages.

In China, the national food safety standard titled "Determination of Caffeine in Beverages" (GB 5009.139-2014) serves as the statutory basis for the quantitative determination of caffeine in beverage matrices. High-performance liquid chromatography (HPLC), owing to its excellent resolution, strong capability to resist matrix interference, and high quantitative accuracy, has become the method of choice for the analysis of complex-matrix samples, including various compound beverages, milk-containing beverages, and herbal functional drinks. This experiment was conducted in accordance with the high-performance liquid chromatography method specified in the national standard GB 5009.139-2014, establishing a sample pretreatment and liquid chromatographic separation/detection protocol for beverage samples, and accomplishing the qualitative identification and quantitative analysis of caffeine in different types of beverages.

Keywords: high-performance liquid chromatography; caffeine; beverages; content determination.

1. Instrument and Reagents

1.1 Liquid Chromatography System Configuration List

Table 1. Instrument Configuration List

1.2 Reagents and Standards List

Table 2. Reagents and Standards List

1.3 Experimental Materials and Auxiliary Equipment

Ultrasonic cleaner;

Vortex mixer;

0.45μm microporous aqueous-phase filter membrane.

2. Experiment Method

2.1 Preparation of Standard Solutions

2.1.1 Caffeine Standard Stock Solution (2.0mg/mL): Accurately weigh 20mg of caffeine reference standard (to the nearest 0.1mg) into a 10mL volumetric flask, dissolve and make up to volume with methanol. Store in a refrigerator at 4 °C. The shelf life is 6 months.

2.1.2 Caffeine Standard Intermediate Solution (200μg/mL): Accurately transfer 5.0mL of the caffeine standard stock solution (2.1.1) into a 50mL volumetric flask, and make up to volume with water. Store in a refrigerator at 4 °C. The shelf life is 1 month.

2.1.3 Caffeine Standard Working Solutions for Calibration Curve: Transfer 0.5 mL, 1.0 mL, 2.0 mL, 5.0 mL, and 10.0 mL of the caffeine standard intermediate solution (2.1.2) into separate 10 mL volumetric flasks, and make up to volume with water. The resulting concentrations of this standard series are 10.0 μg/mL, 20.0 μg/mL, 40.0 μg/mL, 100 μg/mL, and 200 μg/mL, respectively. Prepare immediately before use.

2.2 Sample Pretreatment

2.2.1 Cola-Type Beverages

(a) Degassing: Sonicate the sample using an ultrasonic cleaner at 40 °C for 5min.

(b) Purification: Weigh 5g of the sample (accurate to 0.001g), transfer to a 5mL volumetric flask, and make up to volume with water (ensuring that the caffeine content in the sample solution falls within the calibration curve range). Mix thoroughly. Add 0.5g of magnesium oxide, shake well, and allow to stand. Take the supernatant and filter it through a microporous membrane for subsequent analysis.

2.2.2 Milk-Containing Liquid Coffee and Tea Products

Weigh 1g of the sample (accurate to 0.001g), transfer to a 10mL volumetric flask, and make up to volume with trichloroacetic acid solution (ensuring that the caffeine content in the sample solution falls within the calibration curve range). Mix thoroughly and allow to stand for protein precipitation. Take the supernatant and filter it through a microporous membrane for subsequent analysis.

2.3 Experiment Conditions

2.3.1 Chromatographic Conditions

Column: C18, 4.6 × 250mm, 5μm

Mobile phase: Methanol/water = 24/76 (v/v)

Flow rate: 1.0 mL/min

Column temperature: 25 °C

Detection wavelength: 272nm

Injection volume: 10μL

3. Experiment Result

3.1 Blank

Fig 1. Chromatogram of Blank Solvent

3.2 Linear Range

Fig 2. Linearity Results of Caffeine Reference Standard

Table 3. Linearity Range of Caffeine Reference Standard

3.3 Repeatability

Fig 3. Repeatability Chromatograms of Caffeine Reference Standard (n = 6)

Table 4. Repeatability Chromatographic Parameters for Caffeine Reference Standard (n = 6)

3.4 Sample

Fig 4. Chromatogram of a Certain Brand of Cola Beverage Sample

Table 5. Chromatographic Results for a Certain Brand of Cola Beverage Sample

Fig 5. Chromatogram of a Certain Brand of Milk Tea Beverage Sample

Table 6. Chromatographic Results for a Certain Brand of Milk Tea Beverage Sample

3.5 Spiking Recovery

Fig 6. Chromatogram of Spiking Recovery for a Certain Brand of Cola Beverage Sample (Spiking Concentration: 40μg/mL)

Table 7. Chromatographic Results for Spiking Recovery in a Certain Brand of Cola Beverage Sample (Spiking Concentration: 40μg/mL)

Fig 7. Chromatogram of Spiking Recovery for a Certain Brand of Milk Tea Beverage Sample (Spiking Concentration: 40μg/mL)

Table 8. Chromatographic Results for Spiking Recovery in a Certain Brand of Milk Tea Beverage Sample (Spiking Concentration: 40μg/mL)

3.6 Limit of Detection

Fig 8. Chromatogram for Limit of Detection of a Certain Brand of Cola Beverage Sample

Table 9. Chromatographic Results for Limit of Detection of a Certain Brand of Cola Beverage Sample

Fig 9. Chromatogram for Limit of Detection of a Certain Brand of Milk Tea Beverage Sample

Table 10. Chromatographic Results for Limit of Detection of a Certain Brand of Milk Tea Beverage Sample

4. Conclusion

This experiment was conducted in accordance with GB 5009.139-2014 "National Food Safety Standard — Determination of Caffeine in Beverages," using the Wayeal LC3500 liquid chromatography system equipped with a diode array detector (DAD) for the determination of caffeine in beverage samples. The experimental data demonstrated that the target peak exhibited a symmetrical shape with high theoretical plate numbers. The calibration curve for the standard solution series over the concentration range of 10.0μg/mL to 200.0μg/mL showed a linear correlation coefficient greater than 0.9999. The method exhibited excellent repeatability, with RSD values for both retention time and peak area below 0.3%, indicating good instrument precision. In this analysis, two beverage matrices—cola and milk tea—were tested. Their caffeine contents were determined to be 95.051mg/kg and 260.0456mg/kg, respectively. The spiking recoveries were above 94%, indicating high recovery efficiency, negligible sample loss, and a stable and reliable pretreatment procedure. The limits of detection (LODs) were 0.0446mg/kg for cola-type beverages and 0.75mg/kg for milk-containing tea-based beverages, reflecting the high sensitivity and low baseline noise of the instrument.