LimsPulse
Beverages · Cola and energy drinks, ready-to-drink tea and coffee, brewed coffee

Caffeine in beverages by HPLC

Separates caffeine from the drink matrix on a C18 column and measures it by UV absorbance at about 273 nm against an external caffeine standard. Used to check label declarations and 'high caffeine content' labelling.

  • Food & beverage
  • HPLC (reversed phase, UV detection at 273 nm)
  • HPLC
  • Analytical balance
  • Sonicator
  • Micropipette

Reference: Typical method based on AOAC liquid chromatographic methods for caffeine in soft drinks and ISO 20481 for coffee products. Follow your validated method.

PRINCIPLE

Why this method works

Caffeine is a moderately polar purine alkaloid with a strong UV maximum near 273 nm. On a C18 column with a water–methanol (or water–acetonitrile) mobile phase, slightly acidified, it elutes as a sharp peak well separated from sugars, acids and colours, which elute near the void. The sample peak area is compared with a caffeine standard of known concentration, or with a calibration line, after allowing for the dilution.

Instruments used

MATERIALS

Reagents & standards

ReagentGrade / specification
Caffeine reference standardCertified purity (e.g. ≥ 99%), dried as stated on the certificate
MethanolHPLC grade
Acetic acid or phosphoric acidAR grade, to acidify the mobile phase
WaterHPLC grade
SAMPLE PREPARATION

Preparing the sample

  1. 01

    Degas

    Sonicate carbonated drinks about 10 minutes.

  2. 02

    Pipette

    Pipette 5.0 mL into a 25 mL flask.

  3. 03

    Dilute

    Make up with mobile phase and mix.

  4. 04

    Filter

    Filter through 0.45 µm, discard the first mL.

  5. 05

    Inject

    Inject alongside the caffeine standard.

Procedure

  1. 1Degas carbonated drinks in an ultrasonic bath for about 10 minutes, or until no more bubbles form.
  2. 2Pipette 5.0 mL of the degassed drink into a 25 mL volumetric flask and make up to volume with mobile phase (dilution factor 5). Energy drinks and coffee may need a higher dilution (e.g. 2 mL to 50 mL) to stay within the calibration range.
  3. 3For brewed tea or coffee, cool to room temperature and filter before dilution. For milky drinks, precipitate proteins with Carrez reagents and filter.
  4. 4Filter through a 0.45 µm syringe filter (nylon or PTFE), discarding the first 1 mL, into an HPLC vial.
  5. 5Prepare two independent sample preparations.

Standard preparation

  1. 1Stock: accurately weigh about 50 mg caffeine standard into a 50 mL volumetric flask, dissolve and make up with mobile phase (about 1000 µg/mL).
  2. 2Working standard: dilute 1.0 mL of stock to 50 mL with mobile phase (about 20 µg/mL).
  3. 3For a calibration line, prepare 5 levels covering about 5–50 µg/mL from the stock.

Notes for analysts

  • The chromatogram is illustrative; retention time depends on the column and the exact methanol content.
  • Benzoic acid from preservatives can elute near caffeine in some conditions; adjust the organic content or pH to keep them apart, or quantify both together (see the benzoate–sorbate method).
  • Tea and cocoa drinks also contain theobromine and theophylline, which elute before caffeine on C18.
TEST CONDITIONS

Method parameters (typical)

ColumnC18, 150 × 4.6 mm, 5 µm
Mobile phaseWater containing 0.1–1% acetic acid : methanol, about 75 : 25 v/v, isocratic
Flow rate1.0 mL/min
DetectionUV at 273 nm (DAD 200–400 nm useful for peak purity)
Injection volume10–20 µL
Column temperature30 °C
Run timeAbout 10 minutes (caffeine at roughly 5 minutes)
0100200300400012345678910Unretained matrix (sugars, acids, colour)tR 1.4 minCaffeinetR 5.1 minTime (minutes)mAU (273 nm)
Illustrative chromatogram — your retention time depends on the column and conditions.
BEFORE YOU REPORT

System suitability & sequence

ParameterLimit
%RSD of caffeine area (5 standard injections)Not more than 2.0%
Tailing factorNot more than 2.0
ResolutionCaffeine resolved from the nearest matrix peak (e.g. benzoic acid, theobromine) by at least 1.5

Typical sequence

  1. 1Blank (mobile phase) × 1
  2. 2Standard × 5
  3. 3Sample 1 × 2
  4. 4Sample 2 × 2
  5. 5Bracketing standard × 1
CALCULATION

Calculation with a worked example

Formula
Caffeine (mg/100 mL) = (Aₛₐ / Aₛₜ) × Cₛₜ × DF × 100 / 1000

Where

Aₛₐ
Mean caffeine peak area in the sample solution
Aₛₜ
Mean caffeine peak area in the standard solution
Cₛₜ
Caffeine concentration in the standard (µg/mL), corrected for purity
DF
Dilution factor of the sample (25 / 5 = 5)

Given

Sample
Cola drink, 5.0 mL to 25 mL
Aₛₐ
412,350
Aₛₜ
430,100
Cₛₜ
20.0 µg/mL
DF
5

Working

  1. 1

    Caffeine in sample solution = (412,350 / 430,100) × 20.0 = 19.17 µg/mL

  2. 2

    Caffeine in drink = 19.17 × 5 = 95.87 µg/mL = 95.87 mg/L

  3. 3

    Per 100 mL = 95.87 × 100 / 1000 = 9.59 mg/100 mL

Caffeine = 9.6 mg/100 mL (about 96 mg/L, or about 32 mg per 330 mL can)

ACCEPTANCE CRITERIA

What result passes

  • Compare with the label declaration and product specification. Typical values: cola about 8–12 mg/100 mL, energy drinks about 30–32 mg/100 mL, brewed coffee roughly 40–80 mg/100 mL.
  • Some regulations (for example in the EU) require a 'high caffeine content' statement when a drink contains more than 150 mg/L; check the current rules in your market.
METHOD VALIDATION

Validation parameters (ICH Q2(R2))

ParameterTypical requirement
SpecificityNo interfering peak at caffeine RT in a caffeine-free version of the drink; DAD peak purity passes
LinearityAbout 5–50 µg/mL, r ≥ 0.999
AccuracySpike recovery typically 95–105%
PrecisionRepeatability %RSD typically ≤ 2%
LOQWell below 1 mg/100 mL in the drink (low µg/mL in solution)
In LimsPulse · Sample Management

Run this method in LimsPulse: specifications and STPs under version control, results calculated and checked against limits, and a signed Certificate of Analysis.

See how