Smartphone TLC Fingerprinting of Caffeine in Drinks

ISEF Category: Chemistry

Subcategory: Analytical Chemistry  ·  Difficulty: Intermediate  ·  Setup: School Lab  ·  Time: 1 to 2 Months

The Hook

Your morning drink has a chemical fingerprint. Tea, coffee, and chocolate all carry different mixes of caffeine, theobromine, and theophylline. With TLC and a phone camera, you can turn those mixes into real data. You can also test how brewing changes the result.

What Is It?

This project uses TLC, which stands for thin-layer chromatography. Think of TLC like a race on sandpaper. You place a tiny sample near the bottom of a coated plate, then a solvent climbs upward and carries each compound at a different speed. Compounds that stick more to the plate move less. Compounds that like the solvent move farther.

Your goal is to compare the relative amounts of caffeine, theobromine, and theophylline in drinks and foods. These three molecules are close cousins, but not identical. Caffeine gives the familiar stimulant effect in coffee and tea. Theobromine shows up more in chocolate. Theophylline appears in smaller amounts in some plant sources and can act as a useful marker in mixed samples.

Smartphone densitometry means you use your phone camera to measure how dark each spot looks on the TLC plate. Darker spots usually mean more compound, as long as you keep lighting and camera settings steady. Then you can compare samples and build a standard curve from known references. That lets you move from a picture to numbers.

Why This Is a Good Topic

This makes a strong science fair topic because you can ask clear, testable questions and measure real chemical differences with accessible tools. It connects to a real problem, food labeling, brewing consistency, and how processing changes what ends up in a cup or bar. You can learn chromatography, calibration, image analysis, and basic kinetic modeling without needing a professional lab setup.

Research Questions

• How does brewing time change the TLC-detected caffeine-to-theobromine ratio in tea samples?
• What is the effect of water temperature on the extraction of caffeine from ground coffee?
• Does the type of tea leaf, such as black, green, or oolong, change the relative TLC spot intensity of caffeine-related compounds?
• To what extent does chocolate cocoa percentage affect the theobromine signal measured by smartphone densitometry?
• Which brewing condition produces the fastest rise in measured caffeine signal for a fixed sample mass?
• How does sample dilution affect the linearity of smartphone densitometry for caffeine standards?

Basic Materials

• TLC plates with silica gel coating.
• Capillary spotting tubes or micropipettes.
• Caffeine standard, and if available, theobromine and theophylline standards.
• Clear TLC developing chamber with lid.
• Solvent system appropriate for caffeine-like compounds, chosen from literature.
• UV lamp or iodine chamber for spot visualization, if your school allows it.
• Smartphone with a steady camera app that can lock exposure and focus.
• Ruler or printed calibration scale.
• White light box or consistent desk lamp.
• Filter paper or lint-free wipes.
• Tea bags, coffee grounds, cocoa powder, or chocolate samples.
• Digital kitchen scale with 0.1 g accuracy.
• Glass vials or small beakers.
• Gloves, goggles, and lab coat.

Advanced Materials

• Analytical balance.
• TLC plates with a fluorescence indicator, if needed for visualization.
• Reference standards of caffeine, theobromine, and theophylline.
• UV-Vis spectrophotometer for cross-checking smartphone measurements.
• HPLC access for validation, if available.
• Laboratory oven or drying setup for sample prep consistency.
• Vortex mixer or sonicator for extract preparation.
• Micropipettes with calibrated tips.
• Image analysis workstation.
• Reagents for method development and validation, selected from literature.
• Temperature-controlled water bath or hot plate with stirring for extraction studies.
• pH meter for testing matrix effects.

Software & Tools

• ImageJ: Measures spot intensity from TLC photos and helps you compare samples against standards.
• Python: Organizes image data, fits calibration curves, and tests extraction trends.
• NIH ImageJ macros: Automates repeated densitometry measurements across many TLC plates.
• Google Sheets: Tracks sample metadata, replicates, and basic statistics.
• R: Fits regression models and checks whether brewing variables predict signal changes.

Experiment Steps

1. Define the chemical pair or trio you will compare, then decide whether you are measuring only caffeine or a fuller caffeine-family fingerprint.
2. Select a TLC solvent system and visualization method from published sources, then plan a pilot run to check that the spots separate cleanly.
3. Build a standard curve from known compounds so your phone images can become concentration or signal estimates.
4. Choose one brewing variable to test first, such as time, temperature, or sample mass, and keep the others fixed.
5. Design controls for lighting, camera settings, plate handling, and sample spotting so image intensity stays comparable across runs.
6. Plan the extraction model you will fit, then decide what data shape would support a diffusion-limited or first-order trend.

Common Pitfalls

• Using different lighting for each plate photo, which makes spot intensity look higher or lower for the wrong reason.
• Overloading the TLC plate with too much sample, which causes smeared spots and ruins separation.
• Choosing a solvent system that puts caffeine, theobromine, and theophylline too close together to measure cleanly.
• Comparing raw phone brightness instead of calibrated spot density, which makes the data depend on camera auto-settings.
• Ignoring matrix effects from tea, coffee, or chocolate extracts, which can shift migration distance and distort the fingerprint.

What Makes This Competitive

A stronger version of this project goes beyond simple spot comparison. You would validate your densitometry against standards, control lighting tightly, and test whether the brewing data fit a real extraction model. You can also compare multiple sample classes, not just one drink, and ask whether the same model works across them. Careful statistics and repeat measurements can turn this from a nice demo into a serious analytical study.

Project Variations

• Compare loose-leaf tea, bagged tea, and instant tea to see whether processing changes the caffeine-family fingerprint.
• Test milk chocolate, dark chocolate, and cocoa powder to compare theobromine signal across cocoa-rich foods.
• Swap smartphone densitometry for UV-Vis confirmation on the same extracts to test how well the TLC method agrees with a lab instrument.

Learn More

• PubMed: Search review articles on TLC analysis of caffeine, theobromine, and theophylline in foods and beverages.
• NIH PubChem: Look up compound properties, structures, and related literature links for caffeine, theobromine, and theophylline.
• NCBI Bookshelf: Read free textbook chapters on analytical chemistry, chromatography, and calibration.
• MIT OpenCourseWare: Find free lecture notes on separation methods and analytical chemistry fundamentals.
• Journal of Chromatography A: Search article abstracts on thin-layer chromatography methods for alkaloids and beverage analysis.
• USDA FoodData Central: Check food composition data for cocoa products, tea, and coffee matrices.