# Total Polyphenols (TPP)

> Total polyphenol profiling for black tea - the leading proxy for antioxidant quality and a primary auction pricing signal.

- Canonical URL: https://spectrify.ai/research-pipeline/total-polyphenols-tpp

- Sector: Sector 01 · Food & Agriculture
- Status: Model Ready
- Tags: NIR, Tea, Polyphenols, Auction, PLS-R

Total polyphenol content (TPP) is the most commercially significant quality parameter for black tea at the Colombo auction. High TPP correlates with stronger antioxidant activity, more intense liquor colour, and consistently commands higher lot prices. Our NIR model delivers TPP estimates within seconds from a dry tea powder scan - replacing multi-hour wet chemistry workflows that are impractical at auction or factory pace.

## What We've Established

### Spectral Correlation with Polyphenol Concentration

The aromatic ring system of polyphenolic compounds - catechins, theaflavins, and thearubigins - produces characteristic near-infrared absorption patterns. Combined second-overtone C–H and O–H band interactions provide a discriminative spectral window for TPP quantification across the 4,000–10,000 cm⁻¹ region.

### Calibration Against Folin-Ciocalteu Reference

Reference TPP values were generated using the standard Folin-Ciocalteu colorimetric assay on a curated set of black tea samples spanning 35+ Ceylon tea grades sourced from multiple factories. Cross-validation R² exceeded 0.93 on the calibration set, with RMSECV within acceptable limits for grade-level differentiation.

### Grade Differentiation Capability

The model reliably separates high-grade and standard-grade auction lots by TPP band with accuracy sufficient for practical price guidance - the primary use case identified in early client validation sessions.

## Technical Approach

Black tea powder samples were prepared from authenticated factory-grade lots and scanned using a benchtop NIR spectrometer in diffuse reflectance mode. Spectral preprocessing included multiplicative scatter correction (MSC) and mean centring. PLS-R was applied with optimal latent variables selected via cross-validation. Reference TPP was measured by an accredited external laboratory using the Folin-Ciocalteu method to ISO 14502-1. A held-out validation set from a separate factory harvest was used for final model assessment.

## Where We Stand

The TPP model is complete and validated. We are currently coordinating integration with factory quality management workflows and preparing a summary report for review by industry quality bodies. The model is available for pilot deployment with existing clients who are already using the moisture detection system.

## Next Steps

- Complete review with industry quality management stakeholders
- Integrate TPP output into the multi-parameter scan report alongside moisture and sugar
- Evaluate cross-factory model transferability without recalibration
- Begin planning for the next-level polyphenol fractionation models: Catechins, TF, TR

## Related Projects

- [Moisture Content - Green Tea](https://spectrify.ai/research-pipeline/moisture-content-green-tea) (Model Ready): Real-time MC measurement for green tea. Model trained on TRI-standard data and independently validated in December 2025.
- [Sugar in Tea](https://spectrify.ai/research-pipeline/sugar-in-tea) (Model Ready): Total sugar quantification across black tea grades - a key differentiator for premium blend specification and export documentation.
- [Caffeine Content](https://spectrify.ai/research-pipeline/caffeine-content) (Lite Model Ready): Rapid caffeine screening for black tea blends - lite model delivered, enabling blend certification and label compliance workflows.

[Research Pipeline](https://spectrify.ai/research-pipeline)