Sourcing Decaf Green Tea Extract Below 0.1% Caffeine
The solvent-by-solvent audit guide for procurement teams requiring verified clean-label
A supplement brand specifies "decaffeinated green tea extract, caffeine <0.1%, solvent-free" on their purchase order. The supplier delivers a product testing at 0.08% caffeine — within specification. The COA is silent on decaffeination method. A third-party GC-Headspace screen returns dichloromethane (methylene chloride, DCM) at 380 ppm.
Under ICH Q3C / USP <467>, DCM is a Class 1 solvent — classified as a probable human carcinogen by EPA, WHO, and IARC (Group 2A). The USP <467> pharmaceutical limit is 600 ppm. The FDA food-use limit for DCM in decaffeinated products is 10 ppm (21 CFR §173.255). The supplier's product at 380 ppm is 38× above the applicable food standard, cannot qualify for any clean-label certification, and directly contradicts the "solvent-free" claim on the PO.
This is not an edge case — it is the central compliance failure in decaffeinated green tea extract procurement. The word "decaffeinated" discloses nothing about which process was used, which solvents are residually present, or whether the catechin profile has been compromised. This guide resolves each question with extraction efficiency data, solvent residue limits, HPLC assay parameters, and the supplier audit framework required to source compliantly.
Two genuinely solvent-free decaffeination methods are applicable to commercial green tea extract production: (1) Supercritical CO₂ (scCO₂) — CO₂ above its critical point (T >31.1°C, P >73.8 bar) acts as a tunable, chemically inert solvent. With water as co-solvent at 40°C / 400 bar / 7 wt%, caffeine/EGCG selectivity reaches 2.57 — removing proportionately more caffeine than EGCG per extraction cycle. CO₂ reverts to gas at ambient conditions — zero solvent residue. (2) Hot-water extraction decaffeination — fresh tea leaf at 100°C for 3 minutes (1:20 leaf-to-water ratio) removes 83% of caffeine while retaining 95% of total catechins. No solvents involved at this stage. Both methods require independent HPLC caffeine assay (LOQ ≤0.01%) and GC-Headspace USP <467> Class 1 screening to verify caffeine ≤0.1% and confirmed absence of chlorinated solvent residues. Methylene chloride decaffeination is disqualified for clean-label positioning regardless of residue level.
- Introduction: Clean-Label Demand and the Decaf Disclosure Gap
- Technical Parameter Table: scCO₂ vs. Water vs. DCM vs. Ethyl Acetate
- Supercritical CO₂: Physics, Selectivity Data, and EGCG Retention
- Hot-Water Decaffeination: Parameters and Two-Step Alert
- DCM and Ethyl Acetate: USP Class, Exact Limits, Dual-Standard Trap
- Caffeine HPLC Assay: Verifying <0.1% Residual Caffeine
- USP <467> GC-Headspace Verification Framework
- Supplier COA Audit Checklist and Disqualification Triggers
- Blue Eyes Biotech Solution: Solvent-Free Decaffeinated EGCG 98%
- FAQ
1. Introduction: Clean-Label Demand and the Decaf Disclosure Gap
No US, EU, or international standard currently mandates that a "decaffeinated" ingredient label discloses the decaffeination method or the solvents involved. A product decaffeinated with methylene chloride and one processed entirely by supercritical CO₂ carry legally identical labels. The only mechanism available to procurement teams is contractual method disclosure combined with mandatory third-party analytical verification.
A supplier COA showing "Caffeine: 0.07% (Pass, Spec <0.1%)" contains zero information about: (a) which solvent or process was used to remove caffeine; (b) whether residual organic solvents are present above clean-label thresholds; (c) whether the EGCG profile was compromised during decaffeination. Without explicit method disclosure and independent GC-Headspace data, a "decaffeinated" claim cannot substantiate any "solvent-free," "natural," or "clean label" positioning. Treat undisclosed decaffeination as a disqualification trigger in supplier qualification.
2. Technical Parameter Table: All 4 Decaffeination Methods
| Parameter | Supercritical CO₂ (scCO₂) | Hot-Water Decaffeination | Methylene Chloride (DCM) | Ethyl Acetate (EA) |
|---|---|---|---|---|
| Process Mechanism | CO₂ above Tc = 31.1°C, Pc = 73.8 bar; tunable polarity by T/P; water or ethanol co-solvent improves caffeine selectivity over EGCG | Caffeine selectively leached from fresh tea leaf at near-boiling temperature; caffeine diffuses faster than catechins at short contact times (<5 min) | Chlorinated solvent preferentially bonds to methylxanthine (caffeine); counter-current extraction 12 h; steam-strip residual solvent; air/vacuum dry | Ester solvent (polarity intermediate between DCM and ethanol) selectively dissolves caffeine; requires steam-stripping post-extraction; may hydrolyze to acetic acid in moisture |
| Caffeine Removal Efficiency | 54% per pass (40°C/400 bar/7 wt% water — Kim et al., 2008); 92.8% with ethanol co-solvent at 80°C (Park et al., 2007); multi-pass required for <0.1% | 83% single pass (100°C, 3 min, 1:20 ratio); >90% with sequential passes; insufficient for <0.1% in single step from 3% starting caffeine | 96–97% single pass — most efficient single-pass method | 96–99% single pass — efficient; comparable to DCM |
| EGCG Retention | Water co-solvent at 40°C/400 bar: 79% EGCG retained (21% co-extracted) — highest documented retention; ethanol co-solvent at 80°C: 62.2% retained (37.8% loss) | 95% total catechins retained in single 3-min hot-water pass | EGCG retention depends on process control; extended steaming steps risk catechin hydrolysis | 70–85% EGCG retention typically; intermediate polarity causes some catechin co-extraction |
| Caffeine/EGCG Selectivity | 2.57 (water co-solvent, 40°C/400 bar — Kim et al., 2008); highest documented selectivity; pure CO₂ at 70°C/≥25 MPa highest per 2025 Hansen parameter analysis | ~0.88 with water at comparable conditions; sequential steps improve cumulative selectivity | High caffeine selectivity over EGCG (low DCM solubility for large polyphenols); irrelevant for clean-label | ~0.24 (similar to ethanol); broader polarity extracts catechins alongside caffeine |
| Residual Solvent Class | CO₂ — zero residue; not classified. Ethanol co-solvent (if used): Class 3, 5,000 ppm limit | Water — not classified; no solvent residue limit applies. Downstream ethanol extraction step: Class 3 rules apply | Class 1 — Should Be Avoided. USP <467> pharma limit: 600 ppm. FDA food limit: 10 ppm (21 CFR §173.255) | Class 3 — Low Toxicity. USP <467> limit: 5,000 ppm. Many retailers prohibit EA |
| Clean-Label Compatibility | Fully compatible — "CO₂ decaffeinated," "solvent-free," organic, non-GMO positioning all supported (water co-solvent). Ethanol co-solvent requires disclosure. | Fully compatible when entire process uses only water or CO₂. "Water decaffeinated" supported with process documentation. | Incompatible with all clean-label, organic, or "solvent-free" claims at any residue level | Conditional — USP-compliant at ≤5,000 ppm. "Solvent-free" not defensible. Many premium retailers prohibit EA. |
| Capital Cost | High — pressure vessels >500 bar; CO₂ compressors; recycling systems; not available at commodity extraction facilities | Low-moderate — simple hot-water leaching; downstream catechin extraction adds cost | Low — atmospheric pressure solvent tanks; widely available at commodity processors | Low-moderate — similar to DCM; additional flammability safety controls required |
| Cost Premium to Buyer | $15–35/kg over solvent-decaffeinated grades | Moderate — less premium than scCO₂ | Lowest — cheapest decaffeination method | Low to moderate |
Table 1 — Decaffeination Method Comparison: Efficiency, EGCG Retention, Solvent Class, and Clean-Label Compatibility
3. Supercritical CO₂: Physics, Selectivity Data, and EGCG Retention
CO₂ becomes supercritical above Tc = 31.1°C and Pc = 73.8 bar (7.38 MPa) — simultaneously exhibiting liquid-like density (solvating power) and gas-like diffusivity (mass transfer). Polarity is tunable by adjusting T and P: lower temperature, higher pressure = more polar = more caffeine extraction. Adding water co-solvent (4–7 wt%) at 40°C / 400 bar achieves caffeine/EGCG selectivity of 2.57 (Kim et al., 2008, Journal of Food Engineering) — the highest documented selectivity of any decaffeination method. A 2025 Hansen Solubility Parameter analysis (Food Chemistry, February 2025) confirmed highest selectivity at 70°C / ≥25 MPa pure CO₂. Ethanol co-solvent at 80°C achieves 92.8% caffeine removal but causes 37.8% EGCG loss.
Green tea leaf contains 2.5–4.1% caffeine (dry weight). Achieving ≤0.1% residual requires removing ≥97%. Single-pass scCO₂ with water co-solvent at 40°C/400 bar achieves only 54% removal per pass — leaving approximately 1.15–1.9% residual caffeine. Multi-pass strategies are required: Huang et al. achieved 95.6% caffeine removal with water-modified scCO₂ at 300 bar / ~60°C over 7.5 hours with only 19.8% catechin loss. Commercial scCO₂ decaffeination to <0.1% caffeine from a 3% starting concentration requires 3–5 staged extraction passes with carefully optimized conditions per stage. A supplier claiming single-pass scCO₂ to <0.1% caffeine without multi-stage process documentation is not credible.
Stage 1: T = 40–50°C; P = 380–400 bar; water co-solvent 6–7 wt%; target 50–60% caffeine removal; EGCG co-extraction <22%
Stage 2–3: T = 50–60°C; P = 300–350 bar; water co-solvent 4–5 wt%; cumulative caffeine removal 85–92%
Stage 4–5: T = 60–70°C; P = 250–300 bar; pure CO₂ (per 2025 HSP data for maximum EGCG retention at higher temperature); target residual caffeine ≤0.1%
Post-process verification: HPLC caffeine (LOQ ≤0.01%) + full catechin profile (EGCG ≥70% pre-decaff content) + GC-Headspace USP <467> ND for all Class 1/2 solvents
Facility qualification: scCO₂ systems rated >500 bar with CO₂ recycling/recovery capability are capital-intensive industrial systems. A supplier without documented high-pressure CO₂ extraction infrastructure cannot credibly support an scCO₂ decaffeination claim.
4. Hot-Water Decaffeination: Parameters and Two-Step Alert
Hot-water decaffeination exploits kinetics of differential extraction between caffeine and catechins in aqueous medium at elevated temperatures. Caffeine (MW 194 g/mol; log P −0.07; highly water-soluble) diffuses rapidly from leaf tissue at 100°C. Galloylated catechins (EGCG MW 458; log P 1.55) are larger, more lipophilic, and diffuse more slowly. Published parameters (Camellia sinensis fresh leaf): 100°C / 3 min / 1:20 leaf:water ratio — caffeine decreases 23.7→4.0 mg/g (83% removal); total catechins 134.5→127.6 mg/g (95% retained). Sequential passes achieve >90% caffeine removal with <10% catechin loss.
Hot-water decaffeination operates on the tea leaf — not a finished extract. A separate catechin extraction step follows (ethanol/water, water, or scCO₂). If ethanol is used in this step, the finished extract is NOT solvent-free — ethanol residuals must be tested per USP <467> Class 3 (limit: 5,000 ppm). "Water decaffeinated, solvent-free" is only valid when the downstream extraction also uses water or scCO₂ exclusively. Require a full manufacturing process flowchart — not just the decaffeination step description.
5. DCM and Ethyl Acetate: USP Class, Exact Limits, Dual-Standard Trap
| Solvent | ICH Q3C Class | USP <467> Limit | FDA Food Limit | EU Food Limit | Clean-Label Verdict |
|---|---|---|---|---|---|
| DCM (Methylene Chloride; CH₂Cl₂) | Class 1 — Should Be Avoided — IARC Group 2A carcinogen; CNS depressant; EPA banned from consumer paint strippers 2019 | 600 ppm (pharmaceutical products; USP <467> Option 1) | 10 ppm (21 CFR §173.255 — decaffeinated coffee and tea for direct human consumption) | 10 mg/kg (EU Solvent Directive 2009/32/EC) | Incompatible with all clean-label, organic, "solvent-free" claims at any detectable level |
| Ethyl Acetate (EA; CH₃COOC₂H₅) | Class 3 — Low Toxic Potential — GRAS indirect food additive; fruity/ester odor above ~1,000 ppm; hydrolysis to acetic acid in moisture | 5,000 ppm (USP <467> Class 3 default) | No specific FDA limit; GRAS 21 CFR §173.228 for indirect use | 10–35 mg/kg depending on food category (EU Solvent Directive 2009/32/EC) | Conditional — "solvent-free" not defensible; major retailers (Whole Foods, Costco) prohibit EA in premium products |
| Ethanol | Class 3 — Low Toxicity — GRAS; no carcinogenicity | 5,000 ppm | GRAS (21 CFR §182.1180) | Permitted | Compatible with most clean-label programs when disclosed |
| CO₂ (supercritical) | Not classified — zero residue at ambient | No limit applicable | GRAS | Permitted | Fully compatible with all clean-label, organic, "solvent-free" positioning |
Table 2 — Decaffeination Solvents: ICH Q3C Class, Exact Limits, and Clean-Label Compatibility
Limit A — USP <467> pharmaceutical Class 1: 600 ppm. Applies to drug products and pharmaceutical excipients. A supplier may correctly report "DCM: 380 ppm — Complies with ICH Q3C Class 1" — technically accurate for a pharmaceutical context. Limit B — FDA food limit: 10 ppm (21 CFR §173.255). Governs decaffeinated tea and coffee for direct human consumption — the applicable category for dietary supplement ingredients. A product at 380 ppm DCM is 38× above the FDA food limit while passing the USP pharmaceutical limit. For clean-label and EU compliance, specify DCM as ND at GC-Headspace LOQ ≤1 ppm — not "≤600 ppm."
6. Caffeine HPLC Assay: Verifying <0.1% Residual Caffeine
Column: Reversed-phase C18 (ODS); 250 mm × 4.6 mm; 5 µm; Inertsil ODS-3, Waters Symmetry C18, or equivalent
Mobile Phase A: Water + 0.1% H₃PO₄ or 0.1% formic acid
Mobile Phase B: Acetonitrile (HPLC/gradient grade)
Gradient: 5% B → 25% B over 20–30 min; flow 1.0 mL/min; caffeine elutes ~12–16 min; EGCG elutes ~17–22 min
Detection: UV 272 nm (primary for caffeine); PDA preferred to confirm caffeine UV spectrum identity (λmax ~272 nm)
Column Temperature: 35°C (prevents EGCG epimerization artifacts)
Reference Standard: Caffeine anhydrous; purity ≥99.0%; lot-specific COA with assigned purity factor
Calibration: External standard; 5-point linear from 0.005–0.5%; R² ≥0.9995 (ICH Q2(R1))
LOQ requirement: ≤0.01% (100 ppm w/w equivalent); S/N ≥10 at LOQ
Sample Prep: 100 mg GTE in 100 mL 50% methanol/water; sonicate 15 min; filter 0.45 µm PTFE; inject 10 µL
Reporting: Caffeine ≤0.10% w/w. If ND: report as "ND (LOQ = 0.01%)" — never report ND without stating LOQ
| Component | Typical Retention Time (C18 Acidified Gradient) | Detection λ | MW (g/mol) | In Decaf GTE |
|---|---|---|---|---|
| Gallic Acid | ~5–7 min | 280 nm | 170 | Present |
| EGC | ~9–11 min | 280 nm | 306 | Present |
| Caffeine | ~12–16 min | 272 nm (primary) | 194 | ≤0.1% in decaf grade |
| EC | ~14–17 min | 280 nm | 290 | Present |
| EGCG | ~17–22 min | 280 nm | 458 | Major peak; should be ≥70% pre-decaff content |
| GCG | ~20–24 min | 280 nm | 474 | Present; Rs vs. EGCG ≥2.0 required |
| ECG | ~23–27 min | 280 nm | 442 | Present |
Table 3 — Simultaneous Caffeine + Catechin HPLC Elution Order (C18 Acidified Gradient)
7. USP <467> GC-Headspace Verification Framework
| Decaffeination Method | Solvents to Test | Procedure Required | LOQ Required | Pass / Clean-Label Criterion |
|---|---|---|---|---|
| scCO₂ + water co-solvent only | None expected; CO₂ and water leave no residue | Confirmatory Class 1 + Class 3 screen on qualification batch | ≤500 ppm Class 3 screen; ≤1 ppm for DCM | ND all solvents → fully supports "solvent-free" |
| scCO₂ + ethanol co-solvent | Ethanol (primary); acetone, IPA (cross-contamination) | Class 3 Procedure A per batch | ≤500 ppm ethanol | Ethanol ≤5,000 ppm → USP-compliant; "solvent-free" not supportable; disclose as "scCO₂ + ethanol decaffeinated" |
| Hot-water decaff → ethanol extraction | Ethanol; methanol (industrial ethanol contamination check) | Class 3 Procedure A; Class 2 check methanol (3,000 ppm limit) | Ethanol ≤500 ppm; methanol ≤300 ppm | "Water decaffeinated, ethanol extracted" — "solvent-free" not valid |
| DCM decaffeination | DCM (primary); chloroform; 1,2-dichloroethane | Class 1 Procedure A + confirmatory Procedure B (DB-WAX column) | ≤1 ppm DCM; Rs ACN/DCM ≥1.0 (system suitability) | DCM ≤10 ppm for FDA food compliance; ND for clean-label; any DCM disqualifies "solvent-free" claim |
| EA decaffeination | Ethyl acetate (primary); acetic acid (EA hydrolysis); ethanol | Class 3 Procedure A; GC-MS confirmatory if acetic acid suspected | ≤500 ppm EA; pH screen for acetic acid | EA ≤5,000 ppm → USP-compliant; "solvent-free" not defensible; EA >1,000 ppm may cause off-odor |
Table 4 — USP <467> GC-Headspace Testing Requirements and Pass Criteria by Decaffeination Method
Primary Column (Procedure A): DB-624 = USP Phase G43 = 6% cyanopropylphenyl–94% dimethylpolysiloxane; 30 m × 0.32 mm i.d. × 1.8 µm film
Confirmatory Column (Procedure B): DB-WAX = USP Phase G16 (polyethylene glycol); 30 m × 0.32 mm × 0.25 µm — required for DCM identification when flagged on Procedure A
Headspace: 80°C; 45–60 min equilibration; 10 psi vial pressure; ~1.0 mL injection volume
Oven Program: 40°C hold 20 min → 10°C/min ramp → 240°C final
Carrier: Helium; ~35 cm/s linear velocity; split ratio 5:1
FID: 250–300°C; H₂ 40 mL/min; air 400 mL/min; N₂ makeup 25–30 mL/min
System Suitability (Class 1): S/N for 1,1,1-trichloroethane ≥5; each Class 1 component ≥3; Rs between ACN and DCM in Class 2 Mixture A standard ≥1.0
DCM-Specific LOQ Target: ≤1 ppm (to confirm ≤10 ppm FDA food limit with adequate analytical margin)
8. Supplier COA Audit Checklist and Disqualification Triggers
Required Documentation — Solvent-Free Decaffeinated GTE
- Written decaffeination method declaration: Explicit statement — e.g., "Decaffeinated by scCO₂ at 350–400 bar, 40–50°C, water co-solvent 5–7 wt%; no organic solvents used at any processing stage." Generic "solvent-free" without process description is not auditable.
- Full manufacturing process flowchart: All steps disclosed — decaffeination, catechin extraction, concentration, drying. Not just the decaffeination step alone.
- Caffeine by HPLC, batch-specific: Result ≤0.10% w/w; LOQ ≤0.01% stated; reference standard identified; chromatogram available on request. "ND" without LOQ is analytically non-compliant.
- Full catechin profile by HPLC: EGCG, ECG, EGC, EC, GCG, GC individually quantified; EGCG ≥70% of pre-decaffeination content confirmed.
- GC-Headspace USP <467> residual solvent data: Class 1 panel (DCM, chloroform, benzene): ND at LOQ ≤1 ppm; Class 3 panel (ethanol, EA, acetone): ND or within limit; column type and LOQ per analyte stated on report.
- scCO₂ facility documentation (if claimed): Pressure vessel specification (>400 bar rated); CO₂ recycling system present; plant documentation confirming installation. No documentation = no credible scCO₂ claim.
- Third-party ISO 17025 verification on first qualification batch.
- Batch-specific lot traceability on all COA data; test dates within 12 months of shipment.
Supplier Disqualification Triggers
- Decaffeination method not disclosed — "decaffeinated" only, no process description
- GC-Headspace residual solvent data absent from COA
- DCM detected at any level in a "solvent-free" or "natural" product
- DCM detected above 10 ppm regardless of label claim (above FDA food limit)
- EA above 1,000 ppm in premium or clean-label products
- Caffeine reported as "ND" without stated LOQ
- EGCG below 50% of specified grade after decaffeination — excessive co-extraction
- scCO₂ claimed but no high-pressure equipment documentation available
- Only "total catechins" by UV-Vis — no individual HPLC catechin profile
- GC-Headspace "pass" stated without column type or LOQ
9. Blue Eyes Biotech Solution: Solvent-Free Decaffeinated EGCG 98%
Advanced Physical Separation — scCO₂ + Macroporous Resin + Water Crystallization
Blue Eyes Biotech's decaffeinated green tea extract achieves caffeine <0.1% through a staged process combining scCO₂ decaffeination with macroporous resin purification and water-crystallization finishing — entirely without methylene chloride, ethyl acetate, or any ICH Q3C Class 1 or Class 2 solvent at any production stage.
The water-crystallization step is critical to reaching EGCG ≥95%: EGCG 95.1% purity achievable after macroporous resin fractionation followed by one-step crystallization in distilled water (published literature; recovery rate 88.4%). This physical finishing exploits EGCG's differential solubility — EGCG crystals form preferentially under controlled temperature and concentration conditions without solvent addition.
Full Specification:
- EGCG assay: ≥98% (HPLC; USP <621>; R² ≥0.9991 validated)
- Caffeine: <0.1% (HPLC; LOQ 0.01%)
- Residual solvents (USP <467>): Class 1 — ND (LOQ ≤1 ppm); Class 2 — ND; Class 3 ethanol — ≤500 ppm
- Heavy metals (ICP-MS): Pb ≤0.5 ppm; Cd ≤0.1 ppm; As (inorganic) ≤0.2 ppm; Hg ≤0.1 ppm
- Certifications: ISO 22000, HACCP, FSSC 22000; third-party COA per batch
Why physical separation matters for B2B buyers: DCM or EA decaffeination creates ongoing compliance liability. Physical separation via scCO₂ + resin + crystallization is capital-intensive and process-specific — the higher price premium is the cost of eliminating that liability. For brands building clean-label supplement lines, the total cost of a DCM compliance failure (recall, reformulation, retailer de-listing) far exceeds the per-kilogram premium for verified solvent-free material.
Specify scCO₂ decaffeination when: bulk GTE (50% polyphenols, 40% EGCG) at decaffeinated grade; functional beverages or blended ingredients where ultra-high EGCG purity is secondary to cost at scale.
Specify physical separation (scCO₂ + resin + crystallization) when: finished product requires EGCG 95–98% as primary active; clean-label program demands ND Class 1 solvents; regulatory submissions require full process chemistry disclosure; retailer or certification body has specific EA or DCM exclusion lists.
Request a Free 20g Sample for Lab Testing
Verify caffeine <0.1% by HPLC, full GC-Headspace USP <467> Class 1/2/3 solvent data, and complete catechin profile in your own facility. Batch-specific COA, ICP-MS reports, and full technical dossiers available upon request.
For Direct Inquiries & Sourcing Specifications
Contact our supply chain team for rapid RFQ processing and customized MOQ arrangements for decaffeinated GTE and EGCG 98% grades.
Email: [email protected]
10. FAQ
Not necessarily. "CO₂ extracted" may refer to the catechin extraction step, not a decaffeination step. A product may also be scCO₂ decaffeinated using ethanol as co-solvent — leaving ethanol residuals requiring Class 3 testing. Require: (1) confirmation the decaffeination step used scCO₂ with water as the only co-solvent; (2) confirmation no organic solvents were used at any subsequent processing stage; (3) GC-Headspace USP <467> showing ND for all Class 1/2/3 solvents; (4) facility documentation confirming high-pressure CO₂ extraction equipment is installed and operational. Without these, "CO₂ extracted" is a marketing statement.
No. USP <467> pharmaceutical Class 1 limit is 600 ppm — accurate for pharmaceutical context. The FDA food limit for DCM in decaffeinated tea is 10 ppm (21 CFR §173.255). A product at 380 ppm is 38× above the applicable food standard. Furthermore, any detectable DCM cannot support "solvent-free," "natural," or clean-label claims regardless of limit. Specify DCM as ND at GC-Headspace LOQ ≤1 ppm in your purchase specification.
The conditions that maximize caffeine/EGCG selectivity (40°C/400 bar/water co-solvent) limit single-pass caffeine extraction yield to only 54%. Green tea contains 2.5–4.1% caffeine. Achieving ≤0.1% requires removing ≥97%. Three sequential passes at 54% removal per pass achieves: 54% → 75% → 86% cumulative removal — still insufficient. Commercial scCO₂ decaffeination to <0.1% caffeine requires 4–6 staged extraction passes with optimized conditions per stage, or an additional purification step (macroporous resin + water crystallization). Single-pass claims without multi-stage documentation are not credible.
No. USDA NOP (7 CFR §205.605) does not permit ethyl acetate as a processing aid in organic food or supplement production — regardless of whether the EA is naturally derived from fermentation. For USDA Organic or EU organic certification, only water and CO₂ are acceptable decaffeination solvents. Specify scCO₂ (water co-solvent only) or fully aqueous decaffeination exclusively for organic-certified products.
EGCG 98% decaffeinated requires a full multi-stage physical purification process — not scCO₂ alone. Single-pass or multi-pass scCO₂ decaffeination of tea leaf produces a decaffeinated extract at the GTE 50–60% catechin purity level, not EGCG 98%. Reaching EGCG 98% requires: (1) scCO₂ decaffeination of tea leaf; (2) ethanol/water extraction of catechins from decaffeinated leaf; (3) macroporous resin fractionation to concentrate EGCG from ~40% to ~74–80%; (4) water-crystallization finishing to ≥95%+. A supplier offering EGCG 98% decaffeinated at commodity pricing warrants immediate analytical scrutiny of the EGCG assay, caffeine, and solvent residue data.
Related Blue Eyes Biotech Knowledge Center Articles
- Residual Solvents in Botanical Extracts: Ethanol vs. Ethyl Acetate — USP 467 Full Guide
- HPLC vs. UV-Vis for EGCG: Detecting False Purity Readings
- EGCG 98% Technical Data Sheet — Full Specification and COA Parameters
- Managing EGCG Epimerization in Liquid Formulations
- EGCG Green Tea Extract: Complete B2B Sourcing and Buyer Guide
