TL;DR: A supplier’s Certificate of Analysis is only as useful as the fields it contains — missing OTR, WVTR, or heat-seal initiation temperature data on a COA is grounds to reject the lot before it reaches your line.
TL;DR: In our incoming inspection protocol, we flag any tea bag filter paper lot where the heat-seal bond strength falls below 1.2 N/15mm — at that threshold, bags split under normal brew agitation.
COA Field Requirements for Tea Bag and Infusion Packaging Materials #
Not every COA is equal. We receive COAs from filter paper suppliers, non-woven converters, and overwrap film producers weekly, and the variance in what they actually report is significant. A COA that lists only grammage and tensile strength tells us almost nothing about whether the material will run on our sealing equipment or perform in a consumer’s cup.
For heat-sealable filter paper (flat or pyramid construction), the minimum COA fields we require before releasing a lot to production are:
| Parameter | Required COA Field | Our Acceptance Threshold |
|---|---|---|
| Grammage | g/m² (ISO 536) | 12.0–16.5 g/m² ±0.5 g/m² |
| Heat-seal initiation temperature | °C | 130–160°C (supplier-specific) |
| Heat-seal bond strength | N/15mm | ≥1.2 N/15mm |
| Wet tensile strength | mN/15mm (ISO 3781) | ≥250 mN/15mm |
| WVTR | g/m²/24h (ASTM E96) | Reported — no pass/fail, used for trend tracking |
| Extractable content / food contact compliance | Reference standard | FDA 21 CFR 176.170 or EU 1935/2004 declaration |
| Lot traceability | Batch/reel number | Required — no generic COA accepted |
The grammage tolerance column matters more than it appears. Filter paper at 11.5 g/m² runs through our vertical form-fill-seal equipment at the same tension setting as 13.0 g/m², but the thinner material is consistently prone to pinholes at the corner seals under the same jaw pressure. We adjusted jaw pressure down 8% on one line to compensate and it resolved the issue, but it cost us a half-day of trials. A supplier who ships lots at the low edge of the tolerance window without flagging it on the COA creates production risk that compounds over time.
For non-woven pyramid bag material (PET or PP), the COA must additionally cover fiber orientation consistency (reported as MD/CD tensile ratio) and melt flow index where applicable. We cross-reference melt flow against our sealing jaw temperature log under what we track internally as our Material Compatibility Form MC-04.
Where Supplier Qualification Breaks Down — and What We Check #
The most common qualification failure we see is not outright fraud on the COA. It is structural incompleteness combined with inconsistent lot-to-lot production at the supplier’s end.
Scenario one: a supplier provides a valid COA for their initial qualification sample — grammage correct, bond strength confirmed, food contact declaration attached. Six months into production, a lot arrives with grammage 1.2 g/m² below the declared spec. The COA for that lot still shows the original values because the supplier issued a template COA rather than a lot-specific COA. On our line, the under-weight paper triggered seal failure on roughly 4% of bags in a 200,000-unit run before our inline tension monitoring flagged the anomaly. The check we now require: lot number on the COA must match the reel label barcode exactly. Any mismatch triggers a hold under our QC-12 incoming material quarantine procedure.
Scenario two: WVTR drift in overwrap film. A food-grade BOPP overwrap arrived with a COA showing WVTR of 3.1 g/m²/24h — within range for a standard tea sachet application. After three months in our finished goods warehouse under 70–75% RH conditions, the tea in those sachets had measurably elevated moisture content at outgoing QC. We traced it to seasonal production variation at the film supplier: summer extrusion runs were showing WVTR values of 4.8–5.2 g/m²/24h on our incoming checks, while the COA still reflected winter qualification data. The mechanism is consistent — BOPP barrier performance correlates with extrusion temperature stability and corona treatment level, both of which drift seasonally at mills without closed-loop process controls. What to check: request WVTR data from at least two production seasons during supplier qualification, not just the initial sample batch.
Scenario three: food contact compliance declarations referencing superseded standards. We regularly receive declarations citing EC 1935/2004 compliance without specifying which positive list (EU Regulation 10/2011 for plastics, or the relevant paper and board framework). A declaration that cites the framework regulation without the applicable specific measure is not auditable. For brands selling into the EU, this creates real liability. Our qualification checklist requires the specific regulation number, the substance name and restriction where applicable, and the date of the most recent migration testing. Any supplier who cannot provide migration test results (even a summary) within 45 days of qualification request is removed from our AVL shortlist for food-contact tea packaging.
Does Filter Paper Grammage Affect Brew Quality? #
Yes, and the relationship is direct. Lower grammage means higher porosity and faster infusion, but also lower wet strength and higher risk of bag failure during steeping.
The industry standard for flat tea bags runs 12.5–14.5 g/m², with pyramid bags typically 15.0–17.0 g/m² due to the three-dimensional stress on the gusset seals. Specialty herbal infusion bags, which often carry larger leaf particle sizes, require wet tensile strength of at least 300 mN/15mm per ISO 3781 to resist abrasion from sharp botanical material. For brands specifying a “clean cup” requirement — no fiber migration — we recommend suppliers provide a fiber release test result alongside the standard COA. This is not universally offered, but it should be a qualification ask for premium tier products.
Specification Notes for Brand Partners #
When you brief us on tea bag or infusion packaging, the first thing we need beyond artwork files is the intended market and retail channel. EU market means food contact compliance under EU 1935/2004 is mandatory documentation. US market means FDA 21 CFR 176.170 for paper components and 21 CFR 177-series for any film elements.
We also need the steeping profile: water temperature (80°C green tea versus 100°C black tea or herbal infusion), steep time, and whether the bag is intended for reuse. These parameters directly determine the wet tensile and heat-seal bond specifications we write into the material purchase order.
The most common brief gap we encounter is silence on particle size distribution for the tea blend. Coarse-cut herbals with fibrous stems will stress filter paper seals differently than finely milled CTC black tea. When particle size is unspecified, we default to the higher wet tensile threshold (≥300 mN/15mm) and flag it for confirmation before bulk production.
Our standard pre-production sampling timeline is 15–20 working days from approved material specification. If a COA dispute triggers lot substitution, add 7–10 working days. Brands that provide complete briefs upfront — material, market, steeping profile, fill weight — consistently clear sampling in the shorter window.
Frequently Asked Questions #
What fields on a COA are actually mandatory versus nice-to-have for tea bag filter paper?
Grammage, heat-seal bond strength, wet tensile strength, and food contact compliance declaration are non-negotiable. WVTR and melt flow index are required for non-woven and film materials respectively. Lot-specific batch number is required on every COA — a generic or templated COA is not acceptable as a lot release document.
Our current supplier sends COAs but the values never seem to change lot-to-lot. Should that concern us?
It should. Genuinely lot-specific COA data will show natural variation within the spec window — grammage fluctuating between 13.1 and 13.8 g/m² across lots, for example. Perfectly identical values across 10 consecutive lots usually mean the supplier is issuing template COAs rather than measuring each lot. The practical risk is that when a lot actually falls out of spec, you have no warning signal. Require a lot-number-to-reel-barcode match as a minimum verification step.
What AQL level do you apply to incoming filter paper inspection?
For food-contact packaging materials, we apply AQL 2.5 for major defects (seal bond failures, food contact documentation gaps) and AQL 4.0 for minor defects (cosmetic variation in paper formation) per ANSI/ASQ Z1.4 sampling tables. For a standard incoming lot of 50–100 reels, that means inspecting a sample of 13 reels for major defect classification.
Can a supplier pass initial qualification and still be a risk in ongoing supply?
Yes — this is where most qualification programs have gaps. Initial qualification tests one production batch under controlled conditions. The WVTR drift scenario we track in our incoming records shows that seasonal production variation at film suppliers can shift barrier performance by 30–40% without any COA change. Requalification after supplier process changes, equipment upgrades, or raw material reformulation is the only control. Our protocol requires suppliers to notify us under a 30-day advance change notice; suppliers who cannot commit to this are classified as Category B in our supplier risk matrix and receive 100% incoming lot testing rather than AQL sampling.
What is a realistic lead time for qualifying a new tea bag material supplier from scratch?
It depends on whether the supplier has existing food contact compliance documentation. With documentation in hand, qualification typically runs 30–45 working days: initial COA review and desk audit in week one, physical testing of qualification samples in weeks two and three, line trial and seal validation in weeks four to six. Without existing migration test data, add 60–90 calendar days for third-party testing. Brands planning a launch should build supplier qualification into the project timeline at least 4–5 months before intended bulk production start.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
Grammage variance is where we’ve seen the most avoidable cost creep — a supplier swap to a 13.5 g/m² base paper from our previous 12.0 g/m² spec added roughly €0.004/bag, which sounds trivial until you’re running 180 million bags annually. We didn’t catch it early because the COA listed grammage but no tolerance band, and procurement had approved it on price alone.
The 1.2 N/15mm threshold tracks with what we see — we’ve had lots come in at 1.05–1.08 that passed visual and the supplier signed off on, and they split consistently on pyramid bags during the tumble agitation phase of brew testing.
The food contact compliance column is where compostable material switches get painful — we trialed a PLA-based nonwoven for pyramid bags in 2022 and the EU 1935/2004 declaration from the converter covered the base film but not the bico seal layer, which meant the whole lot was technically uncleared and we had to pull it before the sustainability claim could go on-pack.
Switching seal jaw tooling to handle the 130–160°C initiation range across multiple suppliers cost us more than we expected — we ran two supplier SKUs with a 22°C delta in initiation temp and ended up needing a second jaw set at roughly €1,400 to avoid running everything at the lower supplier’s upper limit and degrading bond strength on the other. Standardizing to a single initiation temp band in our supplier qualification criteria saved that tooling cost entirely on the next line we commissioned.
Wet tensile strength is the one we got burned on — our COA acceptance checklist didn’t include the ISO 3781 test method citation, just the threshold value, and a supplier submitted results from an internal method that used a different dwell time in the water bath, so the 265 mN/15mm figure on their COA wasn’t remotely comparable to our historical data.
Lot traceability field is the one that quietly kills timelines — we’ve had two suppliers in Q3 2023 submit COAs with a single batch number covering what turned out to be three separate production reels, and when one reel failed incoming inspection we had no clean way to quarantine only the affected material. Ended up holding the full 800kg consignment for 11 days while the converter reissued documentation split by reel ID.
The WVTR “trend tracking only” approach works until you’re doing retail-ready shelf displays where the bags sit exposed for weeks — we had to flip WVTR on overwrap film to a hard threshold (≤8 g/m²/24h per ASTM E96) after moisture uptake was visibly caking loose-leaf pyramid bags on shelf in a high-humidity market. Depends entirely on your packaging format and distribution environment, but for anyone moving product into Southeast Asia or similar, reporting-only won’t hold.