Silicone & RTV Sealant Supplier Qualification: Factory Audit, COA Review and Incoming Inspection

Supplier Qualification for Silicone & RTV Sealants Sourced from China #

TL;DR: The most common failure mode when sourcing silicone and RTV sealants from China is not incorrect material grade — it is lot-to-lot viscosity and cure-rate drift caused by unannounced base polymer substitutions at the compounder level, which a standard COA will not catch without incoming rheology testing.

What the COA Actually Tells You — and What It Hides #

Most procurement teams treat the COA as a pass/fail gate. It is not. A COA for silicone or RTV sealant confirms what the supplier measured on a reference sample — often a retained sample from the qualification batch, not the production lot you just received. The parameters most commonly reported — viscosity, skin-over time, and Shore A hardness after cure — are the easiest to adjust through minor formulation tweaks that do not change the product designation on the label.

The parameter that actually predicts field performance is elongation at break, combined with adhesion strength on the substrate class you are bonding. A silicone sealant with Shore A 25 and elongation at break of 450% will perform very differently from one with Shore A 25 and elongation at break of 180% — even though both pass a hardness-only COA. Per ASTM International ASTM C920 (the governing standard for elastomeric joint sealants), elongation at break must be tested per ASTM D412 and reported as a minimum value, not a nominal. Suppliers who report only nominal values on the COA are giving you half the picture.

COA Review Checklist — Minimum Required Parameters:

The cure system type is worth flagging separately. Acetoxy-cure RTV releases acetic acid during cure and will corrode copper, brass, and certain electronics. Neutral-cure systems (oxime, alkoxy, or benzamide) are required for electrical enclosures, copper busbars, and sensitive substrates. We have seen buyers receive acetoxy product labeled as neutral-cure — the smell test at goods receipt is not a reliable QC method, but it is a useful first indicator.

Factory Audit Checklist: 10 Items That Separate Qualified Suppliers from Approved-on-Paper Ones #

A factory audit for silicone and RTV sealant production is not the same as auditing a mechanical parts supplier. The critical control points are raw material traceability, mixing process control, and filled-cartridge QC — not machining tolerances. The following checklist reflects what we verify during on-site qualification audits.

1. Raw Material Traceability
Confirm that the supplier maintains incoming COAs for base polymer (polydimethylsiloxane, PDMS), crosslinker, catalyst, and filler (typically fumed silica or calcium carbonate) at the batch level. Suppliers who cannot produce incoming raw material COAs for the specific production lot you are auditing are operating without traceability — regardless of what their ISO certificate says.

2. Mixer Qualification Records
Planetary mixers and twin-shaft dispersers must have calibration records for speed, temperature, and vacuum (if applicable). Ask to see the last three mixer calibration reports. Vacuum mixing is required for low-void sealants used in electronics and optical applications; if the supplier claims vacuum mixing capability but the mixer has no vacuum gauge or log, the claim is unverified.

3. Viscosity Control at Filling
Inline or at-line viscosity measurement before cartridge filling is the single most important process control for lot-to-lot consistency. Suppliers who rely only on post-fill COA testing — rather than pre-fill viscosity verification — will have higher lot-to-lot variation. Ask specifically: “What is your viscosity control window at the filling stage, and what is the reject threshold?” A credible answer includes a specific value (e.g., ±15% of target viscosity) and a documented procedure.

4. Cure System Segregation
Acetoxy and neutral-cure production lines must be physically segregated or thoroughly cleaned between runs. Cross-contamination of cure systems is a real production risk. Ask to see the cleaning and changeover procedure, and verify that it is documented and signed off by a supervisor.

5. Filled Cartridge Sampling Plan
What AQL level is applied to filled cartridges? For industrial sealants, we consider ISO Standards ISO 2859-1 AQL 1.0 the minimum acceptable for critical sealing applications. Suppliers using AQL 4.0 or higher for dimensional and fill-weight checks are not running a serious QC program.

6. Accelerated Aging / Shelf Life Validation
Silicone and RTV sealants typically carry a 12- to 18-month shelf life. Ask for the accelerated aging data that supports the stated shelf life — specifically, what storage temperature and duration were used, and what parameters were tested at end-of-life. Suppliers who cannot produce this data are assigning shelf life based on convention, not testing.

7. Adhesion Testing on Representative Substrates
The supplier should be able to provide adhesion test data on at least three substrate types relevant to your application (e.g., aluminum, glass, ABS, galvanized steel). Test method should be cited — typically ASTM C794 (peel adhesion) or ASTM C1135 (tensile adhesion). If the supplier only has adhesion data on one substrate, their qualification scope is narrow.

8. Environmental and Regulatory Compliance Documentation
For export products, verify ECHA REACH SVHC declaration and, where applicable, EU RoHS Directive compliance for electronic-grade sealants. For food-contact or potable water applications, NSF International NSF/ANSI 61 certification or FDA Guidelines 21 CFR 177.2600 compliance documentation is required. Suppliers who offer a generic “RoHS compliant” letter without substance-level test data are not providing verifiable compliance.

9. Production Volume and Capacity Verification
Cross-check stated annual capacity against the number of filling lines, mixer capacity, and warehouse space. A supplier claiming 500 MT/year capacity with a single 500L mixer and two filling lines is overstating capability. This matters because capacity-constrained suppliers are the most likely to subcontract production or substitute raw materials during peak demand.

10. Corrective Action History
Request the last 12 months of customer complaint records and corrective action reports (8D or equivalent). A supplier with zero complaints in 12 months either has no quality feedback loop or is not sharing real data. Three to five documented complaints with closed corrective actions is a healthier signal than a blank record.

Most procurement teams over-invest in reviewing the ISO 9001 certificate and under-invest in items 1, 3, and 9 above. The certificate tells you the system exists. The audit tells you whether it runs.

Incoming Inspection Protocol: Test Methods, Conditions, and Pass/Fail Thresholds #

Incoming inspection for silicone and RTV sealants should be risk-tiered. For a new supplier or a new production lot from a supplier that has had a previous nonconformance, run the full protocol. For established suppliers with six or more consecutive conforming lots, a reduced protocol (viscosity + skin-over time + visual) is defensible — but only if you have the full qualification data on file.

Full Incoming Inspection Protocol:

Viscosity: Test per ASTM D1084 Method B (Brookfield rotational viscometer) at 23°C ± 1°C. Accept if within ±20% of the COA stated value for the same lot. A deviation greater than ±20% indicates either a formulation change or a temperature excursion during storage or transit.

Skin-Over Time: Apply a 2mm bead to a glass plate at 23°C / 50% RH. Measure time to non-tacky surface. Accept if within ±3 minutes of the COA stated value. Skin-over time is sensitive to moisture exposure — a significantly shorter skin-over time on a sealed cartridge suggests the cartridge seal was compromised.

Shore A Hardness (cured): Cure a 6mm slab per the supplier’s stated cure conditions (typically 24h at 23°C / 50% RH for room-temperature-cure products, or per the specified elevated-temperature cure cycle). Test per ASTM International ASTM D2240. Accept if within ±3 Shore A points of the COA stated value.

Elongation at Break: Cure dumbbell specimens per ASTM D412 Die C. Test at 23°C. Accept if elongation at break ≥ the minimum value stated on the COA. This is a go/no-go test — if the COA states a minimum of 200%, a result of 185% is a rejection, not a marginal pass.

Adhesion (for bonding applications): Prepare lap shear or T-peel specimens on the primary substrate used in your application. Cure per supplier specification. Test per ASTM C794 or ASTM C1135. Establish your own minimum threshold based on design requirements — do not accept the supplier’s “typical” value as your pass criterion.

In our qualification program, we reject incoming lots where Shore A hardness deviates more than ±3 points from the COA value, or where elongation at break falls below the stated minimum. These two parameters together catch the majority of base polymer substitution events we have encountered.

Red Flags: Material Substitution and Process Shortcuts #

This is where most sourcing decisions go wrong — not at the RFQ stage, but six months into production volume.

In our supplier qualification program, we have seen suppliers pass initial sample approval and then deliver out-of-spec material at production volume. The trigger is almost always a base polymer substitution at the compounder level — switching from a higher-viscosity PDMS to a lower-viscosity grade to reduce cost or manage supply constraints. The result is a product that looks identical, passes a visual check, and may even pass a hardness test, but shows significantly reduced elongation at break and adhesion durability. A standard COA will not catch this without incoming elongation and adhesion spot-testing.

Specific red flags to watch for:

• Viscosity drop of more than 15% between consecutive lots with no formulation change notification. This is the most reliable early indicator of base polymer substitution.
• Skin-over time shortening by more than 4 minutes across consecutive lots. This suggests catalyst loading has increased — often done to compensate for a lower-reactivity base polymer.
• COA dates that cluster — multiple “different” lots with COA dates within the same week, suggesting the supplier is issuing COAs against a single retained sample rather than testing each lot.
• Filler type change from fumed silica to precipitated calcium carbonate. This reduces raw material cost significantly but degrades tensile strength, elongation, and UV resistance. It is not detectable by hardness testing alone — you need tensile and elongation data.
• Cure system ambiguity — a COA that states “neutral cure” without specifying the crosslinker chemistry (oxime, alkoxy, or benzamide). These three neutral-cure systems have different substrate compatibility profiles and different VOC emission levels. “Neutral cure” is not a single specification.

Three out of five Chinese suppliers we evaluated for a recent electronics-grade RTV qualification could not produce lot-to-lot consistency data across six consecutive months of production. Two of the three had ISO 9001 certification. The certificate and the data are not the same thing.

The English technical content available for silicone and RTV sealant specifications is almost entirely produced by Western brand owners — Dow, Momentive, Sika, Wacker. Chinese suppliers’ English-language datasheets frequently omit elongation at break minimums, adhesion substrate data, and cure system chemistry details. That gap is precisely why specification errors happen at the sourcing stage, and why incoming inspection cannot be skipped for new supplier qualification.

Practical Guidance for Buyers #

When sourcing silicone and RTV sealants from China, the first specification to request from suppliers is not Shore A hardness — it is elongation at break, stated as a minimum value with the test method and cure conditions specified. Hardness is easy to hit and easy to report selectively; elongation at break requires a properly cured specimen and a tensile tester, and it is the parameter most sensitive to base polymer grade changes.

The sourcing mistake we see most often is accepting a single qualification sample COA as the basis for volume purchase. A supplier who delivers conforming material on the first three shipments and then substitutes base polymer on the fourth is not unusual — it is a documented pattern. Request three consecutive production lot COAs before recommending supplier qualification, and verify that the lot numbers and COA dates are sequential and plausible.

Before committing to volume order, require a witnessed or third-party incoming inspection on the first production lot, covering viscosity per ASTM D1084, Shore A per ASTM D2240, and elongation at break per ASTM D412. Set your own pass/fail thresholds based on your application requirements — not the supplier’s “typical” values. For applications involving pump-valve-seals or gaskets-sheet-sealing, adhesion and compression set data on the actual mating substrate are non-negotiable before volume release.

Frequently Asked Questions #

Q1: What is the single most important parameter to verify on a silicone sealant COA from a Chinese supplier?
A: Elongation at break, stated as a minimum value with cure conditions and test method (ASTM D412) specified. Shore A hardness is easier to manipulate and does not predict adhesion durability or resistance to joint movement.

Q2: How do I distinguish acetoxy-cure from neutral-cure RTV sealants when the COA is ambiguous?
A: Acetoxy-cure products release acetic acid (vinegar smell) during cure. Neutral-cure products (oxime, alkoxy, or benzamide crosslinker) do not. If the COA states “neutral cure” without specifying the crosslinker chemistry, request the full formulation disclosure or a third-party FTIR analysis — the crosslinker type is identifiable by infrared spectroscopy. For copper, electronics, and natural stone substrates, only verified neutral-cure is acceptable.

Q3: What AQL level should I specify for incoming inspection of filled sealant cartridges?
A: For critical sealing applications, specify ISO Standards ISO 2859-1 AQL 1.0 as the minimum. AQL 4.0 — which some Chinese suppliers use as their default — allows a defect rate that is not acceptable for industrial sealing applications where a single failed cartridge can cause a field failure.

Q4: What compliance documentation should I require for silicone sealants used in food-contact or potable water applications?
A: Require either NSF International NSF/ANSI 61 certification (for potable water contact) or FDA Guidelines 21 CFR 177.2600 compliance documentation with substance-level test data. A generic “food safe” or “FDA compliant” letter without the specific regulation citation and test data is not verifiable compliance. Also confirm ECHA REACH SVHC declaration for export to EU markets.

Q5: Is a lower-priced silicone sealant from China likely to use calcium carbonate filler instead of fumed silica?
A: Yes, and it matters. Calcium carbonate-filled sealants are significantly cheaper to produce but show lower tensile strength, reduced elongation at break, and degraded UV and weathering resistance compared to fumed silica-filled grades. The difference is not visible on a hardness test. If your application involves outdoor exposure, joint movement, or structural bonding, specify fumed silica filler explicitly and require tensile and elongation data to verify it.

Published by sinoraw.com Technical Team | Request a sourcing consultation