TL;DR: Cure-system verification at incoming inspection — not tensile strength or hardness — is the single most predictive indicator of whether a Chinese RTV sealant will perform under thermal cycling in service.
TL;DR: In our qualification program reviewing 31 Chinese silicone and RTV sealant suppliers over three years, fewer than 40% could provide consistent lot-to-lot elongation data within ±10% of their stated spec.
Where Silicone Sealant Qualifications Fail — and What the COA Didn’t Show #
A fluid-systems OEM sourcing acetoxy-cure RTV for gasketing in a generator enclosure assembly discovered the failure mode eighteen months after volume production began. The sealant was passing incoming Shore A hardness checks at 22–25 points. Tensile was within range. The COA looked clean. But field returns showed premature adhesion loss at the metal-silicone interface, concentrated in units assembled during three specific production months.
The root cause, confirmed after supplier audit: the compounder had changed the crosslinker loading during those months to manage a raw material shortage. Cure rate shifted from the qualified 24-hour full cure to approximately 36–40 hours, but the COA continued to report the same 24-hour tack-free time because that test was performed in a controlled 23°C/50% RH environment — not the 18°C/35% RH conditions present in winter production. Nothing on the COA reflected this. The batch hardness was still acceptable because final-state mechanical properties had not changed, only the cure kinetics.
This is the failure mode that qualification programs consistently miss: cure system integrity under boundary conditions. And it is almost entirely absent from the standard COA fields that Chinese suppliers submit.
The Parameters That Actually Predict Field Performance #
Shore A hardness and tensile strength are the two values buyers review first on a silicone sealant COA. Both are weak predictors of field performance in thermal cycling or adhesion-critical applications. The parameters that actually carry signal are tack-free time under controlled humidity, elongation at break, lap shear adhesion to the specific substrate in the assembly, and — for any application above 150°C continuous — thermal stability retention after 168 hours at operating temperature per ASTM D573.
Elongation at break matters more than most procurement specifications acknowledge. A silicone sealant with 250% elongation versus 400% elongation on the same 40 Shore A product will behave completely differently under joint movement. Chinese GB/T standards — specifically GB/T 13477 — permit a testing window that allows elongation to be measured before full cure in some product categories, which means a supplier can report a compliant elongation figure on a product that would test lower if measured after the full 7-day cure period called out in ISO 11600 for building sealants.
Lap shear adhesion is the most commonly under-specified parameter we see in buyer purchase orders. Buyers specify the sealant grade but leave the substrate blank. A silicone RTV formulated for aluminum will test at 1.2–1.8 MPa on anodized aluminum but may drop to 0.4–0.7 MPa on cold-rolled steel without a primer — which is fine if the drawing calls for primer application, but a failure waiting to happen if it doesn’t.
| Parameter | What Chinese COA Typically Reports | What Actually Matters for Qualification |
|---|---|---|
| Shore A Hardness | Single-point value, 23°C | Lot-to-lot variation ≤ ±3 points across 6 consecutive batches |
| Tack-Free Time | Lab condition (23°C/50% RH) | Performance at boundary condition (18°C/35% RH) |
| Elongation at Break | Sometimes pre-full-cure | Post 7-day cure per ISO 37 |
| Tensile Strength | Reported routinely | Weak field predictor — use only as baseline |
| Lap Shear Adhesion | Rarely specified | Substrate-specific, primed vs. unprimed |
| Thermal Aging Retention | Absent from most COAs | Required if service temp > 150°C continuous |
The most commonly overlooked parameter is cure system type verification. Acetoxy, oxime, alkoxy, and neutral-cure silicones share similar mechanical property windows but have fundamentally different compatibility profiles — particularly with copper, electronics, and food-contact surfaces. A supplier switching from oxime to acetoxy cure at the compounder level may produce a product that passes every mechanical test on the COA while failing on contact corrosion of copper bus bars or voiding a food-contact declaration. This substitution has occurred in our supplier review program (we flag it under internal process code CR-04, material substitution risk), and it is not detectable from standard COA data alone.
Decision Framework — How Qualification Depth Should Match Application Risk #
If the application is low-movement, ambient-temperature gasketing with non-critical substrates — HVAC panel assembly, general enclosure sealing — a Tier 1 qualification is proportionate. Request three consecutive lot COAs before approving the supplier. Verify Shore A within ±3 points of spec, tack-free time within ±20% of stated value, and confirm cure system type via FTIR if the purchase volume justifies it. Annual requalification with one batch pull-test is sufficient for stable suppliers at this tier.
If the application involves thermal cycling above 120°C, adhesion to metal substrates without primer, or contact with electronic components, the qualification threshold changes substantially. We require lap shear adhesion data on the actual substrate used in the assembly — not aluminum proxy testing — and thermal aging retention after 168 hours at the service temperature per ASTM D573. Pass threshold in our program: elongation retention ≥ 70% and hardness increase ≤ 8 Shore A points after thermal aging. Suppliers who cannot produce this data before volume commitment are moved to conditional status regardless of price.
For food-contact, potable water, or medical-adjacent applications, cure system type is not optional data — it is a qualification gate. Neutral-cure or platinum-catalyzed addition-cure silicones are required in these environments, and the declaration must be supported by migration testing per FDA 21 CFR 177.2600 or equivalent NSF 51/61 certification from NSF International. We have rejected suppliers at this stage who submitted acetoxy-cure products with food-safe labeling — the label is applied at the distributor level in some Chinese supply chains, not at the formulator, which is where the chemistry decision is made.
A boundary condition worth stating explicitly: for structural bonding applications where the sealant carries mechanical load, silicone RTV is generally not the right specification regardless of supplier qualification status. Structural silicone per ASTM C1184 is a narrower category with higher lap shear minimums (typically ≥ 0.5 MPa sustained at 70°C). If your engineering drawing calls for structural glazing or curtain wall adhesion, the qualification pathway involves third-party testing and this guide’s thresholds are insufficient. The pump-valve-seals and mechanical-seals-packing categories cover related sealing contexts where the structural versus non-structural boundary also applies.
Practical Guidance for Buyers #
When sourcing silicone or RTV sealant from China, the first specification to request is not the product data sheet — it is three consecutive lot COAs for the same product code. The data sheet tells you what the formulation is designed to do. Three consecutive COAs tell you whether the supplier is actually making it consistently. Elongation at break is the most sensitive indicator of lot-to-lot consistency in our experience; hardness is too easily adjusted and too forgiving of compounder variation to serve as a primary quality signal.
The specific risk scenario to plan for is mid-volume raw material substitution. Chinese sealant compounders operate in a silicone polymer market with significant price volatility, and crosslinker or base polymer substitutions are common when spot prices move. A supplier who qualified cleanly on initial samples can deliver meaningfully different cure kinetics by batch six. Incoming cure verification — a simple tack-free time check at your facility’s ambient temperature, not the lab standard condition — takes under two hours and catches this substitution before the sealant enters production.
Before committing to volume, insist on a 10-unit production sample run (not lab samples, which are often made separately) with full COA. Verify elongation at break against spec within ±15%, lap shear on your actual substrate, and tack-free time at the lower end of your facility’s ambient temperature range. Suppliers who object to production-run sample requests at this stage of qualification are flagging a process control problem worth taking seriously. Internal review programs for gaskets-sheet-sealing and related sealing assemblies follow the same incoming verification logic.
What’s the most important field to check on a Chinese silicone sealant COA?
Tack-free time and elongation at break — not hardness. Hardness is the easiest value for a supplier to land within range; elongation is a more sensitive indicator of compounder consistency and cure system integrity. Verify tack-free time at your actual facility ambient conditions, not the 23°C/50% RH lab standard.
How many consecutive lot COAs should I request before qualifying a supplier?
Three is the minimum — it gives you a baseline for lot-to-lot variation on Shore A and elongation. Six months of data is better for high-volume or high-reliability applications. In our review program, roughly 60% of Chinese suppliers cannot produce six-month lot consistency data on request, which is itself a qualification signal.
Can I tell cure system type from the product data sheet alone?
Sometimes, but not reliably. “Neutral cure” is a marketing term applied loosely in the Chinese market. FTIR verification on a production sample is the only method that confirms cure system chemistry definitively. For copper-contact or food-contact applications, we treat FTIR as mandatory, not optional.
Is ISO 11600 the right standard to specify for industrial RTV sealant?
It depends on the application. ISO 11600 was written for building and construction sealants and covers joint movement capability and adhesion on construction substrates. For industrial gasketing or electronic potting applications, ASTM C920 (elastomeric joint sealants) or product-specific standards are more applicable. Specifying ISO 11600 for an industrial RTV application sometimes leads Chinese suppliers to select a building-grade formulation that is technically compliant but wrong for the use case.
We’ve had a supplier pass qualification and then deliver inconsistent batches. What’s the likely cause?
Raw material substitution at the compounder level is the most common trigger, followed by seasonal humidity effects on cure kinetics at the production facility. If the failure pattern correlates with a production date window rather than being random across batches, suspect a raw material event. Request the supplier’s raw material incoming records for the flagged production period — not all suppliers will provide this, but the refusal is informative.
Published by sinoraw.com Technical Team | Request a sourcing consultation
