Industry Standards Explained for Silicone & RTV Sealant

What Happens When the RFQ Standard Field Is Left Vague #

A European industrial equipment manufacturer we were advising in 2023 specified “silicone sealant per ISO standard” in their purchase order. The Chinese supplier shipped product conforming to GB/T 14683, which governs silicone sealants for building construction. The product passed every test the supplier ran. At incoming inspection, the buyer’s lab tested per ISO 11600 Class F-EX-25LM requirements and the material failed the movement accommodation class. The rejection cost seven weeks and a line stoppage.

The root cause was not fraud. The supplier was not substituting material. They were answering to the only standard they understood, because the RFQ gave them room to do so.

This happens because silicone and RTV sealant standards are genuinely fragmented across regions, and the fragmentation is not obvious until you are looking at a COA that uses different test methods than your incoming inspection protocol. There is no single “international silicone standard.” What exists is a patchwork of building sealant standards, industrial assembly sealant standards, food-contact regulations, automotive OEM specs, and electrical insulation requirements — all using different test geometries, conditioning protocols, and acceptance terminology.

The GB/T system compounds this. Chinese suppliers are trained on domestic standards first. When a Western buyer cites an ASTM or ISO standard without specifying the exact clause and grade, a Chinese supplier will often default to the nearest GB/T equivalent — which may be technically correct for their domestic market but misaligned with your engineering drawing.

The Standards That Actually Apply — By Application Domain #

The first thing to establish in any RFQ is the application domain, because this determines which standard family is relevant. Silicone sealant standards divide roughly into four domains: construction/glazing, industrial assembly, food and pharmaceutical contact, and electrical/electronic.

Construction and glazing sealants fall under ISO 11600, which classifies sealants by type (F = facade, G = glazing), movement capability class (7.5 through 50, representing percent movement accommodation), and modulus category. The equivalent in China is GB/T 14683 for single-component silicone building sealants, and GB 16776 for structural glazing sealants. The North American market typically references ASTM C920, which uses a different classification structure: Type S (single-component) or M (multi-component), Grade NS (non-sag) or P (pourable), Class 25 or 50 (movement capability as a percentage of joint width). These are not interchangeable systems. A product rated ISO 11600 Class F-EX-25 does not automatically meet ASTM C920 Class 25 requirements because the test geometries and substrate conditioning differ.

Industrial assembly and gasketing sealants — the RTV products used in engine assemblies, gearboxes, and electronic enclosures — are governed primarily by ASTM D2000 for rubber compound classifications and OEM-specific approvals (Loctite SI, Permatex, Dow Corning equivalents). There is no single ISO standard that directly covers RTV formed-in-place gaskets. This gap is where specification errors concentrate. When we log incoming disputes under our Category B adhesive dispute tracker, roughly 60% of silicone and RTV cases involve industrial assembly products where the buyer cited a building sealant standard by mistake.

Food contact and pharmaceutical silicones require conformance with FDA 21 CFR §177.2600 in the US market, or EU Regulation 10/2011 for food contact materials in Europe. Neither of these is a test standard — they are positive lists of permitted substances and migration limits. The test methods used to demonstrate compliance are typically ISO 10993 series for biocompatibility or extractables testing per the applicable pharmacopoeia. Chinese suppliers often conflate “food grade” with FDA registration, which are different things.

Electrical and electronic applications bring in IEC 60601 for medical devices, UL 94 flammability classifications, and in some cases MIL-A-46146 for military/aerospace conformal coating applications. A silicone that is flame-rated HB under UL 94 does not meet the V-0 rating required for many electrical enclosure applications — and this distinction rarely appears on a generic COA unless you ask for it explicitly.

Which Parameters Each Standard Actually Tests — and What It Misses #

Understanding what a standard tests is as important as knowing its name. Most procurement documents specify a standard without specifying the test conditions, and that is where the acceptance gap opens.

ISO 11600 tests movement accommodation via cyclic extension and compression at ±12.5% or ±25% of joint width, depending on class. What it does not test is long-term thermal degradation. A sealant can pass ISO 11600 Class 25 on day 28 and fail by year two in a facade with large thermal swing. If your application involves operating temperatures above 150°C, the movement class alone is insufficient specification.

ASTM C920 requires adhesion-in-peel testing on at least four substrate types (aluminum, glass, mortar, wood as defaults, but the specification allows substitution). The critical detail that technical buyers miss: ASTM C920 adhesion testing is done at ambient temperature only. If your assembly operates at 80°C continuous, a supplier who passed ASTM C920 at 23°C has not demonstrated adhesion performance under your actual service conditions. Specify ASTM C794 peel adhesion at elevated temperature as a supplementary requirement.

For RTV gasketing sealants, the parameter that most procurement teams over-specify is tensile strength and under-specify is oil resistance after immersion. A formed-in-place gasket in a gearbox will never see tensile loading — it will see sustained compression in a hydrocarbon environment. The relevant test is compression set after 70 hours immersion in reference oil per ASTM D471, with volume swell <25% and hardness change <10 Shore A points as reasonable thresholds for a general-purpose silicone RTV.

The GB/T 14683 standard allows a movement accommodation tolerance that is wider by approximately one class compared to its ISO equivalent. A Chinese supplier shipping “ISO 11600 equivalent” product that was actually tested and classified under GB/T 14683 can deliver material that is technically compliant under Chinese domestic standards but falls short of your ISO 11600 Class 25 acceptance criteria. We see this most clearly when buyers test elongation at break — GB/T allows ≥400% at room temperature, while ISO 11600 Class 25 requires the sealant to survive ±25% joint movement after hot and cold conditioning, which is a more demanding proof of elastic recovery than a simple elongation pull.

I’d prioritize getting the supplier to state, on their COA, which specific standard and which specific clause the product was tested to — not just the standard name. “Tested per ISO 11600” without a class designation is meaningless.

Decision Framework — How to Write the Standard Citation in Your RFQ #

If your application is a building facade or curtain wall, cite ISO 11600 with the full classification: Type (F or G), modulus category (LM, HM), movement class (7.5, 12.5, 20, 25), and substrate types for adhesion testing. Do not leave any of these fields open. Suppliers who cannot fill them in are not qualified for structural glazing work regardless of price.

If your application is an industrial formed gasket or RTV bead — engine covers, gearbox housings, electronic enclosures — there is no single ISO or GB standard to cite. The right approach is to reference ASTM D2000 line call for the cured rubber properties plus your temperature and fluid exposure requirements, and to specify the cure system (acetoxy, oxime, alkoxy) because Chinese suppliers sometimes switch between cure chemistries in the same product line when the acetic acid-releasing cure is causing corrosion issues that the buyer has not connected to the sealant.

If your application involves food contact, the standard citation should reference the regulation and the test method separately: “FDA 21 CFR §177.2600 compliant formulation, confirmed by extractables testing per USP <661> with supplier documentation of substance identity.” Requiring the COA to list major polymer identity and any plasticizer or catalyst residue is not excessive — it is standard practice in medical device and food equipment procurement.

If you are in a region requiring CE marking and the sealant is part of a construction product, EN 15651 applies, not ISO 11600 directly — though EN 15651 references ISO 11600 test methods. This distinction matters for Declaration of Performance documentation under the EU Construction Products Regulation. Chinese suppliers selling into EU construction channels often do not have EN 15651 testing on file, even if they have ISO 11600 COAs. If you need CE-compliant documentation, ask specifically for EN 15651 test reports from a notified body, not just ISO results.

For automotive applications, any OEM-issued material specification (e.g., GM6297M, WSS-M4G323-A) supersedes general standards and should be cited verbatim. If you are a tier-2 or tier-3 supplier using silicone to assemble a sub-component that will feed into an OEM-controlled assembly, the OEM spec governs, and Chinese suppliers will need to demonstrate traceability to that spec explicitly.

One boundary condition worth stating: this decision framework applies to specification-grade purchases. For low-criticality MRO uses — workshop sealing, non-structural gaskets, access panel sealing — over-specifying standards increases cost without proportional benefit. The engineering drawing should dictate which tier of specification applies.

Practical Guidance for Buyers #

When sourcing silicone or RTV sealants from China, the first specification to request is not hardness and not tensile strength. Request the full standard classification with test conditions — specifically, the movement class or the ASTM D2000 line call with the service temperature range, the cure chemistry, and the substrate types tested for adhesion. Hardness and tensile strength appear on every COA because they are easy to measure; they are also the parameters with the weakest correlation to in-service performance for sealants used in dynamic joints or thermal cycling applications.

The risk scenario that causes the most disruption in our QC-07 material risk procedure for this category is a chemistry switch at the supplier level. Oxime-cure and alkoxy-cure silicones look identical on a standard COA. If your application involves electronics or contacts with copper or tin, an undisclosed switch to acetoxy cure will cause corrosion within weeks. Request cure chemistry declaration on every COA, and test for acetic acid release on incoming lots by curing a bead in a sealed container and using pH paper — it takes five minutes and has caught substitutions that formal COA review missed.

Before volume commitment, insist on three consecutive production lot COAs covering movement class or compression set data, not just the initial sample approval report. Require that these lots span at least 60 days of production. Batch-to-batch consistency is where Chinese suppliers with inconsistent polymer blending show up — the qualification sample passes, production diverges.

Frequently Asked Questions

Can I cite ASTM C920 in an RFQ for an industrial RTV gasket application?
No — ASTM C920 is specifically a building sealant standard governing facade and glazing applications. For industrial RTV gasketing, the correct reference is ASTM D2000 for cured compound properties, combined with immersion resistance data per ASTM D471. Citing C920 for an engine gasket application will generate technically incorrect COA data and create incoming inspection ambiguity.

Is GB/T 14683 equivalent to ISO 11600?
Close but not equivalent. GB/T 14683 and ISO 11600 use similar movement class definitions but differ in substrate conditioning requirements and in the specific test geometry for adhesion testing. A product tested under GB/T 14683 Class 25 may not pass ISO 11600 Class 25 acceptance on your incoming test protocol. Always request ISO 11600 test reports from an ISO-accredited laboratory if ISO compliance is the engineering requirement.

What does “food grade silicone” actually mean from a Chinese supplier?
It depends on which market they are targeting. In China, food contact silicone falls under the GB 4806 series. For US export, FDA 21 CFR §177.2600 compliance is required. These are different positive-list systems with overlapping but non-identical permitted substances. “Food grade” without specifying the applicable regulation is not a procurement specification.

How many consecutive lot COAs should I require before qualifying a Chinese silicone supplier?
Three lots minimum, spanning at least 60 days of production. This is based on the production cycle patterns we see in Chinese silicone compounders — reformulation events and polymer batch changes typically show up within that window. One lot COA, however good, only tells you the supplier can hit spec once.

Does EN 15651 replace ISO 11600 for EU construction applications?
It depends on the specific variable of your product’s intended use. EN 15651 is the harmonized European standard for sealants in construction work under the EU CPR, and it references ISO 11600 test methods but has its own classification system and Declaration of Performance requirements. If you need CE marking, you need EN 15651 test reports from a notified body — ISO 11600 results alone are not sufficient for CPR documentation.

Explore related categories: pump valve seals and fluid sealing sourcing and epoxy and anaerobic adhesive standards.

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