Silicone & RTV Sealant Troubleshooting Guide: Common Failure Modes and Root Cause Analysis

TL;DR #

TL;DR: The most common silicone and RTV sealant failures we see in production environments trace back not to application errors, but to specification mismatches at the sourcing stage — wrong cure chemistry for the substrate, or a viscosity grade that was never validated against the actual dispensing equipment.

Failure Mode Identification: What You’re Actually Looking At #

Before diagnosing a sealant failure, the first question to answer is whether the failure is adhesive (at the interface), cohesive (within the sealant body), or substrate-related. That distinction alone eliminates half the diagnostic dead-ends we see procurement teams chase. Most field reports describe “the sealant peeled off” — which tells you nothing about root cause. A clean peel from a metal substrate with no sealant residue left behind is almost always an adhesion failure caused by surface contamination or primer omission. A tear through the sealant body that leaves material on both sides is cohesive failure, which points to cure inhibition or under-specification of elongation at break.

The five failure modes covered here account for roughly 85% of the field complaints we receive from buyers sourcing silicone and RTV sealants from Chinese suppliers. Two of them — cure inhibition and viscosity drift — are directly traceable to sourcing or specification errors that incoming inspection would have caught.

For context on how these sealants fit into broader sealing system design, see our category coverage of gaskets and sheet sealing materials and thread sealants and pipe compounds.

Failure Mode 1 — Cure Inhibition: The Sourcing Risk Nobody Tests For #

Cure inhibition is the failure mode most directly linked to sourcing errors, and it is the one that standard COA review will not catch. It presents as a sealant that remains tacky or uncured at the substrate interface — sometimes days after application — while the exposed surface appears fully cured. The mechanism depends on cure chemistry: condensation-cure RTV systems (acetic acid, oxime, or alkoxy types) are inhibited by certain metals, plasticizers, and sulfur-containing compounds. Addition-cure (platinum-catalyzed) silicones are inhibited by nitrogen, sulfur, tin, and phosphorus compounds at concentrations as low as a few parts per million.

The critical parameter here is the inhibition sensitivity of the specific cure system — and this is almost never stated on a Chinese supplier’s TDS in actionable terms. We have evaluated suppliers whose product data sheets list “suitable for most substrates” without specifying which cure chemistry is in use. That omission is a red flag. A buyer specifying an addition-cure silicone for use on a substrate that has been treated with a tin-based mold release will get complete cure failure at the interface, regardless of how well the sealant performs on its own.

Diagnostic method: Apply a small bead to the suspect substrate and to a clean glass reference surface simultaneously. Cure at 23°C / 50% RH for 24 hours. If the glass reference cures fully (Shore A hardness reaching specified value, typically 20–50 Shore A for general-purpose RTV) but the substrate sample remains tacky, the failure is substrate-induced inhibition, not a product defect. Confirm cure chemistry type from the supplier before proceeding.

Per ASTM International D2240, Shore A hardness measurement on a fully cured 6mm slab is the standard incoming inspection method for cure verification. We use a pass threshold of ±5 Shore A points from the specified grade.

Failure Mode 2 — Adhesion Failure: The Parameter Buyers Consistently Under-Specify #

Most procurement teams specify tensile strength when sourcing silicone sealants. The parameter that actually determines whether the sealant stays bonded in service is peel adhesion — and specifically, whether the failure mode under peel is cohesive or adhesive. A cohesive peel failure (sealant tears, leaving residue on both surfaces) indicates a well-bonded system. An adhesive peel failure (clean interface separation) indicates the bond never formed properly.

We always request peel adhesion data per ASTM International D903 on the actual substrate materials specified in the application — not on the aluminum or glass coupons that Chinese suppliers default to in their standard TDS. The difference in peel strength between a supplier’s standard test substrate and your actual substrate can be 40–60%. That gap is where field failures originate.

Surface energy is the underlying variable. Silicone sealants require a substrate surface energy of at least 38 mN/m for reliable adhesion without primer. Low-surface-energy substrates — PTFE, polyethylene, polypropylene — require either a specialized primer or a different adhesive chemistry entirely. In our qualification program, we reject any supplier that cannot provide substrate-specific peel adhesion data, because a generic TDS value on aluminum tells us nothing about performance on the buyer’s actual assembly.

The primer selection error is the most common sourcing mistake we see in this category. A buyer specifies a silicone sealant with primer, the supplier ships a primer formulated for condensation-cure systems, and the buyer is using an addition-cure product. The primer does not cross-link correctly, and adhesion is worse than without primer. This is not a product defect — it is a specification error that costs real money in rework.

Failure Mode 3 — Viscosity Drift and Premature Skinning: A Storage and Logistics Problem #

Viscosity drift is almost entirely a logistics and storage failure, but it is frequently misdiagnosed as a product quality issue. Silicone RTV sealants are moisture-cure or dual-component systems that begin reacting on contact with atmospheric moisture. A cartridge with a compromised foil seal, stored at temperatures above 25°C during transit from China, will develop a partial skin inside the cartridge that disrupts bead consistency during dispensing.

The specification parameter to verify at incoming inspection is viscosity, measured per ASTM International D1084. For standard gun-grade RTV sealants, viscosity typically falls in the range of 80,000–200,000 mPa·s. A batch arriving at 250,000 mPa·s or above will not dispense consistently through standard 3mm nozzle tips and will produce voids in the bead. We have seen three consecutive shipments from the same Chinese supplier arrive within spec on hardness and elongation but with viscosity 30–40% above the TDS maximum — every time, the root cause was summer transit through Southeast Asian ports without temperature-controlled containers.

The corrective action is not to reject the supplier — it is to specify temperature-controlled logistics for shipments transiting high-humidity, high-temperature routes, and to add viscosity to the incoming inspection checklist. Most buyers do not test viscosity at incoming inspection. They should.

Failure Mode 4 — Compression Set and Creep in Loaded Joints #

Compression set is the failure mode that matters most for static sealing applications — gasketed joints, enclosure seals, flange faces — and it is the parameter most often absent from Chinese supplier COAs. The test method is ASTM International D395 Method B: compress a standard button specimen to 25% of its original thickness, hold at elevated temperature for 70 hours, release, and measure permanent deformation as a percentage of original deflection.

For a silicone sealant used in a static joint at 100°C continuous service, we specify a maximum compression set of 25% after 70h/100°C. General-purpose RTV silicones from Chinese suppliers typically show 20–35% compression set under these conditions — the spread is wide, and the difference between 20% and 35% is the difference between a joint that holds for three years and one that starts weeping at 18 months. Specifying “silicone sealant” without a compression set limit is not a specification — it is a purchase order.

Addition-cure silicone systems consistently outperform condensation-cure systems on compression set. In our comparative testing, addition-cure grades show 12–18% compression set after 70h/100°C, versus 22–35% for oxime-cure condensation systems at the same conditions. If your application involves a loaded joint at elevated temperature, the cure chemistry selection is not interchangeable.

Failure Mode 5 — Cohesive Failure in Dynamic Joints: Elongation Under-Specification #

Dynamic joints — expansion joints, curtain wall glazing, vibrating equipment mounts — require sealants with elongation at break values appropriate for the actual joint movement. The ISO Standards 37 test method measures elongation at break on a dumbbell specimen at 500 mm/min crosshead speed. For joints with movement exceeding ±15% of joint width, elongation at break should be specified at a minimum of 300%.

The sourcing error here is straightforward: buyers specify a general-purpose RTV sealant with elongation at break of 150–200% for a joint that moves ±20% of its width. The sealant cracks at the point of maximum stress concentration — typically at the tooled surface edge — within the first thermal cycle. The failure looks like a quality defect. It is a specification error.

Most Western buyers do not realize that SAC China Standards GB/T 13477 (the Chinese standard governing building sealant testing) uses different specimen geometry and test speed than ISO Standards 37, which means elongation values reported by Chinese suppliers against GB/T may not be directly comparable to ISO values on your engineering drawing. We have seen elongation values diverge by 15–25% between the two test methods on the same product. Always request ISO 37 data specifically if your specification references ISO.

Practical Guidance for Buyers #

When sourcing silicone or RTV sealants from China, the first specification to request is not tensile strength — it is cure chemistry type (condensation vs. addition-cure) and compression set per ASTM International D395 Method B at your operating temperature. Most buyers ask for Shore A hardness because it appears on every TDS. Compression set is the parameter that predicts long-term sealing performance, and it is absent from the majority of Chinese supplier standard data sheets.

The sourcing mistake with the most expensive consequences is primer mismatching: specifying a silicone sealant and a primer from different suppliers, or from the same supplier without confirming cure chemistry compatibility. We have seen this generate 100% rework rates on assembly lines because the adhesion failure is not visible until the joint is loaded or thermally cycled.

Before committing to volume order, require three things: a peel adhesion test report per ASTM D903 on your actual substrate material (not the supplier’s default aluminum coupon), a compression set result per ASTM D395 Method B at your operating temperature with a numeric pass/fail threshold, and three consecutive batch COAs showing viscosity within ±15% of the TDS nominal value. If a supplier cannot provide all three, qualify a different supplier.

Frequently Asked Questions #

Q1: What is the most reliable incoming inspection test for silicone RTV sealants sourced from China?
A: Viscosity per ASTM International D1084 and Shore A hardness per ASTM D2240 on a 24-hour cured slab. Both tests take under 30 minutes and catch the two most common failure modes — viscosity drift from storage and cure inhibition — before the material reaches the production line.

Q2: How do I choose between condensation-cure and addition-cure silicone for a loaded joint application?
A: For any joint under continuous compressive load at temperatures above 80°C, specify addition-cure. In our comparative testing, addition-cure grades show 12–18% compression set after 70h/100°C versus 22–35% for oxime-cure condensation systems — a difference that directly determines joint service life. Refer to the compression set table above and require ASTM D395 Method B data from the supplier before selection.

Q3: Why does my silicone sealant cure on the surface but stay tacky at the substrate interface?
A: This is cure inhibition. It means the substrate contains a compound — tin, sulfur, nitrogen, or certain plasticizers — that is poisoning the cure catalyst at the interface. Run the parallel cure test described in the Failure Mode 1 section: cure on glass and on your substrate simultaneously. If glass cures and substrate does not, the problem is substrate chemistry, not the sealant. Switch cure chemistry type or apply a compatible primer.

Q4: What certifications or test documents should I require before approving a Chinese silicone sealant supplier?
A: Request peel adhesion data per ASTM International D903 on your specific substrate, compression set per ASTM D395 Method B at operating temperature, and three consecutive batch COAs. If the application is food-contact or potable water, also require NSF International NSF/ANSI 61 certification — not just a supplier declaration. For REACH-regulated markets, request the full ECHA REACH SVHC declaration.

Q5: Is a higher Shore A hardness always better for sealing performance?
A: No. Higher Shore A reduces elongation capacity and increases stress on the substrate at the bond line. For dynamic joints, Shore A 20–35 is typically correct. For loaded static joints, Shore A 40–50. Specifying the highest available hardness grade is one of the most common over-specification errors we see.

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