Paintability, exposure and substrate decide the family; the declared movement class and the joint design decide whether the bead survives a Malaysian facade.
Polyurethane and silicone are both elastomeric joint sealants. Both come in a cartridge, both are gunned into a joint and tooled, both stay flexible after curing, and on a shelf they look interchangeable. They are not, and the differences that matter outdoors are not the ones people usually argue about.
The useful way to think about it is that silicone is a weathering material and polyurethane is a construction material. Silicone is exceptionally stable against sunlight and heat, sticks readily to glass and glazed non-porous surfaces, and cannot be painted. Polyurethane is tougher and more abrasion-resistant, bonds well to porous masonry surfaces, takes paint, and is more vulnerable to prolonged sunlight unless it is overcoated. Neither is better. They answer different questions, and the question your joint is asking is usually obvious once you know to ask it.
| Property | Polyurethane | Silicone | What it means on an exterior joint |
|---|---|---|---|
| Paintability | Generally paintable, and often specified because it will be painted | Not paintable — paint beads and peels off it | The single most decisive property on a painted facade. Choose silicone at a joint that will be repainted and you have created a permanent problem |
| Sunlight and heat stability | More vulnerable to prolonged direct sun; commonly overcoated for exposed positions | Very stable, which is why it dominates glazing and exposed non-porous joints | On a fully exposed west-facing elevation this is a real consideration, and the datasheet states what the product is rated for |
| Adhesion to concrete, render, masonry | Generally good, often with a primer on porous substrates | Variable by product; frequently needs a primer and is not always suitable | Ask what the datasheet lists as approved substrates, and whether a primer is scheduled for yours |
| Adhesion to glass, glazed tile, anodised metal | Varies by product; check the substrate list | The usual choice at these surfaces | A window perimeter with a glazed or metal face is a different problem from a rendered one |
| Toughness and abrasion resistance | Tougher, resists tearing and rubbing better | Softer, less resistant to abrasion and tearing | Anywhere a joint gets walked on, brushed past, or hit by debris |
| Dirt pick-up on a facade | Lower, and paint hides the joint anyway | Higher — silicone joints on a light facade often darken into visible lines | An aesthetic issue on a front elevation, and a common source of complaints |
| Movement capability | Declared per product on the datasheet | Declared per product on the datasheet | Do not treat this as a family trait. Two silicones can differ from each other more than a silicone differs from a polyurethane |
| Reversibility | Can be cut out and the joint resealed or painted | Leaves a residue that nothing else bonds to, including new silicone | This decides how easy the next maintenance cycle will be |
Send us a photo of the job on WhatsApp — we reply with an instant quote.
💬 Get Your Instant QuoteThe two-family framing is slightly outdated, because a third option sits between them: hybrid sealants based on modified silane or MS polymer technology. These are marketed as combining the weathering behaviour associated with silicones with the paintability and broad substrate adhesion associated with polyurethanes, often without a primer on many surfaces, and frequently as non-staining.
The honest position is that they are a real and widely used option, that they are not uniformly better, and that the specific product’s datasheet is still the arbiter of what it will do and where. They are usually more expensive per cartridge than a general-purpose tube, which on a facade with a lot of linear metres is a real cost rather than a rounding error. If a contractor proposes one, the questions are exactly the same as for the other two: what movement class does it declare, what substrates is it approved for, does it need a primer on this one, and is it stated as paintable with the paint being used?
This is where most advice goes wrong. People ask which family “moves more”, as though every silicone behaved one way and every polyurethane another. They do not. Movement capability — how much a cured bead can stretch and compress relative to the joint width, repeatedly, without tearing or losing adhesion — is a declared property of the specific product, printed on its datasheet as a class. Low-modulus grades in both families are formulated to move; high-modulus grades in both are formulated to be stiff and strong. Picking the family and ignoring the grade is picking the wrong half of the specification.
So the practical rule is simple and it is a question, not a number: which movement class does this product declare, and is it enough for the movement this joint will actually see? Ask for the datasheet before the work, not after. And notice the second half of that sentence — the joint’s movement is a function of what it is separating, how long that element is, and how much the sun heats it, so a joint between two long render panels on a west elevation is a different problem from a bathroom window reveal on a shaded side.
If a joint is moving enough that this question feels hard, the honest conclusion may be that it is a movement joint rather than a sealed gap, and it belongs in a different specification altogether — see expansion joint leak repair.
If only one thing from this page is remembered, make it this. Ordinary paint does not stick to ordinary silicone. It beads while wet and flakes once dry, and every subsequent repaint on that elevation will do the same thing on that line until someone cuts the sealant out and cleans back the substrate on both sides. Specialist overcoating products and specifically paintable grades do exist, and the datasheet is the only place that claim can be checked — but unless a supplier states it in writing for the exact sealant and the exact paint, treat a silicone joint on an elevation that gets repainted as a permanent problem you have created.
That makes the sequencing question decisive on a painted wall. A joint between rendered surfaces on an elevation that gets repainted every few years is a polyurethane job, because the bead becomes part of the painted surface and disappears. Put silicone there and the painter will either leave a visible untouched line down the facade or waste an afternoon painting something that will peel by the next monsoon.
There are two honest qualifications. First, paintable is not a licence — the sealant has to be fully cured before it is painted, and the paint has to be flexible enough to move with the joint or it will crack in a fine line along the bead. Second, some products are sold as paintable with conditions attached, so this is another datasheet question rather than an assumption. If the facade is being repainted anyway, building facade waterproofing and paint bubbling and peeling cover how the coating and the joints interact.
A Malaysian elevation is a hard place for any organic material. Full equatorial sun, high surface temperatures through the middle of the day, a cold downpour in the afternoon, and that cycle repeating all year without a winter to slow it down. Two consequences follow for sealant selection.
The first is exposure. Silicone’s resistance to sunlight and heat is its main technical advantage, and it is why it holds up in fully exposed positions where other families need help. Polyurethane in permanent full exposure is more commonly specified with an overcoat — which is convenient, because the elevations where you would want a polyurethane are painted ones anyway. What matters is that the decision is made deliberately: an unpainted polyurethane bead sitting in full sun on a west elevation is doing without the protection its specification usually assumes.
The second is movement. Heat is what drives most joint movement on a facade, because materials expand as they warm and the joint is where that expansion is absorbed. A joint that looks static in the morning may be opening and closing measurably by mid-afternoon, and it is doing that every day of the product’s life. That is why the movement class on the datasheet is not a technicality, and why a bead squeezed into a joint too narrow to stretch fails whichever family it came from.
Adhesion is where a joint is won or lost, and it is more specific than the family name suggests. Every sealant datasheet carries a list of substrates it is approved for and a primer schedule saying which of them need priming first. That list, not the family, is the answer.
| Substrate | What usually gets specified | Watch out for |
|---|---|---|
| Cement render, plaster, fair-faced concrete | Polyurethane, commonly with a primer, especially where the joint will be painted | Porous, dusty surfaces need proper preparation; some silicones are not approved on cementitious substrates at all |
| Glass and glazed tile | Silicone | Nothing else bonds as readily, and nothing bonds afterwards to the residue it leaves |
| Aluminium and powder-coated metal | Depends entirely on the product’s approved substrate list and coating type | Coatings vary; a bead that grips one powder coating may not grip another |
| Natural stone and porous cladding | Check for a non-staining grade specifically | Staining is a genuine risk on porous stone; there are products formulated to avoid it and products that are not |
| Bitumen, asphalt or an existing membrane | Compatibility must be confirmed on the datasheet before anything is applied | Some materials are not compatible with bitumen and will soften or fail to bond |
| Polycarbonate and some plastics | Only a product whose supplier explicitly states compatibility | The wrong chemistry can attack the sheet rather than merely fail — see skylight sealant repair |
| Anywhere silicone has been used before | Nothing, until the residue is physically cut out and cleaned off | This is the commonest hidden cause of a resealing job failing on day one |
Three long-term consequences of the substrate decision are worth weighing before it is made for you by whoever is holding the gun.
Silicone comes in two broad cure types and you can tell them apart by nose. Acetoxy-cure silicone smells strongly of vinegar while it cures, because it releases acetic acid. It is the cheap general-purpose tube, and that acid is why it is generally unsuitable on cementitious surfaces and on many metals, where it can corrode or attack the substrate. Neutral-cure silicone has little or no vinegar smell and is the type normally specified for construction and glazing work outdoors. If a tube smells strongly of vinegar and it is about to go against render, concrete or metal, that is worth stopping for.
Polyurethane sealants cure by reacting with moisture in the air. Practically, that means humidity and temperature affect how quickly a bead becomes handleable and how long it takes to reach full properties, and the datasheet states both a skin time and a full cure period. This matters more often than people expect, because two things must not happen before it has cured: it must not be painted, and it must not be washed down or rained on hard. Getting a bead wet before it has skinned is a common way to lose a morning’s work on a facade, and in Malaysia the afternoon storm is a scheduling constraint rather than a possibility.
The instruction that follows from all of that is the same in both families: get the cure and recoat periods from the datasheet, plan the day around them, and do not let anyone talk you into painting a joint the same afternoon it was gunned.
| Exterior joint | Usually specified | Why |
|---|---|---|
| Window or door frame perimeter against render, on a painted wall | Polyurethane | The joint has to be painted with the wall, and it bonds well to render |
| Window frame against a glazed, tiled or metal-faced reveal | Silicone | Non-porous surfaces, no paint involved, full exposure to sun |
| Glass to frame, and glazing generally | Silicone | The family designed for it — see window leak repair |
| Movement joints between render panels or blockwork bays | Depends on the declared movement class, and on whether the wall is painted | This is a movement joint first and a material choice second |
| Roof-to-wall and awning-to-wall junctions | Neither, as the primary seal — the flashing is the primary seal | Sealant alone at a junction like this is the classic failure — see awning joint wall leak |
| Parapet copings and capping joints | Depends on exposure and whether the parapet is painted | Fully exposed, high movement, and a common leak path — see parapet wall leak |
| Aircon pipe and cable penetrations through an external wall | Usually polyurethane on render, with the detail formed correctly first | The sleeve and its fall matter more than the sealant — see aircon trunking penetrations |
| A joint that gets walked on or brushed past | Polyurethane | Toughness and abrasion resistance decide it |
| A hairline crack across a wall face | Neither — this is not a joint | A crack is filled, bridged or injected, not sealed as though it were a designed gap |
Here is the uncomfortable truth of this subject: joint design beats product choice. A correctly designed joint filled with a merely adequate sealant outlasts a badly designed joint filled with the most expensive tube on the shelf, every time. Four principles do almost all of the work, and none of them requires a number to understand.
Preparation sits underneath all four. Joint faces have to be sound, clean, dry and free of dust, laitance, oil, old sealant and loose coating, and primed where the datasheet schedules a primer. A sealant is only ever as good as the two faces it is holding on to.
| Confused with | What is actually different |
|---|---|
| Waterproofing | A sealed joint is a line of defence at one gap. It is not a waterproofing system, and it will not stop water arriving through a wall face — see external wall water seepage |
| A flashing | A flashing directs water away by its shape. Sealant relies on adhesion. Replacing a flashing with a bead removes the geometry that was doing the work |
| Crack repair | A crack is not a designed joint. Filling a moving crack with sealant produces a bead in tension that tears — see injection versus membrane |
| Grout | Grout is rigid and fills between tiles; sealant is elastic and fills a movement gap. They are not substitutes — see grouting versus waterproofing |
| Facade coating | A coating treats the whole wall surface; joints are detailed separately as part of it — see building facade waterproofing |
| Hardware-store DIY sealing | Different job and different scale. The DIY shelf, what it is worth and where it stops is covered in DIY waterproofing sealants |
| What you see | The mechanism | How to avoid it |
|---|---|---|
| The bead has torn down its centre line | Three-sided adhesion, or a joint too narrow for the movement it sees | Backer rod or bond-breaker tape, and a bead wide relative to its depth |
| The bead has pulled cleanly away from one face | Adhesion failure — a dusty or damp face, a skipped primer, or an unapproved substrate | Preparation, and the primer schedule from the datasheet |
| Paint peeling in a line along the joint | Silicone was used where the wall gets painted | Polyurethane or a paintable hybrid on painted elevations |
| The joint has gone hard and crazed | An exposed product ageing in full sun without the protection its specification assumed | Match exposure to the product, or overcoat where the datasheet expects it |
| A dark line down a light facade | Dirt pick-up on a silicone bead | An aesthetic decision to make deliberately before the joint is filled |
| The new sealant would not stick anywhere | Old silicone residue was left in the joint | Cut the old material out mechanically and clean both faces back |
| It leaked again within weeks in the same place | The sealant was never the right answer — water is entering elsewhere and exiting here | Diagnose the entry point first — see why waterproofing fails |
| The bead was applied over an old one | New sealant bonded to old hardened sealant, so it is only as good as what is underneath | Never overlay; always cut out and start from the substrate |
That last question is the one that saves the most money. On many of the junctions where sealant gets used outdoors — roof to wall, awning to wall, a coping joint — the sealant was never meant to be the primary barrier, and a bead is being asked to stand in for a detail that is missing. Questions to ask a waterproofing contractor covers the wider vetting, and waterproofing scams covers the patterns to walk away from.
The cartridge is rarely the cost. Access, preparation, cutting out the old material and the linear metres involved are, and on a facade that is what makes the difference between a morning and a week. Product and pack prices vary by dealer and by grade and are not a useful basis for comparing quotes; what to compare is the metres of joint, the preparation, whether a primer is included and what access is being used.
It is also worth knowing what the alternative looks like if the joints turn out not to be the problem. Where water is arriving through the wall face rather than through a joint, the work is a coated area rather than a sealed line, and that is priced per square foot (market estimate 2026, indicative Klang Valley):
| System | Price (2026, Klang Valley) | Notes |
|---|---|---|
| Acrylic roof coating | RM4 – RM10 per sq ft | an exposed area coating, when the wall face is the problem rather than the joint |
| Cementitious slurry | RM6 – RM10 per sq ft | a rigid system on sound concrete, where the substrate suits it |
Those are installed market ranges for the system, supplied and applied — not a shelf price and not our quoted price. On our own side, ClickBina leak detection and assessment is RM300 – RM800 per visit, waived if you proceed with the repair — that is our price, not a market range, and on a wall that has been sealed twice already it is usually the cheapest money spent. External wall water seepage works through the diagnosis, and how to read a waterproofing quotation covers comparing two quotes line by line.
ClickBina repairs leaks and waterproofs buildings across Kuala Lumpur and Selangor, and on external joints the first job is deciding whether the joint is genuinely the leak path at all — because a wall that has been sealed twice and still leaks is usually telling you the water is coming in somewhere else. We diagnose before quoting, we quote preparation, cutting out, backer rod, materials and access as separate lines so you can see what you are paying for, and we will say plainly when the right answer is a flashing or a coated area rather than another bead. Where the picture is unclear we do a paid detection visit at RM300 – RM800 per visit and waive it if you proceed. Browse all of our waterproofing and leak repair services, or send photos of the joint and the damp patch inside on WhatsApp for a fixed quote.
Tell us what you need — we reply within the hour.