Natural does not necessarily mean durable: why we need to think again about how we specify stone

Natural materials are often assumed to be robust by default. But durability in stone is not automatic. It is the result of appropriate selection, detailing and exposure awareness. Stone is often specified with an assumption of permanence – but permanence is not automatic.

When architects choose stone, they are often choosing continuity, longevity, and a sense of permanence. Yet stone is not a single material. It is a category of materials formed under radically different geological conditions – and those conditions determine how it performs.

Stone is a geological product of its environment. Its density, pore structure, mineral stability and bedding are all consequences of where and how it was formed. Those characteristics are not aesthetic footnotes – they are ultimately what govern service life.

Stone is not a single material

Stone is often perceived as hard, resilient and unyielding, but it is also structurally nuanced. Its performance and appearance are shaped by subtle characteristics such as pore structure, mineral composition, finish and exposure, which reveal both its strength and its sensitivity.

At specification level, “stone” is often treated as a single entity. In reality, it spans three geological families: igneous, metamorphic and sedimentary rocks.

Igneous stones, such as granite and basalt, are formed from cooled molten rock. Their crystalline structure typically results in low porosity and high compressive strength. These stones are often chosen where density and resilience are priorities.

Metamorphic stones, including slate and marble, begin as other rock types and are transformed through heat and pressure. Their internal structure is reconfigured, sometimes producing strong directional characteristics that influence how they should be fixed and oriented.

Sedimentary stones, such as limestone and sandstone, are formed from deposited material, often in layers. Their pore structure and bedding are fundamental to how they behave in wet and freeze–thaw environments.

The important point is not geological classification for its own sake. It is that formation governs pore structure. And pore structure governs water behaviour. And water behaviour governs longevity.

Where stone becomes critically saturated, freeze–thaw cycling can lead to scaling, spalling or delamination. The rate at which this happens depends not on whether a material is “natural”, but on measurable characteristics such as absorption rate and porosity. Treating stone as universally durable overlooks these differences.

Where specification often falls short

Stone is frequently selected visually and justified technically afterwards. Colour, texture and perceived prestige understandably shape early design conversations. A particular limestone may complement glazing. A pale sandstone may sit comfortably within a conservation context. These are valid design considerations, but problems can arise when aesthetic alignment becomes the primary filter, and material performance becomes secondary. In the UK climate, freeze–thaw exposure is not theoretical. Nor is sustained moisture.

Water absorption values, density and flexural strength are published data. Yet they are not always central to early material discussions. By the time technical review occurs, visual decisions are often embedded.

Increasing thickness is sometimes assumed to compensate for porosity. In reality, thickness does not alter intrinsic absorption rate. It may increase structural robustness, but it does not change how the material interacts with water at a fundamental level.

Similarly, sealers are often treated as durability solutions. They can reduce surface absorption and slow moisture ingress. But they are maintenance decisions, not structural corrections. They degrade over time and require reapplication. They do not fundamentally alter the geology of the stone.

When a project underperforms, it is rarely because stone is unsuitable as a material. It is more often because its environmental reality was not fully considered at specification stage.

Finish, detailing and exposure

Performance is influenced as much by detailing as by material selection. In smaller exterior elements, like stone signage and architectural features, detailing often matters more than mass for long-term performance. Appropriate fixings, moisture management, surface treatment and edge refinement frequently determine how gracefully the material ages.

Surface finish alters how water behaves. Highly polished finishes can highlight etching or micro-abrasion in exposed environments, particularly in urban settings with atmospheric pollutants. Honed or textured finishes often weather more evenly because surface change is less visually abrupt. Edge detailing also matters. Sharp arrises are inherently more vulnerable in freeze–thaw conditions than eased or chamfered edges. Minor changes in edge profile can materially affect long-term durability.

Bedding orientation in sedimentary stone is another frequently overlooked factor. Stone laid against its natural bed can behave very differently to stone laid in alignment with it. Over decades, incorrect orientation can accelerate lamination and surface failure.

Fixings and restraint systems must also account for movement, both of the stone and of adjoining materials. Thermal expansion coefficients vary between stone types and metal fixings. While often small, differential movement can accumulate over time. If unaccounted for, it introduces stress that may not become visible for many years.

None of these considerations are dramatic. They are incremental. Yet in practice, longevity is usually governed by precisely these incremental decisions.

Designing for service life

The most successful stone projects are rarely those that simply choose the densest or most expensive material. They are those that understand exposure conditions and design accordingly.

External paving subjected to de-icing salts presents a different challenge to sheltered cladding. Coastal projects introduce salt crystallisation risk. North-facing elevations behave differently to south-facing ones. Even within the relatively small isles of the UK, microclimate matters.

Stone is capable of exceptional longevity, as many historic buildings clearly demonstrate. But those structures also reflect a deep understanding of material behaviour, sometimes born out of centuries of observation, and trial and error.

Having previously worked around construction research and development, I have seen how specification decisions made early under time pressure can ripple into long-term performance challenges. Modern projects operate on shorter timelines and tighter budgets: decisions are made quickly, and materials are often selected from samples viewed indoors, under controlled lighting. But, once installed, those materials are exposed to moisture cycles, pollutants, temperature variation and mechanical wear.

Durability is not an assumption – it is an outcome. And it depends on asking early questions: What is the absorption rate? How will this finish behave after ten winters? Is bedding orientation being respected? Is maintenance part of the design intent, or an afterthought? These questions are not about limiting design ambition. They are about aligning ambition with material reality.

Stone as a long-term material

Stone remains one of the most compelling materials available to architects. It carries depth, texture, geological history and intrinsic beauty, which, I think, cannot be replicated synthetically. Its visual integrity can improve with time when properly specified. But, crucially, natural does not mean invulnerable. 

Understanding the geological origins of a stone, such as its density, pore structure, mineral composition and bedding, allows informed decisions about where and how it should be used. Exposure, detailing and maintenance planning are not secondary concerns; they are part of the specification. Stone can endure for generations. But only when its geology, exposure and detailing are treated as design inputs and not as afterthoughts.