Building with Stone : Slate (Part II)

Barry Hunt concludes the review of slate as a building material that he began in last month’s edition of Natural Stone Specialist. This time he considers the physical properties of slate that make it so good for roofing and building.

The natural mid-tone to dark earthy hues of slate enable it to fit in with most surroundings in most applications. Therefore, when selecting slate it is performance that counts most, which for a roofing slater means ease of laying.

The best roofing slates have the most planar cleavage and minimal thickness variation. This allows the slater’s job to be carried out with relative ease and almost guarantees that the slates lie flat. And the flatter they lie the less likely they are to suffer problems with wind uplift or overly high creep of water due to capillary suction effects between overlapping slates.

The art of good slate roof design is ensuring that the slate layout does not allow the creep of water to reach the nail holes. If it does, it can pass through the slate to the wood below.

Slates are graded on the basis of size and thickness variation and a crate of Penrhyn slate, prized for its colour (often blue and/or purple), flatness and consistent thickness, may consist of slates that are 4-5mm thick. With other slates the range may vary by 4mm, occasionally even more.

With a widely varying slate the skilled slater must undertake time consuming sorting and grading, so that courses are the same thickness. During this process, warped and fractured slates should be removed or saved for other purposes where these problems are irrelevant.

If slates of greatly differing thickness lie beside each other the overlapping slate will not lie flat, dramatically increasing the chances of the roof failing.

Surface variation can present a real problem. It has been dealt with in the USA by the rejection of slates with ‘ribbons’ (rippled surfaces). However, many slates do exhibit slightly rippled surfaces and this is often considered to increase their aesthetic desirability.

Within reason, a good slater should be able to minimise the number and size of gaps to an acceptable level.

The two key elements that influence the durability of a roofing slate are its strength and its water absorption rate. Strength under flexural forces helps to resist breakage, while low absorption prevents damage from frosts and other mechanisms of decay.

Slate is effectively impermeable but water can sit between the slates of a roof in the lapped zone for long periods, which gives the weather an opportunity to do its worst.

It is often the case that old slate roofs appearing to be in good condition are found to have all but turned to dust underneath the laps.

In the UK, until July 2004, the British Standard specification for roofing slate suggested a maximum water absorption of 0.3% or less. The replacement European Standard has increased this maximum value to 0.6%, with slates failing to make this level still able to be used if they do not lose significant flexural strength when they are subjected to a frost cycling test.

Roofing slates should be supplied without holes so that the slater can put the holes in the right place. Slates with holes already in them constrain the user to a specific size of lap and, more importantly, might hide the fact that the slate is of low strength or brittle.

The traditional slater turns a slate over and whacks it with the punch on their hammer to form small, neat fixing holes. Because of the many cleavage planes in good slate, this causes the last few layers of the slate surface to break away, creating the perfect countersink for the nail head.

Slates drilled in a factory often do not exhibit sufficient countersinks and protruding nail heads can create stress points on the overlapping slates.

Furthermore, the act of holing a slate will cause weak or damaged slates to break, something that drilling will not do.

Many slates exhibit what is known as a ‘grain’, which is a secondary splitting direction. The presence of a grain is often identified when hammering a pin into blocks of quarried slate and observing the way the block splits.

It is important to ensure that rectangular slates are prepared so that their long dimension is parallel to the grain. This ensures that if the slate does split for any reason it will do so in such a way that the two separate pieces are held by the two nails used to fix the slate into position. In the unlikely event that a slate splits outside of a nail location, the piece of liberated slate will be relatively small and not a significant risk to safety should it fall to the ground.

The presence of a grain can be checked by strength testing both parallel and perpendicular to the slate length. Microscopic studies can also provide direct evidence. One slate from Spain I have dealt with had a well developed crenulation cleavage that made it almost impossible to hand cut when preparing angles for use at valley gutters.

Anatomy of a roofing slate
The process of cleaving slates to a usable thickness can cause an unusual surface effect as the split jumps between different cleavage planes when the individual slates are eased apart. This creates a series of faint surface lines that focus in towards the point at which the splitting was initiated.

Such variations have caused confusion in practice. The Welsh traditionally split slates from the top, while the Spanish split them from the side. The result is that splitting lines travel down Welsh slates but across Spanish slates. As a consequence, during the Spanish slate boom of the 1990s, many slates were condemned for supposedly exhibiting a transverse grain.

Fitness for roofing
The nature of slate formation means that almost any true slate can be used successfully for roofing and the principal question is: How long will it last?

There are two aspects to durability in this instance. The primary factor being colour stability with weather resistance secondary. This has been recognized for many years and in America roofing slates are classified according to whether they are unfading or not. Although in existence since the Bronze Age, the Italian roofing slate industry never took off because of the proportion of calcite or other carbonate minerals present in the slates from that part of the World.

Under even mildly acidic environmental conditions any carbonate minerals will dissolve and promote surface exfoliation that leads to lightening in colour. This can be dramatic if the original colour of the slate was black. The same problem has blighted some of the Chinese slate imported to the UK, with roofs changing from dark grey to a patchwork of lighter shades to almost white within the space of a few years. Such discolouration does not necessarily lead to a breakdown in weather resistance. For that there is another, more destructive mechanism – pyrite decay.

All slates contain pyrite to some degree – and by pyrite I mean any iron-sulphur-bearing minerals such as marcasite, pyrrhotite and chalcopyrite. The question is: are these minerals reactive and, if so, to what degree?

The former French standard allowed slate with reactive pyrites to be used, even to the extent that it could lead to holes in a slate, although it required the slate to be classified accordingly. The British Standard did not tackle this problem and there were unexpected failures of imported Spanish slate during the 1990s, with most traced to one particular quarry. Sometimes pyrite decay had caused delamination and splitting before the slates were even fixed.

The worst slates exhibited pyrite in combination with carbonates, so that when the pyrite began to react it attacked the carbonates and led to expansive gypsum formation that tore the slates apart.

There might not have been a problem if the purchasers had known what to expect (some people like a rustic roof that appears aged) but the great majority expect an unchanging appearance for possibly the first 30 years in the life of the roof.

You cannot specify pyrite out of slates. In most cases the surface will simply tarnish and that will be the end of it. Delabole slate from Cornwall is a good example where there are sometimes huge nodules of pyrite present that oxidize on the surface and then appear to stabilize.

Pyrite decay can affect some Welsh slates, but these problems are restricted to certain seams or veins, as different deposits are often termed. It should be a simple matter to ensure that such materials end up in the waste pile and not on a roof… but mistakes are made.

Unfortunately such a mistake occurred in a batch sent to Australia to roof the mansion of a television magnate. Within six years in a relatively benign environment some of the slates had expanded to twice their original thickness because of multiple delaminations and many others had discoloured and split.
Thankfully, because of such occurrences and the lessons learnt, there are few problems of this magnitude today.

Once a chemically stable slate is obtained, its strength is the next factor to consider.
Naturally, the higher the strength the greater its resistance to all manner of physical and mechanical effects.

Furthermore, higher strength slates generally have lower absorption and thus are less susceptible to cyclic surface exfoliation from freezing and thawing.

With some slates, the strength becomes a critical factor if they are too thin – especially if they have imperfect cleavage planes that split unevenly. This is usually an indication that the slate formation processes were not complete or further stages of metamorphosis have destroyed the original properties of the slate. A common problem with brittle, irregularly cleaving slates appears to be late hornfelsing (coming under the influence of heat exerted by igneous rock formation).

The European Standard for roofing slate takes on the strength versus thickness issue and uses the strength data to determine the characteristic modulus of rupture, which is the lower value within the expected variation range. It uses this value to calculate the minimum thickness to be used depending on the slate dimensions.

Because of the relationship of strength and durability with absorption, it is often a simple matter to carry out an absorption test to assess the likely durability of a roofing slate.

In practice and as a rule of thumb, the lower the absorption the better the slate.

Using this premise, data now being collected from testing different slates to the European Standard, and previous work by both the French and Americans, the classification of roofing slate quality given in Table 1 has been suggested.

One other aspect that rarely gets taken into account when assessing slate durability is the ability of its surface to hold water (adsorption), which is related to water absorption to some degree but also to the mineralogy and the presence of any swelling minerals. Surfaces that take longer to dry are more able to hold on to dirt and other airborne particulates such as spores.

Wetter surfaces also provide a more desirable environment for the growth of organic matter. Once that is established it creates even longer drying times and promotes further growth. The result can be that slates are gradually eased apart and the water-resisting properties reduced.

Additionally, by holding on to water the slate surfaces become more susceptible to frost attack and other weathering mechanisms.

The solution to this problem is to use copper nails. Copper has many advantages over ferrous metals and one appears to be that copper oxides migrate across slate surfaces and poison organic materials. Using cheaper, galvanized nails rather than copper nails can cost a slate roof many years of life and necessitate regular maintenance.

Other uses for slate
As we have seen in the articles this month and last, roofing dominates the market for slate but slate has a wide range of other uses, where its properties are of great benefit, although its uses can be constrained by the limits of block size. Blocks are typically not much more than the larger standard size of roofing slates: 600mm by 300mm.

When larger blocks become available they are much prized. On roofs they are used as ‘slate-and-a-half’ at verges, ridges and valleys.

More larger blocks have become available in recent times by the change in extraction methods from black powder to wire sawing.

Until quite recently it might have been argued that there are no great slate buildings, only great buildings adorned by slate roofs. This is because slate is not an easy material to work into blocks for walling. But the Millennium Centre in Cardiff Bay changed the thinking about this – and not just around Cardiff.

The Millennium Centre demonstrated the greater versatility of slate when a little imagination is applied. Opened in November 2004, it has been called an icon of ‘Welshness’, not least because of the use of 2,500 tonnes of Welsh slate in the walls. There is purple from Penrhyn, blue from Cwt-y-Bugail, green from Nantlle, grey from Llechwedd and black from Corris representing the five main types of slate traditionally produced in Wales. More Welsh Slate was used in the Walers Assembly building next door to the Millennium Centre and in 2010 it was used to clad the new Welsh Assembly Government building in North Wales at Llandudno Junction.

Other than the issues already covered under the discussion of roofing slate, there is very little that can actually go wrong with slate. Rather it is our own expectations of this incredible material that can create a problem.

Slate is made from what are mostly soft minerals of fine crystal size that are easily scratched or plucked from the surface under direct mechanical action. This is why true slates cannot be polished. Either a honed or riven finish is preferred.

Furthermore, although often dark in colour and exhibiting low absorption characteristics, slate will stain if oily foods are dropped or spilt on its surface and when we use slate for flooring or kitchen worktops we must treat the surface and undertake rigorous maintenance. This usually involves the use of penetrating sealants, although even these cannot stop scratches.

Results for some of the tests applied to slate used for construction purposes have been provided in Table 2.
Because of the low absorption rate, the strength does not vary greatly between oven-dried and water saturated stone. That is normally taken to be a good measure of a stable, weather-resisting stone.

So slate could be used for almost any purpose if it were not hampered by its size and splitting constraints.
The Millennium Centre in Cardiff Bay by architect Jonathan Adams has helped to prove this.

A further example of imaginative slate use is the 14m-high Waterfront Needle in St Helier, Jersey, erected to mark both the Queen’s Golden Jubilee and Jersey’s 800 years of allegiance to the crown.

Without doubt, slate is the unsung hero of stone construction. It protects those who understand its natural gifts and choose it for roofing, normally giving long-lasting and trouble-free performance.

A traditional and narrow short-term view persists that there are cheaper alternatives to roofing slate, but choice has greatly increased while competition has pushed down prices to levels where it makes the selection of other roof coverings seem in some ways perverse.

For other building purposes, slate is less competitive and might initially be outperformed by other types of stone – but few of them will stay the course as well as slate.

Barry Hunt is a chartered geologist, a surveyor and scientist. He has been awarded the designation of European Geologist and is a Corporate Building, Conservation and Specialist Surveyor. He is also a Member of the Chartered Institute of Building and a Fellow of the Royal Microscopical Society.

Barry has served on a number of professional committees, including the Technical Committee of Stone Federation Great Britain (SFGB), which provides advice on all stone construction issues. He is one of the authors of the SFGB team that has published codes of practice for the installation of stone floors and internal stone finishes.

Barry gained 14 years’ experience working as a consultant for two renowned civil engineering materials consultancies before, in 2001, establishing his own consultancy, IBIS, specialising in the investigation of construction materials.

The specialist knowledge and services provided by Barry have allowed him to be instrumental in the resolution of problems ranging from blast damaged claddings in London’s West End to advice on the quarrying and extraction of stone from abroad for import to the UK. Other areas of experience include the investigation of all types of building finishes, specialist advice on remedial treatments and the preparation of advice for potential and actual use in litigation or arbitration.

Having worked for consultancies that both undertook in-house laboratory investigation to UKAS requirements, Barry is also able to conduct or oversee a wide range of on-site and laboratory techniques and ensure they are carried out to traceable standards.

One speciality in all investigations is Barry’s hands-on approach. Being trained in industrial roped access (abseiling) allows him to get close to the problems with external building envelopes quickly, efficiently and cost-effectively.

Throughout his working life Barry has published findings from the many investigations he has undertaken. He has also authored chapters for two books on building stone and is currently engaged in other book projects in this field. His most important contribution is considered to be the chapter on the repair and maintenance of stone in the landmark Geological Society publication Stone.

Barry has also written a regular column covering the full spectrum of natural stone use for Natural Stone Specialist magazine.