OR/13/015 Stone matching for buildings in the Falkirk THI area: Difference between revisions

From MediaWiki
Jump to navigation Jump to search
Line 36: Line 36:
|+ Table 7    Details of collected stone samples.
|+ Table 7    Details of collected stone samples.
| BGS sample no.
| BGS sample no.
| Sampled building
| colspan="2" | Sampled building
| Building stone
| Building stone
| Thin section
| Thin section

Revision as of 09:41, 24 August 2021

Gillespie, M R, Everett, P A, Albornoz-Parra, L J, and Tracey, E A. 2013. A survey of building stone and roofing slate in Falkirk town centre. Nottingham, UK, British geological Survey. (OR/13/015).

Introduction

Several thousand different building stones have been used to create the substantial built heritage of the UK. However, the great majority of quarries that supplied these stones are now closed, and today new building stone sourced in the UK is being supplied by fewer than two hundred quarries. In most cases, therefore, the stone that was used to construct an historical building cannot be used to make repairs to it. Instead, one of the relatively few currently available stones must be selected (or more of the original stone must be salvaged/recycled from elsewhere).

New stone used to repair buildings should match the original stone as closely as possible in order to maximise the likelihood of achieving a successful, long-lasting repair. Comparing the properties of building stones and roofing slates to constrain their sources and identify the closest-matching materials is known as stone matching.

The purpose of this section of the report is to use stone matching to identify currently available stones (section 5.4) and slates (section 5.5) that could be used in future repairs to buildings in the Falkirk THI area.

Stone matching for building stones

Stones with a very wide range of properties have been used in the built environment. Some consist of tightly interlocking crystals (i.e. they are crystalline) and are therefore essentially impermeable - water and air cannot easily get into the fabric of the stone. These include igneous rocks like granite and basalt (the latter sometimes known as ‘whin’), and metamorphic rocks like slate. Selecting good substitutes for these stones usually requires matching the stones on the basis of appearance (colour, texture and distinctive features) and functional requirements (e.g. load-bearing capacity); matching the crystal-scale properties of impermeable stones is relatively unimportant.

Other stones are made of loosely packed grains rather than interlocking crystals (i.e. they are granular) and are therefore commonly permeable - water and air can get into the fabric of many granular stones. Sandstone, the most common and most widespread building stone in Scotland, is granular and usually permeable. Permeable stone is prone to weathering and decay, and it is particularly important that the grain-scale properties of such stones (including the composition of detrital grains and any mineral cement, the character of pore spaces, and the permeability and cohesiveness of the stone) are taken into account during stone matching.

Stone matching in BGS is usually performed in three stages.


  1. The ‘original’ stone is first subjected to a detailed petrographic examination, to establish the range and character of its intrinsic properties.
  2. The range of properties is then compared with those of stone samples held in the BGS Collection of UK Building Stones, to constrain the source of the stone. Historical records (if available), and the likelihood that the stone was sourced locally or imported, are also taken into account.
  3. Finally, the closest-matching currently available stones are identified. If the quarry from which the stone was sourced originally has been identified, and is still open, it will usually provide the closest-matching stone. If the quarry from which the stone was sourced originally has not been identified, or is closed, the closest-matching currently available stones are identified by comparing the properties of the original stone with those of samples of currently available stones held in the BGS Collection of UK Building Stones.

More details of the BGS approach to stone matching for sandstone are presented in Appendix 5.

Samples and stone descriptions

Twenty building stones were recorded in the survey of buildings within Falkirk THI area (section 3 and Appendix 1). Unlike roofing slates, a petrographic description of each building stone is needed for stone matching.

Ideally, a sample of each building stone is collected for petrographic analysis. However, collecting stone samples from buildings is commonly not straightforward: stone suitable for sampling must be reasonably fresh, it must be representative, and it needs to be both accessible and in a discreet location; furthermore, the building owner must give permission for a sample to be collected. The tight timeframe within which this project was undertaken, and the lack of suitable stone in accessible and discreet locations in many of the surveyed buildings, meant that a smaller range of samples was collected than was originally anticipated.

Nineteen samples of stone were collected (Table 7). Twelve of these come from seven different buildings and represent three of the Falkirk building stones: ‘Buff sandstone 1’, Buff sandstone 4’ and ‘Modern buff sandstone 4’. ‘Buff sandstone 1’ is naturally variable in character, and six further samples representing this type of stone were collected from a stockpile of stone (representing a demolished building) adjacent to Tattie Kirk. A single sample was also collected from exposed rock in Viewfield pit (one of the quarries at Bantaskine, on the southern edge of Falkirk). These quarries were an important source of local stone for Falkirk buildings (section 4).

Table 7    Details of collected stone samples.
BGS sample no. Sampled building Building stone Thin section
Name/address PRIMARY_KE
ED11047 30 Newmarket Street 963250 Buff sandstone 4 Yes
ED11048-1
ED11048-2
Burgh Buildings, Newmarket Street elevation 963306 Buff sandstone 1 Yes ED11048-1
ED11049-1
ED11049-2
ED11049-3
The Steeple, High Street (original stone at clock level) 961318 Buff sandstone 1 Yes ED11049-1
ED11050 16 Melville Street 963257 Buff sandstone 1 No
ED11052 Viewfield pit, Bantaskine quarries N/A Buff sandstone 1 Yes
ED11053-1
ED11053-2
ED11053-3
ED11053-4
ED11053-5
ED11053-6
Demolished wall by Tattie Kirk, off Cow Wynd N/A Buff sandstone 1 Yes
ED11053-1
ED11053-2
ED11053-5
ED11054 The Steeple, High Street (cladding at lower levels) 961318 Modern buff sandstone 4 Yes
ED11060-1
ED11060-2
25-29 Newmarket Street 963266 Buff sandstone 4 Yes
ED11060-1
ED11060-2
ED11061 42 Newmarket Street 963242 Buff sandstone 4 Yes
ED11062 6-8 Cow Wynd 961108 Buff sandstone 1 Yes

A full petrographic description of a building stone has two components: a hand specimen (or ‘macro’) description, made from an unaided visual examination of a sample of the stone or from the stone surface as it appears in exposed masonry; and a ‘micro’ description, which is made using microscope examination of a thin section (a slice of the stone mounted on a glass plate and thin enough to be transparent). A hand specimen description was made of all the collected samples, with the exception of three ‘stockpile’ samples. A thin section was prepared from twelve samples; five from buildings, three from the stockpile samples and one from the Bantaskine quarry sample. The petrographic descriptions and thin section images for these twelve samples are presented in Appendix 6. Building stones for which a sample was not collected were examined in situ (i.e. as exposed masonry) in order to obtain the petrographic information required for stone matching. Summary petrographic descriptions obtained in this way for all twenty building stones are presented in Appendix 1. The following points can be made in regard to sampling and to the stone matching results presented in section 5.4. • ‘Buff sandstone 1’ and ‘Buff sandstone 4’ are the most common building stones in the Falkirk THI area, in terms of the number of buildings in which they occur. These stones are well represented amongst the collected samples and thin sections, and the stone matching results are considered to be robust. • Other than ‘Buff sandstone 2’, each of the other buff sandstones appears in only a few of the surveyed buildings. Thin sections of these stones would have helped the stone matching process, but in most cases a good ‘field’ description was obtained from close examination of the stone in buildings. The stone matching results are therefore considered to be reasonably robust. • BGS holds samples of all the ‘modern’ buff sandstones that are commonly used in the UK, and samples of these stones were taken to Falkirk and compared directly with the various modern stones used in surveyed buildings in order to identify the stone (or at least constrain its identity). In most cases, collected samples and thin sections were not essential to produce a robust stone match for the ‘modern’ buff sandstones. • Relatively few orange sandstones have been used for building stone in the UK, and the most widely used orange sandstones have reasonably distinctive ‘macro’ characteristics. There are also very few current suppliers of orange sandstone, and most of the currently available orange sandstones come from the same quarries that supplied the stone historically. This means that a reliable stone match for orange sandstone can usually be performed without a collected sample and thin section, and the stone matching results are therefore considered to be robust. • The single limestone building stone has distinctive ‘macro’ characteristics, and a collected sample and thin section were not needed to produce a robust stone match. • The granite building stones generally display distinctive ‘macro’ characteristics in the buildings in which they occur. There are also very few current suppliers of granite for building stone in the UK. As described in section 5.2, it is generally not necessary to match the grain-scale properties of granite in order to produce a stone match. A collected sample and thin section were therefore not needed to produce robust stone matches for the granites.

5.4 RESULTS The results of the stone matching exercise for building stones are presented in Table 8. The following points should be borne in mind when using the results presented here. • The list of closest-matching stones has been derived by comparing the Falkirk building stones with samples of stone obtained from currently active quarries. The characteristics of stone from a quarry source can vary over time and from place to place within the quarry; there is therefore no guarantee that a sample of quarry stone held by BGS is representative of the stone currently being supplied by the quarry. • The inclusion of any stone within the list of ‘closest-matching stones’ does not guarantee that it will weather sympathetically or co-exist harmoniously with the original stone. BGS stone matching is designed to maximise the likelihood that a replacement stone and the original stone will be compatible. However, the small number and range of currently available stones compared to those that have been used in the past mean that it is commonly not possible to identify an ideal match. Furthermore, several factors - including the highly variable character of natural stone, the wide range of natural and human factors that can influence stone decay, and the wide range of environmental settings and conditions that masonry can be subjected to - mean that it is not possible to predict with certainty how replacement stone will perform in masonry. • Stone is a natural material and as such the character of a single building stone can vary from building to building, and even within the masonry of a single building. Furthermore, the character of currently available building stones can change over time as quarries excavate further into the bedrock. The range of stones that is available at any one time also changes, as quarries close and new ones open. For these reasons, it is recommended that a site-specific stone matching exercise should always be conducted. This should be done immediately before repairs are carried out, using a sample from the masonry that will be replaced. • One or more samples of stone should be obtained from a quarry operator prior to specifying stone for a repair or restoration project, to confirm the appearance and character of the stone currently being supplied.


Table 8 Stone matching results for Falkirk building stones

Building stone Assessment of stone source Closest matching stones Comment Survey code Survey name


SB1

Buff sandstone 1 The ‘local’ Falkirk stone, sourced from sandstone beds in the Carboniferous age (~313 million years old) Scottish Lower Coal Measures Formation to the south of Falkirk. At different times, Bantaskine, Brightons, Callendar Park, Maddiston and Lathallan quarries, amongst others, would have been important suppliers of the stone used in Falkirk town centre buildings. Blagdon sandstone Stone matching based on full petrographic analysis of samples from several buildings and from Bantaskine (Viewfield) quarry. Drumhead Buff sandstone (laminated variant) Blaxter's Northumberland Buff sandstone Bearl sandstone


SB2


Buff sandstone 2 Probably sourced from sandstone beds in the Carboniferous (~320 million years old) Upper Limestone Formation. This unit crops out widely in the Central Belt, and to the east and west of Falkirk. The closest well-known quarries are between Denny and Stirling. Dunhouse Buff sandstone

Stone matching based on macroscopic examination of exposed stone in buildings. Witton Fell sandstone Drumhead Buff sandstone (uniform variant) Prudham sandstone Stainton sandstone


SB3


Buff sandstone 3 Probably sourced from several quarries in different parts of the Central Belt that produced similar-looking sandstone of Carboniferous age. Possible sources include Humbie Quarry in West Lothian (West Lothian Oil Shale Formation), Craigleith Quarry in Edinburgh (Gullane Formation/Strathclyde Group), and Dunmore Quarry south of Stirling (Upper Limestone Formation). Hazeldean sandstone

Stone matching based on macroscopic examination of exposed stone in buildings. Darney White sandstone Cullalo sandstone Drumhead sandstone (white variant)


SB4


Buff sandstone 4 Similar in character to the Glasgow ‘blond’ sandstones (Bishopbriggs and Giffnock types), which were sourced from sandstone beds in the Carboniferous (~320 million years old) Upper Limestone Formation strata that lie beneath Glasgow. The stone used in Falkirk is more likely to have been sourced from sandstone beds in the Upper Limestone Formation between Denny and Stirling. Prudham sandstone

Stone matching based on full petrographic analysis of samples from two buildings. Dunhouse Buff sandstone Drumhead Buff sandstone Witton Fell sandstone Stainton sandstone Darney Cream sandstone


Table 8 continued

Building stone Assessment of stone source Closest matching stones Comment Survey code Survey name



SB5


Buff sandstone 5 The stone has few distinctive features, and could have come from the Carboniferous successions in the Central Belt of Scotland (Stirlingshire or West Lothian) or northern England; the closest-matching stones in the BGS rock collections are from Northumberland. Darney Cream sandstone

Stone matching based on macroscopic examination of exposed stone in buildings. High Nick sandstone Hazeldean sandstone

Alnwick Moor sandstone


SB6

Buff sandstone 6 Almost certainly a local Falkirk stone of Carboniferous age. The relatively poor quality of the stone suggests it was sourced from the Passage Formation or Scottish Lower Coal Measures Formation. Blaxter sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Prudham sandstone Blaxter’s Northumberland Buff sandstone Swinton sandstone


SB7

Buff sandstone 7 Probably from an unidentified quarry in the Carboniferous age Passage Formation (or possibly Scottish Lower Coal Measures Formation) within the Falkirk area. Fletcher Bank Buff sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Lingberry sandstone Naylor Hill's Buff Gritstone Witton Fell Coarse Grit


SBM1

Modern buff sandstone 1 Stanton Moor quarry in the Carboniferous age (Namurian stage, ~320 million years old) Millstone Grit Group of Derbyshire (or another quarry in the same area that produced very similar stone). Stanton Moor sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Peakmoor sandstone (non gritty) Millknock sandstone Birchover Buff sandstone SBM2 Modern buff sandstone 2 Stoke Hall quarry or Peakmoor quarry in the Carboniferous age (Namurian stage, ~320 million years old) Millstone Grit Group of Derbyshire. Stoke Hall sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Peakmoor sandstone (coarser)

SBM3 Modern buff sandstone 3 From an unidentified quarry in the Carboniferous age (Namurian stage, ~320 million years old) Millstone Grit Group of Northumberland. Dunhouse Buff sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Blaxter sandstone Stainton sandstone


Table 8 continued

Building stone Assessment of stone source Closest matching stones Comment Survey code Survey name


SBM4 Modern buff sandstone 4 Probably Crossland Hill quarry in the Carboniferous age (Namurian stage, ~320 million years old) Millstone Grit Group of Yorkshire. Crossland Hill sandstone Stone matching based on full petrographic analysis. Crossland Hill sandstone would provide the closest match. Peak Moor sandstone Stanton Moor sandstone SBM5 Modern buff sandstone 5 Carboniferous sandstone. The stone is unlike any currently available sandstone and has not been matched to a sample in the BGS rock collections. Dovedale sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Millknock sandstone

SBM6 Modern buff sandstone 6 Has some features of the local Falkirk stone (Buff sandstone 1), but probably sourced from the Carboniferous age (Namurian stage, ~320 million years old) Millstone Grit Group in Derbyshire. Stanton Moor sandstone (finer) Stone matching based on macroscopic examination of exposed stone in buildings. Peakmoor sandstone (finer) Birchover sandstone SO1 Orange sandstone 1 Locharbriggs quarry in the Permian age (~280 million years old) Stewartry Group of Dumfriesshire. Locharbriggs sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Corncockle sandstone SO2 Orange sandstone 2 Corsehill quarry and/or Cove quarry in the late Permian to Triassic age (~240 million years old) Sherwood Sandstone Group of Dumfriesshire. Corsehill sandstone Stone matching based on macroscopic examination of exposed stone in buildings. Cove sandstone L1 Limestone 1 Portland limestone from the Purbeck Group in Dorset. The rock was deposited in the Jurassic and Cretaceous periods (~150 to 140 million years ago). Portland Whitbed stone Stone matching based on macroscopic examination of exposed stone in buildings. G1 Granite 1 Kemnay granite from the Ordovician age (~460 million years old) Kemnay Granite Pluton in Aberdeenshire. Kemnay granite Stone matching based on macroscopic examination of exposed stone in buildings.

G2 Granite 2 Dalbeattie granite from the Devonian age (~400 million years old) Criffel Granite Pluton in Dumfriesshire or Rubislaw granite from the Ordovician age (~470 million year s old) Aberdeen Granite Pluton in Aberdeenshire. There may be no good matching stone available in the UK; an overseas source may be required. Stone matching based on macroscopic examination of exposed stone in buildings. G3 Granite 3 Peterhead granite from the Silurian age (~420 million years old) Peterhead Granite Pluton in Aberdeenshire. Peterhead granite (Stirlinghill quarry) Stone matching based on macroscopic examination of exposed stone in buildings.


Table 8 continued

Building stone Assessment of stone source Closest matching stones Comment Survey code Survey name


G4 Granite 4 Possibly Corrennie granite from the Ordovician age (~450 million years old) Corrennie Granite Intrusion in Aberdeenshire or Ross of Mull granite from the Silurian age (~420 million years old) Ross of Mull Granite Pluton on Mull. Corrennie granite Stone matching based on macroscopic examination of exposed stone in buildings.


Contact details for current suppliers of the closest-matching building stones are presented in Appendix 7.

OR/13/015; Final


5.5 STONE MATCHING FOR ROOFING SLATES Slate matching is generally simpler than stone matching because most of the traditionally used slate in the UK has come from a small number of geographically distinct sources, and slate from each source typically has a set of distinctive characteristics by which it can be recognised. Detailed petrographic descriptions of roofing slates are therefore not required for matching. However, restricted visibility during the survey (see section 3) means that in many cases a low- precision name has been assigned (e.g. ‘Welsh slate, undifferentiated’; or ‘Slate, undifferentiated’). In such cases, a closer inspection of the roofing slate (e.g. by cherry picker) than was possible during the survey will be needed to assign the high-precision name (e.g. ‘Welsh slate, grey’) that will enable selection of the correct matching slate. Six different roofing slates were recorded in the survey of buildings within Falkirk THI area (section 3). The slate matching results are presented in Table 9. None of the Scottish slate quarries is in operation today, but several slate quarries in Wales, England and Spain are open. Scottish slate is typically relatively thick and has traditionally been laid in diminishing courses with random widths; in other words, the overall size of the slates increases from the top row (course) to the bottom row on the roof, and the width of individual slates in all rows is variable. Slate from other sources is typically thinner and has traditionally been supplied in more regular sizes, so it can be difficult to replicate the style of a Scottish slate roof using slate from elsewhere. For that reason, Scottish slate re-cycled from buildings undergoing repair or demolition would provide the best source of matching slate for buildings roofed with Scottish slate. Alternatives are suggested in Table 9 if that option is not possible.

OR/13/015; Final



Table 9 Stone matching results for Falkirk roofing slates

Roofing slate Slate source Closest-matching slate Survey code Survey name


SLSHB

Scottish slate, Highland Border

Several quarries that exploited a band of slate developed on the north side of the Highland Boundary Fault (see Figure 3). Notable quarries were located at Luss, Aberfoyle and Dunkeld. Scottish Highland Border slate is not currently being quarried. The possibility of obtaining recycled Scottish slate (from buildings being refurbished or demolished) should be considered for large-scale repairs. Welsh grey slate is the closest-matching currently available slate and could be used in selective repairs to roofs of Scottish slate. Wherever possible, replacement slates should be laid in the typical Scottish roofing style (diminishing courses and random widths).


SLSWH

Scottish slate, West Highland

A cluster of quarries at Easdale and on Luing (on the coast south of Oban) or quarries at Ballachulish. Scottish West Highland slate is not currently being quarried. The possibility of obtaining recycled Scottish slate (from buildings being refurbished or demolished) should be considered for large-scale repairs. Welsh grey slate is the closest-matching currently available slate and could be used in selective repairs to roofs of Scottish slate. Wherever possible, replacement slates should be laid in the typical Scottish roofing style (diminishing courses and random widths).

SLWg Welsh slate, grey Quarries in north Wales, probably mainly those at Porthmadog and Ffestiniog. These quarries have produced mainly grey slate but greenish and purplish slate have also been produced here. Welsh grey slate is currently available.

SLWp Welsh slate, purplish Probably from Penrhyn quarry in north Wales. The slate produced here is mostly purplish, but grey and greenish slate has also been produced. Welsh purple slate is currently available. SLEc English slate English Lake District (Cumbria). There are notable quarries at Kirkby-in-Furness, Honister, Elterwater and Broughton Moor. Blueish grey and greenish grey English Cumbrian slate is currently available. SLS Spanish slate North-west Spain Spanish slate is currently available.


Contact details for current suppliers of the closest-matching roofing slates are presented in Appendix 7.