Sandstone Conservation and Repair

Jamie Coath

 

  Severely decayed sandstone elements in red brick elevation
 
Severe salt crystallisation decay to sandstone features at Wentworth Castle, Barnsley

Graham Lott in his article on Britain’s Building Sandstones has highlighted the importance of selecting compatible replacement stone when repairing sandstone masonry and how this is becoming more difficult as suitable supplies diminish. This makes the effective conservation of our existing sandstone structures more and more significant. This article attempts to provide some brief practical guidance on the common aspects of the conservation and repair of sandstone, including for example cramp repairs and other indenting, the need for cleaning, friability and fissibility issues, porosity and the use of consolidants.

The first step where any masonry is found to be deteriorating is to develop a clear understanding of the principal issues affecting the stonework and the mortar joints, including inherent defects, construction defects, and problems resulting from inappropriate maintenance and repair.

TYPICAL DEFECTS

In sandstones the purity and strength of the constituent minerals and the matrix that bonds them together vary greatly, as does grain size or texture, not only from one geological source to another, but also sometimes between beds in the same quarry. This has a significant effect upon the causes and mechanisms of decay and consequently upon the conservation issues that arise.

There are pure siliceous sandstones in which both the grains and cementitious matrix are silica based. These could be more susceptible to alkaline chemical attack, making them vulnerable to materials used in their conservation and repair, such as pure reactive limes or particular cleaning and ‘degreasing’ chemicals. There are also the more borderline calcareous sandstones which may have siliceous grains but a partially lime based cement matrix, such as some of the regional ‘ragstones’, which may be more vulnerable to acid attack. There are also sandstones with significant clay content either distributed throughout the cement matrix which consequently makes them generally weaker in resisting decay, or in concentrated clay beds which differentially erode, sometimes in alarming fashion due to the action of salt crystallisation and wind abrasion.

  Sandstone Corinthian colonnade and statue of Disraeli, St George's Hall, Liverpool
  St George’s Hall after completion of the conservation and cleaning project
 
  Typical defects at St George’s Hall prior to conservation

The top illustration (at Wentworth Castle) shows a particularly extreme example of salt crystallisation decay caused by over saturation of local sandstone that has high clay content. In this case it was concluded that the masonry was beyond any consolidation as all strength had been lost in the cement matrix, and the sandstone was rapidly reverting to a pile of sand grains. It had to be renewed entirely.

Sandstones generally pose particular challenges when it comes to conservation and repair due to this variation in their constitution. The methods of repair and philosophy of conservation that are appropriate depend, therefore, on a detailed analysis and understanding of the stone characteristics involved, the physical and environmental conditions that prevail and detailed historical research into previous repairs or other influences that may have affected the chemical integrity or equilibrium of the material.

Other factors affected by the physical properties of the sandstone include vulnerability to staining by natural weathering and atmospheric pollution. If they are also vulnerable to inappropriate cleaning techniques, changes to their appearance can be permanent. Some sandstones may also be highly porous, making them vulnerable to oversaturation and subsequent frost damage.

In view of these characteristics and the fact that weaker beds of stone are very vulnerable to salt crystallisation decay and subsequent wind erosion or other chemical attack, it is clear that replacement stone requires careful selection. However, harder beds of stone may be more durable to erosion or chemical attack, but they also tend to be more brittle and consequently more vulnerable to mechanical damage or fracturing, often due to corrosion (rusting) of ferrous materials embedded within them.

This issue of ironwork corrosion within the masonry is a prime example of where it is essential that proper research and analysis is essential so that a full understanding of the risks involved is gained before making any decisions regarding appropriate repair techniques and well before any action is taken. Some particular periods or stylistic traditions in architecture incorporated iron or steel elements in the masonry construction and some didn’t, so an intimate knowledge of their differences is essential when assessing the risk of long term damage and decay.

SANDSTONE CONSERVATION AT ST GEORGE'S HALL, LIVERPOOL

Repairs carried out at St George’s Hall in Liverpool illustrate a pragmatic but informed approach to sandstone conservation. This highly significant 19th century Grade I listed Neo Classical civic building was suffering from all of the most common problems associated with sandstone, and a major conservation and restoration project was carried out in three phases between 1997 and 2008 under the direction of Purcell and with the substantial support of the Heritage Lottery Fund. These included extensive sensitive repair and conservation measures to the entire external fabric, which consisted of mainly sandstone faced masonry.

CRAMP REPAIRS

The illustration to the right shows one of several hundred similar fracture defects before the stonework conservation and repairs. The corroding cramp set into the top face of the ashlar course below the cornice had corroded and shards of hard, brittle stone were about to fall off the building. In contrast however, the moulded course above was selected from a softer bed to make the working of the stone face easier for the banker masons; this course had suffered from salt crystallisation decay around the vertical joint due to over-saturation and continual evaporation cycles caused by an open joint in the sky surface of the cornice above. The black patch was a previous cement repair which had been coloured to match the pre-existing pollution-soiled facade. This patch was almost detached due to the salt action behind and the characteristic permanent picture-framing stain pattern caused by the exacerbated evaporation at the joints can be seen on the ashlar and fascia stone faces.

Repairing cramp damage at St George’s Hall
Prior to conservation Fractured piece released to be reinstated
Corroded cramp and sound stone behind fracture Resulting cramp repair on right hand side

The conventional repair solution adopted widely beforehand, both on this building and many other similar cases where rusting metal cramps need to be removed, is to cut a square opening across the connected stones, replace the cramp, and cover with an indented section of stone. However, this method causes significant conservation issues, particularly where many repairs are required, as it can lead to the loss of the original jointing pattern over time as original stone sizes are reduced or rationalised. In some areas of St George’s Hall (see illustration below) the approach had already resulted in the introduction of more vertical joints, repairs which bridged the original vertical joint entirely, and miss-matching of stone colour or weathering patterns in the indent. The result was a patchwork appearance, with greater loss of original material than was necessary to effect the repairs.

The enormous number of cramp repairs required for the new repair programme raised significant concerns both over the supply of new stone and the visual and structural impact of the work. This led the architects to develop a repair technique that differed from the normal rectangular indent, retaining the original joint lines and as much of the original stone as possible, as illustrated on the left.

The bottom right illustration in the series shows the final repair on the far right of the view as well as a more conventional but angular indent to the centre which was adopted where the fractured stone was in more pieces and could not be pinned back in situ once the cramp been replaced. Also visible in this image is a small grey rectangle on the corner of a stone. This is an example of a hydraulic lime repair which was used for minor edge damage and to consolidate the areas of decay around the cornice joints, after first dressing back the friable stone to a sound surface.

WATER INGRESS

Where decay around the joints of the cornices required ‘plastic’ repairs with lime mortars, the essential requirement was to arrest the water penetration. It was found that the sky surfaces of the enormous coping stones were in perfect condition; even the tooling marks were as crisp as the day in the 1840s when they were cut. The water was just getting in through the open joints between them. Where the copings had been previously covered with lead weatherings, these had caused problems with differential run-off and consequent algal soiling on the facades (illustrated below right). Where weatherings had not been used, the wetting and drying capacity of the exposed copings had moderated and evened out the run-off onto the facades. It was therefore decided to reverse this alteration, keep the sky surfaces exposed and fill the open joints with molten lead (final illustration below) which would resist thermal movement, UV degradation and frost action better than mortar or poly sulphide joints.

 
  Above: Indents gone mad at St George’s Hall prior to conservation. Below: Algal staining down the facade at St George’s Hall prior to conservation
 

There was much debate about whether the areas of friable stone should be treated with a stone consolidant such as an alkoxysilane. However, in this case consolidation was felt to be unnecessary because the original worked face had already been lost and, once the cause of saturation had been rectified, the porosity and sacrificial nature of a mortar repair would be more effective at desalinating the stonework. Had the upper surfaces of the copings shown signs of delamination or of becoming more porous, then the use of a consolidant might have been justified, but thankfully this was not the case. The effectiveness of consolidation of this kind outside the museum environment is still widely held to be unproven and should only be attempted as a last resort.

Descaling is another issue that often requires consideration on sandstone masonry projects. Graham Lott refers to the characteristic of some sandstones to be more fissile along the natural bed due to layers of micaceous particles. This has the benefit of enabling paving flags and thackstone for roofs to be riven rather than sawn, but it can pose a serious problem in masonry walls if the stone is ‘face-bedded’ with its natural bed parallel to the face as it has a tendency to delaminate in scales. (Face-bedding sometimes happens when the quarry has limited bed depth available for the face sizes desired.)

Once delamination starts there is little that can be done to stop it, as moisture penetration leads to freeze thaw action and/or salt crystallisation, causing the stone to cleave along its weak point. It helps to keep the stone as dry as possible from above and below, and by ensuring that the surface remains as porous as possible, as this moderates water vapour pressure in the surface layers of the stone. Consolidation, on the other hand, is virtually impossible to achieve without exacerbating the pressure between the layers of stone. One drastic option is descaling so that water is not retained between the layers, but this is a highly destructive and disfiguring process. Descaling should only be done with considerable care or where there is imminent risk of injury from falling masonry.

CLEANING

 
  Molten lead filling to coping sky joints at St George’s Hall

Finally some points about cleaning of sandstone as this is still a hotly debated subject. There have certainly been some serious errors made in the past and not least at St George’s Hall where previous attempts at both chemical and air abrasive cleaning have left permanent damage. The use of acidic cleaning agents caused etched runs and ferric oxide blotches where mixtures were too strong, dwell times too long and neutralisation not good enough, while the use of abrasives led to ‘leopard skin’ patterning and physical pitting. There was also algal soiling which was partly caused by the use of chemicals of too great a strength.

As the high quality siliceous Derbyshire Gritstone on this building had generally proved to be an extremely resistant and durable stone, it was considered that there was justification on this project to try to rectify some of the previous aesthetic damage. After a great deal of discussion and trial sampling, a combination of hot steam and chemical cleaning was developed. The steam removed and moderated the reappearance of algae, slimes, moulds and lichens, while a series of very weak well neutralised hydrofluoric acid treatments with short dwell times evened out the variation in soiling left by the previous attempts. This combined low-impact cleaning approach has been very successful at St George’s Hall and on similar buildings, but caution should always be applied to the use of ‘standard’ cleaning methods without extensive investigation and trialling due to the variations in the constitution of sandstones.

 

 

The Building Conservation Directory, 2013

Author

JAMIE COATH is a partner with Purcell.

Further information

RELATED ARTICLES

Masonry

Stone

 

RELATED PRODUCTS AND SERVICES

Geological consultants

Masonry services (stone masons)

Stone suppliers

 

 

BuildingConservation.com
Site Map