Long term consequences of stonecleaning

This research project investigated the consequences of previous stonecleaning with respect to its effects on the rate of sandstone decay. Quantification of the rate of decay allows estimation of the likely future costs of façade maintenance and future requirements for stone.

The research programmme was sponsored by the Technical Conservation Research and Education Division of Historic Scotland.

Introduction

Over the past 20-35 years cleaning of stone facades of buildings has been a significant activity, both in terms of financial outlay and the effect on the built heritage of our cities. Removal of the soiling layer has been perceived by the general public and building owners as a "good thing" because of the simplistic notion that a clean, bright facade reflects well on the urban environment in general and on the image of the building occupier in particular.

It is now well known, through research conducted by the Masonry Conservation Research Group of The Robert Gordon University for Historic Scotland and others, that much of this past stonecleaning intervention has been carried out in ignorance of the effects and consequences that this work has had on the stone. The potential for immediate damage to the stone as a result of stonecleaning has been established by previous research and is now generally accepted. In the case of excessively abrasive cleaning methods, there is a noticeable and immediate effect on the stone manifested by grain loss, increased surface roughness, loss of fine detail and erosion of soft beds. In the case of inappropriate chemical methods, damage to sandstone may not be immediately apparent but may become obvious after a number of years if chemical residues in the porous stone form damaging salts below the surface. The longer-term effect on the weathering and decay of stone after stonecleaning has until recently been unexamined.

A limited preliminary study by The Masonry Conservation Research Group (Roberts and Urquhart 1995) investigated the incidence of decay as a result of stonecleaning at four terraces in Edinburgh and Glasgow. Whilst the results of this study were inconclusive due to the small sample size there was nevertheless clear evidence that stonecleaning increased the incidence of granular disintegration on these terraces by 15-20% when compared to uncleaned facades.

It is however evident from a number of examples that stone replacement has been, and is being, carried out on facades that have been cleaned in the past. It is presumed that, at the time of cleaning, the stone would have been left in a sound condition. There is also evidence that buildings cleaned in the past have been re-cleaned, perhaps on several occasions. The longer term effects of multiple cleaning are unclear; however, cumulative damage may be expected to occur.

The need for objective and quantifiable evidence of the impact that stonecleaning has on the incidence of stone decay and stone replacement is required if proper guidance is to be provided, to both practitioners and the stone industry (quarries, suppliers and contractors) on the scale of the problem. This information quantifies the scale of sandstone decay and helps to determine the amount of stone that will in future be required for repair purposes. Data on the expected rate of decay can be used to provide projections with regard to future repair costs.

The following information is a summary from the results in the research report. This information is copyrighted to The Robert Gordon University and Historic Scotland. If quoted the information should be referenced as:
Young, M.E., J. Ball, R.A. Laing, P. Cordiner and J. Scott, 2000.
An investigation of the consequences of past stonecleaning intervention on future policy and resources.
Report to Historic Scotland.

Copies of the research results can be obtained from Historic Scotland, at:

Summary of conclusions

1. Results relating to façade condition are from statistical analysis of a database of pairs of essentially identical building façades. Within each pair, one façade had been stonecleaned and the other had not. All façades were surveyed to determine their condition and map the extent and distribution of decay. Comparison of an essentially identical set of cleaned and uncleaned building façades allows statistical analysis of the condition of stonecleaned façades relative to uncleaned façades.
2. Façades that had undergone stonecleaning have been found to have a significantly higher incidence of stone decay and repair compared to uncleaned façades. On average, the amount of decay and repair on cleaned façades was approximately twice that found on uncleaned façades. The biggest difference between cleaned and uncleaned façades was in the amount of repair that had been carried out, with substantially more repair on cleaned façades. This is thought to indicate a relatively higher level on intervention on these façades.
3. Although there was a large degree of scatter in the data, for uncleaned façades an average rate of decay of approximately 0.6% of surface area per decade was found. For individual sandstone types rates of decay ranged from about 0.05% to 1% surface decay per decade. In extreme individual cases, rates of decay ranging from as low as 0.03% up to 2.4% per decade were measured on uncleaned buildings.
4. Physically cleaned façades had an average decay rate of 9.5% decay of surface area per decade following cleaning. Chemically cleaned façades had an average value of 4.7% decay of surface area per decade. Rates of decay for individual sandstone types were found to range from slower than before cleaning, up to rates that were approximately 10 times as rapid, in some cases even higher.
5. Results indicate that stone decay is most accelerated in the years immediately following cleaning. The effects of stonecleaning on decay rates appear to approach the normal background levels after about 20 years. However, a significant amount of "extra" decay can occur during this 20 year period.
6. Many practitioners have changed their methods of cleaning in the past decade in response to planning department guidelines, information and experience with respect to the potential damage which may be caused to façades. Harsher forms of chemical cleaning and higher pressure abrasive methods are increasingly avoided. The overwhelming majority of practitioners have observed some detrimental effects of stonecleaning including erosion, increased biological growth, discoloration, spalling and efflorescences. Those avoiding stonecleaning often cited how earlier witness to the effects of stonecleaning has led them to believe that the risks outweigh the benefits.
7. The opinion of practitioners regarding replacement or repair to decayed building stone varied markedly. Practitioners involved in stone repair tended to recommend a greater degree of intervention. Among architects, planners and developers there was wide variation in the degree of intervention recommended.
8. The most common reasons given for stone replacement or repair were aesthetic. Problems with the stone's ability to function or potential structural problems were generally only cited where the stone's perceived role in the façade was threatened.
9. Practitioners agreed that re-pointing of façades and replacing of decayed stone with natural stone should be carried out at some stage. Relatively large numbers of practitioners rejected the use of stonecleaning (although some 45% do undertake stonecleaning), water repellents, biocides and plastic repairs.
10. Practitioners attach great importance to correct selection of materials for repair, matching original tooling and original stone colour. Matching the present stone colour was generally not regarded as being important.
11. Evidence from the mapping exercise indicated clearly that stonecleaning methods can accelerate stone decay in the medium term. Such decay can for instance be linked to residues from chemical cleaning agents accelerating decay of the stone through salt crystallisation. In addition, results from the survey of specifiers suggests that only a small amount of decay may result in a desire for the repair and replacement of defective stone. Therefore, a cost for repair work must be considered prior to cleaning taking place.
12. Where cost modelling indicates an overall medium to long term financial loss, stonecleaning must be regarded as inadvisable regardless of possible short term gains. In any case, short term gains incorporated in the model rely on stonecleaning being perceived as a positive alteration. Where cleaning is recognised as having a strong potential to damage a stone façade, gains may be reduced or removed.
13. Where the model indicates an overall medium to long term financial gain results from the mapping exercise should still be used to help plan for the eventuality of stone decay requiring repair. A financial gain (due to market performance) in the short term will reduce over time as a building re-soils and will not in any way avoid associated stone deterioration. The initial costs of cleaning (with all associated costs) must be considered together with medium term repair costs to facilitate proper cost planning. The costs of repair will in many cases be considerable, and the resources required to deal with that work must be anticipated.
14. Stonecleaning must be positioned within a cost model at the beginning of a cycle of work. Stonecleaning is often performed to help realise benefits in market price or marketability, suggesting that different parties will often be required to deal with repair work. Those responsible for cleaning must be willing to take responsibility for the implications of cleaning, which are significant with regard to the townscape and community.
15. Although previous work has shown that increments in market price can be expected following cleaning, due to perceived improvements in appearance, should the results from this and similar studies find common acceptance, it is likely that the association of cleaning with subsequent cost and decay will reduce or remove market increments. Therefore, the consideration of cost planning for the repair of decay may well replace current cost/ benefit analysis.
16. Results from the mapping exercise, laboratory studies and cost modelling produce strong evidence that stonecleaning must not be regarded as a single stage, short term event. The need has emerged for reliable and sustainable sources of stone material and for skilled craftsmen and quarry workers. The long-term effects of stonecleaning on requirements of stone for repair will be significant. The resource implications in terms of material alone will be similarly great and may not be facilitated by current stone resources. Planning for materials sourcing is currently required to ensure that necessary repair is not hindered by a basic lack of resources, funds or skills.
17. In the absence of effects of stonecleaning, northerly facing building façades were found to have a slightly higher incidence of stone decay, algal growth and salt efflorescence than façades with a southerly orientation. This effect is thought to be related to increased duration of moisture retention in relatively shaded areas of stone.
18. Chemical analysis of soluble salts in building sandstones found gypsum (calcium sulphate) to be the most common salt in soiled buildings from urban areas. Sodium chloride (common salt) was more common in coastal areas. In chemically cleaned buildings residues of stonecleaning chemicals (including sodium, phosphate and fluoride) were detected. Relatively more salt efflorescence was present on chemically cleaned façades compared to uncleaned or physically cleaned façades. Indeed, on porous sandstones, chemical residues are inevitable following chemical cleaning as it is not possible to remove all applied chemicals even with the most thorough washing off.
19. In compiling the data necessary to conduct this research project, the greatest difficulties were experienced with gathering information on the cleaning methods and dates of cleaning or repair to building façades. In the absence of detailed and reliable evidence on the cleaning methods employed it has often been possible only to assign façades to broad categories such as "chemical cleaning". This illustrates the need for accurate recording of interventions on building façades. Without this data, future investigations of the consequences of novel treatments or repair methods will not be possible and the potential for long-term investigations and the provision of useful guidance to practitioners will be impaired.

Further publications

Results from this research project and details of the methodologies used are also available in the following publications:

Ball, J., R.A. Laing and M.E. Young (2000).
Stonecleaning: Comparing perceptions with physical and financial implications.
Journal of Architectural Conservation. Vol 6(2). pp 47-62.

Ball, J., M.E. Young and R.A. Laing (2000).
Rapid field assessment of stone decay on buildings and monuments.
In: 9th International Conference on Deterioration and Conservation of Stone. Vol. 2, Ed. V.Fassina. Venice, 19-24th June 2000. Elsevier, Amsterdam. ISBN 0-444-50517-2. pp 13-21.

Laing, R.A., J. Ball, J. Scott and M.E. Young (2000).
The implications of stone cleaning for planned building maintenance.
In: 9th International Conference on Deterioration and Conservation of Stone. Vol. 2, Ed. V.Fassina. Venice, 19-24th June 2000. Elsevier, Amsterdam. ISBN 0-444-50517-2. pp 813-817.

Young, M.E., J. Ball and R.A. Laing (2000).
Quantification of the long-term effects of stonecleaning on decay of building sandstones.
In: 9th International Conference on Deterioration and Conservation of Stone. Vol. 2, Ed. V.Fassina. Venice, 19-24th June 2000. Elsevier, Amsterdam. ISBN 0-444-50517-2. pp 179-186.

Young, M.E., J. Ball and R.A. Laing (in press).
Quantification of the decay of building sandstones.
In: Weathering 2000. Belfast, 26-30th June 2000.

Ball, J. and M.E. Young (2000).
Mapping the decay and weathering of stone: A technique for the assessment of large numbers of buildings.
In: Proceedings of the New Millennium International Forum on Conservation of Cultural Property. 5-8th December 2000. Deajeon, South Korea. pp 134-147.

Young, M.E. and J. Ball (2000).
Effects of conservation treatments on the weathering of sandstone and granite buildings in Scotland.
In: Proceedings of the New Millennium International Forum on Conservation of Cultural Property. 5-8th December 2000. Deajeon, South Korea. pp 12-29.

Young, M.E., J. Ball and R.A. Laing (1999).
Survey results regarding attitude towards stonecleaning of building sandstones.
In: Durability of Building Materials and Components 8. M.A. Lacasse and D.J. Vanier (eds.). Vancouver, Canada. 30th May - 3rd June 1999. NRC Research Press: Ottawa. pp 635-644.

Laing, R.A., A. Al-Hajj, J. Ball, J. Scott and M.E. Young (1999).
Stone cleaning: A life cycle cost model.
In: Durability of Building Materials and Components 8. M.A. Lacasse and D.J. Vanier (eds.). Vancouver, Canada. 30th May - 3rd June 1999. NRC Research Press: Ottawa. pp 1739-1745.

For further information contact Dr. Maureen Young at m.young@mailer.rgu.ac.uk