Cover Image

PAPERBACK
$97.25



View/Hide Left Panel
Click for next page ( 330


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 329
The Evaluation of Stone Preservatives CLIFFORD A. PRICE This paper reviews approaches to the evaluation of stone preservatives and groups them into categones: evaluation without expenment, evaluation by natural weathenng, tests intended primarily for use on untreated stone, tests intended specifically for treated stone, and performance criteria. The paper does not review individual techniques in detail. Instead, it examines the think- ing that underlies the techniques and questions both the validity of some procedures and their relevance to practical conservation problems. It concludes by outlining a policy for the use of stone preservatives. The relentless deterioration of ancient monuments and historic build- ings is a matter of pressing concern throughout the world. Numerous conservation strategies have been proposed, often involving the appli- cation of a preservative. Preservatives typically are colorless liquids applied to the building fabric in an attempt to strengthen that fabric and to protect it against further decay. The search for reliable preser- vatives has extended over many years but has been largely fruitless: No treatments have been found that are widely applicable and that have been proved effective over periods of several decades. Clifford A. Price is Head of the Stone Conservation Section, Building Research Estab- lishment, U.K. Department of the Environment, Watford, England. This paper has been written as part of the research program of the Building Research Establishment of the Department of the Environment, United Kingdom, and is published by permission of the Director (Crown Copyright 1980, Building Research Establishment, Department of the Environment J. 329

OCR for page 329
330 CONSERVATION OF HISTORIC STONE BUILDINGS Me increasing involvement of scientists in conservation problems is giving rise to many new suggestions for preservatives. Some of these materials (organosilicon compounds, for example) have been available only in recent years, and there is no long-term experience of their effectiveness. It would be grossly irresponsible to apply any unproven material to masonry of high artistic or historical importance, and yet it is just such masonry that is in the most urgent need of treatment. Reliable procedures for the rapid evaluation of potential preservatives are therefore essential. The development of evaluation procedures is by no means straight- forward. Historic buildings and monuments are constructed from a bewildering variety of materials and are exposed to a comparable va- riety of environmental conditions. Preservative treatments accordingly must take many forms to counteract a wide range of decay mecha- nisms. Just as there can be no "universal preservative," there can be no universal evaluation procedure. A glance through the proceedings of conservation conferences will reveal the many techniques that have been adopted by researchers for evaluating preservatives. It is not the purpose of this paper to review in detail the techniques themselves. Instead, the paper sets out to examine the thinking that underlies the techniques and to question both the validity of some evaluation procedures and their relevance to practical conservation problems. For convenience, the paper groups current evaluation procedures into a number of different categories, which correspond roughly with increasing levels of sophistication. The paper is unashamedly a personal view, and the categorization is only one of many that would be possible. Inevitably, there are areas of overlap between categories, and there is no suggestion that any one individual or institution will operate deliberately within one category alone. On the contrary, most researchers will draw from most cate- gories during their evaluation of any particular preservative. The evaluation of novel materials and procedures is by no means exclusive to conservation scientists. It is hoped that this overview of the problem as it confronts conservation will stimulate discussion and invite comment from those with relevant experience in other fields. Although the paper deals specifically with the evaluation of pre- servatives for stone, much of it is equally relevant to other porous building materials. EVALUATION WITHOUT EXPERIMENT The evaluation of preservatives without recourse to experiment rep- resents perhaps the lowest level of sophistication. Nevertheless, it can

OCR for page 329
Evaluation of Stone Preservatives 331 play an important role. It is practiced, unconsciously maybe, when a preservative is rejected on grounds of past experience or of existing scientific knowledge. For example, polyiethylene glycoll is sometimes used to consolidate stone carvings before indoor exhibition. It could reasonably be rejected without experiment for outdoor use since it is soluble in water. This type of evaluation leading to rejection is per- fectly acceptable, even if somewhat trivial. On the other hand, eval- uation without experiment is not acceptable when it argues for the acceptance of a preservative. Although this sounds obvious, it is not always clearly seen. In Eng- land, for example, treatments based on lime are being used on several important limestone buildings, notably Wells and Exeter cathedrals. Some protagonists point out that the treatment is based on the very material from which the stone is derived; they argue that the treatment must be safe and effective since it puts back into the stone the same substance that it has lost through decay. (By the same token, they argue that organic impregnants must be hams since they introduce materials foreign to the stone.) The argument has no logical validity, for the microstructure of lime-treated stone may be quite different from that of the original stone. Nevertheless, the argument has enor- mous emotional appeal and can easily sway the nonscientist. This example is not intended to cast a slur on the treatment itself, which has so far proved extremely successful. It is the underlying argument tacitly adopted by some of the treatment's advocates that is under criticism here. The argument is reminiscent of the saying "I've never tried it, because I don't like it," except that here it is put in the converse: "Because I like it, I don't need to test it." EVALUATION BY NATURAL WEATHERING The evaluation of preservatives by exposing treated specimens to the weather is an obvious procedure. IS the long tea it is also the best, for no other procedure can carry 100 percent assurance of demonstrat- ing the performance of preservatives finder natural weathering. Natural exposure trials fall into two categories. On the one hand, small blocks of stone may be exposed to the weather and brought periodically into the laboratory for washing and detailed examination. On the other hand, trials may involve the treatment of stonework that is an integral part of a building. Tests of the first type are surprisingly sensitive and may give a preliminary indication of a treatment's ef- fectiveness after only a few months. On the roof of St. Paul's Cathedral in London, for example, exposure trials are in progress using limestone specimens approximately 100 mm x 80 mm x 60 mm. Untreated

OCR for page 329
332 CONSERVATION OF HISTORIC STONE BUILDINGS blocks lose approximately 1 g (0.1 percent) every 6 months, which is easily detected on a laboratory balance. The particular drawback of this type of test is that it may not accurately reflect the way that the treatment would behave in a building, where joints and structural movement will have an influence. Preservative trials on buildings can range from the treatment of a few blocks to treatment of an entire facade. The condition of the experimental panel is carefully recorded before treatment, as is the condition of an adjacent panel that is to be left untreated. The com- parative condition of the two panels thereafter provides an indication of the treatment's effectiveness. The comparison is where difficulties may arise. Unless the perform- ance of the two panels is so different that there is no possible scope for misinterpretation, a quantitative and objective method is required for monitoring the condition of the panels. Surprisingly little attention has been given to this area. Individual tests, such as rebound tests or measurement of surface hardness, have been used, but the literature has carried few accounts of systematic, quantitative monitoring. One of the reasons may be the frequent requirement that tests be nonde- structive and also the necessity for panels to be large enough to permit statistically meaningful analysis. In practice, monitoring tends instead to be qualitative and somewhat subjective. It will be based, for example, on appraisals of surface hardness made by running the finger across the two panels and comparing the amounts of loose stone that it removes the test is reasonably satisfactory at first but meaningless once all the loose stone has been removed from the control panel. Despite the undeniable benefits of evaluation by natural exposure, there is a further drawback. A single trial can provide, at best, infor- mation on the behavior of preservatives on one type of stone in one particular environment. Extrapolation to other stones and environ- ments is unreliable. To set up a thorough appraisal of a preservative on a range of stones under a variety of conditions of exposure is enor- mously time-consuming and costly. And when the trials have been established, significant results may not emerge for many years- by which time new preservatives will have been developed, and the whole sequence must start again. To summarize, natural exposure has a central part to play in the evaluation of preservatives. Under no circumstances can it be omitted, for it is the only "true" test; Nevertheless, its shortcomings are such that there is an inevitable demand for accelerated laboratory tests. These are the subject of succeeding sections. . .

OCR for page 329
Evaluation of Stone Preservatives TESTS PRIMARILY FOR UNTREATED STONE 333 Many testing procedures are available that were developed originally for use on untreated stone. Some of these procedures measure prop- erties that are directly relevant to the use of stone in buildings. For example, bending strength is relevant to a stone's suitability for use in lintels, and coefficients of expansion are needed to make allowance at the design.stage for movement. On the other hand, most test pro- cedures are intended to provide guidance on a poorly defined and elu- sive property: durability. The procedures either simulate natural decay mechanisms j such as salt crystallization and freezing tests) or measure physical properties that may be expected to have a bearing on durability (such as pore structure). It is understandable that durability tests for untreated stone should be considered when seeking tests for evaluating preservatives. Since it has been shown that coarse-pored stones are generally more durable than fine-pored stones, it is reasonable to expect that a treatment that decreases the proportion of fine pores should improve a stone's weath- ering resistance. Similarly, a treatment that improves a stone's per- formance in a crystallization test might be expected to increase re- sistance to salt crystallization in a building. However, matters are seldom so simple. Almost invariably the treatment will affect the test results in such a way as to make them highly questionable, if mean- ingful at all. As a case in point, porosimetry techniques frequently require knowledge of the contact angle between stone and invading fluid. Unless due allowance is made for the change in contact angle upon treatment, direct comparison of results is meaningless..The inter- pretation of results is further complicated if, as is often the case, the contact angle for the untreated stone lies on one side of 90 and the angle with the treated stone lies on the other. Similarly, water repel- lency in treated specimens can nullify the sigruTicance of the salt crystallization test. It is unfortunate that a great many investigators have overlooked these difficulties when evaluating preservatives. Even when the prob- lems have been recognized and ways of overcoming them proposed, the modified test procedures may still be unsatisfactory. Vacuum sat- uration, for instance, is sometimes proposed to overcome the water repellency of specimens in the crystallization test. However, this ap- proach is unsatisfactory because the solution still will be unable to enter pores with an effective diameter below a certain value. It is essential that a very rigorous scrutiny be made of any test method intended primarily for use on untreated stone if the procedure is to be

OCR for page 329
334 CONSERVATION OF HISTORIC STONE BUILDINGS extrapolated to evaluate treated stone. In some cases the procedure will be equally meaningful on treated and untreated stone. In the majority, it will not. TESTS SPECIFICALLY POR TREATED STONE Tests intended specifically for treated stone may be grouped into three categories: those aimed simply at characterizing the extent and nature of the treatment, those using accelerated aging chambers, and those aimed at assessing the extent to which treatments meet stated objec- tives. They will be considered in turn. Charactenzation of Treatment It is often desirable to record certain characteristics of a treatment even if they provide no direct information on the treatment's effectiveness. These characteristics may include the depth to which the treatment has penetrated the stone, the amount by which it has filled the pore space, the degree of polymerization (if relevant and practicable!, and the effect that the treatment has had on the stone's appearance. Data like these are needed to build up a full picture of a treatment but, because they do not relate directly to effectiveness, they are not con- sidered in detail here. Accelerated Aging Chambers Accelerated aging chambers frequently are used to evaluate the du- rability of organic construction materials like paints and plastics. They have been applied by a number of researchers to the evaluation of stone preservatives. The chambers typically are capable of simulating a va- riety of environmental conditions, either singly or in combination. The conditions may include wetting and drying, heating and cooling, exposure to ultraviolet and infrared radiation, and immersion in dilute acids and salt solutions. The results of tests in these chambers certainly are useful in indicating the relative resistance of different treatments to specific environments, although some of the problems of judging tests for untreated stone may also apply here. The main drawback of accelerated aging chambers is that they can- not give 100 percent assurance that they accurately reflect the proper balance and interaction of natural weathering agents. Nor can there be any reliable indication of the "acceleration factor." In other words, at the end of a test, one is obliged to ask, "So what? Does the test

OCR for page 329
Evaluation of Stone Preservatives 335 really show what would happen to the specimens after n years of natural weathering? And how large is n?" Accelerated aging chambers certainly must not be seen as all-suf- ficient. Instead, they should be seen as one element of the next category of tests assessing how well treatments meet their objectives. In this context such chambers can be very useful, particularly when used to simulates just one set of environmental conditions. Tests for Assessing How Well Treatments Meet Objectives Instead of looking for an overall, nonspecific preservative action, one may more rationally assess the extent to which a treatment fulfills its stated objectives. One is then making a quantitative measurement of a clearly defined parameter, rather than appraising an ill-defined con- cept. Such an approach has the additional advantage that it can directly take into account differing decay mechanisms. Some examples will help to clarify the approach. One of the functions of a stone preservative is to strengthen weath- ered stone, both to protect it against mechanical damage and to make it more resistant to decay mechanisms. One logical element of an evaluation procedure, therefore, must be the measurement of strength, both before and after the treatment is applied. Strength in tension is normally the most appropriate parameter, since the majority of decay mechanisms place the stone in tension. Another property frequently required in a preservative is the ability to protect limestone from attack by acidic air pollutants. The capability of a preservative in this respect may be determined, for example, by exposing treated specimens to moist air containing sulfur dioxide. The amount of calcium sulfite or sulfate formed is an inverse measure of the treatment's effectiveness. A final example concerns the ability of a preservative to prevent damage from salt crystallization. In principle, a suitable test would entail treating salt-contaminated specimens with the preservative and then subjecting the specimens to relative humidities that fluctuated above and below the equilibrium relative humidity for that particular salt. This approach may also be used to assess the vulnerability of pre- servatives to specific risks. One such risk might be degradation of the preservative under ultraviolet light, a risk that may readily be evaluated in an accelerated aging chamber. Another example concerns the pos- sibility of fracture at the interface between treated and untreated stone as a result of differential moisture and thermal movement. This risk

OCR for page 329
336 CONSERVATION OF HISTORIC STONE BUILDINGS can be investigated by treating specimens on one face only and then subjecting them to cycles of oven cIrying followed by rapid immersion in cold water. An associated investigation might entail direct mea- surement of the thermal and moisture expansion of treated and un- treated specimens. This approach was embodied in the early thinking of the RILEM 25 PEM Working Group j Reunion Internationale des Laboratoires d'Essais et de Recherche sur les Materiaux et les Constructions), which met between 1973 and 1978 to establish agreed testing methods for stone. Every test method put forward was considered in the light of questions like "Is this a useful test for measuring the extent to which a treatment strengthens stone?" or "Is this a useful test for measuring the extent to which a treatment immobilizes salts?" In the later stages of the group's work, the purpose of the framework became blurred by the introduction of the all-embracing questions, "Is this a useful test for assessing the durability of treated stone?" The individual questions had been intended originally to build up an overall picture of a treat- ment's effectiveness, and the late introduction of a specific question on durability tended to confuse this intent. The approach is not perfect, not least because it tends to oversimplify decay mechanisms and to treat them in isolation. Nevertheless, it demands a thoughtful analysis of the functions that a preservative is intended to perform in any given situation, and this alone has much to commend it. PERFORMANCE CRITERIA The approach of the preceding section may be given a further degree of sophistication by establishing performance criteria for stone pre- servatives. Although the concept of performance specification has re- ceived considerable attention from the construction industry in recent years, its application to stone preservation appears to have been limited to studies by Gauri and by Sleater Isee Bibliography under Performance Criterial. A performance specification is essentially a statement of what a product is required to do, as opposed to a prescriptive specification, which defines the product in physical, chemical, and geometric terms. In its most elementary form, a performance specification for a stone preservative might read, "The preservative shall lengthen the useful life of stone without affecting its appearance." Such a specification would have no practical value, and a performance specification thus evolves into a set of more specific criteria, such as "Treated calcite

OCR for page 329
Evaluation of Stone Preservatives 337 specimens, exposed to a water-vapor-saturated 10.8 ppm SO2 dynamic atmosphere, shaU not show more than half the reaction of similarly exposed untreated specimens of the same material . . ." or "The surface of the preserved stone should not spell from the stone, decompose, change in appearance, or otherwise deteriorate when exposed to 1200 test cycles of water condensation/evaporation." Such criteria are based on the approach of the preceding section but go one stage further by laying down quantitative limits for the test results. They have an air of arbitrariness, which is inevitable in view of the newness of the approach, but further usage should lend weight to the limits prescribed or else provide evidence to support amended values. It would be tempting to assume that compliance with a compre- hensive set of performance criteria would provide complete assurance that a treatment was effective and risk-free. But how comprehensive must "comprehensive" be? No practicable specification can possibly cover every aspect of performance, and loopholes must inevitably exist. It is particularly difficult to ensure that the test requirements are ad- equate when one does not know the chemical nature of every possible preservative. And it is almost certain that no product could be found that would meet all the criteria; certainly Sleater found no treatment, out of 56 tested, that met his requirements. Nevertheless, as he con- cluded, the assessment was valuable as a preliminary screening of materials for field testing. The wheel has turned full circle, from field testing in an earlier section, through performance testing, and back to field testing. The next section examines some of the implications for both conservators and conservation scientists. DISCUSSION The preceding sections have given credence to the view that field testing, despite its limitations, is ultimately the only valid test for a stone preservative. Laboratory testing is useful for eliminating treat- ments that are unlikely to be effective, but at the present state of knowledge laboratory testing alone is insufficient. This is of little comfort to the conservator faced with pressing practical problems, to whom most of this section is devoted. For the researcher it means that attention must be given to the development of improved monitoring procedures for site trials. Equally important is the need for detailed recording of site trials and for widespread coordination of results. It is easy for the researcher to

OCR for page 329
338 CONSERVATION OF HISTORIC STONE BUILDINGS publicize the results of trials in which the treatment is proved effective. It is less attractive to hull, but no less important, to publicize trials in which disappointing results are obtained. There has been a tendency in the past to let unsuccessful trials fade quietly into obscurity, perhaps to avoid embarrassment. In order to build up a picture of what con- situtes an effective preservative, however, it is essential that knowledge of all site trials be widespread and detailed. But what advice can one give to the conservator, to whom the entire concept of treatment evaluation may seem irrelevant? "The stone is in such appalling condition that whatever I do cannot make it worse" is an argument often used to justify the use of an unproven treatment. fin fact, the argument is overworked and sometimes serves as a salve to the conservator's conscience.) The conservator must be given guid- ance on the scope and limitations of various preservatives, and it is up to-the scientist to provide the technical element of that guidance. In for 'adulating conservation policy, a number of factors must be taken into account. First, stone preservatives are no substitute for common sense. In the case of external stonework, attention must be given to protection against rain, perhaps through the repair of protec- tive features such as canopies or cornices or through the repair of open joints in masonry above. In extreme cases it may be appropriate to move the object in question to a controlled indoor environment. Second, the use of stone preservatives may not be necessary if the stonework is easily accessible to a skilled work force. In such cases it may be preferable to maintain the stonework in sound condition by using adhesives or dowels to secure pieces in danger of falling away, by filling cracks with suitable mortar, by securing loose flakes with a judicious packing of mortar, and by fining water traps. Work of this nature can be undertaken whenever it is seen to be necessary, although it will not be sufficient if the stone is in very friable condition. Third, it must be accepted that stone preservatives, like medicines, do not offer fin ending life. The time inevitably will come when the stone begins to deteriorate -again, and retreatment may become nec- essary. No preservative should ever be used that precludes the possi- bility of eventual retreatment. In deciding whether or not to use a preservative in a particular instance, it may be helpful to draw a distinction between stonework that is exceptionally valuable, either artistically or historically, and stonework that is of a more routine character. In the latter case the conservator may reasonably proceed with the-application of a preser- vative whose record in field and laboratory trials so far has been sat- isfactory. In the case of valuable stonework, it is better for the conser-

OCR for page 329
Evaluation of Stone Preservatives 339 vator to assess whether the object's rate of decay is so great that unacceptable losses wiD take place within the next few years. If it is, the use of a preservative with a good record to date could not be regarded as irresponsible. If, on the over hand, the decay is reasonably slow, it is safer to leave the object untreated for the time being and to wait for experience to accumulate on other objects. This paper inevitably has entailed presenting a lot of uncertainities, but one certainty can be cited. When asked whether one con be sure that a particular preservative will be effective and cause no harm, the answer is a categorical "no." However, it is hoped that this paper will stimulate further work, so that a rider may eventually be added: "No, but the risk of ineffectiveness and harm is so low that the treatment can be used without anxiety." BIBLIOGRAPHY [This bibliography is not intended to be comprehensive. Its purpose is to cite a few papers that contain typical examples of the various evaluation categories considered above.] Evaluation by Natural Weathenng Clarke, B.L., and J. Ashurst. Stone Preservation Experiments. Building Research Estab- lishment: Watford, England, 1972. Moncrieff, A. The treatment of deteriorating stone with silicone resins: Interim report. Studies in Conservation 21:179 {1976J. Tests Primarily for Untreated Stone Arnold, L., and C.A. Price. The laboratory assessment of stone preservatives. Proc. Int. Symposium on the Conservation of Stone, Bologna, 1975, 695. Cormerois, R. Traitments preventif et curatif des structures. Proc. Int. Symposium on the Conservation of Stone, Bologna, 1975, 6.5. Marschner, H. Application of salt crystallization test to impregnated stones. Proc. REHEM/ UNESCO Symposium on Deterioration and Protection of Stone Monuments, Paris, 1976, 3.4. Characterization of Treatments deWitte, E., P. Huget, and P. Van Den Broeck. A comparative study of three consolidation methods on limestone. Studies in Conservation 22:190 {1977J. Accelerated Aging Chambers Cooke, R.U. Laboratory simulation of salt weathering processes in arid environments. Earth Surface Processes 4:347 {1979~. Furlan, V., and F. Girardet. Methode d'essai de vieillissement accelere pour l 'etude des

OCR for page 329
340 CONSERVATION OF HISTORIC STONE BUILDINGS traitements des pierres. Proc. Int. Symposium on the Conservation of Stone, Bologna, 1975, 713. Sleater, G. A laboratory test programme for stone preservatives. Proc. RILEM/U~SCO Symposium on Deterioration and Protection of Stone Monuments, Paris, 1976, 6.13. Tests for Assessing How Well Treatments Meet Objectives Arnold, L., D.B. Honeybome, and C.A. Price. Conservation of natural stone. Chemistry and Industry 17:345 { 1976. Gauri, K.L., et al. Reactivity of treated and untreated marble specimens in an SO2 atmosphere. Studies ~ Conservation 18:25 {1973~. REHEM Working Group 25 PEM. Recornm ended tests to measure the deterioration of stone and to assess the effectiveness of treatment methods. Proc. Int. Symposium on De- terioration and Protection of Stone Monuments, Vol. 5. Paris, 1976. Performance Criteria Gauri, K. L . , J. A. Gwin, arid R. K. Popli . Performance criteria for stone treatment . Proc . 2nd Int. Symposium on the Deterioration of Building Stones, Athens, 1976, 143. Sleater, G.A. Stone preservatives: Methods of laboratory testing and preliminary per- formance criteria. Technical Note 941. National Bureau of Standards: Washington, D.C., 1977. Policy for Use of Stone Preservatives Jedrzejewska, H. Some comments on ethics in conservation of stone objects. Proc. RILEM/ UNESCO Int. Symposium on Deterioration and Protection of Stone Monuments, Pans, 1978, 7.13. Price, C.A. Brethane stone preservative. Building Research Establishment current paper, in press. Torraca, G. Treatment of stone in monuments: A review of principles and processes. Proc. Int. Symposium on Conservation of Stone, Bologna, 1975, 297.