RILEM SEMINAR ON DURABILITY
Durability of building materials: durability research in the United States and the influence of RILEM (1) on durability research (2) James R. Wright Deputy Director National Engineering Laboratory National Bureau of Standards President of RILEM (1982-1985)
G. Frohnsdorff Chief, Building Materials Division National Engineering Laboratory National Bureau of Standards Gaithersburg, Maryland 20899, USA Member of RILEM Technical Advisory Group
This paper is based on the text of a talk given at a RILEM Seminar on Durability of Building Materials. It reviews some of the recent non-proprietary research on the durability of building materials carried out in the US. It also reviews activities in RILEM which have stimulated the generation or dissemination of knowledge on durability of building materials. The paper emphasizes the need for international collaboration in durability research.
1. INTRODUCTION In this age of rapid technological progress, it is easy to underestimate the importance of the changes taking place in building technology. The changes are responses to the needs of society for new construction methods and materials and to other pressures, especially for energy conservation and environmental protection. Examples of changes involving building materials are developments in solar heating and cooling [1], high strength concrete [2], new types of roofing membranes [3], and paints which are free from toxic pigments and from solvents which may contribute to air pollution [4]. While the changes mentioned are generally beneficial, they may cause problems in their implementation. This is often the case when traditional building materials have to be used under new conditions, or when new materials without a long performance history are put into use. The difficulty of evaluating new developments in building (1) InternationalUnion of Testingand ResearchLaboratoriesfor Mat6rials and Structures. (2) Based on a talk by James R. Wright. RILEM Seminaron Durability of BuildingMaterials,Tokyo, Japan,October13, 1984,
technology has led, in the last two decades, to the development of the performance concept ([5], [6], [7]). Two organizations with which both authors are associated, RILEM and the US National Bureau of Standards (NBS), have played important roles in promoting the performance concept and, more generally, in promoting international cooperation in building research. It is hoped this paper will further encourage collaboration in research on the durability of building materials. This paper reviews some of the research on the durability of building materials which has been carried out at the NBS and in other US laboratories in recent years. It also draws attention to R ILEM's role in promoting international collaboration in research on the durability of building materials. It must be noted that the review of US research on building materials durability reflects the authors" perspective and is not claimed to be fully comprehensive. In particular, within the United States, much of the research carried out by private industry is proprietary. The research described in this paper comes only from organizations which make their research results available in the open literature. Appendix I gives names and addresses of several organizations which perform research on building materials.
Mat6riaux ct Constructions 0025-5432/85/03 205 10/$ 3.00/9 BORDAS-GAUTHIER-VILLARS
205
Vol. 18 - N ~ 105 - Mat&iaux et Constructions
DURABILITY INFORMATION AND THE BUILDING CYCLE
_ovol
1. P R O G R A M M I N G 9 Define building
,
6. R E M O V A L OR 9 Decide
--..,
I
on use
9 Identify configuration
- "~3,
DETAILED OEEIGN - Develop plans .and specifications
"- -
5. O C C U P A N C Y . - ~ 9 Maintain ~r~ ~ functionality
~
I
~176
2. P R E L I M I N A R Y ji.
CONSTRUCTION 9 Build building
Fig. 1. - T h e d u r a b i l i t y i n f o r m a t i o n s y s t e m and t h e building cycle. As indicated by the radial arrows, all stages of
the cycle require information on building materials durability and several contribute to the durability information system. Improved feedback could make the system more effective.
2, D U R A B I L I T Y A N D THE B U I L D I N G CYCLE The need to make decisions about the durability of building materials occurs in all stages of the building cycle. The performance concept provides a logical framework for making many of the complex decisions in building technology, but it is difficult to "apply it to durability performance, particularly of new materials. This is because of the difficulty of evaluating durability performance in shortterm tests, and the lack of generally-accepted durability test methods or methods for prediction of service life. The term service life is usually to be preferred to durability; because it is more precise and more in accord with the need for predictions in selecting building materials [8]. While no two buildings are exactly alike, several recognizably similar stages can be seen in their life cycles [9]. A breakdown of the building cycle into six stages is shown in figure 1. The first, Programming, is the stage in which the decision to build is made. It is followed by Preliminary Design, in which the form of the building is decided, and Detailed Design, in which the specifications are completed. Next come the Construction phase, the Occupancy phase and, last, Removal or Rehabilitation. The cycle may be reentered by a return to programming when substantial changes are to be made; an example is in the adaptive reuse of a monumental structure. Figure 1 shows the construction cycle in relation to a "durability information system" which includes all of the available knowledge needed for decisions concerning the durability performance of building materials. It is one of many parts of the total, integrated project information system [9] which the building process requires. The durability information system has been shown in this way to emphasize its importance to decisions in all stages of the building cycle. The radial arrows indicate the flows of information between participants in the cycle and the information system. At this point, it will be helpful to consider types of information and what must be done to use them to full advantage in the building process, including selection of materials for durability. Nine types of information which appear to be more or less idstinguishable are listed in table I in rough order of increasing potential for providing a comprehensive and reliable representation of durability
206
knowledge. The types of expert systems are discussed in Section 9. At present, the first five of these appear to be the basis for most decisions on the durability of building materials. These five will still be needed in the future, but increasing use of the other more systematic types of information, must be expected and encouraged [10].
3. M E T H O D O L O G I E S FOR SERVICE LIFE PREDICTiON In the early 1970"s, because df their interest in generic rather than specific test methods, Masters and his colleagues at NBS sought a general methodology which might be adopted for evaluating the service life of any building material or component under any specific conditions [8]. This led to the establishment of an American Society for Testing and Materials (ASTM) building materials standard of a new type in 1978. The new standard was ASTM E 632, Standard Practice for Developing Accelerated Tests to Aid Prediction of the Service Life of Building Components and Materials. It formalized the logic of accelerated testing and indicated the steps which should be taken in setting up credible tests for service life prediction. ASTM E 632 has been important in drawing attention to the difficulties of predicting the service life of any material, and it has been adopted as one .of the working documents in the joint RILEM/CIB (i) Committee (71PSL/W-80) on Prediction of Service Life. This committee will be mentioned again in Section 10. Most tests of durability of building materials are simple comparisons of the behaviors of similar objects under the same, essentially arbitrary, conditions [11]. Such tests which are represented by the left hand box in figure 2, are of little value for predicting performance under conditions other than the test conditions. They ignore the possibility of designing experiments so as to facilitate prediction of service life under a range of service conditions through mathematical analysis. It must now be emphasized [12] that changing the focus from the standard experiment to the mathematical analysis indicated on the right hand side of figure 2 is a key to improving the selection of materials for durability. This is consistent with the methodology of ASTM E 632. Examples of this approach will be given in discussing the durability of plastics and protective coatings for steel.
DURABILITY A S S E S S M E N T
MEASUREMENT
EVALUATION
(service life prediction)
t i
@
i i
D u r a b i l i t y e v a l u a t i o n . The evaluation is usually governed by the measurements, with the analysis only being considered after the data have been obtained. Much is to be gained by deciding on the details of the analysis before designing the rest of the evaluation procedure.
Fig. 2. -
(1) CIB is the International Council for Building Research. Studies and Documentation.
J. R. Wright - G, Frohnsdorff 4. D U R A B I L I T Y OF PROTECTIVE C O A T I N G S FOR STEEL Concern for protection of the environment is causing changes in the formulations of paints and coatings [4] at the same time as awareness of the large economic losses from corrosion of metals grows [13]. The changes in formulations are in reduced use of pigments containing lead and chromium and of certain solvents which are believed to contribute to air pollution. New paint formulations are normally tested on panels exposed on outdoor sites - a time-consuming and less-than-definitive way of evaluating alternative formulations. A more rapid method with higher potential for use in prediction of service life is needed. Using the methodology of ASTM E 632, Martin and McKnight of NBS have developed an approach to service life prediction of protective coatings for steel. Martin and McKnight use infrared thermography to observe the growth of corrosion spots and blisters under pigmented coatings [14]. The sensitivity a n d precision of the detection of the flaws is much enhanced over direct visual observation, and the method has high potential for automation. Analysis of the results for two different coatings, each on fifteen test panels, showed that they could be fitted well by a Weibull distribution. Further development of this analysis [12] made it possible to predict the probability of failure at any age as a function of temperature. An example of a (probability of failure temperature - time-to-failure) diagram from this work is shown in figure 3.
TABLE I H I E R A R C H Y OF TYPES OF I N F O R M A T I O N W H I C H ARE, OR C O U L D BE, USED IN T H E B U I L D I N G PROCESS
1. 2. 3. 4. 5. 6. 7. 8. 9.
Anecdotal information. Expert opinion. Documented performance. Measured properties and characteristics. Reference data. Predictive models (empirical). Predictive models (theoretical). Expert systems (shallow). Expert systems (deep).
which had been exposed to various combinations of UV radiation, heat and moisture for various times (fig. 4). The analysis showed that the degradation could be represented by an analytical expression (table II), taking into account the effects of temperature, UV radiation and moisture on the number of chain scissions. If a criterion of failure is defined in terms of the molecular weight, the expression makes it possible to predict the service life of the sheets at different temperatures and under different UV exposure conditions. Although time-consuming and requiring a well-equipped laboratory, Martin's approach shows the value of designing experiments to meet the needs of sound mathematical analysis, rather than following a simpler procedure of uncertain relationship to real exposure conditions.
5. D U R A B I L I T Y OF PLASTICS The use of plastics in buildings is growing rapidly [15]. To demonstrate a preferred approach to service life prediction for plastics, Martin of NBS and his colleagues [16] investigated the degradation of a model plastic, poly(methyl methacrylate), ( P M M A ) . In their study, they investigated the degradation of P M M A by heat, ultraviolet radiation (UV), and moisture. Molecular weight profiles across the thickness were obtained for sheets of P M M A
0
i
9
~
=~
i
LU
o
I
-
nO 15 [~
,,=, I
........................... ~ ......~__~
:
O
0. I
/
LU
,,=, 80
ooi
~:._~-
.
.
.
.
.
.
.
.
.
.
e~_....~,~q 100~
lo F - ~
/-
I
== 70
CO
C3
Z
60
o
=o R(t)=O.99_zR( = 0t) , 9 5 - " / '
20
u.
~
115 a
O 15 n-
50
UJ m
10
.~ 40
-
\
',
5'
0
i
0
20
1.5 mm DEPTH IN SPECIMEN FROM FRONT FACE
10 .P 10
I 100
I 1000
I 10000
100000
TIME (h)
Fig. 3. - Probability of failure - t e m p e r a t u r e - t i m e - t o failure d i a g r a m f o r an alkyd coating system, using 0 . 3 % area of corrosion plus blisters as t h e failure criterion.
Fig. 4. - M o l e c u l a r w e i g h t profiles of a sheet o f poly (methyl m e t h a c r y l a t e ) a f t e r exposure to heat and U V radiation f o r various t i m e s in an accelerated aging device. Since the sheet was irradiated from one side only, the profiles strongly suggest that the degradation was dependent on diffusion of some species to or from the surface; since the degradation close to the. surface was much less in a nitrogen atmosphere, it appears that oxygen was necessary for the degradative reaction,
207
Vol. 18 - N~ 105 - Mat6riaux et Constructions
A
Single membrane layer
B
Laboratory rolled seam
C
Laboratory pressed seam
t
>U) Z LU I'-
z
LU I--
<
_J LU n~
D
1
RELATIVE TIME
In service field seam
=
Fig. 5. - Pulse e c h o signals from a d e f e c t i v e s e a m and a satisfactory seam in a s i n g l e - p l y r o o f i n g m e m b r a n e as seen on a c a t h o d e ray t u b e . - Because the signal is reflected from the disbonded region, the echo from the defective seam shows only half the delay observed for the satisfactory seam which is, effectively, twice as thick.
6. D U R A B I L I T Y OF ROOFING M A T E R I A L S The world is seeing a rapid growth in the use of singleply roofing membranes [17]. While some products have performed well under a wide range of exposure conditions, many have shown early failures. In this situation, there is an urgent need for durability performance criteria to aid selection of roofing membranes. Many organizations have responded to the need for improved selection criteria. ASTM Committee D-8 on Roofing, Waterproofing and Bituminous Materials has established a subcommittee on single-ply membranes, and NBS has drawn up and published a comprehensive roofing research plan [18]. In addition, in response to a proposal from the United States, RILEM established Committee 75-SLR on Single-Layer Roof Waterproofing Systems in 1982. No generally-accepted durability tests for single-ply roofing membranes have yet been established in the US. Laboratory and field tests are needed for seams as well as for the sheets, since there is evidence that field-formed seams tend to be sites of early failure. NBS has explored the possible application of infrared thermography and the pulse echo technique (fig. 5) to inspection of seams in the membranes. Both techniques have shown ability to detect flaws in seams on an actual roof. With further development, they could help increase the service life of single-ply roofing membranes by providing a new tool for ensuring the quality of the seams.
7. D U R A B I L I T Y OF W O O D Wood is used extensively for construction in the US, both as framing material for low-rise buildings and as wall, roof, and floor panels. Research on wood and ways of improving its performance is carried out in industrial trade associations and at the Forest Products Laboratory of the US Department of Agriculture.
208
The Forest Products Laboratory (FPL) is concerned with the performance and protection of wood, both below and above ground. In one of its programs, possibilities for protecting wood against termites and decay through chemical modification with permanently-bonded chemicals is being explored [19]. Chemical modification also has potential for providing wood with improved resistance to ultraviolet radiation. Chemicals used to modify wood at the FPL have included acid anhydrides, acid chlorides, carboxylic acids, isocyanates, aldehydes, alkyl chlorides, lactones, nitriles, and epoxides. Some of the treatments have given the wood good biological resistance and much improved dimensional stability (fig. 6). Preservatives of broad-spectrum toxicity, such as arsenic compounds, are often used to protect wood against biological degradation. Concern about the environmental impact of such materials has led to a search for preservatives with higher specificity for the decay organisms [20]. FPL research on ways of exploiting physiological or biochemical differences between wood-degrading and other organisms has tentatively validated the concept of wood preservation through inhibition of chitin synthesis in fungi. This is indicated by the results in table III of the effects of polyoxin mixtures on fungal decay in three or four week soil-block tests and the observed morphological abnormalities of the fungi. The FPL work may lead to a new class of wood preservatives. While preservatives which are highly specific against certain organisms represent one route to less toxic wood preservatives, another route is inorganic preservatives which are less toxic than the widely-used ammoniacal copper arsenate (ACA). On the basis of results such as those shown in table IV for water-leached ACB-treated wood exposed to three different fungi in soil-block tests, FPL scientists now believe that ACB (ammoniacal copper TABLE II EFFECTS OF TEMPERATURE, U L T R A V I O L E T RADIATION A N D M O I S T U R E ON THE N U M B E R OF C H A I N SCISSIONS IN POLY (METHYL METHACRYLATE) (PMMA)
e-Xt (;~t) K P(K=k) = ~ kl
(1)
where* :
~= eXPtR"~--of l l To' I e x P t ~ ) J
(2)
and the in.dividual s y m b o l s are: Symbol
Description
Units
Co
C o n c e n t r a t i o n of w a t e r
%
C1
Empirical constant
--
C2
Empirical c o n s t a n t
E Io
Activation energy Absorbed radiation intensity
no. of p h o t o n s / ( s . c m 3)
la
Reference intensity of radiation absorbed
no, of p h o t o n s / ( s * c m 3)
P(K=k)
P r o b a b i l i t y that exactly k scissions have o c c u r r e d between times 0 a n d t
--
-J/tool
R
Gas constant
TO
Reference absolute temperature
T
Absolute temperature of P M M A film
OK
9
Base of the natural system of Iogarithyms
--
J/(mol'~ OK
k
Number of chain scissions
--
m
Empirical constant having a value 0 < m < l
--
seconds kg/m 3
t
Time
q
Density of the polymer film at temperature t
9~
Expected n~Jmber of scissions per unit time at temperature t and r~diation intensity I a
number of scissions/s ,..
Ao
Constant
number of scissions/s
~The expression in brackets at the right-hand side of equation (2) may be equal to 1. Le.. moisture has no detectable effect on the scission process in PMMA.
J. R. Wright - G. Frohnsdorff borate) could replace ammoniacal copper arsenate as a treatment for wood for ground contact exposure [21]. They also believe that "it is not premature to recommend its use for above-ground exposures". In other research, FPL is gaining improved understanding of the mechanisms of degradation of wood which should lead to treatments to help control the effects of weathering [22]. This research holds high promise of providing practical new ways of improving the durability of wood.
9 (~ TRABEUM_ o L. LEPIDEUS
~ss F
ALLYL METHYL ISOCYANATE ISOCYANATE ,,. _ ~
ALLYL ISOTHIOC'I~NATE
-351--
8. D U R A B I L I T Y OF CONCRETE ~ZO
Concrete is one of the most durable of building materials ([23], [24]). However, the worldwide need for the repair and rehabilitation of the societal infrastructure highways, bridges, dams, water supply systems, and so on - shows that the durability of concrete cannot be taken for granted. In the US, there have been a succession of calls for expansion of research on concrete. For example, in 1980 the Committee on the Status of Cement and Concrete Research [25] recommended increased support for fundamental research, with special attention to be given to "studies of long-term behavior and durability in extreme environments." Also, in 1984, the Transportation Research Board published the report "America's Highways: Accelerating the Search for Innovation" [26] which listed "'cement and concrete in highway pavements and structures" as one of its six areas of highest priority for research. It stated that "'a renewed research program can improve the.., durability and economy of this basic building material". Organizations which have been major contributors to research on concrete durability for many decades include the US Army Waterways Experiment Station, the Bureau of Reclamation, and the Portland Cement Association (PCA). All three of these organizations continue to be active, and NBS also carries out durability-related research on concrete. Examples of the work of the PCA and NBS will be given here. TABLE I1! EFFECT OF POLYOXIN MIXTURE ON FUNGAL DECAY OF WOOD (WEIGHT LOSS)
Concentration (P.g/g) Species
0
10
100
G. trabeum P. placenta I. lacteus. C. versicolor
43 34 38 15
40 22 31 6
18 " 0 13 3
800 1 0 1 1
TABLE IV EFFECT OF AMMONIACAL COPPER BORATE (ACB) ON FUNGAL DECAY OF WOOD AFTER LEACHING (WEIGHT LOSS)
Original retention (before leaching) pcf
L. lepideus G. trabeum P. placenta
s t5 ~ 10
0
5
0.10
0.48
1.01
38 56 57
0 2 55
0 1 10
0 0 0
5 IO 15 o
5 Io 15
WEIGHT GAIN. PCT
Fig. 6. - E f f e c t s o f c h e m i c a l t r e a t m e n t s on t h e d e c a y resistance of wood. The curves relate the weight gains from
the chemical treatments to the subsequent weight losses in 12-week exposures in soil-block wood decay tests. In the case of allyl isothiocyanate treatments, weight gains above 10% produced essentially complete protection of the wood. Over a period of more than 50 years, the PCA's contributions have included development of laboratory tests for freezing and thawing in water; resistance to deicing chemicals applied during exposure to freezing and thawing; resistance to aggressive chemicals such as sulfates; alkaliaggregate reactions; and the corrosion of reinforcing steel. They have also done much important field exposure testing in soils, seawater, and other severe environments. PCA durability research up to 1978 is described by Klieger [27] in the proceedings of the First International Conference on the Durability of Building Materials and Components [28]. More recently, PCA has made important contributions to the technical literature on the durability of concretes containing one or more of fly ash, ground blast-furnace slag, and chemical admixtures. In durability-related research at NBS, the initial research [29] on epoxy coatings for protection of reinforcing steel against corrosion was carried out in the 1970's. This work has led to the establishment of a substantial industry in the coating of reinforcing steel. The current NBS research related to durability is of three types. In the first, Brown and his colleagues are contributing to improved understanding of the fundamental chemistry and physics of cement hydration (fig, 7) and reactions relevant to the chemical degradation of hardened cement paste [30]. The second is improvement of understanding of the mechanisms of crack growth in mortars and concretes under static and fatigue loading [31] (fig. 8). The purpose of this research by Knab et al. is to advance the application of fracture mechanics to concrete. The third area is the development by Clifton of a knowledge-based expert system for computer-assisted decisions concerning the durability of concrete. This will be discussed further in the next section. 9. F E E D B A C K
0
Io 15 0
SYSTEMS
Although every building and structure provides information on the durability of building materials, very little of the information is fed back into the durability information system in a tangible form. It has long been recognized that 209
Vol. 18 - N~ 105 - Mat~riaux et Constructions
Umit of AFM solid solubility C3A. 1 CaS04 .1 Ca(OH)2-nH20 Limit of AFT solid solubility
C3A-2CaSO4.Ca(OH)232H20/ /
C6AS3H32 §
CSH2
Ets.s.
CH
C6AS3H32
Ets.s.
C6A~3H3 2
C6AS3H32 +
C2AH8
C4ASH x + C4AS1/2Hx
+ C4A+SHx
39 tool % CaSO4
:6AS2H32
C2AH8 C4ASH]
~
1c.~o~~ Ca cao 4§ SO4 = 2.564 after Lea
C4AS1/2Hx
6AS2H32
_/ 4AS1/2Hx
4A:.aoHx
C4ASHx C2As,s.Hx
C4As.a.Hx
C4AHx
C2.4AHy
CH
d
C2AH8
C2As.s.
C2.4AHy
C3A
C4AHx
CaA
CH
Fig. 7. - A p o r t i o n of t h e lime-alumina-sulfur trioxide-water phase diagram at 25~
The large number of phase fields indicates the complexity of the system which has to be understood in interpreting microstructure of hardened cement paste and the mechanism of sulfate attack of concrete.
there is a need for feedback systems which can capture the information from those responsible for the maintenance, repair and rehabilitation of facilities. Consequently, feedback systems have been established in several large organizations. One of the most successful of these is the PAVER program developed by the Construction Engineering Research Laboratory of the US Army Corps of Etlgineers [32]. PAVER is a maintenance management system for asphaltic pavements which keeps records of routine surveys of all the pavements in a road system and indicates recommended maintenance priorities and costs and better maintenance of road systems to which it has been applied. PAVER is now generally available in the United States through the American Public Works Association. An analogous program to PAVER, ROOFER, is being developed by the same US Army laboratory for maintenance management of low-sloped roofs. The most ambitious feedback system in the United States, the Architecture and Engineering Performance Information Center (AEPIC) at the University of Maryland, was set up in July 1982 with support from the US National Science Foundation [33]. It compiles information on performance, including durability performances, of materials, elements, systems and processes. AEPIC makes its information available to subscribers for improvement of the performance of the construction industry. The data are stored in the form of: (i) a computerized case file; (ii) a computerized citation file; (iii) a dossier library; (iv) a visual materials library, and (v) a reference library. With the help of two insurance companies, over 40,000 cases have already been entered into the system. It appears that about half of
210
the cases concern water leakage into buildings, whether through roofs, walls or windows. It is now feasible to establish data bases on an international scale to record information on the durability performance of building materials, and AEPIC has been structured so that it can become an international data base. It already has affiliates in Canada and the United Kingdom. This will, undoubtedly, give impetus to durability research by providing better-documented information on the causes of durability-related defects in buildings and other structures and by providing easier access to new research findings relating to durability. One of the most exciting technological developments with relevance to improving durability of building materials is the application of artificial intelligence in the form of knowledge-based expert systems [34]. Expert systems are
Fig. 8. - Crack p a t t e r n near a n o t c h in a p o r t l a n d c e m e n t m o r t a r s p e c i m e n as a r e s u l t of f a t i g u e loading. The crack is filled with a UV-fluorescent dye which enhances
human ability to identify cracks and other features.
J. R. Wright - G. Frohnsdorff computer programs that embody the organized knowledge of human experts (fig. 9) in the form of rules which can be applied to a specially-structured knowledge base containing information about a real or hypothetical situation. By recommending actions which f o l l o w from the rules, the program can provide assistance to human decision-makers by simulating the decision-making behavior of human experts confronting the same situation. The PAVER pavement maintenance management program mentioned earlier has some of the characteristics of an expert system in that it uses its data base and selection rules to recommend courses of action comparable to those which a human expert might recommend. The use of computers for artificial intelligence applications, such as expert system development, has obvious potential for improving the quality of decisions about the durability of building materials. This w o u l d be analogous to the use of expert systems as aids to medical diagnosis [35]. There is an important distinction to be made between expert systems which are based primarily on expert opinion (sometimes referred to as " s h a l l o w " expert systems) and those which make extensive use of calculations based on well-established scientific and engineering principles ("deep" expert systems) [36]. At present, it appears that most expert systems are still of the shallow type though the desirability of developing deep expert systems is recognized. At NBS, Clifton is developing an expert system, DURCON, to demonstrate the feasibility of making computerassisted decisions on the durability of concrete mixtures, both to aid mixture design and to assess the durability of existing concrete (fig. 10). The work is being carried out in collaboration with the American Concrete Institute Committee 201 on Durability. The approach is to analyze the ACI Guide to Durable Concrete [37] to deduce rules which can be included in the expert system computer program. Ultimately, it is possible that the expert system will become an example of a new type of voluntary consensus standard. It seems likely that expert systems will eventually become common aids to the making of decisions concerning building materials, including decisions concerning durability. However, this will not remove the need for continued support of high quality scientific research to provide the technical information upon which computeraided decisions can be based.
TABLE V TYPESOF MATERIALS COVEREDBY RILEM COMMITrEES Concrete (31) (') Cement (2)
Metals (5) Timber (2) Polymeric materials (8) Bituminous materials (2)
Gypsum (1) Stone (3) (*) Number of committees.
TABLE VI SOMERILEM COMMITTEES CONCERNEDWITH DURABILITY Committee No. 71 -PSL 72-SMS 45- LTO 58-VPM 59-TPM 32-RCA 50-FMC 60-CSC 52-RAC 68-MMH 75-SLR 66-BJS 47-SM 46- FST
Name Prediction of service life of building materials and components Stochastic methods in materials and structural engineering Long-term observation of structures Aging tests of stone monuments Treatments of stone monuments Resistance of concrete to chemical attack Fracture mechanics of concrete Corrosion of steel in concrete Resin adherence to concrete Mathematical modeling of cement hydration Single layer roof waterproofing systems Building joint seals Synthetic membranes Fatigue of steel - temperature effects
The need to treat durability in a generic way, rather than handling it differently for different materials, led RILEM to set up, in 1977, Committee 31 -PCM on Performance Criteria for Building Materials and, more recently, Committee 71 -PSL on Prediction of Service Life of Building Materials and Components. Committee 71-PSL was established in 1982, as a joint committee with CIB W-80. Its main objectives are: (1) to identify methodologies for prediction of service life of building materials and components, and (2) to identify areas for improvement in existing methodologies and to stimulate new developments. A state-of-theart report being prepared by the committee will include recommendations on a preferred methodology for service life prediction. The recommendations of the committee Knowledge
Engineer
c@9169
10. D U R A B I L I T Y A C T I V I T I E S IN R I L E M Expert System
So far, this paper has focused on building materials durability research in the US. N o w it will discuss RILEM and its increasingly important role in fostering international collaboration in durability research. Since its founding in 1947, RILEM has been concerned with all aspects of the performance of construction materials, including durability. The types of materials which have been covered by the RILEM technical committees, and the number of committees which have dealt with each material, are shown in table V. The greatest number of committees has been concerned with concrete, but all major materials of construction, except glass, have been addressed. Most committees have included durability among the aspects of performance to be considered [38]. Some of the current committees [39] concerned with durability are listed in table VI.
rence
] "Domain-Specific .
=glne I
[ Knowledge Base
rence ~
'~176
I .... I
Expert
"Expert
Problem-Specific J
Data Base
Fig. 9. - S c h e m a t i c d i a g r a m of an e x p e r t s y s t e m ( f r o m Stahl). The domain-specific and problem-specific knowledge bases are used by the "inference engine" in solving the problem. A knowledge engineer is needed to put the knowledge of experts into the expert system,
211
Vol. 1 8 - N ~ 1 0 5 - M a t 6 r i a u x et C o n s t r u c t i o n s
<
|
o
"~ ~-
'~
o~
O
~
e
~
~,
o ~o
"-~
o~ ~
-
,~
r
~o
~o L., " ~
O~ 9~
e N
9
ZZ ~U~
"~g UJ lIJJ
._~
z~
z
o~
/~
Q
o~N '~-~|
o.
~,<
~ ~.
~2
| ~
o
~
~
~
'c
Z
~
0 u_ c
0 l--
~ m~ z-.
,<
~ ~"
m
i
0 ~=_-
< 7-
~
o
'-
o|
i
o,-~
o
OFo
o.
~ o -
'E o ' ~
I U.l
.9
NI
LLI
rr UO
~_~ <
"s2 oe~ o 9-
~-
B.|
~._
9
~
.~_o
< ~.~ .|
~E~oC oc ~ c~e~ 9
o~
.S N
~ o o
[,
* alew
212
~0 uo!~emle s
I
J. R. Wright - G, Frohnsdorff will be distributed to other committees in RILEM and CIB to help them deal with durability questions in a consistent way. Committees 31-PCM and 71-PSL have established RILEM leadership in promoting exchange of information on service life prediction and in stimulating durability research in several countries. For example, RILEM, with CIB and three North American organizations (ASTM, NBS and the National Research Council of Canada) sponsors the triennial International Conferences on the Durability of Building Materials and Components. The first of these conferences was held in Ottawa in 1978 [28], the second was in Washington in 1981 [40], and the third in Helsinki in 1984 [41]. They have heightened awareness of the complexity of durability problems and the need for them to be treated by materials scientists. The Fourth International Conference on the Durability of Building Materials and Components will be held in 1987. On the broader international level, RILEM influences test methods for materials and structures throug h the Technical Recommendations it publishes. These documents are available to any organization that may want to use them as the basis for its own test methods. About 5 years ago, ISO, the International Organization for standardization, recognized RILEM as a prestanda'rdization organization whose Technical Recommendations should be given special consideration in the development of ISO standards. In 1984, RILEM and ISO further strengthened their ties by preparing a Memorandun:l of Understanding concerning "cooperation at the international level on the utilization of the results of research on materials and structures in the development of International Standards". 1987 will be the fortieth anniversary of the founding of RILEM. To celebrate the occasion, a Fortieth anniversary Congress entitled, "From Materials Science to Materials Engineering" will be held in Paris. Durability of building materials will be a prominent subject in the Congress.
11. OPPORTUNITIES FOR I N T E R N A T I O N A L COLLAB O R A T I O N IN D U R A B I L I T Y RESEARCH In this paper, the need for international collaboration in durability research has been mentioned several times. The problems are great and the numbers of researchers involved in solving the problems are small. Among the problems are development and application of techniques for elucidation of degradation mechanisms, characterization of environments (including microclimates), probabilistic mathematical modeling of degradation, service life prediction, condition assessment of materials in existing structures, and construction of large data bases with evaluated data. RILEM, through many of its 49 technical committees, offers opportunities for international collaboration in dura bility research. In addition, the National Delegates to RILEM's General Council [42] and the Secretary General work together to facilitate international scientific exchanges, including visiting scientist appointments. In the United States, NBS and several other institutions welcome visiting scientists from other countries to work in their laboratories on topics of common interest. We hope these mechanisms will continue to grow in importance in enhancing international collaboration.
A C K N O W L E D G EM ENTS The authors wish to express their appreciation t o Dr. H. Takebayashi, Director, Building Research Institute, Tsukuba, Japan, for the opportunity to present'this paper at the RILEM Seminar on Durability of Building Materials which he organized in Tokyo, October 13, 1984.
APPENDIX
I
SOME ORGANIZATIONS INVOLVED IN RESEARCH ON THE DURABILITY OF BUILDING MATERIALS IN THE UNITED STATES Information about the publications of the organizations listed can be obtained by writing to their Technical Directors. Government laboratories
Bureau of Reclamation, Engineering and Research Center, US Department of the Interior, PO Box 25007, Denver, Colorado 80225. Construction Engineering Research Laboratory, US Army Corps of Engineers, Interstate Research Park, Box 4005, Champaign, Illinois 61820. Fairbanks Highway Research Laboratory, Federal Highway Administration, US Department of Transportation, Washington, DC 20590. Forest Products Laboratory, US Department of Agriculture, North Walnut Street, Madison, Wisconsin 53705. National Bureau of Standards, Center for Building Technology, US Department of Commerce, Gaithersburg, Maryland 20899. Naval Civil Engineering Laboratory, Port Hueneme, California 93043. Waterways Experiment Station, US Army Corps of Engineers, PO Box 631, Vicksburg, Mississippi 39180, Industry associations
American Plywood Association, PO Box 11700, Takoma, Washington 98411. Asphalt Institute, Asphalt Institute Boulevard, University of Maryland, College Park, Maryland 20740. National Ready-Mixed Concrete Association, 900 Spring Street, Silver Spring, Maryland 20910. Portland Cement Association, 5420 Old Orchard Road, Skokie, Illinois 60076. Society of the Plastics Industry, 355 Lexington Avenue, New York, New York 10017. Steel Structures Painting Council, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213. University laboratories
In addition to the laboratories listed above, many of the large universities carry out research on the chemistry and long-term behavior of building materials.
REFERENCES
[1] DIKKERS R.D. Standards for solar energy systems. Proceedings, 1980 ASQC Technical Conference, pp. 201-208, American Society of Quality Control, Milwaukee, Wisconsin, 1980. [2] High strength concrete. SCM 3-83, American Concrete Institute, Detroit, Michigan, 1983, 112 p. [3] MATHEYR.G., ROSSITERW.J. JR. A preliminary evaluation of the tensile and elongation properties of Mngle-ply sheet roofing membrane materials. Preprints, Second International
Conference on the Durability of Building Materials and Components, pp. 442-451, National Bureau of Standards, Gaithersburg, Maryland, September 1981.
213
Vol. 18 - N ~ 105 - Mat(~riaux et Constructions
[4] CAMPBELL P.G., POST M.A. Nontoxic yellow traffic striping. Federal Highway Administration Report FHWA-RD-78-1, January 1978, 74 p.; obtainable from the National Technical Information Service, Springfield, Virginia. [5] FOSTER B.E., ed. Performance concept in buildings. Proceedings of the RtLEM-ASTM-CIB Symposium, 2 vol., Special Publication 361, National Bureau of Standards, Washington, DC, September 1972. [6] Evaluation of the performance of external vertical surfaces of buildings. Proceedings, RILEM/ASTM/CIB Symposium, 3 vol., Technical Research Centre of Finland, Otaniemi, Finland, 1977. [7] Performance concept in buildings. Proceedings, 3rd ASTM/ CIB/RILEM Symposium, 2 vol., National Civil Engineering Laboratory, Lisbon, 1982. [8] FROHNSDORFFG., MASTERS L.W. The meaning of durability and durability prediction, in Durability of building materials and components. P.J. SEREDA,G.G. LffVAN, Eds., ASTM STP 691, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1 980, pp. 17-30. [9] WRIGHTR.N. Computers in building - A strategy for building research. Building Research and Practice, Vol. 1/2, 1984, pp. 14-20. [10] LIDE D.R., Ed. The effect of computers on the generation and use of technical data. NBSIR 84-2907, National Bureau of Standards, Gaithersburg, Maryland, June 1984, 55 p. [11] MASTERSL.W., WOLFEW.C., ROSSITERW.J., SHAVERJ.R. State-of-the art of durability testing of building components and materials. NBSIR 73-132, National~ Bureau of Standards, Gaithersburg, Maryland, june 1973. [12] MARTIN J.W. Service life prediction from accelerated aging test results using reliability theory and life testing analysis. To be published in the Proceedings of the NATO Advanced Research Workshop on Prediction of the Service Life of Building Materials and Components, Saint-R~my-les-Chevreuse~ France, September 10-12, 1984. [13] BENNE'I-rL.H., KRUGERJ., PARKERR.L., PASSAGLIAE., REIMANN C., RUFF A.W., YAKOWlTZH., BERMAN E.B. Economic effects of metallic corrosion in the United States. Part 1, A Report to the Congress by the National Bureau of Standards, NBS Special Publication 511-1, May 1978, 33 p. [14] MCKNIGHT M., MARTIN J. Non-destructive corrosion detection under organic films using infra-red tharmography. Proc. 14th National SAM PE Technical Conference, Atlanta, Georgia, October 1982, pp. 349-358. [1 5] MASTERSL.W., FROHNSDORFFG., EBY R.K. Polymers as materials of construction, in polymer science and engineering: Challenges and opportunities. Report on Panel on Polymer Science and Engineering, National Research Council, Washington, DC July 1981. [16] MARTIN J. A stochastic model for predicting the service life of photolytically degraded polymethyl methacrylate films. J. Applied Polymer Science, Vol. 29, 1984, pp. 777-794. [17] ROSSITERW.J., MATHEY R.G. A methodology for developing tests to aid service life prediction of single-ply roofing membranes. Proceedings 7th Conference on Roofing Technology, National Roofing Contractors Association, Chicago, Michigan, March 1983, pp. 4-11. [18] CULLENW.C., ROSSITERW.J., MATHEYR.G., CLIFTONJ.R. Lowsloped roofing research plan. NBS SP 659, National Bureau of Standards, Washington, DC, July 1983. [19] HOWELL R.M. Chemical modification of wood. Forest Products Abstracts, Vol. 6, 1983, pp. 363-382. [20] JOHNSON B.R. Effects of polyoxin inhibitors of fungal chitin synthesis on the decay of wood. International Biodeter. Bulletin, Vol. 18, Summer 1982, pp. 37-42, [21] JOHNSON B.R., GUTZMER D.I. Ammoniacal copper borate: A new treatment for wood preservation. Forest Products Journal, Vol. 28, 1978, pp. 33-36.
214
[22] FEISTW.C., HON D.N.-S. Chemistry of weathering and protection, in the chemistry of solid wood. R. HOWELL,Ed., Advances in Chemistry Series, No. 207, American Chemical Society, Washington, DC, 1984, pp. 401-454. [23] Durability of concrete Proceedings. Symposium on Durability of Concrete, Special Publication SP-47, American Concrete Institute, Detroit, Michigan, 1975, 385 p. [24] ROSTAMS., Ed. Durability o f concrete structures. Report of the CEB/RILEM International Workshop, Copenhagen, May 1983, Department of Structural Engineering, Technical University of Denmark, DK-2800 Lyngby, Denmark, 1983, 432 pp. [25] RoY D.M., Ed. The status of cement and concrete R and O in the United States. NMAB Report 361, National Materials Advisory Board, National Research Council, National Academy of Sciences, Washington, DC, 1980. [26] KULASH D.J., Ed. America's highways: Accelerating the search for innovation. Special Report 202, Transportation Research Board, National Research Council, National Academy of Sciences, Washington, DC, 1 984, 169 p. [27] KLIEGERP. Durability studies at the Portland Cement Association, in Durability of Building Materials and Components. P.J. SERAOA,G.G. LITVAN, Eds., ASTM STP 691, pp. 282-300, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1980. [28] SEREDAP.J., LiTVAN G.G., Eds. Durability of society materials and components. ASTM STP 691, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1980, 1 031 p. [29] CLIFTON J.R. Protection of reinforcing bars with organic coatings, Materials Performance, Vol. 1 5, 1976, pp. 16-17. [30] BROWNP.W., LlaERMAN L.O., FROHNSOORFFG. The kinetics of the tricalcium aluminate - Calcium sulfate reaction. J. Amer. Ceram. Soc. (in press). [31] KNAB L.I., WALKER H., CLIFTONJ.R., FULLERE.R., JR. Fluorescent thin sections to observe the fracture zone in mortar. Cement and Concrete Research, Vol. 14, 1984, pp. 339-344. [32] SHAHIN M.Y., KOHNS.D. Overview of the "PAVER" management system (and) Economic analysis of field implementing the "PAVER". Pavement Management System, Technical Manuscript M-310, Construction Engineering Research Laboratory, US Army Corps of Engineers, Champaign, Illinois, March 1 982. [33] Loss J. Finally, A database on building performance dysfunctions. Architectural Technology, Fall 1984, pp. 28-35. [34] WINSTON P.H. Artificial intelligence. 2nd Edition, AddisonWesley Publishing Company, Reading, Massachusetts, 1984, 527 p. [35] SHORTLIFFE E.H. Computer-based medical consultations. MYCIN, Elsevier Computer Science Library, Elsevier Scientific Publishing Company, New York, 1976. [36] COHENP.R., FEIGENBAUME.A. The handbook ofartificialintelligence, 3 vol., HeurisTech Press, Stanford, CA, 1982. [37] Guide to durable concrete. ACI 201.2R-77, in ACI Manual of Concrete Practice - 1981, Part 1, American Concrete Institute, Detroit, Michigan, 1981. [38] SNECK T. RILEM and durability. Materials and Structures, Vol. 14, 1981, pp. 379-390. [39] Directory of Technical Committees - 1984, RILEM, Paris, 1984, 108 p. [40] FROHNSDORFFG., HORNERB.J., Eds. The Second International Conference on the Durability of Building Materials and Components. Preprints of the CIB/RILEM/ASTM/NRC (Canada)/NBS Conference, National Bureau of Standards, Gaithersburg, Maryland, September 1981, 471 p. [41] SNECKT., KAARRESALOA., Eds. Proceedings, Third Interna-, tional Conference on the Durability of Building Materials and Components. 3 vol., Technical Research Centre of Finland, Espoo, Finland, 1984. [42] RILEM Directory of Individuals, p. ii, Paris, 1984.