READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
Questions? Contact the NRC Publications Archive team at
PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.
NRC Publications Archive
Archives des publications du CNRC
This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Frost action in porous systems
Litvan, G. G.
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
NRC Publications Record / Notice d'Archives des publications de CNRC:
https://nrc-publications.canada.ca/eng/view/object/?id=0ff9e353-aa17-4a66-8443-f35475649286
https://publications-cnrc.canada.ca/fra/voir/objet/?id=0ff9e353-aa17-4a66-8443-f35475649286
N21d
!
National Research
Conseil national
no. 1176:
c.
2
II
$
Council Canada
de recherches Canada
FROST
ACTION
IN POROUS SYSTEMS
by
G.G.
Litvan
N R C - C I S T I 'IBLDG.
RES.
8
L I B R A R Y
ANALYZED
p!:,,
r_
~ ~ ~ - ~ ~ = f - , - t c_
. . . - . _ _
- I nr:.
-
r - . 3 " : ~ . :- ---.---
-
Reprinted from
Sbminaire: Durabilitb des Betons et des Pierres
Sgminaire organis6 avec la collaboration de I'UNESCO
par
le
Colldge international des sciences de la
construction
Saint-Rbmy-I&-Chevreuse
(France)
du
17
au
19
novembre
1981
p.
95
-
1Q8
DBR Paper No.
1176
Division of Building Research
Sous nos l a t i t u d e s , l ' a c t i o n du g e l e s t s a n s aucun doute l a p r i n c i p a l e cause de d d t e r i o r a t l o n du bston. En d'epit de p r o g r s s c o n s i d ' e r a b l e s r e a l i s S s d a n s l e s methodes de c o n s t r u c t i o n , l e s consSquences du g e l e t du dSgel s o n t encore e r a s p r k c c u p a n t e s e t amsnent p a r f o l s de graves p r o b l b e s . Par exemple, l ' a d o p t i o n il y a quelques a n n h s , de l a p o l i t i q u e "du revstement
1
nu" p a r l a p l u p a r t des s e r v i c e s char& de l ' e n t r e t i e n des r o u t e s a conduitEI
l a d e t e r i o r a t i o n de ponts, de r o u t e s e t de t r o t t o i r s en r a i s o n de l ' u t i l i s a t i o n de grandes quantit'es de s e l s de degivrage. L'emploi dans c e r t a l n e s r e g i o n s , d ' a g r s g a t s de q u a l i t s secondaire, s u i t ea
lli?puisement des materiaux r e s i s t a n t s au g e l , e s t Bgalement3
l ' o r i g i n e d'un c e r t a i n nombre de p r o b l b e s . La p r k e n t e Btude t r a i t e de quelques a s p e c t s des mi?cau+---dll
&
- _ --
-FROST ACTION IN POROUS SYSTEMS
G.G. LITVAN
Division of Building Research, National Research Council of Canada
I. INTRODUCTION
In northern latitudes frost action is undoubtedly the main cause of concrete deterioration. In spite of considerable improvement in building pratices the consequences of freezing and thawing are still of great concern, and at times the situation becomes serious. For example, the adoption, some years ago, of the "bare pavement" policy by most highway authorities has led to deterioration of bridges, highways and sidewalks through the use of de- icing salts in vast quantities. Another difficulty is the marginal quality aggregates that have came into use in some regions after frost resistant supplies have been exhausted. In the present review some aspects of the mechanism of frost action will be discussed.
1I.CHARACTERISTIC FEATURES OF FROST ACTION
The main features of frost action in concrete can be characterized by experimentally established observation:
1. The severity of the mechanical damage caused by frost action is directly proportional to the water content of the concrete and porous solids;
Figure 1 shows the dimensional changes of fully saturated (0.5 and 0.8 water/ cement (w/c) ratio, 3.17 m (0.125 in.) thick) cement specimens during a cooling-warming cycle between +5 and -70°C (1). The specimen which was pre- pared at a 0.8 w/c ratio expanded 0.7% when fully saturated, while a similar specimen conditioned at 84% RH merely contracted 0.12% (Fig. 2) (1). Few if any systems can endure even a single freezing and thawing cycle without injury in the fully saturated state.
2. Comparison of the length changes in a cement specimen 1.27 m thick (Fig. 3) with those of a specimen 3.17 mm thick (Fig. 1) following a freeze-thaw cycle shows clearly the effect of size (1). It is also known that the frost resistance
8 0 0 - ,
,
, I 1 , I I ' fa0-
sOo-
SPECIMEB THICIIBESS, 1 3 5 m m 500-
400-
05-
300 - P O 0-
I I I I I ' 40 20 0 EYPERaTbRE. 'CFig. 3
-
Fractional length changes of vacuum-saturated plain cement specimens determined during temperature cycles. Specimen thickness 0.050 in. and the w/c values are indicated on the curves. For sake of clarity the curves are shifted along the temperature axis. Startingtemperature in each case is + 5°C.
50
T E M P E R A T U R E . ' C
Fig. 4- Fractional length changes (bottom) and thermograms (top) of cement specimen 0.125 in. thick (w/c = 0.7) during a tempe- rature cycle with slow cooling and heating
rate 2.5 'C/h.
porosity and is, therefore, vulnerable unless qecial precautions are taken. Frost susceptibility increases with increasing w/c ratio blcause direct relationship exists between w/c ratio and porosity.
5. Air entrainment, i.e., the addition of a surface-active agent to the plastic mix resulting in the formation. of small air bubbles around which the paste subsequently hardens, has proved to be an excellent method of increas- ing the frost resistance of cement and concrete.
6 . The main features of frost action appear to be common to many various types of porous solids. Figure 5 shows the dimensional changes of porous 96% silica glass subjected to a low temperature cycle (2). The essential
-am
I I-40 -20 0
TEMPERATURE, ' C
Fig. 5
-
Dimensional changes and thermogram of the porous glass-water system : (a) 2 mmthick glass, water saturated, cooling rate
0 . 2 5 'c/min ; (b) 5 mm thick glass, water saturated, 0 . 2 5 *~/min ; (c) 5 mm thick
features,viz. the existence of two freezing ranges both associated with ex- pansion, hysteresis, significant residual expansion after completion of the cycle, sensitivity to sample thickness and cooling rate, are similar to those observed with concrete. The length changes of various stone samples, in- cluding marble, limestone, sandstone, in a freezing-thawing cycle exhibit similar features (Fig. 6).
L I t I t I I 1
- 6 0 - 5 0 - 4 0 - 3 0 - 2 0 - 1 0 0 10
T E M P E R A T U R E , OC
Fig. 6
-
Length changes of stone samples in freeze- thaw cycle.1) Briarhill Sandstone 2) Colorado marble 3) Crab Orchard Sandstone
4)
Minesota Limestone 5) Indiana Limestone (buff) 6) Indiana Limestone (gray) 7) Missouri marble8) Georgia marble 9) Vermont marble
7. The characteristics of frost action with systems containing organic liquids are similar to those observed with water. The length changes in low
or cyclohexanol are shown in Fig. 7, and those with chloroform, m-xylene, or carbon tetrachloride in Fig. 8 (2). When porous silica glass was subjected to a low temperature cycle while immersed in benzene, more than 0.3% residual ex- pansion occurred (Fig. 9), indicating serious permanent damage to the sample by freezing-thawing action (2).
Fig. 7
-
Dimensional changes and thermograms of 5 mm thick porousglass with various adsorbates :
(el
(d) octanol, mp-
16.7 O C ,(e) benzene, mp + 5.5 O C ,
(f) cyclohexane, mp + 6.5 O C
c f ,
L 1 2 0Fig. 8
-
Dimensional changes and thermograms of 5 mm thick porous glass with various adsorbates : (a) chloroform, mp-
63.5 O C ,(b) m-xylene, mp
-
17.9 O C ,( c ) carbon tetrachloride, mp
-
22.9 O C .8. It is well established that repeated freezing and thawing under field conditions result in desiccation of the porous system and in the accmulation of the formerly pore-held liquid outside the porous solid, a phenomenon known as ice lens formation.
9. When a salt solution is contained in the pores the damage from freezing
and thawing is much more severe than it would be with a pure liquid such as water. For example, in a 0.5 w/c ratio cement paste the residual 'expansion in a 5% sodium chloride solution is 1%, whereas with water it is 0.18% (Fig.
10) (3). Again, the phenomenon does not depend on the nature of the solute.
Fig. 9
-
Dimensional changes of5 mm thick porous glass immersed in benzene.
__---
.
_--- _---
I @ % .*__---_
__--
---.
r;
IWO- I r---- I / d I-----
---,-
--
- Fig. 10
-
Fractional length changesof 0.5 water-cement ratio cement pas-
::
\
$_-: tes (concentrations impregnated with as indicated) NaCl solutions duringI I temperature cycles. 100-
.
I - 0 % 0-
I I , I I I I - -no -60 -w -2a I) ra 'C 101111. MECHANISM
Frost action also occurs with organic liquids that normally contract on freezing, clearly indicating that the 9% increase in volume associated with the freezing of bulk water cannot in itself explain all the phenomena.
3.1
-
ABSENCE OF CRYSTALLIZATION IN PORESThe inability of water to crystallize in the confined spaces of small pores, documented in isothermal adsorption studies, may be responsible for the chain of events leading to mechanical failure (2). When, at temperatures below O°C, a porous solid contains liquid water instead of ice, complete saturation would take place only if the vapour pressure above the solid were to become equal
to that of supercooled water. In practice, :.owever, this cannot be realized because the vapour pressure of supercooled water is greater than that of ice; any attempt to elevate the vapour pressure above the level of that of ice will result in the formation of ice deposition outside the pores of the solid. As a consequence, complete saturation of porous solids with water cannot be achieved at temperatures below O°C. Furthermore, because the difference in the vapour pressures of supercooled water and of ice increases with decreasing temperature the equilibrium water content of porous bodies falls short of complete saturation, with an increasing margin the lower the temperature is below O°C.
3.2 - MOISTURE CONTENT AFFECTED BY TEMPERATURE CHANGE
The question arises of what happens if a porous solid, fully saturated at OOc, is cooled to lower temperatures? The water in the pores at, say, -2'C is in a liquid-like state and its vapour pressure is thus higher than that of ice. This is a non-equilibrium condition and can, in the absence of freezing, be eliminated by one of the following processes.
3.2.1 Ideal Case
In the absence of freezing (which would reduce the vapour pressure from that of undercooled water to that of ice) equilibrium can be, and often is established in nature by a very simple mechanism: part of the water contained in the pores migrates to and freezes at locations where the effect of the sur- face is not felt. Not all the water has to leave the porous solid: as already stated, only sufficient for the fraction remaining in the pores to be under menisci, with an appropriate curvature under which the water has a vapour pres-
sure equal to that of ice. Because the free energy of liquid is less if it has a concave instead of a planar surface, equilibrium can be established between the external ice (having relatively low vapour ~ressure) and the unfrozen water, with reduced vapour pressure in the partially filled pores.
As a result of this process, the porous body becomes partially desiccated and consequently contracts; ice accumulates in the large pores and on external surfaces. Significantly, no mechanical damage occurs. On further cooling to
temperatures lower than - Z ° C , this sequence of events repeats itself because the difference between the vapour pressures of ice and undercooled water increases with decreasing temperature. Thus, on cooling, the radius of curvature of the menisci decreases, the moisture content of the porous solid decreases, and the amount of external ice increases. According to the outlined mechanism, cooling of saturated porous bodies produces moisture transfer but not necessarily mechanical damage.
3.2.2. Practical Case
In nature and in laboratory experiments large step changes are frequently imposed on the system and, although moisture redistribution takes place due to the continuously increasing demand for mass transfer, the system moves through a succession of non-equilibrium states. In the terminology of thermodynamics, the changes are not reversible.
Reversibility is possible if (a) the amount of water that has to be re- distributed is small (that is, the porosity or degree of saturation, or both,
are low), (b) the cooling rate is low, (c) the permeability of the system is high, (d) the migratory path is short, and (e) the mobility of the water is high, that is, the viscosity is low or the temperature is high. It is known from field experience and laboratory studies that under these conditions damage is, indeed, minimal. In contrast, if one or several of the parameters are un- favorable, mass transfer does not take place at the required rate and mechanical damage ensures.
IV. MECHANICAL DAMAGE
Solids can suffer mechanical damage in non-equilibrium freeze-thaw cycles by one or several of the following mechanisms:
1. It follows from the described theory that if, on cooling, the porous body loses water to the environment, the reverse process will take place on warming: water from the external surface and cracks migrates back to the interior. Whether all the water can be re-absorbed before the next cooling period depends on various factors. Usually the re-saturation process is incomplete, and because
/
of the reduced water content in the interior the effect of frost action is less severe. The water accumulated in the fine cracks, however, freezes in the next cooling phase and the accompanying 9% volume increase may lead to propagation of the crack. The space so enlarged will attract more water from the interior in subsequent cycles and further damage will occur. This mechanism can account for the destructive effect of repeated freeze-thaw cycling.
2 . The volume of porous solids that contain adsorbed water is affected by
environmental changes. For example, cooling causes shrinkage, as determined by the coefficient of thermal expansion and, in addition, results in an altered moisture content that affects the dimensions of the porous body. If cooling is rapid, significant temperature and moisture gradients are created, causing considerable stress.
In Fig. 11 the changes in the relative dynamic modulus of elasticity of
l
a neat cement prism (2.5 by 2.5 by 15 cm) formed at w/c ratio of 0.4 are shown
I
during drying at room temperature and exposed to (a) 50% RH, and to (b) 84 and
subsequently 66% RH. The modulus was determined according to ASTM C215
-
60(76),Test for Fundamental Transverse, Longitudinal and Torsional Frequencies of Concrete Specimens.
I
-
0 . 2 0-
O R A P I D D R Y I N G ( 5 0 % R H )-
. S L O W D R Y I N G ( 8 4 - 6 0 % RH) 0 . 1 0 h k I I I 1 . U . 9 9 - 9 8 . 9 7 .96 - 9 5 . 9 4Fig. 1 1
-
Relative dynamic Young'smodulus of elasticity of a hydrated neat cement paste bar (water-cement ratio 0.4) during first drying.
RELATIVE W E I G H T . W T / O R I G . W l
It may be seen that on rapid drying the dynamic modulus of elasticity undergoes a transient decrease to a minimum of 33% of its original value; on slow drying the minimum value of the modulus is 65% that of the initial value. This transient decrease of the modulus of elasticity, then, depends on the
drying rate, a characteristic feature consistent with the process of frost action. The change in the modulus cannot be attributed to shrinkage or loss of water because similar transient reduction occurs also during wetting when the system expands.
Inhomogeneity created by rapid wetting and drying is known to create stresses that result in cracking. In frost action the damage is made more serious by the very high rate of drying during cooling to temperatures below 0°C.
3. In a frost-susceptible porous solid the rate of water movement out of the
pores declines with decreasing temperature to a level much less than that re- quired to maintain equilibrium. With the excess of water increasing with cool- ing, the movement of water ceases until, eventually, a noncrystalline amorphous solid forms.
V. FACTORS AFFECTING FROST RESISTANCE
Recommendations in practical terms can be made for achieving good re- sistance to freezing and thawing. The most beneficial is to keep the water content low. Moisture content in a porous solid will be low if either porosity is very low, i.e., there is no space for water to accumulate, or if it is very coarse. In the latter case water will accumulate in the pores only on exposure for extended periods to a very high relative humidity; and when it decreases by a small amount below 99% most pores will quickly empty.
Low porosity in concrete can be achieved by keeping the w/c ratio low.
Concrete with a 0.4 w/c ratio has such good frost resistance that even the need
for air entrainment may be questioned. Unfortunately, it cannot be said that concrete made with a very high w/c ratio is also durable; although the porosity increases very significantly, it does not reach a sufficiently high value. In
fact, 0.6 or 0.7 w/c concrete is highly frost susceptible because considerable
amounts of water can be contained in the pore system; yet the permeability is not very high and the average pore diameter still quite small.
Reducing moisture loss from porous bodies can also lead to a high degree of saturation and, because of this, to frost damage. Painting a concrete or brick wall with a "non-breathing" oil paint can transform a well functioning structure to one with a durability problem. Special mention has to be made of waterproofers, sealants, and impregnants. Reduction of porosity and a in- gress of water is very beneficial in principle. Often, however, these agents reduce the rate of evaporation very substantially while not preventing all water from entering the network. Silicon, for example, prevents rain penetration but not condensation of moisture; at the same time, by drastically reducing water loss it promotes high water content and, thus, frost damage. This ill- ustrates the need for a careful analysis of the situation before changes are made in a system.
Architects, engineers and designers can do a great deal to minimize frost damage by providing means to avoid accumulation of water in roads, building walls, and bridge decks. Horizontal surfaces should have slopes, camber or flashings where practical.
Permeability, too, is an important parameter since it affects the amount of water accumulated before freezing and the rate with which excess water can be expelled during freezing, and therefore, frost resistance itself. Material properties that have a bearing on diffusion rate also influence frost action. Among these the length of the diffusion path or size of the porous solid is of great significance.
VI. SALT SCALING
Apart from exceptional cases, de-icing salts do not attack concrete through chemical action. The aggravating effect of de-icing salts on frost action results from the fact that solutions have lower vapour pressure than pure
liquids ; for a given water pressure or relative humidity, therefore, a higher
degree of saturation of the concrete results than would in the absence of any solutes. Conversely, .if concrete or another porous body is saturated with a solution other than water, desiccation or evaporation will commence at lower relative humidity than would be the case with pure water in the pores. For
example, a porous solid containing a 26 % NaCl solution will remain saturated
until the relative pressure of the water vapour over it falls below 77 %. The
importance of the degree of saturation in frost action has already been em- phasized. High moisture content can thus account for the aggravating effect of
de-icing salts. ^
VII. AIR ENTRAINMENT
The method of improving frost resistance of concrete by incorporating air voids in the paste at regular and frequent intervals has been used with great success for many decades. The protective effect can be ascribed to the provision of a refuge for the excess water, eliminating the need for water to migrate to
external surfaces or, conversely, avoiding the~~accumulation of excess water in
the pores. It is obviously very important for the voids to be distributed even- ly and to be present in large numbers.
Air entrainment is such an effective means of frost protection that failure in cement paste was attributed until recently almost automatically, and with a fair degree of certainty, to insufficiency of air entrainment. Durability of concrete containing high range water reducers have shown that present opinion regarding the relation between frost resistance and character- istics of air entrainment is questionable. In such concrete the recommended
0.2 mm spacing cannot normally be achievedbut, in spite of thi~~performance in
freezing and thawing is quite good, certainly better than expected on the basis of paste characteristics. This puzzling phenomenon has been discussed at con-
siderable length (4) without consensus. One explanation may bethat superplasti-
cizers,because of high fluidity and antifoaming characteristics, tend to
eliminate large voids, > 200 pm, but not alter the number of the smaller pores
of 1
-
2 um diameter. These are apparently the ones necessary for good frostprotection. Verification of this theory is in progress.
Air entrainment is most commonly achieved by the addition of a surface- active agent to the plastic mix but the same effect can be achieved by the addition of preformed voids. Hollow microspheres (HMS) made of plastic (5) and,
recently, of inorganic materials have also been found effective (6). Those
having densities similar to that of cement can be easily and uniformly dis- tributed in the mix. Some of the porous inorganic materials are inexpensive and seem to be quite promising.
The performance of a given concrete cannot be predicted unless the characteristics of the environment to which it will be exposed are known. The term "frost resistant concrete" is somewhat misleading because it implies that the concrete will withstand the action of freezing and thawing under any con- ditions. The term "frost resistant" should mean that only under conditions to which the concrete is normally exposed, will no noticeable damage occur.
Testing of frost resistance requires the estimation of the conditions that will determine the state of the concrete during freezing and thawing. As discussed, the most important of these is moisture content. It is n6t usually realized how difficult it is to estimate correctly the actual degree of satu- ration during freezing and this uncertainty is the major source of error in predicting frost resistance. The test procedures can be divided into direct and indirect methods.
8.1 DIRECT METHODS FOR TESTING FROST RESISTANCE
The specimens are subjected to accelerated freezing and thawing cycles. Notwithstanding the uncertainty concerning the appropriateness of the moisture content of the specimens during test, it is possible to arrive at some meaning- ful evaluation provided two conditions are met: that all essential parameters of the test are vigorously controlled and good reproducibility is achieved; and that field experience involving the intended application is known and can be expressed in terms of parameters determined in the accelerated freeze-thaw test. For example, if concrete that performed well as highway pavement ex- pands 0.01% after being subjected to 300 freeze-thaw cycles, testing of new concrete mixes for this application can be reduced to comparing them to the standard. It is a reasonable assumption that concrete expanding less than 0.01% when subjected to 300 freeze-thaw cycles in the same apparatus will also perform well.
Unfortunately, direct experience is often lacking and it is difficult to make predictions with a reasonable degree of confidence. Still, by keep- ing the stated principles and the features of the mechanism in mind the direct method can be, and has been, used as a basis for useful evaluation of frost resistance.
8.2 INDIRECT METHODS FOR ASSESSING FROST RESISTANCE
These are based on determination of parameters such as air content, spacing factor, critical degree of saturation (7), pore size distribution, and surface area (8). Each test gives valuable information on some aspect of freeze- thaw durability. Unfortunately, none of the tests will provide clear-cut answers, so that the need for careful analysis of the material properties and enviromnent- a1 characteristics remains imperative.
REFERENCES
(1) G. LITVAN
-
Phase Transitions of Adsorlltes IV. Mechanism of Frost Actionin Hardened Cement Paste, J. Am. Ceram. Soc.
55
(1). 38 (1972).(2) G. LITVAN
-
Phase Transitioas of Adsorbates 111. Heat Effects andDimensional Changes in Nonequilibrium Temperature Cycles, J. Coll. Inter-
face Sci.
38
(1). 75 (1972).(3) G. LITVAN
-
Phase Transitions of Adsorbates VI. Effect of De-icing Agentson the Freezing of Cement Paste, J. Am. Ceram. Soc.
58
(1-2), 26 (1975).( 4 ) Developments in the Use of Superplasticizers, V.M. Malhotra Ed., American
Concrete Institute, Detroit, SP-68, p. 189, 215, 253, 269, 359, 515, (1981).
(5) H.A. SOMMER
-
New Method of Making Concrete Resistant to Frost and De-icingSalts. Betonwerk und Fertigteil Techn., No. 9, 476 (1978).
(6) G. LITVAN, P. SEREDA
-
Particulate Admixture for Enhanced Freeze-Thaw Re-sistance of Concrete. Cement and Concr. Res.
g,
53 (1978).(7) G. FAGERLUND
-
The Critical Degree of Saturation Method of Assessing theFreeze-Thaw Resistance of Concrete. Mat. & Struct. 10 (58). 205 (1977).
(8) G. LITVAN
-
Testing the First Susceptibility of Bricks, ASTM SpecialPublication 589, 123 (1975).
BIBLIOGRAPHY
Proceedings of the International Colloquium on Frost-Resistance of Concrete, Vienna, June 1980. Austrian Cement Manufacturers Association, Vienna, Austria.
*********
This paper is a contribution from the Division of Building Research, National Research Council of Canada, and is submitted with the approval of the Director of the Division.
T h i s paper, w h i l e being d i s t r i b u t e d i n r e p r i n t form by t h e D i v i s i o n of B u i l d i n g Research, remains t h e c o p y r i g h t of t h e o r i g i n a l p u b l i s h e r . It should n o t be reproduced i n whole o r i n p a r t w i t h o u t t h e permission of t h e p u b l i s h e r . A l i s t of a l l p u b l i c a t i o n s a v a i l a b l e from t h e D i v i s i o n may be o b t a i n e d by w r i t i n g t o t h e P u b l i c a t i o n s S e c t i o n , D i v i s i o n of B u i l d i n g R e s e a r c h , N a t i o n a l R e s e a r c h C o u n c i l of C a n a d a , O t t a w a , O n t a r i o ,
