Publisher’s version / Version de l'éditeur:
Materials Research and Standards, 4, 12, pp. 663-666, 1965-03-01
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
Delayed hydration in white-coat plaster
Ramachandran, V. S.; Sereda, P. J.; Feldman, R. F.
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=d3631a4c-14fb-47cb-a910-cde22ae3cbbc https://publications-cnrc.canada.ca/fra/voir/objet/?id=d3631a4c-14fb-47cb-a910-cde22ae3cbbc
Authorized Reprint from the Copyrighted Materials Research & Standards, Vol. 4, No. 12, December 1964 Published b y the American Society for Testing and Materials, Philadelphia 3, Pa.
Delayed Hydration in White-Coat Plaster
By V. S. RAMACHANDRAN, P. J. SEREDA, and R. F. FELDMANThe Case of the Falling Plaster-some careful detective work nabs the culprit, unhydrated magnesium oxide.
S E V E R A L I N V E S ~ G A T O H S have re-
ported the occurrence of "popping" or 'lbulging" in white-coat plaster prepared froin a liine putty gauged with plaster of paris [I-91.1 I n these plasters (this term is used here for set plasters), t,he onset of slight bulging was noticed a t periods ranging from 5 to 15 years, with progressive increase and final separatioil from the wall. Attempts have been made to settle the controversial question of whether the delayed hy- dration of nlagnesiunl oxide is responsible for this type of deterioration. The presence of a great nuin- ber of constituents such as free water, CaSO4.2I-Iz0, Ca(0I-I) 2, RIIg(OI1)2, hIIgO, CaCOa, CaS04, SiOz,
A1203, Fe203, and possibly other coinpounds of nlagnesiunl in the plaster maltes identification diffi- cult.
'
The italic nunibers in brackets refer t o the list of references appended t o this paper.'I'he present investigatio~l confirins that delayed hydratioil of MgO is responsible for failure of a white-coat plaster that exhibited l'bulgir~g" nearly eight years after application. A new technique using conlpacts has been applied to study the un- soundness of plasters, and it is hoped to extend this nlethod to other related building materials. Procedure
Plaster specin~erls were obtained from the malls of the Westminster Hospital, London, Ontario, Canada with the ltind cooperation of the Canadian Departnlent of Veterans Affairs. The white-coat plaster on the walls shows extensive "bulging" after a period of about eight years. The specimens were collected in the form of flaltes of uneven thickness vaiying from 0.15 to 0.35 in. Bulges of various sizes and shapes were noticed which, when removed, exposed irregular areas of the base coat measuring from 2 in. to over 2 ft in any direction. Fragments
Table 1 -Some Physico-Chemical Constants for the Reactions Occurring in White-Coat Plaster.
M o l a r N a m e of Cliernical Molecular D e n s i t y , Volume, Compoileilt Composition Weight g / m l a n ~ l
Possible Cliemical Transformation M o l a r Volume Change, Per Cent
Plaster of p a r i s . . . . . .CaS04.%1-1?0 145.15 2 . 7 j b 52.796 CaS04.x.iHaO + CaSO4.2II?O 41.13
G y p s u m . . . . CaSO,. 2H?O 172.18 2.32 74.218
L i m e (calcia) . . . CaO 56.08 3.37 1 G . 642 C a O -+ C a ( O l I ) ? 90.03 H y d r a t e d lime.. . . .Cn(OH)? 74.10 2.343 31.623 Ca(0E-I)z-+ CaCO3 1G.75 Calcite. . . .CaC03 100.01) 2.711 36.92 CaO -+ CaCOr 121.87
Periclase. . . B r u c i t e . . . . M g O 40.32 3.58 11.261 M g ( O l I ) ? 58.34 2.385 24.462 a See Ref. [ I f 1. "ee llef. [ 1 8 ] .
of white-coat plaster froin which the paint was scraped were used for the study. The standard chen~icals used for autoclave expansion studies in- cluded precipitated CaS04.21-120, iVIg(OI-I)2,
and CaC03.
The differential thernlal analysis (DTA) of the specimens was carried out by an autoinatic ap- p a r a t ~ s . ~ The rate of heating was 10 C per min, and the sensitivity for full-scale deflection on the differential recorder lvas 100 pv. I n each case 0.125 g of the specimen was placed in the holder to allow a reasonable assessnlent of the relative intensities of various therlnal effects.
The specimens Itrere ground to pass a 100-mesh sieve, and conlpacts were forined from the powder with a steel mold of 1.25-in. diameter a t a load of 10,000 or 20,000 lb. The diameter of the conlpact was deterinined by a traveling inicroscope to an ac- curacy of 0.001 cm. Details of the method are covered in a previous paper [ l o ] .
A special holder containing several coi~lparti~leilts of copper gauze was used to hold the colnpacts for autoclave treatment. Specimens were autoclaved in a standard autoclave a t 300 psi for 3 hr.
Weight loss values were obtained by heating a known amount of the specimen in a sintered aluinina crucible for 12 hr a t 200 C and 12 hr a t 500 C. Discussion
The inain constituents in a gypsum-gauged white- coat plaster are gypsum, formed by the hydration of plaster of paris (hemihydrate), and calcium hy- droxide, formed by the hydration of calciuni oxide, provided the linie is obtained froin high-calcium lime- stone. I n a lime obtained by calcining dolomitic limestone the principal constituents before applica- tion on the wall are calciuln hydroxide, magnesium hydroxide, and considerable amount;~ of unhydrated niagnesium oxide (periclase), especially if the liine- stone is produced a t high temperatures and is not pressure-hydrated. The norinal doloniitic hydrated lime (type
N)
belongs to this category. Some calciuni carbonate is forined by the carbonation of calcium hydroxide, and may also be present as anM o d e l 12BC. T h e R o b e r t L. Stone Co., Austin, Tex.
impurity in the gauged plaster or as a core material. Snlall amounts of calcium sulfate (anhydrite), silica, alumina, and ferric oxide also occur in the plaster. There is little or no evidence of the formation of inagnesium carbonate.
A gypsum-gauged plaster containing (mainly) plaster of paris, calciuln oxide, and magnesiun~ oxide would transforni to gypsum, Ca(OH),, CaC03, and Mg(OH), after being treated with water and sub- jected to normal exposure.
Table 1 gives the molar volunies of various con- stituents and the theoretical nlolar volunle chanees L 3 for sonie of the possible reactions. I n a hydration reaction, which can be represented as R
+
H -+ RH,where R = anhydrous material, I1 = water, and R H = hydrate, volume changes occur that can be defined as follows:
( 1 ) The niolar volulne change refers to the differ-
ence between the inolar volume of the hydrate and the niolar volume of the anhydrous niaterial brought about by the changes in the lattice parameters as the crystal of anhydrous material converts to a crystal of the hydrate. This change is found by calculation
V. S. RAMACHANDRAN received a B.Sc. from Mysore University, an M.Sc. in physical chemistry from the Banaras Hindu University, and a D.Phil. from the Calcutta University of India. From 1949 to 1956 he was engaged in research on electrical conductivity in gases and on catalysis. He joined the Central Building Research Institute o f India in 1956. Since 1 9 5 6 he has been working in clay mineralogy and lime, gypsum, and cement chemistry. At present he is a guest worker a t the Division o f Building Research, National Research Council, Ottawa, Ont., Canada.
PETER J. SEREDA received his M.Sc. in chemical engineering from the University o f Alberta. From 1 9 4 4 to 1948 he worked at the Atomic Energy o f Canada, Ltd., at Chalk River. He joined the Division of Building Research, National Research Council, Canada in 1948. He has been engaged in research on corrosion, electrostatic hazards, and properties o f building materials. He has been an active member o f ASTM Committee 8-3 and also represents the National Research Council on Committees C-1 1 and E-1
.
R. F. FELDMAN was granted his M.A.Sc. b y Toronto University in 1959 and has since been employed a t the National Research Council. He has been engaged in research on many physical and chemical phenomena related to materials, their formation, and their deteriora- tion. He has also been involved in a basic study o f sorption and length changes on various building materials.
or determination of the nlolecular weight and density.
( 2 ) The apparent volun~e change can be defined
as the difference between the apparent volu~ne of the hydrate and the apparent volunle of the anhydrous material. The apparent volume is that bounded by the external surfaces and includes the space and the liquid between the particles of the material. The measure~nent of the length changes of molded materials falls in this category.
( 3 ) The absolute volume change represents the
difference between the absolute volume of the hy- drate and the sum of the absolute volu~nes of the anhydrous material and the water reacting with it. This difference is usually a snlall negative quantity. Dilatonletric nleasurenle~lts arc used to find this quantity.
The increase in nlolar volunle inay bc several times greater than the apparent volume change; for ex- ample, it has been found thal a calcium oxide com- pact hydrated in water increases by 22.8 per cent in terms of apparent voluine change, whereas the theoretical nlolar volunle change should be as high as 90.03 per ccnt. Obviously, several factors such as pore sizc and volume, shape and size of ciystals, and mechanism of reaction play a part in the apparent volume change pl~enomenon.
On hydration, plaster of paris converts to CaSO.,. 2H20, with an increase in molar volume of 41.13 per cent,. The hydration reaction is sufficiently fast as to be allnost conlplete during the setting period. Hence bulging in the plaster cannot be attributed to this expansion.
The volu~ne increase resulting fro111 the formation of Ca(OI-I)2 from CaO is as high as 90.03 per cent; but, because CaO reacts rapidly with water-even before setting arid hardening of the plaster-it does not contribute to delayed expansion. Stutterheim and co-worlmrs [9] found that, in dolon~ites calcined in the range fro111 600 to 1400 C, calciuin oxide hydrated rapidly and co~npletely for all tempera- tures and times studied.
The formation of CaC03 from Ca(OR)2 is attended by a voluine change of about 16.75 per cent. This reaction is extremely slow under normal conditions of exposure. Wells ct a1 [6] have shown that carbona- tion of Ca(OIl)2 is far froin complete even after a number of years. Although it involves a large vol- unle increase of 121.87 per cent, the reaction of CaO with COz to form CaC03 is improbable because the presence of free CaO is unlilcely in the plaster; even if free CaO were present, hydration mould be the preferred reaction.
h4gO converts to Mg(OH)2 by a volume increasc of 117.22 per cent. Magnesia is very liliely to be present in an overburnt form in dolomitic limes. I n an overburnt for111 iUgO hydrates extreinely slowly, so that even after the setting of a finish coat there is a considerable quantity of l a g 0 in the plaster. There is every reason to believe that the gradual hydration of MgO is attended by a corresponding ex- pansion. Campbell [I31 found that MgO prepared from MgCO3 a t 800 C hydrated completely in three
days, that prepared a t 1000 to 1100 C hydrated m about three months, and that formed a t 1200 C re- quired three years. I-Iydration was not conlplete even after six years in MgO prepared a t 1300 C. Wells [5] reported that anlong 27 regularly hydrated dolomitic limes an average of only 15.7 per cent of the total i\/IgO was hydrated in each.
Experimental Results
Thc spcci~nens were subjected to differential thermal analysis to identify the conlponents present in the plaster. Figure 1 shows the results. Plaster exhibits an endothermic doublet with peaks a t 142 and 158 C, owing to stepwise dehydration of gypsum. An endothermic peak a t 391 C corresponds to the deconlposition of I\Ig(OH)2; the other, with a pcak a t 464 C, should be the result of &(OH)? decompo- sition. Pure Ca(OH), exhibits an endothermic peali above 500 C, and the temperature of the peak is low- ercd by the presence of other constituents. A large endothermic effect with a peali a t 800 C is caused by decarbonation of CaC03 formcd over several years by thc reaction of &(OH), with atmospheric COS The endothernlic peak area resulting from CaCOj de- co~nposition is several times greater than that owing to decomposition. A co~nparison of the areas mould not give a direct indication of the rela- tive amounts of &(OH), and CaC03 in the speci- lnen, because the heat of reaction per nlole of CaC03 is about three times that of Ca(OH)2 decon~position. The DTA results show that the specinlen under question is gypsum-gauged lime, possibly obtained from doloinitic limestone.
The presence of Ca(OH), and i\iIg(OH)2 in the failcd plaster indicates the possibility of the presence of unhydrated MgO in the sample. I t is difficult to prove its presence by chemical analysis. Another approach is to hydrate the MgO undcr severe condi- tions and study the fornlation of >Ig(OH)z by mcas- uring thc dilnensional changes or by lneasuring the intensity of the endothermic effect resulting from hIg(0I-I), decomposition a t about 400 C by DTA.
An unreactive RIgO can be hydrated by subject- ing it to an autoclave treatnlcnt in steal11 a t 300 psi for 3 hr. Pieces of the failed white-coat plaster 1 in. square wcre trcatcd in an autoclave, but the speciineils showed unequal expansion and such warping that expansion nleasureinents could not be carried out by this method. An alternative method of forming the specinleris for autoclave trcatnlent was morlccd out.
White-coat plaster was ground to pass through a 200-mesh sieve, and circular conlpacts of a nominal diameter of 1.250 in. were formed a t a load of 10,000 or 20,000 lb. The two autoclave-treated compacts showcd uniforin linear expansions of 12.22 and 14.67 pcr cent, respectively. I n order to verify whether any other constituent in tlie plaster com- pact contributed to this expansion, coinpacts mere forined from pure Ca(OH),, Mg(OH),, CaS04.2Hz0, and CaC03 and were subjected to autoclave treat- ment. I n none of these coinpacts could measurable expansion be observed. Ilence, it was inferred that
the expansion was nlainly owing to the forination of Mg(OH)2 from MgO.
The DTA of the autoclaved sanlple shows endo- thermic peaks corresponding to Ca(OH)2, CaC03, and Mg(OH)2 (Fig. I ) . The endothermic effect owing to Mg(OH), decomposition, however, is much more intense for the autoclaved than the untreated specimen, showing that the autoclave treatment re- sults in the formation of R ~ l g ( 0 H ) ~ fro111 unreactive MgO. I t is surprising that an autoclaved specinlen fails to show endothermic effects for the presence of gypsum. I t was thought that the gypsum is either decomposed or leached away by this treatment. Hence, the effect of autoclave treatment on a com- pact of precipitated CaS04.21-120 was studied. The DTA results are presented in Fig. 1. The unreacted material shows two endothermic pealrs a t 153 and 188 C. The autoclaved specimen shows alnlost com- plete absence of endothermic effects, indicating that during autoclave treatment gypsum is deconlposed to CaS04.
The above results are confirmed by weight-loss measurements. The untreated CaS04.21-120 speci- men heated a t 200 C showed a loss of 20.57 per cent, whereas the autoclaved specimen sho~ved a loss of only 0.02 per cent. These results indicate that the loss of 20.57 per cent corresponds to the dehydration of 2 moles of water from gypsum. The failed white- coat plaster heated to 200 C exhibited a loss of 7.10 per cent. This loss corresponds to a gypsum content of 33.97 per cent, which is about the quantity present in nlany gypsum-gauged white-coat plasters. The formation of Mg(OH), after autoclave treatment can also be verified by weight-loss measurements. The difference between the loss a t 200 C and that a t 500 C should represent the loss of (OH) water from Ca- and @(OH2). An autoclaved specinlen showed a loss of 7.33 per cent between 200 and 500 C,
whereas the untreated white-coat plaster showed a loss of 2.18 per cent. The 5.15 per cent difference in weight loss can be attributed to the formation of Mg(OIl)2 in an autoclave treatment.
Differential thenilla1 analysis, weight loss, and autoclave experinlents have established that the delayed hydration of h4gO is mainly responsible for bulging in the white-coat plaster investigated. Study also indicates that the use of compacts shows great promise as a method of determining unsound- ness in mortars, plasters, and cements. The ob- vious advantages of the method are the elimination of water and binding agent for the formation of a compact, magnification of expansion resulting from dense packing of particles, the small amount of material required, the possibility of carrying out tests on several specimens a t a time, quickness, and reliability.
Aclcnowledgment
The authors thaitlr Mr. L. J. O'Byrne for his val- uable experimental assistance. This paper is a contribution from the Division of Building Research, National Research Council, Canada, and is published with the approval of the director of the Division.
Fig. 1-Differential thermal curves of plaster and gypsum before ond after outocloving.
[I] J. A. Rlurray and H. C. Fischer, "A Study of White Coat Plaster by Differential Thermal Analysis," Pro-
ceedings, Am. Soc. Testing Mats., Vol. 51, 1951, p. 16.
[2] E. S. Newman, "Investigation of Plaster Failure,"
Bulletin, Am. Ceramic Soc., Vol. 26, 1947, p. 117.
[S] D. F. itichardson, "Analysis of Hydrated Limes by s
Thermochernical Method," Industrial & Engineerinq Chemistry, Vol. 19, 1927, p. 625.
[Q] L. S. Wells and I<. Taylor, "Hydration of Magnesia in Dolomitic Limes," Journal of Research, National Bureau of Standards, Vol. 19, 1937, p. 215.
[5] L. S. Wells, W. F. Clarke, and E. M. Levin, "Elpansive Characteristics of Hydrated Limes and the Development of an Autoclave Test for Soundness," Jozlnlnl of Research, National Bureau of Standards, Vol. 41, 1945, p. 179.
[ 6 ] L. S. Wells, W. F. Clarke, E. S. Newman, and D. L. Bishop, "Investigation of Failures of White Coat Plns- ter," Building iMate~ials and Stlucture Report iYo. 121,
National Bureau of Standards, 1051, p. 42.
[ 7 ] E. L. Fox and F. C. Mathers, "Determination of tlie Degree of Hydration of Magnesia in Dolomitic Limes,"
Proceedings, Indiana Academy of Science, Vol. 43, 1934,
p. 113.
[8] F. C. Welch, "The Hydration of Mrtgnesian Lime and the Autoclave Test," Syrnposiuln oil Li~tze, AST'ild
S T P 40, Am. Soc. Testing ~Mats., 1939, p. 10.
[9] N. Stutterheim, T. L. Webb, and B. Uranovsky, De-
velopnzents in Research on the Burning and Hydration of Li~rte and on i t s Use in Building, Building itesearch
Congress, London, 1951, p. 120.
[lo]
P. J. Sereda and R. F. Feldinan, "Compacts of Powdered Materials as Porous Bodies for Use in Sorption Studies,"Jozirnal of Applied Cl~ctnistry, Vol. 13, No. 4, 1963, p. 150. [Ill C. D. Hodgman, Handbook of Clte~nistry and Physics, Chemical Rubber Publishine Co.. Cleveland. Ohio.
-
1954.[I21 B. M. O'Kelly, "Physical Changes in Setting Gjq~sum Plaster," Bulletin, Am. Soc. Testing Mats., No. 237, April, 1950, p. 55.
[IS] E. D. Campbell, "On the Influence of the Temperature of Burning on the Rate of Hydration of Magnesium Oxide," Industrial Engineering Chelnistry, Vol. 1, 1909, p. 665; Vol. 10, 1915, p. 595.
