G^uk SaaXLOU^ STATE OF ILLINOIS WILLIAM G. STRATTON, Governor DEPARTMENT OF REGISTRATION AND EDUCATION VERA M. BINKS, Director ***** Effects of Hydration Procedures and Calcination in the Presence of NaCI on the Properties of Lime Hydrates D. L. Deadmore J. S. Machin DIVISION OF THE ILLINOIS STATE GEOLOGICAL SURVEY JOHN C. FRYE, Chief URBANA CIRCULAR 270 1959 Effects of Hydration Procedures and Calcination in the Presence of NaCI on the Properties of Lime Hydrates jff JN°'S STATE GEOLOGICAL SURVEY 3 3051 00003 8426 Effects of Hydration Procedures and Calcination in the Presence of NaCI on th Properties of Lime Hydrates D. L. Deadmore and J. S. Machin ABSTRACT This investigation is part of a study of factors that affect the properties of lime putties. The influence of salt (Na CI) content and calcining temperature on the surface area and on the reactivity of the calcinates of 1) dolomite, 2) calcium carbonate (CaCOJ, and 3) basic magnesium carbonate was studied. It was found that both the presence of NaCI and increased calcining temperatures acted to decrease the surface area and reactivity of all three materials. The influence of NaCI content and calcining temperature of the calcinate on the plasticity of the dolomitic hydrate was inves- tigated. Two hydration procedures were used to convert the calcin- ates to the hydrated form - an atmospheric pressure and a high- pressure method. The results indicate that neither the presence of NaCI nor increased calcining temperatures had any beneficial influ- ence on the plasticity of the low-pressure hydrate. However, NaCI appears to improve slightly the plasticity of the high-pressure hy- drate. It was found that calcium hydroxide (Ca(OH) ) was consid- erably coarser and of lower plasticity than magnesium hydroxide (Mg(OH) ). Both are composed of thin hexagonal platelets. INTRODUCTION The addition of salt (NaCI) to CaCO-, before calcination, accelerates the crystal growth of the CaO and decreases its reactivity toward water, according to Noda and Kan (19 37). Noda and Oka (1938) noted that the growth of MgO crystals was accelerated when NaCI was added to the carbonate before calcination. Briscoe and Mathers (1927) reported that when dolomites naturally con- tained more than 0.07 percent chloride ion, a plastic hydrate was produced, where- as dolomites containing less than this amount yielded low plasticity hydrates. Lamar and Shrode (1953) have shown that the liquid inclusions in dolomites contain Na + and Cl~ ions, among others. These reported influences of sodium chloride on the properties of the com- ponents of calcined dolomite, and the fact that dolomites often contain liquid in- clusions rich in NaCI, suggested that it might be of interest to measure some of the properties of the calcinates made from carbonate starting materials that con- tained varied amounts of NaCI added before calcination. Because plasticity is closely related to the workability of masonry mortars, the property of plasticity and the basic factors that affect it are the focus of the present study of factors in- fluencing the properties of hydrated limes. [1] 2 ILLINOIS STATE GEOLOGICAL SURVEY RAW MATERIAL SELECTION Four hundred pounds of crushed Niagaran dolomite were kindly furnished by the Marblehead Lime Company for this study. The material had been crushed and sized at the plant,, The particle-size distribution of the dolomite was: Weight Screen +0.371 inch -0. 371 inch + 4 mesh per inch -4 + 10 mesh -10 + 14 -14 + 20 -20 + 28 -28 percent 10. ,8 48. ,2 35. ,7 3. ,0 1. ,0 0. .3 1. ,0 No chemical analysis of this material was made, but it is considered to be similar to samples from the same quarry analyzed previously at the Illinois Geologi- cal Survey. The following chemical analysis is taken from Lamar (19 57): Si0 2 0.11 Na 2 0.06 A1 2 3 0.30 C02 47.05 Fe 2 3 0.19 Ign. loss 47.87 CaO 31.20 so 3 0.10 MgO 20.45 MnO 0.015 The sum of CaC0 3 and MgC0 3 is 98.45 percent, which shows that this is a rather pure stone from the industrial viewpoint. The stone was used as received with no further treatment. The raw material sources of CaO and MgO were CaC0 3 (precipitated) and basic magnesium carbonate.respectively. Both were very fine powders. The mag- nesium carbonate was labeled "U.S.P,, Heavy Powder, MgC0 3 Formulae Weight, 84.33, " but x-ray diffraction and ignition loss proved it to be the basic MgC0 3 (5MgO • 4C0 2 • 5H 2 0). The CaC0 3 was precipitated, U.S.P. (light). Its chem- ical analysis was given by the supplier as Si0 2 - 0.03 Ti0 2 - 0.09 A1 2 3 - 0.06 Fe 2 3 - 0.05 MgO - 0.38 CaO - 55.53 Na 2 - 0.06 K 2 - 0.09 The NaCl used was reagent grade crystals from Baker and Adamson. PROPERTIES OF LIME HYDRATES 3 EXPERIMENTAL PROCEDURES Sample Preparation for Calcination The desired amount of carbonate starting material with the same particle- size distribution as given above (1400 grams of dolomite, or 600 grams of CaC03 [powder] or basic MgC03 [powder]) was weighed into a 1500 milliliter beaker, to which was added 450 ml of distilled water containing in solution the desired amount of NaCl. This was placed on a steam bath and the water evaporated overnight. Even the samples that contained no added salt received the same pretreatment. The NaCl contents reported are in weight percent based on the uncalcined carbon- ate starting material. Calcination The pretreated samples were placed in shallow refractory saggers (4 inches high x 8 inches in diameter) to a depth of approximately three inches. Two such samples, each with the same amount of NaCl, were calcined at the same time. A muffle-type Globar-heated furnace was used. The two saggers occupied most of the furnace's floor space, but there was 2 to 3 inches overhead clearance. A tube was inserted through the rear of the furnace into the muffle at a level of approximately one inch above the saggers. By means of this tube a slow current of air was flushed through the furnace during the entire calcination cycle. A hole in the furnace door and various other openings allowed the air to escape. After the cold furnace had been loaded and the stream of air started, the power to the furnace was turned on. The heating rate was controlled manually so that the desired maximum temperature was reached in seven to eight hours. A platinum-platinum +10 percent rhodium thermocouple inside a porcelain protection tube was placed two inches above the saggers for temperature measurement. The time at this maximum temperature was two hours, unless otherwise specified. After this soaking period the power was shut off and the furnace and its contents, with the air stream still on, was allowed to cool for 16 hours. The material, still at 350° C, was removed from the furnace, immediately placed in two-quart Kerr jars, and the self-sealing lids were tightened. As the contents of the jars cooled, a vacuum was produced in the jars. These were stored until needed. Reactivity of Calcined Carbonates with Water To establish the reactivity of the calcined oxides with water an apparatus similar to that described by Murray, Fischer, and Sabean (1950) was assembled. Figure 1 is a schematic drawing of this apparatus. The function of the apparatus is to measure the temperature rise in a given amount of water when a given amount of calcinate is added. A ratio of 7 parts of distilled water to 1 part of calcinate by weight was used. Operation is as follows: 200 grams of distilled water at 26° C is placed in the Dewar flask, the stirrer is started, the recorder is calibrated against the poten- tiometer, then the potentiometer is removed from the circuit and the recorder con- nected directly to the thermocouple in the Dewar calorimeter. Now 28. 6 grams of the calcinate, ground to a powder with a mortar and pestle, is added through the powder funnel; the temperature rise of the water vs. time is recorded by the recorder. The values reported, for comparative purposes, are the temperature rises after 1000 seconds or AT^QOO' tne average value of three runs is reported. The reproducibility is ± 15 percent. ILLINOIS STATE GEOLOGICAL SURVEY _300rpm constant speed f synchronous motor Powder funnel \ Cu leads Water Calorimeter Fig. 1. - Reactivity apparatus. Surface Area Surface areas were determined by low temperature, low pressure adsorption of nitrogen, after the method of Brunauer, Emmett, and Teller (19 38). The surface measurements of those samples having areas in excess of one m /g (square meters per gram) is reproducible to about ± 5 percent. Hydration Two means of hydrating the calcinates were used- an open-dish, atmospher- ic pressure method, and a closed-vessel, high-pressure method. The operation of the atmospheric pressure method involves weighing 700 to 800 grams of the calcined stone into a 12-inch diameter evaporating dish (this cal- cinate has essentially the same size distribution as the uncalcined stone). Dis- tilled water in an amount equal to 30 percent of the weight of the stone was placed in a separatory funnel, which had been previously adjusted to deliver at the rate of 7.0 ml per minute. The stone and water were stirred with a steel rod during the entire course of the hydration. The maximum temperature of the calcinate-water mixture (the "bed" temperature) was observed by stirring at frequent intervals with a thermometer. After all the water had been added, the moist mixture was placed in an open, two-quart Kerr jar, which was set in an oven and dried for 16 hours at 105° C. After the drying period, a self-sealing Kerr lid was tightened on the jar, and after cooling to room temperature the hydrate was ball-milled in a two-gallon porcelain mill for one hour. The milled material was returned to the Kerr jar for storage. After considerable experimentation with various setups in an attempt to hydrate the calcined dolomite almost completely, the apparatus shown in figure 2 was finally used for all pressure hydrations unless otherwise specified. The procedure used for the pressure hydration of the calcined dolomite was as follows: 625 grams of calcined solids, having roughly the same particle size as the original stone, were weighed into the inner can and the bomb was closed and placed in a cold oven. A vacuum was drawn on the bomb through the release PROPERTIES OF LIME HYDRATES 0-200 psi pressure gauge Oven thermometer Oven wall Pressure releose volve 2-quart steel can, perforated, open at top and bottom Screen-wire support Fig. 2. - Pressure hydration apparatus valve and 475 ml of water (2.8 times the theoretical amount necessary for complete hydration) was drawn in on top of the calcined stone. The valve was closed and the oven turned on. In about three hours the pressure in the bomb rose to 140 psi gauge (the maximum pressure of all hydrations unless otherwise specified). This pressure was maintained for 4 1/2 hours. The release valve was then opened and the pressure became atmospheric. The bomb was then opened and its contents dried in a two-quart jar at 105 °C for 16 hours. After drying, a self-sealing lid was tightened on the jar. This material was then ball-milled for one hour. In the hydration of CaO and MgO samples, the procedure described above was followed, except that 200 grams of solids and four times the amount of water necessary for complete hydration of the oxide were used. Plasticity Determinations The visco-plastic properties of the hydrate pastes or suspensions were ex- plored by means of the conventional Emley plasticimeter. This apparatus was con- structed at the Geological Survey after A. S.T.M. specifications (1949). Porcelain base plates were used. The plates were made in the University of Illinois Ceramic Engineering Department to conform to the prescribed absorption rates. The plastic- ity determinations were made according to the standard procedures described by the A. S.T.M. method C-l 10-49. The plasticity was determined immediately after tem- pering with water and again after soaking for 24 hours. The water at standard con- sistency, as given in the text, is the weight percent of water in the total water- solid mixture necessary to produce a paste of standard penetration on the penetrom- eter (standard consistency). In most cases only one determination of plasticity was possible, due to the lack of hydrate. Where duplicate runs were made, the reproducibility was within ± 15 percent of the average value. Drying of Soaked Hydrates In order to study the hydrates after they had been soaked, to see what alter- ations had taken place during soaking, it was necessary to dry the samples in such a way as to minimize any alteration caused by the drying itself. The method used to dry the sample is similar to that given by Wells and Taylor (19 37). 6 ILLINOIS STATE GEOLOGICAL SURVEY After the Emley value of the soaked paste was determined, a 75-gram sam- ple of the wet putty was placed in 150 ml of absolute ethanol and shaken for several minutes. The sample was then transferred to a Buchner vacuum filter, the liquid removed by suction, and the cake washed with ethanol and ethyl ether. The nearly dry cake was placed in a vacuum desiccator, which was continually pumped for 24 to 48 hours to remove the last ethanol and ether. The dried samples were studied by various techniques, as will be described. Carbon Dioxide Determinations The method used to determine carbon dioxide depended on the acid evolution of the CO2 with adsorption on ascarite. The procedure and apparatus are similar to those given by Hillebrand and Lundell (1953). Ignition Loss and Composition Estimation The ignition loss was determined by heating weighed samples in platinum crucibles to 975° C for two hours. From the ignition loss of the dolomitic hydrate the composition of the hydrate was calculated. The calculations were based on the assumptions that all the CaO had been hydra ted in the hydration process, that there was no carbonation of the hydrate, and that the original dolomite had the theoretical dolomite composition. The validity of these assumptions was checked. The first assumption is widely accepted because CaO hydrates so readily. A rough check of this assumption comes from an examination of the x-ray powder patterns of the hydrates which never showed any CaO lines. The validity of the second assumption was checked by determining the CO2 content of some hydrates. It was found to be less than one percent (tables 6, 7). A theoretical dolomite, CaMg(C03)2, contains 21.9 percent MgO and 30.4 percent CaO. The chemical analysis given previously, which is assumed to represent the stone used here, showed 20.45 percent MgO and 31.2 percent CaO. For our purposes this is a rather close approach to the theoretical composition of dolomite. It is evident that the compositions of the hydrates calculated from the ig- nition loss are not highly accurate, and it is estimated that the amount of any component reported may be in error by at least ±3 percent of the reported value. X-Ray Identifications All x-ray diffraction powder patterns were made on a G. E. XRD-3, Record- ing Spectrometer, with Cu Ka radiation. A shallow Al sample holder was used. The powdered samples were placed in the holder and smoothed off with a glass slide. Particle-Size Analysis The particle-size distribution of some of the hydrates was determined by dispersing the hydrate in anhydrous n-butanol in a thermostatted cylinder and with- drawing portions at intervals. Anhydrous n-butanol was used on the recommenda- tion of Bishop (19 39). The portions were evaporated to dryness and the weight of the dry residue was determined. X-ray diffraction patterns of these residues were made. PROPERTIES OF LIME HYDRATES 7 Water Retention of the Hydrates It was observed, in the Emley plasticity test, that those limes that appeared moist for the longest time always had the highest plasticity values. It was therefore thought desirable to make a measurement that would reflect the water-retention ability of the lime putties and substantiate the visual observations of water reten- tion in the Emley test. Pelaez and Murray (1956) devised a penetration test to measure water re- tention. With a penetrometer they measured the thickness of nonpenetrable cake built up on a vacuum filter as a function of time of evacuation. They found that the tendency to retain water was in a linear relationship to the Emley plasticity. The more tenaciously a lime clings to the water in the paste, the greater is its Emley plasticity. Rather than measure the penetration of a filter cake, the apparatus shown in figure 3 was set up so that the volume of water removed from the lime putty as a function of time could be measured directly. The sample of hydrate Rubber seal Coors No. 0, Buchner funnel \ Rubber vacuum tubing \ Air leak, glass capi lube (to give 100 mm. Hg ) ary V. To vacuum pump £3 Mercury manometer was tempered with water to the standard consistency used for the Emley determination of plas- ticity. With the system at at- mospheric pressure, the paste was placed in the Buchner fun- nel that contained a moistened No. 1 Whatman filter paper, and the paste was struck off level with the rim of the funnel. The timer was started when the first paste was placed in the funnel. After the funnel was filled, the vacuum pump was turned on at one minute after the first paste was placed on the funnel. In less than 0.2 minute a vacuum of 100 mm of Hg was attained. The volume of water removed was recorded as a function of time. The test was discontinued when the manometer indicated the vacuum had been lost. The loss of vacuum due to cracking of the cake or the cake pulling away from the funnel wall occurred rather abruptly (in less than one minute). The results are reported as the empirical ratio R, where: -Rubber support Fig. 3. - Apparatus for measuring water retention. R = Total volume of water removed until the vacuum is lost _ ml Total time until vacuum is lost min This term gives a rate of drying out or removal of water from the paste and will be referred to later as " the water removal factor. " Electron Micrographs All electron micrographs were made in the Chemistry Department of the Uni- versity of Illinois by A. E. Vatter. The carbon replica technique, as first described by Bradley (1954), was used. In brief, the method is as follows: a glass slide is coated with a film of collodion, then the sample is spread on the collodion film, and a film of carbon is shadowed over the specimen. Now the composite layer is floated away from the glass slide in water, then transferred to acetone where the collodion is removed. Next the sample is removed from the carbon replica by floating in dilute acid solution. The clean carbon replica is then mounted on a grid for observation. The shadowing angle was 20°. ILLINOIS STATE GEOLOGICAL SURVEY EXPERIMENTAL RESULTS Some Properties of the Calcined Materials Calcined Dolomite The treatment of the Thornton, Illinois, dolomite and the reactivity and sur- face area of the resulting products are shown in table 1. It will be noted in table 1 that the samples calcined at 825° C were held at the maximum temperature for seven hours, but that at all other temperatures the re- tention time at the maximum temperature was fixed at two hours. The longer reten- tion time at 825° C was necessary to complete the decarbonation of the sample, as explained below. The results of calcining samples of Thornton dolomite containing 0.0, 1/2, and 1 percent NaCl at 825 °C for two hours are interesting. X-ray diffraction pat- terns of these calcined samples showed lines due to CaO, MgO, and CaC03 but no lines due to dolomite. The intensity of the CaC03 lines decreased as the NaCl content increased. The ignition loss was determined on these calcinates. The sample that contained no NaCl lost 21.05 percent, the one containing 1/2 percent NaCl lost 10.17 per- cent, and the sample containing 1 percent NaCl lost only 5.01 per- cent. This indicates that under the conditions used here the NaCl greatly aided in the decarbonation of the dolomite. However, even in the presence of 1 percent NaCl not all the carbonate had been decom- posed. However, after seven hours of retention at 825°C no carbonate lines could be observed in the x-ray pattern. So all samples prepared at 825° C were held for this period of time to insure complete decarbona- tion. X-ray diffraction patterns of the samples calcined at 875 °C and higher for a period of two hours, under conditions used here, showed no lines due to carbonate materials. Therefore, this retention period was used for all samples prepared at temperatures of 875 °C and higher. In figure 4 it will be noted that the surface area decreases with both increasing temperature of cal- cination and with increasing salt content. The influence of tempera- ture on the surface area, especially in the region of 900° to 1000 °C, is 800 900 Maximum 1100 1200 Temperature 300 1000 Calcining Fig. 4. - Influence of NaCl on the surface area of calcined dolomite from Thornton, Illinois Time at maximum temperature was two hours PROPERTIES OF LIME HYDRATES Table 1. - Properties of Calcined Dolomite from Thornton, Illinois Maximum Total time NaCl added Reactivity Surface calcining at (wt. percent 1000 area temp. max. temp. based on un- (m /gram) No. (°c) (hrs.) burned stone) (°C) 8 825 7 16.4 3.55 9 ■• It 0.1 15.1 - 10 H H 0.4 9.7 3.10 11 ■i II 0.60 7.2 - 12 M tl 0.90 6.2 - 13 ■1 It 1.25 4.9 2.71 14 H ll 1.75 4.4 - 15 ■I It 2.50 4.2 2.53 16 n It 3.50 4.1 2.45 17 ■I M 5.00 4.1 - 18 875 2 17.8 5.15 19 It It 0.50 9.7 3.60 20 It tt 1.00 7.0 3.24 21 925 2 17.1 3.12 22 It II 0.50 8.8 2.43 23 It It 1.00 6.8 2.35 24 1025 2 13.0 1.62 25 •t •• 0.50 8.0 1.37 26 ■■ •• 1.00 4.0 1.17 27 ii n 2.00 3.2 - 28 H M 3.00 2.3 - 29 ii II 4.00 2.2 1.01 30 1140 2 12.3 1.14 31 ■I It 0.50 7.6 0.93 32 N II 1.00 2.8 0.68 33 1250 2 10.5 0.69 34 n It 0.50 7.0 0.60 10 ILLINOIS STATE GEOLOGICAL SURVEY 45 200 400 600 800 Time - Seconds 1000 1200 Fig. 5. - Temperature rise in calorimeter vs. time for dolomite calcined at various temperatures and containing percent NaCl. greater than that of salt. In the region of 1000° to 1200°C, 1 percent salt has ap- proximately the same effect on the surface area as an increase in temperature of about 100°C. Figure 5 shows some tracings of the calorimeter curves of dolomite calcined at various temperatures. At time zero, when the solids were added, the temperature instantaneously jumped 3° to 8°C, depending on the calcining temperature. In the case of those calcined at 925°C and higher, there was then a period of 100 to 150 seconds of very little activity, after which the temperature began to rise more rap- idly with time. The shapes of these curves are very similar to those given by Knibbs (19 37). When the calcined dolomite comes in contact with liquid water, under the conditions used here, there is an initial, very steep rise in temperature amounting to as much as 50 percent of the total rise. Knibbs calls this the initial liquid adsorption rise; the next period, of little activity, he calls the period of quiescence. Then the main reaction of the calcined dolomite and water proceeds for a considerable period of time. With samples calcined at 875 °C there is no period of quiescence, and after the initial sharp rise in temperature the reaction proceeds quite rapidly. PROPERTIES OF LIME HYDRATES 11 40 36 - 32 - O 28- 24 u o O) rr CaO 0%NaCI a D "\ .. 5%NaCI \ Thornton, Dolomite \0%NaCI (run in salt soln.) 500 600 700 800 900 1000 1100 1200 1300 Maximum Calcining Temperature °C Fig. 6. - Influence of NaCl on the reactivity of CaO, MgO, and dolomite from Thornton, Illinois. Time at maximum temperature, two hours. All run in distilled water except where noted. Figure 6 shows the influence of salt content and calcining temperature on the reactivity of CaO, MgO, and dolomite so that these materials may be compared directly. The properties of MgO and CaO are discussed later in the text. The re- activity of the MgO decreases rapidly with increasing temperature of preparation from the basic carbonate, and by 9 25°C its reactivity is immeasurably small on our apparatus. The addition of 1/2 percent salt decreases the reactivity at low temper- atures, but its influence is very slight at the higher calcination temperatures. The reactivity of CaO without salt is influenced only slightly by tempera- ture of calcination up to approximately 1100°C; it then falls rather rapidly as the temperature rises. The presence of salt seems to decrease the reactivity some- what at temperatures of less than 1100° C. For dolomite the decrease in reactivity is continuous as the temperature of calcination increases. Salt has a very large effect on the reactivity; 1/2 percent of salt is approximately equivalent in its effect on reactivity to a 400° C increase in temperature of calcination. 12 ILLINOIS STATE GEOLOGICAL SURVEY Table 2. - Properties of CaO Maximum Total time NaCl added Reactivity Surface calcining at (wt. percent AT 1000 area temp. max. temp. based on un- (m /gram) No. (°c ) (hrs • ) fa urned CaCO ) (°C ) 33.8 35 825 7 4.42 36 It n 0.50 29.7 0.66 37 II ii 1.0 29.3 _ 38 II •i 2.0 28.4 - 39 II M 3.0 25.9 0.51 40 •I II 4.0 27.2 _ 41 If If 5.0 - - 42 925 2 33.5 3.70 43 " M 0.50 32.0 1.04 44 1025 2 33.8 1.29 45 •1 n 0.50 30.4 0.62 46 1080 2 33.8 1.31 47 II n 0.50 31.0 0.82 48 1140 2 27.4 _ 49 II •i 0.50 30.4 0.58 50 1250 2 23.6 0.42 To determine the influence of the salt present in the calcined stone on the measured reactivity of the calcined dolomite, some samples containing zero NaCl were run in water containing 0.55 gram NaCl per 200 grams of water rather than in distilled water. Figure 6 shows that in the salt solution the measured reactivity was slightly higher than in distilled water. In comparing the reactivity of calcined dolomite with that of its component oxides, it can be seen that MgO decreases in reactivity very rapidly as the temper- ature of preparation rises and by 9 25°C it is very unreactive, while CaO is still very reactive even at the highest temperature employed. Since dolomite is approx- imately half MgO and half CaO on a molar basis, and the reactivity of the MgO is nearly zero at the temperatures used to decarbonate the dolomite, then it follows that the MgO is acting as a more or less inert diluent and the reactivity is due mainly to the CaO content. One would consequently expect the reactivity of the dolomite to be approximately half that of the CaO. The data confirm this expecta- tion. Properties of CaO The properties of CaO prepared by decomposition of CaC03 containing var- ious amounts of NaCl are given in table 2. Figure 7 shows that at the lower temperatures NaCl decreases the surface area to a marked extent. However, at the higher temperatures the influence of salt on the surface area is much less. For comparison purposes the surface areas of calcined Missouri limestone, as determined by Staley and Greenfeld (1949), are included. Above 1050° C there is some agreement; at less than 1050° C the agree- ment is not as good. PROPERTIES OF LIME HYDRATES 13 The reactivity values are shown in figures 6 and 8. In figure 8 the surface area was plotted vs. the reactivity, with- out regard to either the salt con- ey tent or calcining temperature. It ^ appears as though the surface , area of the CaO must be reduced g to less than about 1 mvgram be- < fore there is any appreciable de- £} crease in the reactivity. How- *j ever, beginning at about 1 mVg Jj the rate of decrease in reactivity becomes very rapid. The reac- tivity seems to be independent of how the surface area decrease is produced, whether by increase in temperature of preparation or by addition of NaCl. 4.0 3.0 2.0 1.0 0%NaCI x Values of Staley ft Greenfield (1949) for 2 hrs. calcination of | — l" Mo. limestone pebbles, 0%NaCI. •• Present data, for 2 hr. calcina- tion of precipitated CaCOy i%NaCI x _i_ _i_ 800 900 1000 1100 1200 1300 Maxfmum Calcinatfon Temperature °C Fig. 7. - Influence of NaCl on the surface area of CaO. Time at maximum temperature was two hours. Properties of MgO Some properties of MgO produced by decomposition of the basic magnesium carbonate are given in table 3. Figure 9 shows that the sur- face area decreases rapidly as the calcination temperature increases, and that salt decreases the surface area considerably at the lower tem- peratures, but has a smaller effect at the higher temperatures. Some data of Livey, Wanklyn, Hewitt and Murray (1957) on the surface area of MgO produced by decomposition of Mg(OH)2 containing no salt are given for comparison. Considering the different starting materials and preparation methods, the agreement is not bad, especially at the higher temperatures. Figures 6 and 10 show the reactivity of MgO. From figure 10 it appears that the reactivity falls Surface Area- m/g. Fig. 8. - Reactivity of CaO with various surface areas. 14 ILLINOIS STATE GEOLOGICAL SURVEY Table 3. ■ - Prop erties of Mg< D Maximum Total time NaCl added Reactivity Surface calcining at (wt. % based 1000 area temp. max. temp. on unburned (m /gram) No. (°c ) (hrs.) basic MgC0 3 ) (°C ) 51 500 2 14.8 104.3 52 M H 0.5 6.9 78.2 53 625 2 1.5 _ 54 it H 0.5 1.1 - 55 720 2 0.9 32.85 56 •■ II 0.5 0.7 28.08 57 925 2 0.3 8.00 58 ft •■ 0.5 0.3 6.46 59 1250 0.0 1.44 rather rapidly with a decrease in surface area down to approximately 60 mVgram, then the reactivity falls off slowly with further decrease in surface area. This re- lationship between surface area and reactivity for MgO is in contrast with that for CaO. Some Properties of Laboratory Prepared Hydrates Dolomitic Hydrates Table 4 includes, in addition to the Emley plasticity values, the amount of water necessary to produce a hydrate putty of standard consistency and the maxi- mum bed temperature (see section on hydration) attained during the atmospheric pressure hydrations. The water content at standard consistency for both high and low pressure hydrates decreases slightly as the salt content increases at a particu- lar temperature. The high-pressure hydrates required 3 to 5 percent more water to form a paste of standard consistency than the low-pressure hydrates. The bed temperature, in general, decreases with an increase in both cal- cining temperature and salt content of the raw stone. This agrees with the decrease in reactivity measured in the calorimeter as these two factors increased. From table 4 it can be seen that on soaking, the plasticity of the high- pressure hydrates changes very little. They develop nearly their maximum plastic- ity almost immediately on tempering with water. However, the low-pressure hy- drates, in most cases, show a marked increase in plasticity on soaking in water. The plasticity of the high-pressure hydrates reported in table 4 were all produced by hydrating the calcined dolomite at 140 psi (gauge) for4| hours with sufficient excess water 1 so that some liquid water was present in the vessel during this period. Before it was realized that the presence of liquid water was necessary to get a high degree of hydration of the MgO in pressure hydration, some work was carried out using essentially dry steam. In one such experiment a steam generator was set up and essentially dry steam at 60 psi (gauge) was delivered to another container holding the calcined dolomite, which was at a temperature PROPERTIES OF LIME HYDRATES 180 15 160 140 120 I — \ _ x Present data for the decomposition of basic MgC0 3 in air. (Time 2 hr.) — •Data of Livey et al. (1957) for MgO produced by decomposition of Mg(0H) 2 in air 0%NaCI). (Time I hr at 500° 8 700° C if hrs. at 1000° 8 I380°C ) 500 700 900 1100 1300 1500 Maximum Temperature of Calcination °C Fig. 9. - Surface area of MgO vs. temperature of preparation. sufficient to maintain a pressure of 60 psi (guage) . After three hours of this treatment the unsoaked plasticity was only 14 6 and the soaked value was 200. X-ray diffraction patterns showed that very little of the MgO had been hydrated. Knibbs and Gee (1952), among others, have pointed out the necessity of having liquid water present during pressure hydration in order to effect a high degree of hydration in a short time. It is evident from table 4, and the above discussion, that the plasticity is strongly dependent upon the method and procedure of hydration, as well as calcin- ing temperature, salt content, soaking, etc. Table 5 gives surface area values for some of the dolomitic hydrates. The most interesting feature of table 5 is the large increase in the surface area of the low-pressure hydrates upon soaking, and the rather small increase in the area of the high-pressure hydrates. It appears that a more or less extensive interaction of the low-pressure hydrates with water takes place, but that the interaction with the high-pressure hydrates is less extensive. The presence of salt in the raw 16 ILLINOIS STATE GEOLOGICAL SURVEY Table 4. - Emley Plasticity Values of Laboratory Prepared Dolomitic Hydrates 8 High-pressure hydration Low-pressure hydration Water at Emley plasticity , standard No. Unsoaked SoakecP consistency Emley plasticity, Unsoaked Soaked Water'at standard consistency Maximum bed temperature (°c) 242 238 198 190 190 296 309 274 232 238 41.3 39.8 37.1 38.2 37.4 90 85 75 62 55 189 217 38.1 67 193 245 37.6 62 226 132 126 335 246 229 40.3 38.5 38.5 88 79 75 8 326 320 45.3 9 388 392 42.8 10 416 455 42.3 11 434 423 43.1 12 435 455 42:2 13 - - - 14 395 424 44.0 15 - - - 16 432 430 43.4 17 - - - 18 _ _ _ 19 - - - 20 - - - 21 326 388 44.7 22 351 424 41.2 23 403 408 42.7 24 320 353 43.0 25 304 355 42.2 26 284 334 41.2 27 278 277 40.0 28 346 396 40.5 29 321 360 39.2 30 232 261 42.8 31 255 262 41.2 32 259 278 41.0 33 208 203 36.9 34 208 206 40.6 178 250 41.1 81 75 77 78 74 69 68 150 206 39.0 148 189 35.7 147 173 36.4 131 147 34.9 144 176 34.2 141 199 34.4 128 160 38.9 122 163 36.5 124 152 35.5 a For compositions, calcination temperatures, etc., see table 1, £ -b- Percent water by weight in standard consistency paste. g -c- Soaking period 24 hours. 73 68 64 Illinois State Geological Survey Circular 270 — Plate 1 .Ji> ■*s 4?_ ©* %> Sv<*" (A * *'y SS fsT^c?! ^ p ' *L* « ^'J* Electron Micrographs A — Magnesium hydroxide crystals B — Calcium hydroxide crystals Magnification approximately 20,000 diameters PROPERTIES OF LIME HYDRATES 17 Table 5. - Surface Area of Laboratory Prepared Dolomitic Hydrates (meters /gram)' High-Pressure Hydration Low-Pressure Hydration No. Unsoaked Soaked Unsoaked Soaked 8 12.37 - 14.04 9 _ 10 - - 11 12.88 - 11.09 12 - - - • 13 - - - - 14 - - 15 - - - - 16 - - - - 17 - - - 18 - - 14.46 24.47 19 11.25 27.55 20 - - 10.95 21 13.04 14.65 12.96 23.55 22 11.90 23 14.90 - - - 24 10.81 11.90 25 10.05 12.30 26 9.67 11.32 27 28 29 30 9.52 11.53 31 9.12 10.84 32 9.18 10.28 33 11.44 34 9.54 10.68 26.71 9.48 22.35 8.20 19.47 9.61 22.69 9.28 17.66 9.40 16.53 a For compositions, calcination temperatures, etc., see table 1. 18 ILLINOIS STATE GEOLOGICAL SURVEY 100 Fig. Surface Area-m/g. 10. - Reactivity of MgO vs. surface areas, stone or the increase in calcining temperature, in general, results in decreased surface area of both the high- and low-pressure hydrates. Another interesting result is that the surface areas of the unsoaked, high- and low-pressure hydrates are very nearly identical for the same temperature of preparation and the same salt content. Figures 11, 12 and 13 show the relationships between the sur- face area and the Emley plasticity for unsoaked and for soaked dolo- mitic hydrates produced by high- and low-pressure hydration. Figure 11 shows that for the high-pressure hydrates there is an increase in plasticity as the surface area of the hydrates in- creases. The same relationship holds for both soaked and unsoaked values. In figure 12, for the low-pressure hydrates, there is considerable scatter, but there appears to be a relationship between surface area and plasticity, and the plasticity increases with the surface area. It will be noticed that the soaked specimens have higher surface areas than the unsoaked. In figure 13 the plasticity-surface area relationship for high- and low- pressure hydrates in the unsoaked condition is compared directly. It appears that the rate of increase of plasticity with increasing surface area is greater for high- pressure hydrates than for low-pressure hydrates. Even though both the high- and low-pressure hydrates have nearly the same surfaces, the high-pressure hydrates are usually more plastic. Bishop (19 39) calculated the surface area from particle size distribution curves for a number of commercial hydrates and compared them with the Emley plasticity values. He stated that no relationship existed between plasticity and surface area or particle size distribution of hydrates. This is at variance with the present data shown in figures 11, 12, and 13, which strongly suggest that a rela- tionship does exist between surface area and plasticity. Bishop further stated that the lack of correlation may have been due to the fact that he could not deter- mine the size distribution of particles of less than 2 microns and that these parti- cles may have a large influence on plasticity. He showed that the plasticity of fractions finer than 2 microns from highly plastic hydrates was very great but sim- ilar fractions from low plasticity hydrates were not very plastic. The method of surface area determination used here measured the surface area of the entire sample and not just a fraction of it, as was the case in Bishop's method. Also, an examination of figures 11, 12, and 13 shows that the relationship between surface area and plasticity depends on the method of hydration and the subsequent treatment of the hydrates. The relationship would probably break down if one tried to relate various hydrates with no knowledge of their preparation and pretreatment. This is what Bishop did and may be a part of the reason for his find- ings. PROPERTIES OF LIME HYDRATES 19 Tables 6 and 7 give the compositions of the hydrates as determined from the ignition loss and also some CO2 contents. 400 During soaking the hydrates pick up some CO2, but the total amount in the samples is less than 1 percent. For the samples on which the CO2 content was determined, the compositions calculated from the ignition losses were corrected for the CO2 picked up. ^ The correction was very small. This fact gives us .2 confidence that the compositions calculated from the g 300 ignition loss data are roughly correct. ^ Table 6 indicates that the high-pressure hy- drates are all very nearly completely hydra ted regard- less of the calcining temperature or salt content of the £ raw stone. Furthermore, on soaking there is no sig- nificant increase in Mg(OH)2 content. Table 7 shows that the low-pressure hydrates are far from being completely hydrated. However, lower calcining temperatures and salt contents seem to favor a slightly greater degree of hydration. In most cases soaking of these hydrates appears to in- crease the degree of hydration, especially at lower salt contents. The proportion of Ca(OH)2 in the low- pressure hydrates appears to be larger than in the high- iqo pressure hydrates. From compositions given in tables 6 and 7, the percentage of the total possible Mg(OH)2 was cal- p^g culated. This was then plotted vs. the unsoaked 200 o Soaked Values • Unsoaked Values _L 6 8 10 12 14 16 Surface Area-m/g. 11. - Surface area of high- pressure dolomitic hydrates vs. Emley plasticity. Surface Area- m/g. Fig. 12. - Surface area of low-pressure dolomitic hydrates vs. Emley plasticity. Emley plasticity values, as shown in figure 14. It will be noted that the plasticity increases with increases in the amount of Mg(OH)2. As the calcining tem- perature rises, the plasticity de- creases, even when very nearly all the possible Mg(OH) 2 is pres- ent. This indicates that in- creased amounts of Mg(OH)2 produce better plasticity, and that any way of producing Mg(OH)2 from the MgO in the calcined Thornton, Illinois, dolomite will improve the plasticity. All indications are that the presence of MgCOHK improves the plasticity. In the hope of throw- ing more light on the reasons for 20 ILLINOIS STATE GEOLOGICAL SURVEY Table 6. - Composition of Laboratory Prepared High-Pressure Hydrated Dolomitic Limes (Calculated from Ignition Loss) Ignition Loss Composi"t ;ion (S C0 o Content 2 (%) Un soaked Soaked No. Unsoaked Soaked Ca(0H) 2 Mg(0H) 2 MgO Ca(OH) 2 Mg(0H) 2 MgO Unsoaked Soaked 8 9 25.7 - 57.0 38.2 4.8 - - - - 10 11 12 13 14 15 16 17 18 19 20 21 27.0 - 56.0 43.0 1.0 _ - - i i i i i i i i i i i i i i i 26.5 26.6 56.4 41.2 2.4 56.3 41.5 2.2 - 22 25.4 - 57.3 36.7 6.0 - - - - 23 27.1 - 56.0 43.4 0.6 - - - - 24 26.0 26.4 56.8 39.2 4.0 56.6 40.0 3.4 0.218 0.772 25 25.7 25.9 57.0 38.1 4.9 56.8 38.8 4.4 - 26 26.1 26.2 56.7 39.6 3.7 56.6 40.0 3.4 - 27 - - - - - - - - - 28 - - - - - - - - - 29 - - - - - - - - - 30 25.7 26.2 57.0 37.8 5.2 56.8 39.3 3.9 0.216 0.761 31 25.7 26.1 57.0 37.8 5.2 56.7 39.6 3.7 - 32 26.0 26.1 56.8 39.2 4.0 56.7 39.7 3.6 — ™ a For compositions, calcination temperatures, etc., see table 1, b See "Experimental Procedure." PROPERTIES OF LIME HYDRATES 21 Table 7.- Composition of Laboratory Prepared Low-Pressure Hydrated Dolomitic Limes (Calculated from Ignition Loss) b Ignition Loss Compos it ion (%) C0_ Con 2 (%) tent Un soaked Soaked No. Unsoaked Soaked Ca(0H) 2 Mg(0H) 2 MgO Ca(0H) 2 Mg(0H) 2 MgO Unsoaked Soaked 8 9 16.2 - 64.1 1.9 34.0 - - - - - 10 11 12 13 14 15 16 17 18 19 20 17.8 - 62.9 8.1 29.0 - - - - - 16.4 16.7 16.9 19.2 19.5 64.0 63.7 63.6 2.6 3.9 4.5 33.4 32.4 31.9 61.9 61.6 13.6 14.6 24.5 23.8 - - 21 22 23 24 25 26 27 28 16.3 19.1 64.0 2.4 33.6 61.9 13.0 25.1 - - 16.2 15.8 15.9 19.4 16.7 16.3 64.2 64.4 64.3 1.5 0.4 0.9 34.3 35.2 34.8 61.8 63.8 64.0 13.4 3.4 2.3 24.8 32.8 33.7 0.262 0.765 0.712 29 30 31 32 15.5 15.8 15.6 19.0 18.1 16.9 64.6 64.6 64.6 35.4 35.4 35.4 62.4 62.8 63.7 11.8 8.4 4.4 25.8 28.8 31.9 0.304 0.331 0.811 0.861 a For compositions, calcination temperatures, etc. see table 1. b See "Experimental Procedure." 22 ILLINOIS STATE GEOLOGICAL SURVEY this, particle size distributions were determined on three samples containing different amounts of Mg(OH)2 produced by different methods of hydration and soaking. The three samples investigated were all de- rived from the calcinate from run No. 24 by the fol- lowing treatments: A) Low-pressure hydrated, unsoaked. B) Low-pressure hydrated, soaked. C) High-pressure hydrated, unsoaked. Figure 15 shows the particle size distribu- tion of these three samples. If we consider the re- gion of less than one micron, it appears that C has the greatest amount of fine material, followed next by B, and that A has the least amount. On compar- ing their plasticity values, C has the highest plas- ticity (320), followed by B (206), and then A (150). It appears that the samples with the greatest amount of fine material had the highest plasticities. 400 g 300 200 400 T3 a> o o with B intermediate. Inasmuch as larger amounts of fine material are associated with higher Mg(OH) 2 contents, then the Mg(OH) 2 should tend to be a fine material. To check this, x-ray dif- fraction patterns of various size fractions were made. The results are summarized in table 8, from which it appears that the proportion of Mg(OH)2 steadily increases in the finest fractions of B and C. PROPERTIES OF LIME HYDRATES 23 100. 80. 60. 40. 20. -, 10. in % «■ 1 •■ a> a> E o b o s. 0.8 0.6 0.4 0.2 0.1 4 >K A— Low-Pressure Hydrate (unsoaked) B - Low-Pressure Hydrate (soaked) C - High-Pressure Hydrate (unsoaked) 20 40 60 80 100 Percent Finer Than Fig. 15. - Particle size distribution of laboratory-prepared dolomitic hydrate No. 24. 24 ILLINOIS STATE GEOLOGICAL SURVEY Table 8. - X-Ray Identification and Intensity of Certain Lines for Laboratory-Prepared Dolomitic Hydrates Intensity of X-ray Lines Size (cm above bac kground) Ratio fraction (D* (2) (3) (4) Int. Mg(0H) o (18.5°) Int. Ca(0H) 2 (18.0°) Hydrate (microns) Ca(0H) 2 Mg(0H) 2 Mg(GH) 2 MgO No. 24 Low-pressure 4 4.6 7.5 hydration 2.2 4.7 9.3 (unsoaked) 1.2 5.1 11.0 (Sample A) 0.86 4.6 10.7 0.70 4.6 9.7 0.62 4.6 8.5 No. 24 Low-pressure 20 7.5 2.1 1.6 7.0 0.28 hydration 4 7.5 2.0 1.6 8.3 0.27 (soaked) 2.2 7.6 2.6 1.6 7.3 0.34 (Sample B) 1.2 6.7 2.4 1.9 8.6 0.36 0.86 6.6 2.9 2.1 9.0 0.44 0.70 5.7 3.0 2.9 10.3 0.53 0.62 5.3 3.1 2.1 8.6 0.58 0.46 5.2 2.7 1.6 7.4 0.52 No. 24 High-pressure 60 8.4 4.7 4.5 1.0 0.56 hydration 20 8.7 5.6 4.7 0.9 0.64 (unsoaked) 4 9.3 5.8 5.9 0.6 0.62 (Sample C) 2.2 9.8 7.2 7.0 1.2 0.74 1.2 9.6 9.3 8.3 1.0 0.95 0.86 9.6 10.9 10.3 1.3 1.14 0.70 7.3 9.8 10.1 0.8 1.34 0.62 5.2 9.3 8.8 0.5 1.79 0.46 4.5 8.5 8.7 0.3 1.89 *(1) Ca(0H) 2 line at 18.0° in 2 9 (2) Mg(0H) 2 line at 18.5° in 2 9 (3) Mg(0H) 2 line at 37.8° in 2 9 (4) MgO line at 42.8° in 2 9 From this discussion It seems likely that the reason for the increase in plasticity as the Mg(OH)2 increases is that the Mg(OH)2 forms as very fine particles. Bishop (19 39) observed that the fractions less than 2 microns from highly plastic limes were very plastic, but that the same fractions from low plasticity limes were not plastic. This can now be explained from the above data; that is, the high plasticity limes have greater amounts of very fine material which is richer in Mg(OH)2 than the nonplastic hydrates. Mg(OH) 2 and Ca(OH) 2 Some of the MgO samples previously discussed (table 3) were pressure hy- drated by the usual procedure (140 psi for 4^ hours), then the surface area and plasticity was measured. The results are given in table 9. PROPERTIES OF LIME HYDRATES 25 Table 9. - Some Properties of Mg(OH) 9 a Water at Surface area Eml ey standard 2 plastici ty consistency (X) (m /gram) No. Unsoaked Soaked Unsoaked Soaked 55 645 625 63.8 36.65 - 57 468 480 50.0 12.85 17.52 58 400 410 50.2 - - 59 468 468 43.5 7.78 - a For compositions, calcination temperatures, etc., see table 3. Table 10. - Some Properties of Ca(0H) 2 a Water at Surface area Emley standard , 2 , plastici ty consistency (%) (m /qramj No. Unsoaked Soaked Unsoaked Soaked 42 101 93 54.5 7.37 7.84 44 172 102 47.9 7.04 - 45 173 71 46.2 - - 50 202 246 44.0 11.84 - a For compositions, calcination temperatures, etc., see table 2. The plasticity, surface area, and water content at standard consistency for No. 55, prepared from MgO formed at 720° C, are very high. As the temperature of preparation of the MgO rises to 925°C, the plasticity, surface area, and water necessary to form a paste of standard consistency decrease. The presence of salt seems to decrease the plasticity somewhat. The MgO formed at 1250° C produced a hydrate of approximately the same plasticity as that produced from MgO prepared at 9 25°C, both without salt. Some of the CaO samples given in table 2 were pressure -hydra ted by the usual procedure, then the surface area and plasticity were measured. The results are given in table 10. The unsoaked and soaked plasticities are all rather low. The increase in both the soaked and unsoaked plasticities as the temperature of preparation of the CaO increases is unexpected, and no explanation is apparent at present. The particle size distribution of pressure-hydrated Ca(OH) 2 and Mg(OH) 2 , prepared from their respective oxides, which were formed at 9 25°C, are given in figure 16. It will be noticed that the Mg(OH) 2 has a much larger proportion of fine material (less then 2\±) than the Ca(OH) 2 . If we compare the plasticities (tables 9 and 10), it is apparent that Mg(OH) 2 is much more plastic than Ca(OH)2. 26 ILLINOIS STATE GEOLOGICAL SURVEY 100.0 80.0 - 60.0 - 40.0 - 20.0 ~ 10.0 o 8.0 4.0 O E UJ 200 100 • Dolomitic Hydrates o Mg(0H) 2 * Ca(0H) 2 0.4 2.0 2.4 OS 1.2 1.6 R (ml. /min.) Fig. 17. - Water removal factor Rvs. Emley plasticity. Electron micrographs of the same samples that were used for the particle size distribution are shown in plate 1. It is evident once again that the Mg(OH)2 contains more fine crystals than the Ca(OH)2. In both cases, however, many of the particles are rather thin hex- agonal platelets. The water removal factor, R (see fig. 3 and section on water retention of hydrates), in relation to the plasticity is shown in figure 17. It will be noted that the fine particle sized samples (as Mg(OH)9) show a low rate of water removal and high plasticities, but that the larger particle sized Ca(OH)2 shows a lower plastic- ity and relatively high water removal rates. The dolomitic hydrates lie between these two extremes and their position probably depends mainly on the proportion of Mg(OH)2 present in the hydrate. 28 ILLINOIS STATE GEOLOGICAL SURVEY DISCUSSION It has been demonstrated by experimental data that the presence of NaCl decreases the reactivity and surface area of calcined dolomite, CaO and MgO. Noda (19 38, 19 39) has shown that the presence of NaCl in limestone or magnesite caused increased grain growth and decreased reactivity of the CaO and MgO formed on calcination of the carbonates. Atlas (1957) has shown that the lithium halides also catalyze the sintering or grain growth of MgO. Staley and Greenfeld (1947) observed that limes prepared from limestones to which 0.5 percent NaCl had been added had low surface areas and that the sodium chloride apparently acted as a fluxing material, increasing the mobility of the calcium and oxygen ions sufficiently to permit them to pack more rapidly than in its absence. Noda and Kan (19 37) also found that the addition of NaCl acceler- ated crystal growth of CaO. Tacvorian (1952, 1954) has suggested a mechanism by which the surfaces of refractory oxide grains may be activated and thus cause an acceleration of the sintering process. He believed that if the refractory oxide is mixed and heated with a compound which is a weakened model, a relatively concentrated solid solu- tion may be formed in the surface of the refractory grain. The surface solid solu- tion will have lower activation energies and therefore higher ionic diffusion rates than the pure refractory oxide. This increase in the ionic diffusion rate causes crystal growth to occur at a greater rate, since at the point of contact of two small crystals the atoms of the two crystals interdiffuse more readily and the net result is the production of a larger crystal. The application of Tacvorian' s ideas to the systems NaCl-CaO and NaCl- MgO would require NaCl to be a weakened model of CaO and MgO. Especially for the MgO, due to the large difference in size of the Mg ++ and Na + ions, it is not easy to think of NaCl as such a model. However, Atlas (1957) showed that al- though LiCl and LiBr were not truly weakened models of MgO (due to the discrep- ancy between the radii of CI" and Br~ and O ions), they were nevertheless as effective in catalyzing the sintering of MgO as was the true weakened model, LiF. To him this suggested the importance of the cation, since in the three halides the cation was the same. He suggests that the Li + ion from the chloride and bromide can be incorporated into the surface layer of MgO and that this solution may be aided by partial thermal decomposition of the halide to Li20. Since Li20 does not have the same cation to anion ratio as MgO, solution of Li20 in MgO must be ac- companied by the formation of defects, as either O vacant sites or interstitial Li + ions. Such defects will increase the rate of diffusion or material transport and thereby the sintering or grain growth rate. If one applies these ideas, then in the case of CaO, the Ca" 1 " 4 " and Na + ions are nearly the same size, and if some of the NaCl is partially thermally decomposed, then the Na20 could go into solution in the surface of the CaO crystals and cause defects, which will increase the rate of diffusion and promote sintering of the CaO. However, in the case of MgO there is such a large difference in the radii of Mg ++ and Na + ions that if the above conceptions of the mechanism of promotion of sinter- ing are valid, then Na + should not be an effective aid in the promotion of sintering of MgO. If one uses the surface area as a measure of grain growth, then from figures 7 and 9 it can be seen that NaCl causes a greater decrease in the surface area of CaO than in the surface area of MgO when compared on a percentage basis. So in dolomite the influence of NaCl on grain growth should be the greatest for the CaO. PROPERTIES OF LIME HYDRATES 29 400 Fig. 18. - Emley plasticity of low-pressure hydrates made from calcined dolomite from Thornton, Illinois, vs. reac- tivity (see " Experimental Procedures") of the calci- nates. o o in 300 >» o 0. 200 E UJ 100 4% No CI nt Noci^// ^>"/ _i L J i- 8 12 16 Reactivity, AT |000 °C 20 300 - o o ~ 2 00 in a a> E UJ 100 - ^ ^ S y /-. . ■■/ ./■ A 875° cn A 925° [Calcining Time • 1025° 2 hrs O 1140° J • x 825° — 7 hrs _i _ i i . Fig. 19. - Emley plasticity of low-pressure hydrated calcined dolomite from Thornton, Illinois, vs. the surface area of the calcinate. l.O 2.0 3.0 4.0 5.0 Surface Area of Calcinate - m 2 /g 30 ILLINOIS STATE GEOLOGICAL SURVEY 200 1.0 2.0 3.0 4.0 Surface Area of Calcinate- m 2 /g. The over-all effect of NaCl on the oxides appears to be the same as the effect of thermal treatment. That is, it promotes sintering. A small amount of NaCl will affect sintering to about the same degree as a small increase in tem- perature . Next one must consider what in- fluence the NaCl enhanced growth of the oxide crystallites has on the properties of the hydrates, particularly the plastic- ity. Because the plasticity depends strongly on the method of hydration, the results given are relative and may not apply to methods other than those used here. To this end consider the data in figures 18, 19, and 20, which show cor- relations between the oxide properties and plasticity of the hydrates. In the case of low-pressure hy- dration, figure 18 shows that at any given temperature the plasticity de- creases as the salt content rises. So the decrease in the reactivity of the oxide, Fig. produced by NaCl-enhanced sintering, results in a less plastic hydrate when low-pressure hydration is used. Figure 19 shows that as the surface area of the oxides, derived from cal- cining dolomite, decreases (no distinction being made between the effect of salt and temperature of calcination), the plasticity of the low-pressure hydrates also decreases. Figure 20 shows that for the high-pressure hydrates the plasticity of the hydrate is increased by the addition of salt to the raw stone at all surface areas, especially at the higher surface areas. In the low-pressure hydrates, the presence of salt during calcination does not increase the plasticity. Salt appears to help produce a more plastic lime in high-pressure hydrates, but even the least plastic pressure-hydrated lime is of such a high plasticity that increasing it further by adding salt seems to be of ques- tionable value. The results of Briscoe and Mathers (1927) seem to be somewhat at odds with the present results. This may be due to different stones, different hydration methods, different calcining procedures, etc., none of which they described. They stated that pressure hydration produces low plasticity hydrates and that the addition of salt to their dolomite produced a highly plastic hydrate by low-pressure hydration. In their conclusions they stated that "the plastic properties of hydrated limes seem to be determined by the extent to which hydration progresses during the soaking of the lime to form putty. If the lime is completely hydrated at the time of formation of the dry product, it will be a non-plastic lime. The quicklime must be produced so that during hydration the CaO hydrates slowly, but yet the MgO is reactive and on soaking the MgO will hydrate extensively and produce a highly 20. - Emley plasticity of high-pressure hydrated calcined dolomite from Thornton, Illinois, vs. the surface area of the calcinate. PROPERTIES OF LIME HYDRATES 31 plastic lime. " They believed that salt decreases the activity of the CaO and not that of the MgO so that on low-pressure hydration the CaO hydrates slowly, leav- ing the MgO unhydrated but yet reactive, and on soaking the hydrate a high plas- ticity results due to the hydration of the MgO. The present information seems to confirm Briscoe and Mathers' (19 27) con- clusion that the influence of salt on the sintering of CaO is greater than it is on MgO. There seems to be agreement that the plasticity is greatly increased in dol- omitic hydrates if most of the MgO is converted to Mg(OH) 2 . However, the pres- ent data indicate that the method of conversion has little influence. Webb and Sampson (1957) and a number of patents by Corson (Corson, 1946) show the great increase in plasticity produced by pressure hydration in which a high proportion of the MgO is converted to Mg(OH)2» This brings us to a consideration of plasticity itself. As already shown, the presence of increased amounts of MgfOHjo in dolomitic hydrates increases the plasticity, whether the Mg(OH)2 is formed by pressure hydration of the oxides or by soaking a low-pressure hydrate. It was further demonstrated that Mg(OH)2 is very finely divided compared to Ca(OH)2 and that it has, in the pure state, a much higher plasticity than Ca(OH)2. Electron micrographs indicate that both pure Mg(OH)2 and Ca(OH)2, pressure-hydrated, are thin hexagonal platelets. Also the Mg(OH) 2 tends to retain its water with greater tenacity than Ca(OH) 2 or a mixture of Ca(OH)2 and MgO such as exists in partly hydrated dolomitic limes. The greater ability of the lime putty to retain its tempering water is impor- tant, since in the Emley plasticimeter, water is being removed by the porous base plate and the more slowly this water is removed the greater will be the plasticity figure . Since Ca(OH)2 and Mg(OH)2 have the same particle shape (pi. 1), one is led to believe that the large difference in their plasticity is, to a considerable ex- tent, due to the smaller crystal size of Mg(OH)2 as compared to Ca(OH)2- The rate of withdrawal of water from the putty on the Emley apparatus is a capillary phenomenon in which the capillaries in the putty tend to resist the with- drawal. Small particles make for small capillaries and therefore greater resistance to water removal. It appears that any situation which will promote the conversion of the free MgO in calcined dolomites to Mg(OH)2 will increase the plasticity of the lime pro- duced. This is believed to be due to the fine particle size of the Mg(OH)2 which tends to increase the ability of the putty to retain water. CONCLUSIONS 1) The addition of salt to dolomite before calcining in the manner described, and then hydrating this calcined dolomite at low pressures by the method described, does not increase the plasticity of the lime produced. NaCl appears to increase slightly the plasticity of the pressure-hydrated dolomitic limes. 2) Mg(OH) 2 is considerably finer than Ca(OH)2 produced under the same conditions. Both show thin, hexagonal platelets at high magnification. The Mg(OH)2 retains water much better than Ca(OH)2 and shows a higher Emley plastic- ity. 3) Any method used in this investigation (such as soaking at atmospheric pressure or pressure hydration of the calcined dolomite) that increased the amount of Mg(OH)2 also resulted in increased plasticity of the lime. 32 ILLINOIS STATE GEOLOGICAL SURVEY 4) The results of this work indicate that, in general, the method of manu- facture of the hydrate is more important than the presence of NaCl, as far as the production of high plasticity is concerned. REFERENCES A.S.T.M. (1949), Standard methods of physical testing of quicklime and hy- drated lime: in A.S.T.M. Standards, Pt. 3, p. 254-262. Atlas, L. M., 1957, Effect of some lithium compounds on sintering of MgO: Jour. Am. Cer. Soc, v. 40, p. 196. Bishop, D. L., 19 39, Particle size and plasticity of lime: lour. Res. National Bur. of Standards, v. 23, p. 285. Bradley, D. E., 1954, Evaporated carbon replica technique for use with the elec- tron microscope: Brit. Jour. Applied Physics, v. 5, p. 96. Briscoe, H. T. , and Mathers, F. C. , 1927, Plasticity of finishing limes: Ind. Eng. Chem., v. 19, p. 88. Brunauer, S., Emmett, P. H. , and Teller, E., 1938, Adsorption of gases in multi- molecular layers: Jour. Am. Chem. Soc, v. 60, p. 309. Corson, B. L., 1946, Methods of conditioning and treating lime and product thereof: U. S. Patent 2,409,546, Oct. 15, 1946. Hillebrand, W. F., and Lundell, G. E. F. , 1953, Applied inorganic analysis: (2nded.), John Wiley and Sons, New York, New York, p. 768. Knibbs, N. V. S., 1937, Some chemical reactions and compounds of lime: Cement and Lime Manuf . , July, 1937, p. 195. Knibbs, N. V. S. , and Gee, B. J., 1952, Lime and limestone, Pt. I: H. L. Hall Corp. Ltd., Toronto, Canada, 113 p. Lamar, J. E., 1957, Chemical analyses of Illinois limestones and dolomites: Illinois Geol. Survey, Rept. Inv. 200, p. 9. Lamar, J. E., and Shrode, R. S., 1953, Water soluble salts in limestones and dolomites: Econ. Geol., v. 48, p. 100. Livey, D. T., Wanklyn, B. M., Hewitt, M. , and Murray, P., 1957, The proper- ties of MgO prepared by the decomposition of Mg(OH) : Trans. Brit. Cer. Soc, v. 56, p. 217. Murray, J. A., Fischer, H. C, and Sabean, D. W. , 1950, Effect of time and tem- perature of burning on the properties of quicklime prepared from calcite: Proc Am. Soc. for Testing Materials, v. 50, p. 1263. Noda, T., and Kan, H. , 1937, Effects of the addition of common salt during the calcination of lime, Pt. V - Rate of hydration, the microscopic and X-Ray examination of pure calcium oxide calcined under various conditions: Jour. Soc Chem. Japan, v. 40, Suppl. Binding, p. 196B. PROPERTIES OF LIME HYDRATES 33 Noda, T., 1939, On the calcination of lime: Jour. Soc. Chem. Ind. (Japan), v. 42, p. 265. Noda, T., and Oka, M. , 1938, Effects of the addition of salts on the crystal growth of periclase: Jour. Soc. Chem. Ind. (Japan), v. 41, Suppl. Bind- ing, p. 74B. Pelaez, R. V., and Murray, J. A., 1956, Report on water retention and plasticity of lime pastes: Report to the National Lime Assoc, June 1, 1956, 3 p. Staley, H. R. , and Greenfeld, S. H., 1947, Surface areas of high-calcium quick- limes and hydrates: A. S.T.M. Proc, v. 47, p. 958. Staley, H. R. , and Greenfeld, S. H., 1949, Surface areas of quicklimes: Ind. Eng. Chem., v. 41, p. 520. Tacvorian, S., 1952, Acceleration of sintering in a single phase; consideration of mechanism of minor additions: Compt. rend., v. 234, p. 2363. Tacvorian, S., 1954, Sintering by surface activation: Bull. Soc. Franc. Ceram., v. 23, p. 3-8. Webb, T. L., and Sampson, V., 1957, Pressure hydration of dolomitic lime: Pit and Quarry, p. 106, Oct.; p. 136, Nov. Wells, L. S., and Taylor, K. , 1937, Hydration of magnesia in dolomitic hydrated limes and putties: Jour. Res. National Bur. of Standards, v. 19, p. 215. Illinois State Geological Survey Circular 270 33 p., lpl., 20 figs., 10 tables, 1959 CIRCULAR 270 ILLINOIS STATE GEOLOGICAL SURVEY URBAN A :•;-; ...