Você está na página 1de 76

MIRDHA

GROUP

Baldev Plaza (Orbit Mall), Ajmer Road, Jaipur - 302 006, Rajasthan, INDIA Tel: +91-141-222 20 33 ~ Fax: +91-141-222 22 33 E-mail: jaipur@mirdhagroup.com

Group Companies Marwar Chemicals Pvt. Ltd. MCPL, India FACTORY National Highway No. 65, Village - Khimsar, District - Nagaur, Rajasthan, INDIA Tel: +91-1585-204 331 E-mail: mcpl@mirdhagroup.com

Calcined Lime
Introduction Lime and lime-based products are used in a wide variety of traditional and emerging applications because of their chemical and physical properties. Calcined Lime or Quicklime (or Calcium Oxide or Burnt Lime) is obtained by calcining (controlled heating - time and temperature) limestone at temperatures above 900C. This highly reactive product is essential in many industrial processes. We are one of the leading manufacturers & suppliers of highgrade calcined lime (burnt lime / quicklime) which is manufactured from 'gas fired vertical shaft lime kiln' to enable higher and sustained quality of calcined lime for specific uses. This new technology of calcined lime produces a high-grade product, which is devoid of any external

Lime Kiln

impurities (such as carbon, silica, sulphur, etc.) added in case of mix-fuel domestic kilns, where limestone is burnt with coal. The project is set up near village Khimsar, District Nagaur, Rajasthan on NH 65. Chemical formula: CaO Process The production of high calcium quicklime (calcium oxide) requires a large amount of heat, which is generated in the kiln environment. The quarried and sized high calcium limestone travels through a vertical shaft lime kiln and is subjected to these high temperatures where the calcium carbonate begins to dissociate with the resultant formation of calcium oxide. The minimum temperature for the dissociation of calcium carbonate is 1648F (898C). For practical production purposes, however, the kiln temperature range is from an initial temperatures of about 1750F (954C) to a final temperature of about 1950F (1066C). These temperatures can vary dependent upon the nature of the limestone being calcined. "High Calcium" Limestone Calcination: CaCO3 + Heat ---> CaO + CO2 1750F to 1950F 954C to 1066C Specifications CaO :: SiO2 :: MgO :: LOI :: 95% 0.50% 1.00% 2.00% Sizes Powder 250 Mesh 10 to 30 MM 30 to 60 MM 40 to 80 MM

Preparation method of calcium oxide expansion agents (14-Jul-2010)

Front Page

Options

More Like This View Original

PRC English Translation (Source: SIPO_EN) Identifying No. 101774777 A dated 14-Jul-2010 Application No. CN 201010101520 filed on 26-Jan-2010 Inventors Guo Fei [+7] Applicants Jiangsu Bote New Materials Co., Ltd. [+2] Abstract The invention relates to a preparation method of calcium oxide expansion agents. The method comprises the following steps: mixing and grinding lime stone and composite mineralizer to obtain raw material, calcining at 1100-1300 DEG C to obtain calcium oxide expansion agent chamotte, and mixing and grinding the chamotte with disperse carrier to obtain the calcium oxide expansion agents, wherein composite mineralizer is composed of 50-70wt% of gypsum and 30-50wt% of copper slag, the weight ratio of composite mineralizer to lime stone is 5:95-10:90, and the weight ratio of calcium oxide expansion agent chamotte to disperse carrier is 75:25-90:10. The calcining heat of the calcium oxide expansion agents of the invention is relatively lower, calcium oxide is not easy to overheat; the water requirement of hydrated product is small, the expansion efficiency is high, the expansion rate is high; and cement concrete can realize effective expansion in dry environment. Classifications International: C04B 22/06

Method of producing expansive and high strength cementitious pastes, mortars and concretes
Patent 4046583 Issued on September 6, 1977. Estimated Expiration Date: 6, 1994. Abstract Claims Description Full Text
Ads by Google
Calcium Oxide Suitable for all Edible & Industrial applications. www.dehramineral.com

September

Patent References
2751308 2880101 3676541

Inventor

Collepardi, Mario

Assignee

Emesa Aktiengesellshaft

Application
No. 05/684727 filed on 05/10/1976

US Classes:
106/725, 106/726, 106/727

Examiners
Primary: Poer, J.

Attorney, Agent or Firm

Weaver; D. Paul

Foreign Application Priority Data


1975-05-09 IT
Ads by Google
L-Camphorsulfonic acid L-Camphorsulfonic acid ,Highquality suppliers from China. en.jlhggroup.com/ CNC Right Angle Heads Woodworking Aggregate Heads A new standard in reliability www.hsdusa.com Rambus: 1990 - 2011 Learn more about Rambus' award-winning technology www.rambus.com/inventor

Abstract
Ads by Google
Elementar www.elementar.de German High Tech Manufacturer of Elemental & TOC Analyzers

Expansive and high strength cementitious pastes, mortars and concretes are produced by adding to a mix of water, cement and aggregates a mixture containing calcium oxide and/or magnesium oxide, a water-reducing agent and an additive for reducing or eliminating bleeding of the fresh mix.

Claims
What is claimed is: 1. A method for producing expansive and high strength cementitious pastes, mortars, and concretes, comprising adding to a mix of water, cement and aggregates a mixturecontaining: a compound selected from the group consisting of calcium oxide, magensium oxide or mixture thereof, and a water-reducing agent capable of reducing the water/cement ratio of the mix at least about 15% at constant workability. 2. The method in accordance with claim 1 wherein the said oxide consists of commercial lime. 3. The method in accordance with claim 1 wherein the water-reducing agent comprises a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series, with formaldehyde. 4. The method in accordance with claim 1 wherein the water-reducing agent comprises a polymer obtained by polycondensation of melamine with formaldehyde and containing

free or salified sulphonic groups in the heterocyclic rings. 5. The method as defined in claim 3 wherein the water-reducing agent consists of the polymer obtained by polycondensation of free or salified -naphthalenesulphonic acid with formaldehyde. 6. The method as defined in claim 3 wherein the water-reducing agent consists of a polymer obtained by polycondensation of free or salified phenolsulphonic acid, with formaldehyde. 7. The method in accordance with claim 3 wherein the water-reducing agent consists of a polymer obtained by polycondensation of a free or salified benzene sulphonic acid, with formaldehyde. 8. The method in accordance with claim 3 wherein the water-reducing agent consists of a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series with formaldehyde, and containing methyl groups. 9. The method as defined in claim 8 wherein the water-reducing agent consists of a polymer obtained by polycondensation of a sodium salt of a sulphonic acid of the aromatic series, with formaldehyde. 10. The method in accordance with claim 1 wherein the percentage of the mixture is 0.540% by weight of said cement. 11. The method in accordance with claim 1 comprising adding to the mixture an additive for reducing the bleeding of the resulting mix. 12. The method in accordance with claim 11 wherein said additive is silica (SiO2) having a specific surface area of not less than 2 104 cm2 /gr determined by isothermal adsorption of a gas. 13. The method in accordance with claim 12 wherein the mixture has the following composition by weight: calcium oxide 45-65%; silica 30-50%; and water-reducing agent 1-12%. 14. The method in accordance with claim 12 wherein the percentage of the mixture is 0.540% by weight of cement.

Description
This invention relates to a method of producing expansive and highstrength cementitious pastes, mortars and concretes. As known in the art, cement conglomerates (concretes, mortars or grouts) notably undergo

contraction during hydration of the cement due to the evaporation of the water and/or the sedimentation of the inerts. In certain applications thisphenomenon must be avoided as, for example, in sealing work in order to avoid the formation of cracks and to ensure maximum adherence of the mix to the confining perimeter. In the anchoring of machines to foundations this must be avoided in order toguarantee maximum support capacity and uniform static load distribution. To attain this object it is known to add to the cement mix an expansive component capable of producing an increase in volume approximately equal to or greater than the contraction in volume which occurs during shrinkage. In this manner it ispossible to prepare shrinkage-compensating or self-stressing concretes or mortars. Known expansive component normally used for this purpose including the following: CALCIUM ALUMINATES OR SULPHOALUMINATES, CAPABLE OF PRODUCING AN EXPANDING COMPONENT, NAMELY ETTRINGITE, OF COMPOSITION 3CaO.Al2 O3.3CaSO.sub.4.32H.sub.2 O, by means of a hydration reaction, such as: the compound obtained, namely ettringite, occupies a greater volume than the reacting solid. METALLIC IRON, WHICH DURING HARDENING OF THE MIX IS TRANSFORMED INTO FERRIC OXIDE AND CONSEQUENTLY CAUSES AN INCREASE IN VOLUME OF THE ENTIRE MIX. Other expansive agents theoretically usable are calcium oxide (CaO) and magnesium oxide (MgO), as both form the corresponding hydroxides by reaction with water, leading to increase in volume. Calcium oxide in particular would have the advantageof being particularly economical as it is produced on a large scale for other uses. In order for an expansive component to be advantageously used in a concrete or mortar, the expansion must take place within a certain time interval and in a sufficiently opposing environment. These circumstances depend upon the degree ofhardening of the hydraulic binder, for example Portland cement, with which the expansive component is mixed. If expansion takes place too early with respect to the concrete hardening process, it is not able to induce any stress within the structure, asthis latter becomes too deformed or cracked. On the other hand if expansion takes place too late with respect to the concrete hardening process, i.e. when this has attained high strength, the results are equally ineffective (P. Mehta and M. Polivka:"Expansive Cements" Acts of the VI International Congress on Cement Chemistry, pg. 15, Moscow, September 1974). In addition, most of the expansion should terminate within a relatively short period since in practice the moist curing periods -- necessaryto the expansion process -- are usually short for practical reasons (M. Polivka "Factors influencing expansion of expansive cement concretes", American Concrete Institute, Special Publication 38, pg. 241). Up to the present time neither calcium oxide nor magnesium oxide have been applied on a

large scale because of the difficulty of controlling the oxide hydration and hence the expansion process in relation to the hardening of the hydraulic binder. In U.S. Pat. No. 3,649,317, for example, it is proposed that this problem could be solved by delaying the expansion of calcium oxide by a pretreatment of the expanding agent with "Vinsol" (dodecylbenzenz sulphonate). However, by this treatmentthe above-mentioned drawbacks concerning a too long curing are not avoided. The object of the invention is to provide a method of producing expansive and unusually high strength cementitious pastes, mortars and concretes by using calcium and/or magnesium oxide as the expansive agent. According to the invention this object is attained by adding to a mix of water, cement and aggregates a mixture containing: calcium oxide and/or magnesium oxide, and a water-reducing agent capable of reducing the water/cement ratio of the mix at least about 15% at constant workability, additives being also provided for reducing or completely eliminating the bleeding of the fresh mix. Advantageously the water-reducing agent may consist of a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series with formaldehyde, and/or the polymer obtained by polycondensation of melamine withformaldehyde and containing free or salified sulphonic groups in the heterocyclic rings. According to the invention the admixture may comprise silica (SiO2) with a specific surface area of not less than 2104 cm2 /g determined by isothermal adsorption of a gas. The present invention is further describedin detail hereinafter and also by way of specific working embodiments, which are given for illustrative purposes only, and with reference to the accompanying figures, in which: FIG. 1 is a graph which shows the progress with time of the percentage expansion of samples prepared with an expansive admixture according to the invention and, for comparison purpose, of samples prepared with known expansive admixtures, and FIG. 2 is a graph which shows the progress with time of the percentage expansion of samples prepared with further expansive admixtures according to the invention and, for comparison purpose, of a sample prepared without expansive admixture. As stated, the admixture according to the invention contains the calcium and/or magnesium oxide and a water-reducing agent capable of reducing the water/cement ratio of the mix of water, cement and aggregates at least about 15% at constantworkability.

The action of the calcium and magnesium oxide is to provide the expansion which can be advantageously utilized anticipating the hardening of cement by strongly reducing the water/cement ratio of the mix by the presence of the above-mentionedwater-reducing agent. Furthermore, it has been discovered that the simultaneous presence of the two compounds in accordance with the present invention give an unexpected synergistic effect on the development of the strength. The admixture so obtained is mixed with a hydraulic binder and inerts, and this may be done either at the place of use or preferably at the place of production, in order to supply the user with a premixed product of ready use and rigorouslyconstant composition. In any case, the admixture comprising the above mentioned expansive oxide(s) and the water-reducing agent, hydraulic binder and inerts, is mixed with water at the time of use. The water-reducing agent according to the invention can be, for example, one of the following compounds: polymer obtained by polycondensation of melamine with formaldehyde and containing free or salified sulphonic groups in the heterocyclic rings and represented by the formula: ##STR1## where n = 1,2,3 . . . polymer obtained by polycondensation of free or salified -naphthalenesulphonic acid with formaldehyde and represented by the formula: ##STR2## where n = 1,2,3 . . . As stated, the admixture according to the invention are mixed for use with a hydraulic binder and inerts. The hydraulic binder preferably consists of Portland cement, but pozzolanic, slag or aluminous cement may be used instead of it. The percentage of the admixtures according to the invention relative to the weight of hydraulic binder to be used to make up a paste or a mortar or a concrete may vary from a minimum of 0.5% to a maximum of 40% according to the expansion or thestrength to be obtained. The inerts may be natural (for example sand) or metal based on iron, and the choice of one or other type depends on various factors and in particular on the characteristics which the cement conglomerate is to have. Thus if for example impactresistant characteristics are required, the use of metal inerts is preferred, while if wear resistant characteristics are required, natural inerts may be used, which are also of lower cost. In the cases in which a very flowable mix is required, the presence of high specific surface silica can improve the performance of the mixture containing calcium and/or magnesim oxide and the water-reducing agent. In this ternary mixture calciumor magnesium oxide reacts with the water in the mix to form hydroxide, which in its turn reacts with the water and silica to form calcium or magnesium silicate hydrate in accordance with the reactions: where x varies from 0.8 to 2.

The introduction of silica into the expansive admixture and the consequent conversion of the calcium hydroxide into calcium silicate hydrate in accordance with the reaction given heretofore enable calcium oxide to be more advantageously used asan expansive agent, for the following reasons: while calcium hydroxide has poor cementing properties, calcium silicate hydrate with a molar ratio of CaO/SiO2 between 0.8 and 2, i.e. similar to that obtained in the hydration of Portland cement, has excellent binding characteristics; while calcium hydroxide is easily washed away by water, especially if it contains carbon dioxide, calcium silicate hydrate is considerably less soluble and therefore more resistant to water action. Instead of silica, a natural or artificial material of high specific surface area, in which silica is the predominant compound, could be used. The ternary mixture containing calcium or magnesium oxide, water-reducing agent and silica may be used for anchoring machines (rotating, pulsating, vibrating) to foundations, for anchoring pins, for sealing column bases, for sealing plates,joints and pipes, for filling cavities and crevices, for laying bricks, tiles and floor tiles, for anchoring iron mesh for floors, for repairing holes. The advantage of using specific surface silica is due to the fact that it considerably reduces bleeding when the mix is still fresh. The absence of bleeding is particularly important in all the above-mentioned applications. The following examples will serve to further illustrate the invention, but are not intended to limit the characteristics of the admixtures prepared according to the invention and of the mixes obtained with them. EXAMPLE 1 This example shows the unespected synergistic effect on the strength of the admixture according to the invention. Furthermore only in the admixture according to the invention the hydration of calcium oxide can be advantageously utilized as theexpansion takes place in a strongly opposing environment. A mix containing 40% of Portland cement, 56% of sand and 4% of admixtures A, B, C according to table 1 has been prepared. TABLE 1: ______________________________________ Strength and expansion of mortars ______________________________________ Composition A B C of admixture Expansion 0,10 0,00 0,06 (%) Strength Flexural 9 17 60 (Kg/cm2) Compressive 77108 402 ______________________________________ A = 100% of CaO B = 100% of polymer obtained by polycondensation of sodium salt of -naphtalensulphonic acid with

formaldehyde C = 82% of CaO and 18% of the same polymer as in B All the mortars were mixed with enough water to obtain the same workability, as evaluated by measuring their drop table (140 mm). The percentages of mixing water were 22,0%, 13,0% and 13,5% for the mixes containing admixtures A, B, Crespectively. The fresh mortars were cast into metal cylinders of 10 cm in height and 5 cm in diameter for subsequent measurement of the dimensional changes in the test pieces according to the Corp of Engineering Standard CRDC-589-70. The values ofexpansion shown in table 1 are measure of change in height of the samples from the final setting -- evaluated by means of the soil test -- and 1 day, after which no significant expansion took place. The strength was measured on prismatic specimens (4416cm) cured at 20 C. for 1 day. The results reported in table 1 show: (1) the synergistic effect on the strength of the admixture C according to the invention; (2) only in the admixture C according to the invention the hydration of calcium oxide is advantageously utilised as the expansion takes place in a strongly opposing environment due to the synergistic effect on the strength. EXAMPLE 2 An admixture of the following composition by weight was prepared: commercial lime (max. diameter 0.15 mm): 55% amorphous silica with a specific surface area of 15104 cm2 /g: 35% polymer obtained by polycondensation of formladehyde with sulphonated melamine: 10% Samples were then prepared by mixing Portland cement, the above mentioned admixture, sand and/or particles of metallic iron, as indicated in Table 2. TABLE 2: ______________________________________ Percentage composition by weight of the prepared samples. ______________________________________ Sample Cement Admixture Sand Iron ______________________________________ 1 46.3 -53.7 -- 242.3 4.0 53.7 -- 3 38.3 8.0 53.7 -- 4 46.3 -- 24.3 29.4 5 42.3 4.0 24.3 29.4 6 38.3 8.0 24.3 29.4 ______________________________________ The Portland cement is of the high strength type (425 Kg/cm2 at 28 days) in accordance with Italian standards for hydraulic binders. The sand is of siliceous type with a maximum particle size of 1.2 mm. The maximum particle size of themetallic inert is also 1.2 mm. The samples shown in table 2 were mixed with water to prepare mortars all of the same workability, as evaluated by measuring their drop table (140 mm).

Three expansive commercial mortars were also examined, here called M1, M2, M3. These were also made up with the required amount of water to obtain a drop table flow of 140 mm. The expansion and the strength values were obtained according to the methods described in Example 1. The expansion results are shown in FIG. 1 in which the ordinate represents percentage expansion and the abscissa represents time in hours. The data show that the mortars prepared with samples No. 2, 3, 5, 6 of table 2, all according to theinvention behave as expansive mortars. The results concerning the strength are shown in table 3. These data indicate that mortars prepared from samples 2, 5 and 6 of table 2, all according to the invention, have a strength which is considerably greated both than mortars without the expansive admixture (Samples 1 and 4) and mortars M1, M2, M3commercially available. For example, after a curing of one day, the strength of mortars prepared from the admixture according to the invention is two to three times greater than that of all mortars. TABLE 3 ______________________________________ Flexural (F) and compressive strength (C) of mor- tars prepared from the samples indicated in table 2 and from certain commercially available products. ______________________________________strength ______________________________________ 1 day 3 days 7 days 28 days ______________________________________ F C F C F C F C ______________________________________ Sample 1 26 98 58 296 75 434 88 536 Sample 2 70 317 85 632 88 805 101 1009 Sample 3 71 318 85 463 89 600 95 702 Sample 4 25 101 56 290 77 480 91 609 Sample 5 78 339 98 663 118 830 131 1020 Sample 6 67 254 95 580 115 670 120 980 M 1 26 93 62 320 68 489 104 671 M 2 26 105 56 284 74 444 88 570 M 3 44 178 82 447 75 559 90708 ______________________________________ EXAMPLE 3 Four expansive admixture mixtures were prepared having the composition in table 4 and using the materials described in example 2 for lime, silica and polymer B and in example 1 for polymer A. TABLE 4 ______________________________________ Percentage composition by weight of the expansive admixtures used for the concretes described in example 3 ______________________________________ Expansive admixture 1 2 3 4 ______________________________________ Lime 40 55 80 58 Silica 50 35 10 38 Polymer A 10 -- -- -- Polymer B -- 10 10 4 ______________________________________ Polymer A Polymer obtained by

polycondensation of a sodium salt of -naphtalensulphonic with formaldehyde. Polymer B Polymer obtained by polycondensation of melamine with formaldehyde and containing free or salified sulphonic groups in the heterocyclic rings. Concretes having the following composition were prepared with expansive admixtures: normal Portland cement: 320 Kg/m3 expansive admixture: 30 Kg/m3 inerts having a maximum diameter of 2.5 cm and containing 40% by weight sand: 1900 Kg/m3 Analogous concrete was prepared without the expansive mixture and containing 350 Kg/m3 of the same normal Portland cement. The quantity of mix water used was such as to obtain the same workability (8 cm slump) of all concretes. Prismatictest pieces of size 5530 cm were prepared from these concretes. Each test piece comprised two reference points spaced approximately 25 cm apart to enable the percentage change in length to be measured. Cubic test pieces (15 cm side) werealso prepared for measuring the compressive strength after curing for 1, 7 and 28 days. One half of the prismatic test pieces and all the cubic test pieces were preserved at 20 C. in an environment saturated with water vapour (relative humidity100%). The remaining prismatic test pieces were preserved under the conditions heretofore described for 7 days and at 20 C. and relative humidity 50% for the remainder of the time. FIG. 2 shows the dimensional changes in the prismatic test pieces as a function of time. The dashed curves refer to test pieces preserved in air at a relative humidity of 50% while the continuous curves refer to test pieces preserved in anenviroment saturated with water vapour. The results show that during the first 7 days in a relative humidity of 100% the concretes containing the expansive admixture of table 4, all according to the invention, show a greater expansion than the concretewithout the expansive admixture. After the first seven days, all the concretes contract because of water evaporation. However those containing the expansive admixture show a dimension greater than the initial dimension because of the greater expansionwhich had occurred during curing in humid conditions, to the extent that net shrinkage during evaporation did not occur. This is indicated by the fact that the curves remain above the zero line. However the concrete without the expansive admixtureunderwent a contraction such that after only one day from the beginning of evaporation, a material shrinkage occurred. The results relative to the compressive strength of the concretes are shown in table 5. Table 5 ______________________________________ Compressive strength (Kg/cm2) of the concretes using the expansive mixtures of table 4. ______________________________________ Expansive Curing Mixture used 1 day 7 days 28 days ______________________________________ -- 94 311 438 1 139 400 540 2 168 412 570 3 149 401 530 4 103 320 444

______________________________________ These results show that it is possible to prepare concretes with the admixtures according to the invention which are expansive and which have compressive strength greater than that of the analogous concrete without the expansive admixture. The following specific compounds will serve to further illustrate examples of waterreducing agents that may be employed in the practice of the invention: a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series, with formaldehyde a polymer obtained by polycondensation of melamine with formaldehyde and containg free or salified sulphonic groups in the heterocyclic rings a polymer obtained by polycondensation of free or salified -naphthalenesulfonic acid with formaldehyde a polymer obtained by polycondenstion of free or salified phenolsulphonic acid, with formaldehyde a polymer obtained by polycondensation of free or salified benzene sulphonic acid, with formaldehyde a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series with formaldehyde, and containing methyl groups. a polymer obtained by polycondensation of a sodium salt of a sulphonic acid of the aromatic series, with formaldehyde. This invention relates to a method of producing expansive and high strength 5 cementitious pastes, mortars and concretes. Cement conglomerates (concretes, mortars or grouts) notably undergo contraction during hydration of the cement due to the evaporation of the water and/or the sedimentation of the inerts In certain applications this phenomenon must be avoided, for example in sealing work in order to avoid formation of cracks and ensure maximum 10 adherence of the mix to the confining perimeter and in the anchoring of machines to foundations, in order to guarantee maximum support capacity and uniform static load distribution. To attain this object it is known to add to the cement mix an expansive component able to produce an increase in volume approximately equal or greater than 15 the contraction in volume which occurs during shrinkage It is possible in this way to prepare shrinkagecompensating or self-stressing concretes or mortars. The known expansive components normally used are the following:

2-calcium aluminates or sulphoaluminates, capable of producing an expanding 2 component, namely ettringite, of composition 3 Ca O A 1203 3 Ca SO, 32 HIO, by means of a hydration reaction, such as: 3 Ca O A 100, + 3 Ca SO 4 2 H 2 Ol+ 26 H 2 O-> 3 Ca O Al Q, 3 Ca SO 4 32 HO,, in which the compound obtained, namely ettringite, occupies a greater 25 volume than the reacting solids. -metallic iron, which during hardening of the mix is transformed into ferric oxide and consequently causes an increase in volume of the entire mix. Other expansive agents theoretically usable are calcium oxide (Ca O) and magnesium oxide (Mg O), as both form the corresponding hydroxides by reaction with 30 water, leading to increase in volume Calcium oxide in particular would have the advantage of being particularly economical as it is produced on a large scale for other uses. In order for an expansive component to be advantageously used in a concrete or mortar, the expansion must take place within a certain time interval and in a suffi 35 ciently opposing environment, and these circumstances depend upon the degree of hardening of the hydraulic binder, for example Portland cement, with which the expansive component is mixed If expansion takes place too early with respect to the concrete hardening process, it is not able to induce any stress within the structure, as this latter becomes too deformed or cracked On the other hand if expansion takes 40 place too late with respect to the concrete hardening process, i e when this has attained high strength, the results are equally ineffective (P Mehta and M Polivka: "Expansive Cements" Acts of the VI International congress on Cement Chemistry, pg. 15, Moscow, September 1974) In addition, most of the expansion should terminate within a relatively short period since in practice the moist curing periods-necessary to the expansion process-are usually short for practical reasons (M Polivka "Factors influencing expansion of expansive cement concretes", American concrete Institute, Special Publication 38, pg 241). Up to the present time neither calcium oxide nor magnesium oxide have been 5 applied on a large scale because of the difficulty of controlling the oxide hydration and hence the expansion process in relation to the hardening of the hydraulic binder. In the US Patent No 3,649,317, for example, it was tried to solve this problem delaying the expansion of calcium oxide by a pretreatment of the expanding agent with "Vinsol" (Registered Trade Mark) (dodecylbenzene sulphonate) However, by 10 this treatment the above mentioned drawbacks -concerning a too long curing are not avoided. The object of the invention is to provide a method of producing expansive and unusually high strength cementitious pastes, mortars and concretes by using calcium and/or magnesium oxide as the expansive agent 15 According to the invention this object is

attained by providing a mix of water, a silicous or aluminous hydraulic cement, (in the case of mortars and concretes) aggregates, and a mixture containing: -calcium oxide and/or magnesium oxide, and -a water-reducing agent able to reduce the water/cement ratio of the mix, which 20 would otherwise be required in the absence of the agent, at least by 15 % at constant workability. Advantageously the -water-reducing agent may consist of a polymer obtained by polycondensation of a free or salified sulphonic acid of the aromatic series with formaldehyde, and/or the polymer obtained by polycondensation of melamine with 25 formaldehyde and containing free or salified sulphonic groups in the heterocyclic rings. Advantageously the mixture according to the invention may also comprise an additive for reducing or completely eliminating the bleeding of the fresh mix. According to the invention the mixture may comprise silica (Si O,) with a specific surface area of not less than 2 1 G 4 cm 2/g determined by isothermal adsorption 30 of a gas. The general method for carrying out the present invention is further described hereinafter and is also illustrated by three examples with reference to the accompanying figures, in which: Figure 1 shows the progress with time of the percentage expansion of samples 35 prepared with an expansive mixture according to the invention and, for comparison purpose, of samples prepared with known expansive mixtures, and Figure 2 shows the progress with time of the percentage expansion of samples prepared with further expansive mixtures according to the invention and, for comparison purposes, of a sample prepared without expansive mixture 40 As stated, the additive mixture used in the method according to this invention contains the calcium and/or magnesium oxide and a waterreducing agent able to reduce the water/cement ratio of the mix of water, cement and aggregates at least by 15 % at constant workability. The action of the calcium and magnesium oxide is to provide the expansion which 45 can be advantageously utilised anticipating the hardening of cement by strongly reducing the water/cement ratio of the mix by the presence of the above mentioned water-reducing agent Furthermore it has been found that the simultaneous presence of the two compounds gives an unexpected synergistic effect on the development of the strength 50The additive mixture so obtained is mixed with a hydraulic binder and aggregates, and this may be done either at the place of use or preferably at the place of production, in order to supply the user with a premixed product for ready use and rigorously constant composition. In any case, the additive mixture comprising the above mentioned expansive 55 oxide(s) and the water-reducing agent, hydraulic binder and aggregates, is mixed with water at the time of use.

The water-reducing agent according to the invention can be, for example, one of the following compounds: -polymer obtained by polycondensation of melamine with formaldehyde and 60 containing free or salified sulphonic groups in the heterocyclic rings and 4 represented by the formula: 1,551,118 2. HO-Ch 2-NH-\ C-NH-CH O-CH -NH-C C-NH-CH 0-CH 2-NH-C C-NH-CH 2-OH 22 I 2 21 I H N N N C, y Sn XC NH-CH 2-SO 3 Na NH-CH 2-SO Na N H-C 2 -SO 3 Na where n= 1, 2, 3 -polymer obtained by polycondensation of formaldehyde with an aromatic sulphonic acid, free or salified, selected from at least one of the group comprising naphthalenesulphonic acid, phenolsulphonic acids; benzenesulphonic 5 acids, methylbenzenesulphonic acids; preferably this polymer is obtained by polycondensation of free or salified 6-naphthalenesulphonic acid with formaldehyde and represented by the formula: CCH-C Ha CH-CH / X _ / \-_ CH C-C 2 /C-H CN XC / CH C CH /CH C Ha-CCH-C 503 Na 503 Na where n= 1, 2, 3 10 As stated, the additive mixtures used in the method according to the invention are mixed for use with a hydraulic binder and aggregates The hydraulic binder preferably consists of Portland cement, but pozzolanic, slag or aluminous cement may be used instead of it. The percentage of the additive mixture used in the method according to the 15 invention relative to the weight of hydraulic binder to be used to make up a paste or a mortar or a concrete may vary from a minimum of 0 5 % to a maximum of % according to the expansion or the strength to be obtained. The aggregates may be natural (for example sand) or metal based on iron, and the choice of one or other type depends on various factors and in particular on 20 the characteristics which the cement conglomerate is to have Thus if for example impact resistant characteristics are required, the use of metal aggregates is preferred. While if wear resistant characteristics are required, natural aggregates may be used, which are also of lower cost. In the cases in which a very flowable mix is required, the presence of high specific 25 surface silica can improve the performance of the additive mixture containing calcium and/or magnesium oxide and the water-reducing agent In this ternary mixture, the general composition of which is the following (by weight): calcium oxide and/or magnesium oxide 45-65 % silica having a specific surface area of not less than 2 104 cm 2/g 30 determined by isothermal adsorption of a gas 30-50 % water-reducing agent 1-12 % calcium or magnesium oxide reacts with the water in the

mix to form hydroxide, which in its turn reacts with the water and silica to form calcium or magnesium silicate hydrate in accordance with the reactions: 35 x Ca O+ H 20 O->x Ca(OH), x Ca(OH)2 +Si On+y H 20 a-+x Ca(OH), Si O 2 y H 2,O where x varies from 0 8 to 2 The introduction of silica into the expansive mixture and the consequent conversion of the calcium hydroxide into calcium silicate hydrate in accordance with 40 the reaction given heretofore enable calcium oxide to be more advantageously used as an expansive agent, for the following reasons: 1,551,118 -while calcium hydroxide has poor cementing properties, calcium silicate hydrate with a molar ratio of Ca O/Si O, between 0 8 and 2, i e similar to that obtained in the hydration of Portland cement, has excellent binding characteristics; -while calcium hydroxide is easily washed away by water, especially if it contains 5 carbon dioxide, calcium silicate hydrate is considerably less soluble and therefore more resistant to water action. Instead of silica, a natural or artificial material of high specific surface area, in which silica is the predominant compound, could be used. The ternary mixture containing calcium or magnesium oxide, water-reducing 10 agent and silica may be used for anchoring machines (rotating, pulsating, vibrating) to foundations, for anchoring pins for sealing column bases, for sealing plates, joints and pipes, for filling cavities and crevices, for laying bricks, tiles and floor tiles, for anchoring iron mesh for floors, and for repairing holes. The advantage of using high specific surface silica is due to the fact that it 15 considerably reduces bleeding when the mix is still fresh The absence of bleeding is particularly important in all the above mentioned applications. The examples given hereinafter further illustrate but do not limit the characteristics of the mixtures prepared according to the invention and of the mixes obtained with them 20 EXAMPLE 1 This example shows the unexpected synergistic effect on the strength of the additive mixture used in the method according to the invention Furthermore only in the additive mixture can the hydration of calcium oxide be advantageously utilised as the expansion takes place in a strongly opposing environment 25 A mix containing by weight 40 % of Portland cement, 56 % of sand and 4 % of mixtures A, B, C according to Table 1 has been prepared. TABLE 1 Strength and expansion of mortars Composition of mixture A B C Expansion (%) 0 10 0 00 0 06 Flexural 9 17 60 Strength (Kg/cm 2) Compressive 77 108 402 A= 100 % of Ca O B= 100 % of polymer obtained by polycondensation of sodium salt of 6naphthalenesulphonic acid with formaldehyde C= 82 % by weight of Ca O and 18 % by weight of the same polymer as in B All the mortars were mixed with enough water to

obtain the same workability, as evaluated by measuring their drop table ( 140 mm) The weight percentages of mixing water were 22 0 %, 13 0 % and 13 5 % for the mixes containing mixture A, B, C respectively The fresh mortars were cast into metal cylinders of 10 cm in height and cm in diameter for subsequent measurement of the dimensional changes in the test pieces according to the Corp of Engineering Standard CRD-C-589-70 The values of expansion shown in Table 1 are measure of the change in height of the samples from the final setting evaluated by means of the soil test-and 1 day, after which no significant expansion took place. The strength was measured on prismatic specimens ( 4 x 4 X 16 cm) cured at 200 C for 1 day. The results reported in Table 1 show: 1) the synergistic effect on the strength of the mixture C according to the invention; 2) only in the mixture C according to the invention is the hydration of calcium oxide advantageously utilised as the expansion takes place in a strongly opposing environment due to the synergistic effect on the strength. 351,551,118 1,551,118 5 EXAMPLE 2 A mixture of the following composition by weight was prepared: -commercial lime (max diameter 0 15 mm) 55 % -amorphous silica with a specific surface area of 15 104 cm 2/g 35 % -polymer obtained by polycondensation of formaldehyde with 5 sulphonated melamine 10 % Samples were then prepared by mixing Portland cement, the above mentioned mixture, sand and/or particles of metallic iron, as indicated in Table 2. TABLE 2 Percentage composition by weight of the prepared samples 10 Sample Cement Mixture Sand Iron 1 46 3 53 7 2 42 3 4 0 53 7 3 38 3 8 0 53 7 4 46 3 24 3 29 4 42 3 4 0 24 3 29 4 6 38 3 8 0 24 3 29 4 The Portland cement is of the high strength type ( 425 Kg/cm 2 at 28 days) in accordance with Italian standards for hydraulic binders:The sand is of siliceous type with a maximum particle size of 1 2 mm The maximum particle size of the metallic inert is also 1 2 mm. The samples shown in Table 2 were mixed with water to prepare mortars all of the same workability, as evaluated by measuring their drop table ( 140 ' mm). Three expansive commercial mortars were also examined, here called M 1, M 2, M 3 These were also made up with the required amount of water to obtain a drop table flow of 140 mm. The expansion and the strength values were obtained according to the methods described

in Example 1. The expansion results are shown in Fig 1 in which the ordinate represents percentage expansion and the abscissa represents time in hours The data show that the mortars prepared with samples No 2, 3, 5, 6 of Table 2, all according to the invention, behave as expansive mortars. TABLE 3 Flexural (F) and compressive strength (C) of mortars prepared from the samples indicated in Table 2 from certain commercially available products Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6 M 1 M 2 M 3 strength 1 day 3 days 7 days 28 days F C F C F C F C 26 9 R 58 2 75 434 S RR 53 317 85 71 318 85 101 56 78 339 98 67 254 95 26 93 62 26 105 56 44 178 82 632 463 290 663 580 320 284 447 88 805 89 600 77 480 118 830 6,70 68 489 74 444 559 u,. 101 91 131 104 88 1009 702 609 1020 980 671 570 708 These data indicate that mortars prepared from samples 2, 3, 5 and 6 of Table 2, all according to the invention, have a strength which is considerably greater than both mortars without the expansive mixture (samples 1 and 4) and mortars M 1, M 2, M 3 commercially available For example, after a curing of one day, the strength of mortars prepared from the additive mixture used in the method according to i the invention is generally two to three times greater than that of the other mortars. EXAMPLE 3 Four expansive mixtures were prepared having the composition indicated in Table 4 and using the materials described in Example 2 for limne, silica and polymer B and in Example 1 for polymer A. TABLE 4 Percentage composition by weight of the expansive mixtures used for the concretes described in Example 3 Expansive mixture 1 2 3 4 Lime 40 55 80 58 Silica 50 35 10 38 Polymer A 10 Polymer B 10 10 4 Polymer A: Polymer B: Polymer obtained by polycondensation of a sodium salt of /3-naphthalenesulphonic acid with formaldehyde. Polymer obtained by polycondensation of melamine with formaldehyde and containing free or salified sulphonic groups in the heterocyclic rings. Concretes having the following composition were prepared with expansive mixtures:

-normal Portland cement -expansive mixture -aggregates having a maximum diameter of 2 5 cm and containing 40 % by weight of sand 320 Kg/ma Kg/m' 1900 Kg/m 3 Analogous concrete was prepared without the expansive mixture and containing 350 Kg/mn 3 of the same Portland cement The quantity of mix water used was such as to obtain the same workability ( 8 cm slump) of all concretes Prismatic test 1,551,118 7 1,551,118 7 pieces of size 5 X 5 X 30 cm were prepared from these concretes Each test piece comprised two reference points spaced approximately 25 cm apart to enable the percentage change in length to be measured Cubic test pieces ( 15 cm side) were also prepared for measuring the compressive strength after curing for 1, 7 and 28 days. One half of the prismatic test pieces and all the cubic test pieces were preserved at 5 C in an environment saturated with water vapour (relative humidity 100 %) The remaining prismatic test pieces were preserved under the conditions heretofore described for 7 days and at 200 C and relative humidity 50 % for the remainder of the time. Fig 2 shows the dimensional changes in the prismatic test pieces as a function of 10 time The dashed curves refer to test pieces preserved in air at a relative humidity of % while the continuous curves refer to test pieces preserved in an environment saturated with water vapour The results show that during the first 7 days in a relative humidity of 100 % the concretes containing the expansive mixture of Table 4, all according to those used in the method of the invention, show a greater expansion 15 than the concrete without the expansive mixture After the first seven days, all the concretes contract because of water evaporation However those containing the expansive mixture show a dimension greater than the initial dimension because of the greater expansion which had occurred during curing in humid conditions, to the extent that net shrinkage during evaporation did not occur This is indicated by 20 the fact that the curves remain above the zero line However the concrete without the expansive mixture underwent a contraction such that after only one day from the beginning of evaporation, a material shrinkage occurred. The results relative to the compressive strength of the concretes are shown in Table 5 25 TABLE 5 Compressive strength (Kg/cm 2) of the concretes using the expansive mixtures of Table 4. Expansive Curing mixture used 1 day 7 days 28 days 94 311 438 1 139 400 540 2 168 412 570 3 149 401 530 4 103 320 444 These results show that it is possible to prepare concretes with the mixtures used in the method according to the invention which are expansive and which have compressive strength greater than that of the analogous concrete without the expansive mixture.

EXPANSIVE ADDITIVE FOR LIME CEMENT AND PROCESS FOR THE PREPARATION OF THE SAME
United States Patent 3801339

An expansive additive composition for use in controlling the expansion of cement on aging comprising a sintered mixture of lime and at least one additive selected from the group consisting of ferric oxide, calcium fluoride and calcium sulfate is disclosed. The process for preparing the expansive additive composition is also disclosed. CROSS REFERENCE TO RELATED APPLICATIONS This application is a division of application Ser. No. 43,050, filed June 3, 1970, now abandoned. Claims: What is claimed is 1. A cement composition consisting essentially of a cement and from 3 to 10 percent by weight of a crushed sintered mixture of lime and at least one additive selected from the group consisting of ferric oxide, calcium fluoride and calcium sulfate, said crushed sintered mixture being obtained by heating said mixture to a temperature of from about 1,200 to about 1,400C. for from about 1 to 2 hours followed by cooling and crushing. 2. The cement composition of claim 1, wherein said additive is ferric oxide and said ferric oxide is present at a level of from 3 to 40 percent by weight based on the weight of the crushed sintered mixture. 3. The cement composition of claim 1, wherein said additive is calcium fluoride and said calcium fluoride is present at a level of from 3 to 60% by weight based on the weight of the crushed sintered mixture. 4. The cement composition of claim 1, wherein said additive is calcium sulfate and said calcium sulfate is present at a level of from 30 to 85% by weight based on the weight of the crushed sintered mixture. Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an expansive additive for lime system cements and to a process for preparing an expansive additive for lime system cements. More particularly, this invention relates to a process for preparing an expansive additive for lime system cements by adding to lime at least one of the ferric oxides, calcium fluoride, and calcium sulfate and sintering or melting the mixture. 2. Description of the Prior Art A cement containing free lime or calcium oxide frequently expands during the setting and hardening thereof. Hence, such a cement has hitherto been considered to be undesirable

since the expansion causes a lack of stability in quality. On the other hand, however, were it possible to control the expansion characteristics of such a cement so that the expansion of the cement occurred at a proper period after hardening of the cement, it might be quite preferable. In the case of compounding ordinary calcium oxide prepared by sintering limestone at a temperature of about 1,200C to 1,400C in a cement, it expands rapidly immediately after setting. Hence, such a cement cannot be used practically. Investigations of a sintering process for lime, capable of maintaining a mild expansion characteristic of the cement up to a suitable time after hardening of the cement by improving the rapid expansion of the free lime, resulted in the discovery of the process of this invention. SUMMARY OF THE INVENTION The present invention is a process for the preparation of an expansive additive for lime system cement which comprises adding at least one member selected from the group consisting of ferric oxide (Fe2 O3), calcium fluoride (CaF2), and calcium sulfate (CaSO4) to lime, crushing the mixture, and sintering or melting the mixture at a temperature at which the additive or additives are not decomposed. The clinker thus obtained effectively controls the occurrence of the rapid expansion accompanied by the use of lime alone to cause a mild and prolonged expansion of a cement containing the clinker at the proper time during the progress of hardening of the cement. Hence, it is used as a very effective inflating agent for cement. DETAILED DESCRIPTION OF THE INVENTION As the lime used in this invention as the main raw material, any material which can provide a CaO source, such as limestone, calcium oxide or calcium hydroxide, may be employed. The ferric oxide used as one of the additives to the lime in this invention is mainly composed of Fe2 O3 and can contain impurities but it is necessary that such impurities be inactive toward lime. In particular, mill scale is preferably utilized as the ferric oxide. Also, the calcium fluoride and the calcium sulfate to be used as the additives in this invention can contain impurities which are inactive toward lime. Generally, commercially available materials can be employed as such additives. The amounts of ferric oxide, calcium fluoride, and calcium sulfate to the lime will be explained by relating the amounts added in terms of CaO and to the expansion effect of a mortar prepared by compounding the expansive additive formed by sintering the mixture of lime and the additives in a cement as shown in the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a graph showing the characteristics concerning the expansion coefficient of a cement, having added thereto a cement expansive additive containing lime and ferric oxide, with age. FIG. 2 is a graph showing the characteristics concerning the expansion coefficient of a cement having added thereto a cement expansive additive containing lime and calcium fluoride, with age. FIG. 3 is a graph showing the characteristics concerning the expansion coefficient of a cement, having added thereto a cement expansive additive containing lime and calcium sulfate, with age. The cement expansive additive prepared by mixing a powdered limestone crushed to a size smaller than 88 microns and a ferric oxide powder (Fe2 O3) crushed into a size smaller than 88 microns in the weight ratio shown in the table below, heating the mixed powders in an electric muffle furnace for 2 hours at temperatures of 1,400-1,450C to provide a clinker, and crushing the clinker to a size smaller than 149 microns was compounded in a cement in an amount of 6 percent by weight. The relation between the expansion characteristics of the cement and the curing period (in days) is shown in FIG. 1. Test No. CaO Fe2 O3 ______________________________________ A 95 5 B 90 10 C 85 15 D 75 25 E 63 35 F(control) 100 0 ______________________________________ For comparison, calcium oxide (CaO), prepared by baking the powder of limestone for 1 hour at 1,400C without added Fe2 O3, was added to a cement in an amount of 3 percent by weight and the relation of the expansion characteristics of the cement and the curing period is shown in FIG. 1 as Curve F. Also, the expansive additive prepared similarly by mixing powdered limestone crushed smaller than 88 microns and powdered calcium fluoride (CaF2) crushed smaller than 88 microns in the weight ratio shown in the following table, sintering the mixture by heating the mixture in an electric muffle furnace under the same conditions as used for FIG. 1 to provide a clinker, and crushing the clinker smaller than 149 microns was compounded in a cement in an amount of 4 percent by weight. The relation of the expansion characteristics of the cement and the curing period in days is shown in FIG. 2. For comparison, the use of powdered calcium oxide alone is also shown in FIG. 2 as Curve F. Test No. CaO CaF2 ______________________________________ G 80 20 H 70 30 I 60 40 J 50 50 F (control) 100 0 ______________________________________ Moreover, the expansive additive prepared by mixing powdered limestone crushed smaller than 88 microns and powdered calcium sulfate (CaSO4) crushed smaller than 88 microns in the mixing ratio shown in the table below, sintering the mixture by heating it

in an electric muffle furnace for 2 hours at 1,200-1,300C to provide a clinker, and crushing the clinker smaller than 149 microns was compounded in a cement in an amount of 5 percent by weight. The relation between the expansion characteristics and the curing period is shown in FIG. 3. For comparison, use of powdered calcium oxide alone is shown in the same FIG. 3 as Curve F. Test No. CaO CaSO4 ______________________________________ K 45 55 L 35 65 M 20 80 F (control) 100 0 ______________________________________ The mortar samples used above were prepared by the JIS R-5201 method and the expansion characteristics of the mortar were obtained by means of comparative measurements of the change of the surface length during curing in water at 20C using the surface length of sample just after coagulation and release of the mold as the basis. As is clear from the results shown in FIGS. 1 to 3, the mortar containing calcium oxide alone with no expansion controlling agent such as ferric oxide, calcium fluoride or calcium sulfate, suffered rapid expansion at the beginning of aging, while the mortar containing calcium oxide together with ferric oxide, calcium fluoride or calcium sulfate expanded gradually during aging, as shown by the slowly changing curve, and became stable after about 3 weeks. The results obtained also show that the expansion characteristics of the mortar are influenced by the amount of expansion controlling agent added and an increase in the amount of the additive, results in a tendency for the expansion of the mortar to occur later. As far as the results shown in the graphs are concerned, the effect of the addition of ferric oxide, calcium fluoride or calcium sulfate, which contributes to the control of the expansion of the cement, is at a maximum in Curve E, J, or M respectively. In this respect, further investigations were made into the effective amount of the aforesaid expansion controlling agent together with the minimum amount thereof capable of showing the expansion controlling effect. It was discovered that the critical characteristics were shown by Curves R and S in FIG. 1, Curves T and U in FIG. 2, and Curves V and W in FIG. 3. Accordingly, the effective amount of ferric oxide to be added is about 3-40% by weight of the total weight of the sintered product, that of calcium fluoride is about 3-60% by weight, and that of calcium sulfate is about 30-85% by weight. In particular, a preferred amount of ferric oxide is from 5 to 35% by weight, that of calcium fluoride is from 20 to 50% by weight, and that of calcium sulfate is from 55 to 80% by weight. The mixture of limestone and the additive must be heated at temperatures higher than 1,200C until the mixture becomes a homogeneous sintered state or semi-molten state. If the mixture is insufficiently sintered, the effect obtained with the sintered product is the same as that obtained with calcium oxide alone in the expansion characteristics of the

cement to which the clinker is added. In other words, no effect of the additive is obtained. Therefore, when the heating period for sintering the mixture is about 1 to 2 hours, the proper heating temperature is about 1,450C where ferric oxide is used as the additive, about 1,400C where calcium fluoride or a mixture of ferric oxide and calcium fluoride is used as the additive, and about 1,200-1,300C where calcium sulfate is used as the additive. In addition, the same experiments described above were repeated using industrial raw materials such as calcium oxide prepared by baking limestone, mill scale, fluoride, and gypsum by-products in the production of hydrofluoric acid instead of the limestone, ferric oxide, calcium fluoride, and calcium sulfate respectively. Almost the same results as described above were obtained. The amount of the expansive additive in this invention to be compounded with the cement will be explained below. An expansive additive generally shows different expansion characteristics with the passage of time depending on the composition thereof. Also, when the composition of an expansive additive remains constant, the expansion characteristics thereof were, of course, higher as the amount of the agent to be compounded in cement is increased. Furthermore, since the proper compounding ratio of the expansive additive in this invention to cement necessary for causing an expansion of from 1 to 1.5%, by which the shrinkage of an ordinary cement can be completely prevented, depends upon the composition of the expansive additive. It is generally impossible to define numerically the compounding ratio thereof but it is usually preferable to add the expansive additive to a cement in an amount of from about 3 to 10% by weight based on the weight of the cement. If the amount of the expansive additive is less than about 3% by weight, it cannot provide effective expansion characterics. While, if the amount is larger than 6% by weight, the mechanical strength of the cement after setting is reduced. For example, the most preferred compounding ratio of the expansive additive is about 6% by weight when the ratio of the calcium oxide to the ferric oxide of the inflating agent is 85:15, about 4% by weight when the ratio of the calcium oxide to the calcium fluoride is 70:30, about 5% by weight when the ratios of the calcium oxide to the ferric oxide to the calcium fluoride are 70:15:15, and about 6% by weight when the ratio of the calcium oxide to the calcium sulfate is 45:55. The invention has been described in terms of blending of mortar but is also applicable to the blending of concrete. From the above descriptions it will be understood that in the present invention, ferric oxide, calcium fluoride, and calcium sulfate exhibit an effective and a similar function for controlling the expansion characteristics of calcium oxide to cement when they are employed individually or in combination together with calcium oxide. This invention will be explained further by reference to the following examples.

EXAMPLE 1 A mixture of 91 parts by weight of limestone having a purity of 98% and 9 parts by weight of mill scale was crushed and baked for 2 hours at 1,450C to provide a clinker for use as an expansive additive. The clinker was crushed and added to a cement in an amount of 5% by weight or 6% by weight to provide cement and cement , respectively. For comparison, a conventional cement containing no such expansive additive was prepared. By using the cement prepared above a 10 10 50 cm3 concrete sample was prepared. The composition of the concrete per cubic meter thereof was as follows: Blend ______________________________________ Blend Blend ______________________________________ Expansive Additive of this Invention 168 kg 191 kg Ordinary Portland Cement 299 kg 296 kg Water 218 kg 218 kg Gravel (2.5 mm in diameter) 955 kg 955 kg Sand 876 kg 876 kg Total 2364 kg 2364 kg ______________________________________ The expansion coefficient of the concrete during curing in water at 20C, measured by the above-described method employed with the mortar samples, is shown in the following Table 1. TABLE 1 ______________________________________ Expansion Coefficient ( 103%) Type Age (day) ______________________________________ 1 2 4 7 14 21 28 ______________________________________ Cement 12 19 31 48 64 78 79 Cement 18 23 39 63 81 101 101 Conventional -2 3 -1 2 1 3 2 ______________________________________ EXAMPLE 2 A powdered mixture of 70 parts by weight of calcium oxide prepared by baking limestones at 1,200C and 30 parts by weight of fluoride was heated for 2 hours at 1,400C to provide a clinker as an expansive additive. The clinker was crushed smaller than 149 microns and compounded in a cement in an amount of 4% by weight to provide cement . Separately, a powdered mixture of 70 parts by weight of calcium oxide, 15 parts by weight of mill scale, and 15 parts by weight of fluoride was heated for 2 hours at 1,400C to provide a clinker as an expansive additive. The clinker was crushed into a size smaller than 149 microns in diameter and compounded in a cement in an amount of 5% by weight to provide cement . The expansion coefficients of the cements prepared above were measured by the same test procedure as used in Example 1 and the results are shown in Table 2. TABLE 2

_______________________________________________________________________ ___ Expansion Coefficient (x 10-3 %) Type Age (day) _______________________________________________________________________ ___ 1 2 4 9 14 21 28 _______________________________________________________________________ ___ Cement 122 180 195 202 212 223 223 Cement 54 81 120 168 187 196 198 _______________________________________________________________________ ___ EXAMPLE 3 A powdered mixture of calcium oxide having a purity of 98 percent and a gypsum byproduct obtained in the production of hydrofluoric acid and having a purity of 90 percent mixed in a weight ratio of 40:60 or 20:80 and was baked for 2 hours at 1,200C to provide a clinker as an expansive additive. The clinker was crushed to smaller than 149 microns and was added to a cement in an amount of 5 percent by weight in the former (40:60 mixture) and 8 percent by weight in the latter (20:80 mixture) to provide cement and cement . The expansion coefficients of these cements were measured by the same test method as used in Example 1 and are shown in the following table. TABLE 3 _______________________________________________________________________ ___ Expansion Coefficient (x 10-3 %) Type Age (day) _______________________________________________________________________ ___ 1 2 4 7 14 21 28 _______________________________________________________________________ ___ Cement 100 154 190 211 220 225 224 Cement 94 135 161 182 189 195 196 _______________________________________________________________________ ___ antz, Delbert E. Assistant Examiner: Hellwege, James W. Attorney, Agent or Firm: Caesar, Rivise, Bernstein & Cohen Claims: I claim 1. An additive for use in hydraulic cements, comprising a major amount of sodium nitrite in admixture with a minor amount of calcium formate, said major amount of sodium nitrite comprising about 52% to 81% by weight of said additive and said minor amount of calcium formate comprising about 19% to 48% by weight of said additive. 2. An additive according to claim 1, which also contains triethanolamine or sodium benzoate.

3. An additive according to claim 2, wherein the triethanolamine or sodium benzoate is present in a proportion of up to 40% by weight of the additive. 4. An additive according to claim 2 wherein the triethanolamine or sodium benzoate comprises about 0.1 to 10% by weight of said additive. 5. An additive according to claim 1 and water. 6. An additive according to claim 5 in which the weight ratio of water to additive is in the range of 4:1 to 1:9. 7. An additive according to claim 6 wherein the weight ratio of water to additive is in the range 2:3 to 3:2. 8. An additive according to claim 1, and a dispersant to present the additive in liquid, paste or powder form. 9. An additive according to claim 1 comprising about 48% calcium formate, and about 52% sodium nitrite. 10. An additive according to claim 2 comprising about 19.5% calcium formate, about 77.9% sodium nitrite, and about 2.6% triethanolamine. 11. A cementitious composition incorporating an additive according to claim 1. 12. A cementitious composition according to claim 11, in which the weight of additive in the composition is up to about 20%. 13. A cementitious composition according to claim 12, in which the weight of additive in the composition is about 1% to about 6%. Description: BACKGROUND OF THE INVENTION This invention relates to additives for hydraulic cements and, more particularly, additives for concretes which are to contain metal reinforcement. Such additives may be added to a hydraulic cement at a cement manufacturing plant or on a construction site. The term "hydraulic cement" means a cementitious composition capable of being set and hardened by the action of water; examples are Portland cement, sulphate-resisting cements, blast-furnace cements, and pozzolanic cements. When, for example, a Portland cement, (mainly calcium silicate 3Ca0. Si02 and a minor amount of tricalcium aluminate 3Ca0. Al2 03), is mixed with water, setting takes place in

only a few hours. However the hardening of the cement takes much longer; in some cases final hardness is only achieved after weeks. The setting is due to the initial reaction of water with the tricalcium aluminate and the hardening is mainly due to the reaction of water with the tricalcium silicate. Ultimately, a gelatinous hydrated product, (probably gelatinous calcium silicate) forms and probably binds the particles together. The rate of setting is dependant on the temperature, the setting reaction proceeding very slowly at lower temperatures. It is desirable to accelerate setting. As on-site heating is beset with difficulties, it is general practice to add to the water/hydraulic cement mix, a setting accelerator, which serves to decrease the setting time both at high and at low temperatures. Such accelerators include chlorides, carbonates, hydroxides or fluorsilicates. Calcium chloride is widely used but the chloride ions react with atmospheric moisture and oxygen to form an acidic mixture which will attack a metal reinforcement. Unless the reinforcement is completely sealed against the atmosphere it can, in time, be seriously corroded. Hence calcium chloride is not preferred for use with metal (especially steel) reinforced concrete. SUMMARY OF THE INVENTION It is an object of the invention to provide an additive for a hydraulic cement usable with metal reinforcement. It is an object of the invention to provide an additive for a hydraulic cement which can be used at low temperatures. It is another object of the invention to provide an additive for hydraulic cement which is free of chloride. It is yet another object of the invention to provide an additive for hydraulic cement which gives improved concrete compressive strength. These and other objects of the invention are obtained using an additive which comprises a major amount of sodium nitrite in admixture with a minor amount of calcium formate. Optionally the additive may also contain a third constituent, for example triethanolamine or sodium benzoate, e.g. in a proportion up to 40% by weight. A preferred dry additive comprises 19 to 48% calcium formate, 81 to 52% sodium nitrite, and 0.1% to 10% of triethanolamine or sodium benzoate. Water may be added in a weight ratio of water to additive of up to 0.8:1, preferably 0.4 to 0.6 to form a liquid additive. The additive may be admixed with the cement just before use on site or dry additive may be admixed with the cement in manufacture, and water added on site. Preferably the additive is dissolved in the gauging water (water used for admixture with the cement) which is then added to cement.

The additive may be added to the cementitious composition in amounts up to 20% by weight, preferably 1-6% by weight. The following examples will serve to illustrate the invention. In these examples, use is made of two specific additives (parts are by weight) Additive A calcium formate 48% sodium nitrite 52% (Application rate 1.8 lb/cwt dry cement) Additive B calcium formate 19.5% sodium nitrite 77.9% triethanolanium 2.6% (Application rate 2 lbs/cwt dry cement.) The following test procedures were used in the Examples. Compressive strength was measured on a concrete containing by weight, one part test cement (Portland or sulphateresisting) 2.4 parts sand and 3.6 parts 3/4 inch to 3/16 inch Thames gravel, these being mixed according to British Standard BS 1881. The 4 inch cubes for compressive strength determinations were cured under BS 12 conditions (i.e. 20 C) or immediately placed at +6 C and then cycled every 24 hours between +6 C and -6 C. Water Reduction was determined by the following test. If the quantity of water required to give a control concrete a degree of workability is 100%, the water content required by cement with admixture for the same workability expressed as a percentage of the control concrete water requirement is termed "water reduction" (for example if without admixture 60 lbs of water is required, but with admixture only 54, water reduction is [(60-54)/60] 100 = 10%). The Proctor Needle and Vicat tests to determine setting times were performed as described in ASTMC. 40.3 with 1:2:4 mortars and British Standard BS 12 with neat cement, respectively. The slump test was measured as described in British Standard BS 1881 and BS 12. EXAMPLE I Samples of ordinary Portland cement and sulphate-resisting cement were made up into a

concrete as indicated, with and without the additives as indicated in Table I. The products were cured at 20 C. and measured and the results are shown in Tables I and II below. The results of Tables I and II show that both additives give a much higher compressive strength in all stages of setting and hardening than the control. TABLE I _______________________________________________________________________ ___ Sample Water/ Slump Water Density Compressive Strength (p.s.i.) Cement (ins.) Reduction (lb/cu.ft.) 18hr 20hr 24hr 3day 7day Ratio _______________________________________________________________________ ___ Ordinary Portland Cement Control 0.625 17/8 -- 146.2 730 970 1540 2660 3750 Additive B (1.2 lbs/cwt) Cement 0.625 17/8 -- 146.2 1470 1760 1680 3470 4060 Additive A (11/2 lb/cwt) 0.625 13/4 -- 146.9 1270 1260 1820 3440 4800 Control 0.625 21/8 -- 148.6 1050 1190 1680 2730 3850 Additive B (2.4 lb/cwt) 0.605 17/8 3.2 148.1 1820 2100 2500 3640 4740 Additive A (3 lb/cwt) 0.600 21/8 4.0 147.7 1470 1540 1970 2450 4070 Sulphate resisting Cement Control 0.610 2 -- 147.0 670 980 1150 3080 3550 Additive B (1.2 lb/cwt cement) 0.605 23/4 0.8 145.0 900 1060 1400 3470 4190 Additive A (11/2lb/cwt) 0.559 21/4 1.8 146.0 830 870 1370 3080 4170 Control 0.591 17/8 -- 147.7 700 1120 1050 2370 3640 Additive B (2.4 lb/cwt cement) 0.575 17/8 2.7 146.4 1470 1750 1890 3500 5020 Additive A (3 lb/cwt) 0.545 21/8 7.8 147/1 1330 1680 1960 3500 _______________________________________________________________________ ___ 4970 ##SPC1## EXAMPLE II A series of tests were set up to test the efficiency of additive B at various addition rates and temperatures. The samples were tested by measurement of the Proctor Needle Setting times on two samples of 1:2.4 mortars (sulphate resisting and ordinary Portland cement). The results are given in Tables III and IV. EXAMPLE III A further set of comparative tests was set up under various conditions using the following: Additive C comprised 75% calcium formate, 25% sodium nitrite Additive D comprised 48% calcium formate, 52% sodium nitrite Additive E comprised 25% calcium formate, 75% sodium nitrite The cement used was ordinary Portland cement.

The results are given below in Table V in the terms of the compressive strength in p.s.i. and in percent of control after the times indicated. EXAMPLE IV The compressive setting time of the compositions of Example III was determined on Portland Cement, the samples being cured at 20 C. The Vicat and Proctor Needle determinations in minutes are shown in Table VI. From the results of the Tables, it is clear that additives of this invention give positive strength increases and acceleration of setting times over the controls and compare favourably with CaCl2 with regard to strength and setting times. Further tests have shown that the resultant concretes show no negative results with regard to shrinkage and sulphate resistance, and have a positive corrosion inhibition effect. TABLE III _______________________________________________________________________ ___ Using Portland Cement Rate of Temperature Water/ Sample Addition (C) Cement Setting Times(hrs) _______________________________________________________________________ ___ Ratio Initial Final _______________________________________________________________________ ___ Control - 20 0.408 3.1 5.0 Additive B 1.2 lb 20 0.406 2.5 3.7 Additive A 11/2 lb 20 0.407 2.7 4.0 _______________________________________________________________________ ___ Control - 20 0.415 3.3 5.1 Additive B 2.4 lb 20 0.393 2.4 3.3 Additive A 3 lb 20 0.422 2.4 3.5 _______________________________________________________________________ ___ Control - 6 0.455 6.5 10.6 Additive B 1.2 lb 6 0.400 3.8 6.0 Additive A 11/2 lb 6 0.410 3.9 6.5 _______________________________________________________________________ ___ Control - 6 0.400 5.0 10.0 Additive B 2.4 lb 6 0.400 3.0 6.2 Additive A 3 lb 6 0.400 3.0 6.3 _______________________________________________________________________ ___ TABLE IV _______________________________________________________________________ ___ Using Sulphate resisting Cement Rate of Temperature Water/ Setting Times (hrs) _______________________________________________________________________ ___ Sample Addition Cement Initial Final (lb) (C) Ratio _______________________________________________________________________ ___ Control - 20 0.397 4.8 6.8 Additive B 1.2 20 0.403 4.7 6.5 Additive A 11/2 20 0.305 4.0 5.5 _______________________________________________________________________

___ Control - 20 0.393 4.7 6.3 Additive B 2.4 20 0.393 4.2 5.7 _______________________________________________________________________ ___ Control - 6 0.395 11 - Additive B 1.2 6 0.375 8 - Additive A 11/2 6 0.375 8 _______________________________________________________________________ ___ Control - 6 0.40 10.8 19 Additive B 2.4 6 0.40 6.2 11.3 Additive A 3 6 0.40 4.0 10.0 _______________________________________________________________________ ___ These results show that the additives decrease the setting time of both Portland Cement and sulphate-resisting cement. TABLE V _______________________________________________________________________ ___ Sample Curing W/C Slump Comp. Strength (p.s.i.) _______________________________________________________________________ ___ Ref. Cond's. Ratio (ins) 1day 3Day 7day 28day _______________________________________________________________________ ___ Control -5C last 0.718 21/4 840 750 700 24 hrs in Water at 19C 2% Mix C 0.679 21/2 830 690 570 2% Mix D 0.680 21/4 1130 770 870 2% Mix E 0.670 21/4 850 1000 2% CaCl2 0.660 2 1110 880 740 Flake _______________________________________________________________________ ___ Control Cured 0.680 2 180 2040 3520 outdoors 2% Mix D Max.Temp 0.670 2 630 3080 5270 11C 2% CaCl2 Min. Temp 0.603 2 720 3160 5030 -3C Flake Mean Temp 4C _______________________________________________________________________ ___ Control BS.12 0.676 2 1130 3840 5030 (Std. 2% Mix D Lab. 0.660 2 1730 5040 6340 Cond'ns.) 2% CaCl2 0.610 2 1550 4590 5950 Flake _______________________________________________________________________ ___ Control BS.12 0.673 2 1260 4240 5840 (Std. 1% Mix D Lab. 0.640 2 1750 4690 6250 Cond'ns.) 1% Mix E 0.640 2 1620 4790 6080 1% CaCl2 0.646 2 1780 4990 6210 Flake _______________________________________________________________________ ___ TABLE VI _______________________________________________________________________ ___ Vicat Proctor Needle _______________________________________________________________________ ___ Sample Initial Set Final Set Initial Set Final Set Control 115 185 310 430 2% Mix C 100 135 - - 2% Mix D 75 120 170 290 2% Mix E 65 115 - - 2% CaCl2 Flake 55 90 150 250 _______________________________________________________________________ ___ Control 140 220 310 430 1% Mix D 115 165 220 300 1% Mix E - 14 220 280 1% CaCl2 Flake 80 105 160 250 _______________________________________________________________________ ___

Without further elaboration, the foregoing will so fully illustrate my invention that others may, by applying current or future knowledge, readily adapt the same for use under various conditions of service. Expansive cement compositions and novel expansion agents containing free lime for use in such compositions. The expansion agents are prepared by heating lime of particle size less than 250 microns in an atmosphere comprising water vapor at 100.degree. to 400.degree. C. or carbon dioxide at 350.degree. to 850.degree. C., or by heating a partially hydrated lime at a temperature of at least 100.degree. C. The encapsulated limes produced in this way are useful as expansion agents in hydratable cement compositions, the amount thereof generally being less than 30%, preferably 3 to 10%, based on the weight of the cement and additive. Cement compositions containing the novel expansion agents are useful in a wide variety of uses where shrinkage of the composition on drying has to be prevented or reduced. Claim: We claim: 1. A pulverulent composition which is useful as an expansion additive for hydratable cement compositions; which consists essentially of particles having a size less than 100 micronsand comprising a core of calcium oxide and a protective coating around said core of calcium carbonate; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%. 2. A composition according to claim 1 which comprises 40 to 91% by weight of calcium oxide and wherein said protective coating consists essentially of calcium carbonate. 3. A composition according to claim 2 which comprises 64 to 91% by weight of calcium oxide. 4. A composition according to claim 1 wherein said particles have a size of 1 to 100 microns. 5. A composition according to claim 4 wherein said particles have a size of 5 to 60 microns. 6. A composition according to claim 1 having an Expansion Factor of 0.07 to 0.4%. 7. A process for the production of a composition useful as an expansion additive for hydratable cement compositions, which process comprises heating particles (a) having a size less than 100 microns and (b) initially comprising at least 80% byweight of CaO at a temperature of 100 to 400.degree. C. in an atmosphere comprising water vapor for a time sufficient to increase the weight of the particles by at least 2%, said heating being carried out under conditions such that the product contains36 to 95% by weight of CaO and exhibits an Expansion Factor of at least 0.06%. 8. A process according to claim 7 wherein the weight of the particles is increased by 4 to 12%.

9. A process according to claim 8 wherein the weight of the particles is increased by 5 to 11%. 10. A process for the production of a composition useful as an expansive additive for hydratable cement compositions, which process comprises heating particles (a) having a size less than 100 microns and (b) initially comprising at least 80% byweight of CaO at a temperature of 350.degree. to 850.degree. C. in an atmosphere comprising CO.sub.2 for a time sufficient to increase the weight of the particles by at least 2%, said heating being carried out under conditions such that the productcontains 36 to 95% by weight of CaO and exhibits an Expansion Factor of at least 0.06%. 11. A process according to claim 10 wherein the weight of the particles is increased by 4 to 16%. 12. A process for the production of a composition useful as an expansion additive for hydratable cement compositions, which process comprises heating particles (a) having a size less than 100 microns and (b) initially comprising at least 80% byweight of CaO, the particles being maintained at a temperature of at least 100.degree. C. in an atmosphere which comrpises CO.sub.2 and water vapor for a time sufficient to increase the weight of the particles by 4 to 14%, said heating being carried outunder conditions such that the product contains 36 to 95% by weight of CaO and exhibits an Expansion Factor of at least 0.06%. 13. A process for the production of a composition useful as an expansion additive for hydratable cement compositions, which process comprises heating hydrated lime particles at a temperature of 100.degree. to 400.degree. C., said particles (a)having a size less than 100 microns and (b) comprising at least 36% by weight of CaO and 8 to 64% by weight of Ca(OH).sub.2, said heating being carried out under conditions such that the product contains 36 to 95% by weight of CaO and exhibits anExpansion Factor of at least 0.06%. 14. A process according to claim 13 wherein said particles initially contain 16 to 50% by weight of Ca(OH).sub.2. 15. A cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and a protective coating around saidcore of calcium carbonate; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount less than 30%, based on the combined weight of the cementand the pulverulent composition. 16. A cement composition according to claim 15 which comprises 3 to 10% of said pulverulent composition, based on the combined weight of the cement and the pulverulent composition.

17. A cement composition according to claim 15 wherein said pulverulent composition comprises 40 to 91% by weight of calcium oxide and said protective coating consists essentially of calcium carbonate. 18. A cement composition according to claim 17 wherein said pulverulent composition comprises 64 to 91% by weight of calcium oxide. 19. A cement composition according to claim 15 wherein said particles have a size of 1 to 100 microns. 20. A cement composition according to claim 19 wherein said particles have a size of 5 to 60 microns. 21. A composition according to claim 15 wherein said pulverulent composition has an Expansion Factor of 0.07 to 0.4%. 22. A process for the production of a cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and aprotective coating around said core of calcium carbonate; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount less than 30%, based on thecombined weight of the cement and the pulverulent composition; which process comprises (a) mixing together at least part of the Portland cement and at least part of the water; and (b) adding the pulverulent composition to the mixture produced in step (a). 23. A pulverulent composition which is useful as an expansion additive for hydratable cement compositions; which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and a protectivecoating around said core of calcium hydroxide; which exhibits an Expansion Factor of at least 0.06%. 24. A pulverulent composition which is useful as an expansion additive for hydratable cement compositions; which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and a protectivecoating around said core of calcium hydroxide and calcium carbonate; which exhibits an Expansion Factor of at least 0.06%. 25. A cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and a protective coating around saidcore of calcium hydroxide; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount less than 30%, based on the combined weight of the cementand the pulverulent composition.

26. A cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and a protective coating around saidcore of calcium hydroxide and calcium carbonate; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount less than 30%, based on the combinedweight of the cement and the pulverulent composition. 27. A process for the production of a cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and aprotective coating around said core of calcium hydroxide; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount less than 30%, based on thecombined weight of the cement and the pulverulent composition; which process comprises (a) mixing together at least part of the Portland cement and at least part of the water; and (b) adding the pulverulent composition to the mixture produced in step (a). 28. A process for the production of a cement composition comprising a Portland cement, water and a pulverulent composition which consists essentially of particles having a size less than 100 microns and comprising a core of calcium oxide and aprotective coating around said core of calcium hydroxide and calcium carbonate; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor of at least 0.06%; said pulverulent composition being present in amount lessthan 30%, based on the combined weight of the cement and the pulverulent composition; which process comprises (a) mixing together at least part of the Portland cement and at least part of the water; and (b) adding the pulverulent composition to the mixture produced in step (a). 29. A composition according to claim 23 which comprises 36 to 92% by weight of calcium oxide and wherein said protective coating consists essentially of calcium hydroxide. 30. A composition according to claim 29 which comprises 50 to 84% by weight of calcium oxide. 31. A composition according to claim 23 or 24 wherein said particles have a size of 1 to 100 microns. 32. A composition according to claim 31 wherein said particles have a size of 5 to 60 microns.

33. A composition according to claim 23 or 24 having an Expansion Factor of 0.07 to 0.4%. 34. A cement composition according to claim 25 or 26 which comprises 3 to 10% of said pulverulent composition, based on the combined weight of the cement and the pulverulent composition. 35. A cement composition according to claim 25 or 26 wherein said pulverulent composition comprises 36 to 92% by weight of calcium oxide and said protective coating consists essentially of calcium hydroxide. 36. A cement composition according to claim 35 wherein said pulverulent composition comprises 50 to 84% by weight of calcium oxide. 37. A cement composition according to claim 25 or 26 wherein said particles have a size of 1 to 100 microns. 38. A cement composition according to claim 37 wherein said particles have a size of 5 to 60 microns. 39. A composition according to claim 25 or 26 wherein said pulverulent composition has an Expansion Factor of 0.07 to 0.04%. Description: BACKGROUND OF THE INVENTION 1. Field of the Invention This invention relates to expansive cement compositions. 2. Summary of the Prior Art Compositions which contain hydratable cements (generally Portland cement) tend to shrink on drying. It is known to counteract this undesirable tendency by including in the composition an expansion additive which expands at the correct stageduring the drying of the composition [see for example Gen. Civ. 109, 285 (1936, H. Lossier), U.C. SESM Report No. 72-13 (1973, G. Komendant et al.), Plan. Bau 2, 351 (1951, H. Bickenbach), Concrete Technology and Practice, 3rd Edn, 359 (1969, W. H.Taylor) and U.S. Pat. Nos. 3,519,449, 3,649,317, 3,801,339, 3,883,361, 3,884,710, 3,947,288 and 4,002,483]. In many of the known expansion additives, the active ingredient is free lime (CaO), also known as quick lime, which expands by about 100% involume when hydrated, the other ingredients being present to prevent the hydration taking place until the desired stage in the drying of the cement composition. Free lime is one of the ingredients of Portland cement, but an excess of it is normallyregarded as undesirable, and the addition of free lime to cement compositions is avoided. Free lime does not serve as an expansion additive because it is very rapidly hydrated and therefore expands far too early in the setting process to be effective. It is also known (see for example U.S. Pat. No. 3,106,453

and German Pat. No. 1,216,753) to subject free lime to certain specific treatments with water and/or CO.sub.2 to produce a lime which hydrates at a slower rate, but the products obtained areineffective as expansion additives. Likewise the products obtained by exposing lime to the atmosphere, which products contain Ca(OH).sub.2 and CaCO.sub.3, are ineffective as expansion additives. It has also been proposed in U.S. Pat. No. 1,732,409 tomake quick-setting lime products which contain a high proportion of lime, at least some of the lime being in the form of a product obtained by heating finely divided lime in an atmosphere containing CO.sub.2 at a temperature of 500.degree. to850.degree. C. until the increase in weight is 3 to 40% based on the weight of the starting material. SUMMARY OF THE INVENTION We have now discovered that excellent expansion additives for cement compositions can be made by heating particles comprising CaO in the presence of water vapor and/or CO.sub.2, or simply by heating particles comprising CaO and a sufficientamount of calcium hydroxide which has been formed by absorption of water vapor. In one aspect, the invention provides a pulverulent composition which is useful as an expansion additive for hydratable cement compositions; which consists essentially of particles (a) having a size less than 250 microns and (b) comprising a coreof calcium oxide and a protective coating around said core of calcium hydroxide or calcium carbonate or both; which contains 36 to 95% by weight of calcium oxide; and which exhibits an Expansion Factor (as hereinafter defined) of at least 0.06%. Suchcompositions can be made by a process which comprises heating particles (a) having a size less than 250 microns and (b) comprising at least 36% by weight of CaO, with the remainder (if any) preferably consisting essentially of Ca(OH).sub.2 and/orCaCO.sub.3, the particles being maintained at a temperature of at least 100.degree. C. and, if the particles initially contain less than 8% by weight of Ca(OH).sub.2, the particles being maintained at a temperature of 100.degree. to 400.degree. C. inan atmosphere which comprises water vapor or at a temperature of 350.degree. to 850.degree. C. in an atmosphere which comprises carbon dioxide, said heating being carried out under conditions such that the product contains 36 to 95% by weight of CaOand exhibits an Expansion Factor (as hereinafter defined) of at least 0.06%. The invention also includes the use of such lime-containing compositions as expansion additives in hydratable cement compositions, in particular cement compositions comprising a Portland cement, water and a lime-containing composition as definedabove in amount less than 30%, based on the combined weight of the cement and the limecontaining composition. The invention further includes a method of making a shaped article which comprises casting a cement composition as defined above. DETAILED DESCRIPTION OF THE INVENTION Except where otherwise noted, parts and percentages in this specification are by weight. The Expansion Factor of a lime-containing composition is defined herein as the % increase in length of a test prism (test bar) which has been cured in waterfor 7 days after

casting, according to ASTM C490-70 and C157-69T, from a mortar prepared according to ASTM C305-65 and containing 0.1 part of the lime-containing composition, 0.9 part of Type III Portland cement, 2 parts of No. 20 washed sand, and 0.4part of water. The Expansion Factor of the treated limes of the invention is at least 0.06%, generally 0.07 to 2.5%, preferably 0.07 to 0.4%. The Expansion Factor is a measure of the effectiveness of the protective coating in delaying access of waterto the calcium oxide core, and inter alia also of the calcium oxide content of the composition. The larger is the particle size of the treated lime, the greater is its lime content for a given thickness of the protective coating. For this reason the particle size is generally at least 1 micron. On the other hand, if particles of sizegreater than 250 microns are used, they cause an undesirable reduction in strength. Larger particle sizes also make it more difficult to obtain a uniform distribution of the treated lime in a cement composition. The treated limes preferably have aparticle size of 1 to 100 microns, especially 5 to 60 microns. The starting materials used in this invention comprise at least 36% of CaO, with the remainder preferably consisting essentially of Ca(OH).sub.2 and/or CaCO.sub.3. The term "consisting essentially of" is used to indicate that the startingmaterial may contain minor amounts, generally less than 10%, of other ingredients which do not prevent formation of an effective protective coating, e.g. impurities such as silicates and oxides of iron, magnesium and aluminum. When the treated lime isprepared by heating the starting material in the presence of water vapor and/or carbon dioxide, the starting material preferably contains at least 80%, especially at least 90%, of CaO. Thus dolomitic limes, which contain lesser quantities of CaO, arenot preferred starting materials. When the atmosphere comprises water vapor, but no carbon dioxide, the temperature should be 100.degree. to 400.degree. C., preferably 150.degree. to 400.degree. C., especially 150.degree. to 250.degree. C. For a given residual CaO content,the higher the temperature of the treatment, the higher the expansion activity of the product. At temperatures below 100.degree. C., the lime becomes hydrated but no effective protective coating is formed. At temperatures above 400.degree. C.,calcium hydroxide is not formed, or if present initially is at least partially decomposed. To ensure an adequate protective coating, heating in an atmosphere comprising water vapor should preferably be continued for a time such that the particlesincrease in weight by at least 2%, e.g. 2 to 16%, particularly 4 to 12%, especially 5 to 11%, based on the weight of the starting material. For a starting material which consists essentially of CaO, these weight increases will result in a product whichcontains at most 92%, e.g. 92 to 36%, particularly 84 to 50%, especially 84 to 50% of residual active lime as the core and a protective coating which consists essentially of calcium hydroxide. When the atmosphere comprises carbon dioxide, a temperature of 350.degree. to 850.degree. C. is needed in order to convert the CaO to CaCO.sub.3. At temperatures below 350.degree. C. the reaction between CaO and CO.sub.2 is very slow, and noeffective protective coating is formed. At temperatures above 850.degree. C. CaCO.sub.3 is not formed or if present initially is decomposed. To ensure an adequate

protective coating, heating in an atmosphere comprising CO.sub.2 should preferably becontinued for a time such that the particles increase in weight by at least 2%, e.g. 2 to 28%, particularly 4 to 16%, especially 5 to 16%, based on the weight of the starting material. For a starting material which consists essentially of CaO, theseweight increases will result in a product which contains at most 95%, e.g. 36 to 95%, particularly 64 to 91%, especially 64 to 89%, of residual active lime as the core and a protective coating which consists essentially of calcium carbonate. It is also possible to form a protective coating comprising both CaCO.sub.3 and Ca(OH).sub.2 by heating the starting material in an atmosphere comprising CO.sub.2 and water, preferably at a temperature of 350.degree. to 400.degree. C. Undersuch conditions, the heating should preferably be continued for a time such that the particles increase in weight by 4 to 14%, especially 5 to 12%, these weight increases corresponding to (for a starting material which consists essentially of CaO) finalproducts containing 91 to 40%, especially 80 to 50% of residual active lime. As noted above, limes which have been partially hydrated and/or carbonated at temperatures below 100.degree. C. do not have an effective protective coating thereon and are not useful as expansion additives. Useful expansion additives can beobtained from such partially hydrated and/or carbonated limes by using them as starting materials in the processes described above which involves heating in an atmosphere comprising water vapor and/or CO.sub.2, provided that the product has an adequateresidual calcium oxide content. However, we have discovered that partially hydrated limes can also be converted into useful expansion additives by heating them at temperatures of at least 100.degree. C., preferably less than 500.degree. C., e.g.200.degree. to 400.degree. C., in an atmosphere which contains no water vapor or carbon dioxide; in this embodiment of the invention, the starting material can comprise 8 to 64%, preferably 16 to 50%, of Ca(OH).sub.2, with the balance preferablyconsisting essentially of CaO. For a given CaO content, the higher is the temperature of the treatment and the longer its duration, the greater is the expansion activity of the product. Treated limes which already have some expansion activity can be further treated by one of the processes described above to make them more active. However, it should be noted that such further treatment, especially when carried out in anatmosphere comprising water vapor at relatively low temperatures, e.g. below 200.degree. C., especially below 150.degree. C., can reduce the residual CaO content of the treated lime to an extent such that activity is reduced. The heat treatments described above can be carried out in any convenient way which does not involve grinding of the particles which would disrupt the protective coating. For small quantities of product, the starting material can simply be spreadout on a pan. For larger quantities fluidised bed procedures may be used. The novel expansion additives can be incorporated into hydratable cement compositions in any convenient way, but we have found that optimum expansion activity is obtained if the additive is added to a mixture of at least part of the cement and atleast part of the

water. The cement compositions can of course contain sand, aggregate and other conventional additives, but the presence of additional quicklime, hydrated lime or carbonated lime is preferably avoided. The novel additives exhibit adelayed reactivity with water, measured by ASTM Test C110-71, Section 9, of up to 100 minutes, but this does not account for the fact that in cement compositions they show a much greater delayed reactivity. It is thought that this unexpectedly delayedreactivity is due to some form of interaction between the protective coating and components present in the cement slurry. The amount of the additive used will depend inter alia on its CaO content and the strength and expansion desired in the finalproduct, and will generally be 3 to 30%, preferably 3 to 10%, especially 5 to 7%, based on the combined weight of the hydratable cement and the additive. Especially when the treated limes have been thoroughly treated, their presence has little or nodeleterious effect on such important properties of the cement composition as water requirements, consistency, open time, and tendency to "false set". The invention is illustrated in the following Examples. The expansion activities of the varioussamples prepared in Examples 1-7 are shown in Table 1, which shows the % increase in length of test prisms (i.e. test bars) which (except where otherwise noted) were cured in water after casting according to ASTM C490-70 and C157-69T from mortarsprepared according to ASTM C305-65 and containing 1 part of a mixture of the treated lime sample and a Type III Portland cement (Norcem Rapid), 2 parts of No. 20 washed Monterey sand, and 0.4 parts of water. The amount of the treated lime sample in thetest prism and the number of days of curing are given in parentheses after each % expansion figure. The samples with a "C" after the sample number in Table 1 are comparative examples, not in accordance with the invention. EXAMPLE 1 Portions of CaCO.sub.3 having a particle size less than 44 microns were heated in a muffle oven at 1000.degree. C. for the time necessary to produce samples comprising CaO in amount 100% (Sample 1), 93.6% (Sample 2), 90.2% (Sample or 72.3%(Sample 4). Portions of Sample 1 were heated in a muffle oven at 580.degree. C. in an atmosphere of CO.sub.2 until the CaO content was 94.5% (Sample 5), 89.8% (Sample 6) and 80.5% (Sample 7). Portions of a commercially available lime (Flintkote)having a particle size less than 44 microns were heated in a muffle oven at 580.degree. C. in an atmosphere of CO.sub.2 until the CaO content was 94.8% (Sample 8), 87.5% (Sample 9) or 74.8% (Sample 10). EXAMPLE 2 Portions of a commercially available lime (Flintkote) containing about 95% CaO, with the remainder Ca(OH).sub.2 and a trace of CaCO.sub.3, and having a particle size less than 44 microns, were exposed to water vapor at 22.degree. C. (Sample 11),60.degree. C. (Sample 12), 100.degree. C. (Sample 13), 150.degree. C. (Sample 14) or 200.degree. C. (Sample 15) until the CaO content had been reduced to the level shown in Table 1. EXAMPLE 3

A treated lime having expansion activity, containing 60% CaO, 27% Ca(OH).sub.2 and 13% CaCO.sub.3, and having a particle size less than 100 microns (Sample 16) was exposed to water vapor at 22.degree. C. (Sample 17), 60.degree. C. (Sample 18),100.degree. C. (Sample 19), 150.degree. C. (Sample 20) and 200.degree. C. (Sample 21) until the CaO content was 31% (Sample 17) or 40% (Samples 18-21). EXAMPLE 4 A treated lime having expansion activity, containing 85% CaO, 5% Ca(OH).sub.2 and 10% CaCO.sub.3, and having a particle size less than 100 microns, was exposed to water vapor at 110.degree. C. until the CaO content had been reduced to 64%(Sample 22). Portions of Sample 22 were heated at 110.degree. C. for 180 min. (Sample 23), at 200.degree. C. for 15 min. (Sample 24), at 200.degree. C. for 60 min. (Sample 25), at 300.degree. C. for 15 min. (Sample 26), at 340.degree. C. for 30min. (Sample 27), at 340.degree. C. for 50 min. (Sample 28), at 400.degree. C. for 90 min. (sample 29) and at 455.degree. C. for 15 min. (Sample 30). Samples 29 and 30 lost 2% and 3% in weight, respectively, during the heat treatment, due todecomposition of calcium hydroxide. The same treated lime was exposed to water vapor at 200.degree. C. until the CaO content had been reduced to 64% (Sample 31). A commercially available lime (Diamond Springs) consisting essentially of CaO and havinga particle size less than 100 microns was exposed to water vapor at 25.degree. C. until the CaO content was 64% (Sample 32). A portion of Sample 32 was heated at 340.degree. C. for 60 minutes (Sample 33). Another sample of the same commerciallyavailable lime was exposed to water vapor at 200.degree. C. until the CaO content had been reduced to 58% (Sample 34). EXAMPLE 5 Portions of a commercially available lime (Flintkote) as used in Example 2 were exposed to water vapor at 260.degree. C. until the CaO content was 85% (Sample 35) or 58% (Sample 36). These Samples were made into mortars as described aboveexcept that a Type 1-2 cement (Kaiser Permanente) was used and the amount of water was 0.36 part. A Control mortar containing no added lime was also prepared. The prisms cast from these mortars were cured both in water and in 50% relative humidity air. EXAMPLE 6 A treated lime having expansion activity, containing 84.5% CaO, 6.2% Ca(OH).sub.2 and 9.3% CaCO.sub.3, and having a particle size less than 100 microns was used to prepare mortars containing 0.9 part of a Type 1-2 cement (Kaiser Permanente), 0.1part of the treated lime, 2 parts of No. 20 washed Monterey sand and 0.36 part of water. The mortars were prepared by three different methods. Method 1 (ASTM C305-65) A. The treated lime and cement were dry mixed. B. The water was added and mixed at

slow speed for 30 seconds. C. The sand was added and slow speed mixing continued for another 30 seconds. D. The mixer was increased to medium speed for 30seconds. E. Pause for 90 seconds. F. Mix at medium speed for 1 min. Method 2 A. Mix the cement and water for 30 seconds. B. Add the treated lime and mix for 30 seconds. C-F. Same as in Method 1. Method 3 A. The treated lime and water were mixed for 30 seconds, then the cement was added and mixed for 30 seconds. D-F. Same as in Method 1. The % expansions of prisms cast from these mortars are shown in Table 1 under Samples 37, 38 and 39 (Methods 1, 2 and 3 respectively). EXAMPLE 7 A commercially available lime (Flintkote) was fractionated into a first fraction in which the particles were 45 to 425 microns in size and a second fraction in which the particles were less than 45 microns in size. Both fractions were heated at950.degree. C. for 1 hour and then treated with water vapor at 200.degree. C. until the CaO content was 69.5% (first fraction, Sample 40) or 73% (second fraction, Sample 41). TABLE 1 _______________________________________________________________________ ___ Ex. Sample No. No. % CaO % Expansion (parts of sample in mortar, days) _______________________________________________________________________ ___ 1 1(C) 100 0.042 (0.1, 7) 2(C) 93.6 0.031 " 3(C) 90.2 0.031 " 4(C) 72.3 0.029 " 5 94.5 0.234 " 6 89.8 0.188 " 7 80.5 2.3 " 8 94.8 0.077 " 9 87.5 0.150 " 10 74.8 0.757 " 2 11(C) 49 0.037 " 12(C) 53 0.043 " 13 53 0.125 " 14 57 0.212 " 15 530.347 " 3 16 60 0.61 (0.1, 14) 0.067 (0.04, 16) 17(C) 31 0.134 " 0.046 " 18 40 0.186 " 0.045 " 19 40 0.330 " 0.044 " 20 40 0.413 " 0.058 " 21 40 2.8 " 0.092 " 4 22 64 0.077 (.08, 1) 0.147 (.08, 7) 0.156 (.08, 14) 23 64 0.068 " 0.131 " 0.138 " 24 64 0.065 " 0.104 " 0.110 " 25 64 0.211 " 0.332 " -- " 26 64 0.063 " 0.212 " 0.218 " 27 64 0.025 " 0.556 " 0.560 " 28 64 0.020 " 0.651 " 0.656 " 29 72 0.043 " 0.394 " 0.401 " 30 76 0.084 " 0.358 " 0.363 " 31 64 0.020 "0.619 " 0.625 " 32(C) 64 0.019 " 0.022 " 0.022 " 33 64 0.176 " 0.632 " 0.635 " 34 58 0.155 " 0.347 " 0.352 " 5 35 83 0.10 (.07, 1) 0.21 (.07, 7) 0.23 (.07, 40) 36 58 0.02 " 0.20 " 0.22 " Control 0.01 (0, 1) 0.015 (0, 7) 0.02 (0,40) 35 83 -0.01 (.07, 1) 0 (.07, 7) 0.47 (.07, 40) cured 36 58 -0.01 " -0.02 " -0.025 " at Control -0.01 (0, 1) -0.035 (0, 7) -0.055 (0, 40) 50% R.H. 6 37 84.5 0.331 (.1, 11) 38 84.5 0.409 " 39 84.5 0.246 " 7 40 69.5 0.09 (.06, 2) 0.15 (.06, 3) 0.165 (.06, 10) 0.18 (.06, 13) 41 73 0.03 " 0.04 " 0.10 " 0.12 " 40 69.5 0.20 (.1, 2) 1.16 (.1, 5) 1.21 (.1, 6) 1.24 (.1, 18) 41 73 0.05 " 0.60 " 0.72 " 0.75 " _______________________________________________________________________ ___

EXAMPLE 8 A commercially available lime (Cementa) was treated with water vapor at 250.degree. C. until the CaO content was 80%, and was then fractionated into a first fraction in which the particles were less than 74 microns in size and a second fractionin which the particles were 75 to 125 microns in size. The first fraction was used to prepare mortars containing 0.954 part of a Type 1 cement (Norcem PC 300), 0.046 part of the treated lime, 3 parts of No. 20 washed Monterey sand, and 0.5 part ofwater. The second fraction was used to prepare mortars in the same way except that 0.96 part of the cement and 0.04 part of the treated lime were used. The compressive strength of prisms cast from these mortars was measured after curing in water forthe indicated number of days, and is shown in Table 2. Table 2 ______________________________________ Compressive Strength (kg/cm.sup.2) after Particle Size 1 day 3 days 7 days ______________________________________ 0-74 105.5 240 300 75-125 103.5 195 248 ______________________________________ EXAMPLE 9 Portions of a commercially available lime (Flintkote) as used in Example 2 were treated with water vapor at 200.degree. C. until the CaO content was 91% (Sample 42), 70% (Sample 43) or 49.5% (Sample 44) or with CO.sub.2 at 580.degree. C. untilthe CaO content was 88% (Sample 45) or 68% (Sample 46). Part of Sample 45 was treated with water vapor at 200.degree. C. until the CaO content was 54% (Sample 47). Cement pastes were made by dry mixing 0.1 part of the sample and 0.9 part of a Type 1-2cement (Kaiser Permanente); water (0.36 part) was then added and the mixture was mixed in a Hobart Mixer for 20 seconds at low speed followed by 20 seconds at medium speed. The pastes were placed in a Halliburton Consistometer, and their consistencies,expressed in Halliburton Consistency units, are shown in Table 3 below (which also shows results for a paste containing 1 part of the Type 1-2 cement and no treated lime). Table 3 _______________________________________________________________________ ___ Consistency after indicated time (mins.) Sample No. 0 20 40 60 80 100 120 140 160 180 200 220 240 _______________________________________________________________________ ___ 42 .sup.x 100 45 45 47 49 53 64 82 100 43 40 40 38 38 40 42 43 45 48 52 58 63 70 44 32 33 32 34 36 38 40 42 45 50 56 63 75 45 .sup.x 100 42 40 42 42 43 47 5258 63 70 79 91 46 56 50 47 45 45 46 46 50 51 53 56 60 67 47 55 38 36 36 36 36 38 40 41 43 45 48 54 Control 52 38 36 36 36 36 36 37 38 39 40 42 43 _______________________________________________________________________ ___ .sup.x shows"false set"***** Claim: We claim:

1. An expansive cement composition made of argillaceous and calcareous materials and containing from about 1 to about 70% by weight thereof of free CaO and from about 2 to about 7% byweight thereof of MgO uniformly distributed therein, with the remaining portion of said materials being primarily tricalcium silicate and containing substantially no dicalcium silicate. 2. The expansive cement composition of claim 1 wherein said MgO exists in the form of a solid solution and of crystal particles, and wherein at least 75% of said crystal particles are no larger than about 10 microns. 3. The expansive cement composition of claim 1 having a particle size no larger than about 80 mesh (Tyler) and a fineness no greater than about 4700 cm.sup.2 /gm. 4. The expansive cement composition of claim 3 having from about 5 to about 30% by weight thereof of free CaO. 5. The expansive cement composition of claim 3 having a particle size that passes 100 mesh (Tyler) and a fineness no greater than 3700 cm.sup.2 /gm. 6. The expansive cement composition of claim 5 having a fineness between about 2000 and 3700 cm.sup.2 /gm. 7. An expansive cement composition made by burning mixtures of magnesia supplying materials and argillaceous and calcareous reactant materials containing an excess of CaO relative to silica under conditions to thereby convert substantially allof the silica into tricalcium silicate and uniformly distribute said magnesia in said composition, said composition containing from about 1 to about 70% by weight of free CaO and from about 2 to about 7% by weight of MgO with the remaining portion beingprimarily tricalcium silicate and containing substantially no dicalcium silicate. 8. The expansive cement composition of claim 7 further comprising a liquid phase which forms during the burning thereof which is in the range of from about 15 to about 35% by weight of said composition. 9. The expansive cement composition of claim 8 wherein said magnesia is present in said composition in solid solution in said liquid phase and the silicate phase thereof and in crystals wherein at least about 75% of the crystals have a size nolarger than about 10 microns. 10. The expansive cement composition of claim 7 having a particle size no larger than about 80 mesh (Tyler) and a fineness no greater than about 4700 cm.sup.2 gm. 11. The expansive cement composition of claim 10 containing from about 5 to about 30% by weight of free CaO. 12. The expansive cement composition of claim 10 containing from about 5 to less than

about 40% by weight of free CaO. 13. The expansive cement composition of claim 10 having a particle size that passes 100 mesh (Tyler) and a fineness no greater than 3700 cm.sup.2 /gm. 14. The expansive cement composition of claim 13 having a fineness of between 2000 and 3700 cm.sup.2 /gm. 15. An expansive cement composition made by burning mixtures of magnesia supplying materials and argillaceous and calcareous reactive materials containing an excess of CaO relative to silica and burned under conditions to thereby convertsubstantially all of the silica into tricalcium silicate, and uniformly distribute MgO therewithin, said composition containing from about 1 to about 70% by weight thereof of free CaO and a minor but effective amount of MgO which is sufficient toincrease the ability of the resulting composition when hydrating to grab steel reinforcement members therewithin, and the remaining portion of said composition being primarily tricalcium silicate and containing substantially no dicalcium silicate. 16. An expansive cement comprising a mixture of: a. a hydratable cement; and b. an expansive composition made from argillaceous and calcareous materials and containing from about 1 to about 70% by weight of free CaO, and from about 2 to about 7% by weight thereof of MgO uniformly distributed therein, with the remainingportion being primarily tricalcium silicate and containing substantially no dicalcium silicate. 17. The expansive cement of claim 16 wherein said hydratable cement is Portland cement. 18. The expansive cement of claim 17 wherein said mixture is made by intergrinding a Portland cement clinker and a clinker of said expansive cement composition. 19. The expansive cement of claim 16 wherein said expansive composition contains from about 5 to less than about 40% by weight of CaO. 20. The expansive cement of claim 19 wherein said expansive composition contains from about 5 to about 30% by weight of CaO. 21. The expansive cement of claim 16 wherein said expansive composition has a particle size no larger than about 80 mesh (Tyler) and a fineness no greater than about 4700 cm.sup.2 /gm. 22. The expansive cement of claim 21 wherein said expansive composition has a particle size that passes 100 mesh (Tyler) and a fineness no greater than 3700 cm.sup.2 /gm.

23. The expansive cement of claim 22 wherein said expansive composition has a fineness of between 2000 and 3700 cm.sup.2 /gm. 24. The expansive cement of claim 16 containing from about 1 to about 99% by weight thereof of Portland cement and from about 99 to about 1% by weight thereof of said expansive composition. 25. The expansive cement of claim 24 wherein said expansive composition contains from about 5 to less than about 40% by weight of said free CaO. 26. The expansive cement of claim 25 wherein said expansive composition contains from about 5 to about 30% by weight of said free CaO. 27. The expansive cement of claim 16 wherein said hydratable cement is a pozzolanic cement. 28. A method of producing an expansive cement composition comprising: a. introducing magnesia supplying materials and argillaceous and calcareous reactant materials to the feed end of a kiln, said calcareous reactant materials containing sufficient CaO forming materials to completely react with the argillaceousmaterials in the kiln and to provide an excess of unreacted CaO in the burned clinker in the range of from about 1 to about 70% by weight thereof, said magnesia supplying materials supplying said MgO in an amount in the range of from about 2 to about 7%by weight of said composition; b. burning said materials in a burning zone in said kiln at a temperature in the range of from about 2500.degree. F to about 2900.degree. F and for a sufficient time to convert substantially all of the silicate in said argillaceous materials totricalcium silicate and yield substantially no dicalcium silicate, but yield from about 1 to about 70 weight percent of free CaO in the resulting burned clinker, and to uniformly disperse said MgO in the resulting burned clinker; and c. grinding said clinker to produce a powdered hydraulic expansive cement composition. 29. The method of claim 28 wherein said magnesia supplying materials and said argillaceous and calcareous reactant materials fed to said kiln have a particle size such that no less than about 97% thereof passes through a 50 mesh (Tyler) screen,and no less than about 80% thereof passes through a 200 mesh (Tyler) screen. 30. The method of claim 28 wherein a liquid phase forms during said burning of said materials and is in the range of from about 15 to about 35 percent by weight of said resulting burned clinker. 31. The method of claim 30 wherein said burning results in said MgO being uniformly distributed throughout said composition in the form of a solid solution and in the form of crystals and wherein at least 75% of said crystals have a size of nomore than about 10

microns. 32. The method of claim 31 wherein said grinding yields particles having a size no larger than about 80 mesh (Tyler) and a fineness no greater than about 4700 cm.sup.2 /gm. 33. The method of claim 32 wherein said free CaO is present in said composition in the range from about 5 to less than about 40% by weight thereof. 34. The method of claim 33 wherein said CaO is present in said composition in an amount in the range of from about 5 to about 30% by weight thereof. 35. The method of claim 33 wherein said clinker is ground to a particle size that passes 100 mesh (Tyler) and a fineness no greater than 3700 cm.sup.2 /gm. 36. The method of claim 35 wherein said clinker is ground to a fineness of between 2000 and 3700 cm.sup.2 /gm. Description: BACKGROUND OF THE INVENTION This invention relates to expansive cements. In another aspect, this invention relates to novel expansive compositions which have an improved ability to stress steel reinforcing members which are positioned within plastic concretes containingsuch compositions. Portland cement concrete is one of the most widely used construction materials; however, it possesses the inherent characteristic of shrinkage on drying. This drying shrinkage can cause cracks in the resulting concrete. Cracks of varying sizecan generally be found in most reinforced concrete structures. In essence, when shrinkage of the curing concrete occurs, the concrete is placed under tension and if the tensile forces produced exceed the tensile strength of the material, a crack isproduced. As a consequence, expansive cements were developed so that concrete made from these cements would undergo a volumetric expansion during the curing cycle. The forces generated by this volumetric expansion of the concrete are harnessed in a mannerthat makes it possible to utilize these forces to control the cracking of the concrete caused by the drying shrinkage. One technique for utilization of the volumetric expansion of concrete as a means of controlling cracking in the concrete is known as shrinkage compensation. This technique requires that the rate and amount of expansion in the concrete occursimultaneous with and in the same magnitude as the drying shrinkage. This balance of forces would eliminate the tensile stresses which normally give rise to cracking in concrete. In practice, such a match of expansion and drying shrinkage to achieve anonshrinking or dimensionally stable concrete has been difficult to achieve and control as a great variety of materials and widely varying ambient conditions of temperature and relative humidity are encountered in concrete practice. Another technique for utilization of volumetric expansion of concrete as a means of controlling cracking in concrete is to capture the force generated by the concrete

expansion in reinforcing steel in such a manner that permits use of theseforces to oppose the tensile stress generated by the drying shrinkage. This technique is referred to as stress induction. Reinforcing steel is placed in the fresh plastic concrete. During the setting reaction, the concrete forms a bond to or grabs thereinforcing steel. As the concrete expands, it carries the steel with it and thus creates a tensile stress in the steel. These tensile stresses in the reinforcing steel place the concrete under restraint and accordingly when expansion occurs, it is ofa lesser magnitude and of a more controlled rate, and produces a stronger, denser concrete. This chemical stressing of the steel is analogous to mechanical prestressing of steel in concrete members in which a steel cable is placed and held in tensionuntil the concrete has set and achieved a certain minimum strength. The cable is then released but within the concrete member it remains in a state of tension and places the concrete in compression. In both cases, the tensile stresses within the steel reinforcement members exert compressive stresses on the concrete. The tensile stresses generated in concrete when drying shrinkage occurs are neutralized or balanced (because the concrete isin compression) until the magnitude of the drying shrinkage is greater than the magnitude of the tensile stresses of the steel reinforcement. It has been long assumed that as long as the measured expansion of the concrete exceeds any later dryingshrinkage, the embedded steel reinforcement will maintain an overall state of compression within the concrete and no cracking will occur. It has recently been found that all concrete expansion cannot be directly related to the induced tensile stresses within the reinforcing steel. It has been found that some slippage between the concrete and the steel occurs during expansion andtherefore only a portion of the concrete expansion is utilized to stretch the steel. For example, in a case where complete bonding occurs between the expansive concrete and the steel reinforcement bar, the measured expansion of the concrete is equal to the tensile strain on the steel. In this situation, the steel will keep theconcrete in compression and for all amounts of drying shrinkage less than the maximum expansion prevent shrinkage cracking due to tensile forces set up by the decreasing volume. At the other extreme, if no bond occurs between the concrete and the steel,the steel bar remains slack and exerts no compressive stresses on the concrete. Subsequent drying shrinkage will place the concrete in tension and cause cracking. In this respect, the presence of the steel reinforcement contributes only to structuralrequirements and nothing to crack prevention. In actual practice, expansive cement concrete will exhibit behaviors somewhere between these two extremes. Thus, there is needed an effective expansive cement whose degree of expansion can be easily controlled but yet will bond or grab steelreinforcement members more effectively such that a greater degree of concrete expansion can be directly related to the induced tensile stresses within the reinforcing member. The earliest and most common type of expansive cements are the so called "Type K" compositions which are based upon a sulfoaluminate expansive mechanism. Examples of such compositions are disclosed in U.S. Pat. No. 3,251,701 and U.S. Pat. No. 3,303,037.

The Type K expansive cements are produced by burning of a special clinker containing the proper amount of tetracalcium trialuminate sulfate (C.sub.4 A.sub.3 S). It is noted in cement nomenclature that C = CaO; S = SiO.sub.2 ; A =Al.sub.2 O.sub.3 ; F = Fe.sub.2 O.sub.3 ; and S = SO.sub.3. After hydration, this material generally forms ettringite and is accompanied by a concurrent increase in volume. Another expansive cement composition based upon a sulfoaluminate expansion mechanism is the Type S expansive cement, which is a Portland cement containing a large amount of C.sub.3 A and modified by an excess of calcium sulfate above the usualamount found in Portland cement. Still another conventional expansive cement is the "Type M" cement which is either a mixture of Portland cement, calcium aluminate cement, and calcium sulfate, or an interground product made from the above respectivecement clinkers. The above expansive cement compositions have met with only a limited success due to the fact that the amount and rate of expansion have been difficult to control within acceptable parameters. Furthermore, even when such cements have exhibitedcontrollable expansion, the poor ability to bond to or grab reinforcing steel has prevented the successful placement of expansive cement concrete. Recently, expansive cements have been developed based upon calcium oxide reactions. Such expansive cements typically are formed by burning a clinker of argillaceous and calcareous materials to a degree such that the silicate is in the form ofC.sub.3 S and substantially no C.sub.2 S exists in the composition, and an excess of free uncombined calcium oxide exists in the composition together with the other components in the form of a liquid phase containing C.sub.4 AF and C.sub.3 A.Compositions of this type are disclosed in U.S. Pat. No. 3,785,844 and copending patent application Ser. No. 404,934 filed Oct. 10, 1973, now U.S. Pat. No. 3,884,710. These expansive cements can be controlled to yield a wide range of expansionallowing them to be more effective than the sulfoaluminate cements for both shrinkage compensating and for stress inducing applications. However, even with this improved formulation, it was found that only a small percentage of the expansive potentialwas being used in concrete because of a partial bond to or grab of the reinforcing steel. SHORT STATEMENT OF THE INVENTION In accordance with the invention, it has been discovered that the addition of a minor but effective amount of MgO to a free lime expansive cement composition increases the ability of the composition when hydrating in a mortar paste or concretemixture to stress steel reinforcement members therewithin. Thus, an improved expansive cement composition is provided which exhibits improved ability to stress steel reinforcement members during the expansion process which comprises an effective amountof a free CaO as the expansive agent (generally from about 1 to about 70% by weight of the composition), and an effective amount of MgO (generally from about 2 to about 7% by weight thereof) uniformly distributed within the composition for increasing theability of the concrete to stress the reinforcing member, and wherein the silicate in the composition

is predominantly in the form of C.sub.3 S with substantially no C.sub.2 S. According to one embodiment of the subject invention, the expansive cement composition of the subject invention is produced by a process which includes introducing an intimate mixture of argillaceous and calcareous reactant materials, andmagnesia (magnesium oxide) containing material to the feed end of a kiln, said materials containing magnesium oxide in the range of from about 2 to about 7% by weight thereof and sufficient calcium oxide or materials which will form calcium oxide insitu, to fully stoichiometrically react with the remaining materials in the kiln and provide an excess of unreacted calcium oxide in the burned clinker in the range of from about 1 to about 70 weight percent thereof; and thereafter burning the reactantmaterials in a burning zone in the kiln for a sufficient time and temperature to completely react the silicate material with calcium oxide such that substantially all of the silicate is present in the form of tricalcium silicate and the compositioncontains substantially no dicalcium silicate, and thereafter grinding the clinker to produce an expansive cement composition. According to still another embodiment of the invention, an expansive cement is provided which comprises a mixture of the above described expansive cement composition and a hydratable cement such as Portland cement. DETAILED DESCRIPTION OF THE INVENTION Thus, we have discovered that the addition of a small amount of MgO to a free lime expansive cement composition will result in that composition having an improved ability to grab steel reinforcing members. The free line cement composition mustbe one in which substantially all of the silicate present in the composition is in the form of C.sub.3 S and there is substantially no C.sub.2 S therewithin (no more than about 5% by weight of C.sub.2 S therewithin). The composition must contain aneffective amount of free lime or uncombined lime which will impart expansive properties thereto. Generally, from about 1 to about 70 weight percent of the composition must be free CaO. The composition can be manufactured from conventional ingredientsthat are normally fed to a Portland cement kiln. Therefore, constituents other than free lime and tricalcium silicate and MgO can be present within the composition. For example, the composition will contain aluminate generally in the form of tricalciumaluminate (C.sub.3 A) and iron generally in the form of tetracalcium aluminoferrite (C.sub.4 AF). In addition, the composition can contain minor quantities of other impurities normally found in feed to Portland cement kilns such as phosphates andalkalies. Furthermore, mineralizers such as fluorspar can be added. As will be discussed hereinbelow, the composition of the subject invention will generally have a liquid phase which forms in the burning of the clinker in an amount in the range offrom about 15 to about 35% by weight of the composition. It is not understood why the small amount of MgO in the free lime expansive cement composition imparts the improved ability of the composition to grab the steel reinforcing members. We have found that when a threshold amount of MgO is present inthe free lime expansive cement composition, the amount of grab or the ability for the composition to bond to and stretch the steel reinforcing member is greatly increased. It is necessary that

the MgO be uniformly distributed throughout the composition. Generally, we have found that where the magnesium oxide content in the expansive cement is less than about 2% by weight thereof, the probability of obtaining a low percent grab is greatly increased. When this threshold amount of MgO is exceeded, thepercent grab is greatly improved. Furthermore, we have found in view of the limited solubility of magnesium oxide in the liquid phase and silicate phase of the clinker, if too great an amount of periclase (free crystalline MgO) is formed in the clinker,it can have adverse effects, particularly if it is poorly distributed throughout the clinker and has a large particle size. Therefore, it is preferred that the MgO be uniformly distributed throughout the clinker as a solid solution in the liquid phase,and diffused in the silicate phase, and in the form of crystals or crystallites of MgO, substantially all of which are of a size of 10 microns or less. Generally, it is preferred that at least 75% by weight of the MgO crystals or crystallites present inthe composition have a size of 10 microns or less and more preferably at least from about 90% by weight thereof have a size of 10 microns or less. The presence of greater quantities of MgO crystals of a larger particle size will generally have no effecton the ability of the expansive composition to grab the steel reinforcing members, but may result in unsound cements or concretes because such larger crystals of MgO are known to cause delayed and possibly disruptive expansion in mortars and concrete dueto expansion which accompanies slow hydration of the large periclase particles. The smaller particles of MgO used in the scope of the subject invention do not cause delayed and disruptive expansion. It is preferred that the MgO not be present in thecomposition in amounts greater than about 7% by weight thereof, even though under certain conditions greater quantities may be used. The most preferred range of MgO within the clinker is between about 3 and about 5 weight percent thereof. The effectiveness of MgO content in enhancing the grab effect of a free lime type expansive cement composition varies somewhat with the percent of free lime in the expansive composition. The MgO within the preferred ranges stated above is mosteffective when the free lime content of the expansive composition varies between about 5 and about 30% by weight thereof. When the free lime content reaches about 40% by weight or greater, the formulation containing the preferred amount of MgO stillexhibits a superior performance to a similar free lime containing expansive composition having a lesser amount of MgO, although the performance of the latter will have improved. The expansive cement composition of the subject invention is preferably made from argillaceous and calcareous reactant materials which are normally used in the production of Portland cement clinker, and magnesia (magnesium oxide) supplyingmaterials. The source of the magnesium oxide can be in the calcareous and argillaceous reactant materials, as it is well known that depending upon geographic location, such materials may be quarried which naturally have high amounts of magnesium oxide,frequently in the form of the mineral brucite. In the case of other argillaceous and calcareous reactant materials which contain substantially lesser amounts of magnesium oxide, as is described in this subject invention, materials must be added which have a high magnesium oxide content andthus

raise the total magnesium oxide content of the expansive cement composition to a range of about 2 to about 7% by weight thereof. Materials which can be used include any argillaceous, calcareous or magnesia reactant materials which have acomposition compatible with the composition of Portland cement. Examples of suitable materials include dolomite, magnesia limestone, magnesium hydroxide, magnesium oxide, magnesium carbonate, and certain magnesium silicates such as serpentine, mixturesthereof, and the like. Generally, the argillaceous and calcareous reactant materials and magnesia supplying materials are reacted in a conventional cement kiln such as rotary kiln which is operated under such conditions that the starting silicate materials within theargillaceous materials will react with the lime within the calcareous material to the extent that the stoichiometric laws will allow and form primarily tricalcium silicate and the excess amount of free lime. This process will inherently distribute theMgO uniformly throughout the clinker. This process is carried out by operating the kiln under "normal" to "hard burn" conditions. The kiln is operated at a temperature in the range of from about 2500.degree. F to about 2900.degree. F and preferablyat a temperature between about 2650.degree. F to about 2750.degree. F, and the materials are reacted in the burning zone of the kiln for a time sufficient to convert all, or substantially all, of the silicate materials to tricalcium silicate. Thiswill inherently distribute the MgO throughout the clinker and allow it to enter the liquid phase as a solid solution, diffuse into the silicate phase, and become diffused throughout the composition in the form of very fine crystallites or crystals ofpericlase, the vast majority of which have a size no greater than about 10 microns. Additionally, the clinker should be cooled no more slowly than free lime expansive cement clinkers are conventionally cooled. A slower cooling of the clinker may causeexcess quantities of periclase crystals to form which have a size larger than about 10 microns. To assist burnability of the above mentioned calcareous, argillaceous, and magnesia materials, the raw feed which enters the kiln is preferably ground to a sufficient fineness so that complete reaction occurs between all of the reactants withinthe burning zone of the kiln. For example, if dolomite is used as the magnesia supplying material, its diffusion and solubility throughout the clinker matrix is enhanced by a fine particle size as this material is generally quite hard in comparison toother argillaceous and calcareous reactant materials, and insufficient grinding or screening will allow large particles of dolomite to pass into the kiln. It has been found that these larger particles tend to give an excessive inhomogeneity of periclasewithin the clinker. In similar fashion, large quartz or silica particles within the reactant mixture will react with surrounding lime and form C.sub.2 S and then C.sub.3 S. However, because of the large particle size of such materials, the immediatelysurrounding area will become depleted in lime, and if insufficient time is available for diffusion of calcium ions to the site, the result will be what is termed a "belite nest" which is a ring of C.sub.2 S surrounding the residual quartz particle. Toavoid these complications, the raw feed materials should preferably have a particle size such that no less than about 97% by weight thereof passes through a 50 mesh (Tyler) screen, and no less than about 80% by weight thereof passes through 200 mesh(Tyler) screen.

The improved expansive cement composition of the subject invention will have a liquid phase from about 15 to about 35% by weight thereof and more preferably from about 25 to about 30% by weight thereof as calculated by the well known method ofLee and Parker, which is set forth in Building Research Technical Paper No. 16, His Majesty's Stationary Office, London, 1935, which is herein incorporated by reference into this application. For example, when the composition of the subject invention ismade at the preferred temperature of between about 2650.degree. F and about 2750.degree. F, the percent liquid phase at this temperature based on the percentage contents of Al.sub.2 O.sub.3, Fe.sub.2 O.sub.3, MgO, K.sub.2 O and Na.sub.2 O in theclinker will be 3.0 (%Al.sub.2 O.sub.3) + 2.25 (%Fe.sub.2 O.sub.3) + %MgO + %K.sub.2 O + %Na.sub.2 O. As the percentage liquid phase formed depends upon the burning temperature and is lower when the burning temperature is also lower, it is noted thatwhen the composition of the subject invention is made at a temperature of about 2550.degree. F, the percent liquid phase will be 2.95 (%Al.sub.2 O.sub.3) + 2.20 (%Fe.sub.2 O.sub.3) + %MgO + %K.sub.2 O + %Na.sub.2 O. The value of MgO shall not exceed 2%in the above liquid phase calculations. After the clinker of the subject invention is burned, it can either be ground separately or interground with a conventional Portland cement clinker, as desired. When ground separately, the clinker should be ground to a particle size no largerthan about 80 mesh (Tyler) and a Blaine fineness of no greater than about 4700 cm.sup.2 /gm as determined by ASTM C204-68. More preferably, the expansive composition of the subject invention has a particle size that substantially passes 100 mesh (Tyler)and a fineness in the range of from about 2000 to 3700 cm.sup.2 /gm. If the expansive composition of the subject invention is ground to a very low fineness, it will hydrate at a reduced rate and result in rather high and somewhat delayed or extended rates of expansion. On the other hand, when the composition ofthe subject invention is ground to a fineness much higher than about 4700 cm.sup.2 /gm, the expansion occurs more rapidly and is of lower magnitude even though in both of these cases the percentage of free lime available is the same as one ground to thepreferred fineness set forth above. The expansive cement composition of the subject invention can be utilized alone as an expansive cement composition. In such a case, conventional amounts of gypsum can be ground therein to achieve a measure of set control. In addition, theexpansive cement composition of the subject invention can be blended with conventional cement compositions to impart expansive qualities to such conventional compositions. Furthermore, as explained herein below, the expansive cement composition clinkerof the subject invention can be interground with clinkers of the conventional cement compositions. For example, the expansive cement composition of the subject invention can be blended or interground with all types of Portland cement, including ASTMTypes I through V, masonry Portland cement, special colored Portland cement, and pozzolanic cement, such as ASTM Type IP. Such pozzolanic cements include cements containing fly ash, slag, calcined clays or shales, or other materials often used as ablend in various Portland cements. In addition, the cement composition of the subject invention can be used with non-Portland cements such as calcium aluminate cements or gypsum cements.

It is noted that the free calcium oxide in the expansive composition is an extremely strong desiccant and the expansive composition must be protected from humid air and other sources of moisture. This is difficult to achieve in practice as evenwater of crystallization in gypsum which is interground with the expansive clinker to achieve a measure of set control may be slowly removed and adsorbed by the free calcium oxide particles. Such process may be enough to cause some degree of fracturingof the ground cement particles. This so-called decrepitation or reduction of particle size has been found to slowly occur in stored free-lime containing expansive cements, particularly that stored in large silos or bins at a cement plant. For example,over a period of several months, the average particle size of the expansive composition expressed in times of Blaine fineness may increase by as much as a 1000 cm.sup.2 /gm. Accordingly, it is desirable to grind the expansive clinker to as low a surfacearea as can provide adequate expansion and rate of expansion, so as to allow sufficient latitude for possible loss of performance upon storage. In accordance with another embodiment of the subject invention, it has been discovered that if the expansive clinker of the subject invention is interground with Portland cement clinker to produce an expansive cement composition which can be useddirectly, without dilution or the addition of further Portland cement, the increase in fineness (decrepitation) during storage appears to be lessened and storage life is therefore prolonged. Generally, when the expansive cement composition of the subject invention contains less than about 10% by weight of free lime therein, it can be used directly as an expansive cement. When used in a mixture with Portland cement compositions, theexpansive cement compositions of the subject invention preferably contains from about 5 to about 45% by weight of free CaO, and more preferably from about 5 to about 30% by weight of free CaO and generally from about 2 to about 7% by weight MgO andpreferably from about 3 to about 4% MgO. The expansive cement composition of the subject invention can be combined with any conventional Portland cement at any suitable portion, e.g., from about 1 to about 99% by weight of expansive cement compositionin the resultant blend. The expansive cement composition of the subject invention can be tailored for expansiveness not only by controlling the amount of free lime and MgO therewithin, but also by the control of the particle size thereof. Generally,the fineness of the composition will regulate the rate of hydration and expansion. Very simple tests for particle size versus expansion time can be made with the cement for any one particular situation. In addition to the ability of the expansive cement composition of the subject invention and concrete mixes containing such composition to effectively bond to and stress steel reinforcement members in an improved manner, the expansive compositionof the subject invention shows a marked improvement over many conventional expansive compositions in its ability to perform well under elevated temperature and extended mixing time conditions. One of the defects of the earlier sulfoaluminate typeexpansive cement compositions is that under hot weather conditions where the temperature of the concrete is in excess of 100.degree. F, virtually all of the expansive potential is lost, and the concrete fails to perform in its expected manner. Incontrast, the expansive cement composition of the subject invention exhibits both good expansive characteristics and

improved ability to stress steel reinforcing members under the elevated temperature mixing and placement conditions which are normal andcharacteristic in the Southern and Western states in this country during the Summer months. The expansive cement composition of the subject invention can be worked in such warm conditions with usual construction practices which include retempering themix at the job site with additional water in order to achieve a more workable fluid in the concrete. The improved qualities of the expansive cement composition of the subject invention are illustrated in greater detail in the following examples. These examples are given to better facilitate the understanding of the subject invention and are notintended to limit the scope thereof. EXAMPLES In the following examples, the apparatus and method for testing the ability of an expansive cement composition to bond to and stretch steel reinforcing members which are disclosed in copending U.S. patent application Ser. No. 546,705 filed Feb. 3, 1975, and entitled EXPANSIVE CEMENT MEASUREMENT were utilized, and such application is herein incorporated by reference into the subject application. This copending application discloses a special test specimen which was developed in order todetermine the degree of bondability of concrete to a reinforcement steel, and accordingly, the degree that such steel reinforcement is stretched in relation to the degree of expansion of the concrete. This specimen consists of a cast concrete slab 5.times. 14 .times. 37 inches in size, containing a length of No. 3 deformed reinforcement steel bar, or any other desired size steel bar, placed longitudinally within the slab 21/2 inches below the top surface thereof, and 21/2 inches from one side. The steel bar has a sufficient length to extend longitudinally through the slab and extend from either end thereof. Rectangular risers are welded at both ends of the reinforcement bar which extend from the ends of the specimen and each riser contains atapped recess on 40 inch centers. Additionally, brass inserts are embedded in the unset concrete positioned at 10 inch intervals across the surface of the slab aligned with the longitudinal axis of the reinforcement bar. The positioning of the insertand the tapped recesses on the risers are such that subsequently inserted Whittemore stainless steel gauge plugs would lie along a line coincident with the longitudinal axis of the reinforcement bar and any adjacent pair of such plugs would beapproximately 10 inches, one from the other. In such a specimen, after the concrete has obtained a final set, the form is removed and initial measurements are made with a Whittemore Strain Gauge across the four 10 inch spans from one end of the reinforcement bar to the other. Atappropriate time intervals, the span lengths are again measured, and differential expansion of each is recorded. The expansion of the steel reinforcement bar is calculated taking the sum of the length changes across the four spans and dividing by 3.7(of the nominal 40 inch length of the steel bar, 37 inches is embedded within the concrete). The expansion of the concrete itself directly over the steel reinforcement bar is determined by taking the average differential expansion of the two centerspans, which consists of the three gauge plugs embedded 10 inches apart within the concrete surface. The ratio of steel expansion to concrete

expansion is termed grab and is a measure of the degree of stretching of the steel by the expanding concrete. Complete bonding would give a grab of 100% while total slippage of the steel or lack of bonding will yield a grab of 0%. In other words grab is equal to ##EQU1## Additionally, brass inserts and gauge plugs are positioned within the concrete surface in a row parallel to the axis of the reinforcement bar, but at a distance 5 or 10 inches away on the surface of the slab, or both. These gauge plugs allowmeasurement of concrete expansion at varying distances from the steel bar and indicate how far the restraining effect of the steel is exerted on the concrete. In effect, such measurements would indicate the "severe influence" the steel bar has on theconcrete in the proximity thereof. EXAMPLE I In this Example, conventional sulfoaluminate and free CaO type expansive cement compositions were compared with the improved composition of the subject invention as to expansive performance and ability to grab steel reinforcing members. Thesulfoaluminate expansive cements were used directly as commercially obtained. The CaO expansive cement and the CaO-MgO improved expansive cement of the subject invention were blended with Type I Portland cement at such a ratio as to produce the desiredamount of expansion. Expansive Composition No. 1 was an expansive cement of the sulfoaluminate type sold by General Portland Inc. under the name "Crakgard". Expansive Composition No. 2 was expansive cement of the sulfoaluminate type sold by TexasIndustries Inc. under the name of "ChemComp". Expansive Composition No. 4 was an experimental free calcium oxide expansive cement manufactured by General Portland Inc. Expansive Composition No. 3 was CaO-MgO expansive cement made in the scope of thesubject invention. The chemical composition of each of the above expansive cement formulations is set forth in Table 1 below: Table 1 ______________________________________ Chemical Composition of Expansive Cement Formulations ______________________________________ Expansive Composition No. Chemical (percent by weight) Composition 1 2 3 4 ______________________________________ SiO.sub.2 17.4 18.9 14.1 14.1 Al.sub.2 O.sub.3 7.8 6.9 4.9 4.9 Fe.sub.2 O.sub.3 2.2 2.4 2.7 3.6 CaO 63.0 63.3 72.5 71.8 MgO 1.0 0.8 2.5 1.2 SO.sub.3 6.1 6.3 2.5 2.5 Ign. Loss 1.1 1.1 1.0 1.4 Free CaO 0.91.2 22.6 21.5 Blaine Fineness 3440 3640 2200 1930 ______________________________________ Expansive Composition No. 3 and Expansive Composition No. 4 were each blended at a ratio of 25% expansive component to 75% by weight of Type I Portland cement. Thereafter, each of the four expansive compositions were mixed using a standard 51/2sack per yard mix design consisting of 44 parts by weight of cement, 103.8 parts by weight of sand, 175.3 parts by weight of aggregate, and sufficient water to produce a 5.5 inch slump as determined by the method of ASTM C143-69. The resulting concretemixes containing Expansive Compositions Nos. 1, 2, 3 and 4 were labeled "Concrete Mixture Nos. 1, 2, 3 and 4", respectively. Each of the concrete mixtures was molded into the 5

.times. 14 .times. 37 inch slabs containing the 40 inch No. 3 deformedreinforcing steel bar with the rectangular risers molded to the ends thereof and also having the tapped recesses and Whittemore gauge plugs positioned on the risers and on the surface of the slab as described in the introductory paragraphs of theexamples above. Each of the test specimens was allowed to cure under ambient conditions and the expansion of the steel reinforcement bar in the slab; the expansion of the concrete in the slabs (a) adjacent to steel reinforcement bar, (b) 5 inches fromthe axis of the steel reinforcement bar, and (c) 10 inches from the axis of the steel reinforcement bar was measured; and thereafter the grab of the concrete on the reinforcement bar at various distances was calculated as defined in the introductoryparagraphs of the examples. The results are set forth in Tables 2 through 6 below: Table 2 ______________________________________ Expansion* of Steel Reinforcement Bar in Slabs(%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 3 6 12 ______________________________________ 1 .010 .000 -.006 -.010 -.013 -- 2 .009 . 001 -.007 -.012 .005 -.002 3 .018 .026 .032 .071 .070 .084 4 .018 .029 .028 .020 .026 . 032 ______________________________________ *Negative sign indicates that steel inslab became compressed. Table 3 ______________________________________ Expansion* of Concrete in Slabs Adjacent to Steel Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 3 6 12 ______________________________________ 1 .013 .002 -.004 -.001 -.004 -- 2 .015 . 006 -.006 -.004 .006 .005 3 .022 .032 .046 .090 .090 .095 4 .064 .076 .074 .080 .079 .084 ______________________________________ *Negative sign indicates that concrete inslab underwent contraction. Table 4 ______________________________________ Expansion* of Concrete in Slabs 5-Inches from the Axis of the Steel Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 36 12 ______________________________________ 1 .015 . 000 -.003 .004 -.002 -- 2 .015 .008 -.005 -.001 .006 .003 3 .028 .037 .058 .115 .106 .117 4 .070 .099 .098 .102 .109 .113 ______________________________________ *Negative sign indicates thatconcrete in slab underwent contraction. Table 5 ______________________________________ Expansion* of Concrete in Slabs 10-Inches from the Axis of the Steel Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 36 12 ______________________________________ 1 .023 . 012 .009 .012 .008 -- 2 .017 .009 -.003 .000 .007 .007 3 .027 .039 .062 .135 .124 .138 4 . 101 .115 .118 .119 .126 .133 ______________________________________ *Negative sign indicates that concretein slab underwent contraction. Table 6 ______________________________________ "Grab" of Concrete on Reinforcement Bar at Various Distances* (%)

______________________________________ Concrete Distance Measurement Intervals Mixture from Days Months No. Bar 1 7 28 3 6 12 ______________________________________ 1 0 77 0 0 0 0 -- 5 67 0 0 0 0 -- 10 43 0 0 0 0 -- 2 0 60 17 0 0 0 0 5 60 12 0 0 0 0 10 53 11 0 0 0 0 3 0 81 82 68 79 77 88 5 63 70 54 62 66 71 10 66 65 51 57 57 60 4 0 28 38 38 26 32 39 5 26 29 28 20 2429 10 18 25 24 17 20 24 ______________________________________ *Distances are measured on surface of the concrete slab and perpendicularly to the projected axis of the bar. As can be clearly seen in Tables 2 through 6 above, Concrete Mixture No. 3 containing the improved expansive cement component of the subject invention exhibited superior performance characteristics in that it showed a substantial level expansionand stressing of steel reinforcement, and therefore a high calculated grab over a time period of 1 year. During this time period, drying shrinkage was negligible. EXAMPLE II In this Example, four free CaO expansive cement compositions were prepared in which the major compositional difference was the amount of MgO contained within each. Each composition was prepared by burning ground argillaceous and calcareous rockat a temperature between 2700.degree. F and 2800.degree. F in a laboratory pilot rotary kiln, and air quenching the resultant hard densified clinker product. The resulting expansive composition No. 5 clinker, the control composition, contained about1% by weight MgO; Expansive Composition No. 6 clinker contained about 3% by weight MgO; Expansive Composition No. 7 clinker contained about 5% by weight MgO; and Expansive Composition No. 8 clinker contained 7% by weight MgO. The analyses of theseclinker products are given in Table 7 below: Table 7 ______________________________________ Chemical Analyses of Expansive Clinkers Containing Varying Amounts of Magnesium Oxide ______________________________________ Expansive Composition No. (Percent by Weight) Composition 5 6 7 8 ______________________________________ SiO.sub.2 14.2 13.8 14.1 14.0 Al.sub.2 O.sub.3 5.6 5.3 4.4 4.8 Fe.sub.2 O.sub.3 3.2 3.1 2.9 2.9 CaO 76.0 75.0 71.4 70.7 MgO 0.9 3.0 5.4 7.0 SO.sub.3 0.2 0.1 0.0 0.0 Ign. Loss 0.7 0.3 1.3 1.0 Na.sub.2 O 0.10.0 0.0 0.0 K.sub.2 O 0.5 0.3 0.2 0.1 Free CaO 25.6 25.4 24.0 22.6 ______________________________________ Next, all the expansive compositions 5 through 8 clinkers were ground in a steel ball mill with gypsum in the quantity of 24 parts by weight clinker and 1 part by weight of gypsum. Each of the ground products was then passed through a 100 meshscreen (Tyler) and all particles which were retained on the screen were discarded. The ground cement of clinker composition No. 1 had a Blaine surface area of 3245 cm.sup.2 /gm; the ground cement of clinker composition No. 2 had a Blaine surface area of3040 cm.sup.2 /gm; the ground cement of clinker composition No. 3 had a Blaine surface area of 3054 cm.sup.2 /gm; and the ground cement of clinker composition No. 4 had a Blaine surface area of 3054 cm.sup.2 /gm all determined by ASTM C204-68.

An expansive cement composition was made from each of the ground cements of Expansive Composition Nos. 5, 6, 7 and 8 by blending 25% by weight of each of such compositions with 75% by weight of Type I Portland cement. Concrete Mixture Nos. 5,6, 7 and 8 were prepared which contained the ground expansive cement composition Nos. 5, 6, 7 and 8, respectively, by mixing each of the above described expansive cement mixtures with aggregate and water in a standard mix design containing a cementfactor of 51/2 sacks of cement per cubic yard of finished concrete. Each concrete sample made comprised 44 parts by weight of cement, 103.8 parts by weight of sand, 175.3 parts by weight of coarse aggregate (crushed limestone), and sufficient water(about 28 parts by weight), so that the resulting sample mixes had a slump of 51/2 inches as determined by the method of ASTM C143-69. Portions of each of the concrete mixtures 5, 6, 7 and 8 were then individually cast into a wooden mold to form the above described 5 .times. 14 .times. 37 inch slab containing the length of No. 3 to form reinforcement steel bar placedlongitudinally within the slab as described in the introductory paragraphs of the examples, specifically, the 40 inch steel bar was placed longitudinally within the slab 21/2 inches below the top surface thereof and 21/2 inches from one side thereof. Rectangular risers were securely welded to both ends of the reinforcement bar and each riser contained a tapped recess on 40 inch centers. Additionally, as described in the introductory paragraphs of the examples, brass inserts were embedded in theunset material positioned at 10 inch intervals across the surface of the slab aligned with the longitudinal axis of the reinforcement bar. The positioning of the inserts and the tapped recesses on the riser was such that subsequently inserted gaugeplugs would lie along a line coincident with the longitudinal axis of the reinforcement bar and any adjacent pair of such plugs would be approximately 10 inches one from the other. After pouring, the 5 .times. 14 .times. 37 inch specimens were allowed to set under damp burlap for about 6 hours and at that time, the mold was removed. The steel gauge plugs were then secured in place by insertion into the appropriate tappedrecesses as described in the introductory paragraphs of the examples. The initial measurements were then taken with a "Whittemore" strain gauge and the slabs were coated with a curing compound to retard moisture loss from the slab. The slabs remainedon the laboratory floor at approximately 75.degree. C and 50% relative humidity for about 1 month and then were removed to a location outside the laboratory where they were subjected to ambient weather conditions. Measurements were taken at weeklyintervals for the first month, monthly for the next three months, and then every third month thereafter for a total period of 1 year. The data showing the expansion of steel reinforcement bar in each slab is set forth in Table 8 below, and the expansionof the concrete in the specimen slab adjacent to the steel reinforcement bar is set forth in Table 9 below, and the grab of the concrete on the reinforcement bar was calculated by the method set forth in the introductory paragraph of the examples and isset forth in Table 10 hereinbelow. Table 8 ______________________________________ Expansion of Steel Reinforcement Bar in Slabs (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 3 6 12

______________________________________ 5 .018 .029 .028 .020 .026 .032 6 .011 . 019 .038 .058 .075 .090 7 .005 .014 .017 .036 .052 .059 8 .024 .037 .038 .048 .075 .082 ______________________________________ Table 9 ______________________________________ Expansion of Concrete in Slabs Adjacent to Steel Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 3 6 12 ______________________________________ 5 .064 .076 .074 .080 .079 .084 6 .018 . 030 .051 .074 .092 .108 7 .017 .025 .032 .052 .066 .071 8 .036 .050 .054 .070 .098 .104 ______________________________________ Table 10 ______________________________________ "Grab" of Concrete on Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 3 6 12 ______________________________________ 528 38 38 26 32 39 6 63 64 74 78 82 84 7 27 54 54 71 79 82 8 65 74 70 68 77 79 ______________________________________ Tables 8, 9 and 10 clearly show that once a threshold value of magnesium oxide in the composition has been surpassed, the grab is greatly increased. More specifically, there are relative minor differences in grab between concrete mixtures 6, 7and 8 even though the magnesium oxide content varied between 3 and 7% in the expansive component and the free CaO content of the expansive component varied between 22.6 and 25.4 weight percent thereof. Concrete Mixture No. 1 which contained less than 1%of the MgO in the expansive component thereof exhibited much lower grab than any of the other compositions. A portion of each of the concrete mixtures 5, 6, 7 and 8 were individually poured into standard 3 .times. 3 .times. 10 inch molds, one-half containing a 1/4 inch threaded steel reinforcing rod extending longitudinally through the mold andconnected to two steel plates on the opposite ends thereof and one-half containing no steel reinforcement. After casting and setting for about 6 hours, the 3 .times. 3 .times. 10 inch bars were demolded, initially measured with a length comparator and then placed in a moist room where they remained throughout the test period at a temperature of73.degree. F and a relative humidity of approximately 100%. The 3 .times. 3 .times. 10 inch test bars were measured daily for a period of 1 week, and then weekly until a month had elapsed. At this point, they were placed in a 180.degree. F steamcabinet for a period of three days. Upon removal, they were cooled overnight in a moist room and measured to determine length changes due to steam treatment. Next, the steam treated bars were autoclaved for 3 hours at 3000 psi, again cooled in a moistroom overnight, and measured. The results are as set forth in Tables 11 and 12 below. Table 11 ______________________________________ Restrained Expansion of 3 .times. 3 .times. 10-Inch Bars ______________________________________ (%) Concrete Measurement Interval Mixture Days Condition After 28 Days No. 1 7 28 SteamAutoclave ______________________________________ 5 .027 .040 .095 .120 .

097 6 .032 .052 .072 .179 .208 7 .015 .024 .051 .116 .160 8 .020 .068 .066 .175 .211 ______________________________________ Table 12 ______________________________________ Unrestrained Expansion of 3 .times. 3 .times. 10-Inch Bars ______________________________________ (%) Concrete Measurement Interval Mixture Days Condition After 28 Days No. 1 7 28 SteamAutoclave ______________________________________ 5 .087 .101 .112 .145 . 172 6 .048 .077 .125 .485 .535 7 .024 .067 .087 .232 .301 8 .082 .126 .140 .441 .522 ______________________________________ Data set forth in Table 11 as compared to Table 9 indicates the nature of the differences experienced between the small cast bars maintained at approximately 100% humidity, and the more realistic larger slab specimens. Over the measurementperiod of 28 days, the fully restrained bars exhibited a greater expansion than the slabs which were restrained only to the degree indicated by the amount of grab of concrete to steel reinforcement bar. Furthermore, as shown in Table 12 the levels ofexpansion are even greater where the bars underwent complete unrestrained expansion. The expansion due to steam treatment and autoclaving indicate the maximum possible expansion of the high temperature conditions, and as this expansion value changes little with respect to the level of magnesium oxide, it is clear that suchexpansion is mostly due to the presence of free lime, and possibly due to recrystallization of very small calcium hydroxide crystallites and to larger crystals. Crystal coarsing resulting from heat treatment is well known in the chemical literature. Thus, differences in the expansion of the four concrete mixes are within expected experimental variation and show no definite trend due to progressive compositional changes. This is further brought out by the data set forth in Tables 8 through 10 which show the expansion of concrete slabs stored out-of-doors and subject to all variations of winter cold, summer heat, and wide ranges of ambient moisture content werestabilized by the first six month period and very little change in dimension had occurred during the next six month period. Thus, this example clearly shows not only the beneficial effect of the magnesium oxide in the free calcium oxide expansivecomposition, but also the long term stability of such compositions. EXAMPLE III Still another expansive composition of this invention was produced by burning an argillaceous and calcareous raw mix containing ground dolomite (calcium magnesium carbonate) in a kiln. The proportions of said raw materials were controlled suchthat the resulting mix would produce about 40% by weight free CaO and about 4% by weight MgO therewithin. A second control composition was formulated in the same fashion, but lacked any additional source of magnesium oxide. This composition wascontrolled such that the resulting mix would produce about 40% by weight free CaO and about 1% by weight MgO therewithin.

Both mixtures were passed into a kiln and heated to a temperature of about 2650.degree. F to 2700.degree. F. The clinker from the kiln containing about 4% by weight MgO was ground with gypsum at the ratio of 24 parts by weight expansive clinkerto 1 part by weight of gypsum to a Blaine fineness of 2810 cm.sup.2 /gm and the resultant expansive cement component was called Expansive Composition No. 9. The control clinker, which contained about 1% by weight MgO, as ground with gypsum in the sameproportion to a Blaine fineness of 2720 cm.sup.2 /gm, and the resultant expansive cement component was called Expansive Composition No. 10. The two expansive cement compositions were analyzed and the results of the analyses are set forth in Table 13. Table 13 ______________________________________ Chemical Analysis of Expansive Cement Components Containing Two Levels of Magnesium Oxide ______________________________________ Expansive Composition No. Chemical (Percent by Weight) Composition 9 10 ______________________________________ SiO.sub.2 10.4 10.9 Al.sub.2 O.sub.3 3.6 3.9 Fe.sub.2 O.sub.3 2.9 2.9 CaO 75.5 77.9 MgO 3.7 1.0 SO.sub.3 2.1 2.3 Ign. Loss 1.4 1.8 Free CaO 40.0 41.9 ______________________________________ Two expansive concrete mixtures were prepared, which each comprised 44 parts by weight of cement (consisting of 15% of the above described expansive compositions and 85% of standard Type I Portland cement), 103.8 parts by weight of sand, 175.3parts by weight of coarse aggregate (crushed limestone), and sufficient water (about 28 parts by weight), so that the resulting sample mixes had a slump of 51/2 inches as determined by the method of ASTM C143-69. The concrete prepared from ExpansiveComposition No. 9 is referred to as Concrete Mixture No. 9, and the concrete prepared from Expansive Composition No. 10 is referred herein as Concrete Mixture 10. Both expansive compositions were cast into 5 .times. 14 .times. 37 inch slabs, each containing a length of No. 3 deformed reinforcement steel, and each carried the brass inserts as described in the introductory paragraphs of the Examples. Next, each slab was cured and measured as was described in Example II. Initial measurements were taken with a Whittemore strain gauge on both slabs at the age of about six hours when the forms were removed, and then subsequently measured periodically up to six months. Measurements of the 5 .times. 14 .times. 37inch slabs are given in Tables 14, 15 and 16 below, in which Table 14 shows the expansion of the steel reinforcement bar within each slab, Table 15 shows the expansion of the concrete adjacent to the reinforcement bar, and Table 16 shows the percent grabof the concrete on the reinforcement bar and additionally at distances at 5 and 10 inches from the longitudinal axis of the reinforcement bar. Table 14 ______________________________________ Expansion of Steel Reinforcement Bar in Slabs (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 9 .050 .054 .046 .047 .078 .072 10 .023 . 027 .027 .033 .039 .028 ______________________________________

Table 15 ______________________________________ Expansion of Concrete in Slabs Adjacent to Steel Reinforcement Bar (%) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 9 .058 .062 .056 .056 .079 .072 10 .030 . 033 .032 .038 .041 .032 ______________________________________ Table 16 ______________________________________ "Grab" of Concrete on Reinforcement Bar at Various Distances* (%) ______________________________________ Concrete Distance Measurement Interval Mixture from Days Months No. Bar (in.) 1 7 282 3 6 ______________________________________ 9 0 87 89 83 84 99 100 5 80 76 68 67 82 83 10 62 60 54 53 68 67 10 0 75 83 85 88 94 88 5 47 49 51 53 58 48 10 35 38 37 42 46 36 ______________________________________ *Distances are measured on surfacesof slab and perpendicularly to the projected axis of the bar. As can be seen in the data presented in the above Tables 14-16, as the free CaO increases to 40% and above, the effect of the MgO on grab or the gripping of the steel by the concrete at a point adjacent the steel diminishes. That is to say, thegrab adjacent the steel is excellent for both cases, with and without added magnesium oxide. However, in such cases of higher free lime content, the addition of magnesium oxide to the expansive composition aids in extending the "sphere of influence" ofthe steel reinforcement bar in restraining the expanding concrete at greater distances therefrom. In the case described in this example, although directly adjacent to the steel, the variation in grab between Concrete Mixtures 9 and 10 are negligible,but it becomes more significant at distances of 5 and 10 inches from the longitudinal axis of the reinforcement bar. Thus at an age of six months, the grab of Concrete Mixture No. 9 which contains almost 4% magnesium oxide, at a distance of 5 inchesfrom the axis of the steel bar is 42% greater than the grab of Concrete Mixture No. 10, which contains only 1% magnesium oxide. Accordingly, at a distance of 10 inches from the longitudinal axis of the reinforcement bar, the grab of Concrete Mixture No.9 is 46% greater than the corresponding grab of Concrete Mixture No. 10. Therefore, it is apparent in this example, the effect of magnesium oxide additions to a free calcium oxide expansive cement composition made according to the subject invention is still useful at free lime levels of 40% and up, and in this case,the major effect is not only to maintain the expected increases and high level of percent grab of the concrete on the steel reinforcement bar, but also to extend the sphere of influence on the expanding concrete by said reinforcement bar. EXAMPLE IV Another expansive cement composition of this invention was produced by burning an argillaceous, calcareous and magnesia raw mix in a laboratory rotary kiln. The ratio of reactants was designed such that the resulting mix would have about 28% byweight MgO therewithin. The mixture was burned in two stages. First, half of the raw mix was passed

into the kiln and heated to a temperature of about 2450.degree. F. The clinker from the kiln was cooled and was subsequently ground with gypsum at aratio of 23.75 parts of expansive clinker to 1.25 parts of gypsum to a Blaine surface area of 2810 cm.sup.2 /gm. This ground cement was called Expansive Composition No. 11. The remaining half of the original raw mix was likewise passed into the kiln and heated to a temperature of about 2700.degree. F. The clinker was air cooled and ground with gypsum at the same ratio as Expansive Composition No. 11 to a Blainesurface area of 2790 cm.sup.2 /gm. This ground cement was called Expansive Composition No. 12. A second raw mix was designed, which by proper proportioning of the argillaceous, calcareous, and magnesia reactants, would produce upon burning in a laboratory rotary kiln a clinker having about 28% by weight free lime and about 4% by weightmagnesium oxide therewithin. The raw mix was also burned in two stages. The first half was passed into the kiln and heated to a temperature of about 2450.degree. F. The resulting clinker was air cooled, ground with gypsum in a ball mill at the ratioof 23.75 parts of clinker to 1.25 parts of gypsum to obtain a Blaine surface area of 2810 cm.sup.2 /gm. This ground cement was called Expansive Composition 13. The remaining half of the raw feed was burned in the kiln at an increased temperature ofabout 2700.degree. F. The resulting clinker was similarly air cooled, ground with the same proportion of gypsum as the first half of the lower temperature burned clinker to a Blaine surface area of 2810 cm.sup.2 /gm. This ground cement was calledExpansive Composition No. 14. The four compositions were analyzed and the results of those analyses are set forth in Table 17 below. Table 17 ______________________________________ Chemical Analysis of Expansive Cement Components Burned at Two Levels of Kiln Temperature Expansive Composition No. Chemical (Percent by Weight) Composition 11 12 13 14 ______________________________________ SiO.sub.2 12.5 12.0 14.0 13.7 Al.sub.2 O.sub.3 3.9 4.7 3.1 3.1 Fe.sub.2 O.sub.3 4.4 4.6 3.1 3.0 CaO 72.8 72.6 71.8 70.7 MgO 3.2 3.2 3.8 3.8 SO.sub.3 2.3 2.5 2.1 4.3 Ign. Loss 1.3 1.4 2.1 1.5 Na. sub.2 O 0.1 0.0 0.1 0.1 K.sub.2 O 0.3 0.1 0.1 0.2 Free CaO 30.4 29.1 31.0 26.8 ______________________________________ Concrete 5 .times. 14 .times. 37 inch test slabs as described in the introductory paragraphs of the Examples were prepared for expansion test using Expansive Composition No. 11 and Expansive Composition, No. 12. Concrete Mixture No. 11 wasprepared by using 25% of Expansive Composition No. 11 and 75% of a Type I Portland cement to comprise the cement portion of the mix. Concrete Mixture No. 12 was prepared by using 25% of Expansive Composition No. 12 and 75% of a Type I Portland cement tocomprise the cement portion of this mix. Concrete Mixture No. 13 was prepared by using 25% of Expansive Composition No. 13 and 75% of Type I Portland cement to comprise the cement portion of the mix. Similarly, Concrete Mixture No. 14 was prepared byusing 25% of Expansive Composition No. 14 and 75% of Type I Portland cement to comprise the cement portion of the mix.

In all four cases, the final concrete mix consisted of 44 parts by weight of the above described cement blends, 103.8 parts by weight of sand, 175.3 parts by weight of coarse aggregate, and sufficient water (about 28 to 29 parts by weight) toproduce a concrete slump of 51/2 inches. The resultant 5 .times. 14 .times. 37 inch slabs, each containing a length of No. 3 deformed rebar, were cast, cured, and measured as previously described in Example II. The expansions of the steel reinforcement bar within the test slab aregiven in Table 18, the expansions of the concrete adjacent to the reinforcement bar are given in Table 19 and the percent grab of the concrete on the reinforcement bar is shown in Table 20. Table 18 ______________________________________ Expansion of Steel Reinforcement Bar in Slabs (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 11 .034 .054 .084 .180 . 181 .192 12 .030 .037 .044 .085 .091 .105 13 .071 .082 .092 .260 .358 .368 14 .090 .088 . 088 .081 .088 .077 ______________________________________ Table 19 ______________________________________ Expansion of Concrete in Slabs Adjacent to Steel Reinforcement Bar (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 11 .038 . 064.094 .200 .206 .226 12 .044 .056 .063 .108 .114 .130 13 .049 .054 .063 .230 .277 .281 14 .054 .050 .048 .042 .050 .050 ______________________________________ Table 20 ______________________________________ "Grab" of Concrete on Reinforcement Bar (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 11 88 84 90 88 88 85 12 69 66 70 79 80 80 13 69 67 68 89 78 76 14 58 57 54 51 56 65 ______________________________________ As can be seen, the two groups of expansive concrete, each consisting of an expansive cement component burned at a lower and a high temperature, have roughly the same expansive characteristics up to about 28 days. Beyond this point in time, theconcrete which contained the expansive cement component (Concrete Mixtures Nos. 11 and 13) which was ground from a clinker burned at a lower temperature, underwent a very rapid increase in expansion, and thus if not adequately restrained, might possiblyreach a level which could be considered disruptive or at least result in a loss of mechanical strength due to microcrack formation. In all four cases illustrated in this example, due to the beneficial effect of an elevated magnesium oxide level practiced according to the subject invention, the percent grab remained high. This occured even in those cases of delayed highexpansion. EXAMPLE V Another expansive cement composition of the subject invention was produced by burning

an argillaceous, calcareous, and magnesia raw mix in a large commercial rotary kiln. The ratio of said reactants was controlled such that the resulting mixwould have been 26% by weight and 27% by weight free lime and 4% by weight magnesium oxide therewithin. The mixture was passed into the kiln and heated to a temperature of about 2750.degree. F. The clinker from the kiln passed into a grate cooler, wasair cooled and subsequently was conveyed into a finish mill and storage tank. It was ground in a closed circuit grinding system with gypsum to a Blaine fineness of 3270 cm.sup.2 /gm, and 99.3% by weight of the particles passed through a 100 mesh (Tyler)screen, 95.0% by weight of the material passed through a 200 mesh (Tyler) screen and 80.9% by weight passed through a 325 mesh (Tyler) screen. The resulting ground cement composition had 26.1% free lime, 3.7% magnesium oxide, and the remainingconstituents were tricalcium silicate, tricalcium aluminate, and tetracalcium aluminoferrite. This cement was called Expansive Composition No. 15. The composition was analyzed, and the results of the analysis are set forth in Table 21 below. Table 21 ______________________________________ Chemical Analysis of a Ground Expansive Cement Component Burned in a Commercial Rotary Kiln ______________________________________ Chemical Expansive Composition No. 15 Composition (Percent byWeight) ______________________________________ SiO.sub.2 12.9 Al.sub.2 O.sub.3 3.8 Fe.sub.2 O.sub.3 4.0 CaO 71.0 MgO 3.7 SO.sub.3 2.6 Ign. Loss 2.3 Free CaO 26.1 ______________________________________ This Expansive Composition No. 15 was utilized to determine the effect of extended elevated temperature concrete mixing conditions, which are commonplace in many geographical areas during the summer months of the year. Three expansive concretemixtures were prepared, each of identical composition, and were mixed under varying conditions of time, temperature, and water content. They were subsequently cast into 5 .times. 14 .times. 37 inch test slabs, as described in the introductoryparagraphs in the Examples. The concrete mixture utilized contained 44 parts by weight of cement, which consisted of 20% by weight of Expansive Composition No. 15 and 80% by weight of Type I Portland cement, 103.8 parts by weight of sand, 175.3 parts by weight of coarseaggregate, and sufficient water to obtain a 51/2 inch slump. Additionally, a control mixture was prepared in which the cement fraction of the concrete consisted of 100% Type I Portland cement. No slab was cast with this concrete. The concrete mixer used for these tests was preconditioned by being allowed to stand for several hours in the hot summer sun. Additionally, the sand and coarse aggregate used was conditioned in an oven to a temperature of 100.degree. F. Allwater used was preheated to a temperature of 90.degree. F. The first concrete mix, referred to as Concrete Mixture No. 15, was made using a cement portion consisting of 100% Type I Portland cement and the other heated materials previously described. The slump was adjusted to 51/2 inches with hot water and the temperature of the concrete was measured. The hot concrete was mixed for five minutes, and then the mixer was stopped for a period of four minutes. Mixing was resumed for one minute, and atthis point, the concrete allowed to

rest for an additional four minutes. Then after one minute of mixing, the slump was determined and the concrete temperature was measured. This cycle of the concrete being allowed to rest four minutes, mixed oneminute, was continued and both slump and temperature were determined every 10 minutes until a total of 40 minutes had elapsed. At this time, Concrete Mixture No. 15 was discarded. Expansive Concrete Mixture No. 16 was prepared in the same fashion as Concrete Mixture No. 15, but it contained a cement fraction consisting of 20% of Expansive Composition No. 15 and 80% Type I Portland cement. It too was initially adjusted toa slump of 51/2 inches and the temperature measured. In a similar fashion to the previous test, after being mixed for five minutes, it was allowed to rest four minutes, mixed for one minute, rest for five minutes, mix for one minute and then the slumpand temperature measured. Again, this procedure was continued for a total of 40 minutes, at which time Expansive Concrete Mixture No. 16 was cast into a 5 .times. 14 .times. 37 inch slab containing a length of No. 3 deformed reinforcement bar asdescribed in the introductory paragraphs in the Examples. Concrete Mixture No. 17 was identical in composition to Expansive Mixture No. 16. It was mixed with hot starting materials and alternate resting and mixing for a total period of one hour. At this time, it was retempered with hot water to regainthe initial slump of 51/2 inches, and subsequently cast into a 5 .times. 14 .times. 37 inch slab containing a length of No. 3 deformed rebar as described in the introductory paragraphs in the Examples. Concrete Mixture No. 18 was similarly composed of a cement fraction consisting of 20% Expansive Composition No. 15 and 80% Type I Portland cement, and the proportion of sand and coarse aggregate previously described. This expansive concrete mixwas produced within the laboratory with room temperature (about 75.degree. F) conditioned components, and was subsequently mixed for the usual five minute or so mixing time. Sufficient water was added during the mixing procedure to produce a 51/2 inchslump. The standard procedure used in producing this concrete was similar to that described in Example II. Concrete Mixture No. 18 was then cast into a 5 .times. 14 .times. 37 inch slab containing a length of No. 3 deformed rebar as described in theintroductory paragraphs in the Examples. The effect of elevated temperature and extended mixing time on Concrete Mixture Nos. 15, 16 and 17 are shown in Table 22 below. Table 22 _______________________________________________________________________ ___ Temperature and Slump of Concrete Mixes Produced at an Elevated Temperature Concrete Mixture No. 15 16 17 Time After Temp. Slump Temp. Slump Temp. Slump Initial Mixing (.degree. F) (in.) (.degree. F) (in.) (.degree. F) (in.) _______________________________________________________________________ ___ 0 117 5.5 118 5.5 118 5.5 10 114 2.5 112 3.5 116 5.0 20 110 3.5 109 3.25 110 4.5 30 105 3.0107 3.0 103 3.75 40 102 2.5 104 2.25 101 2.75 50 (Discarded) (Slab Cast) 97 2.5

60 96 2.25 5.5* (Slab Cast) _______________________________________________________________________ ___ *Retempered with 90.degree. F water. The three 5 .times. 14 .times. 37 inch concrete slabs, each containing a length of No. 3 deformed reinforcement steel which were cast from Expansive Concrete Mixture Nos. 16, 17 and 18 were cured and measured as previously described in ExampleII. The expansions of the steel reinforcement bar within the test slab are given in Table 23, the expansions of the concrete adjacent to the reinforcement bar are given in Table 24, and the percent grab of the concrete on the reinforcement bar is shownin Table 25. Table 23 ______________________________________ Expansion of Steel Reinforcement Bar in Slabs (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 16 .011 .012 .001 .008 . 006 .001 17.020 .028 .027 .029 .035 .024 18 .026 .023 .018 .024 .017 .012 ______________________________________ Table 24 ______________________________________ Expansion of Concrete Slabs Adjacent to Steel Reinforcement Bar (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 16 .015 .017 . 010.019 .020 .014 17 .032 .043 .050 .052 .052 .044 18 .036 .036 .029 .036 .029 .026 ______________________________________ Table 25 ______________________________________ "Grab" of Concrete on Reinforcement Bar (%) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 6 ______________________________________ 16 72 68 5 44 32 2 17 64 65 55 56 66 54 18 71 64 62 64 58 47 ______________________________________ As can be seen in Table 22, the slump loss of the two expansive concrete mixes (Concrete Mixture Nos. 16 and 17) is no different than that which occurs in an ordinary Portland cement concrete (Concrete Mixture No. 15), when subjected to elevatedtemperatures and extended mixing times. These conditions are characteristic of summer weather conditions and usual transport times from a concrete batching plant to a particular job site. As can be seen in Table 22, the slump has dropped by about halfin all three cases from its original value. Customary practice, when concrete is used in the field, is to add additional mix water in order to increase the slump and therefore the fluidity and ease of placement of the concrete. Concrete Mixture No. 17was therefore retempered with hot water to raise the slump back to its initial value, and determine subsequently what effect this action would have on the expansion properties and percent grab of a 5 .times. 14 .times. 37 inch specimen prepared withthis concrete. Tables 23, 24 and 25 show the results of comparing the three expansive concrete slabs formulated with the same components in the same proportions, but differing in temperature. time of mixing, and also water required to achieve an initial slumpof 51/2 inches. Concrete Mixture No. 16 which was mixed for a prolonged period at elevated temperatures initially required about 30 parts by weight of water. Concrete Mixture No.

17, which also underwent hot extended mixing and was subsequentlyretempered to regain the original slump, initially required about 33 parts by weight of water. Finally, Concrete Mixture No. 18, which was mixed for a short period under standard room temperature conditions, initially required about 28 parts by weightof water to reach the required 51/2 inches of slump. Concrete Mixture No. 18 exhibited the usual good grab expected of a concrete prepared from an expansive cement made by the subject invention. Additionally, the level of expansion achieved was also at the level expected for an expansive cementmixture containing about 20% of an approximately 26% free lime component and 80% of a Type I Portland cement. In the case of Concrete Mixture No. 16, however, the effect of placement of a low slump hot concrete resulted in a reduced level of expansion, and also a poorer and more erratic percent grab. At several times of measurement (that is, 28 days and6 months), the steel reinforcement bar showed no expansion at all, and was completely slack. However, in the case of Concrete Mixture No. 17, the percent grab was excellent throughout the entire measurement period and the levels of both steel and concrete expansion were higher than that exhibited by the normally mixed concrete,exemplified by Concrete Mixture No. 18. This example also clearly shows that the expansive composition of the subject invention can be usefully utilized under conditions of elevated temperature and extended mixing times. It actually shows a superior performance under such temperatureand mixing conditions if retempered with water, as is the common custom followed in field practice. This behavior is in direct opposition to the now commonly used sulfoaluminate expansive cements which generally show a greater than normal slump lossduring elevated temperature mixing conditions, and almost completely lose their ability to expand under such elevated temperature mixing conditions. EXAMPLE VI Still another expansive cement composition of the subject invention was produced by intergrinding quarry rock, high-grade limestone, mill scale and dolomite sand in a commercial two compartment ball mill to produce an argillaceous, calcareous andmagnesia containing raw slurry, containing about 38% moisture. The ratio of the reactants utilized in this mix was controlled such that the resulting product would have between 16 and 18% by weight free lime and between 3 and 4% by weight magnesiumoxide therewithin. The mixture was passed into a large commercial rotary kiln and heated to a temperature of about 2650.degree. F to 2700.degree. F. The clinker exiting from the kiln passed into a grate cooler and was subsequently conveyed into a finish mill storage tank. The composition of this clinker is shown in Table 26 below.

Table 26 ______________________________________ Chemical Analysis of an Expansive Cement Clinker Burned in a Commercial Rotary Kiln Chemical Composition Percent by Weight ______________________________________ SiO.sub.2 16.0 Al.sub.2O.sub.3 3.9 Fe.sub.2 O.sub.3 4.8 CaO 69.8 MgO 3.4 SO.sub.3 0.2 K.sub.2 O 0.4 Na.sub.2 O 0.1 Ign. Loss 1.9 Free CaO 15.7 ______________________________________ The clinker was interground in a closed circuit grinding system with gypsum and a Type I Portland cement clinker in such proportion that the resultant interground expansive cement had a free lime content of about 7 to 8% by weight, and a SO.sub.3content of about 2.2% by weight. The centrifugal separator into which the mill discharge was fed was adjusted such that the Blaine fineness was about 3200 cm.sup.2 /gm and 99.3% by weight of the particles passed through a 100 mesh (Tyler) screen, 94.2%by weight of the material passed through a 200 mesh (Tyler) screen, and about 80% by weight passed through a 325 mesh (Tyler) screen. This cement was called Expansive Composition No. 19. The freshly ground expansive composition was conveyed to a storage silo. Subsequently, four weeks after the grind, the silo was opened and a sample of expansive cement was removed. This cement sample was called Expansive Composition No. 20. Again, at a period of time ten weeks after the initial grind, another sample of cement was withdrawn from the silo. This cement was called Expansive Composition No. 21. The Blaine surface area and free lime content of each Expansive Composition Nos. 19, 20 and 21 is presented in Table 27 below. Table 27 ______________________________________ Surface Area and Free Lime Content of Interground Expansive Cement Stored for Various Periods of Time Expansive Composition Storage Surface No. Time (Weeks) Area (cm.sup.2 /gm) Free CaO(%) ______________________________________ 19 0 3190 7.1 20 4 3290 8.4 21 10 3330 6.8 ______________________________________ At the time each cement was obtained, it was taken to the laboratory and concrete mixes were prepared. The concrete mix made with Expansive Composition No. 19 is called Concrete Mixture No. 19, that made from Expansive Composition No. 20 iscalled Concrete Mixture No. 20, and finally, that made with Expansive Composition No. 21 is called Concrete Mixture No. 21. Each concrete mix prepared contained 44 parts by weight of expansive cement, 103.8 parts by weight of sand, 175.3 parts by weight of coarse aggregate (crushed limestone) and sufficient water to produce a slump of 51/2 inches. Concrete MixtureNo. 19 required 28.0 parts by weight of water, Concrete Mixture No. 20 required 29.0 parts by weight of water, and Concrete Mixture No. 21 required 28.8 parts by weight of water. Each of the three concrete mixtures, at the time of their preparation, was cast into a 5 .times. 14 .times. 37 inch slab containing a length of No. 3 deformed reinforcement steel as described in the introductory paragraphs in the Examples. Theremaining concrete was

cast into standard 3 .times. 3 .times. 10 inch molds, some containing a 1/4 inch threaded steel rod extending longitudinally through the mold and fastened to two steel plates on the opposite ends thereof, and other moldscontaining no steel reinforcement. The 5 .times. 14 .times. 37 inch slabs and the 3 .times. 3 .times. 10 inch bars were subsequently cured and measured as was described in Example II. Initial measurements were taken on the slabs with a Whittemore strain gauge at the age of about six hours when the forms were removed and then periodically up to three months. The expansion of the steel reinforcement bar within each slab isshown in Table 28. The concurrent expansion of the concrete adjacent to the steel reinforcement bar is shown in Table 29. The calculated percent grab of the concrete on the steel reinforcement bar is shown in Table 30. Table 28 ______________________________________ Expansion of Steel Reinforcement Bar in Slabs (Percent) ______________________________________ Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 ______________________________________ 19 .033 .036 .042 .032 .034 20 .032 . 032 .024 .030 .018 21 .034 .036 .023 --* --* ______________________________________ *Data not avilable. Table 29 ______________________________________ Expansion of Concrete in Slabs Adjacent to Steel Reinforcement Bar (Percent) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 ______________________________________ 19 . 049.053 .058 .052 .054 20 .053 .053 .048 .056 .044 21 .048 .057 .042 --* --* ______________________________________ *Data not available. Table 30 ______________________________________ "Grab" of Concrete on Reinforcement Bar (Percent) Concrete Measurement Interval Mixture Days Months No. 1 7 28 2 3 ______________________________________ 19 68 68 72 62 62 20 61 61 50 54 38 21 71 64 55 --* --* ______________________________________ *Data not available. In a similar fashion, the 3 .times. 3 .times. 10 inch concrete bars were demolded at an age of six hours and initial measurements were made with a length comparator. Subsequent measurements were made periodically up to an age of one month. Each bar was then placed in a steam cabinet at a temperature of 180.degree. F for a period of three days, cooled overnight in the moistroom and then measured. Steam-treated bars were further placed in an autoclave and heated for three hours at a steampressure of 300 psi. They were subsequently removed, cooled overnight in the moistroom and measured. The restrained expansion values of concrete bars over the time period of a month and subsequent steam-treatment conditions described are presented in Table 31. Similar expansion data for unrestrained concrete bars are given in Table 32. Table 31 ______________________________________ Restrained Expansion of 3

.times. 3 .times. 10 Inch Bars ______________________________________ (%) Concrete Measurement Interval Mixture Days Condition After 28-days No. 1 7 28 SteamAutoclave ______________________________________ 19 .050 .074 .090 .107 . 142 20 .064 .080 .070 .099 .137 21 .029 .057 .056 .076 .105 ______________________________________ Table 32 ______________________________________ Unrestrained Expansion of 3 .times. 3 .times. 10 Inch Bars ______________________________________ (%) Concrete Measurement Interval Mixture Days Condition After 28-days No. 1 7 28 SteamAutoclave ______________________________________ 19 .165 .207 .216 .268 . 331 20 .234 .274 .266 .289 .343 21 .092 .117 .121 .141 .163 ______________________________________ The significance of this example is that it demonstrates the stability of the expansive composition of the subject invention under normal conditions of plant silo storage. Table 27, in which the storage age of the interground expansive cementvaries from freshly ground material to that which is ten weeks old shows the normal variation in free lime content due to the expected performance of the grinding equipment and the variability of the expansive clinker which was interground with the TypeI Portland clinker. However, in terms of the Blaine fineness of the cement, it is noted that there is a progressive increase in surface area, due most likely to a slow fracturing of cement particles or decrepitation resulting from unavoidable moistureadsorption. The free lime particles in the expansive composition are reactive toward moisture, and any moisture available from the dehydration of the interground gypsum or from air seepage into the silo will result in a premature hydration of the freelime. Thus, we may reasonably expect that upon long term storage, the effectiveness of the expansive composition will tend to somewhat diminish. This will be apparent in the increase of measured Blaine fineness and in a reduced expansive capability ofany concrete prepared with the expansive cement composition. Tables 28, 29 and 30 show the effect of silo storage on the expansive effectiveness of the subject cement. Little change in expansive capability appears to have occurred, although at later ages there may be a tendency for the percent grab todecrease. In all cases, the percent grab is high due to the beneficial effects of the magnesium oxide in the expansive clinker prepared according to the composition of the subject invention. It is noted that a greater degree of decrepitation and lossof expansive capability has been in the past noted in stored samples of pure expansive cement component of high free lime content rather than the interground cement described in this example. A loss of expansive performance, however, is noted in the 3 .times. 3 .times. 10 inch bar data presented in Tables 31 and 32. At the end of one month, restrained concrete bars prepared from 10 week old expansive cement suffered about a 38%decrease in expansive capability when compared with restrained concrete bars prepared from freshly ground expansive cement. In a similar fashion, unrestrained concrete bars prepared from 10 week old expansive cement had a measured expansion which wasabout 44% lower than unrestrained concrete bars prepared with freshly ground expansive cement. Although this

loss was not as obvious in the same concrete cast into 5 .times. 14 .times. 37 inch slabs, and may be exaggerated somewhat by the differentgeometry and smaller size of the 3 .times. 3 .times. 10 inch bars, it indicates that a reduction of expansive potential does occur slowly upon storage, but is much less pronounced with an interground expansive cement than with a pure expansivecomponent. Concrete, at an age of 28 days, which is further subjected to a sequential 180.degree. F steam and then 300 psi autoclave treatment shows the expected additional expansion. This expansion resulting from steam and autoclave treatment also tendsto diminish as the expansive cement used in the tested concrete had been stored for increasing lengths of time. Thus, the same trends of lower expansion noted for concrete prepared from freshly ground expansive cement compared to that sampled four andten weeks later, is equally applied to the magnitude of expansion measured when these same bars were subjected to steam and autoclave treatment. In addition to the foregoing, a comparison of Concrete Mixture No. 19 (freshly ground and commercially burned expansive cement) in Tables 31 and 32 with Concrete Mixture Nos. 6 through 8 (laboratory ground and burned in a small laboratory kiln)in Tables 10 and 11, shows that the extended burning time and mixing of the former composition has resulted in an improved homogeniety and reactivity of clinker minerals, particularly free lime and periclase crystallites, so that the additional expansionobtained due to steam treatment and autoclaving the concrete made with a commercially burned expansive cement was significantly less than that obtained from the steam treatment and autoclaving of concrete made with a laboratory produced expansive cementof the subject invention. In both cases, however, the expansions and percent grab measured by the 5 .times. 14 .times. 37 inch slab technique were superior to those obtained from concrete specimens prepared with expansive compositions not prepared bythe methods of the subject invention described herein. While the invention has been described in relation to its preferred embodiments, it is to be understood that various modifications thereof will now be apparent to one skilled in the art upon reading this specification and it is intended to coversuch modifications as fall within the scope of the appended claims.

Você também pode gostar