CEMENT, may be any bonding agent such as glue, mucilage, certain plastics, putty, solder, asphalt, and hydraulic cement. However, the use of hydraulic cements in the construction and roadbuilding industries has become so important that the unmodified term “cement” now refers almost exclusively to these products.

A hydraulic cement is a bonding agent that reacts with water to form a hard stonelike substance that is resistant to disintegration in water. Most hydraulic cements are specific combinations of silicates and aluminates of lime. Many other combinations of mineral oxides would qualify as hydraulic cements but cannot compete for use in ordinary construction because of their cost. Most hydraulic cement is used in the form of concrete, which consists of cement, water, sand, and gravel or crushed stone. The cement is the bonding agent, and the other rock materials, which are called aggregates, act as filler.

Three classes of cements have been developed commercially: natural cements (including hydraulic limes), aluminous cements, and portland cements.

NATURAL CEMENTS

History. In the United States, the need for a waterresistant mortar became imperative with the development of canals as major arteries of transportation. With the beginning of construction of the Erie Canal in 1817, a search for natural cement rock resulted in the discovery of a suitable deposit near Fayetteville, N. Y. Canals were being built in many parts of the country, and other deposits were discovered and processed in widely scattered locations, including Pennsylvania, Maryland, West Virginia, Virginia, Kentucky, Wisconsin, and Illinois. A mili built in 1828 in Rosendale, Ulster county, N. Y., became the center of the industry, and the term “Rosendale cement” came to be synonymous with natural cement.

The first natural cement was made in small, upright, woodburning kilns that were fired for about a week, after wlıich the clinker was ground between millstones by waterpower. In 1899 nearly 10 million barrels (1.7 million metric tons) of natural cement were produced in the United States, but because of the increased production of portland cement, production of natural cement had dropped by 1918 to less than half a million barrels (85,000 metric tons). Since then production has again increased and is now more than 3 million barrels (513,000 metric tons) annually.

Manufacture. When carbon dioxide is removed from pure limestone (calcium carbonate) by prolonged heating, a process known as calcining, the resulting quicklime (calcium oxide) slakes rapidly in water with the evolution of considerable heat, and the product (calcium hydroxide) forms a putty that does not set under water. Its use as a plaster or mortar is dependent on its interaction with carbon dioxide in the air and the resulting formation of a moderately hard bond of calcium carbonate. The calcined product of such a limestone, which has a high calcium content, is a natural cement called fat lime.

When the limestone contains up to 25% of an argillaceous (claycontaining) material, such as shale, the calcined product reacts slowly with water, there is no rapid evolution of heat, and a hard product is formed that does not disintegrate under water. The calcined products with a relatively low 10 to 20% silica and alumina content are usually called hydraulic limes, while those with a silica and alumina content of 20 to 35% are referred to simply as natural cements. Magnesia may be present in both kinds in concentrations of 10 to 25%.

ALUMINOUS CEMENT

The commercial development of aluminous or highalumina cement is associated principally with the work of J. Bied of France, during the first quarter of the 20th century. This research was initiated in the hope of finding a cement that would be resistant to groundwaters rich in sul~ fates, such as gypsum. A product eventually was obtained that not only possessed the desired properties of sulfate resistance but also hardened more rapidly than the portland cement of that period. The cement was first put on the market in 1918. Aluminous cement is made by heating a mixture of limestone and bauxite until it is molten. The finely ground product, consisting principally of aluminates of lime, has the property of reacting rapidly with water to form a hard rrıass that is resistant to water and sulfate solutions. Its rapid rate of hardening, faster than ordinary portland cement, makes it particularly suitable for use in repair of roads where traffic diversion must be as brief as possible. Its resistance to the action of salt has led to its use in roads that are exposed to ice and snow. Another important application is in the insulation of furnaces where high temperatures are encountered.

PORTLAND CEMENT

History. About 98% of the cement produced in the United States is portland cement, which is not a brand name but a type of hydraulic cement. The name was given in 1824 by Joseph Aspdin, a bricklayer of Leeds, England, to a hydraulic lime that he patented, because when set with water and sand, it resembled a natural linıestone quarried on the isle of Portland in England.

At about the same time it was discovered that an excellent cement could be made by pulverizing the nodules, called grappiers, which occasionally became sintered (that is, formed into a nonporous solid without melting) when hydraulic lime was fired. The resulting cement, produced from the formerly discarded grappiers, was of much higher quality than that obtained from the unsintered material. This fact was firmly established by the English cement manufacturer L. C. Johnson in 1845, and the ternı “portland cement” has since been applied solely to the cement made from the sintered material. This period marks the real beginning of the portland cement industry.

The first portland cement made in the United States was produced in 1876 by David Saylor at Coplay, Pa. It was made in vertical kilns similar to those used for burning lime. The increasing demand for both quantity and quality led to the introduction of the rotary kiln in 1899. This kiln had been invented in 1885 by Frederick Ransome in England but had not been enthusiastically receivea there.

The production of portland cement is a major industry in the United States, increasing from 8 million barrels (1.4 million metric tons) in 1900 —when it trailed natural cement slightly in output—to alnıost 400 million barrels (68.4 million metric tons) annually. (A 376poımd, or 171kg, barrel is the standard unit of weight for hydraulic cement in the United States, even though no cement, except for export, is ııoıv shipped in barrels. The 94pound, or 42.7kg, bag now in general use contains one fourth of a barrel.) The leading cement producing countries are the United States, Russia, Germany, Japan, and France.

Manufacture. The raw materials of portland cement consist principally of a limecontaining material such as limestone, marl, chalk, or shells, and an argillaceous material such as clay, shale, ashes, or slag. The concentration of the constituent calcium oxide, silica, alumina, and ferric oxide must be exact within very narrowly defined limits. These restrictions sometimes necessitate the introduction of other types of rock besides those immediately available, such as a highcalcium limestone, sandstone, or iron ore. On other locations, the available material must be treated to remove excessive amounts of silica, iron oxide, carbon, magnesia, alumina, or alkalies.

Ali raw materials must be ground to a fine powder intimately mixed before burning. In a modern cement plant this is a prodigious task: the rock has to be blasted out of the mountain, fed into a crusher that takes blocks as large as a grand piano, and broken, hammered, and pulverized to the fineness of flour. The clay or shale and other materials then must be thoroughly mixed with the lime rock. Blending of the rock may begin in the quarry and continue as the raw materials flow into each crusher or mili. In the dry process, ali grinding and blending is done with dry materials, and the final mixing is accomplished chiefly in the grinding milis. In the ıvet process, the final grinding and blending is carried out in a water slurry, and the mixing is accomplished both in the grinding milis and by stirring in large vats. In both processes, rigid control of the composition of the final kiln feed is attained through chemical analysis of the raw materials at various stages and the subsequent blending of mixtures.

The water is partially removed from slurries by various processes: by gravitysettling of the solids in a thickener; by filtıation of the water through canvascovered rotating drums, the interiors of which are under reduced pressure; and by evaporation of the water in various kinds of heat exchangers. Often, however, the slurry is fed directly into the kilns, and the water is removed by evaporation, aided by chains or baffle plates inseıted in the back ends of the kilns.

In Europe, the shaft or vertical kiln is stili used extensively because it can be operated with greater fuel economy than the rotary kiln. It is claimed that improvements have been made that make the shaft kiln competitive with the rotary kiln with respect to uniformity and quality of clinker produced. A sintergrate kiln is also used in which the nodulized charge containing coal or coke is burned on a traveling grate. Ignition is produced by a downblast of burning oil and continued by drawing hot air down through the moving charge by means of an exhaust fan.

The rotary kiln has completely replaced the vertical kiln in the United States. The kiln is usually between 300 and 400 feet (92 to 122 meters) in length, although there are some that are more than 600 feet (183 meters) long. The kiln is set at an inclination of about Vz inch per foot (4.3 cm per meter) and rotated at a speed of between 30 and 90 revolutions per hour, causing the load to work its way downward toward the discharge end. There heat is introduced, usually by a blast of ignited powdered coal and air, or less commonly with fuel oil or gas. During the passage of the mixture down the length of the kiln, several reactions take place at various temperature levels.

The liquefaction of some compounds during the burning process causes the charge to agglomerate into nodules of various sizes (usually Y4 to 1 inch in diameter) that are characteristically black, glistening, and hard. This material is known as portland cement clinker. The charge drops from the end of the kiln into some form of cooler and is then ground, usually with 4 to 5% gypsum (hydrous calcium sulfate), to a powder so fine that most of it will pass through a sieve that will retain water. More specifically, the fineness is specified by a minimum average surface area of 1,600 to 1,800 sq cm per gram, depending on the type of cement being made. The purpose of the gypsum is to make the mixture of the cement with water and aggregate (as in fresh concrete mix) remain fluid and workable over a period of several hours. Without the addition of gypsum, the mixture might set before it had been properly placed in the forms.

In the modern cement plant the many operations involved in cement production are inereasingly defined and controlled by instrumentation and automatioıı. From the quarry to the milis to the kilns, the process is in the hands of the operator in the control room, surrounded by dials, flow meters, and closedcircuit television consoles.

Chemistry of Cement Production. The heart of the manufacturing process consists of a number of chemical reactions brought about by the high temperatures in the kiln. By these reactions, the compounds emerging from the kiln in the clinker are totally different from those of the raw material. Dııring the 2 to 4 hours of passage through the kiln, the temperature of the charge gradually increases until it reaches about 1425 to 1540°C (2600 to 2800°F).

Evaporation of free water from the raw materials begins immediately on entering the kiln, but combined water in the clay is retained up to a temperature of about 540°C (1000°F). Magnesium carbonate decomposes to magnesium oxide and carbon dioxide at about 600 °C (1100°F), calcium carbonate to calcium oxide and carbon dioxide at about 900°C (1650°F). The initial interactions of the component oxides— principally lime, silica, alumina, and ferric oxide —begin at the surface of the grains even before any liquid has formed, but such action is mjnimal. Liquid begins to appear at temperatures just under 1320°C (2400°F), the amount and composition depending on the temperature and composition of the raw mix. At maximum temperatures, about 20 to 30% of the charge is in the form of a liquid, which consists principally of the alumina and ferric oxide components, together with the alkalies and portions of the lime, silica, and magnesia.

Interaction between the lime and silica becomes rapid at 1375°C (2500°F) with the formation of dicalcium silicate, which is slowly converted in part to tricalcium silicate. If the temperature fails to exceed 1375 °C, the dicalcium silicate may invert on cooling to an inactive form which, because of a large increase in volume, results in a “dusting” of the clinker with consequent loss in hydraulic value.

If the clinker is cooled slowly, the alumina and ferric oxide components have time to crystallize, and they emerge in the form of tricalcium aluminate and tetracalcium aluminoferrite. Magnesia appears as the oxide, and alkalies as sulfates. With rapid cooling, the liquid may solidify as “undercooled” liquid or glass. Hence, the rate of cooling determines the relative amounts of crystalline and amorphous phases present.

It is essential that the components of the cement raw mix be accurately pıoportioned because each compound or phase in the clinker exerts its peculiar influence on the final properties of the cement when used in mortar or concrete.

Composition. The principal compounds in portland cement are: tricalcium silicate ($\displaystyle 3CaOSi{{O}_{2}}$), which is chiefly responsible for initial set and early strength of the cementwater paste; dicalcium silicate ($\displaystyle 2CaOSi{{O}_{2}}$), which hardens slowly but contributes notably to strength at ages over a month; tricalcium aluminate ($\displaystyle 3CaOA{{l}_{2}}{{O}_{3}}$), which liberates a large amount of heat during the first few days of hardening and is rapidly attacked by sulfate solutions; the ironcontaining phase (a solid solution that approaches the composition $\displaystyle 4CaO\cdot A{{l}_{2}}{{O}_{3}}\cdot F{{e}_{2}}{{O}_{3}}$), which is valuable as a flux in manufacture; magnesia (MgO), which, if present in excessive amount, may cause expansion of structures exposed to moisture after a number of years; and calcium oxide or free lime (CaO), which results from incomplete reaction in the kiln and, if present in amounts over 2 or 3%, may cause unsoundness and expansion in the cement paste. In addition to the above clinker compounds, gypsum or calcium sulfate hydrate ($\displaystyle CaS{{O}_{4}}\cdot 2{{H}_{2}}O$) is interground with the clinker to control the rate of set of the cementwater paste.

Hydration of Cement. The usefulness of portland cement for making concrete depends on a series of reactions between its components and water. The most important of these reactions is the hydration of the calcium silicates to form a colloidal gel of calcium silicate hydrate that solidifies to a hard mass. This material fornıs a continuous phase that surrounds and encloses each piece of aggregate in concrete, and bonds the whole into a rocklike structure. The behavior of the material will be determined by many factors, including the composition and fineness of the cement; the ceraıentwater ratio of the paste; the grading and nature of the aggregate; the time, temperature, and manner of curing; and the presence of entrained air.

Types of Portland Cement. Five types of portland cement are included in the standard specifications of the American Society for Testing Materials and the Federal Specifications Board. The properties of each depend in great part on the relative proportions of the compounds described in the preceding section.

Type I is for use in general concrete construction where the special properties specified for Types II, III, IV, and V are not required. Type II is for use in general concrete construction exp_osed to moderate sulfate action or where moderate heat of hydration is required. Type III is for use where high early strength is required. Type IV is for use where low heat of hydration is required. Type V is for use where high sulfate resistance is required.

Ali types of portland cement may be obtained with or without a specific agent that allows air to be entrained in the paste when the cement is mixed with water and aggregate. The purpose of the entrained air is to improve the durability of the concrete, especially under conditions where cycles of freezing and thawing are encountered.

In addition to the standard types of portland cement, many modified cements have been manufactured. White portland cement is made for special architectural uses and differs from regular portland cement principally in having a low content of ferric oxide. Oilwell cement, made for sealing oil wells, must be slowsetting and able to resist high temperatures and pressures. The substitution of some iron oxide for clay has been found to improve the resistance to sulfate waters, and such cements are manufactured under the names of Ferrari and ironore cement.

Most important of the modified portland cements are the slag and pozzolanic cements made by intergrinding from 15 to 85% of granulated blast furnace slag or pozzolanic material with the portland cement clinker or in some cases with slag and lime or slag and anhydrite.

Pozzolana was named for a fine volcanic roftk from Mount Vesuvius that the ancient Romans found valuable in improving the quality of mortars. The name is now given to a variety of naturally occurring materials such as volcanic ash, trass, Santorin earth, and pumicite, and it is also applied to artificially prepared materials, such as burned clays or diatomaceous earth, that have the capacity to combine in water with lime to form a calcium silicate.