Popular Science Monthly/Volume 60/December 1901/Cement for a Modern Street
| CEMENT FOR A MODERN STREET. |
By Dr. S. F. PECKHAM.
A MODERN street consists of a concrete foundation which extends from curb to curb, upon which is laid a wearing surface of asphalt, brick or other material.
The use of these concretes has an instructive history, which might be profitably preceded by a discussion of the uses of mortars and cements in antiquity, did space permit. So far as I have been able to learn, all the different varieties of cementing materials, including ordinary lime mortars, have been experimented with for the construction of concrete foundations. It is therefore proper that these different materials should be briefly described.
Mortar, in the ordinary sense of the term, designates a mixture of lime and sand. The lime is prepared by heating limestone in kilns, until the carbonic acid of the limestone is expelled and oxide of calcium remains, which readily absorbs water and slacks, as it is termed, and in time reabsorbs the carbonic acid that was driven off. The lime is mixed with the water when it is slacked and the carbonic acid is absorbed from the atmosphere. When mortar is made, the lime is first made into a thin paste with water, and sand is added until the mass ceases to be sticky. Such mortar acquires strength slowly. The excess of water first dries out and then the lime by slow absorption of carbonic acid forms thin particles of limestone between the grains of sand, until the mortar becomes a coherent mass. That this process goes on very slowly is shown by the fact that the mortar between the bricks of chimneys centuries old is found to contain a considerable percentage of unchanged lime. This mortar, when first laid, will not bear wetting, and will set only in dry air.
The Romans had learned before the Christian Era that the addition to lime mortar of volcanic ash or pozzuolana would make the mortar set under water and with additional strength. The so-called Roman cement was noted in antiquity for its superior strength when compared with ordinary lime mortar. Where they could not obtain the pozzuolana they used pulverized brick and pottery.
During the middle ages, for more than a thousand years, the art of making hydraulic cement was lost, and, with every other art, the art of making good mortar declined until the beginning of the eighteenth century, when attempts were made to revive the art of making Roman cement, but with only slight success.
In the year 1756, the celebrated engineer, John Smeaton, was wrestling with the problem of constructing the Eddystone Lighthouse, While experimenting with different varieties of mortar, he discovered that certain limestones produced an hydraulic lime. He found that a mortar made of pure lime and pozzuolana or powdered bricks gave only unsatisfactory results, but when an impure lime from the 'Alberthaw' was used the hydraulic properties were more fully developed. Continuing his experiments, he at length announced that only limestones containing clay produced a lime of satisfactory hydraulic properties.
Speaking of this discovery, Smeaton says in his 'Narrative of the Eddystone Lighthouse':
It is easy to add clay in any proportion to pure lime, but it produces no such effect; it is easy to add brick dust, either finely or coarsely powdered, to such lime in any proportions also; but this seems unattended with any other effect than what arises from other bodies, becoming porous and spongy and therefore absorbent of water as already hinted and excepting what may reasonably be attributed to the irony particles that red brick dust may contain.
In short, I have as yet found no treatment of pure calcareous lime that renders it more fit to set in water than it is by nature, except what is to be derived from the admixture of trass, pozzuolana and some ferruginous substance of similar nature.These investigations and the conclusions that he drew from them led Smeaton to use in the construction of the Eddystone Lighthouse a mortar or cement composed of hydraulic lime from the Alberthaw and Italian pozzuolana. A step or two farther in his investigations, which he did not take and which were not taken until the middle of the last century, would have led to the Portland Cement of the present time.
In 1796, a Mr. Parker, of London, patented a process for what he called 'Roman Cement." He used for this purpose certain nodules of limestone containing clay that were found along the coasts of the Isle of Sheppy and certain parts of Kent and Essex. These nodules were first calcined and then reduced to fine powder in mills. The result was a cement of a better quality than Smeaton's. In 1818, one Canvass White discovered and patented in the United States a process for making cement from a similar rock, found at Payetteville, in central New York. Large quantities were manufactured and used in the construction of locks on the Erie Canal, which was then being built. The State of New York purchased the patent and made it public property. This laid the foundation of a great industry, which is known generally as the 'Natural Cement Industry' but locally in the neighborhood of New York City as the Rosendale Cement Industry.
Continuing these experiments it was found that the rocks found in a few localities produced cements of a quality superior to those in general use. This was particularly true of a stone found in the Island of Portland. At length artificial mixtures of limestone and clay were made and burned under such conditions that the resulting cement very closely resembled the natural cement made from the Portland rock. These results led to the adoption of the name of Portland for all cements of this class whether made in England, where the name originated, or elsewhere. The first attempts to prepare a cement by artificial mixtures, in imitation of the natural Portland rock, were made in France about 1802.
Portland cements, as at present manufactured, were first made in England by a process that was patented in 1824, although there had been several patents for 'Portland Cements' previously issued. In the patent specifications of 1824 occurs the following description of the process used:
It will thus be seen that at the middle of the last century there were in use: Common lime mortar, consisting of slacked lime and sand; made all over the world and used for common masonry and plastering.
Also Roman cement, made by mixing common lime and some dry aluminous material, like pulverized tufa, brick or slag. This was stronger than common mortar and slightly hydraulic.
Also, natural cement, called Rosendale cement in the United States, made by burning and grinding a natural limestone containing clay. These natural cements are of very varying quality and are hydraulic, i. e., will set or harden under water.
Also, Portland cements, called also Artificial cements, made by grinding limestone or marl, both of which are nearly pure carbonate of lime, and mixing it in proper proportions with ground clay, which is a silicate of alumina containing a variable proportion of the oxides of iron. This mixture is calcined at a temperature that will produce semifusion and the resulting clinker is ground to a fine powder. The powder is 'aged' in order to partially slack the lime. The powder is mixed with sand and water to the consistency of mortar and used. Like natural cement, Portland cements are hydraulic, and they make the strongest mortars known.
A modern street consists of a foundation of broken stone that is formed into a concrete or solid mass of masonry by admixture with mortar. The character and quality of this mortar are a matter of the greatest importance. All of the four kinds of mortar mentioned above have been used for this purpose.
An experiment was made in London by laying in Holborn, opposite Gray's Inn, nine inches of lime mortar concrete with a floating on top of 3⁄4 inches of lime mortar. Upon this foundation was laid a surface of Val de Travers asphalte. When the concrete was ready to receive the
asphalte, a fire broke out in Holborn; the place was flooded with water, the engines drove over the concrete and the population of Gray's Inn trampled it down. It was subsequently made good and the asphalte spread. Tor five or six years the road was kept up at considerable expense and then relaid. On removing the asphalte, it was found that the lime concrete had never set, that the mortar floating had never adhered to the concrete, but was mostly in powder, produced either by the action of the rammers or by the traffic afterwards.
Roman cement was tried in Paris and condemned for street foundations. There remains for use for this purpose natural cement and Portland cement. The details of the process of manufacture will now be given.
In the United States the manufacture of natural cements is chiefly carried on in the Lehigh Valley, near Louisville, Ky., at Akron, Ohio, at Milwaukee, Wis., and at Glens Falls and other points in the State of New York. While the rocks occurring at these different points are not identical, either in geological age or in chemical composition, they are in many respects similar. In geologic age they are of carboniferous age or older and in chemical composition they consist of limestones in which clay occurs, either uniformly disseminated throughout the rock, forming a very intimate admixture, or else interstratified with the layers of limestone, so that when the rock is broken up and burned, the resulting mixture of the constituents of the rock is very intimate. Yet intimate as the mixture is, both before and after burning and grinding, in all the ledges the rock has to be sorted and mixed in the same quarry with the greatest care, in order to insure a uniform product from the same works. From the different localities the output is sufficiently different to give the Louisville, Milwaukee and other brands distinctive, though unimportant, characteristics.
At all the natural cement works substantially the same method of manufacture is followed, although the details are modified to suit different localities.
One of the most extensive natural cement plants in the country is that of the Milwaukee Cement Co., at Milwaukee, Wis., the officers of which have kindly furnished the accompanying illustrations. Fig. 1 shows the tramway approaches to the kilns, which are arranged in a double set of ten each. The rock is quarried in the immediate neighborhood and is run in tram-cars, which are seen in the middle foreground, up the inclines to the top of the kilns into which the rock is thrown. The trestle on the right is the dump for coal, which is also loaded into tram cars, one of which is seen at the chute, and run up the incline to the kilns. The rock and fuel are thus conveniently supplied to the kilns at the top, while the burned cement is removed from the kilns at the bottom. Fig. 3 shows one of the kilns on the left; the grinding and shipping house in the center, with the inclines up which the burned cement is hauled and the railroad tracks over which the cement is shipped in all directions from Milwaukee. Fig. 3 represents the grate at the bottom of the kiln, from which the burned cement is removed, while fresh rock and fuel are supplied at the top, thus making the action of the kilns continuous.
Two obstacles make it impossible to prepare a theoretically perfect cement from the natural rock. The first is a lack of uniformity in the rock itself as it occurs in the quarry. This difficulty is obviated as nearly as is possible by careful sorting. By which the least desirable rock 
Fig. 2. Inclines entering grinding House and Tracks connecting with all Railroads entering Milwaukee.
Fig. 3. Bottom of Kiln, Campbell Patent, showing Calcined Material
is rejected and the varied masses selected are carefully mixed so as to ensure a uniform grade. The second is the difficulty of burning the rock in kilns and in large masses so uniformly as to ensure complete burning and no considerable amount of overburning. If natural cement rock is overburned or fused, it becomes a slag, and loses its hydraulic properties. It is not surprising, therefore, that a considerable lack of uniformity exists in the quality of the natural cement found upon the market. The best of them contain a considerable amount of impurity, or material that is not cement, that exists in the rock before it is burned, and also a considerable amount of unburned rock, which together serve to dilute the cement proper, as if a certain amount of sand had already been added to the cement before it is used. These impurities that are inherent in the nature of the materials from which the natural cement is made and also in the process of manufacture that is of necessity followed, result in a cement that can be made and sold at a less price than Portland cement and that is inferior to it for many purposes, while, on the other hand, for a great many purposes natural cements have been found to answer every requirement and are made and used in enormous quantities.
For Portland cements, either a very pure natural limestone or marl is selected and brought into a very finely pulverized condition. Lime- stones are selected as free as possible from every impurity except clay. ]Magnesia is never absent, and at best is an inert impurity, but the amount present should not exceed five per cent. Marl is frequently used and is generally purer than limestone. In England chalk is gen- erally used. In Germany chalk and a limestone, locally known as 'mergel,' which is soft and contains clay, are employed.
The following table, No. I., taken from 'Cement Industry,' page 1 2, gives the composition of the carbonate of lime in use in some of the leading Portland cement manufactories in the United States: .
Table I.
| Limestones and Marls. | Chalk, England, (Reed). |
Cement Rock, La Salle, Ills. |
Cement Rock, Phillips- burg, N. J. |
Cement Rock, Siegfried, Penna. |
Marl, Sandusky, Ohio. |
Marl, Syracuse Ind. |
| Calcium carbonate | 98.57 | 88.16 | 70.10 | 68.91 | 91.77 | 88.49 |
| Magnesium" | 0.38 | 1.78 | 3.96 | 4.28 | 0.53 | 2.71 |
| Calcium sulphate | 3.19 | 1.58 | ||||
| Silica | 0.64 | 8.20 | 15.05 | 17.32 | 0.22 | 1.78 |
| Alumina | 0.16 | 1.00 | 9.02 | 7.07 | 1.22 | 0.91 |
| Iron oxide | 0.08 | 0.30 | 1.27 | 2.04 | 0.40 | 0.30 |
The clay should be highly siliceous, but should be free from grains of sand. Clays containing carbonate of lime or marl are softer and more easily mixed with the other materials. Clays containing 70%, or more, of silica stand firing without fusing, produce a clinker that is easy to grind and yield a cement that sets slowly and gains strength over a long period. On the contrary, highly aluminous clays give a fusible clinker and quick-setting cement. A high authority states that the percentage of silica in the clay should be three times the percentages of the alumina and iron taken together. The less iron the clay and limestone contains the lighter colored will be the cement.
The following table, No. II., also taken from 'Cement Industry,' page 13, gives the composition of the clays in use in several Portland cement manufactories:
Table II.
| Clays. | Medway, England. |
Harper, Ohio. |
Sandusky, Ohio. |
La Salle, Ills. |
| Silica | 70.56 | 51.50 | 65.41 | 54.30 |
| Alumina | 14.52 | 13.23 | 16.54 | 19.33 |
| Iron oxide | 3.06 | 3.30 | 6.06 | 5.57 |
| Lime | 4.43 | 11.52 | 2.22 | 3.29 |
| Magnesia | 3.45 | 1.88 | 2.57 | |
| Carbonic acid | 3.48 | 12.85 | ||
| Alkalies | 3.95 |
A large part of the Portland cement manufactured in the United States is made from natural cement rock, that is, from a rock that contains both the carbonate of lime and clay, very intimately mixed in a natural rock. The best cements, however, are made from an artificial mixture of either limestone or marl and clay. The proportions are determined after very careful chemical analysis in such manner that the several ingredients that form cement shall not only be free from harmful substances, but shall combine to produce theoretical chemical compounds in certain quantitative relations.
Although much has been written for many years concerning the chemistry of hydraulic cements, it is only within about 25 years that researches have been conducted in such a manner that by constructing the compounds possessing hydraulic properties from pure elementary materials, much light has been thrown upon the problem. The French chemist Vicat suggested an 'hydraulic index' to designate the hydraulic properties of different cements, which is a figure representing the number of parts of silica and alumina combined with 100 parts of lime and magnesia.
In 1872 Erdmenger showed that in commercial Portland cements the ratio of lime to the acid constituents, silica, alumina and iron oxide taken together, averaged 1.9. At about the same date, Michaelis determined the ratio, as between 1.8 and 2.2, and called it the 'hydraulic modulus.' These ratios no longer represent the composition of Portland cements as with improved methods of manufacture the proportion of lime has steadily increased, until the 'hydraulic modulus' is no longer applicable to the varying conditions and materials under which the cements are now manufactured.
Since 1890 great progress has been made in the United States in the application of theoretical and scientific principles to the technology of Portland cements, and the result has been an enormous expansion of the business with an improved quality of the product.
The original method pursued in England, and largely adopted elsewhere, was to grind the materials very wet, floating off the fine particles to a large tank where they were allowed to settle. The settling and drying required a great deal of time. This method was followed by a dry process in which the materials were ground together dry and then moistened sufficiently to be molded into briquettes. The briquettes were then dried and stacked in a kiln and burned. The introduction of rotary kilns rendered the molding and drying unnecessary as either dry or wet materials, as a dry powder or wet mud are fed directly to the kilns. The composition of the materials, whether it
be cement rock or mixed materials, must be maintained by constant chemical analysis, as the percentage of carbonate of lime should not vary by more than 1⁄2 per cent, from that found correct for the other materials used.
From the foregoing statements it will be observed that two quite different methods of manufacture are followed. In the first the cement 'mix' is molded into briquettes, which are dried, stacked in a kiln and burned. For the burning by this method three different forms of kiln are used: 1. The intermittent dome kilns that resemble in their operation common lime kilns. 2. Continuous kilns, of the Ditzsch or Schœfer patterns. 3. The Hoffman ring-kiln. These kilns are economical in fuel, but expensive in time and labor. The dome-kiln was the first used, but has now disappeared from the United States, and survives in Europe only in a few localities. The continuous kilns require highly skilled labor, and are used only to a limited extent either in the United States or Europe. The Hoffman ring-kiln is widely used in Europe, but has found few patrons in the United States. The second method of manufacture of Portland cement is by the use of the rotary furnace, into which the mix is fed as a dry powder or ad a wet mud. Although the rotary cement furnace was originally patented in England, by Frederic Ransome, in 1885, it has been improved in the United States to such an extent that it has become practically
an American invention. Ransome's patent required the use of gas for fuel and the first rotary kilns installed in the United States required gas, but gas was soon replaced by crude petroleum. This fuel took the place of gas entirely, and was exclusively used until it was replaced by pulverized coal, blown into the kilns and burned in a jet like gas or sprayed petroleum. The current or blast of air which carries the powdered coal into the kiln furnishes the oxygen for its complete combustion. This is a convenient method for burning the cheapest fuel on the market, and while it is not economical of fuel it is very economical of labor and is very uniform in its action, which is an extremely desirable condition in the cement industry.
I am indebted to the obliging courtesy of the officers of the Virginia Portland Cement Co., of Craigsville, Augusta Co., Va., for the accompanying illustrations of a Portland Cement Plant.
Fig. 4 is a general view of the works, which it will be seen at once are very unlike the natural cement works, at Milwaukee. All the operations of a Portland cement works are under cover. Fig. 5 represents
the stone house where the materials are received and sorted, preparatory to being finely ground. A great variety of mills are used for grinding both the crude materials and the cement clinker. So far as the making of the cement is concerned, it does not matter in what kind of a mill the ingredients may be ground, provided they be ground fine and thoroughly mixed in the right proportions. If the mixture is burned dry, the mixing is accomplished by the use of screens and sieves; if it is burned wet, the grinding is done in a wet mill, the paste being floated off and allowed to settle in large tanks. The dry materials are blown into the kiln. The wet mud is allowed to drip into the upper end of the kiln as it is forced in by a pump.
Fig. 6 represents the rotary kiln. It consists of a slightly inclined steel cylinder, about 60 feet in length and 6 to 7 feet in diameter, lined with fire-brick, and revolving by means of powerful gears at the rate of one revolution in from one to three minutes. Fig. 7 shows the arrangement of apparatus for injecting at the center of the kiln an air blast which carries with it the powdered coal, received from the hopper shown on the right. The rotation of the kiln keeps the 'mix' in constant motion as it passes through the kiln, when it is first dried, then deprived of its carbonic acid and then vitrified or partially fused in such manner as to insure the proper chemical reaction between the basic lime and the acid silica, alumina and iron. Only that skill that is determined by experience can direct the burning at such a temperature that the continuous operation of the kilns will result in a clinker that is neither underburned nor overburned.
So far as the chemistry of cement burning is understood, it appears that at a red heat the water is expelled from the clay; the carbonic acid is then driven from the lime, and it escapes. The silica, alumina and iron of the clay then combine with the lime, first forming fusible glasses and then taking on more lime; at length the tri-calcium silicate informed with the alumina and iron as calcium alumino-ferrite.
Properly burned clinker is in hard rounded grains about the size of dried peas and of a greenish-black color. If it is underburned, it is light colored and soft. If it is overburned it becomes like slag. If it is burned too long, it falls to powder on cooling. Uneven burning is more common in vertical than in rotary kilns, hence the product of rotary kilns is more uniform, ensuring a better cement as the burning is performed quickly and is quickly arrested when the process is completed.
Experience has proved that the burning of cement is an important factor in its manufacture as determining its qualities. A cement that is properly proportioned, thoroughly mixed and well burned and finely ground will set in about two hours after mixing with water, will harden well after setting and will continue to harden through long periods, even to several years. Ground gypsum is frequently added to cement to control the setting.
Cement should be finely ground. Ninety-two to ninety-three per cent, should pass a sieve of 100 meshes to the linear inch.
Until within the last decade, the Portland cements manufactured in the United States were generally inferior to the best English, German, Belgian and French brands. While these foreign cements have been maintained with the highest degree of excellence the American brands have been greatly improved until at the present time there are no better cements made anywhere in the world than in the United States.