www.mrs.org/publications/bulletin HISTORICAL NOTE

Hydraulic Cement: Lost and Found In 1756, John Smeaton, a British engi- continued until the 18th century. using equal parts of siliceous limestone neer, was presented with the challenge of So Smeaton knew of the ancient from the Blue Lias formation at Aberthaw building a sturdy lighthouse on a site hydraulic cements at the time of his in South Wales, and pozzolana from where two earlier lighthouses had fallen. Eddystone assignment, but was generally Civita Vecchia in Italy. But the quest to The location was Eddystone, off the coast surrounded by inferior nonhydraulic understand and ultimately to manufac- of Cornwall, where navigating conditions varieties. The only contemporary example ture synthetic cements had just begun. were treacherous and shipwrecks fre- he could follow was that of Holland, a Spurred by the industrial revolution and quent. The previous lighthouses, each notoriously water-logged country that laws like the London Building Act of 1774 made of wood, had collapsed in stormy had succeeded in building sea defenses that mandated the use of fire-resistant weather, and Smeaton was determined to by combining “tarras” (a local type of stuccos and concretes in place of exposed use stone this time. His problem was find- pozzolana) with lime to form a hydraulic timbers in new construction, the search for ing a cement that would endure under cement widely known as “tarras mortar.” cementitious raw materials flourished. In water—a hydraulic cement. Since the Before beginning his experiments, 1796, James Parker of Northfleet found method of making such a material was Smeaton toured Holland in 1754–1755. He that the nodules of limestone called “sep- not well understood, Smeaton undertook saw that the Dutch formulation worked, taria” found mainly along the coast of the a systematic study of hydraulic cements. but wanted to know why it worked. Isle of Sheppey made an excellent cement The first hydraulic cements are attrib- when burned at high temperatures and uted to the Greeks, whose addition of a ground into powders. This material, volcanic tuff from the Island of Santorini which Parker patented under the name (so-called Santorin earth) to burned lime- The first hydraulic cements “Roman Cement,” enjoyed great populari- stone produced cement that could harden are attributed to the Greeks, ty until mining of the septaria within 50 under water. The ancient Romans added whose addition of a volcanic feet of the clay cliffs was banned in 1845 a tuff called “pozzolana” from the region due to the environmental damage it was of Pozzuoli near Naples at the base of tuff from the Island of causing. Other sources, such as Medina Mt. Vesuvius. This “pozzolana cement” Santorini (so-called Santorin cement from the Isle of Wight and was used to hold up structures that are Calderwood cement from Calder Glen in still standing such as the Coliseum in earth) to burned limestone Scotland, were also exploited until natural Rome and the Pont du Gard aqueduct at produced cement that could deposits were depleted. Nîmes in France. In general, the use of a harden under water. But chemistry was now sufficiently strong mortar made the construction of advanced to provide insight. In 1813, a thinner walls possible. Cement can thus be scientist named Descotels wrote, “It credited with inspiring some of the grace- Smeaton studied the effects of four vari- appears very probable that the essential ful architecture of Roman civilization. ables on cement quality: the type of lime, condition for a limestone to furnish good But no one knew why pozzolana the type and quantity of tuff added, the lean lime is that it contain a large quantity cement worked as it did. Clearly, some use of fresh or salt water in the mix, and of siliceous material disseminated in very material in the volcanic-earth additive the addition of varying proportions of fine particles.” turned a nonhydraulic cement into a gypsum. After formulating an experimen- Five years later, L.J. Vicat in France and hydraulic one, but chemistry had not pro- tal mix, he formed the mortar into a ball J.F. John in Holland independently sum- gressed far enough to provide an analysis. two inches in diameter, then let it sit in air marized the accumulated data and A Roman writer named Vitruvius to set before being submerged in water. expressed what was becoming obvious: described the Roman methods of making “What happened to the ball in this In the absence of natural deposits of lime- cement, but his speculations on the mech- state,” he wrote, “was the criterion by stone having the necessary proportions of anisms at work were burdened by the which I judged of the validity of the com- clay compounds, an artificial mixture of alchemical theories popular at the time. position for our purposes.” clay and limestone would suffice. Nonetheless, his surviving writings are The results were surprising. When he “We see that, by being able to regulate credited by some for the later revival of began his experiments, it was generally the proportions, we can give to the face- hydraulic cements. considered that the harder, purer lime- tious lime whatever degree of energy we When the Roman Empire fell, the stones were preferable over the softer, please, and cause it at pleasure to equal “secret” of hydraulic cement disappeared clayey ones for use as mortars. But or surpass the natural limes,” Vicat with it. Cements continued to be used, but Smeaton found just the opposite: the wrote. This fundamental understanding the omission of volcanic earth, and the clayey limestones formed much stronger of the composition and reactions of calci- general lack of care in burning the lime to cements than the hard limestones. um-, silica-, and alumina-containing raw a sufficiently high temperature, rendered “An admixture of clay in the composi- materials freed manufacturers from seek- these mortars nonhydraulic. The quality tion of a limestone might be the most cer- ing better limestone quarries and tuff of the cement generally depended on the tain index of the validity of a limestone for deposits; they could instead concentrate chemistry of the limestone in the local aquatic buildings,” Smeaton concluded. on optimizing formulations and manu- quarry. If by chance the limestone con- Later, chemists would show that it is facturing processes. tained some of the silica and alumina precisely the hydrated aluminosilicate Joseph Aspdin, a British bricklayer compounds found in the volcanic tuff, the clays that provide the necessary reactants from Leeds, in 1824 patented the first cement might be of high quality, and at to turn the calcium carbonate into the “Portland Cement,” so named because least semihydraulic. But those with little of strong calcium aluminate and calcium sil- when it hardened it resembled the stone these cementitious materials would not icate cementitious materials. Smeaton obtained from the quarries in Portland, hold up under water. This state of affairs built his Eddystone lighthouse in 1760 England. But Aspdin’s product appears to

726 MRS BULLETIN/SEPTEMBER 2001 HISTORICAL NOTE

be no more than a minor variant on the samples made of these vitrified materials Raw Materials, Manufacture, Testing, and standard Roman cements of the time, and proved to be superior in color and hard- Analysis, Third Edition (The Chemical certainly not deserving of the reputation ness. Variations on these original mix- Publishing Co., Easton, PA, 1926); that later Portland cements have gained. tures and further experimentation led to Norman Davey, A History of Building Rather, it was probably I.C. Johnson to the practice of “incipient vitrification” Materials (Phoenix House, London, 1961); whom credit for the discovery of the key that is the basis for the high quality of Robert H. Bogue, The Chemistry of to modern Portland cement should be modern Portland cements. Portland Cement (Reinhold Publishing given. In 1845, Johnson’s experiments TIM PALUCKA Corp., New York, 1955); and Edison with higher calcining temperatures pro- Portland Cement Company, The Romance duced a vitrified material that was initial- FOR FURTHER READING: Richard K. of Cement (Livermore and Knight Co., ly discarded as unusable; later, cement Meade, Portland Cement: Its Composition, New York, 1926).

MRS BULLETIN/SEPTEMBER 2001 www.mrs.org/publications/bulletin 727