What if some things you hear about portland cement, supplementary cementitious materials, cement paste, entrapped and entrained air, aggregates—coarse and fine—mineral and chemical admixtures, durability, concrete longevity, and strength are not true or only partially true? For example, we all know what causes carbonation and attendant shrinkage but harder paste, the drop in pH from about 12½ to about 9, and the easy way to identify the carbonation zone using phenolphthalein. But what carbonates? Many report only the calcium hydroxide (Ca(OH2) component of cement hydration reacts during the carbonation process—not completely true because all of the cement hydration products react to final destinations; the calcium silicate hydrates to calcium carbonate (CaCO3) and silica gel; the calcium aluminate hydrates to calcite and alumina gel; ettringite to calcium carbonate, alumina gel, and gypsum (CaSO4·2H2O); and the ferrite phase to calcium carbonate, iron oxide (Fe2O3), and alumina gel if alumina is present. Note they all have one common denominator—one of the main end products of carbonation, calcium carbonate (calcite). More recent is physical salt attack, also known as salt hydration distress, which disrupts concrete surfaces initially giving them fine to coarse sandpaper textures.
However, this attack is progressive. The salts usually involved are sodium sulfate and sodium carbonate, which at ambient temperatures and at varying humidity, cycle to different hydration states. When sulfate salts are involved, it is incorrectly called a sulfate attack that implies it is a result of chemical phenomena. This is not true.
The alkalies, sodium and potassium, are thought to be directly responsible for alkali silica aggregate reactions. These alkalies are vehicles through which damage is done, and actually the hydroxyl ions (OH-) with them precipitate the reactions Frequently, “complete hydration of portland cement” is referenced in discussions and in articles, and by some, thought to have occurred by 28 days. Not so—if it ever occurs. Portland cement is a fine powder but, invariably, contains relatively coarse particles that incompletely hydrate and linger forever. The 28-day pick-off point is about where hydration reactions taper off and strength development slows. In these days of very low water-cement ratios, there usually is not enough water to “completely hydrate” the cement Often the term “unhydrated cement” (with respect to portland cement-based products) is bandied about. No such thing exists because when portland cement meets water a chemical marriage always results leaving behind residual cement particles, no matter how incomplete the cement reactions. Once, a concrete dry dock was constructed that, shortly after completion, began to crack. A detailed investigation ensued finding sulfate attack, aggressive bicarbonate attack, and alkali-silica reactivity were to blame. The latter was thought to be the main culprit. For years, crack gages were used to monitor crack widths, and a report concluded that if the expansion continued, the structure would, in effect, self-destruct in 25 years. Coincidently, about 25 years later out of the briny depths came one of the world's first nuclear subs for rehabilitation—at that dry dock. Records and history of the construction and performance were screened and the self-destruct comment noted. A frenzied investigation began using all of man's field and laboratory ingenuity, which included sonic evaluations, compressive strengths, and petrographic examinations. Six-foot-long cores were taken and wax-sealed to keep the “evil” within from escaping. Findings were the same as the original 25-year-old findings, plus new petrographic information showed there was more than ordinary amounts of: (a) epizet (strained calcium hydroxide, CaOH2) from hydration of uncombined lime—free lime); and (b) magnesium hydroxide (brucite, MgOH2) from hydration of magnesium oxide, MgO. Each contributed to the expansion and, in time, as their anhydrous parents were consumed during continued hydration, the expansions abated. The expansions could not be the result of the localized sulfate or aggressive bicarbonate attack because of the massiveness of the dry dock—but could result from the alkali silica aggregate reactions or excessive amounts of free lime and periclase. The dry dock worked well during the repair period and the sub returned to the sea. The dry dock still stands, a tribute to the permanence of concrete even in the face of adversity.
Bernard Erlin is president of The Erlin Co. (TEC), Latrobe, Pa., and has been involved with all aspects of concrete for more than 48 years.
William Hime is a principal with Wiss, Janney, Elstner Associates and began working as a chemist at PCA 54 years ago.