UNITED STATES DEPARTMENT OF COMMERCE Maurice H. Stans, Secretary NATIONAL BUREAU OF STANDARDS Lewis M. Branscomb, Director Exploratory Studies of Early Strength Development In Portland Cement Pastes and Mortars Exploratory Studies of Early Strength Development R. L. Blaine* and L. A. Tomes* A modified vane-shear apparatus was used to measure the shear resistance of neat cement Key words: Cement; cement mortar; cement paste; early strength; false set; hardening of 1. Introduction There is extensive literature dealing with the early hydration reactions of portland cements and the development of structure which causes stiffening and hardening [1-6]. The plasticity as well as the rate of stiffening and hardening of cement pastes, mortars, and concretes are of great importance in the construction industry. The rheology of freshly mixed cementwater pastes, mortars, and concretes have been studied by various means, using viscometers [7], shear apparatus [8], penetrometers [9] as well as pull-out pins [10], soniscope [11], and other devices. A discussion of the various methods was presented by Kelly [12] and the reactions involved have been reviewed by Steinour [13] and more recently by Kondo and Ueda [6]. The Vicat and Gillmore needles are also used for determining time of set, that is, the time at which the neat cement pastes have attained strength adequate to prevent a penetration of the Gillmore needles or a limited penetration of the Vicat needle. It has been stated [3] that "Published information on the viscosity or consistency of cement pastes during the early period of hydration is scanty, however, and somewhat lacking in agreement . . .". One of the problems has been the lack of apparatus sensitive enough to measure small differences in rheological properties but with a sufficient range to evaluate any strength from the time of mixing to the time of hardening. The present exploratory studies are reported at this time to indicate the application of a modified vane-shear apparatus in obtaining information on changes of rheological properties and apparent development of structure of cement pastes and mortars during the first two to three hours after mixing. The vane-shear apparatus has been used in measuring the rheological properties of soils and many articles have been published with respect to its use [14-18]. These articles also give reference to many other publications in the field. One of the problems associated with its use has been the fact that the soil sample is disturbed when the vane is inserted. This would also happen if the vane were inserted in hardening cement paste. It was therefore necessary to modify the apparatus and techniques for use in studying the early hardening of cement pastes and mortars as indicated in the following sections and in the appendix A. The springs were calibrated using a cord attached to a horizontal rod 10 cm from the center of the spring. (see appendix B and Figure 2b) The cord passed over a' pulley to a balance cradle and pan and a series of weights were placed on the pan noting the number of degrees of rotation required to bring the spring and weights to equilibrium. Plots of the rotation values versus the weights resulted in straight lines passing through the origin with each of the springs. The constants for the springs used in this study were calculated in terms of the grams per degree rotation at 1 cm radius. The weakest spring had a value of 1.91 g/deg rotation, the second spring 21.4 g/deg and the strongest spring 44.9 g/deg. The vanes were made of 1.6 mm steel sheets bent at right angles and silver soldered together as indicated in figure 2. They were then machined to size ±0.05 mm and the side edges and bottom edges beveled (see figure 2a) to relieve any friction between the edges and mortar or paste when the vane was rotated. Vanes were made having a width or diameter of 1, and 2 cm. The 1 cm vanes were made for use in a 2.5 cm-deep mold* and the 2 cm vanes for use in a 4 cm-deep mold.** The lengths of the vanes were about 1 cm greater than the depth of the respective molds. The equation used to compute the shear values was as follows: S= If the depth in mortar and paste is constant and only one size vane used, the shear equals a constant times the degrees of rotation. The constants used with the different springs would be different depending on their calibration factor. 3. Cements Tests were made on six portland cements which had previously been used in the Cement and Concrete Reference Laboratory (CCRL) interlaboratory reference sample program. These included two type I (CCRL Nos. 3 & 4), two type IA (CCRL Nos. 5 & 6), and two type III (CCRL Nos. 9 & 10) cements. The chemical composition and results of some of the ASTM Designation C190, Tensile Strength of Hydraulic Cement Mortars. ** ASTM Designation C348, Flexural Strength of Hydraulic Cement Mortars. hysical tests as determined in Cement and Concrete eference Laboratory interlaboratory tests are preented in table 1. These values are the averages of alues reported by all laboratories after outlying reults from a few laboratories had been eliminated. There were 85 to 145 laboratories reporting chemical esults and 104 to 162 laboratories reporting physical esults. The estimated standard deviations of the ime of set values determined by the different laboraories ranged from 14 to 28 min for Vicat time of et, from 22 to 36 min for Gillmore initial set and from 4 to 56 min for Gillmore final set. 4. Methods of Tests A number of preliminary tests were made using the lifferent sized vanes in different molds before adopt ing the techniques presented in this report. Further information about the preliminary tests will be presented in the discussion. Tests were made on neat cement pastes mixed in a mechanical mixer in accordance with specification ASTM Designation C305-65 except that 1400 g of cement were used. The percentage water was that given in table 1. After mixing, a scoop of paste was placed in each of 3 molds of a 3-gang briquet mold and this was then vibrated sufficiently to cause the paste to flow to the edges,-about 5 to 10 s. The excess paste was cut off,-the mold trowelled once and then cut off with a sharp trowel using a sawing motion. One of the 1 cm vanes was placed in each half of the briquet at about the center of the broadest dimensions making six vanes for each mold. The mold with paste and vanes was then given 2 or 3 TABLE 1. Average values for chemical analyses and physical tests reported for Cement and Concrete Reference Laboratory interlaboratory tests on six cements |