Journal of Materials in Civil Engineering

Many permeation methods for characterizing the near-surface quality of concrete claim to have the added advantage of being applicable in situ. This paper attempts to highlight some of the aspects associated with using such tests in situ, and deals specifically with experience in using the covercrete absorption test (CAT) for in-situ tests on concrete structures. Some of the better known permeation test methods are listed, and reasons are given for selecting the CAT for this investigation, in preference to other methods. Results of CAT tests performed in the laboratory on laboratory cast samples showed the CAT can be sensitive to variations in water/cement ratio and curing time, moisture content at the time of test, and surface orientation. Good correlations between covercrete absorption and other durability index parameters were obtained for the laboratory-cast samples, but these correlations were poor or even absent when applied to in-situ tests on mature concrete structures. From these tests it appears that permeation methods have little value for assessing the quality and state of durability of concrete structures.

The repair performance of resinous and cementitious materials exposed to static and fluctuating temperature conditions has been evaluated in this study, which is of importance for the repair of concrete structures in the Arabian Gulf region. Specimens were exposed to a total of 90 thermal cycles between 25°C and 70°C, simulating the temperature variation of concrete surfaces on typical summer days in the Arabian Gulf region. The slant shear bond strengths as well as failure characteristics have been observed at 0, 60, and 90 cycles. The results show that the shear slant bond strength undergoes significant reduction with thermal fluctuations due to the thermal incompatibility between the concrete and the repair materials. It is found that for resinous materials the reduction varied from 9.3–20.47% for 60 cycles, and from 18.98–36.43% for 90 thermal cycles. For the cementitious materials, the corresponding values were 3.2–17.46% for 60 cycles, and 8.07–34.80% for 90 cycles. It is also seen, in general, that the mode of failure of the test specimens changed from crushing of concrete to combined crushing-joint failure at 60 cycles, and then to a distinct joint failure at 90 thermal cycles.

A unified size-effect model based on the weakest-link brittle fracture theory has been developed to represent statistical size effects in visually graded structural lumber. Test results from published studies and new results from a comprehensive study of strength properties of Canadian visually graded lumber have been used to develop the model parameters. Analyses have shown that size-effect factors are not statistically different across grades, species, or strength levels. For international harmonization of size-effect adjustment procedures for characteristic lumber strength properties, factors have been suggested for (1) Length and width effects for tension and bending members to equal 0.2; and (2) length and width effects for compression members to equal 0.1.

A procedure for defining the appropriate relative quantity for the large molecular size (LMS) fraction in asphalt was developed based on the correlation between chromatogram and selected physical properties. The physical properties and chromatogram of two viscosity grades of three asphalt cements, conditioned at three artificial aging times, were analyzed. Each chromatogram was divided into 30 equal-time slices and the relative quantity of each slice was calculated. Correlation between relative quantity of LMS and each physical property was evaluated through regression analysis. The relative quantity of individual slices was also analyzed to establish correlation with each physical property. The results of analyses using the procedure suggested in this study indicated that in the front part of the chromatogram there was a portion that had a high correlation with selected asphalt properties. It was therefore suggested that generally accepted LMS should be defined by this portion. It was also found that there was an individual slice, within the LMS portion, by which the three properties could be best predicted.

Long-term age-hardening behavior of asphalt cements was simulated by the thin-film-oven test procedure with extended oven durations. Seven AC-20 asphalt cements from different crude sources were aged for 5, 14, and 24 h. Testing of selected physical properties (e.g., absolute and kinematic viscosity) and high-performance gel permeation chromatography (HP-GPC) analyses were conducted on the virgin and aged asphalt cements. HP-GPC profiles were partitioned into three, four, and 10 segments (HP-GPC parameters) based on equal elution time. Statistical analysis conducted on HP-GPC profiles indicated that long-term laboratory aging produced significant differences among HP-GPC parameters. It also indicated that significant compositional differences existed among the AC-20 asphalt cement sources. Regression models were developed for predicting physical properties of asphalt cements based on HP-GPC parameters. Strong relationships ( R² > 0.75) were established between four physical properties and HP-GPC partitions into tenths. Partitioning of chromatograms into 10 segments provided a better resolution of the data than the division into thirds and fourths.

A finite-element program to evaluate the stresses found in finger-jointed lumber subjected to uniaxial tension has been developed and verified. This program exploits a unique contact element useful in modeling a thin, isotropic boundary (glueline) between two anisotropic adherents. A parameter study was conducted using the aforementioned model to evaluate the influences of certain key finger-joint properties on the stress distributions induced in the lumber. Three joint geometries (step, scarf, and finger), three combinations of adherents (isotropic materials on both sides, orthotropic materials on both sides, and an isotropic material on one side and an orthotropic material on the other), and two glueline thicknesses (0.01 and 0.001 times the joint length) were studied. Glueline thickness is directly related to glueline stiffness, which is known to influence the stresses in the glueline, and consequently, in the adherents. The results showed that maximum adherent stresses were developed at the interface with the glueline. For isotropic materials with constant ratios of the elastic parameters between adherent and adhesive (e.g., E_adherent /E_adhesive), stress distributions are identical, as expected. Stiff gluelines, relative to the adherent stiffness, tend to increase stress concentrations at the edges of all three types of joints. In step joints, the large fingertip width was the most influential factor in accounting for the high stress concentrations. In contrast to the step joint, stress concentrations in scarf joints were highly sensitive to differences in material properties between adherents. It was concluded in this study that the development of a finger-joint manufacturing process that produces fingertips of near-zero width is desirable.

In this paper, Schapery's nonlinear elastic-viscoelastic correspondence principle based on pseudostrain concept is introduced as a means of separately evaluating important mechanisms of asphalt concrete under cyclic loading, including time dependence, damage growth, and fracture healing. Earlier work by Kim and Little and the findings from this present study have demonstrated that the correspondence principle can be applied to most of the testing conditions encountered in testing of asphalt concrete. The importance of using proper analytical representation of relaxation modulus in course of calculating pseudostrain is demonstrated using the cyclic data. At the loading frequencies typically used in asphalt-concrete testing (faster than 10 Hz), the creep or relaxation data must be represented in a form that can accurately describe the short time behavior, such as the modified power law. It is concluded that the correspondence principle provides a means of more accurately modeling the damage growth and fracture healing of asphalt concrete under complex cyclic loading.

The plastic sheeting approach for negative skin friction reduction in end-bent piles behind proprietary mechanically stabilized earth walls was investigated along with the conventional bitumen coating method. A number of concrete blocks were prepared and furnished with different arrangements of polyethylene sheets and bitumen materials, and tested in the laboratory using a direct-shear apparatus. Shearing resistance of the concrete-bitumen-soil samples was influenced by temperature, shear rate, and normal stresses. Bitumen-coated samples tested with crushed limestone exhibited the highest shearing resistance. Test results showed that wrapping concrete piles with polyethylene sheets reduced the skin friction by 78%. Additional reduction was achieved by lubricating the polyethylene sheets with oil. The effectiveness of oil-lubricated polyethylene sheets in reducing skin friction was about 94%. Factors that affect the shearing behavior of bitumen coating did not influence the friction characteristics of the polyethylene sheeting. Based on the results of this study, two formulas were suggested for estimating the shearing resistance of the bitumen-soil matrix and lubricated sheets.

In our previous research of freeze-thaw cycle corrosion tests on rebars with various deicing chemicals, the observed freezing points of some chemicals were not low enough to be effective deicers. Also, published data on ice undercutting rates show variation depending on test conditions. Thus, it is necessary to accurately determine ice undercutting rates of these chemicals. In this study, a simple laboratory method was developed to determine the ice undercutting rates of deicing chemicals. The ice undercut rate decreased as the test temperature decreased. The undercutting rate decrease varied with type of deicing chemical. The lowest temperatures of CaCl ₂ , MgCl ₂ , NaCl, and calcium magnesium acetate (CMA) to be able to undercut ice in test tubes was −35, −22, −21, and −10°C, respectively. Freezing temperatures of solutions of various concentrations of deicing chemicals were measured using a procedure modified from ASTM D-1177. The freezing temperatures of deicing chemicals decreased as the concentrations of deicing chemicals increased.