The Effect of Coarse/Fine Aggregate Ratios on the Properties of Additively Constructed Concrete (2024-09)¶

Coarse aggregates are still uncommon in additive construction [1] with concrete extrusion despite potential benefits, such as, reduced paste, cost, and shrinkage [2]. This is due to limited understanding of the effect of aggregates on printable concrete mixes. While attempts have been made to characterize effects of aggregates, previous attempts have not used tests that are field deployable [2, 3]. This study investigated how varying the ratio of coarse to fine aggregates affects various fresh and hardened properties of concrete, focusing on characterization techniques including the unconfined compression and flow table tests. Such tests have the advantage of being deployable in a field setting, making it possible to obtain properties live for quality assurance and quality control methodologies during the printing process to allow for mix adjustments as needed. The US Army Engineer Research and Development Center investigated 6 different coarse/fine (C/F) ratios (0, 0.2, 0.3, 0.4, 0.5, 0.6). It was hypothesized that as the packing density of the aggregates increases, the fresh concrete will exhibit higher strength and its flow will decrease. This is based on literature that suggests that the thixotropic buildup rate is higher for mixes with a higher packing density [4] and the Chateau-Overlez-Trung model that suggests the yield stress is maximized when the packing density is maximized [5]. From the Krieger-Doughterty model, the viscosity of a mix is maximized at the maximum packing density [5] and therefore the flow should be maximized. For each C/F, the bulk density (ASTM C29) was measured prior to mixing [6]. The unconfined compression tests were performed (modified ASTM D2166 [7]), and flow table tests (ASTM C1437) immediately and 15, 30, 45, 60, 75, 90, and 120 minutes after mixing [8, 9]. From results the stress-strain behavior of fresh concrete samples, and the elastic and plastic modulus, were determined following a similar process to Negron-McFarlane et al. [7]. Investigation of the relationship between the C/F and the elastic and plastic modulus and limits was conducted. The sample horizontal deformation was determined by measuring the diameter of the sample at the end of the unconfined compression test using a caliper and was confirmed using a digital image correlation. In addition to tests on the fresh properties of concrete, compressive strength testing using 50 mm cast cubes was conducted in accordance with ASTM C109 [10]. Based on results from others, a noticeable impact was not expected on the hardened compressive strength of the concrete [2] but was tested for confirmation. As expected, a noticeable impact of the C/F on the hardened strength was not observed. Building on previous experiments [7], results indicate an exponential growth in both the elastic and plastic modulus at very-early ages (0 – 120 min). Using an exponential growth model, 𝑦 = 𝐴𝑒𝐵𝑡, an initial modulus, 𝐴, and modulus growth rate, 𝐵 can be determined. Results from testing and analysis resulted in no obvious relationship between the packing density and modulus growth rate or initial modulus. However, it was observed that there is an inverse relationship between the initial modulus and modulus growth rate for both the elastic and plastic modulus. A fit of the flow table data using the Krieger-Doughtery model incorporating data from this study and previous studies was performed. Results indicate that the flow of fresh concrete decreases linearly, which agrees with previous internal experiments completed by this research group and others. These models are expected to be used to model the stress and modulus development and the flow given a packing density and time since mixing providing a theoretical foundation for future projects.