Mix Design, Printability and Structural Buildability of Biopolymer and Fiber Modified Earthen Composites for Extrusion based 3D Printing in Construction

Author

• Suvechha Dhakal, Southern Illinois University CarbondaleFollow

Date of Award

8-1-2026

Degree Name

Master of Science

Department

Civil Engineering

First Advisor

Tiwari, Nitin

Abstract

The global construction industry is responsible for approximately 39% of total carbon dioxide emissions and 50% of all raw material consumption worldwide, with ordinary Portland cement production alone contributing 8–10% of anthropogenic CO2 emissions annually. In response to this environmental crisis, earth-based construction materials have re-emerged as a compelling low-carbon alternative, offering minimal processing requirements, inherent recyclability, and favorable indoor environmental qualities such as passive moisture buffering and volatile organic compound absorption. The integration of extrusion-based 3D printing with earthen materials promises to further amplify these sustainability advantages by eliminating formwork waste, accelerating construction timelines, and enabling precise automated deposition from digital models. However, the direct application of raw earth in additive manufacturing is constrained by significant material limitations: clay-rich soils inherently exhibit low tensile strength, low bearing capacity, and pronounced swelling-shrinkage behavior that compromise both the fresh-state printability and the hardened structural integrity of printed elements. To address these challenges, biopolymers and reinforcing fibers must be incorporated into earth-based mixtures, yet these additives simultaneously alter rheological properties in complex and often competing ways. Achieving the delicate balance between extrudability, the capacity for continuous, dimensionally consistent extrusion through a nozzle, and buildability, the ability of deposited layers to sustain the gravitational stress of subsequent layers without collapse, remains a critical and poorly understood design challenge.This thesis presents a systematic experimental investigation into the printability and buildability of locally sourced earth-based composites reinforced with biopolymers and fibers, with the goal of developing a cement-free, structurally sound formulation suitable for extrusion-based 3D printing. The study used a high-plasticity inorganic clay (classified as CH under the Unified Soil Classification System) excavated from Carbondale, Illinois, blended with fine white silica sand at a fixed 80:20 soil-to-sand ratio by dry weight. Two biopolymers, Xanthan gum and Locust bean gum, were each evaluated at concentrations of 0.75%, 1.0%, 1.5%, and 2.0% by dry weight, and two fiber types, natural hemp fiber and synthetic polypropylene fiber, were each tested at dosages of 0.25%, 0.50%, 0.75%, and 1.0% by dry weight in a full factorial design, yielding seventeen distinct mixture formulations.A multi-scale characterization program was employed to evaluate each formulation. Geotechnical characterization of the base soil confirmed a liquid limit of 61%, a plasticity index of 41, and a median particle diameter of 0.0043 mm, underscoring the high expansivity and water sensitivity of the material. Fresh-state assessment encompassed modified flow table testing, mini-slump testing, rotational viscometry, and setting time measurement. Printability was evaluated through a triangular-bag extrudability test employing a 10 mm nozzle and a quantitative filament width criterion of 7–13 mm, as well as multilayer manual extrusion and green strength testing using a height retention index threshold of 30%. Full-scale automated printing trials were conducted using a customized gantry printer with four interchangeable nozzle diameters (5 mm, 10 mm, 15 mm, and 20 mm). Hardened-state performance was assessed through unconfined compressive strength and flexural strength testing after 28 days of curing, dimensional shrinkage measurement across three axes after 30 days of drying, and a surface water resistance evaluation via spray testing.The experimental results demonstrated that biopolymer type was the single most decisive variable governing printability. Xanthan gum at a concentration of 1% by dry weight consistently produced sufficient cohesion and thixotropic recovery to enable smooth, continuous extrusion without water segregation, with viscosity measurements reaching 5.39 Pa·s. In contrast, Locust bean gum-modified mixtures exhibited cracking, structural disintegration, and excessive water demand across all tested concentrations, with water-to-dry ratios ranging from 1.06 to 2.0, resulting in oversized filament widths and inadequate green strength. Hemp fiber caused consistent nozzle clogging in all formulations due to its average fiber length of 19–38 mm exceeding the 10 mm nozzle diameter, confirming that nozzle diameter should be increased for natural fibers to remain viable as reinforcement agents.The optimal formulation identified was CSXP2, comprising 80% soil, 18.5% sand, 1% Xanthan gum, and 0.5% polypropylene fiber at a water-to-dry ratio of 0.635. This mixture satisfied all defined printability criteria: it achieved consistent machine extrusion with an average filament width of 10.13 mm, a green strength of 39%, and a maximum of 10 stable buildable layers under machine printing conditions. A revised empirical flowability range of 25–30% was identified as appropriate for this class of high-plasticity, xanthan gum-modified earth mixture, substantially different from the 67–89% threshold established in existing literature for other earth-printing systems, highlighting that printability thresholds are material-specific and system-specific and cannot be transferred across different soil types, biopolymer chemistries, or printer configurations without empirical validation.Post-drying dimensional assessment revealed that the CSXP2 specimens exhibited a 12.66% reduction in total specimen height after 30 days of ambient drying, a 26.26% reduction in the thickness of the bottommost layer due to overburden compressive stress, and a 25.98% reduction in longitudinal dimension attributable to the widening of interlayer gaps and microcracks during shrinkage. These values were substantially lower than those of the CSLP2 formulation (the best-performing Locust bean gum variant), which exhibited a 57.9% total height reduction and severe surface cracking. Unconfined compressive strength testing showed that fiber and biopolymer inclusion increased compressive strength by a factor of 1.43 relative to unreinforced specimens, with CSXP2 achieving a UCS of 8.2 MPa, a value consistent with non-structural earthen construction standards. The nozzle sensitivity study confirmed an inverse relationship between nozzle diameter and maximum buildable layer count, with the 10 mm nozzle producing 10 stable layers and the 20 mm nozzle producing only 6, because larger nozzles reduce extrusion-induced shear thinning and permit greater lateral spreading upon deposition. Surface water resistance testing demonstrated that CSXP2 exhibited only a 0.33% mass reduction following spray exposure, representing a 59% improvement over CSLP2.This study establishes that high-plasticity clay soils from the midwestern United States can be transformed into structurally stable, cement-free printable earth composites through targeted biopolymer and short synthetic fiber modification, supporting the feasibility of site-specific earth printing as a low-carbon construction strategy. The findings provide actionable mix design criteria, including a revised flowability threshold and quantitative biopolymer and fiber dosage recommendations, that advance the scientific understanding of earth-based additive manufacturing and lay the groundwork for future investigations into mechanical anisotropy, freeze-thaw durability, long-term weathering resistance, and full-scale wall-segment printing.