Patrick Sturm

In this work, one-part geopolymers were synthesized by mixing solid silica and sodium aluminate with water. Two industrial silicas, a microsilica and a residue of the chlorosilane production, as well as two differing batches of biogenic silica (rice husk ash) have been used as silica feedstocks (Chapter 2). The alumina feedstock incorporated the alkaline activator, thus, just water must be added to start the reaction. Therefore, the handling and alteration of highly alkaline activator solutions, such as sodium hydroxide solutions or sodium silicate solutions is avoided on site. Viscometer and spread flow tests have been conducted on the one-part mixes to evaluate the general rheological properties of the OPG pastes and mortars. The fresh pastes appeared to be stiffer than OPC-based pastes, even when higher water-binder ratios (w/b by mass) were applied to the OPG pastes. This is mainly attributed to the viscosity of the pore solution and the lower densities of the solid feedstocks. At equal w/b by mass, the viscosity of the OPG fresh pastes increased with increasing silica content up to a certain content, while the yield stresses increased continuously due to the simultaneous decrease of the w/b by volume (Chapter 3). Several superplasticizers were tested and promising results were found for lignin sulfonates. The workability of the OPG fresh mortars is very variable by adjusting the paste content. Depending on the paste content the fresh mortars can be applied manually or by spraying in various layer thicknesses (Chapter 6).

Pastes were cured at elevated temperatures (60–90 °C) and relative humidity (r.H.) of 80–98 %. After curing the pastes at 80 °C and 80% r.H for one day, the reaction virtually ceased. No significant further changes of the phase assemblage have been observed for longer curing times. For MS- and CR-based OPG, the reaction ceased, when a SiO 2/Al 2O 3-ratio of ~2 mol/mol was reached in the reaction products, i.e. in MS/CR-based OPG mixes with an initial SiO 2/Al 2O 3 > 2 mol/mol, always a fraction of the silica feedstock remained unreacted. CR- and MS-based OPG mixes formed geopolymer- zeolite composites. Zeolite A and hydrosodalite were the dominant crystalline phases. Depending on the molar SiO 2/Al 2O 3 ratio of the starting mixes their relative amounts varied. For CR-based mixes, zeolite A was always the dominant crystalline phase and more diverse zeolites formed. In contrast to that, RHA-based OPG achieved higher degrees of incorporation of silicon into the aluminosilicate network and can provide zeolite free geopolymeric gel formation, depending on the pre-treatment of the silica feedstock. The compressive strength of the geopolymer-zeolite composites was lower as the compressive strength of a pure geopolymer (Chapter 4), mainly due to a significantly denser and glassy microstructure with less interfaces.

The major part of the thermal dehydration occurred between 60 °C and 200 °C. Up to 700–800 °C only minor changes of the phase assemblage have been observed for the composites. Furthermore, only low to moderate linear thermal shrinkage was observed. The samples gained compressive strength and stayed above 100% of the as-cured state. At higher temperatures, the samples underwent new phase formation. That was connected to partial melting, sintering and the loss of the mechanical strength. Depending on the paste composition either ceramic or amorphous phases from during exposure to 1000 °C (Chapter 5).

MS-based mortars provided a significantly faster hardening than the pure pastes (Chapter 7), due to the induced nucleation sites from the aggregates. By adding CaO up to a certain content the hardening can be further accelerated, without a fundamental change of the forming reaction products. Compared to other AAM, very low ambient drying/wetting shrinkage/expansion was observed for the mortars (Chapter 8). This is mainly attributed to the initial curing at elevated temperatures and the mainly physical bond water in the composites (zeolitic water).

Regarding the very promising bond behavior between mortars and concrete substrate, curing only at ambient temperatures can be beneficial. Depending on the processing of the silica feedstock, gas forming agents can be included as traces in the silica feedstocks. Depending on the curing temperature and the correlating reaction speed, those gas forming agents can introduce microcracks. In such a case the bond between OPG mortar and concrete is stronger than the actual tensile strength of the OPG mortar.

The figure below gives a summary of the general relationships of the factors influencing the properties of the considered OPG, synthesized with silica and SA and thus excluding the additive manufacturing.

General relationships, observed in this work, for the properties of silica-based OPG pastes and mortars, activated with SA. Big black arrows represent major influences, whereas, grey arrows represent minor influences.

The OPG mortars provide very high resistance against sulfuric acid (pH 1) and provided residual compressive strengths up to > 77 % of the reference. The acid initially attacks parts of the paste and a secondary silica gel precipitates at the acid-mortar interface. This gel protects the remaining reaction products and decreases the further corrosion speed. Furthermore, huge parts of the corroded layer are not dissolved from the specimen and can still provide protection for a potential substrate.

In terms of sulfuric acid resistance, above a critical CaO content from potential additives, the formation of gypsum is introduced. This causes expansion, cracking and the decrease of the sulfuric acid resistance. (Chapter 9).

First investigations on the additive manufacturing by LIS revealed promising behavior. In this context lithium aluminate revealed significantly lower viscosities of the fresh pastes and the possible manufacturing of complex structures (Chapter 10).