Influence of Phosphates on Phase Formation and Pore Solution in Alkali-activated MgO-Al2O3-SiO2-P2O5 Cements
Open Access
- Author:
- Reed, Titus
- Graduate Program:
- Materials Science and Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 20, 2025
- Committee Members:
- John Mauro, Program Head/Chair
Yashar Mehmani, Outside Unit & Field Member
Juan Pablo Gevaudan Burgos, Co-Chair & Dissertation Advisor
John Mauro, Co-Chair & Dissertation Advisor
Maziar Montazerian, Major Field Member - Keywords:
- alkali-activated materials
phosphate
layered double hydroxides
cement phase formation
phosphates
XRD
cement
phase formation
pH - Abstract:
- The disposal of high-level nuclear waste (HLW) is still an unsolved problem. Many countries are working on designing deep geological repositories (DGR) for the permanent disposal of HLW. Ordinary Portland cement (OPC) is a popular material for some nuclear waste disposal applications. However, the high pH of OPC pore solution can cause degradation of the minerals present in the geological formation hosting a DGR. Yet, the high pH of OPC creates an environment that promotes the growth of a protective layer of oxides and hydroxides that prevents the corrosion of steel. As steel canisters are part of most DGR designs, this is desirable. Therefore, there is a need for a low pH cement (pH < 12.5) that is capable of balancing the competing requirements of maintaining the integrity of the geological formation while still preventing corrosion of embedded steel. One pathway for achieving such a material is to include a corrosion inhibitor such as phosphate in a low-pH cement. Phosphates have been shown to improve the protective quality and lower the pH required for a protective passivation layer on the steel. Additionally, due to their polyprotic nature when combined with sodium hydroxide, phosphates can form a pH buffering solution effective over a range of pH values and have been used commercially to buffer pH in high-pressure boilers. In calcium-based cement, such as OPC, phosphates form insoluble precipitates with calcium, hindering the effectiveness of a corrosion inhibitor. However, the influence and compatibility of phosphates on alkali-activated materials (AAM) have been largely unexplored. In this work, for the first time, the effect of phosphates on the phase assemblage and pore solution of a calcium-free high-magnesium NaOH-activated AAM is established. The polymer-assisted sol-gel process was used to produce a series of MgO-Al2O3-SiO2-P2O5 precursors. The previously established sol-gel process was improved to reduce the time needed to produce a batch of cement precursors from as long as 6 days to no more than 48 hours. The modified process was used to create a series of precursors that were activated with NaOH solutions cured in sealed molds at 35°C and examined after 1 to 470 days of curing. X-ray diffraction (XRD) and magic angle spinning-nuclear magnetic resonance (NMR) are used to examine the crystalline and amorphous phases that form. The XRD results confirm that Al is preferentially incorporated into hydrotalcite-like layered double hydroxides (LDH) over zeolites. Zeolites form when more Al is present than can be incorporated into the LDH. NMR results show that phosphate is not incorporated into aluminosilicate networks formed in these materials (such as zeolites or disordered aluminosilicate hydrate). Instead, it is shown that the phosphates react with sodium, forming soluble sodium phosphates. The NMR data shows that phosphate is adsorbed onto the surface of the aluminum-containing phases present in the material, forming inner and outer-sphere complexes. It is shown that increasing phosphate content slows (seen at 19.1 wt. % P2O3 in precursor) and eventually suppresses (seen at 26.6 wt. % P2O3 in precursor) the crystallization of zeolites. Cold water extraction (CWE) and Gibbs energy minimization (GEM) modeling were used to measure the pore solution composition and pH of samples cured for 28 days. CWE has not previously been used for cements containing soluble hydrates. To enable its use in this system, a methodology was developed to account for the difference in the water extracted from soluble hydrates during the CWE procedure and the water removed during drying (which is used to calculate the concentration of the pore solution). The CWE and GEM results show that sodium and phosphate species are the primary constituents of the pore solution. This is unlike a high-calcium AAM, where phosphates are removed from the pore solution and do not significantly impact pH. It is shown that high (>4) soluble Na/P ratio phosphates cause a moderate decrease (less than one pH unit) in the pH of the pore solution. However, when the pore solution is saturated with respect to sodium phosphate and the soluble Na/P ratios between 2 and 3, as is observed in the high-phosphate AAM composition studied, not only is the pH dramatically decreased (multiple pH units). The resulting pore solution and a precipitated mixture of di and trisodium phosphate have a buffering effect on the pore solution's pH and Na+, keeping the pH and Na+ near 11.0 and 1.0 mol/kg, respectively. These results provide evidence that phosphates are compatible with low Ca AAMs, which is consequential as there is a growing interest in both the use of AAM and phosphate-based corrosion inhibiter in steel-reinforced concrete. It is demonstrated that phosphates can create a low pH (pH = 11) AAM within the pH range desired for HLW applications. Furthermore, the provided evidence that phosphates are not incorporated into the phase typically found in low-calcium AAM provides essential information for the further development of AAM from phosphate-containing raw material. Additionally, a systematic study of the effect of relative humidity on the interlayer spacing of a phosphate-intercalated magnesium aluminum LDH (Mg, Al-PO4 LDH) was conducted. This study confirms a distinct high- and low-humidity basal spacing. The XRD results are the first to show that the transition from the low- to high-humidity form begins between 23 and 33% relative humidity at 25°C. XRD and thermogravimetric analysis provide strong evidence that the change in basil spacing is due to a transition from a single layer to a double layer of interlayer water and is not solely due to the reorientation of the phosphate ion. This information will help researchers identify proper drying or hydration steps needed to examine Mg, Al-PO4 LDH under the desired form. It provides the essential understanding of Mg,Al-PO4 LDH interaction with water, which is needed to allow for a complete understanding of the behavior of these LDHs in many proposed applications such as advanced fertilizers, wastewater management, and corrosion control.