Effects of NaOH amount on condensation mechanism to form aluminosilicate, case study of geopolymer gel synthesized via sol–gel method

Abstract

To balance cost and performance of geopolymers, alkalinity of activating solution is critical. Alkalinity affects condensation that determines the final gel structures, but this effect is confounded by dissolution and is not understood from direct experimental evidence. In this study, we investigated effects of alkalinity on condensation for gels synthesized via a sol–gel method that eliminates dissolution process. As alkalinity increased, particle sizes of the gels increased as indicated by SEM, Si/Al ratios of the gels decreased but polymerization extent increased as supported by FTIR, ²⁷Al and ²³Na NMR, and composition analysis. The mechanism for the effects of alkalinity was proposed accordingly: (1) increasing alkalinity lowers the Si/Al ratio (i.e., more incorporation of Al) of the resulting products probably by affecting charging conditions of the Si and Al units; (2) the presence of Al(OH)₄⁻ units promotes their condensation with nearby species to increase the extent of polymerization; (3) enhanced condensation increases particle sizes of the gels even at microstructural level. This understanding on condensation independent of dissolution provides ways to control gel structures and Si/Al ratios and thus tailor properties accordingly, as well as to suggest a strategy (by altering Si/Al ratios during condensation) to develop kinetics-controlling admixtures.

Highlights

• Condensation of geopolymer gel was studied independently of dissolution.

• Increasing alkalinity lowers Si/Al ratio (i.e., more incorporation of Al) of gel.

• More incorporated Al units enhance condensation with Si species.

• Enhanced condensation increases particle sizes of the gel at microstructural level.

Appendix

Figure 8²⁹Si NMR spectra of empty NMR glass tube (a), glass tube filled with sodium silicate solution (b), and the sodium silicate solution (c) by subtracting spectra (a) from (b). The broad hump centered at −109.8 ppm seen in both the spectra (a) and (b) is attributed to signal from the glass tube. The peak at −72.6 ppm seen in the spectrum of the sodium silicate solution (c) indicates the aqueous Si is present as Q⁰, i.e., the Si monomers.

The solid-state MAS ²⁷Al (shown in Fig. 9) and ²³Na (shown in Fig. 10) NMR spectra of the phases that were removed by water during the washing process (i.e., residual phases by drying filtrates that were separated from the gel–water suspensions during washing). In the residue from suspension of the 3 mL gel in Fig. 9a, the Al mainly exists as aluminosilicate, as indicated by the sharp peak at 59.3 ppm. This presence of aluminosilicate is further confirmed by the peak at −4.1 ppm in the corresponding solid-state MAS ²³Na NMR spectrum shown in Fig. 10a. To the contrary, in addition to the peak at around 56.6 ppm in the ²⁷Al NMR spectrum, the spectrum of the residue from washing the 1 mL gel shows an evident at 3.9 ppm, which is attributed to a 6-coordianted peak. This observation is consistent with the strong peak at −0.5 ppm in the corresponding ²³Na NMR spectrum shown in Fig. 10b, which is assigned to the free Na that is not bound to aluminosilicates. The presence of the strong 6-coordianted Al peak and the intense single form the free Na indicated that much of the Al has not participated in the reaction to form aluminosilicates when 1 mL of the NaOH solution was added during the gel synthesis, probably because the pH of 4.1 in the synthesizing mixtures was much lower than that is needed for geopolymerization.

Figure 11 shows the SEM images using the LEI detector for the same samples as in Fig. 1 (with the SEI detector). The morphology of the flocculated particles in Fig. 11 is clearly revealed. While the sizes of the agglomerates increase with the increasing alkalinity as shown in Fig. 1, the flocculated particles (as elements that form the agglomerates) follow the same trend as shown here in Fig. 11. Specifically, while the size of the flocculated particles is too small to be discernible under such magnification (×20,000) for the 5 mL gel (Fig. 11a), it becomes larger for the 10 mL gel (Fig. 11b), and it increases further as circled in Fig. 11c for the 20 mL gel.

Fig. 8

Solution-state ²⁹Si NMR of empty NMR glass tube (a), glass tube filled with sodium silicate solution (b), and sodium silicate solution (c) by subtracting spectra (a) from (b)

Fig. 9

Solid-state MAS ²⁷Al NMR of phases removed during washing with water (i.e., residual phases by drying filtrates that were separated from the gel–water suspensions during washing) for the 3 mL (a) and 1 mL (b) gels

Fig. 10

Solid-state MAS ²³Na NMR of phases removed during washing with water (i.e., residual phases in filtrates that were separated from the gel–water suspensions during washing) for the 3 mL (a) and 1 mL (b) gels

Fig. 11

SEM micrographs using the LEI detector for gels synthesized using 5 mL (a), 10 mL (b), and 20 mL (c) of additional 10-M NaOH solution