The question of Al content and distribution
The content of Al is 6.86 wt.% Al2O3 (average of 8 analyses; Lykova et al., Reference Lykova, Rowe, Poirier, Friis and Barnes2024) in natromelansonite and 6.29 wt.% Al2O3 (average of 8 analyses; Lykova et al., Reference Lykova, Rowe, Poirier, Friis and Barnes2024 [note there is a typo in Lykova et al., Reference Lykova, Rowe, Poirier, Friis and Barnes2024: 5.89 instead of 6.89; so the correct range is 5.53–6.89] and 6.32 wt.% Al2O3 (average of 15 analyses; Gore and McDonald, Reference Gore and McDonald2023) in melansonite. These numbers are very consistent and correspond to ~1 atoms per formula unit of Al.
In the structures of both melansonite and natromelansonite one of the Si-centred tetrahedra is larger than others which is reflected in lower bond-valence sums for one of the Si sites in both melansonite (Si3, 3.63 vu; Gore and McDonald, Reference Gore and McDonald2023) and natromelansonite (Si4, 3.65 vu; Lykova et al., Reference Lykova, Rowe, Poirier, Friis and Barnes2024).
These data are consistent with our model of ordering of Al atoms at one of the Si sites and support the Y3+ + Si4+ ↔ Zr4+ + Al3+ substitution mechanism to describe the relationships between melansonite/natromelansonite and monteregianite-(Y). Such ordering has also been described in the other members of the rhodesite mero-plesiotype series: delhayelite, hydrodelhayelite (Pekov et al., Reference Pekov, Zubkova, Chukanov, Sharygin and Pushcharovsky2009) and fivegite (Pekov et al., Reference Pekov, Zubkova, Chukanov, Zadov and Pushcharovsky2011).
The model supported by Gore and McDonald does not predict these data nor do they provide an alternative explanation for it. Furthermore, U-rich melansonite that Gore and McDonald used to illustrate variability of Al content in melansonite (4.21 wt.% Al2O3; Lykova et al., Reference Lykova, Rowe, Poirier, Friis and Barnes2024) is, in fact, a Ln-rich phase, which corroborates with the Y3+ + Si4+ ↔ Zr4+ + Al3+ substitution mechanism.
Na in the structure of melansonite
Gore and McDonald did not provide the refined occupancy factor for the partially-occupied Na site in melansonite in either the original publication (Gore and McDonald, Reference Gore and McDonald2023) or the Comment (Gore and McDonald, Reference Gore and McDonald2024). Furthermore, they stated that the occupancy at the site was fixed to satisfy the charge balance by the substitution mechanism Y3+ + Na+ ↔ Zr4+ + □. Thus, the number was fixed to fit their model; therefore, it cannot be used to test either of the models.
Conclusion
Based on the available data, Al should be considered a species-defining element in both natromelansonite and melansonite. The model based on the Y3+ + Na+ ↔ Zr4+ + □ substitution mechanism does not predict the observed phases or the existing data. A new model or/and new data is required to challenge our interpretation of natromelansonite and melansonite.
Competing interests
The author declares none. Note: This response is from I. Lykova, there were additional co-authors on the original manuscript.
