73 meV above the hull
Fe12BW4 phase diagram 2
Phase diagramPhase diagram of Fe12BW4 with Orb v3 conservative inf MPA; eabovehull: 0.103858 eV/atom; predicted_stable: False
Fe12BW4 phase diagram 1
.htmlPhase diagram of Fe12BW4 with Orb v3 conservative inf MPA; eabovehull: 0.103858 eV/atom; predicted_stable: False
Phonon band structure — BFe12W4 (P4/mmm)
ImagePhonon dispersion (supercell [3, 2, 2]); freq range [0.2668, 26.0664] THz
BFe12W4 (P4/mmm) - 3x3x3 supercell
.cifSupercell 3x3x3 of Fe12BW4 (Space group: P4/mmm, 432 symmetry operations)
Fe12BW4 phase diagram
.htmlPhase diagram of Fe12BW4; eabovehull: 0.103858 eV/atom; predicted_stable: False
BFe12W4 (P4/mmm) - phonon dispersion
ImagePhonon band structure (supercell [3, 3, 3], Δ=0.01 Å); imaginary modes detected; min freq = -5.24 THz
Actually is dynamically stable! But poor magnetization.
Short answer: no. Nothing at Fe22W6B2 stoichiometry (Fe11W3B reduced) has ever been reported as made. I searched the exact formula in both orderings and then read the Fe-W-B phase-equilibria literature, which is thorough enough that an Fe-rich ternary boride there would have shown up.
What the experimental Fe-W-B system actually contains. Three independent equilibria studies agree there are exactly two ternary compounds:
WFeB (TiNiSi-type, Pnma) and W2FeB2 (W2CoB2-type, Immm). Rogl's 1323 K isothermal-section reinvestigation found only these two, and explicitly no ternary tungsten boride above 40 at% B. Raghavan's assessments say the same (JPE 24, 457 (2003), MSIT chapter, 2008).
The most careful modern work confirms it: Kirkovska, Homolová, Zobač and Zemanová arc-melted B-Fe-W alloys and annealed them 2000 h at 1050 °C and 4224 h at 677 °C before CALPHAD modeling (J. Phase Equilib. Diffus. 2021); Ouyang et al. mapped it experimentally and thermodynamically (Calphad 63, 212 (2018)).
At Fe73W20B7 (the Fe22W6B2 composition) the equilibrium assemblage is (Fe,W) solid solution + Fe2B + mu-Fe7W6. There is no Fe-rich ternary boride field.
Closest things anyone has actually made:
W2FeB2 and FeWB powders by reactive synthesis from FeB + W at 1150 °C / 3 h (93 wt% W2FeB2; 77 wt% FeWB, with W0.14Fe1.86B and W0.7Fe0.3 as intermediates): Zhou et al., RSC Adv. 2022, Li, Li, Liu 2018. FeWB also goes to bulk by SPS at 91.8 HRA (Li et al. 2015).
Both ternaries also precipitate during pack boronising of Fe-W alloys (Surface Engineering 2017), and alpha-Fe + FeWB + Fe7W6 crystallize out of Fe-W-B amorphous powders.
Metastable W-doped Fe3B: (W0.2Fe2.8B) in the Ti3P type, stable only in the narrow 1353-1383 K window (same 1323 K isothermal-section work). Rod-like (Fe,W)3B and reticular (Fe,W)2B appear as metastable precipitates in Fe-3.5B-W alloys (study here).
The 23:6 family, which is the nearest structural relative of an "Fe22 + metal + B6" idea: binary Fe23B6 (Cr23C6-type) is metastable and forms only from undercooled or rapidly quenched Fe-B (undercooled Fe83B17); the substituted siblings Fe22PdB6 and Fe22PtB6 are likewise metastable quenched phases (
One identity note. I cannot find a Materials Project entry with the formula Fe22W6B2. The published MP dataset contains exactly four pure Fe-W-B entries: W2FeB2 (mp-1068296), a second W2FeB2 cell (mp-613327), FeWB (mp-1103093), and WFe22B6 (mp-1224794, cubic F-43m, 29 atoms, DOI 10.17188/1654288), which is one W on a metal site of an Fe23B6-type cell. If the entry you saw is WFe22B6, the answer is the 23:6 paragraph above: not made, and W has never been placed on that site experimentally. If instead you mean the Fe22W6B2 composition in magnes's GGen B-Fe-W screen
Bottom line: Fe22W6B2 is genuinely unmade in any ordering, and it would be new if it exists. The equilibrium routes that produced Fe-W-B ternaries all converge on WFeB and W2FeB2; the metastable Fe-rich borides (Fe23B6-type, Ti3P-type Fe3B) have only ever been reached by non-equilibrium processing like melt spinning, undercooling, or rapid quench. That is also where the Fe-W-B magnet question ends: