An unexpected softening from WB ₃ to WB ₄

Abstract

We report a drastic reduction of hardness of about 61% from WB3 to WB4. The three-dimensional covalent network consisting of boron honeycomb planes interconnected with strong zigzag W-B bonds underlies the high hardness of WB3. Despite the strong intralayer and interstitial B-B bonds, the interlayer B-B nonbonding and the considerably weak zigzag W-B bonding allow the layers of WB4 to cleave readily, which results in the anomalous softening of WB4. The results provide robust evidence that the highest boride of tungsten, which is characterized experimentally by an inexpensive superhard material, should be stoichiometric WB3, not WB4.

Crystal Structure of W1−x B3 and Phase Equilibria in the Boron-Rich Part of the Systems Mo-Rh-B and W-{Ru,Os,Rh,Ir,Ni,Pd,Pt}-B

“…[20][21][22][23][24][25][26] Zhang et al [24] reopened the question on the correct crystal structure of WB 4 or WB 3 , as the calculated indentation hardness of WB 4 is lower than that of ReB 2 in contrast to the experimental result and the calculated normalized c/a ratio of WB 3 exhibits a negative pressure dependence, inconsistent with the observed trend. Li et al [25] in their latest study presented new thermodynamically stable structures for WB 3 (R 3m-''6u'') and WB 4 (P6 3 /mmc-''2u'').…”

“…However, more recent theoretical calculations of band structure and physical properties (density functional theory calculations DFT) of this compound have inferred some confusion on the true crystal structure of ''WB 4 ''. [20][21][22][23][24][25][26] Whereas the first studies by Chretien and Helgorsky [15] reported a tetragonal unit cell for ''WB 4 '' (a = 0.634 nm, c = 0.450 nm), Rudy et al [16] claimed a hexagonal cell (a = 0.3004 nm, c = 0.3174 nm) but proposed ''WB 12 '' as a more appropriate composition. A first structure model was derived by Romans and Krug [17] from x-ray powder data backed by single crystal Laue and precession photographs indexed on a hexagonal lattice (a = 0.5200 nm, c = 0.6340 nm).…”

“…The theoretical DFT studies on the structural assignment and stability of WB 4 or MoB 4 (WB 3 or MoB 3 ) [20][21][22][23][24][25][26] claimed that WB 3 (MoB 3 ) with a two-dimensional boron net is thermodynamically more stable than WB 4 (MoB 4 ) with a three-dimensional boron skeleton. [20][21][22][23][24][25][26] Zhang et al [24] reopened the question on the correct crystal structure of WB 4 or WB 3 , as the calculated indentation hardness of WB 4 is lower than that of ReB 2 in contrast to the experimental result and the calculated normalized c/a ratio of WB 3 exhibits a negative pressure dependence, inconsistent with the observed trend.…”

Structure of superhard tungsten tetraboride: A missing link between MB 2 and MB 12 higher borides

“…Within the past five years, there has been a rapid succession of computational papers with the goal of identifying the structural origin of the properties of WB 4 (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28). Although the correctness of the Romans and Krug model was initially assumed (24), it was quickly noticed that such a structure should be unstable (26).…”

“…Although the correctness of the Romans and Krug model was initially assumed (24), it was quickly noticed that such a structure should be unstable (26). This lead to the theoretical confirmation that the structure and properties of the highest boride of tungsten are better accommodated if the B 2 dimers are removed (21,23,27), producing a composition of WB 3 . Most recently, several exotic models have appeared with larger unit cells (18,29) and varying stacking orders (28) of the metal layers.…”