Science1 publisher2 min readPublished
Calculations put hexagonal diamond's electron mobility at 28,473 cm2V-1s-1 along the c axis
An ab initio study in npj Computational Materials traces the figure to symmetry-enforced phonon selection rules. The effective masses it computes are comparable to cubic diamond's, so mass cannot account for the gap.
The Scientist · Science desk
What happened
- Ab initio calculations published in npj Computational Materials put hexagonal diamond's room-temperature electron mobility at 12,339 cm2V-1s-1 perpendicular to the c axis and 28,473 cm2V-1s-1 along it.
- The calculated hole mobilities come out almost identical in the two directions, 6000 cm2V-1s-1 perpendicular to c and 6024 cm2V-1s-1 parallel to it.
- A second effect, a real-space mismatch between electronic wavefunctions and phonon scattering potentials, weakens acoustic scattering for electrons and suppresses out-of-plane longitudinal acoustic scattering for holes.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision Anyone ranking wide-bandgap candidates on bandgap and effective mass would score the two diamond phases as near-equals, so crystal symmetry and phonon polarisation have to enter the shortlist criteria to catch a difference like this.
- capability Because electrons move 2.3 times faster along c than across it in the calculation, crystal orientation becomes a design variable for any future hexagonal diamond device rather than a fabrication detail.
- precedent If symmetry-enforced selection rules are doing the suppressing, the same check is now worth running on other hexagonal polytypes whose cubic siblings already have measured mobilities to compare against.
Transverse acoustic phonons are the scattering channel that predominates in cubic diamond around room temperature, according to He, Gao and Chen [7]. In the hexagonal phase, selection rules enforced by the lattice symmetry suppress that scattering [6]. The second effect they report is geometric: the electronic wavefunctions and the phonon-induced scattering potentials do not sit in the same place in real space, and the paper calls the result electron-phonon decoupling [8]. For holes it shows up as the suppression of out-of-plane polarised longitudinal acoustic scattering, and for electrons as a systematic weakening of acoustic scattering [9].
The directional split is large for electrons and nearly absent for holes. Dividing 28,473 by 12,339 gives 2.31, so the calculated electron mobility along the c axis is more than twice the value across it [14]. The two hole figures differ by 24 cm2V-1s-1, or 0.4 percent [15]. Along c, electrons come out 4.7 times more mobile than holes [16].
The abstract says these values are significantly superior to those of most known semiconductors, including cubic diamond, and does not give a number for the cubic phase [4]. So the size of the margin over cubic diamond is not something a reader can take from the paper's summary. The part that is stated plainly is the negative result underneath it: the small effective masses in hexagonal diamond are comparable to those in the cubic phase, and therefore cannot explain the higher mobilities [5]. If you rank candidate wide-bandgap semiconductors by bandgap and carrier effective mass, these two crystals look close to a tie.
Both of the effects named act on carrier-phonon coupling [6][8][9]. I would treat the numbers as intrinsic room-temperature ceilings for a clean crystal, which is a different quantity from what a doped, contacted wafer will measure on a Hall bench. The thing this does not tell you is how much of the advantage survives the dopants and defects a device needs. All of it comes from ab initio calculation, not from a sample [3].
Lonsdaleite has been of interest as a wide-bandgap semiconductor mostly for its thermal conductivity and hardness [10]. This paper is the transport argument, and it is a calculation with a testable claim attached: change the symmetry, change which phonons are allowed to scatter carriers. The work was supported by research funds from Shanghai Advanced Silicon Technology Co., Ltd. and by the Natural Science Foundation of Shanghai under grant 23ZR1403300, and the authors declare no competing interests [12][13].
What to watch
- A measured Hall mobility on a real hexagonal diamond sample, at any temperature, to test the calculated ceilings.
- An independent calculation using a different electron-phonon treatment that either reproduces or fails to reproduce the 28,473 cm2V-1s-1 figure.
- Whether the same symmetry selection-rule argument is applied to other hexagonal-phase semiconductors in follow-up work.