LiFeAs: An intrinsic FeAs-based superconductor with Tc=18 K
Summary (2 min read)
I. INTRODUCTION
- Quantum-well ͑QW͒ states in nanometer thick metal films have been associated with a number of interesting properties, such as magic layer numbers in thin-film growth, [1] [2] [3] oscillatory magnetic interlayer coupling in magnetic multilayers, [4] [5] [6] work-function modulations, 7 and periodic anomalies in thinfilm conductance.
- In particular, in the case of Mg͑0001͒ films grown epitaxially on W͑110͒, well resolved series of QW states and resonances have been reported in several recent photoemission studies. [10] [11] [12].
- In particular, the changes observed in the initial oxidation rate-when most of the film was still metallic, were found to be dramatic.
- The results are at variance with a recent interpretation of the surface states splitting in the photoemission spectra.
II. COMPUTATIONAL DETAILS
- The calculations were performed within densityfunctional theory, using the Perdew-Burke-Ernzerhof exchange-correlation functional.
- The Mg 3s, 3p, 3d orbitals and the W 5d, 6s, 6p orbitals were treated as valence states, and the authors used the nonlinear core correction.
- The epitaxial films were modeled using slab geometries in a supercell.
- For thicknesses above 2 monolayers ͑ML͒, Mg films are known to grow epitaxially on W͑110͒ with lattice parameters corresponding essentially to their bulk values.
- The self-consistent calculations were carried out using a ͑20,20,1͒ Monkhorst-Pack k-point grid 24 and a Gaussian electronic-level smearing of 0.02.
A. Quantum-size effects on the decay length
- The behaviors are very similar in the two cases.
- In the calculations, the first QW state with energy higher than the SS states enters the occupied-state spectrum at a Mg thickness of 8-9 ML, and the second one at a thickness of 16-17 ML.
- The same systematic shift is observed between the calculated maxima in and the experimental maxima in the reactivity.
- The authors note that the calculated energy levels, in Fig. 2 , compare relatively well ͑to within ϳ0.2 eV͒ with the results of recent pseudopotential calculations performed for the free-standing Mg͑0001͒ films.
B. SS, QW, and interface states
- In the freestanding Mg͑0001͒ film, the Shockley states of the two surfaces strongly overlap, and their interaction gives rise to a split pair of even-and odd-symmetry states relative to the midslab reflection ͑right-hand-side panels of Fig. 3͒ .
- And therefore cannot describe the Rashba splitting taking place at k ʈ 0, they are expected to describe well the Shockley surface-interface band splitting, especially at ⌫ ¯.
- In this figure the authors also display the edges of the valence gaps in the calculated W͑110͒ bulk projected band structure ͑PBS͒ at the k points considered.
- This is in qualitative agreement with the experimental observations, 11, 12 although an additional smaller splitting was observed in Ref. 12.
- The state SI has a dominant interface state character and originates mainly from the interaction of the W͑110͒ d-derived surface state 33 with the Mg͑0001͒ SS states.
C. Effect of strain
- In order to understand the shift, between calculation and experiment, in the number of layers for QW states to cross the Fermi energy, the authors have investigated the effect of strain on free-standing Mg͑0001͒ films.
- In fact, a recent low energy electron diffraction ͑LEED͒ study has indicated that the first Mg atomic layer, closest to the interface, is strongly contracted along the inplane W͓11 ¯0͔ direction ͑b axis͒, relative to the bulk Mg͑0001͒ lattice.
- Focusing thus on Mg films with thicknesses of 6-7 atomic layers-in the range where experimentally the first Fermilevel crossing takes place-an 11% contraction of the first Mg layer along the b axis corresponds to an average lateral strain ⑀ yy Ϸ −0.02 over these films.
- Results are presented both for films which have been elongated in response to the actual inhomogeneous lateral contraction of the film, and films whose length has been kept unchanged.
IV. SUMMARY AND CONCLUSIONS
- The authors find that the decay length in vacuum, , of the electronic local density of states at the Fermi energy exhibits pronounced oscillations with film thickness.
- The decay length is maximal when a quantum-well state passes through the Fermi energy.
- The investigation of the effect of strain, in particular, provides a possible explanation for the shift in the number of layers observed between calculated ͑unstrained͒ and experimental spectra.
- Comparison with the photoemission spectra provides an unambiguous identification of the corresponding two-band splitting.
- Furthermore, the investigation of the electronic states of the film with wave vector k ʈ along the ⌫S direction of the surface Brillouin zone helps settling a controversy concerning the microscopic origin of two bands of Mg surface-related states located in a pocket gap of the W͑110͒ projected bulk band structure.
Did you find this useful? Give us your feedback
Citations
1,373 citations
969 citations
Additional excerpts
...Після відкриття надпровідності в арсенідах та фосфідах феруму(ІІ) стало зрозуміло, що різноманіття потенційних ВТНП не обмежується лише купратною керамікою та похідними фулеренів [62-70]....
[...]
...Для подвійного арсеніду літію і феруму(ІІ) величина Тс за різними даними складає 16 – 18 K [66-67], а для ізоструктурного NaFeAs Тс = 9 K [68]....
[...]
640 citations
603 citations
590 citations
References
6,643 citations
1,456 citations
1,441 citations
1,399 citations
1,325 citations
Related Papers (5)
Frequently Asked Questions (12)
Q2. How does the reactivity of the O2 molecules change?
16 Assuming a transfer rate by tunneling proportional to e−d/ , with d the distance between the metal surface and the center of mass of the molecule, and considering a typical distance d of 3.5 Å—within the expected physisorption range of the O2 molecules,16,27 a 10% variation in produces a 100% change in the transfer rate.
Q3. What is the effect of the tunneling on the electron transfer rate?
The changes in the decay length reported here are expected to have a direct, exponential impact on the electron transfer rate by resonant tunneling—from the metal to the O2 molecule—which has been proposed to control the initial sticking of the oxygen molecules impinging on the surface, via the attractive image charge potential on the ionized O2 − molecule.
Q4. What is the effect of the changes in in the tunneling rate in the electron transfer mechanism?
The changes in are expected to have a major impact on the tunneling rate in the electron transfer mechanism, which is believed to control the initial sticking of O2 in the oxidation process.
Q5. What is the effect of the large lattice misfit?
The strain due to the large lattice misfit between Mg and W is expected to be only partially released in the Mg layers closest to the interface.
Q6. How many atomic layers can be derived from the Bohr-Sommerfeld?
For a discrete number of atomic layers, the periodicity of the crossing of EF can be derived from the Bohr-Sommerfeld rule, which for Mg 0001 yields a periodicity of 7.7 ML.10,13
Q7. What is the main origin of the observed splitting?
As the spin-orbit interaction is negligible in the case of Mo,33 this demonstrated that the Rashba effect is not the main origin of the observed splitting.
Q8. What is the effect of strain on free-standing Mg 0001 films?
C. Effect of strainIn order to understand the shift, between calculation and experiment, in the number of layers for QW states to cross the Fermi energy, the authors have investigated the effect of strain on free-standing Mg 0001 films.
Q9. What is the atomic-scale effect of the QW states on the Mg 0001?
With increasing film thickness an unoccupied QW states, in Fig. 2, is found to cross the Fermi energy at 9 layers and a second one at 17 layers, which exactly coincides with the local maxima observed in the decay length in Fig.
Q10. What is the effect of a lateral contraction of the Mg layers?
This will be discussed in Sec. III C, where the authors show that a lateral contraction of the Mg layers induces the entrance of a QW state in the valence spectrum at a lower number of Mg layers, as observed experimentally.
Q11. What is the probability density of the QW states in the W substrate?
In the presence of the W substrate, the SS states persist as strong surface/interface resonances, with maxima in the probability density located both in the region of the surface and of the interface Mg layer left-hand-side panels of Fig. 3 .
Q12. What is the reason for the oscillations in the decay length of the local density of Cu?
The authors note that the oscillations the authors find in the decay length of the local density of035438-3states near EF may also be related to a recent observation of quantum-size effects on the chemisorption properties of Cu 001 thin films.