• No results found

Electronic structure, bonding characteristics, and mechanical properties in (W2/3Sc1/3)(2)AIC and (W2/3Y1/3)(2)AIC i-MAX phases from first-principles calculations

N/A
N/A
Protected

Academic year: 2021

Share "Electronic structure, bonding characteristics, and mechanical properties in (W2/3Sc1/3)(2)AIC and (W2/3Y1/3)(2)AIC i-MAX phases from first-principles calculations"

Copied!
8
0
0

Loading.... (view fulltext now)

Full text

(1)

Journal of Physics: Condensed Matter

PAPER • OPEN ACCESS

Electronic structure, bonding characteristics, and

mechanical properties in (W

2/3

Sc

1/3

)

2

AlC and

(W

2/3

Y

1/3

)

2

AlC i-MAX phases from first-principles

calculations

To cite this article: Martin Dahlqvist et al 2018 J. Phys.: Condens. Matter 30 305502

View the article online for updates and enhancements.

Related content

Stability, electronic structures, and mechanical properties of Fe–Mn–Al system from first-principles calculations

Ya-Hui Liu, Xiao-Yu Chong, Ye-Hua Jiang et al.

-Phase stability, elastic anisotropy and electronic structure of cubic MAl2 (M = Mg, Ca, Sr and Ba) Laves phases from first-principles calculations

Yuanyuan Kong, Yonghua Duan, Lishi Ma et al.

-Anisotropic elastic properties and ideal uniaxial compressive strength of TiB2 from first principles calculations

Min Sun, Chong-Yu Wang and Ji-Ping Liu

(2)

1. Introduction

Mn+1AXn (MAX) phases, where M is an early transition metal,

A is an A-group element (mostly fro m group 13 and 14) and

X is C and/or N, are a family of atomically layered ceramics that consist of Mn+1Xn sheets sandwiched in between one atom thick A-layers [13]. To date ~70 ternary MAX phases have been synthesized. The large interest in these materials stems from their unique combination of metallic and ceramic attributes [3, 4], with properties such as oxidation resistance and self-healing characteristics [5, 6], reversible deformation [7], and magnetism [810]. In recent years, an increased focus

has been directed towards using MAX phases as parent mat-erial for its two-dimensional (2D) derivative, MXene, realized from selective etching of the A-element [11, 12]. Even though MXenes are a comparatively young family of 2D materials, it has shown high promise for, e.g. energy storage [13] and electromagnetic interference shielding [14].

Adding a fourth element to the MAX phase by alloying on the M-, A- and/or X site allows for even more elemental combinations. Historically, alloying through solid solutions is the primary route investigated [8, 1521] to, e.g. incor-porate magnetic elements or tailoring thermal expansion. However, recent findings demonstrate chemically ordered MAX phases, where the underlying crystal lattice or sym-metry governs the stoichiosym-metry. The first example is out-of-plane ordered MAX phases (o-MAX) like Cr2TiAlC2

[22], Mo2TiAlC2 [23, 24], Mo2Ti2AlC3 [24], Mo2ScAlC2 Journal of Physics: Condensed Matter

Electronic structure, bonding

characteristics, and mechanical properties

in (W

2/3

Sc

1/3

)

2

AlC and (W

2/3

Y

1/3

)

2

AlC i-MAX

phases from first-principles calculations

Martin Dahlqvist , Andreas Thore and Johanna Rosen

Thin Film Physics Division, Department of Physics, Chemistry and Biology (IFM), Linköping University, SE-581 83 Linköping, Sweden

E-mail: martin.dahlqvist@liu.se and johanna.rosen@liu.se Received 11 April 2018, revised 7 June 2018

Accepted for publication 12 June 2018 Published 3 July 2018

Abstract

With the recent discovery of in-plane chemically ordered MAX phases (i-MAX) of the general formula (M1

2/3M1/32 )2AC comes addition of non-traditional MAX phase elements. In

the present study, we use density functional theory calculations to investigate the electronic structure, bonding nature, and mechanical properties of the novel (W2/3Sc1/3)2AlC and

(W2/3Y1/3)2AlC i-MAX phases. From analysis of the electronic structure and projected crystal

orbital Hamilton populations, we show that the metallic i-MAX phases have significant hybridization between W and C, as well as Sc(Y) and C states, indicative of strong covalent bonding. Substitution of Sc for Y (M2) leads to reduced bonding strength for W–C and Al–Al

interactions while M2–C and M2–Al interactions are strengthened. We also compare the

Voigt–Reuss–Hill bulk, shear, and Young’s moduli along the series of M1 = Cr, Mo, and W,

and relate these trends to the bonding interactions. Furthermore, we find overall larger moduli for Sc-based i-MAX phases.

Keywords: atomic laminate, MAX phase, chemical order, electronic structure, first-principles calculations, bond analysis

M Dahlqvist et al Printed in the UK 305502 JCOMEL © 2018 IOP Publishing Ltd 30

J. Phys.: Condens. Matter

CM

10.1088/1361-648X/aacc19

Paper

30

Journal of Physics: Condensed Matter IOP

Original content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

2018

1361-648X

https://doi.org/10.1088/1361-648X/aacc19

(3)

M Dahlqvist et al

2

[25], defined by alternating M-layers of two different M ele-ments, where each layer is based on one M-element only. The second example is in-plane ordering (i-MAX), shown for (Mo2/3Sc1/3)2AlC [26, 27], (Mo2/3Y1/3)2AlC [27, 28],

(V2/3Zr1/3)2AlC [28], (Cr2/3Sc1/3)2AlC [29], (Cr2/3Y1/3)2AlC

[29], and most recently (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC

[30]. The i-MAX phases are defined by a 2:1 ratio of two dif-ferent M elements with the general formula (M1

2/3M1/32 )2AC.

Moreover, it has been demonstrated that by detailing the chemical etching of an i-MAX phase, one can obtain either a chemically ordered alloy MXene from selectively etching Al only [31], or a MXene with ordered vacancies by etching Al as well as Sc/Y. Examples of the latter are W1.33C [30] and

Mo1.33C [26] MXene.

In this work, we have used first-principles calculations to investigate the electronic, vibrational, and mechanical properties of (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC i-MAX

phases, motivated by their recent discovery, and being the first W-based MAX phase materials. Both (W2/3Sc1/3)2AlC

and (W2/3Y1/3)2AlC are stable with a calculated formation

enthalpy of −27 and −22 meV/atom, respectively [30]. However, neither of the ternary MAX phases W2AlC, Sc2AlC

or Y2AlC have been synthesized, explained by Meshkian et al

showing that these are far from being theoretically stable, with a calculated formation enthalpy of +148, +118, and +185 meV/atom, respectively [30]. The finding of i-MAX phases, allowing introduction of non-traditional MAX phase elements like Sc, Y and W, may alter or introduce new prop-erties as compared to previously known MAX phases. This motivates their exploration, for fundamental understanding, and potential future property tailoring.

2. Computational methods

All calculations were performed within the framework of density functional theory as implemented in the Vienna

ab initio simulation package [3234], combined with the Perdew–Burke–Ernzerhof generalized gradient approx-imation (PBE-GGA) [35] and the projector augmented wave (PAW) method [36, 37]. The unit cells of both compounds were converged to an accuracy of 0.1 meV/atom, using a Monkhorst–Pack [38] 13 × 7 × 5 k-point grid and a plane wave cutoff energy of 400 eV. To calculate the density of states (DOS) and the projected crystal orbital Hamilton popu-lation (pCOHP) for each alloy, we used the LOBSTER code [39]. Dynamic stability in terms of phonon dispersion for both i-MAX phases was calculated from 1 × 2 × 1 super-cells using the finite displacement method. PHONOPY [40] was used both to create the displacements and for the post-processing analysis.

Using the energy-strain method [41], we derived and cal-culated the single crystal elastic constants, where a number of different strains are applied to the crystal lattice, followed by a calculation of the energy associated with each strain. For the monoclinic C2/c structure of (W2/3Sc1/3)2AlC and

(W2/3Y1/3)2AlC, 14 different strains are needed to obtain the

13 independent elastic constants Cij,. Here we used strain parameters of 0, ±0.01, and ±0.02. From Cij we calculated the bulk modulus (B), shear modulus (S), and Young’s mod-ulus (E) within the Voigt (V) and Reuss (R) model as expressed by the equations BV= 19(C11+ C22 + C33) +29(C12+C13+C23), (1) BR=(S 1 11+S22+S33) +2(S12+S23+S13), (2) GV= 151 ((C11+ C22 + C33− C12− C13− C23) +3 (C44+ C55 + C66)), (3) GR= 15 4 (S11+S22+S33)− 4 (S12+S23+S13) +3(S44+S55+S66), (4) EV,R= (3B9BV,RGV,R V,R+GV,R), (5) where the Cij’s are elastic constants and the Sij’s are compli-ance constants given by an inversion of the elastic-constant matrix. The upper boundary (Voigt) is found assuming that the strain is everywhere uniform while the lower boundary (Reuss) is found assuming that the stress is everywhere uni-form. Arithmetic averages of Voigt and Reuss moduli are inter-preted as the ratio of average stress and average strain within the composite. The stress and strain are generally unknown in the material and are expected to be nonuniform.

Schematics were produced with VESTA [42].

3. Results and discussion 3.1. Structural properties

(W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC crystallize in a monoclinic C2/c structure [30], like the isostructural (Mo2/3Sc1/3)2AlC

[26, 27], (Mo2/3Y1/3)2AlC [27, 28], (V2/3Zr1/3)2AlC [28],

(Cr2/3Sc1/3)2AlC [29], and (Cr2/3Y1/3)2AlC [29] i-MAX

phases. It should, however, be noted that a closely related orthorhombic Cmcm structure have also been identified for (Mo2/3Y1/3)2AlC, in line with observations for, e.g. the

(Cr2/3Y1/3)2AlC i-MAX [29]. The close relation between the C2/c and Cmcm structures have also been shown to be almost degenerate in energy [28, 29, 43].

The monoclinic unit cell of (W2/3Sc1/3)2AlC and

(W2/3Y1/3)2AlC contains 48 atoms, 12 formula units (Z = 12).

Table 1 show the calculated lattice parameters together with atomic positions. Our calculated results are in good agreement with the experimental values with deviations for (W2/3Sc1/3)2AlC lattice parameters a, b, c, and β as −0.45%,

−0.13%, +0.06%, and +0.36%, respectively. From table 1

we observe that the exchange of Sc for Y results in increased lattice parameters, as expected with larger metallic radius for Y (1.80 Å) as compared to Sc (1.62 Å).

(4)

3.2. Electronic structure and bonding analysis

The electronic structure and nature of bonding is essen-tial to explain and understand many physical properties of materials. In addition to calculations of the electronic band structure, see figure 1 and an evident metallic character of both compounds, the bonding characteristics of monoclinic (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC have been analyzed in

terms of the DOS and the pCOHP, as shown in figure 2. In order to facilitate interpretation and to preserve the analogy for crystal orbital overlap population (COOP) analysis, the results are here presented as –pCOHP, rather than pCOHP. Except for the W–W (2) interaction across the Al-layer, see notation in figure 2, only nearest-neighbor interactions are considered in the pCOHP analysis, depicted in figures 2(g) and (h), since these are by far the strongest. Both DOS and pCOHP display five main regions; (i) from −13 to −11 eV, showing localized C-s states interacting mainly with W and Sc or Y, (ii) between −9 and −2 eV, from localized Al-s states in Al–Al interactions, (iii) between −7 and −3.4 eV, with Al-p states in Al–Al interactions as well as the bonding (interacting) states of M-d (W, Sc, Y) and C-p, (iv) from −3.4 to −1 eV, from M-d and Al-p states in bonding Al–Al, W–Al, Sc–Al, Y–Al interactions, (v) and states from −1 eV up to the Fermi level (Ef), mainly dominated by W-d with some

contrib-ution from Sc-d or Y-d. The last region shows a significant number of electrons, (N[Ef] = 2.25 states/fu for M2 = Sc and

2.08 states/fu for M2 = Y) despite the low contrib ution from

pCOHP, which mainly consists of W–Sc and W–Y interac-tion. This is indicative of non-bonding electrons, mainly from the transition metals W and Sc or Y. The non-zero contrib-ution at Ef is also a strong indication of metallic character of

these i-MAX phases.

From pCOHP we also find anti-bonding interactions below

Ef; around −5 to −4 eV by C–C and from −2.4 eV up to Ef

by W–C. This is an indication of a non-optimized electronic structure which could be related to an unbalanced number of electrons. Here we have assumed complete occupation of the carbon sites, however, theoretical as well as experimental

investigation of C occupancy is the scope of future work. Since the contribution from anti-bonding interactions is rather small, a small amount of, e.g. carbon vacancies could possibly counteract this interaction. In related MAX phase materials, i.e. V4AlC3 [44, 45] and Nb4AlC3 [46], carbon vacancies have

been shown theoretically and experimentally to be possible, and to stabilize the materials.

Table 1. Calculated and experimental structural parameters and atomic positions for (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC with C2/c

symmetry. Structural parameters

(W2/3Sc1/3)2AlC (W2/3Y1/3)2AlC

Calculated data Experimental data [30] Calculated data Experimental data [30]

a (Å) 9.326 82 9.368 74 9.543 59 9.515 48 b (Å) 5.397 68 5.404 56 5.519 23 5.499 32 c (Å) 13.968 92 13.960 75 14.127 72 14.227 57 α (°) 90.000 00 90.0000 90.000 00 90.0000 β (°) 103.458 70 103.0908 103.623 40 103.3302 γ (°) 90.000 00 90.0000 90.000 00 90.0000 W1 (8f) (0.772 08, 0.421 65, 0.079 58) (0.7725, 0.4250, 0.0821) (0.770 10, 0.423 93, 0.077 05) (0.2679, 0.3677, 0.0863) W2 (8f) (0.110 26, 0.405 80, 0.079 80) (0.1111, 0.4149, 0.0832) (0.110 00, 0.402 89, 0.077 02) (0.1113, 0.4664, 0.0724) M2 (8f) (0.458 23, 0.419 13, 0.110 74) (0.4529, 0.3878, 0.0907) (0.461 06, 0.419 47, 0.118 76) (0.9500, 0.2906, 0.1411) Al1 (8f) (0.241 17, 0.153 92, 0.251 43) (0.2373, 0.1236, 0.2535) (0.247 71, 0.160 01, 0.251 77) (0.2403, 0.1305, 0.2400) Al2 (4e) (0.000 00, 0.431 39, 0.250 00) (0.0000, 0.3506, 0.2500) (0.000 00, 0.420 14, 0.250 00) (0.0000, 0.4117, 0.2500) C1 (8f) (0.916 46, 0.250 25, 0.000 09) (0.9316, 0.2530, 0.0000) (0.913 55, 0.259 19, 0.000 07) (0.9221, 0.2600, 0.0000) C2 (4c) (0.250 00, 0.250 00, 0.000 00) (0.2500, 0.2500, 0.0000) (0.250 00, 0.250 00, 0.000 00) (0.2500, 0.2500, 0.0000)

Figure 1. Calculated electronic bandstructure (a) (W2/3Sc1/3)2AlC

(5)

M Dahlqvist et al

4

By integrating pCOHP up to Ef, it is possible to get a rough

estimate of the relative bond strengths within each i-MAX phase. IpCOHOP in figures 2(c) and (f) suggests the following order, in terms of bond strength for interactions defined in figures 2(g) and (h), for (W2/3Sc1/3)2AlC: C–W > Al–

Al > C–Sc ≈ Al–W > W–W (2) ≈ W–Sc > W–W (1) ≈ Al–Sc > C–C. For (W2/3Y1/3)2AlC the corresponding

order is C–W > Al–Al > C–Y > Al–W ≈ Al–Y > W–W (2) ≈ W–Y > W–W (1), C–C. The two phases thus differ pri-marily in that the Al–Sc bonds are significantly weaker rela-tive to most other bonds in (W2/3Sc1/3)2AlC, in contrast to the

corresponding Al–Y bonds in (W2/3Y1/3)2AlC. Moreover,

sim-ilar to regular MAX phases [47], the M–X bonds are stronger than the M–A bonds. On the other hand, the in-plane nearest neighbor bonds are comparatively weak, with the exception for the Al–Al nearest neighbor bonds.

3.3. Phonon dispersion and density of states

The phonon dispersion between high symmetry points in (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC are shown in figures 3(a)

and (c), with only positive frequencies implying their dynamic stability, i.e. stability with respect to lattice vibrations. Their monoclinic crystal structure contains 48 atoms, giving rise to totally 144 phonon branches out of which three are acoustic

modes and 141 optical modes. Both i-MAX phases exhibit similar phonon dispersions, which can be related to the sim-ilar bonding characteristics discussed in a previous section, with a band gap around 13 THz separating high-frequency contribution, that mainly comes from the light carbon atoms (figures 3(b) and (d)), from the heavier transition metals and Al below the band gap.

Figures 3(b) and (d) shows the phonon partial density of states (PHDOS) of (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC where

lowest frequency peaks are mainly attributed to W followed by Y(Sc) and Al. The contribution from W is similar for both

i-MAX phases. However, upon substitution of Sc for Y the phonon DOS contribution from Al and Y is shifted to lower frequencies. The observed shift for Y, compared to Sc, can be attributed to the fact that it is heavier than Sc. The shift for Al can be attributed to the stronger Al–Y bonds as compared to Al–Sc, seen in figure 2.

3.4. Mechanical and elastic properties

The mechanical and elastic properties are dependent on the crystal structure and bonding between atoms. In table 2 we present calculated single crystal elastic constants Cij at 0 GPa for (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC along with

polycrys-talline Voigt moduli; bulk modulus B, shear modulus G, and

Figure 2. Calculated (a) and (d) electron density of states, (b) and (e) projected crystal orbital Hamilton population (pCOHP), and (c) and (f) integrated projected crystal orbital Hamilton population (IpCOHP) in (a)–(c) (W2/3Sc1/3)2AlC and (d)–(f) (W2/3Sc1/3)2AlC. Schematic

along (g) [0 1 0] and (h) [1 1 0] zone axes for considered interactions used in pCOHP analysis.

(6)

Young’s modulus E as derived from Cij using equations (1), (3) and (5). In addition, we also present the Reuss moduli, cal-culated from the compliance constants obtained by inverting the 6 × 6 elastic-constant matrix, derived using equations (2), (4) and (5), together with the arithmetic means of the Voigt and Reuss moduli, i.e. the Voigt–Reuss–Hill (VRH) averages. Both i-MAX phases investigated herein fulfill the mechanical stability criterion for monoclinic structures [48].

First to note is that substitution of Sc for Y in general leads to decreased moduli. This decrease is greater for the Reuss moduli than for the Voigt moduli. We also note that the Reuss moduli are significantly lower than the Voigt moduli, in par-ticular when it comes to the bulk modulus: BR is 37% lower

than BV for (W2/3Sc1/3)2AlC, and 45% lower than BV for

(W2/3Y1/3)2AlC. A similarly large difference is seen for the

two i-MAX phases (Mo2/3Sc1/3)2AlC and (Mo2/3Y1/3)2AlC,

which suggests that it is important to calculate both Voigt and Reuss moduli for i-MAX phases [27].

Comparison of calculated moduli of (W2/3Sc1/3)2AlC with

the hypothetical W2AlC (BV = 214 GPa, GV = 110 GPa, EV = 280 GPa) and Sc2AlC (BV = 88 GPa, GV = 57 GPa, EV = 140 GPa) in [49] shows that all moduli are in-between the two M2AX phases. Both GV and EV are strengthened by

~10% as compared to the arithmetic mean for the two M2AX

phases. Generally, a large shear modulus G of a material is an indication of well-defined directional bonding between atoms. Here, GV is 100 and 99 GPa for (W2/3Sc1/3)2AlC and

(W2/3Y1/3)2AlC, respectively. This indicates rather similar

bonding, or more likely, a comparable total bond strength in the two i-MAX phases, which is confirmed by the bonding anal-ysis in figure 2 where W–C and Al–Al is found to be slightly stronger for (W2/3Sc1/3)2AlC compared to (W2/3Y1/3)2AlC.

This is, however, compensated by the latter having stronger interactions between C–Y and Al–Y as compared to C–Sc and Al–Sc.

3.5. Comparison of i-MAX phases where M1 = Cr, Mo, and W

To investigate the effect of M-element on bonding and related properties in i-MAX, we depict IpCOHP for five selected interactions and the VRH moduli as function of M1

in (M1

2/3Sc1/3)2AlC and (M2/31 Y1/3)2AlC for M1 = Cr, Mo,

and W. VRH values for Cr- and Mo-based i-MAX phases are from [27, 29] while IpCOHP values for Mo-based i-MAX phases are taken from [27]. All interactions, except for Al–M2 and C–M2, are stronger when M2 = Sc. For both

Sc and Y, we find that both C–M and Al–M interactions are strengthened when going from Cr to Mo to W. For the C–M2, Al–Al, and Al–M2 there is a decrease when going

from Cr to Mo followed by no change or a slight increase in their interaction when going from Mo to W. These trends in interaction strengths are reflected in the trends found for the moduli in figure 4(b). The moduli for Cr-based i-MAX are in most cases higher than for Mo-based ones, which can be related to the comparatively stronger Al–Al and C–M2 Figure 3. Phonon dispersion (left panels) and total and atomic phonon density of states (right panels) in (a) and (b) (W2/3Sc1/3)2AlC and (c)

(7)

M Dahlqvist et al

6

interactions. The increased moduli when going from Mo to W can be explained by the steady decrease in Al–M inter-action strength and the dominating C–M interaction while the other interactions are unaffected. These non-linear trends

found for moduli and for some of the bonding interactions for Cr, Mo, and W may be a result from size differences for

M1 and M2 as well as different electronegativities. The size

of the M1 atom (1.28 Å for Cr and 1.39 Å for Mo and W)

may explain why selected bonding interactions are changed from Cr to Mo but not from Mo to W. This is most notable for the Al–Al interaction. Moreover, the electronic properties will also influence the bonding interactions where both Mo (2.16) and W (2.36) are significantly more electronegative than Cr (1.66). Finally, comparing the moduli with a well-known traditional MAX phase, Ti2AlC with B = 138 GPa, G = 113 GPa, E = 267 GPa [50], we find similar B but lower

G and E for the i-MAX phases.

4. Conclusions

Two new atomically laminated W-based compounds, (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC, belonging the family

of i-MAX phases, have recently been synthesized, and are herein theoretically explored in terms of electronic structure, bonding characteristics, and mechanical properties. Both phases are described by a monoclinic structure of space group

C2/c (#15) and are dynamically stable. They also display a clear metallic character. Substitution of Sc for Y (M2),

how-ever, leads to reduced bond strength for W–C and Al–Al inter-actions, while M2–C and M2–Al interactions are strengthened,

resulting in slightly decreased moduli; from B = 143 GPa,

G = 94 GPa, and E = 231 GPa to B = 130 GPa, G = 91 GPa, and E = 219 GPa, respectively. Extending the comparison to other Y-based i-MAX phases, the moduli increases along the series where M1 = Cr, Mo, and W.

Acknowledgments

The calculations were carried out using supercomputer resources provided by the Swedish National Infrastructure for Computing (SNIC) at the National Supercomputer Cen-tre (NSC), the High Performance Computing Center North (HPC2N), and the PDC Center for High Performance Com-puting. We acknowledge support from the Knut and Alice Wal-lenberg (KAW) Foundation for a Fellowship Grant and Project funding (KAW 2015.0043), and from the Swedish Foundation for Strategic Research (SSF), program EM16-0004.

ORCID iDs

Martin Dahlqvist https://orcid.org/0000-0001-5036-2833 References

[1] Barsoum M W 2000 Prog. Solid State Chem. 28 201 [2] Barsoum M W 2013 MAX Phases, Properties of Machinable

Ternary Carbides and Nitrides (New York: Wiley)

[3] Eklund P, Beckers M, Jansson U, Högberg H and Hultman L 2010 Thin Solid Films 518 1851

[4] Barsoum M W and El-Raghy T 2001 Am. Sci. 89 334

Table 2. Calculated elastic constants Cij and Voigt (V), Reuss (R)

and VRH moduli (in GPa) of (W2/3Sc1/3)2AlC and (W2/3Y1/3)2AlC.

The Reuss moduli have been calculated from the compliance constants obtained by inverting the 6 × 6 elastic-constant matrix.

(W2/3Sc1/3)2AlC (W2/3Y1/3)2AlC C11 319 293 C12 117 114 C13 117 111 C15 85 89 C22 301 295 C23 85 80 C25 65 74 C33 318 309 C35 78 86 C44 112 110 C46 −0.74 −1.26 C55 91 92 C66 92 94 BV 175 167 BR 112 92 BVRH 143 130 GV 100 99 GR 88 83 GVRH 94 91 EV 253 247 ER 209 192 EVRH 231 219

Figure 4. (a) IpCOHP for five selected interactions and (b) Voigt Reuss–Hill (VRH) moduli as function of M in (M2/3Sc1/3)2AlC and

(M2/3Y1/3)2AlC where M = Cr, Mo, and W [27, 29].

(8)

[5] Farle A-S, Kwakernaak C, van der Zwaag S and Sloof W G 2015 J. Eur. Ceram. Soc. 35 37

[6] Sloof W G et al 2016 Sci. Rep. 6 23040

[7] Barsoum M W, Zhen T, Kalidindi S R, Radovic M and Murugaiah A 2003 Nat. Mater. 2 107

[8] Ingason A S et al 2013 Phys. Rev. Lett. 110 195502 [9] Mockute A, Dahlqvist M, Emmerlich J, Hultman L,

Schneider J M, Persson P O Å and Rosen J 2013 Phys. Rev. B 87 094113

[10] Dahlqvist M et al 2016 Phys. Rev. B 93 014410 [11] Naguib M, Kurtoglu M, Presser V, Lu J, Niu J, Heon M,

Hultman L, Gogotsi Y and Barsoum M W 2011 Adv. Mater.

23 4248

[12] Naguib M, Mochalin V N, Barsoum M W and Gogotsi Y 2014

Adv. Mater.26 992

[13] Anasori B, Lukatskaya M R and Gogotsi Y 2017 Nat. Rev.

Mater.2 16098

[14] Shahzad F, Alhabeb M, Hatter C B, Anasori B, Man Hong S, Koo C M and Gogotsi Y 2016 Science 353 1137

[15] Meng F L, Zhou Y C and Wang J Y 2005 Scr. Mater. 53 1369 [16] Meshkian R, Ingason A S, Arnalds U B, Magnus F, Lu J and

Rosen J 2015 APL Mater. 3 076102

[17] Halim J, Chartier P, Basyuk T, Prikhna T, Caspi E A N,

Barsoum M W and Cabioc’h T 2017 J. Eur. Ceram. Soc. 37 15 [18] Jiang C and Chroneos A 2017 Phys. Chem. Chem. Phys.

201173–1180

[19] Saltas V, Horlait D, Sgourou E N, Vallianatos F and Chroneos A 2017 Appl. Phys. Rev. 4 041301 [20] Cabioch T, Eklund P, Mauchamp V, Jaouen M and

Barsoum M W 2013 J. Eur. Ceram. Soc. 33 897 [21] Nechiche M, Cabioc’h T, Caspi E N, Rivin O, Hoser A,

Gauthier-Brunet V, Chartier P and Dubois S 2017 Inorg.

Chem.5614388–14395

[22] Liu Z et al 2014 Acta Mater. 73 186

[23] Anasori B, Halim J, Lu J, Voigt C A, Hultman L and Barsoum M W 2015 Scr. Mater. 101 5

[24] Anasori B et al 2015 J. Appl. Phys. 118 094304 [25] Meshkian R, Tao Q, Dahlqvist M, Lu J, Hultman L and

Rosen J 2017 Acta Mater. 125 476

[26] Tao Q et al 2017 Nat. Commun. 8 14949

[27] Thore A and Rosen J 2017 Phys. Chem. Chem. Phys.

19 21595

[28] Dahlqvist M, Lu J, Meshkian R, Tao Q, Hultman L and Rosen J 2017 Sci. Adv. 3 e1700642

[29] Lu J, Thore A, Meshkian R, Tao Q, Hultman L and Rosen J 2017 Cryst. Growth Des. 17 5704

[30] Meshkian R et al 2018 Adv. Mater. 301706409 [31] Persson I, Ghazaly A E, Tao Q, Halim J, Kota S,

Darakchieva V, Palisaitis J, Barsoum M W, Rosen J and Persson P O Å 2018 Small 141703676

[32] Kresse G and Hafner J 1993 Phys. Rev. B 47 558

[33] Kresse G and Furthmüller J 1996 Comput. Mater. Sci. 6 15 [34] Kresse G and Furthmüller J 1996 Phys. Rev. B 54 11169 [35] Perdew J P, Burke K and Ernzerhof M 1996 Phys. Rev. Lett.

77 3865

[36] Blöchl P E 1994 Phys. Rev. B 50 17953

[37] Kresse G and Joubert D 1999 Phys. Rev. B 59 1758 [38] Monkhorst H J and Pack J D 1976 Phys. Rev. B 13 5188 [39] Maintz S, Deringer V L, Tchougreeff A L and Dronskowski R

2016 J. Comput. Chem. 37 1030

[40] Togo A, Oba F and Tanaka I 2008 Phys. Rev. B 78 134106 [41] Wang J and Sehitoglu H 2014 Int. J. Plast. 61 17

[42] Momma K and Izumi F 2011 J. Appl. Crystallogr. 44 1272 [43] Dahlqvist M, Petruhins A, Lu J, Hultman L and Rosen J 2018

in preparation

[44] Etzkorn J, Ade M and Hillebrecht H 2007 Inorg. Chem.

46 7646

[45] Dahlqvist M, Alling B and Rosén J 2010 Phys. Rev. B

81 220102

[46] Zhang H, Hu T, Wang X, Li Z, Hu M, Wu E and Zhou Y 2015

Sci. Rep.5 14192

[47] Dahlqvist M and Rosen J 2015 Phys. Chem. Chem. Phys.

17 31810

[48] Mouhat F and Coudert F-X 2014 Phys. Rev. B 90 224104 [49] Cover M F, Warschkow O, Bilek M M M and McKenize D R

2009 J. Phys.: Condens. Matter 21 305403

[50] Cover M F, Warschkow O, Bilek M M M and McKenzie D R 2008 Adv. Eng. Mater. 10 935

References

Related documents

Skolan där vi gjorde vår studie är ett bra exempel eftersom de har gått igenom de olika problemen, båda tekniska som pedagogiska, som ”Att skriva sig till läsning” med datorn

All, except the last one, are concerned with the manipulation of Boolean functions typically used to model problems at the logic synthesis step of the integrated circuit design

This study evaluated the use of multiple- locus variable- number of tandem repeats analysis (MLVA) and compared results with antimicrobial resistance (AMR) determina- tions for 100

Denna frihet inom äktenskapet som verkar rymmas inom det habitus som han har, ger en möjlighet för jämställdhet utifrån att båda kan förvärva kulturellt kapital och

Vid vår datainsamling uppmärksammade vi exempelvis flera studier angående sambandet mellan lässvårigheter hos barn och våld i hemmet, som kan ses som en indikator på våld

By computing a perceptual linearization of the parameter space of a video operator, the user can explore this interpolated space to find the best set of parameters in a robust

Figure 1: Familywise error rates for SPM8 as function of paradigm and repetition time, when global normalization was turned o ff and motion regressors not were included.. The

När det gäller åtgärder i elevens relation till övriga förekommer åtgärden att ingå i liten grupp i fyra av fem åtgärdsprogram skrivna i matematik