See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/1859185

First-principles thermoelasticity of bcc iron under pressure

Article  in  Physical Review B · December 2006

DOI: 10.1103/PhysRevB.74.214111 · Source: arXiv

CITATIONS READS
62 106

2 authors:

Xianwei Sha Ronald E Cohen

Engility Corporation University College London and Carnegie Institution for Science
53 PUBLICATIONS   1,962 CITATIONS    355 PUBLICATIONS   18,521 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hydrogen properties under extreme (Pressure-temperature) conditions View project

All content following this page was uploaded by Xianwei Sha on 19 March 2015.

The user has requested enhancement of the downloaded file.

 First-principles thermoelasticity of bcc iron under pressure

Xianwei Sha and R. E. Cohen

Carnegie Institution of Washington, 5251 Broad Branch Road, NW, Washington DC 20015, USA

We investigate the elastic and isotropic aggregate properties of ferromagnetic bcc iron as a function

of temperature and pressure, by computing the Helmholtz free energies for the volume-conserving

strained structures using the first-principles linear response linear-muffin-tin-orbital method and the

generalized-gradient approximation. We include the electronic excitation contributions to the free

energy from the band structures, and phonon contributions from quasi-harmonic lattice dynamics.

All the elastic moduli increase with increasing pressure, and decrease with increasing temperature.

The isotropic aggregate sound velocities obtained based on the calculated elastic moduli agree well

with available ultrasonic and diamond-anvil-cell data. Birch’s law, which assumes a linear increase

in sound velocity with increasing atomic density, fails for bcc Fe under extreme conditions. First-

principles linear response lattice dynamics is shown to provide a tractable approach to examine the

elasticity of transition metals at high pressures and high temperatures.

PACS number(s): 62.20.Dc, 46.25.Hf, 31.15.Ar, 71.20.Be

1
I. Introduction

Information on the influences of pressure and temperature on the elastic moduli and

related aggregate properties of single crystals plays an essential role in predicting and understanding

the interatomic interactions, strength, mechanical stability, phase transition mechanisms and

dynamical response of materials. During the past several decades, considerable experimental efforts

have been devoted to examine the elasticity of bcc iron, as well as its temperature and pressure

dependences.1-10 There are also some first-principles data available, but these calculations only

focus on elasticity at zero temperature, without any thermal effects included.11, 12 Here we have

performed first-principles quasiharmonic lattice dynamics study to examine the elastic moduli of

bcc Fe with pressures and temperatures, using the full-potential linear response linear-muffin-tin-

orbital (LMTO) method.

II. Theoretical methods

For a cubic crystal, the three elastic moduli C11, C12 and C44 fully describe its elastic

behavior. C11 and C12 can be determined from the bulk modulus K and shear constant Cs

K=(C11+2C12)/3 (1)

Cs=(C11-C12)/2 (2)

In order to make direct comparisons to the ultrasonic experimental measurements, one

should use the adiabatic bulk modulus Ks13

Ks=(1+α γ T)*KT (3)

where α is the thermal expansion coefficient, γ is the Grüneisen parameter, and T is the temperature.

The isothermal bulk modulus KT was determined according to the Vinet equation of state (EoS).14,
15
We recently reported first-principles thermal EoS properties for ferromagnetic bcc Fe, including

2
α, γ and KT as a function of temperature and pressure,16 from which we obtained the adiabatic bulk

modulus Ks using Eq. (3).

For cubic crystals, the shear modulus Cs describes materials resistance to shear

deformation across the (110) plane in the [1 1 0] direction, and C44 is the resistance to shear

deformation across the (100) plane in the [010] direction. We obtained both the shear moduli by

shearing the cubic lattice at constant volumes.17 The elastic moduli we present are those that appear

in the equations of motion and directly give sound velocities.18-20 The following tetragonal strains

are applied to obtain Cs

⎛δ 0 0 ⎞
⎜ ⎟
ε = ⎜0 δ 0 ⎟ (4)
⎜ 0 0 (1 + δ ) −2 − 1⎟
⎝ ⎠

where δ is the strain magnitude. The Helmholtz free energy of the strained structure F(δ) is related

to δ as:

F (δ ) = F (0) + 6C sVδ 2 + O(δ 3 ) (5)

with V as the volume, and F(0) as the free energy for the unstrained structure.

In a similar way, C44 is calculated by applying the volume-conserving orthorhombic

strain:

⎛0 δ 0 ⎞
⎜ ⎟
ε = ⎜δ 0 0 ⎟ (6)
⎜ 0 0 δ 2 /(1 − δ 2 ) ⎟
⎝ ⎠

Here the free energy is:

F (δ ) = F (0) + 2C 44Vδ 2 + O(δ 4 ) (7)

For many metals and alloys, the Helmholtz free energy F can be accurately separated

as13

3
 F(V,T,δ)=Estatic(V,δ)+Fel(V,T,δ)+Fph(V,T,δ) (8)

Estatic(V,δ) is the energy at zero temperature, Fel(V,T,δ) is the thermal free energy arising from

electronic excitations, and Fph(V,T, δ) is the phonon contribution. We obtain both Estatic(V,δ) and

Fel(V,T,δ) from first-principles calculations directly, and use linear response lattice dynamics to

examine the lattice vibrational contribution in the quasi-harmonic approximation. Our

computational approach is based on the density functional theory and density functional

perturbation theory, using multi-κ basis sets in the full potential LMTO method.21, 22 The induced

charge densities, the screened potentials and the envelope functions are represented by spherical

harmonics up to lmax =6 within the non-overlapping muffin-tin spheres surrounding each individual

atom, and by plane waves in the remaining interstitial region with cutoff corresponding to the

16×16×16 fast-Fourier-transform grid in the unit cell of direct space. The k-space integration

needed for constructing the induced charge density is performed over the 16×16×16 grid. We use

the Perdew-Burke-Ernzerhof (PBE) generalized-gradient approximation (GGA) for the exchange

and correlation energy.23 When calculating Fel(V,T), we assume temperature-independent

eigenvalues for given lattice and nuclear positions.16, 24, 25 We determine the dynamical matrix for a

set of irreducible q points at the 8×8×8 reciprocal lattice grid. The perturbative approach is

employed for calculating the self-consistent change in the potential.26, 27 Careful convergence tests

have been made against k and q point grids and many other parameters. See Ref. 16 for more

computational details.

We calculate the elastic moduli at six different volumes 65, 70, 75, 79.6, 85 and 90

bohr3/atom, and apply several different volume-conserving strain magnitudes varying from 1 to 3%

at each given volume. For each configuration we perform first-principles linear response LMTO

calculations to obtain the band structure and lattice dynamics information, and then computed the

Helmholtz free energies for temperatures from 0 K to 2000 K in intervals of 250 K. We derived the

4
elastic moduli Cs and C44 by fitting the calculated Helmholtz free energies at each given volume and

temperature to Eqs. (5) and (7), respectively. We obtain the related pressure information at the

given volume and temperature from the first-principles thermal EoS.16

III. Results and Discussions

We present the calculated static elastic moduli of ferromagnetic bcc Fe as a function of

volume (pressure) in Fig. 1 and Table 1. Several sets of experimental moduli have been published,

measured by inelastic neutron scattering,9 synchrotron x-ray diffraction,8 low-frequency torsional,10

and ultrasonic techniques such as rectangular parallelepiped resonance (RPR)7 and phase

comparison method.6 At both the ambient and high pressures, the first principles calculations

generally overestimate the elastic moduli. One of the major reasons for the differences here is due to

the errors in the thermal equation of state. GGA approximation usually leads to smaller equilibrium

volumes for bcc Fe.11, 12, 16, 28 The agreement between the calculated moduli and the experiment is

better if compared at constant volumes, as shown in Fig. 1. A couple of first-principles GGA

calculations were previsously performed using the plane-wave VASP code to examine the elasticity

of bcc Fe at zero temperature,11, 12 which underestimated C12 and C44, but overestimated C11. There

are several aspects to account for the differences between the current and pervious theoretical

calculations: the current calculation use the full potential LMTO method, and earlier calculations

used ultrasoft pseudopotentials12 and the projector-augmented-wave (PAW) method11, respectively;

both the previous calculations used the PW91 GGA functional,29 while current calculations used the

more recent PBE functional;23 different Eos formulations might have been used to obtain the bulk

modulus; and the equilibrium volumes at zero temperature are slightly different.

Although the pressure dependences of the elastic moduli of bcc Fe were studied using

ultrasonic methods almost forty years ago,1, 5

earlier studies were generally limited to very low

5
pressures (~1GPa). Only within the past decade, new experimental techniques such as synchrotron

x-ray diffraction8 and inelastic neutron scattering9 have been applied to examine the elastic

properties at much higher pressures. All the calculated elastic moduli show a strong increase with

the increase of pressure, except for Cs at very high pressures. The softening of Cs indicates the

dynamic instability of bcc structure, consistent with earlier theoretical predictions.28, 30 Despite the

offset in the values, the calculated slopes of the elastic moduli with respect to the volume shows

good agreements with experiment. The agreement between current and earlier PAW calculations11

are good, except at large volumes. Some of the discrepancies might be attributed to the different

GGA functional used. The large differences for Ks at large volumes mainly come from the different

equation of state formulations. Caspersen et al. used the Murnaghan EoS, and we employ the Vinet

EoS. Cohen et al. examined the accuracy of various EoS formulations, and found the Vinet

equation to be the most accurate.31 We also compared with the full potential linear-augmented-

plane-wave (LAPW) calculated bulk modulus using the third order Birch-Murnaghan EoS.28

Although the LAPW and PAW calculations used the same PW91 GGA functional, the LAPW bulk

modulus agrees better with current studies at large volumes.

For volumes less than ambient equilibrium value, the anisotropy ratio A=C44/Cs shows a

strong increase with increasing pressure. This is different from bcc Ta, where A first decreases for

P<100GPa, and then increases with further increase in pressure.17

Ultrasonic techniques have been widely used to examine the temperature dependence of

the elastic moduli for single-crystal bcc Fe at ambient pressure.4, 6, 7, 32

In Fig. 2 we show the

calculated moduli C11, C12 and C44 as functions of temperature at pressures from 0 to 40 GPa in

intervals of 10 GPa, in comparison to these experimental data. Although the current first principle

calculations overestimated C11, both C12, C44 and the calculated temperature dependences of the

elastic moduli show good agreements with the experiments at ambient pressure. The calculated

6
moduli decrease with temperature in a quite linear manner at all pressures, consistent with earlier

ultrasonic measurements by Leese and Lord Jr.4 At temperatures above 750 K, the ultrasonic

measured moduli show a more rapid drop than the initial linear slope, which might be associated

with the effects of changing spin order on the elastic properties at high temperatures.7 Dever

reported that the magnetic contributions to C11, Cs and C44 at Debye temperature Tθ (473 K) are 6%,

22%, and 4%, respectively6. Isaak and Masuda found that magnetic contributions account for 28%,

44% and 12% to the change in the C11, Cs and C44 moduli from Tθ to the Curie temperature Tc

(1043 K).7 Although we performed spin-polarized total-energy calculations to study the properties

for ferromagnetic bcc Fe, we have not included the contributions from magnetic fluctuations to the

free energies and elastic moduli. This might be the main reason for the bigger differences between

the calculations and the experiment as approaching the Curie temperature.

The temperature dependence of elastic moduli is often expressed as (1/Cij)(dCij/dT). At

ambient conditions, our calculations give the following values for bcc iron:

1 dC11
= −2.1 × 10 −4 K −1
C11 dT

1 dC 44
= −1.6 × 10 −4 K −1 (9)
C 44 dT

1 dC '
= −2.2 × 10 −4 K −1
C ' dT

Where C’=(C11+C12+2C44)/2. These are in good agreement with ultrasonic data by Leese and Lord

Jr,4 -2.52, -1.90 and -2.20×10-4 K-1 for C11, C44 and C’, respectively.

Sound velocities in solids are related to the elastic moduli according to the Christoffel

equation.17 For cubic polycrystalline sample, the average isotropic shear modulus G can be

determined from single crystal elastic moduli according to the Voigt-Reuss-Hill scheme13

7
 1
G V = (2C s + 3C 44 )
5

1 2 3 −1
GR = [ ( + )] (10)
5 C s C 44

1 V
GH = (G + G R )
2

We show the temperature dependences of the calculated adiabatic bulk modulus Ks and

Hill-averaged G at different pressures in Fig. 3. Compared to the experiments, first principle

calculations overestimated both Ks and G, which can be largely attributed to the errors in the EoS.

The calculated temperature dependences show good agreements between the theory and the

ultrasonic experiment by Leese and Lord Jr.4 The ultrasonic data of Dever6 and Isaak and Masuda7

show a more rapid nonlinear drop at high temperatures, which might be associated with the degree

of ferromagnetic ordering. Low-frequency torsional measurements gave much smaller values for G

at high temperatures, 29 and 38 GPa at frequencies of 0.01 and 1 Hz at TC, respectively.10 These

values are around 14-29 GPa less than the ultrasonic measurements, and the differences might be

attributed to the significant viscoelastic relaxation in bcc Fe at high temperature.7 The torsional

oscillation and creep tests at seismic frequencies and high temperatures reveal intense viscoelastic

relaxation in bcc Fe.10

The three isotropically averaged aggregate sound velocities can be derived from bulk

modulus K and shear modulus G

4
v P = [( K + G ) / ρ ]1 / 2
3

v B = ( K / ρ )1 / 2 (11)

v S = (G / ρ )1 / 2

8
Where vP, vB and vS are the compressional, bulk and shear sound velocities, respectively, and ρ is

atomic density.

For [110] wave propagation direction in a cubic lattice, the longitudinal mode is

ρv 2 = (C11 + C12 + 2C 44 ) / 2 (12)

and the two transverse modes are

ρv 2 = (C11 − C12 ) / 2 = C s
(13)
2
ρv = C 44

We show the calculated sound velocities of bcc Fe as a function of atomic density in Fig.

4, which agree within 3% with the ultrasonic1, 5 and recent inelastic X-ray scattering33 experimental

data. The velocities at a given volume agree much better with the experiment then the moduli at a

given pressure due to two major reasons: the errors in the thermal equation of state; and the fact that

the sound velocities are the square root of the bulk and shear moduli. We show the calculated shear

wave velocities as a function of atomic density at three different temperatures in Fig. 5. When the

atomic density varies from 7.5 to 9 g/cm3, the computed values closely obey Birch's Law, which

predicts that the velocity of each material is linear with atomic density. The linear slope of the

Birch’s plots shows an increase with increasing temperature, similar to what has been observed in

hcp Fe in recent inelastic x-ray scattering experiment.34 However, with further compression, the

computed velocities show a strong deviation from Birch’s Law, and the effects are more

pronounced at high temperatures. The large deviation occurs at high pressures (P> 25GPa), where

bcc is still dynamically stable, but not thermodynamically stable.16 At room temperature, bcc Fe

transforms to hexagonal-close-packed (hcp) structure at ~ 11GPa.35 Recent experiments show that

the compressional and shear wave velocities of hcp Fe at high pressures and temperatures could not

be fitted to Birch’s law either.34

9
IV. Conclusions

The elasticity and sound velocity of bcc Fe are presented from first-principles linear

response calculations. Generally the calculated moduli are in fairly good agreements with

ultrasonic, inelastic neutron scattering and x-ray diffraction measurements. However, there are

some systematic shifts in moduli and the thermal equation of state from experiment, so that an

improved density functional for ferromagnetic bcc Fe is desirable. Elastic moduli normally show an

increase with increasing pressure and a quite linear decrease with temperature. The temperature and

pressure dependences of the calculated moduli agree with experiment. The isotropic aggregate

sound velocities are obtained based on the calculated elastic moduli, which show good agreements

with available ultrasonic and diamond-anvil-cell data. The isotropic wave velocities follow the

Birch’s law only when the pressure is less than 25 GPa, with an increase in the slope of the Birch’s

lines with increasing temperatures.

ACKNOWLEDGEMENTS

We thank S. Y. Savrasov for kind agreement to use his LMTO codes and many helpful discussions.

This work was supported by DOE ASCI/ASAP subcontract B341492 to Caltech DOE w-7405-

ENG-48. Computations were performed on the Opteron Cluster at the Geophysical Laboratory and

ALC cluster at Lawrence Livermore National Lab, supported by DOE and the Carnegie Institution

of Washington.

10
REFERENCES
1
G. Simmons and H. Wang, Single crystal elastic constants and calculated aggregate

properties: a handbook (M.I.T. Press, Cambridge, Mass., 1971).

2
A. E. Lord Jr. and D. N. Beshers, J. Appl. Phys. 36, 1620 (1965).
3
C. A. Rotter and C. S. Smith, J. Phys. Chem. Solids 27, 267 (1966).
4
J. Leese and A. E. Lord Jr., J. Appl. Phys. 39, 3986 (1968).
5
M. W. Guinan and D. N. Beshers, J. Phys. Chem. Solids 29, 541 (1968).
6
D. J. Dever, J. Appl. Phys. 43, 3293 (1972).
7
D. G. Isaak and K. Masuda, J. Geophys. Res.-Solid Earth 100, 17689 (1995).
8
A. K. Singh, H. K. Mao, J. F. Shu, and R. J. Hemley, Phys. Rev. Lett. 80, 2157 (1998).
9
S. Klotz and M. Braden, Phys. Rev. Lett. 85, 3209 (2000).
10
I. Jackson, in High Pressure Science and Technology, Colorado Springs, 1993, edited by

S. C. Schmidt, J. W. Shaner, G. A. Samara and M. Ross (Am. Inst. of Phys., New York,

1994), p. 939.
11
K. J. Caspersen, A. Lew, M. Ortiz, and E. A. Carter, Phys. Rev. Lett. 93, 115501 (2004).
12
L. Vocadlo, G. A. de Wijs, G. Kresse, M. Gillan, and G. D. Price, Faraday Discuss. 106,

205 (1997).
13
O. L. Anderson, Equations of state of solids for geophysics and ceramic science (Oxford

University Press, New York, 1995).

14
P. Vinet, J. H. Rose, J. Ferrante, and J. R. Smith, J. Phys.-Condes. Matter 1, 1941 (1989).
15
P. Vinet, J. Ferrante, J. H. Rose, and J. R. Smith, J. Geophys. Res. 92, 9319 (1987).
16
X. W. Sha and R. E. Cohen, Phys. Rev. B 73, 104303 (2006).
17
O. Gulseren and R. E. Cohen, Phys. Rev. B 65, 064103 (2002).

11
18
T. H. K. Barron and M. L. Klein, Proc. Phys. Soc. London 85, 523 (1965).
19
E. Gregoryanz, R. Hemley, H. Mao, and P. Gillet, Phys. Rev. Lett. 84, 3117 (2000).
20
G. Steinle-Neumann, L. Stixrude, and R. E. Cohen, Phys. Rev. B 60, 791 (1999).
21
S. Y. Savrasov and D. Y. Savrasov, Phys. Rev. B 46, 12181 (1992).
22
S. Y. Savrasov, Phys. Rev. B 54, 16470 (1996).
23
J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett. 77, 3865 (1996).
24
X. W. Sha and R. E. Cohen, Phys. Rev. B 73, 104303 (2006).
25
E. Wasserman, L. Stixrude, and R. E. Cohen, Phys. Rev. B 53, 8296 (1996).
26
S. Y. Savrasov and D. Y. Savrasov, Phys. Rev. B 54, 16487 (1996).
27
S. Y. Savrasov, Phys. Rev. Lett. 69, 2819 (1992).
28
L. Stixrude, R. E. Cohen, and D. J. Singh, Phys. Rev. B 50, 6442 (1994).
29
J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh,

and C. Fiolhais, Physical Review B 46, 6671 (1992).

30
P. Soderlind, J. A. Moriarty, and J. M. Wills, Phys. Rev. B 53, 14063 (1996).
31
R. E. Cohen, O. Gulseren, and R. J. Hemley, Am. Miner. 85, 338 (2000).
32
J. A. Rayne and B. S. Chandrasekhar, Phys. Rev. 122, 1714 (1961).
33
G. Fiquet, J. Badro, F. Guyot, C. Bellin, M. Krisch, D. Antonangeli, H. Requardt, A.

Mermet, D. Farber, C. Aracne-Ruddle, and J. Zhang, Phys. Earth Planet. Inter. 143-144, 5

(2004).
34
J. F. Lin, W. Sturhahn, J. Y. Zhao, G. Y. Shen, H. K. Mao, and R. J. Hemley, Science

308, 1892 (2005).

35
C. S. Yoo, J. Akella, A. J. Campbell, H. K. Mao, and R. J. Hemley, Science 270, 1473

(1995).

12
 600
3.0

A=C44/Cs
500 C 11 2.5

2.0
Elastic Moduli (GPa)

400 K
1.5
60 70 80
C 12 Volume (Bohr3 )
300

C 44
200

Cs
100

0
65 70 75 80 85
V (bohr3 )
Fig. 1. The calculated static elastic constants for ferromagnetic bcc Fe as a function of

atomic volume (solid lines) at ambient temperature. Open circles, filled circles and open

diamonds with error bars are experimental data from inelastic neutron scattering (Ref. 9),

ultrasonic (Ref. 5) and X-ray diffraction (Ref. 8) measurements, respectively. Zero-

temperature theoretical PAW moduli (Ref. 11) and LAPW bulk modulus (Ref. 27) are

shown in dashed and dotted lines. The inset shows the volume dependence of the

anisotropy ratio.

13
 500
C11(GPa) 400
300
200

300
C12(GPa)

200

100

200
C44(GPa)

150

100

50
500 1000 1500
T (K)
Fig. 2. The calculated temperature dependences of the elastic moduli for ferromagnetic

bcc Fe, at pressures from 0 (lowest curve) to 40 GPa (uppermost curve) with 10 GPa

internal. Experimental data at ambient pressure from ultrasonic measurements (cross,

Ref. 6; open diamonds, Ref. 4; filled circles, Ref. 7) are also presented.

14
 400

320
Ks (GPa)

240

160

160
G (GPa)

120

80

40

500 1000 1500

T (K)
Fig. 3. The adiabatic bulk modulus Ks and isotropic shear modulus G of bcc Fe as a

function a temperature, at pressures from 0 (lowest curve) to 40 GPa (uppermost curve)

with 10 GPa internal. Experimental data at ambient pressure from ultrasonic (cross, Ref.

6; open diamonds, Ref. 4; filled circles, Ref. 7) and low-frequency torsional

measurements ( filled stars, Ref. 10) are also plotted.

15
 8.0

7.0 VP
Sound velocity (km/s)
6.0
VB
5.0
[001]
4.0
Vs
3.0 [1-10]

2.0
7.0 7.5 8.0 8.5 9.0 9.5
Density (g/cm3)

Fig. 4. Sound velocities of bcc Fe at ambient temperature calculated from elastic

constants. The isotropic aggregate velocities are shown by solids lines, with vP, vB and vS

standing for the compressional, bulk and shear sound velocities, respectively. Dotted

lines represent the longitudinal and two transverse sound velocities in the [110] direction,

with brackets showing the polarization of the shear waves. Experimental isotropic

compressional and shear velocities, measured by ultrasonic (open diamonds, Ref. 1; filled

circles, Ref. 5) and inelastic X-ray scattering (Cross, Ref.33) are also shown.

16
 4.5

4.0
vs (km/s)

3.5

3.0

2.5

2.0
7 8 9 10
Density (g/cm3)
Fig. 5. Compressional wave velocity for bcc Fe as a function of atomic density. Birch

law (linear dependence of the longitudinal acoustic sound velocity with the atomic

density, shown as solid, dashed and dotted lines for 250, 1000 and 2000 K results,

respectively) clearly fails when the atomic density is less than 7.5 g/cm3, showing by the

significant discrepancies from the computed data (filled circles, 250K; cross, 1000K; and

open diamonds, 2000K). The slope of the Birch lines shows an increase with temperature.

17
 TABLE I. The static elastic, bulk and isotropic shear moduli of ferromagnetic bcc Fe.

Moduli are given in units of GPa, temperature T are given in Kelvin, and volume V are

given in bohr3/atom.

T V C11 C12 C44 Ks G

0 GPa

LMTO-GGA 0 75.1 303 150 126 201 103

PAW-GGA [11] 0 77.1 271 145 101 187 84

PP-GGA [12] 0 76.9 289 118 115 175 102

ultrasonic [1] 4 79.0 243.1 138.1 121.9 173 87

LMTO-GGA 250 75.4 297 148 123 198 100

ultrasonic [4] 300 - 226 140 116 169 78

ultrasonic [6] 300 79.5 232.2 135.6 117 168 82

RPR [7] 300 79.7 230.5 133.3 116.3 166 82

neutron [9] 300 - 223 127 122 159 84

4.6 GPa

LMTO-GGA 250 73.8 326 167 137 220 110

X-Ray [8] 300 - 260 154 153 189 100

9.8 GPa

LMTO-GGA 250 72.1 360 188 152 245 120

neutrons [9] 300 - 283 167 146 206 101

18

View publication stats