scieee Science in your language
[en] (orig)

On the constitution and thermodynamic modeling of the phase diagrams Nb-Mn and Ta-Mn

Abstract

The constitution of the two phase diagrams Nb-Mn and Ta-Mn has been determined from light optical and transmission and scanning electron microscopy (LOM, TEM and SEM) with energy dispersive (EDX) as well as wavelength dispersive (WDX) X-ray spectroscopy, X-ray powder (XPD) and single crystal diffraction (XSCD), differential thermal analysis (DTA) and/or differential scanning calorimetry (DSC). The Laves phases NbMn2 and TaMn2 are the only binary compounds in these systems. High-temperature differential thermal analyses revealed congruent melting for NbMn2 with T,(NbMn2) = 1515 +/- 15 degrees C, whereas TaMn2 melts incongruently with T-m(TaMn2)= 1797 +/- 40 degrees C close to a depleted peritectic reaction. Both Laves phases engage in eutectic reactions l <-> (Mn) + Nb(Ta)Mn-2 (T-eut = 1220 +/- 10 degrees C at 4.9 at% Nb and T-eut = 1234 +/- 10 degrees C at 0.7 at% Ta, respectively). NbMn2 also forms a eutectic with (Nb): l <-> (Nb) + NbMn2 at T-eut = 1493 +/- 15 degrees C and 53.2 at% Nb. Mn shows remarkably large maximum solid solubilities of 19.4 at% Mn in (Nb) as well as of 21.3 at% Mn in (Ta). Detailed atom site distribution has been established for the Laves phases by means of temperature dependent X-ray single crystal data (both C14 - MgZn2-type). Combined data from XPD, EDX/WDX and SEM microstructure indicate that for both Laves phases extended homogeneity regions exist: Nb1+xMn2+x (62.5-73.0 at% Mn at 950 degrees C: -0.19 <= x <= 1.125) and Ta1+xMn2-x (59.5-68.5 at % Mn: -0.055 <= x <= 1.215). Density functional theory (DFT) calculations favor Nb(Ta)/Mn antisite occupation rather than defects. The phases, "NbMn" and "TaMn", adopted earlier in the literature as binary system inherent compounds, were shown (TEM, WDX electron microprobe data and X-ray Rietveld refinements) to be oxygen stabilized phases of the Ti4Ni2O type (so-called eta(eta)-phases) with modified Nb(Ta)/Mn site substitution to comply with the formula Nb(Ta)(3-x)Mn3+xO1-y (defect eta-W3Fe3C-type). From magnetic susceptibility and magnetization measurements, both oxide stabilized eta phases eta-Nb3Mn3O1-y and eta-Ta3Mn3O1-y were found to order ferromagnetically below T-c similar to 77 K, but the Laves phases NbMn2, TaMn2 reveal weakly temperature dependent paramagnetism. No trace of the rhombohedral kyphase (W6Fe7-type) has been encountered in our investigation of the two binary phase diagrams. Thermodynamic and transport properties (specific heat, electrical resistivity and magnetic susceptibility/magnetization) classify the Laves phases with metallic behavior whilst mechanical properties (elastic moduli from DFT and nanoindentation as well as hardness and thermal expansion) group both Laves phases among rather hard and brittle intermetallics. Based on (i) the experimentally derived constitution of the Nb-Mn and Ta-Mn systems, and (ii) on new own DFT data of the energy of formation of the Laves phases, a CALPHAD (CALculation of PHAse Diagrams) calculation of both systems was made providing a complete set of optimized thermodynamic data. Furthermore, the DFT calculations provided information on the instability of the eta-Ta3Mn3 structure and the atom-site specific stabilization effect of oxygen.

Read accessible full text

On the constitution and thermodynamic modeling of the phase diagrams Nb-Mn and Ta-Mn

Author: Yan, Xinlin
Publisher: Elsevier
Year: 2021
DOI: 10.1016/j.jallcom.2021.158715
Source: https://dspace.vsb.cz/bitstreams/99a9b000-1578-4df5-bf75-b5e56718a509/download
Con en s lis s a ailable a ScienceDi ec
Jou nal o Alloys and Compounds
jou nal homepage: www.else ie .com/loca e/jalcom
On he cons i u ion and he modynamic modeling o he phase diag ams
Nb-Mn and Ta-Mn
☆
Xinlin Yan
a,1
, Pa el B ož
b
, Jan Vřešťál
b
, Jiří Vlach
b
, Jiří Bu šík
c
, Ma ina Mazalo á
b
, Jana Pa lů
b
,
Bedřich Sme ana
d
, Ge da Rogl
a
, Ma kus Eibe ge
a
, And iy G y si
a
, He wig Micho
e
,
He be Mülle
e
, Ge ald Gies e
, Pe e Rogl
a,⁎
a
Ins i u e o Ma e ials Chemis y, Uni e si y o Vienna, Waeh inge s asse 42, A-1090 Wien, Aus ia
b
Depa men o Chemis y, Facul y o Science, Masa yk Uni e si y, Ko lářská 2, 611 37 B no, Czech Republic
c
Ins i u e o Physics o Ma e ials, Czech Academy o Sciences, Žižko a 22, 6160 00 B no, Czech Republic
d
Facul y o Ma e ials Science and Technology, VSB-TU Os a a, 17. lis opadu 15, 708 33 Os a a, Czech Republic
e
Ins i u e o Solid S a e Physics, TU-Wien, Wiedne Haup s asse 8-10, A-1040 Wien, Aus ia
Ins i u e o Mine alogy and C ys allog aphy, Uni e si y o Vienna, Al hans asse 14, A-1090 Wien, Aus ia
a icle in o
A icle his o y:
Recei ed 3 No embe 2020
Recei ed in e ised o m 5 Janua y 2021
Accep ed 9 Janua y 2021
A ailable online 22 Janua y 2021
Keywo ds:
In e me allics
C ys al s uc u e
Phase diag ams
The modynamic modeling
Magne ic measu emen s
The mal analysis
abs ac
The cons i u ion o he wo phase diag ams Nb-Mn and Ta-Mn has been de e mined om ligh op ical and
ansmission and scanning elec on mic oscopy (LOM, TEM and SEM) wi h ene gy dispe si e (EDX) as well
as wa eleng h dispe si e (WDX) X- ay spec oscopy, X- ay powde (XPD) and single c ys al di ac ion
(XSCD), di e en ial he mal analysis (DTA) and/o di e en ial scanning calo ime y (DSC). The La es phases
NbMn
2
and TaMn
2
a e he only bina y compounds in hese sys ems. High- empe a u e di e en ial he mal
analyses e ealed cong uen mel ing o NbMn
2
wi h T
m
(NbMn
2
) = 1515 ± 15 °C, whe eas TaMn
2
mel s in-
cong uen ly wi h T
m
(TaMn
2
) = 1797 ± 40 °C close o a deple ed pe i ec ic eac ion. Bo h La es
phases engage in eu ec ic eac ions ℓ ↔ (Mn) + Nb(Ta)Mn
2
(T
eu
= 1220 ± 10 °C a 4.9 a % Nb and
T
eu
= 1234 ± 10 °C a 0.7 a % Ta, espec i ely). NbMn
2
also o ms a eu ec ic wi h (Nb): ℓ ↔ (Nb) + NbMn
2
a
T
eu
= 1493 ± 15 °C and 53.2 a % Nb. Mn shows ema kably la ge maximum solid solubili ies o 19.4 a % Mn in
(Nb) as well as o 21.3 a % Mn in (Ta). De ailed a om si e dis ibu ion has been es ablished o he La es
phases by means o empe a u e dependen X- ay single c ys al da a (bo h C14 - MgZn
2
- ype). Combined
da a om XPD, EDX/WDX and SEM mic os uc u e indica e ha o bo h La es phases ex ended homo-
genei y egions exis : Nb
1+x
Mn
2−x
(62.5–73.0 a % Mn a 950°C: −0.19≤x≤0.125) and Ta
1+x
Mn
2−x
(59.5–68.5 a
% Mn: −0.055≤x≤0.215). Densi y unc ional heo y (DFT) calcula ions a o Nb(Ta)/Mn an isi e occupa ion
a he han de ec s. The phases, “NbMn” and “TaMn”, adop ed ea lie in he li e a u e as bina y sys em
inhe en compounds, we e shown (TEM, WDX elec on mic op obe da a and X- ay Rie eld e inemen s) o
be oxygen s abilized phases o he Ti
4
Ni
2
O ype (so-called e a(η)-phases) wi h modi ied Nb(Ta)/Mn si e
subs i u ion o comply wi h he o mula Nb(Ta)
3−x
Mn
3+x
O
1−y
(de ec η-W
3
Fe
3
C- ype). F om magne ic sus-
cep ibili y and magne iza ion measu emen s, bo h oxide s abilized e a phases η-Nb
3
Mn
3
O
1−y
and
η-Ta
3
Mn
3
O
1−y
we e ound o o de e omagne ically below T
c
~ 77 K, bu he La es phases NbMn
2
, TaMn
2
e eal weakly empe a u e dependen pa amagne ism. No ace o he hombohed al μ-phase (W
6
Fe
7
- ype)
has been encoun e ed in ou in es iga ion o he wo bina y phase diag ams. The modynamic and anspo
p ope ies (speci ic hea , elec ical esis i i y and magne ic suscep ibili y/magne iza ion) classi y he La es
phases wi h me allic beha io whils mechanical p ope ies (elas ic moduli om DFT and nanoinden a ion
as well as ha dness and he mal expansion) g oup bo h La es phases among a he ha d and b i le in-
e me allics. Based on (i) he expe imen ally de i ed cons i u ion o he Nb-Mn and Ta-Mn sys ems, and
h ps://doi.o g/10.1016/j.jallcom.2021.158715
0925-8388/© 2021 The Au ho s. Published by Else ie B.V.
CC_BY_4.0
☆
The pape is dedica ed o P o . D . Wol gang Jei schko, in memo iam.
]]]]
]]]]]]
⁎
Co esponding au ho .
E-mail add ess: pe e [email p o ec ed] (P. Rogl).
1
Cu en add ess: Ins i u e o Solid S a e Physics, TU-Wien, Wiedne Haup s asse 8-10, A-1040 Wien, Aus ia.
Jou nal o Alloys and Compounds 865 (2021) 158715
(ii) on new own DFT da a o he ene gy o o ma ion o he La es phases, a CALPHAD (CALcula ion o PHAse
Diag ams) calcula ion o bo h sys ems was made p o iding a comple e se o op imized he modynamic
da a. Fu he mo e, he DFT calcula ions p o ided in o ma ion on he ins abili y o he η-Ta
3
Mn
3
s uc u e
and he a om-si e speci ic s abiliza ion e ec o oxygen.
© 2021 The Au ho s. Published by Else ie B.V.
CC_BY_4.0
1. In oduc ion
In di e se echnological applica ions Nb, Ta and Mn a e playing
an impo an ole such as o ins ance: (i) addi ions o Mn a e able o
inc ease he yield s eng h o Ti-Nb-Ta-Mn alloy oams o biome-
dical implan s [1,2], (ii) niobium-manganese composi e elec odes
we e ound o be mo e sui able in supe capaci o s han niobium
elec odes [3]; (iii) Mn, Nb and bo on addi ions maximize s eng h
and oughness in ma ensi ic mic o-alloyed s eels o hea y-du y
engine connec ing ods [4]; acili a e he p oduc ion o seamless
s eel ubes [5]; Ta inc eases he pi ing co osion in supe duplex
s ainless s eels o ming (Ta,Mn) - oxysul ide [6], and (i ) Ha ano [7]
s udied he coun e ac ing e ec s o Nb and Mn on mic os uc u e
and oughness o 590 MPa class low ca bon baini ic s eels. Nb(Ta)-
o Mn-doping has been employed o imp o e he he moelec ic
beha io o pe o ski e mangani es [8,9], o NbFe
2−x
Mn
x
Al Heusle
alloys [10] and o highe manganese silicides (Nowo ny chimney
ladde s uc u es) [11].
Al hough he knowledge o phase diag ams and he modynamic
p ope ies is essen ial in de ining p ocessing condi ions o op imal
enginee ing p ope ies, eliable phase diag am in o ma ion on he
basic bina y sys ems Nb-Mn and Ta-Mn is s ill sca ce. Fo a de ailed
summa y o he expe imen al indings, see he compila ion o bina y
phase diag ams in Massalski [12]. The limi ed expe imen al da a
a ailable wi hin he pa ial phase diag am Nb-Mn conce n he e-
gion a ound he La es phase (50–80 a % Mn [13]) and he Mn- ich
liquidus/solidus, which has been de i ed om he mal analysis in
he egion om 88 o 100 a % Mn [14]. In e es ingly he in es iga ion
o he sys em Nb-Mn-B a 800 °C e ealed a bina y NbMn phase a
45–50 a % Mn (s uc u e unde e mined, W
6
Fe
7
- ype assumed [15]).
Such a phase was la e bu unknowingly accoun ed o as a μ-phase
(Nb
6
Mn
7
) in he he modynamic modeling o he Nb-Mn bina y as
pa o he Nb-Fe-Mn sys em [16].
F om he phase diag am o he Ta-Mn sys em hi he o only he
egion om 67 o 100 a % Mn has been expe imen ally de i ed [17]
wi h some addi ional da a on he ha dness o he TaMn
2
-La es phase
[13]. Being unawa e o an expe imen al diag am, Kau man [18]
p oduced a i s he modynamic calcula ion o he en i e phase
diag am Ta-Mn using s uc u e-insensi i e hea o o ma ion da a
de i ed om Miedema’s model [19] and ea ly ab ini io da a by Co-
line [20]: besides TaMn
2
, also a compound TaMn was shown o
exis . Howe e , he Gibbs ene gies o phases we e no based on
s anda d SGTE una y da a [21]. Al hough expe imen al da a a e only
a ailable o he Mn- ich pa and he ype and empe a u e o
mel ing o TaMn
2
emained unclea [17], he modynamic modeling
o he en i e phase diag am was pe o med by C. Wang e al. [22]
esul ing in a Mn- ich and a Ta- ich eu ec ic besides cong uen ly
mel ing TaMn
2
(T
m
= 1670 °C). A phase “TaMn” was no conside ed.
The La es phases in bo h sys ems ecei ed mo e a en ion: om
high- empe a u e di ec syn hesis calo ime y measu emen s
Meschel e al. [23] claimed a s anda d en halpy o o ma ion o
NbMn
2
o −31.2 ± 8.1 kJ/mol. .u. and −43.5 ± 7.5 kJ/mol. .u. o TaMn
2
.
This expe imen al alue o NbMn
2
, howe e , appea s signi ican ly
lowe han he Δ
298
H
° = −45.05 ± 3.64 kJ/mol. .u. ex ac ed om
d op isope ibolic calo ime y in a Ni ba h [24], which p o ed o be
consis en wi h a densi y unc ional heo y (DFT) calcula ion by Yan
e al. [24], who a i ed a −46.5 kJ/mol. .u. These alues compa e
well wi h olde ab ini io da a by Coline e al. (−42 kJ/mol. .u. o
NbMn
2
and −24 kJ/mol. .u. o TaMn
2
[20]), which we e e alua ed
wi hin a igh -binding scheme o he d band – an app oxima e
model calcula ion o es ima e he en halpies o o ma ion no
dealing wi h s uc u e and magne ic e ec s. Mo e ecen ab ini io
calcula ions o Yan e al. on NbMn
2
also p o ided an eDOS ( om
which a gamma alue o γ = 10.5 mJ/mol.K
2
is de i ed), as well as a
ull se o elas ic moduli [24] ( o de ails see Sec ion 4.4. Ha dness
and mechanical p ope ies).
Wi h espec o (a) he absence o eliable phase diag am da a in
bo h sys ems Nb-Mn and Ta-Mn as well as (b) conce ning he in-
consis encies on he exis ence o he phases "Nb
7
Mn
6
" and "TaMn",
and (c) inconsis encies in he hea o o ma ion and he lack o
physical p ope y da a o he La es phases, he aim o he p esen
pape is mani old: (i) o check on he c ys al s uc u e, he o ma ion
and s abili y o he sys em inhe en phases pa icula ly on he
"NbMn" (Nb
7
Mn
6
) and "TaMn" phases, (ii) o es ablish eliable phase
ela ions o he en i e phase diag ams Nb-Mn and Ta-Mn, (iii) o
p o ide physical p ope y da a o a de ailed cha ac e iza ion o he
La es phases, (i ) o calcula e ia ab ini io me hods he ene gy o
o ma ion o he La es phases, and ( ) o check on he DFT s abili y
o he "Nb(Ta)Mn" phases (pa icula ly on η-Ta
3
Mn
3
O
1−x
), as well as
( i) o p o ide a eliable se o he modynamic da a ia CALPHAD
modeling o bo h sys ems.
2. Expe imen al de ails
2.1. Syn hesis and cha ac e iza ion o physical p ope ies
The s a ing ma e ials we e o a minimal pu i y o 99.9 mass%: Ta
and Nb in he o m o ingo , oil o wi e om Good ellow, UK, and
elec oly ically deposi ed Mn pla ele s (>99.95%, Al a Aesa , D),
which we e su ace cleaned in concen a ed HNO
3
immedia ely
p io o use. Sample specimens (1–2 g each) we e p epa ed om
elemen al pieces by ei he a gon a c mel ing on a wa e -cooled
coppe hea h o on a wa e -cooled Hukin c ucible o a high e-
quency (HF) u nace in Ti-ge e ed a gon. To ensu e homogeniza ion,
all alloys we e e-mel ed se e al imes. A sligh excess o Mn was
used o allow o e apo a ion du ing usion in o de o keep he o al
mass loss o he sample a e mel ing below 0.5 mass%. Each alloy
was weighed ca e ully a e mel ing and, i necessa y, Mn was added
un il he nominal composi ion was achie ed a e inal mel ing. Due
o he high di e ences in he mel ing poin s o Nb(Ta) and Mn in
combina ion wi h he high apo p essu e o Mn, some o he spe-
cimens a e mel ing and annealing s ill u ned ou o be in-
homogeneous. The e o e, alloy specimens wi h mo e han 90% Mn
and a o al weigh o 2 g we e p epa ed om well blended powde
mix u es, which we e compac ed in s eel dies (Φ = 10 mm, wi hou
lub ican ) a a p essu e o 50 kg/cm
2
(~5 MPa). Nb(Ta) powde s o
99.9 mass% we e pu chased om Sigma-Ald ich, D; Mn powde was
ob ained om clean Mn-pieces eshly c ushed in a WC mo a (see
abo e).
A pa o each sample (wi hin an Al
2
O
3
c ucible) was sealed in a
silica capsule unde 280 mba A and hea - ea ed a 950 °C o
abou 120 h and quenched. A maximum annealing empe a u e he
A -p essu e inside he silica capsule eached abou 1 ba and
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
2
e icien ly supp essed Mn-e apo a ion. Fo annealing a 750 °C,
700 °C and 650 °C he ampullae we e kep a empe a u e o 1
mon h.
X- ay powde di ac ion (XPD) da a om as-cas and annealed
alloys we e collec ed employing a Guinie -Hube image pla e sys em
wi h monoch oma ed Fe K
α1
o Cu K
α1
adia ion (8°≤2θ≤100°).
P ecise la ice pa ame e s we e calcula ed by leas -squa es
i s o he indexed θ- alues wi h Ge as in e nal s anda d
(a
Ge
= 0.5657906 nm). Rie eld e inemen s we e made wi h he
FULLPROF p og am [25]. Single c ys als (SC) we e mechanically
isola ed om a c ushed alloy. Inspec ion on an AXS D8-GADDS
ex u e goniome e assu ed high c ys al quali y, uni cell dimensions
and Laue symme y o he single c ys al specimens p io o X- ay
in ensi y da a collec ion on a ou -ci cle APEX II di ac ome e
equipped wi h a CCD a ea de ec o and an Incoa ec Mic o ocus
Sou ce IμS (30 W, mul ilaye mi o , Mo-K
α
; λ = 0.071069 nm; de-
ec o dis ance o 3 cm; ull sphe e; 2°≤2θ ≤72°). Whe eas he SC o
TaMn
2
was s udied a oom empe a u e (RT), o NbMn
2
we col-
lec ed X- ay da a a ou empe a u es: 100 K, 150 K, 200 K and
300 K, cooled by a con inuous s eam o ni ogen gas enclosing he
c ys al a p ese empe a u e. Besides he gene al ea men o ab-
so p ion e ec s using he mul i-scan echnique (SADABS; e-
dundancy o in eg a ed e lec ions >8) [26], no indi idual abso p ion
co ec ion was necessa y because o he a he egula c ys al
shape and small dimensions o he in es iga ed specimens
(40 × 45 × 60 µm³). The c ys al s uc u e was sol ed applying di ec
me hods (P og am SHELXS-97) and e ined agains F
2
(P og am
SHELXL-97-2) wi hin he p og ams OSCAIL o WINGX [27]. Finally,
he c ys al s uc u e was s anda dized wi h he p og am S uc u e
Tidy [28].
All as-cas and annealed samples we e g ound on SiC pape s and
polished wi h Al
2
O
3
powde s (down o 0.3 µm) ia s anda d p oce-
du es and ha e been examined by ligh op ical me allog aphy (LOM)
and scanning elec on mic oscopy (SEM). The mic os uc u e and
chemical composi ion o he alloys we e analyzed by SEM on a Zeiss
Sup a 55 VP equipped wi h an ene gy dispe si e X- ay (EDX) de ec o
ope a ed a 20 kV. The non-me al con en in oxygen/ni ogen/ca bon
s abilized impu i y phases was de e mined om wa eleng h dis-
pe si e X- ay (WDX) analyses in a Jeol JSM-6460 scanning elec on
mic oscope ope a ed a 20 kV equipped wi h an Ox o d Ins umen s
mic oanalyse using Nb/Ta-L
α
, Mn-K
α
, O-K
α
, C-K
α
, Ν-K
α
and Si-K
α
a-
dia ion and spec ome e c ys als LiF ( o Mn), PET (Nb/Ta), LSM60
(O, C) and LSM80N (N) (PET = Pen ae y h i ol, LSM60 = W-Si supe -
la ice, LSM80N = Ni-C supe la ice). Quan i a i e e alua ion o com-
posi ions was pe o med wi h he INCA - so wa e [29].
Thin lamellae (la e al dimensions abou 10 × 7 µm
2
) o he TEM
s udy we e p epa ed om he Ta45Mn55 alloy a e long annealing
(2 mon hs a 1000 °C) using a ocused ion beam (FIB) echnique in a
TESCAN LYRA 3 XMU FEG/SEM×FIB scanning elec on mic oscope. A
Philips CM12 ansmission elec on mic oscope ope a ed a 120 kV
was used namely in di ac ion mode.
Elec ical esis i i y was measu ed om 4.2 K o oom empe a-
u e in a con en ional
4
He c yos a , elying on an in-house equip-
men (e o <3%). The speci ic esis ance was ob ained ia a dc ou -
poin echnique using a Lake Sho e Resis ance B idge 370 AC.
Fo speci ic hea measu emen s, we employed a Quan um Design
PPMS in he empe a u e ange om 2 o 300 K using Apiezon-N
g ease o ensu e a good he mal con ac be ween sample and sample
pla o m. Tempe a u e and ield dependen magne iza ion da a we e
collec ed a empe a u es anging om 3 o 298 K using a 6T CRY-
OGENIC SQUID magne ome e . Tempe a u e dependen ac suscep -
ibili y measu emen s we e ca ied ou om 4.2 o 150 K wi h a
e ised Lakesho e 7000 AC Suscep ome e [30], applying an ac ield
wi h an RMS ampli ude o 400 A/m and a equency o 200 Hz.
The mal expansion om 4.2 K o 300 K was measu ed in a min-
ia u e capaci ance dila ome e , using he il ed pla e p inciple [31].
Th ee di e en ypes o equipmen se ed o ob ain ha dness da a
(HV): (i) a mic oha dness es e , AD Paa MHT-4 moun ed on a Zeiss
Axioplan op ical mic oscope, employing loads o 0.1, 0.5, 1, 1.5 and
2 N, applying a a e o 0.1 N s
−1
and a loading ime o 10 s, e e ed o
as HV s a ic, (ii) a mic oinden e MHT4 wi h a Zeiss mic oscope,
e e ed o as HV dynamic (MI), using loads o 0.1, 0.5, 1, 1.5, 2, 3, 5 N
and a a e o 0.1 N s
−1
p o iding in pa allel o HV he Young’s mod-
ulus, E, om he inden a ion expe imen s, using he Poisson’s a io ν
(ν = 0.36 was aken om DFT calcula ions o NbMn
2
[32]) and (iii) a
nanoinden e ASMEC Una wi h QCSM module (Vicke s Ha dness
Tes V1, ISO 14577 s anda d me hod), e e ed o as HV dynamic (NI),
measu ing he inden a ions (a leas 30 imp in s pe load and
sample) wi h a load o 100 mN, a load a e o 100 mN/20s, an un-
loading a e o 10 mN/15s and p o iding in addi ion he Young’s
modulus o a gi en ν. To e alua e he ha dness da a o (i), he di-
agonal leng h, 2ℓ is measu ed and HV is calcula ed acco ding o:
=
×
=
°
HV 0.102 2Fsin
(2 )
0.1891F
(2 )
136
2
2 2
(1)
wi h F as inden a ion load. To ge eliable esul s, a leas 10 di e en
imp essions pe load we e e alua ed and he e o was calcula ed.
The e o o (ii) and (iii) was abou 5%.
2.2. Di e en ial he mal analysis (DTA) and di e en ial scanning
calo ime y (DSC) measu emen s
DTA and DSC measu emen s we e pe o med on annealed sam-
ples in a Ne zsch STA 409 CD/3/403/5/G appa a us wi hin sealed
qua z and/o Al
2
O
3
c ucibles, espec i ely, unde a s eam o 6 N
a gon. Sealed qua z ampules we e used o p e en e apo a ion o
ola ile Mn om he samples du ing he measu emen ( echnique
discussed, e.g. in [33]). The equipmen was calib a ed in he em-
pe a u e ange om oom empe a u e o 1400 °C agains pu e
me al s anda ds supplied by Ne zsch wi h he accu acy o be
wi hin ± 1 °C. The measu emen s we e pe o med a hea ing and
cooling a es o 10 K/min o which he bes signals s. hei se-
pa a ion we e ound. As a negligible in luence o Mn e apo a ion on
measu ed da a was ound and consis ency be ween DTA and DSC
da a was obse ed, he DSC echnique allowing measu emen up o
1400 °C was p e e ed. Sample specimens, o which mel ing was
an icipa ed abo e 1400 °C, we e measu ed in Al
2
O
3
and/o Z O
2
c ucibles co e ed by a hin sp ayed-on laye o Y
2
O
3
unde high
pu i y a gon (6 N) a a hea ing a e o 15 K/min in ei he a DTA Se-
a am SETSYS 18
TM
, a 3D DSC Se a am MHTC Line 96, o a DTA
NETZSCH STA 449 F3 Jupi e .
2.3. DFT s abili y calcula ions
The ab ini io calcula ions we e pe o med using he Vienna ab
ini io Simula ion Package (VASP) [34,35] wi hin he amewo k o
DFT. The pseudopo en ials we e cons uc ed acco ding o he p o-
jec o augmen ed wa e me hod [36,37], whe e he Pe dew, Bu ke
and E nze ho app oxima ion was used o ea he exchange-co -
ela ion e m [38,39]. The alence s a e con igu a ion o he con-
s uc ion o he pseudopo en ials included he 3d and 4s s a es o
Mn, 4p, 5s, 4d o Nb, 6s, 5d o Ta and 2s, 2p o O. In case o he
con igu a ions NbMn
2
, Nb
2
Mn, TaMn
2
, and Ta
2
Mn o he C14 La es
phase, he op imiza ion calcula ions we e made wi h he e imag-
ne ic (FIM) a angemen o magne ic momen s (he e, he magne ic
momen s o Nb and Ta ha e opposi e di ec ion and smalle alues
han hose o Mn), as his a angemen o magne ic momen s is
mo e s able in compa ison wi h he nonmagne ic (NM) a angemen
[40]. The cu -o ene gy es ic ing he numbe o plane wa es in he
basis se was 550 eV o TaMn
2
, Ta
2
Mn, Ta
3
Mn
3
, Ta
48
Mn
48
O
8
,
Ta
48
Mn
48
O and he co esponding pu e elemen s, i.e. body-cen e ed
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
3
e agonal Mn (bc , being equi alen o ace-cen e ed e agonal),
body-cen e ed cubic (bcc) Ta and he dimolecule o oxygen. Fo
NbMn
2
and Nb
2
Mn, bcc Nb and bc Mn (used o he e alua ion o
ene gies o o ma ion in he Nb-Mn sys em), a cu -o alue was se
a 650 eV. He e, he bc s uc u e co esponds o he bcc one wi h
he c la ice pa ame e sho e han a and b. The con e gence es s o
o al ene gies wi h espec o he numbe o k-poin s showed ha
he 21 × 21 × 21 k-poin g id in he i educible pa o he B illouin
zone is op imal o TaMn
2
and Ta
2
Mn, whils a g id 13 × 13 × 7 is
app op ia e o NbMn
2
and 23 × 23 × 13 o Nb
2
Mn. Fo Ta
3
Mn
3
,
Ta
48
Mn
48
O
8
and Ta
48
Mn
48
O, he k-poin g id was 6 × 6 × 6. Fo pu e
elemen s, he g id o k-poin s 31 × 31 × 31 was used o bcc Ta and
19 × 19 × 19 o bc Mn, bo h in he Ta-Mn sys em, and he g id
35 × 35 × 37 and 29 × 29 × 29 we e used o bc Mn and bcc Nb in he
Nb-Mn sys em. Fo oxygen, he k-poin g id was 8 × 8 × 8. This e-
sul ed in well-con e ged o al ene gies and equilib ium s uc u al
pa ame e s, i.e. la ice pa ame e s and in e nal a omic posi ions.
The ene gies o o ma ion we e calcula ed as he di e ence
be ween he DFT o al ene gy o he co esponding compounds and
he weigh ed a io o o al ene gies o pu e elemen s in hei s an-
da d elemen e e ence (SER) s a es, i.e. αMn, bcc Nb, bcc Ta and O
2
.
In o de o a oid he ab ini io calcula ion o he complex magne ic
s uc u e o αMn, we used he o al ene gy o he bc Mn s uc u e
wi h an an i e omagne ic a angemen (an ipa allel o ien a ion o
magne ic momen s) plus he o al ene gy di e ence be ween α and
bc Mn (1.82 kJ/mol), as calcula ed by Chen e al. [41], o e alua e he
o al ene gy o he SER s a e o αMn.
2.4. CALPHAD modeling
Fo he modynamic and phase diag am calcula ions as well as o
op imiza ion o he modynamic pa ame e s based on he CALPHAD
me hod, he Panda so wa e package [42] was used. The mo-
dynamic modeling o phases exis ing in he bina y sys ems Nb-Mn
and Ta-Mn elies on he well-known Compound Ene gy Fo malism
(CEF) enabling us o espec he eal c ys allog aphic s uc u e o a
phase by means o a subla ice desc ip ion [43]. Fo he he mo-
dynamic desc ip ion o elemen s i in he phase ϕ, he commonly
used polynomial o m was applied:
=
= + + + =
G (T) G (T) H (298.15 K)
A BT CT ln T D T ,
i
0
ii
SER0
n 2 nn
(2)
whe e A, B, C, D
n
and n ( ypically equal o 2, 3, and −1) a e
cons an s cha ac e is ic o he pa icula s uc u e o he elemen i
in a gi en empe a u e in e al in Kel in [21]. The Gibbs ene gy o a
gi en e na y phase ϕ is he e exp essed as a sum o se e al con-
ibu ions:
= + + +
=
G x H (298.15 K) G G G G ,
mi Mn,Nb,Ta ii
SER0
m
e mm
id
m
ex
(3)
whe e
Gm
e
is he e e ence le el o he mola Gibbs ene gy o a
gi en phase ϕ,
Gm
means Gibbs ene gy o o ma ion,
Gm
id
desc ibes
he mola Gibbs ene gy o ideal mixing o componen s (non-
s oichiome ic case) and
Gm
ex
is he mola excess Gibbs ene gy de-
sc ibing a non-ideal beha io o componen s due o hei mu ual
in e ac ions [44].
The Gibbs ene gy o o ma ion o he phase ϕ is exp essed by he
equa ion
=G H T S,
m
(4)
which is used o he desc ip ion o he Gibbs ene gies o com-
pounds. The alue o ΔH was based on DFT alues o ene gy o o -
ma ion o he phase and op imized on phase equilib ium da a,
simila ly as ΔS. The la ice s abili ies o me as able o uns able
phases a e calcula ed om i s -p inciples now; his idea was p e-
sen ed i s in 2001 in pape s [45,46]. Fo he desc ip ion o solu-
bili ies o componen s in phases, he subs i u ional model o he
Compound Ene gy Fo malism (CEF) Eq. (3) is used. The o mula o
ideal mixing:
=
=
G RT x ln(x )
m
id
k i,j k k
(5)
and he Redlich-Kis e polynomial o excess con ibu ion om
in e ac ion o componen s o he Gibbs ene gy:
=G RT L (x x)
m
ex
nnijn
(6)
a e used. In Eqs. (5) and (6) x
i
, x
j
, a e he mola ac ions o
componen s and n is he summa ion index o pa ame e s L, which
may be empe a u e dependen .
3. Resul s and discussion
3.1. C ys al s uc u e o he La es phases NbMn
2
and TaMn
2
Bo h La es phases NbMn
2
and TaMn
2
a e known o c ys allize
wi h he hexagonal MgZn
2
- ype s uc u e (C14- ype). A summa y o
all XPD da a a ailable in he li e a u e can be ound in he compi-
la ion by P. Villa s and K. Cenzual [47], howe e , ecen single c ys al
X- ay wo k on NbMn
2
and Nb
0.87
Mn
2.13
(Nb29Mn71 a %) by D.
G üne [48] was no included (de ails will be discussed below). So
a , no de ailed single c ys al s uc u e de e mina ion has been
published o TaMn
2
.
Fo he p esen wo k, small single c ys als we e b oken om a c
mel ed alloys wi h he s oichiome ic nominal composi ion {Nb,Ta}
Mn
2
. The X- ay in ensi y pa e ns in bo h cases we e ully indexed
and we e unambiguously consis en wi h hexagonal symme y wi h
space g oup P6
3
/mmc and la ice pa ame e s: a = 0.48898(1) nm,
c = 0.79970(1) nm o he Nb-c ys al and sligh ly smalle alues
a = 0.48708(1) nm and c = 0.79718(1) nm o he Ta-c ys al. S uc u e
solu ion by di ec me hods yielded a ully o de ed a om a ange-
men o MgZn
2
- ype o NbMn
2
, whe eas a mino bu signi ican
andom dis ibu ion o 0.96(1) Mn1 + 0.04 Ta1 in he 2a si e
was obse ed o Ta(Ta
x
Mn
1−x
)
2
(x = 0.01). The e inemen s wi h
aniso opic a om displacemen pa ame e s (ADPs) con e ged o
R
F2
= 0.0110 wi h esidual elec on densi ies smalle han ± 0.51
e
-
/10
−3
nm
3
o NbMn
2
(a RT) and R
F2
= 0.0244 wi h esidual elec on
densi ies < ± 2.40 e
-
/10
−3
nm
3
o Ta(Ta
x
Mn
1−x
)
2
; x = 0.01, espec-
i ely. C ys allog aphic da a o bo h single c ys als a e summa ized
in Tables 1 and 2. In e a omic dis ances a e shown in Fig. 1 o
Ta(Ta
x
Mn
1−x
)
2
wi h 0.2970 ≤ d
Ta-Ta
≤0.2990 nm, 0.2848 ≤ d
Ta-Mn
≤0.2858 nm, 0.2360 ≤ d
Mn-Mn
≤ 0.2510 nm and a e a he consis en
wi h he sum o CN12 me al a om adii (R
Nb
= 0.1468 nm,
R
Ta
= 0.1467 nm and R
Mn
= 0.1304 nm [49]). Whe eas dis ances Ta-Ta
a e sligh ly longe han he sum o adii, dis ances Mn-Mn o si e 6h
a e sho e by abou 10% and o Mn-Mn (2a o 6h) a e sho e by
abou 5% indica ing a s ong Mn-Mn in e ac ion. Such a beha io is
also inhe en o NbMn
2
. Ou oom empe a u e a om pa ame e s o
NbMn
2
a e essen ially consis en wi h he indings o G üne [48],
al hough his a om pa ame e s a e sligh ly highe and his la ice
pa ame e s a e somewha lowe .
E alua ion o he empe a u e dependen X- ay single c ys al
in ensi y da a se s in Table 2 clea ly documen ha NbMn
2
adop s
he MgZn
2
-La es ype a all he ou empe a u es measu ed,
namely: 100 K, 150 K, 200 K and 300 K. In combina ion wi h XPD
da a up o he mel ing poin we obse e ha no phase ansi ion
appea s wi hin he en i e empe a u e in e al (100–1788 K). TaMn
2
displays analogous beha io (298–2070 K).
The empe a u e dependence o he la ice pa ame e s o NbMn
2
o he empe a u e ange om 100 o 300 K, as depic ed in Fig. 2
( op), shows a a he smoo h a ia ion wi hin ~2 pe mille. Fig. 2
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
4
(middle) deno es he a ia ion wi h T o he a om pa ame e s in he
a ious Wycko si es o NbMn
2
. Al hough he z pa ame e o he
Nb-a oms in 4 seems o cons an ly ise wi h empe a u e, he
x pa ame e o he Mn-a oms in he 6h si e passes h ough a max-
imum a 150 K.
Fig. 2 (bo om) (100–300 K) po ai s he empe a u e depen-
dence o he in e a omic dis ances in NbMn
2
, which me ely e lec s
he a ia ion o la ice pa ame e s in combina ion wi h he a om
pa ame e a ia ion s. T. Consequen ly, all dis ances be ween Mn
a oms ise a he mono onically wi h empe a u e wi hin 2 pe mille.
Fo he e alua ion o he Debye empe a u e om he empe a u e
dependen ADPs see Sec ion 4.2. Speci ic hea .
In his con ex , i is in e es ing o no e ha D. G üne [47] ob-
se ed a weak 2a×2a×2c supe s uc u e o he MgZn
2
- ype on
polyc ys alline esidues, which we e ex ac ed by dilu ed HCl om
a c mel ed Mn- ich alloys Nb10Mn90 (in a %), sealed in Nb- essels
and annealed a 1100 °C. No s uc u al de ails we e epo ed, bu i
was said ha single c ys als Nb
0.87
Mn
2.13
(Nb29Mn71 a %), ex ac ed
in he same way om Mn- ich alloys, did no show any supe -
s uc u e e lec ions and clea ly e ealed iso ypism wi h he MgZn
2
-
ype. The andom dis ibu ion o (3.48Nb+0.52Mn) in he 4 -si es in
he Mn- ich single c ys al Nb
0.87
Mn
2.13
[48] is in line wi h andom
dis ibu ions o (0.1Nb+1.9Mn) in he 2a si es and (0.22Nb+5.78Mn)
in he 6h si es o he Nb- ich alloy Nb
1.08
Mn
1.92
(Rie eld e inemen
da a in Table 2). I should be men ioned he e, ha ou XPD nei he
in Nb-Mn no he Ta-Mn sys em de ec ed any supe s uc u e
e lec ions.
3.2. The oxygen-s abilized phases "NbMn" (^ Nb
3
Mn
3
O
1-y
) and "TaMn"
(^ Ta
3−x
Mn
3+x
O
1-y
)
Ou ecen a emp s o ge a phase-pu e compound NbMn
2
[24],
e ealed ha in Nb- ich samples small amoun s o a second com-
pound iche in Nb caused he appea ance o weak e omagne ism.
EDX measu emen s yielded a phase composi ion close o NbMn.
Fu he expe imen s ga e clea hin s ha he pe cen age o his phase
in Nb-Mn and also in Ta-Mn alloys inc eases wi h he non-me al
impu i y con en ( om WDX p edominan ly oxygen, no ni ogen, no
ca bon) in he aw elemen s (such as Mn o Nb/Ta-powde ). The e-
o e, he new seconda y phase was concluded o be an impu i y
s abilized e na y phase. Indeed, pains aking a emp s o educe he
oxygen-le el o he s a ing ingo ma e ials as well as in he p e-
pa a ion (mel ing unde high pu i y a gon, handling ma e ials ex-
clusi ely in a <3 ppm (O+H
2
O) glo e box sys em) we e inally
success ul yielding phase pu e and pa amagne ic La es phases NbMn
2
[24] and TaMn
2
( his wo k). La es phase samples wi hou any ace-
able amoun s o he magne ic phase ha e been ob ained wi h an
addi ion o 1–2 mass% o lan hanum me al as an in e nal oxygen
ge e ob iously cap u ing he oxygen and o ming a „nonmagne ic“
LaMn
x
O
y
phase, which did no dis u b he magne ic cha ac e iza ion
o he La es phases (see below Sec ion 4.1 Suscep ibili y).
In his pape we adop ed a “di y way” o inc ease he amoun o
his impu i y s abilized phase by mel ing powde compac s in he
egion Nb o NbMn (and Ta o TaMn) in an a gon a c-mel e wi h
in e mi en c ushing he bu ons in ai and adding Mn o com-
pensa e usion losses. Va ious oxygen con en s we e in oduced by
adding p ope amoun s o MnO
2
o Nb
2
O
5
and Ta
2
O
5
. In a ew cases,
we also ied o s a om ine powde compac s ha we e eac ed
in a sealed qua z ube o 3 × 36 h a 950 °C wi h in e mi en
c ushing in ai and e-compac ing. Al hough he new phase o med
in la ge quan i ies, we we e unable o syn hesize a single-phase
sample in bo h sys ems. The spec a usually con ained he La es
phase as he dominan phase and he “impu i y phase” in addi ion.
A his s age, a se ies o selec ed a ea elec on di ac ion (SAED)
pa e ns in Fig. 3 we e ob ained ia TEM om hin lamellae p epa ed
in SEM by FIB om he whi e phase in he cas alloy Ta50Mn50,
which was annealed o 2 mon hs a 1000 °C. Thei analyses
p omp ed a ace-cen e ed cubic ( cc) la ice wi h a la ice pa ame e
a ~ 1.12 nm (a e aged om a se o axes, [100], [110], [111], [210],
[310] and [510]). On he basis o his in o ma ion, he indexa ion o
he X- ay powde in ensi y pa e n o he new phases in bo h sys-
ems was success ul. A sea ch o he s uc u e ype in c ys al da-
abases such as in Pea son’s C ys al Da a [47] and in ICSD [50],
employing cc c ys al symme y wi h a ange o la ice pa ame e s
a = 1.13 ± 0.1 nm p omp ed di ec ly he CdNi- ype (o H Mn), which
in ac is an occupa ion a ian o he Ti
2
Ni- ype (e a(η)-phase).
The mode o a om si e occupa ion and illing he oc ahed al oids
in he a ious composi ion a ian s o he so-called η-phases de-
i ing om he pa en Ti
2
Ni- ype was al eady ex ensi ely discussed
by Rogl e al. [51]. Re e ing o his analysis, he phases Nb
~1
Mn
~1
O
y
and Ta
~1
Mn
~1
O
y
ha e o be classi ied as a illed me al hos la ice
s uc u e whe e he manganese a oms in 16c o space g oup Fd
3
m
(o igin a cen e o symme y) ill he cen e s o me al icosahed a
(each o med by six Nb(Ta)/Mn a oms om he si es 48 and 32e)
yielding a o mula Nb(Ta)
~3
Mn
~3
O
y
. Gene ally in e a-phases, non-
me al a oms such as C, N o O a e assumed o occupy he oc ahed al
oids in 16d, bu may also en e a second se o oc ahed al oids in
Wycko si e 8b (⅜,⅜,⅜). I may be no ed he e ha Wycko si e 8b
changes o si e 8a (⅛,⅛,⅛) in he non-s anda dized s uc u e (using
Fd
3
m wi h he o igin a −43m). We used his s anda dized s a ing
model o Rie eld e inemen s o elucida e he inal a om dis-
ibu ion in he no el compounds Nb
~1
Mn
~1
O
y
and Ta
~1
Mn
~1
O
y
. As
he si ua ion is qui e simila o bo h sys ems Nb(Ta)-Mn, we may
u he on ocus he e on he alloys om he Ta-Mn sys em.
Table 1
S uc u al da a o Ta(Ta
x
Mn
1−x
)
2
(x = 0.01) om X- ay single c ys al measu emen a
300 K. Pa ially o de ed MgZn
2
- ype; space g oup P6
3
/mmc; No. 194; s uc u e s an-
da dized wi h p og am S uc u e Tidy [28]. Aniso opic a omic displacemen pa a-
me e s U
ij
in [10
2
nm
2
].
Pa ame e /compound C ys al da a
Phase composi ion (EDX, a %) Ta35Mn65 ^Ta
1.05
Mn
1.95
Re inemen composi ion (a %) Ta
1.02
Mn
1.98
^Ta(Ta
x
Mn
1−x
)
2
;
x = 0.01
S uc u e ype MgZn
2
- ype
θ ange (deg) 5.1 ≤ 2θ≤ 72.5
C ys al size 65 × 70 × 80 µm
3
a = b (nm) 0.487077(7)
c (nm) 0.797175(11)
Re lec ions in e inemen 182 ≥ 4σ(F
o
) o 182
Numbe o a iables 12
Mosaici y <0.49
R
F2
= Σ|F
2
o
-F
2
c
|/ΣF
2
o
0.0244
wR2 0.0582
R
In
0.069
GOF 1.278
Ex inc ion (Zacha iasen) 0.0011(8)
M1 in 2a (0,0,0); occ. 0.96(1) Mn1 + 0.04 Ta1
U
11
= U
22
; U
33
; U
12
; U
23
= U
13
= 0 0.0029(8); 0.0028(9); 0.0015(4)
Ta2 in 4 (⅓, ⅔, z); occ. z = 0.56376(6); 1.00(1)
U
11
= U
22
; U
33
; U
12;
U
23
= U
13
= 0 0.0027(2); 0.0036(3); 0.0014(1)
Mn2 in 6h (x, 2x, ¼); occ. x = 0.1718(2); 1.00(1)
U
11
; U
22
; U
33
; U
12
; U
23
= U
13
= 0 0.0009(4); 0.0005(5);
0.0025(5); 0.0003(3)
Residual elec on densi y; max; min in
(elec on/nm
3
) × 10
3
2.40 (0.134 nm om Mn2);
−1.76
In e a omic dis ances < 0.3 nm (s anda d de ia ion <0.0004)
Ta – 3 Mn2 0.2848
Ta – 6 Mn2 0.2853
Ta – 3 Mn1 0.2858
Mn1 – 6 Mn2 0.2464
Mn1 – 6 Ta 0.2858
Mn2 – 2 Mn2 0.2360
Mn2 – 2 Mn1 0.2464
Mn2 – 2 Mn2 0.2510
Mn2 – 2 Ta 0.2848
Mn2 – 4 Ta 0.2853
Ta – 1 Ta 0.2970
Ta – 3Ta 0.2990
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
5

Rie eld e inemen o an a c mel ed alloy Ta43Mn57 (see
Fig. 4) clea ly showed ha he composi ion o he new phases
appea s close o a a io (Nb,Ta):Mn ~ 1:1 a he han o a a io
(Nb,Ta):Mn ~ 4:2 such as is ypical o many oxygen-s abilized
compounds de i ing om he Ti
2
Ni- ype (Ti
4
Ni
2
O) (e a(η)-phase;
o ep esen a i es see Pea son’s C ys al Da a [47]). Al hough he
hexagonal La es phase Ta
0.9
Mn
2.1
(MgZn
2
- ype) wi h 85 ol% is he
dominan phase in he X- ay powde di ac ion pa e n, he sec-
onda y phase o his pa e n can clea ly be indexed on an cc la ice
(a = 1.14546(1) nm), ully consis en wi h he s uc u e model
discussed abo e. The esul s o he e inemen a e lis ed in Table 3.
These esul s e eal only sligh de ia ions om a ull a om o de
wi h a andom a om dis ibu ion o (43.6 Ta+4.4 Mn) a oms in he
48 si e, (1.0 Ta+31.0 Mn) a oms in he 32e si e and exclusi ely Mn
in he 16c si e. In e es ingly he e inemen e uses any oxygen in
he Wycko si es 8a and 8b bu eadily accep s 5.34 O a oms in he
16d si e o he space g oup Fd
3
m (o igin a cen e ). F om his a om
dis ibu ion, we a i e a a chemical o mula Ta
2.78
Mn
3.22
O
0.33
(Ta
3−x
Mn
3+x
O
1−y
; x = 0.22, y = 0.67), which di ec ly complies wi h
he p o o- ype o W
3
Fe
3
C as one o he s uc u e a ian s o he
Ti
4
Ni
2
O amily [47,51]. The small de ia ions om ull s oichio-
me y (Ta
2.78
Mn
3.22
O
0.33
^ Ta43.9Mn50.9O5.2 in a %, i.e. a a a io
Ta:Mn = 46.3:53.7) a e well e lec ed in he EDX/WDX-composi-
ions de i ed by he X- ay mic oanalyses. The in ensi y simula-
ions o he SAED-pa e ns, calcula ed o he s uc u e model o
Ta
2.78
Mn
3.22
O
0.33
(plo ed in Fig. 3 in he JEMS so wa e [52,53])
a e ully consis en wi h he obse ed pa e ns. I should be
men ioned he e ha he appea ance o oxygen exclusi ely in
Wycko si e 16d ules ou he s uc u e ype o W
6
Fe
6
C (whe e he
nonme al a om only en e s he 8b si e bu lea es he 16d si e
emp y).
Table 2
S uc u al da a o NbMn
2
om X- ay single c ys al measu emen a ou di e en empe a u es. Pa ially o de ed MgZn
2
- ype; space g oup P6
3
/mmc; No. 194; o igin a cen e ;
s uc u e s anda dized wi h p og am S uc u e Tidy [28]. Aniso opic a omic displacemen pa ame e s U
ij
in [10
2
nm
2
]. In e a omic dis ances < 0.3 nm.
Pa ame e /Alloy composi ion NbMn
2
Tempe a u e Room empe a u e 200 K 150 K 100 K
a=b (nm); c (nm) 0.48898(1); 0.79970(1) 0.48834(1); 0.79853(2) 0.48824(1); 0.79823(1) 0.48801(1); 0.79798(1)
Da a collec ion, 2Θ ange 2 < 2Θ < 72.73 2 < 2Θ < 72.39 2 < 2Θ < 72.41 2 < 2Θ < 72.44
Re lec ions in e inemen 173 F
o
> 4σ(F
o
) o 184 172 F
o
> 4σ(F
o
) o 182 175 F
o
> 4σ(F
o
) o 181 170 F
o
> 4σ(F
o
) o 182
Mosaici y < 0.43 < 0.43 < 0.43 < 0.43
Numbe o a iables 11 11 11 11
R
F2
= Σ|F
02
-F
c2
|/ΣF
02
0.0110 0.0210 0.0158 0.0118
R
In
0.0265 0.0296 0.0299 0.0269
GOF 1.901 2.072 2.178 2.183
Ex inc ion (Zacha iasen) 0.014(1) 0.015(1) 0.015(1) 0.016(1)
Nb1 in 4 (1/3,2/3,z); occ. z = 0.56349(3); 1.00 Nb z = 0.56347(4); 1.00 Nb z = 0.56339(4); 1.00 Nb z = 0.56337(4); 1.00 Nb
U
11
= U
22
; U
33
;
U
12
0.0051(1); 0.0051(1);
0.0025(1)
0.0038(1); 0.0036(2);
0.0019(1)
0.0031(1); 0.0030(1);
0.0015(1)
0.0027(1); 0.0025(1);
0.0014(1)
Mn1 in 2a (0,0,0); occ. 1.00 Mn 1.00 Mn 1.00 Mn 1.00 Mn
U
11
= U
22
; U
33
;
U
12
0.0051(2); 0.0043(3);
0.0026(1)
0.0038(2); 0.0029(3);
0.0019(1)
0.0030(2); 0.0023(3);
0.0015(1)
0.0026(2); 0.0022(3);
0.0013(1)
Mn2 in 6h (x,2x,¼); occ. x = 0.17128(4); 1.00 Mn x = 0.17136(5); 1.00 Mn x = 0.17139(5); 1.00 Mn x = 0.17133(4); 1.00 Mn
U
11;
U
22
;
U
33
; U
12
0.0048(1); 0.0045(2);
0.0051(3); 0.0023(1)
0.0036(2); 0.0035(2);
0.0034(2); 0.0017(1)
0.0030(2); 0.0030(2);
0.0029(2); 0.0015(1)
0.0026(2); 0.0026(2);
0.0025(2); 0.0013(1)
Residual elec on densi y; max; min
in (e/nm
3
) x 1000
0.51; − 0.50; 0.097 nm
om Mn1
1.11; − 1.12; 0.135 nm
om Mn2
0.98; − 0.79; 0.149 nm
om Nb1
0.72; − 0.67; 0.160 nm
om Mn2
Nb1 – 3 Mn2 0.2858 0.2854 0.2852 0.2851
Nb1 – 6 Mn2 0.2864 0.2860 0.2860 0.2859
Nb1 – 3 Mn1 0.2868 0.2865 0.2864 0.2862
Mn1 – 6 Mn2 0.2470 0.2467 0.2466 0.2465
Mn1 – 6 Nb1 0.2868 0.2865 0.2864 0.2862
Mn2 – 2 Mn2 0.2377 0.2373 0.2372 0.2372
Mn2 – 2 Mn1 0.2470 0.2467 0.2466 0.2465
Mn2 – 2 Mn2 0.2513 0.2510 0.2510 0.2508
Mn2 – 2 Nb1 0.2858 0.2854 0.2852 0.2851
Mn2 – 4 Nb1 0.2864 0.2860 0.2860 0.2859
Nb1 – 2 Nb1 0.2983 0.2979 0.2979 0.2979
Nb1 – 2 Nb1 0.3000 0.2996 0.2995 0.2994
Fig. 1. C ys al s uc u e o TaMn
2
(Ta
1.02
Mn
1.98
) in h ee-dimensional iew wi h bond
dis ances (in Å). A oms a e displayed wi h hei aniso opic he mal displacemen
ellipsoids as de i ed om X- ay single c ys al e inemen (see Table 1). Mn1 a oms in
si e 2a a e yellow (0.96 Mn1 + 0.04 Ta1); Mn2 a oms in si e 6h a e oche ; Ta2 a oms in
si e 4 a e ed. (Fo in e p e a ion o he e e ences o colou in his igu e legend, he
eade is e e ed o he web e sion o his a icle.)
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
6
A e annealing a 950 °C o 4 days he XPD pa e n o he cas
alloy Ta43Mn57 has changed, i.e. besides he La es phase as he
dominan phase, we now encoun e wo o he cc e a-phases in he
SEM analysis as well as om Rie eld e inemen (see Table 3). In-
e es ingly he La es phase (~81 ol%) appea s p ac ically s oichio-
me ic bu wi h sligh ly lowe la ice pa ame e s. The seconda y
phases can bo h be desc ibed wi h he W
3
Fe
3
C- ype bu con ain
a he small con en s o oxygen: Ta
2.84
Mn
3.16
O
0.23
(14 ol%,
a = 1.14088(1) nm) and Ta
3.01
Mn
2.99
O
0.13
(4 ol%, a = 1.12859(2) nm).
Random dis ibu ions o Ta/Mn a oms only appea in he 48 si e and
a e o smalle ex en , he e o e he o mulae a e close o ull o de ;
he main di e ence is he oxygen con en e lec ed also by he di -
e en uni cell dimensions. I is unclea i he sample e eals a
ansi ion s a e o inal equilib ium o i he wo e a-phases a e
equilib ium phases o he Ta-Mn-O sys em.
The e omagne ic impu i y phase, which appea ed in small
quan i ies du ing he a emp s o p epa e single phase NbMn
2
was
equally well desc ibed by a W
3
Fe
3
C- ype e a-phase ( o de ails see
he Rie eld e inemen o alloy Nb40Mn60, annealed a 900 °C, in
Table 3 and Fig. S1 o Supplemen a y Ma e ial). Rie eld and EDX
analyses ag ee on a o mula Nb
2.86
Mn
3.14
O
0.31
(Nb
3−x
Mn
3+x
O
y
;
x = 0.14, y ~ 0.69) wi h a la ice pa ame e a = 1.13415(2) nm.
Finally, a sea ch o ep esen a i es o he W
3
Fe
3
C ype in
Pea son’s C ys al Da a [47] indeed e ealed he e a-phase Ta
3
Mn
3
O,
as epo ed by Schönbe g [54]. Al hough Schönbe g in es iga ed
he Ta-Mn-O sys em (<50 a % O), he was unable o ob ain he
e a-phase in a pu e condi ion bu a he “in he p esence o ai ly
la ge amoun s o o he phases”. Howe e , nei he he Ta-Mn-O
phase equilib ia no a s uc u e e inemen o he e a-phase ha e
been gi en; only he ange o la ice pa ame e s was lis ed om
1.115 o 1.118 nm [54], which, howe e , is much smalle han he
uni cells de i ed om ou samples. In e es ingly, Schönbe g’s
in es iga ion o he Ta-Mn-N sys em [55] did no yield any η-phase,
in con as o indings o Holleck e al. [56], who was able o syn-
hesize bo h e a-ni ides: Nb
3.5
Mn
2.5
N (a = 1.142 nm) and Ta
3
Mn
3
N
(a = 1.135 nm). He e i should be emphasized ha ou WDX es on N
in ou e a-phases was always nega i e.
The o ma ion o e a-phases in Ta-Mn alloys can be bes seen in
Fig. 5 om an HF mel ed alloy TaMn
2
, p epa ed om Ta- oil and
pieces o Mn. The e alua ion o he mic os uc u e by means o SEM-
EDX and WDX e ealed a pa ially mel ed Ta- oil su ounded by li-
quid Mn, which solidi ied in laye s o e a-phases and inally he
La es phase ( o de ails see Fig. 5).
3.3. Cons i u ion o he sys ems Nb-Mn and Ta-Mn
In o de o p o ide a ho ough cons i u ional basis o p ope
he modynamic modeling, we ha e s a ed wi h DTA/DSC in-
es iga ions. As he hi he o known in o ma ion sugges ed mel ing
empe a u es o he sec ions T - TMn
2
(T = Nb,Ta) abo e 1400 °C, we
applied high- empe a u e DTA wi h Al
2
O
3
and/o Y
2
O
3
-su ace
coa ed Z O
2
c ucibles om which we de ined he ollowing
cha ac e is ics: cong uen mel ing o NbMn
2
a T
m
(NbMn
2
) =
1515 ± 15 °C, incong uen mel ing o TaMn
2
a T
m
(TaMn
2
) =
1797 ± 40 °C and a p ac ically iden ical mel ing empe a u e o alloy
Ta40Mn60, whe eas alloy NbMn displayed la ge p ima y dend i es
o NbMn
2
and a eu ec ic ℓ = (Nb) + NbMn
2
a T
eu .
= 1493 ± 15 °C.
Some o he high- empe a u e DTA uns a e summa ized in Fig. S2
and Fig. S3 o Supplemen a y Ma e ial. This in o ma ion on he li-
quidus cu es, complemen ed by SEM mic og aphs and EDX/WDX
analyses (see Fig. 6 and Tables 4 and 5), con i med (i) cong uen
mel ing o NbMn
2
(S echniko e al. ga e T
m
= 1500 °C [13]) and (ii)
a a he deple ed pe i ec ic mel ing o TaMn
2
a 1797 °C, consis en
wi h (a) he no e o Sa i skii e al. ha TaMn
2
“mel s abo e 1670 °C”
[17] and (b) S echniko [13], who simply lis ed (wi hou commen )
he mel ing o TaMn
2
a 1800 °C. I should be emphasized, ha in
none o ou samples om bo h sys ems (Nb-Mn and Ta-Mn) we
obse ed any ace o he hombohed al μ-phase (W
6
Fe
7
- ype; space
g oup
R m3
; hex. axes: a ~ 0.47 nm, c ~ 2.58 nm). The e o e, and om
he discussion (i) in Sec ion 3.2. on he oxygen s abilized e a-phases
as well as (ii) on he DFT-ins abili y o e a-Ta
3
Mn
3
, (iii) he Rie eld
e inemen s o he alloys Nb40Mn60 (as-cas a e anneal a 900 °C)
and Ta43Mn57 (as-cas and a e anneal a 1000 °C), and (i ) he
Fig. 2. S uc u e pa ame e s o NbMn
2
in he empe a u e ange om 100 o 300 K
( om X- ay single c ys al da a, Table 2). Top: he mal expansion o he la ice pa a-
me e s; middle: a om pa ame e s o he a ious Wycko si es; bo om: in e a omic
dis ances. The dis ance d
Mn2–2Mn2
= 0.25130 nm ( owa ds he neighbo ing uni cell) a
300 K wi h almos no a ia ion is no shown he e.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
7
solidi ica ion mic os uc u es in Fig. 6 (Nb-Mn) and Fig. 6l (Ta-Mn),
we may sa ely exclude he phases “NbMn” [16] and “TaMn” [18] om
he co esponding bina y phase diag ams.
Whils he homogenei y egions o he wo La es phases will be
ea ed below, he maximum solubili y o Mn in (Nb) and (Ta), as
de i ed om he cas alloys and he alloys om DTA (see also he
mic os uc u es in Fig. 6 and Tables 4 and 5) we e 19.4 a % Mn in
(Nb) and e en 21.3 a % Mn in (Ta). In addi ion, i should be men-
ioned ha o he (Nb/Ta)- ich pa o he phase diag ams, DTA did
no e eal any ans o ma ions om RT o he mel ing.
Con en ional DTA (alloys sealed in qua z) and DSC (Al
2
O
3
-c u-
cibles unde a gon) we e chosen o elucida e he Mn- ich pa o he
phase diag ams, o which liquidus empe a u es we e all expec ed
o eside below 1250 °C. Whe eas he cas alloy Nb5Mn95 u ned ou
o be ully eu ec ic (T
eu
= 1220 ± 10 °C), he si ua ion is less
s aigh o wa d in he sys em Ta-Mn e ealing an almos deple ed
eu ec ic a abou 0.7 a % Ta ((δ-Mn) + TaMn
2
a T
eu
= 1234 ± 10 °C)
(see Fig. 6). The maximal solid solubili y o (Nb) and (Ta) in δ-Mn
appea s o be small and is abou 0.5 a % Mn o bo h Nb and Ta. In
con as o ha , Hellawell [14] (on he basis o mic os uc u es, TA
and XPD da a; bu no EPMA) epo ed he eu ec ic a 3.6 a % Nb
(1224 °C) and a maximum solubili y o 2.4 a % Nb in δ-Mn, a alue
which in iew o ou in es iga ions (see Fig. 6a and Table 4) seems o
be exagge a ed. In con as o ou indings, he da a o he Ta-Mn
sys em by Sa i skii [17] exclude δ-Mn om he eu ec ic, which was
claimed a 2.6 a % Ta a 1175 °C and o med by (γ-Mn) + TaMn
2
. Ou
DTA and DSC analyses in combina ion wi h SEM-EDX/WDX on a
se ies o alloys wi h 1, 2, 3, 4, 5 and 10 a % o he T-me al se ed o
de ine he phase ela ions in he Mn- ich pa o bo h sys ems Nb-
Mn and Ta-Mn, which a e cha ac e ized by he 4 c ys al s uc u e
modi ica ions o pu e Mn and hei phase ansi ions [12]:
° ° ° °
liquid Mn Mn Mn Mn.
C C C C1246 1138 1100 727
Al hough he highe ansi ion empe a u es a e alike, he lowe
wo ansi ion empe a u es in he compila ion om A. Dinsdale [21]
( he modynamic da a o una y phases) show a disc epancy as high
as abou 20 °C (707 °C and 1087 °C).
EXD and WDX da a in e ed ha o bo h phase diag ams he
solubili ies o he T-me al a e below ~0.8 a % T o all modi ica ions
excep o (γMn) and (αMn), he la e yielding a signi ican inc ease
o i s T-solubili y close o i s pe i ec oid decomposi ion on hea ing
(see Fig. 6 and Tables 4 and 5), hand in hand wi h a p onounced
inc ease o he (αMn) o (βMn) ans o ma ion empe a u e on
solu ion ( o de ails see he phase diag ams below in Figs. 8 and 9). I
should be men ioned he e, ha also Sa i skii e al. [17] om he mal
analyses concluded a pe i ec oid decomposi ion on hea ing o (αMn)
a ~1.5 a % Ta and 750 °C, qui e simila o ou DTA and EPMA da a.
DTA/DSC cu es o selec ed Mn- ich alloys a e p esen ed in Fig. S4
and Fig. S5 o Supplemen a y Ma e ial.
3.3.1. The homogenei y egions o he La es phases NbMn
2
and TaMn
2
The expe imen ally de i ed homogenei y egion o he La es
phase Nb
1+x
Mn
2−x
(62.5–73.0 a % Mn a 950 °C: −0.19 ≤x≤ 0.125) (see
Fig. 7a) is ai ly consis en wi h he epo s by D ys [57] (64–69 a %
Mn ^ −0.07 ≤x ≤ 0.08 a 800 °C), S echniko e al. [13] (62–70 a % Mn
a 1000 °C) and Blazina [58] (62–69 a % Mn a 1000°C), which was
Fig. 3. SAED pa e ns o he cubic “TaMn” e a(η)-phase a a ious sample il s (uppe panel) oge he wi h esul s o kinema ic simula ion (lowe panel). Weake e lec ions 00k
o k = 4n + 2 obse ed in he cen al ho izon al line o he [110] pa e n (compa e he SAED and i s simula ion below) a e caused by dynamic e ec s (double di ac ion).
Fig. 4. Rie eld e inemen o as-cas alloy Ta43Mn57. The excluded egion a ound
2θ≈ 22° belongs o a small peak s emming om he sample ca ie oil.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
8
also used by Gup a e al. [59] o he Nb-Mn-Ni sys em a 1000 °C.
Ou s udy o he la ice pa ame e s con i med he homogenei y e-
gion ob ained om EDX measu emen s wi hin a small ma gin o
± 0.5 a % Mn (see Fig. 7a). A simila ly wide homogenei y egion was
ound om XPD o Ta
1+x
Mn
2−x
(59.5–68.5 a % Mn o 900 °C and
cas alloys: −0.055 ≤x ≤ 0.215; see Fig. 7b), which is ai ly consis en
wi h he EPMA alues.
A his poin , i should be emphasized ha bo h sys ems Nb-Mn
and Ta-Mn a e cha ac e ized by only one bina y compound, a La es
phase wi h MgZn
2
- ype wi hou any o he s uc u e modi ica ions in
he en i e composi ion/ empe a u e ange o exis ence. Fo NbMn
2
his has been p o en by empe a u e dependen XPD (4.2–300 K), X-
ay SC di ac ome y om 100 o 300 K and by XPD up o he
mel ing ange on quenched alloys. Fig. 7a and b display he la ice
pa ame e s o Nb
1+x
Mn
2−x
and Ta
1+x
Mn
2−x
as a unc ion o compo-
si ion and o a se o empe a u es. Fo compa ison, all da a a ail-
able in he li e a u e a e plo ed ( aken om [47]), al hough many o
hem lack a de ined composi ion and can only be placed a s oi-
chiome ic TMn
2
. I is in e es ing o see ha he wid h o he
homogenei y egion Nb
1+x
Mn
2−x
is p ac ically in a ian om 950 °C
o he mel ing ange: − 0.19 ≤ x ≤ 0.125. Simila beha io is also en-
coun e ed o Ta
1+x
Mn
2−x
.
F om he ene ge ics o poin de ec o ma ion in a DFT supe cell
app oach o he la ge homogenei y egion o he Z Mn
2
La es
phase, Chen e al. [41] clea ly de i ed ha an isi e occupa ion is
mo e a o able han acancy o ma ion. Pa icula ly he low ene gy
o Mn a Z si es in e s a b oad nons oichiome y ange owa d he
Mn- ich side and wi h inc easing empe a u e he minimum o he
ee ene gy o he La es phase is shi ed owa d Mn- iche compo-
si ions (a 32.8 a % Z ) concomi an wi h he occu ence o he
cong uen mel ing poin no a he s oichiome ic bu a a sligh ly
Mn- iche composi ion [41]. Simila a gumen s may also hold o
bo h La es phases NbMn
2
and TaMn
2
.
3.4. DFT s abili y o in e me allic phases
The ab ini io calcula ed equilib ium la ice pa ame e s o he
FIM a angemen o he a ious con igu a ions o C14 La es phases
a e p o ided below. This ype o magne ic a angemen was chosen,
as i shows o be he mos s able (a de ailed s udy o his opic can be
ound in Re . [40]). The la ice pa ame e s o NbMn
2
(a = 0.48087 nm,
c = 0.79061 nm) and TaMn
2
(a = 0.47899 nm, c = 0.78874 nm) ag ee
Table 3
XPD Rie eld e inemen ( oom empe a u e da a) o alloys Nb40Mn60 (as-cas a e anneal a 900 °C) and Ta43Mn57 (as-cas and a e anneal a 1000 °C); he a ange
14 ≤ 2Θ≤ 95°); all s uc u es s anda dized wi h p og am S uc u e Tidy [28]. Iso opic a omic displacemen pa ame e s B
iso
in [10
2
nm
2
].
Pa ame e /sample (a %) Nb40Mn60; 900 °C Ta43Mn57; a c Ta43Mn57; 1000 °C
P o ile pa ame e s
Numbe o a iables 24 24 26
R
P
= Σ|y
oi
-y
ci
|/Σ|y
0i
| 0.0150 0.0348 0.0638
R
wP
= [Σw
i
|y
oi
-y
ci
|
2
/Σw
i
|y
oi
|
2
]
½
0.0198 0.0459 0.0871
R
e
= [(N-P + C)/Σw
i
y
2
oi
)]
½
0.0132 0.0085 0.0081
La es phase Nb
1.08
Mn
1.92
; 86 ol% Ta
0.9
Mn
2.1
; 85 ol% TaMn
2
; 81 ol%
Space g oup, P o o ype P6
3
/mmc, MgZn
2
P6
3
/mmc, MgZn
2
P6
3
/mmc, MgZn
2
a; c (nm) 0.489376(4); 0.800616(9) 0.491376(4); 0.803961(8) 0.48680(1); 0.79713(2)
R
F
= Σ|F
o
-F
c
|/ΣF
o
0.0654 0.0322 0.0435
R
I
= Σ|I
o
-I
c
|/ΣI
o
0.0826 0.0450 0.0464
Si e 2a (0,0,0); B
iso
0.36(2); 0.10(2) Nb + 1.90 Mn 0.20(3); 0.09(1) Ta + 1.91 Mn 0.21 2 Mn
Si e 4 (⅓,⅔,z); z; B
iso
0.56276(5); 0.49(3); 4 Nb 0.56381(2); 0.42(4); 3.51(1) Ta + 0.49 Mn 0.56483(3); 0.41(4); 4 Ta
Si e 6h (x,2x,¼); x; B
iso
0.1733(3); 0.57(4); 0.22(2) Nb + 5.78 Mn 0.1697(1); 0.21(4); 6 Mn 0.1778(2); 0.40(6); 6 Mn
η (e a) phase #1 Nb
2.86
Mn
3.14
O
0.31
; 14 ol% Ta
2.78
Mn
3.22
O
0.33
; 15 ol% Ta
2.84
Mn
3.16
O
0.23
; 14 ol%
Space g oup, P o o ype Fd
3
m, W
3
Fe
3
C Fd
3
m, W
3
Fe
3
C Fd
3
m, W
3
Fe
3
C
a (nm) 1.13415(2) 1.14546(1) 1.14088(3)
R
F
= Σ|F
o
-F
c
|/ΣF
o
0.096 0.0985 0.0980
R
I
= Σ|I
o
-I
c
|/ΣI
o
0.114 0.0995 0.1041
Si e 48 (x,⅛,⅛); x; B
iso
0.4308(2); 0.63; 42.6(3) Nb + 5.4 Mn 0.4325(1); 0.55; 43.6(1) Ta + 4.4 Mn 0.4288(2), 0.2; 45.4(1) Ta + 2.6 Mn
Si e 32e (x,x,x), x; B
iso
0.2064(3); 0.56; 3.1(2) Nb + 28.9 Mn 0.2120(3); 0.12; 1.0(1) Ta + 31.0 Mn 0.2038(3), 0.4; 32 Mn
Si e 16c (0,0,0) B
iso
16 Mn; 0.51 16 Mn; 0.12 16 Mn; 0.4
Si e 16d (½,½,½) B
iso
5.0(1) O; 0.61 5.34(8) O; 0.41 3.61(1) O; 0.55
η (e a) phase #2 no no Ta
3.01
Mn
2.99
O
0.13
; 4 ol%
Space g oup, P o o ype – – Fd
3
m, W
3
Fe
3
C
a (nm) – – 1.12859(2)
R
F
= Σ|F
o
-F
c
|/ΣF
o
– – 0.1020
R
I
= Σ|I
o
-I
c
|/ΣI
o
– – 0.1141
Si e 48 (x,⅛,⅛); x; B
iso
– – 0.4288(-); 0.2; 48 Ta
Si e 32e (x,x,x), x; B
iso
– – 0.2038(-); 0.4; 0.2 Ta + 31.8 Mn
Si e 16c (0,0,0) B
iso
– – 16 Mn; 0.4
Si e 16d (½,½,½) B
iso
– – 2.05(6) O; 0.55
Fig. 5. SEM-BSE mic og aph o HF mel ed alloy Ta39.8Mn60.2 (o e all EDX): b igh
Ta99.7Mn0.3, ligh g ay band adjacen o b igh (un eac ed) oil: Ta48Mn52
(Ta44.8Mn52.0O3.2 ≡ η-phase Ta
3−x
Mn
3+x
O
1−y
om WDX), da k g ay band second
nex o Ta- oil: Ta46Mn54 (second η-phase) and da k La es phase Ta40.6Mn59.4.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
9
E = 233.9 ± 16 GPa, espec i ely, e alua ed om he nanoinden e
measu emen s as well as wi h he DFT alue E
H
= 197 calcula ed o
NbMn
2
[32] ( o de ails see Table 7).
Fo he calcula ions o he bulk modulus, B, and he shea mod-
ulus, G, he ollowing equa ions we e applied:
= =
+
BE
3(1 2 ) , G E
2( 1) .
(8)
The esul s o TaMn
2
, B ~ 277 GPa, G ~ 86 GPa, a e sligh ly highe
han hose o NbMn
2
wi h B ~ 242 GPa, G ~ 75 GPa ( o de ails see
Table 7). This bulk modulus alue o NbMn
2
is ai ly consis en wi h
he alues ob ained om DFT calcula ions o a p e ious wo k o he
au ho s [24] wi h B ~ 250 GPa o he pa amagne ic and
B = 200–220 GPa o e omagne ic and an i e omagne ic NbMn
2
.
Acco ding o he c i e ion o Pugh [65] ha ma e ials wi h
<G/B 1.75
a e b i le, bo h compounds NbMn
2
and TaMn
2
wi h
=G/B 0.309
classi y as b i le.
Fo iso opic compounds, i is possible o de e mine he Debye
empe a u e, θ
D
om he sound eloci y, using Ande son’s
Eq. (9) [66]:
=
h
k
3nLd
4M ,
D
B
1/3
m
(9)
whe e h is he Plank’s cons an , k
B
is he Bol zmann’s cons an , L is
Loschmid ’s numbe , d is he densi y, M is he molecula weigh , and
n is he numbe o a oms. The mean sound eloci y
m
ollows om
ela ions (10) [66]:
= +
=
+
=
1
3
2
1
wi h 3B 4G
3d
and G
d
m
T
3L
3
1/3
L
1/2
T
1/2
(10)
Whe eas o NbMn
2
Debye empe a u es o θ
D
= 423 K (MI) and
θ
D
= 424 K (NI), we e calcula ed, we a i e o TaMn
2
a somewha
smalle alues o θ
D
= 378 K (MI) and θ
D
= 377 K (NI), espec i ely.
F ac u e oughness is a measu e o he esis ance o a ma e ial o
c ack p opaga ion. As can be seen in he inse o Fig. 17, c acks we e
p oduced om he imp in , om which we calcula e he inden a ion
ac u e oughness K
IC
using Eq. (11):
=KE
HV
F
c,
IC
1/2
3/2
(11)
whe e E is he Young’s modulus, HV is he ha dness, F is he in-
den a ion load, c is he adial c ack leng h om he cen e o he
Vicke s inden a ion and β is a unc ion o he inden e angle, which
o Vicke s inden a ion is β = 0.016(4). The inden a ion ac u e
oughness usually is subs i u ed o ac u e oughness in case o
b i le ma e ials, o o samples no big enough o quali y o he
s anda d me hods o de e mine ac u e oughness (like he Che on
No ched Flexu e Specimen me hod, he Single-Edge P ec acked
Beam me hod o he sha p ‘V′ no ch beam me hod). The esul s a e
Fig. 15. Tempe a u e dependen elec ical esis i i y, ρ(T), o NbMn
2
(Nb34Mn66) and
TaMn
2
(Ta33Mn67). Solid lines indica e en a i ely assigned esidual esis i i y plus
Bloch-G üneisen elec on-phonon sca e ing con ibu ions (see ex ).
Fig. 16. Es ima ed esis i i y con ibu ions, Δρ(T), due o elec on-elec on and elec-
on-pa amagnon sca e ing ob ained by sub ac ing o he con ibu ions as en a-
i ely indica ed by dashed lines in Fig. 15. Solid lines a e powe law i s o he low-
empe a u e beha io (see ex ).
Fig. 17. S a ic and dynamic (MI) ha dness, HV, o NbMn
2
and TaMn
2
s. load. The solid
lines a e guides o he eyes. The scale on he y-axis ( igh ) ans e s HV alues o GPa.
Inse : Imp in wi h c ack p opaga ion lines.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
16

almos alike o hose o measu ed ac u e oughness da a. E alu-
a ing many c ack leng hs o a ious loads om s a ic and dynamic
ha dness imp in s, yields a K
IC
= 1.1 ± 0.5 MPa m
1/2
o NbMn
2
and
K
IC
= 1.5 ± 0.2 MPa m
1/2
o TaMn
2
. F om σ ≈ E
1/2
a ough es ima ion o
he ac u e s eng h can be de i ed, which e eals o NbMn
2
σ ≈ 14 GPa
1/2
and o TaMn
2
σ ≈ 15 GPa
1/2
.
The mal expansion o NbMn
2
was measu ed in he empe a u e
ange o 4.2–300 K (Fig. 19). The co esponding
T( )
0
s. T cu e
shows a s ic ly linea beha io abo e 150 K. Consequen ly, a linea i
in he empe a u e ange o 150–300 K e eals a he mal expansion
coe icien α = 10.9 × 10
−6
Κ
−1
. Besides ex ac ing he he mal expan-
sion coe icien , he semi-classical ea men by Mukhe jee e al. [67]
was applied o analyze he he mal expansion as a unc ion o em-
pe a u e, aking in o accoun he Debye model o acous ic phonons
and he Eins ein app oxima ion o he op ical modes.
The leng h change
T( )
0
is gi en by:
= = +
= +
( ) ( ) ( )
, x T [ G F ]
3k T ,
T x( )
x x
T2
23g
4c
2 3
3
pBT3
0
z dz
e 1
p 3
p
k
e 1
T T
0
0
02
D
D
T3
zB E
D/T
(12)
whe e δ is he elec onic con ibu ion o he a e age la ice dis-
placemen , θ
D
is he Debye empe a u e, θ
E
is he Eins ein em-
pe a u e, and p is he a e age numbe o phonon b anches ac ually
exci ed o e he empe a u e ange. G, F, c, and g a e u he ma e ial
dependen cons an s. The Debye and Eins ein empe a u e, ex-
ac ed ia a leas -squa es i o he expe imen al da a we e
θ
D
= 413 K and θ
E
= 87 K, wi h θ
D
in e y good ag eemen wi h he
da a gained ia elas ic moduli and elec ical esis i i y.
5. Summa y/conclusion
Based on a de ailed e-in es iga ion, he cons i u ion o he wo
phase diag ams Nb-Mn and Ta-Mn has been es ablished om 600 °C
o he mel ing ange: he sys ems a e cha ac e ized by MgZn
2
- ype
La es phases (C14), which a e he only bina y compounds in
hese sys ems bu e eal ex ended homogenei y egions (each in-
cluding he s oichiome ic composi ion): Nb
1+x
Mn
2−x
(62.5–73.0 a %
Mn a 950 °C: −0.19 ≤ x ≤ 0.125) and Ta
1+x
Mn
2−x
(59.5–68.5 a %
Table 7
Elas ic cons an s c
ij
o NbMn
2
(all in GPa; densi y d = 8.15 g/cm
3
) and o TaMn
2
(d = 11.86 g/cm
3
; c
66
= (c
11
-c
12
)/2); shea modulus G; bulk modulus B; Young’s modulus
E; subsc ip s V, R, H e e o Voig , Reuss limi s and Hill a e age; Poisson’s a io ν; mean sound eloci y
m
in ms
−1
; Debye empe a u e θ
D
in K; e e ences ( .w. = his wo k);
alues in cu si e sc ip we e comple ed in his wo k om he C
ij
ma ix o he au ho [24,32].
c
11
E
(010)
c
12
c
13
c
33
E
(001)
c
44
G
V
G
R
G
H
B
V
B
R
B
H
E
V
E
R
E
H
ν
m
θ
D
Re .
NbMn
2a
346 218 180 384 68 73 71 72 248 248 248 200 194 197 0.36 3214 410 [24,32]
341# 75# 75# 242# 203# 3407# 423# .w.
TaMn
2
391# 86# 86# 277# 233# 0.36 3025# 378# .w.
a
Fo pa amagne ic C14- ype; # da a ex ac ed om mic oinden a ion and nanoinden a ion measu emen s.
Table 8
P ope ies (s a ic ha dness (HV
0.1
(s a .) in GPa), dynamic ha dness ia mic oinden e
(HV
0.1
(MI) in GPa), dynamic ha dness ia nanoinden e (HV
0.01
(NI) in GPa), Bulk
modulus B (DFT) in GPa, Debye empe a u e calcula ed wi h E (MI) (θ
D
(MI)) in K,
ac u e esis ance, (K
IC
) in MPam
1/2
, Debye empe a u e calcula ed wi h E (NI) (θ
D
(MI)) in K, Debye empe a u e ia Mukhe jee- i (θ
D
(MUK)) in K, Eins ein empe a u e
ia Mukhe jee- i (θ
E
) in K, he mal expansion coe icien (α)*10
−6
in K
−1
.
P ope ies NbMn
2
TaMn
2
Re .
HV
0. 1
(s a .) 823 (9.2) 900 (10.1) w.
HV
0.1
(MI) 735 (7.5) 815 (7.9) w.
HV
0.01
(NI) 724.7 (7.1) 815.9 (8.0) w.
HV – 730 [17]
HV 1040–1250 1020 [13]
B (DFT), pa amagne ic 250 – [24]
B (DFT), e omagne ic 200–220 – [24]
B (DFT), an i e omagne ic 220 – [24]
θ
D
(C
p
) ~430 ~370 w.
θ
D
(U
ij
) 397 – w.
θ
D
( he m. exp.) 413 – w.
θ
E
( he m. exp.) 87 – w.
θ
E
(C
p
) ~166 ~135 w.
α 10.9 – w.
K
IC
1.1 1.5 w.
Fig. 18. Elas ic modulus, E, s. load o NbMn
2
and TaMn
2
. The solid lines a e guides
o he eye.
Fig. 19. Tempe a u e dependen dila ome ic he mal expansion o NbMn
2
. S a s e-
p esen SC X- ay da a om Table 2. Fo he i see ex .
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
17
Mn: −0.055 ≤ x ≤ 0.215). Abou 20 samples ha e been p epa ed in
each sys em, mainly by a c- o high equency mel ing elemen al
ingo s, in some cases also cold compac ed powde blends. The in-
es iga ion o he cons i u ion comp ised ligh op ical and ans-
mission and scanning elec on mic oscopy (TEM and SEM) wi h
ene gy dispe si e (EDX) as well as wa eleng h dispe si e (WDX) X-
ay spec oscopy, X- ay powde (XPD) and single c ys al (XSCD)
di ac ion, di e en ial he mal analysis (DTA) and/o di e en ial
scanning calo ime y (DSC).
Whe eas NbMn
2
mel s cong uen ly a T
m
(NbMn
2
) = 1515 ± 15 °C,
TaMn
2
mel s incong uen ly wi h T
m
(TaMn
2
) = 1797 ± 40 °C close o a
deple ed pe i ec ic eac ion. Bo h La es phases engage in eu ec ic
eac ions wi h manganese: ℓ ↔ (Mn) + TMn
2
(T
eu
= 1220 ± 10 °C a
4.9 a % Nb and T
eu
= 1234 ± 10 °C a 0.7 a % Ta, espec i ely).
NbMn
2
also o ms a eu ec ic wi h (Nb): ℓ ↔ (Nb) + NbMn
2
a
T
eu
= 1493 ± 15 °C a 53.2 a % Nb. In bo h sys ems, Mn e eals e-
ma kably la ge maximum solid solubili ies (a he eac ion iso-
he ms 1493 °C (Nb) and 1797 °C o (Ta)): 19.4 a % Mn in (Nb) and
21.3 a % Mn in (Ta).
De ailed in es iga ion (SEM-WDX, TEM, Rie eld-XPD) on he
s abili y and s uc u e o he phases, “NbMn” and “TaMn”, adop ed
ea lie in he li e a u e as bina y sys em inhe en compounds,
clea ly e ealed hese phases o be oxygen-s abilized adop ing he
Ti
4
Ni
2
O ype (so-called e a(η)-phases) wi h modi ied Nb(Ta)/Mn si e
subs i u ion o comply wi h he o mula Nb(Ta)
3−x
Mn
3+y
O
1−y
(W
3
Fe
3
C- ype). We also emphasize ha no hin s we e ound in ou
analyses o he exis ence o a μ-phase Nb
6
Mn
7
o Ta
6
Mn
7
(W
6
Fe
7
-
ype; space g oup R
3
m). Magne ic suscep ibili y and magne iza ion
measu emen s documen ed ha bo h e a-phases η-Nb
3
Mn
3
O
1−y
and
η-Ta
3
Mn
3
O
1−y
a e e omagne ic below T
c
~ 77 K, whe eas he La es
phases NbMn
2
, TaMn
2
a e empe a u e independen pa amagne ic.
Fu he mo e, ou DFT calcula ions de ine ins abili y o he e a-phase
Ta
3
Mn
3
wi hin he bina y sys em (wi h espec o he weigh ed a io
o he C14 La es phase TaMn
2
and pu e bcc Ta). Addi ionally, he ab
ini io calcula ions con i med (i) he s abilizing e ec o oxygen on
he e a-phase and (ii) he p e e ence o oxygen a oms o subla ices
in he sequence 16d (W
3
Fe
3
C- ype)> 8b> 8a (W
6
Fe
6
C- ype) in space
g oup Fd
3
m.
The expe imen ally de i ed cons i u ion o he Nb-Mn and Ta-
Mn sys ems se ed as he basis o CALPHAD calcula ions esul ing
in a comple e se o op imized he modynamic da a o each
sys em. The CALPHAD op imiza ion was suppo ed by DFT calcu-
la ed hea s o o ma ion o a ious con igu a ions o he C14 La es
phases.
Tempe a u e dependen X- ay single c ys al da a (100–300 K)
o NbMn
2
as well as oom empe a u e SC da a o Ta
1.05
Mn
1.95
p o ided de ails on a om si e dis ibu ion and he mal expansion
(CTE-NbMn
2
= 10.9 × 10
−6
K
−1
). The modynamic and anspo
p ope ies (speci ic hea , elec ical esis i i y and magne ic sus-
cep ibili y/magne iza ion, om 2 o 300 K) classi y bo h La es
phases wi h me allic beha io whils mechanical p ope ies (elas ic
moduli om DFT and nanoinden a ion as well as ha dness and
he mal expansion) g oup bo h La es phases among a he ha d
(HV-NbMn
2
= 800, HV-TaMn
2
= 900) and b i le in e me allics
(K
IC
= 1.1 ± 0.5 MPa m
1/2
o NbMn
2
and K
IC
= 1.5 ± 0.2 MPa m
1/2
o
TaMn
2
).
CRediT au ho ship con ibu ion s a emen
All au ho s ha e con ibu ed equally o he wo k.
Decla a ion o Compe ing In e es
The au ho s decla e ha hey ha e no known compe ing i-
nancial in e es s o pe sonal ela ionships ha could ha e appea ed
o in luence he wo k epo ed in his pape .
Acknowledgemen
This esea ch was suppo ed by he Czech Science Founda ion
unde p ojec s GA 17-12844S and CZ.02.1.01/0.0/0.0/17_049/
0008399. Thanks a e also due o he OEAD o suppo ia he bi-
la e al WTZ CZ02 and o he MEYS CR ia he bila e al Mobili y
8J19AT011. Compu a ional esou ces we e supplied by he p ojec
"e-In as uc u e CZ" (e-INFRA CZ LM2018140) p o ided wi hin he
p og am “La ge In as uc u es o Resea ch, Expe imen al
De elopmen and Inno a ions“ suppo ed by he Minis y o
Educa ion, You h and Spo s o he Czech Republic.
Appendix A. Suppo ing in o ma ion
Supplemen a y da a associa ed wi h his a icle can be ound in
he online e sion a doi:10.1016/j.jallcom.2021.158715.
Re e ences
[1] C. Gue a, C. Aguila , D. Guzman, M. A ancibia, P.A. Rojas, S. Lascano, L. Pe ez,
P oduc ion and cha ac e isa ion o mechanical p ope ies o Ti-Nb-Ta-Mn alloys
oams o biomedical applica ions, Powde Me all. 58 (1) (2015) 12–15.
[2] C. Gue a, M. Sancy, M. Walczak, D. Sil a, C. Ma inez, B. T ibolle , C. Aguila ,
E olu ion o oxide ilm on he in e nal po osi y o Ti-30Nb-13Ta-2Mn alloy oam,
Elec ochim. Ac a 283 (2018) 676–682.
[3] Y. Sun, W.W. Zhang, H.Q. Li, P epa a ion and p ope ies o niobium-manganese
composi e elec ode ma e ial, Diandu Yu Tushi 35 (1) (2016) 28–32.
[4] Y.S. Ko, H. Pa k, S.C. Jung, S.D. Wang, De elopmen o high s eng h ma ensi ic
mic oalloyed s eel o hea y-du y engine connec ing ods, in: J.C. Zhao, M.
Fah mann, T. Pollock (Eds.), P oeedings o Symposium on Ma e ials Design
App oaches and Expe iences, Indianapolis, IN, US, No embe 4–8 (2001)
151–16.
[5] P. Palze , M. O o, T. E e z, P oduc ion o seamless medium manganese s eel
ubes, PCT In e na ional Applica ion (2018), WO 2018083028 A1, May 11, 2018.
[6] M. Kawamo i, J. Kinugasa, Y. Yonenaga, Y. Fuku a, M. Shimamo o, T. Sugimu a,
T. Sa o, N. Nishizawa, M. Nagao, E ec o an alum addi ion on pi ing co osion
beha io o supe duplex s ainless s eels - de elopmen o high co osion e-
sis an duplex s ainless s eels by inclusion con ol, Zai . Kankyo 64 (7) (2015)
320–323.
[7] H. Ha ano, E ec s o Nb and Mn on mic os uc u e and oughness o simula ed
HAZ o 590 MPa class low ca bon baini ic s eels, Te su Hagane 91 (12) (2005)
875–881.
[8] E. Eks oem, A. Le Feb ie , F. Bou geois, B. Lundq is , J. Palisai is, P.O.Å. Pe sson,
O. Caballe o-Cale o, M.S. Ma in-Gonzalez, J. Kla b ing, S.I. Simak, F. E iksson,
B. Paul, P. Eklund, The e ec s o mic os uc u e, Nb con en and seconda y
Ruddlesden-Poppe phase on he moelec ic p ope ies in pe o ski e
CaMn
1-x
Nb
x
O
3
(x = 0 - 0.10) hin ilms, RSC Ad . 10 (13) (2020) 7918–7926.
[9] M. Nabi, T.M. Bha , D.C. Gup a, E ec o p essu e on elec onic, magne ic, he -
modynamic, and he moelec ic p ope ies o an alum-based double pe -
o ski es Ba
2
MTaO
6
(M = Mn, C ), In . J. Ene gy Res. 43 (9) (2019) 4229–4242.
[10] B. Hamad, Ab ini io in es iga ions o he s uc u al, elec onic, and he mo-
elec ic p ope ies o Fe
2
NbAl-based alloys, J. Ma e . Sci. 51 (2016) 10887–10896.
[11] N. Pa se, A. A anusilp, W. Silpawilawan, K. Ku osaki, S. Pini soon o n, Enhancing
he moelec ic p ope ies o highe manganese silicide (HMS) by pa ial Ta
subs i u ion, J. Elec on. Ma e . 49 (5) (2020) 2726–2733.
[12] T. Massalski, Bina y Alloy Phase Diag ams (second ed., ASM In e na ional,
Ma e ials Pa k, Ohio, USA. ol. 3, 1990, p. 2578; see also upda e by H. Okamo o,
Supplemen al li e a u e e iew o bina y phase diag ams: Ag-Sn, Al-Pd, Ba-Gd,
Ba-P , Cu-P, Dy-Ni, Ga-Mn, Gd-Sb, Gd-Z , Ho-Te, Lu-Sb, and Mn-Nb, J. Phase
Equilib. Di us. 35 (1) (2014) 105–116.
[13] V.N. S echniko , V.V. Pe 'ko , Fo ma ion o La es phases in alloys o manganese
wi h ansi ion me als o g oups IV-Va (in Russian) Me allophysika, Akad. Nauk
Uk . SSR Ins . Me allo iz. 64 (1976) 24–28.
[14] A. Hellawell, The cons i u ion manganese base alloys wi h me als o he second
ansi ion se ies, J. Less Common Me . 1 (1959) 343–347.
[15] B. Kuz'ma Yu, Phase equilib ia in e na y ha nium-cobal -bo on, ha nium-nickel-
bo on, and niobium-manganese-bo on sys ems, Visnik L' i s'kogo Uni . Se iya
Khimichna 14 (1972) 20–22 (in Uk ainian).
[16] S. Liu, B. Halls ed , D. Music, Y. Du, Ab ini io calcula ions and he modynamic
modeling o he Fe–Mn–Nb sys em, Calphad 38 (2012) 43–58.
[17] E.M. Sa i skii, C.V. Kope zkii, Phase diag am o he manganese- an alum sys em,
Russ. J. Ino g. Chem. 5 (11) (1960) 1274–1275 ansla ed om Zh. Neo g. Khimii
64 (1960) 363-364.
[18] L. Kau man, Coupled he mochemical and phase diag am da a o an alum
based bina y alloys, Calphad 15 (1991) 243–259.
[19] F.R. DeBoe , R. Boom, A.R. Miedema, En halpies o o ma ion o liquid and solid
bina y alloys based on 3d me als II. Alloys o ch omium and manganese, Physica
B+C l13 (1982) 18–41.
[20] C. Coline , A. Pas u el, P. Hic e , T ends in cohesi e ene gy o ansi ion me al
alloys, Calphad 9 (1985) 71–99.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
18
[21] A.T. Dinsdale, SGTE da a o pu e elemen s, Calphad 15 (1991) 317–425. O iginal
da a o Mn om A. Fe nandez Guille me , W. Huang, In . J. The mophys.
11 (1990) 949–969.
[22] C. Wang, C. Zhao, Z. Lin, X. Liu, Expe imen al de e mina ion and he modynamic
calcula ion o he phase equilib ia in he Co-Mn-Ta sys em, In . J. Ma e . Res. 105
(12) (2014) 1179–1190.
[23] S.V. Meschel, P. Nash, X.Q. Chen, The s anda d en halpies o o ma ion o bina y
in e me allic compounds o some la e 4d and 5d ansi ion me als by high
empe a u e di ec syn hesis calo ime y, J. Alloy. Compd. 492 (12) (2010)
105–115.
[24] X. Yan, X.Q. Chen, H. Micho , W. Wol , V.T. Wi usiewicz, E. Baue , R. Podloucky,
P. Rogl, S uc u al, he modynamic, and elec onic p ope ies o La es-phase
NbMn
2
om i s p inciples, x- ay di ac ion, and calo ime ic expe imen s,
Phys. Re . B 97 (2018) 1–10 125110.
[25] J. Rod iguez-Ca ajal, Recen de elopmen s o he p og am FULLPROF, Physica B
55 (1993) 192 and T. Roisnel, J. Rod iguez-Ca ajal, Ma e . Sci. Fo um 118 (2001)
378–381.
[26] Nonius Kappa CCD, P og am Package COLLECT, DENZO, SCALEPACK, SORTAV,
Nonius Del , The Ne he lands.
[27] G.M. Sheld ick, SHELX – P og am o c ys al s uc u e e inemen , Ac a
C ys allog . A 64 (2008) 112–122; Windows e sion OSCAIL by P. McA dle, K.
Gilligan, D. Cunningham, R. Da k, and M. Mahon, Na l. Uni . I eland, Galway:
C ys . Eng. Comm. 6 (2004) 303–309. Windows e sion WINGX by L.J. Fa ugia,
J. Appl. C ys allog . 32 (1999) 837–838.
[28] L.M. Gela o, E. Pa hé, STRUCTURE TIDY-a compu e p og am o s anda dize
c ys al s uc u e da a, J. Appl. C ys allog . 20 (1987) 139–143; see also P og am
Typix, Gmelin Handbook o Ino ganic and O ganome allic Chemis y,
ol. 1, Sp inge , N.Y., 1993, p. 186.
[29] INCA Ene gy-300 and INCA Wa e 700, Ox o d Ins umen s Analy ical L d., UK,
2000.
[30] G. Binde , S udies o Spin-glass Ma e ials wi h AC Suscep ibili y Measu emen s
(in Ge man) (Mas e hesis), TU-Wien, Aus ia, 2019.
[31] M. Ro e , H. Mülle , E. G a z, M. Doe , M. Löewenhaup , A minia u e capaci-
ance dila ome e o he mal expansion and magne os ic ion, Re . Sci. Ins um.
69 (7) (1998) 2742–2746.
[32] X.Q. Chen, Ab-ini io S udies o La es Phase In e me allic Compounds (Ph.D.
hesis), Uni e si y o Vienna, Aus ia, 2003.
[33] P. B ož, F. Zelenka, J. Sopoušek, M. Hejduko á, T. Kuběno á, O. Zobač, The mal
analysis and Knudsen e usion mass spec ome y combined in a specially-
adap ed comme cial skimme coupled ins umen (Ne zsch), Calphad 65 (2019)
86–92.
[34] G. K esse, J. Fu hmülle , E icien i e a i e schemes o ab ini io o al-ene gy
calcula ions using a plane-wa e basis se , Phys. Re . B 54 (1996) 11169–11172.
[35] G. K esse, D. Joube , F om ul aso pseudopo en ials o he p ojec o aug-
men ed-wa e me hod, Phys. Re . B 59 (1999) 1758–1775.
[36] P.E. Blöchl, P ojec o augmen ed-wa e me hod, Phys. Re . B Condens. Ma e
50 (20) (1994) 17953–17979.
[37] P.E. Blöchl, J. Käs ne , C.L. Fö s , Chap e 1: Elec onic s uc u e me hod: aug-
men ed wa es, pseudopo en ials and he p ojec o augmen ed wa e me hod, in:
Handbook o Ma e ials Modeling, Sp inge , The Ne he lands, 2005, pp. 93–119.
[38] J.P. Pe dew, K. Bu ke, M. E nze ho , Gene alized g adien app oxima ion made
simple, Phys. Re . Le . 77 (18) (1996) 3865–3868.
[39] J.P. Pe dew, M. E nze ho , K. Bu ke, Ra ionale o mixing exac exchange wi h
densi y unc ional app oxima ions, J. Chem. Phys. 105 (22) (1996) 9982–9985.
[40] S. Ko aříko á Oweis, M. Mazalo á, J. Pa lů, P. Rogl, Ab ini io s udy o Mn-based
La es and e a phases, o be published.
[41] X.Q. Chen, W. Wol , R. Podloucky, P. Rogl, M. Ma sman, Ab ini io s udy o g ound-
s a e p ope ies o he La es-phase compound Z Mn
2
, Phys. Re . B 72 (2005)
1–11 054440.
[42] Panda so wa e, e sion 2014 2.0, CompuThe m LLC, Madison, Wisconsin, USA,
2014, 〈h p://www.compu he m.com/〉.
[43] H.L. Lukas, S.G. F ies, B. Sundman, Compu a ional The modynamics, Camb idge
Uni e si y P ess, Camb idge, 2007.
[44] H.S. Liu, J. Wang, Z.P. Jin, The modynamic op imiza ion o he Ni–Sn bina y
sys em, Calphad 28 (2008) 363–370.
[45] J. Vřešťál, Recen p og ess in modelling o sigma-phase, A ch. Me all. 46 (3)
(2001) 239–247.
[46] J. Ha ánko á, J. Vřešťál, L.G. Wang, M. Šob, Ab ini io analysis o ene ge ics o
σ-phase o ma ion in C -based sys ems, Phys. Re . B 63 (2001) 174104.
[47] P. Villa s, K. Censual, Pea son’s C ys al Da a: C ys al S uc u e Da abase o
Ino ganic Compounds, ASM In e na ional, Ma e ials Pa k, Ohio, USA, Release,
2019/20.
[48] D. G üne , In es iga ions on he Na u e o La es Phases in T ansi ion Me al
Sys ems (in Ge man) (Ph.D. hesis), TUD esden, Ge many, 2007.
[49] W.B. Pea son, The C ys al Chemis y and Physics o Me als and Alloys, Wiley-
In e science, New Yo k, USA, 1972. The 12-coo dina ed me allic adii in Tables
4–4 o his book ha e been aken om: E. Tea um, K. Gschneidne , and J. Wabe ,
Repo LA-2345, US Depa men o Comme ce, Washing on, D.C., USA, 1960.
[50] ICS da abase, 2019/20: Fachin o ma ionszen um Ka ls uhe, Ge many.
[51] P. Rogl, H. Nowo ny, The mode o illing he oids in η-phases, Mona she e
Chem. 108 (1977) 1167–1180 (in Ge man).
[52] P.A. S adelmann, JEMS: Ja a Elec on Mic oscopy So wa e, Swi ze land, 2020;
〈h p://www.jems-swiss.ch〉.
[53] P.A. S adelmann, EMS - a so wa e package o elec on di ac ion analysis and
HREM image simula ion in ma e ials science, Ul amic oscopy 21 (1987)
131–145.
[54] N. Schönbe g, Me allic e na y phases in he Mn-Ta-O sys em, Ac a Me all.
3 (1955) 14–16.
[55] N. Schönbe g, An X- ay in es iga ion on e na y phases in he Ta-Me-N sys ems
(Me = Ti, C , Mn, Fe, Co, Ni), Ac a Chem. Scand. 8 (1954) 213–220.
[56] H. Holleck, F. Thümmle , Te na y complex-ca bides, -ni ides, and -oxides wi h
pa ially illed Ti
2
Ni ype s uc u e, Mona she e Chem. 98 (1967) 133–134
(in Ge man).
[57] M. D ys, Phase equilib ia in he niobium-gallium-manganese sys em a 900 °C,
J. Less Common Me . 75 (1980) 261–265.
[58] Z. Blazina, R. T ojko, S uc u al in es iga ions o he Nb
1-x
Si
x
T
2
and Nb
1-x
Al
x
T
2
(T = C , Mn, Fe, Co, Ni) sys ems, J. Less Common Me . 119 (1986) 297–305.
[59] P. Gup a, S.K. Singh, The manganese-niobium-nickel sys em, J. Phase Equilib.
16 (1995) 320–323.
[60] A.U. Khan, P. B ož, M. P emo ić, J. Pa lů, J. Vřešťál, X. Yan, D. Macciò, A. Saccone,
G. Gies e , P. Rogl, The Ti–Mn sys em e isi ed: expe imen al in es iga ion and
he modynamic modelling, Phys. Chem. Chem. Phys. 18 (2016) 23326–23339.
[61] J. Pa lů, J. Vřešťál, M. Šob, Re-modeling o La es phases in he C -Nb and C -Ta
sys ems using i s -p inciples esul s, Calphad 33 (2009) 179–186.
[62] S. Khmele skyi, p i a e communica ion, TU-Wien, Aus ia, 2020.
[63] A. Junod, D. Bichsel, J. Muelle , Modi ica ion o acous ic phonon spec a as a
unc ion o elec on densi y o s a es in he supe conduc o s o ype A15, Hel .
Phys. Ac a 52 (1979) 580–596 (in F ench).
[64] A. Junod, T. Ja lbo g, J. Muelle , Hea -capaci y analysis o a la ge numbe o A15-
ype compounds, Phys. Re . B 27 (1983) 1568–1585.
[65] S.F. Pugh, Rela ions be ween he elas ic moduli and he plas ic p ope ies o
polyc ys alline pu e me als, Philos. Mag. 45 (1954) 833–838.
[66] O.L. Ande son, A simpli ied me hod o calcula ing he Debye empe a u e om
elas ic cons an s, J. Phys. Chem. Solids 24 (1963) 909–917.
[67] G.D. Mukhe jee, C. Bansal, A. Cha e jee, The mal expansion s udy o o de ed
and diso de ed Fe
3
Al: an e ec i e app oach o he de e mina ion o ib a ional
en opy, Phys. Re . Le . 76 (11) (1996) 1876–1879.
X. Yan, P. B ož, J. Vřešťál e al. Jou nal o Alloys and Compounds 865 (2021) 158715
19