Reduced ternary rare-earth-transition metal tellurides for the smaller rare-earth elements. An exploration and an explanation of the marked stability differentiation among the rare-earth elements in these phases

Article abstract of DOI:10.1002/zaac.200801399

The existence of further metal-rich condensed cluster compounds in R-Z-Te systems has been synthetically explored for R = Sc, Y, Pr, Dy, Er, Tm, Yb, Lu and, mainly, Z = Ru, Rh, Pd, Ag, Ir, Pt, Au. Ten new examples of orthorhombic Er₇Ni₂

Full text of DOI:10.1002/zaac.200801399

ARTICLE
DOI: 10.1002/zaac.200801399
Reduced Ternary Rare-Earth؊Transition Metal Tellurides for the Smaller
Rare-Earth Elements. An Exploration and an Explanation of the Marked
Stability Differentiation among the Rare-Earth Elements in These Phases
Nina Herzmann,[a] Shalabh Gupta,[b] and John D. Corbett*[b]
Dedicated to Professor Gerd Meyer on the Occasion of His 60th Birthday
Keywords: Solid-state reactions; Rare earth compounds; Phase stabilities; Stability trends; Metal interstitials
Abstract. The existence of further metal-rich condensed cluster
compounds in RϪZϪTe systems has been synthetically explored
pears to be limited to those for R ϭ Sc, Y, and DyϪLu, a trend
that is parallel to but more emphatic than those variations found
for R ϭ Sc, Y, Pr, Dy, Er, Tm, Yb, Lu and, mainly, Z ϭ Ru, Rh, among parallel cluster halide systems. Stability trends among the
Pd, Ag, Ir, Pt, Au. Ten new examples of orthorhombic Er7Ni2Te2-
type (Imm2) have been identified and that for Dy7Ir2Te2 has been
refined. Seven new examples of other ternary structure types plus
X-ray powder pattern evidence for 14 unknown phases have also
been identified. To date the family of ternary RϪZϪTe phases ap-
halide and especially the telluride series follow the I1 ϩ I2 sums
for the R elements well, the larger values of which are considered
to reflect better mixing of R and Z valence d orbitals in the more
stable phases.
bility of these phases. Condensation of the rare-earth-metal
clusters to chain, sheet, or network structures in various
systems naturally follows a lower number of valence elec-
trons and decreases of X:M reaction proportions.
Introduction
The rare-earth elements R (Sc, Y, LaϪLu) have in recent
decades provided a notable variety of otherwise unpre-
cedented metal-rich cluster phases in which the clusters are
centered by a late transition metal heteroatom Z. The clus-
ter phases were first identified among the halides, initially
for zirconium and later with the rare-earth metals, as more
or less stoichiometric cluster phases with carbon, nitrogen,
oxygen, boron, silicon, etc. interstitials (Z), often as com-
mon impurity atoms [1, 2]. The fundamental cluster center-
ing condition appears to originate with the otherwise elec-
tron-poor character of the M6X12 cluster shells (M ϭ Zr,
R; X ϭ Cl, I) relative to the familiar tantalum and niobium
analogues [3]. Clusters of the earlier metals are naturally
larger and afford more room for the interstitials. The center-
ing of these nonmetal components also continues the theme
of more-or-less alternating metalϪnonmetal shells. The ad-
ded bonding, principally via nd on M and s and p orbitals
on the interstitial, certainly contributes to the overall sta-
A substantial advance in the concepts of this solid state
cluster chemistry occurred with the discovery that both
rare-earth metal (and some zirconium) cluster halides could
be alternatively centered by metal atoms instead, namely, a
considerable number of the later transition metals, viz.,
3dϪ5d metals in groups 7Ϫ11 [4, 5]. Orbital bonding
schemes in such discrete clusters differed from those for
centered nonmetals mainly in the introduction of metal d
orbitals, but the occurence of such diverse examples of
heteroatomic RϪZ intermetal bonding was novel and sig-
nificant. This feature appears to follow on earlier general
concepts developed by Brewer and co-workers [6] regarding
the particularly strong bonding present in alloys between
early and late transition metals. The dependencies of sta-
bility of particular compositions and structure types on the
locations of R and Z in the periodic Table are not well
studied or clear, but Mullikan electronegativity differences
[7] give some rough idea of the relative heteroatom bonding
components. However, stability intercomparisons are more
complicated than that because stability is always relative to
those of other competitive phases in each system. The latter
may change from one Z element to the next, even among
the binary products.
* Prof. Dr. J. D. Corbett
Fax: ϩ1-515-294-6789
E-Mail: jcorbett@iastate.edu
[a] Institute of Inorganic Chemistry
Cologne University
Greinstraße 6
50939 Cologne, Germany
[b] Ames Laboratory and Department of Chemistry
Iowa State University
The extent of cluster condensation and the varieties of
structure types among the rare-earth-metal cluster halides
appear to be relatively limited by the size and number of
353 Spedding Hall
Ames, IA 50011 USA
848
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2009, 635, 848Ϫ854

Reduced Ternary Rare-EarthϪTransition Metal Tellurides
halides that in effect sheathe the metal clusters. A consider- R7Z2Te2-type-Phases
able gain in novelty and variety of compounds formed and
All compounds were found to form during arc-melting, suggesting
in the degrees of condensation is found on change of the
electronegative component from halide to, especially,
telluride. This probably occurs because half as many anions
are then necessary to yield roughly the same electron count
per atom (ϳ2) in the metal network. Two- and three- di-
mensional networks are now common [8], with the separate
telluride anions bound within isolated cavities, and the
strength and numbers of metalϪmetal interactions in these
appear to be less limited by “matrix” (packing) effects and
more nearly intrinsic than with halides. Such effects can be
judged in terms of variations of COHP values (ϳ overlap
populations) for “bonds” versus observed distances among
various metal atom pairs and the deleterious effect that
neighboring telluride has on the former [5, 9, 10].
The number of RϪZ systems among the tellurides is con-
siderable, ϳ 240 for the practical R and the Z listed above,
and most of the novel results obtained to date [8, 10Ϫ12]
have been found in relatively scattered systems rather than
during systematic explorations. Moreover, negative results
are seldom reported. Therefore, we have undertaken a more
systematic survey of such ternary tellurides for a range of
R ϭ Sc, Y, Pr, Dy, ErϪLu and Z ϭ RhϪAg and IrϪAu
and with a fairly regular synthetic procedure. The results
soon focused on new examples of the orthorhombic
R7Z2Te2 structure [13, 14], although a few hexagonal
R6ZTe2 examples (Fe2P derivatives) [15Ϫ17] were also
found. The choice of R metals reflects our earlier obser-
vations that the more unusual products appear to occur
principally with Sc, Y, and the heavier lanthanides. This as-
pect will be discussed later in terms of some fundamental
properties of the R elements.
these were among the higher melting phases in these systems. For
good quality single crystals, the crushed samples were subsequently
annealed between 935 °C and 1250 °C over 4 days to 4 weeks.
These were sealed into tantalum tubes, and these sealed into silica
jackets for reactions run in tube furnaces below 1100 °C. Other
samples in Ta containers alone were annealed above 1100 °C in a
graphite-heated high vacuum furnace. The yields of the 7:2:2 com-
pounds on annealing did, in general, not change much compared
with those after arc-melting.
The targets of exploratory syntheses were of course always some-
thing new, and so many loaded compositions did not match the
stoichiometry R7Z2Te2. In detail, samples yielding this structure
were loaded as follows: Y7Au2Te2 formed in a sample with a loaded
Y17Au6Te3 composition (following Er17Ru6Te3 [11]). Annealing at
1100 °C for seven days and cooling to 850 °C at 10 °C/h led to
ϳ75 % of Y7Au2Te2 plus an unknown phase, which grew to ϳ40 %
at 1200 °C. A yield of 30 % Dy7Rh2Te2 formed from a Dy16Rh3Te3
composition annealed for one week at 1200 °C. Loading
Dy20Pd6Te3, Dy5IrTe or Er20Ag6Te3 and tempering each at four
weeks at 935 °C led to 80 % Dy7Pd2Te2, 20 % Dy7Ir2Te2, and nearly
single-phase Er7Ag2Te2, respectively. However, the last was ob-
tained only when the sample is quenched from 1000 °C. Single crys-
tal diffraction data were collected for Dy7Ir2Te2. The Dy7Pt2Te2
was obtained from samples loaded as Dy16Pt3Te3, Dy20Pt6Te3, or
Dy5PtTe that were heated to 1000 °C or 1100 °C for more than one
week. The powder pattern of a Dy5PtTe composition showed al-
most only Dy7Pt2Te2, whereas Er7Rh2Te2, Er7Pd2Te2 and Er7Pt2Te2
crystallized in ϳ90 % yields from the corresponding Er12Z3Te2
samples annealed at 1250 °C for four days. The first was also
gained in a 65 % yield from a Er20Rh6Te3 sample annealed at
1000 °C for 9 days, with Er7Rh3 as a byproduct. Furthermore, 70 %
Er7Pt2Te2 was obtained from an Er16Pt3Te3 sample tempered at
1100 °C for 13 days. Er7Ir2Te2 was synthesized in a 60 % yield from
samples loaded as Er12Ir3Te2 or Er5IrTe and annealed at 1200 °C
for one week. A composition of Lu20Rh6Te3 annealed at 1000 °C
for one week showed ϳ70 % Lu7Rh2Te2.
Experimental Section
Syntheses
Other Product Types
All materials were handled in a nitrogen or helium-filled glove box.
The starting rare-earth materials were pieces of Sc, Y, Pr, Dy, Er,
Tm, Yb, or Lu (Ames Lab, 99.99 % total), the Z components Ru,
Rh, Pd, Ag, Os, Ir, Pt or Au as pieces or foil (Alfa, >99.5 %),
and Te ingots (Aldrich, 99.99 %). The syntheses started with the
preparation of the corresponding RTe, for which small pieces of R
and Te in 1:1 proportions were sealed in evacuated silica tubing
and heated over 16 h to 475 °C, held for 6 h, and then raised over
4 h to 850Ϫ900 °C, where they were held for 6h. Guinier X-ray
powder diffraction data showed only the target phases. Mixtures of
R, RTe, and Z were pelletized with a hydraulic press, and then arc-
melted with a 20Ϫ30 ampere current for 20Ϫ30 s per side. The
weight losses during arc-melting were < 2 %. The buttons gained
from arc-melting were crushed into smaller pieces, and portions
ground into fine black powders for powder diffraction analyses.
Our intent was to explore phase space around R7Z2Te2 compo-
sitions; thus some reactions were loaded with that composition
whereas some 7:2:2, and other products were also identified follow-
Some systems yielded only new examples of known hexagonal 6:1:2
[15] or the 17:6:3 [11] structures. Reactions with erbium and
osmium gave Er6OsTe2 in a 80Ϫ90 % yield from Er12Os3Te2 or
Er7Os3Te compositions tempered at 1250 °C for four days. More
than 80 % R6RuTe2 was obtained from reactions loaded with
R20Ru6Te3, R ϭ Dy, Er, and run at 935 °C for four weeks. The
orthorhombic Er6AgTe2 was obtained under similar conditions.
The 6:1:2 phase formed similarly from Lu20Ru6Te3 or Lu17Ru6Te3
samples that were annealed at 1000 °C for at least one week.
Lu20Ru12Te3 that was tempered at 1200 °C for one week gave 85 %
of an unknown phase. However, the Dy20Ru6Te3 composition an-
nealed at 1000 °C for 9 days and quenched gave the new version
of a structure known with Er [11], Dy17Ru6Te3, in ϳ80 % yield, but
an unknown product was obtained on slow cooling, without
quenching.
Direct arc-melting of thulium and ytterbium test samples was not
feasible because of the substantial volatility of those elements.
ing exploratory reactions starting with other compositions. None Therefore, samples of Tm10Ru2Te3 were reacted at 1125 °C for two
of the samples showed sensitivity to moist air, at least over one day. weeks, which gave ϳ70 % Tm6RuTe2. The same procedure with a
Z. Anorg. Allg. Chem. 2009, 848Ϫ854
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
849

N. Herzmann, S. Gupta, J. D. Corbett
Tm5Ru3Te composition led to TmTe and ϳ15 % of an unknown chromatized Mo-Kα radiation. Data reduction and a numerical ab-
ARTICLE
phase, but annealing these at 1000 °C for two weeks also led to
sorption correction according to crystal shape optimization were
applied utilizing the STOE program package. The WinGX suite of
programs [18], including SHELXS-97 and SHELXL-97 [19], were
employed for structure solution and refinement. The last refine-
ment cycles included atomic positions and anisotropic displace-
ment parameters for all atoms. Some data collection and refine-
ment parameters for Dy7Ir2Te2 are listed in Table 1. Further details
of the crystal structure of Dy7Ir2Te2 are available from the Fach-
informationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen,
Germany, on quoting the depository number CDS-420220, the
names of the authors, and the citation of the paper, E-mail:
crysdata@fiz-karlsruhe.
80 % Tm6RuTe2. sample of Tm5Rh3Te so yielded 20 %
A
Tm6RhTe2, 35 % of an unknown, and TmTe. The compositions
Yb10Ru2Te3 and Yb5Ru3Te reacted either at 1000 °C or at 1125 °C
for two weeks exhibited little reaction except for small amounts of
the 6:1:2 phase.
Some samples showed neither the 6:1:2 nor the 7:2:2 phase. These
were all samples that could not be arc-melted because of high metal
volatility, namely for thulium and ytterbium. Yb10Pd2Te3 and
Yb5Pd3Te compositions annealed at 1000 °C or 1125 °C for two
weeks led to 40 % Yb3Pd and 10 % Yb5Pd2. The powder pattern
of Ho3Pd2 accounted for at least 30 % of the patterns from
Tm10Rh2Te3, Tm10Pd2Te3, and Tm5Pd3Te samples. About 45 % of
an unknown phase forms from Tm10Pd2Te3 at 1125 °C. About
30 % TmAu appears when Tm5Au3Te is annealed at 1125 °C, but
Table 1. Some X-ray data collection and refinement parameters
for Dy7Ir2Te2.
ϳ45 % of an unknown appears after
a
1000 °C reaction.
Empirical formula
Dy7Ir2Te2
YbϪAuϪTe samples show YbTe as main product and 20 % of the
same unknown as from the Tm reactions. This is also true for a
Tm10Au2Te3 composition. The reaction temperatures used for the
Tm and Yb systems were probably too high.
Formula weight /g·molϪ1
Crystal system
1777.10
orthorhombic
Imm2 (No. 44), 2
3.9552(8)
15.645(3)
9.392(2)
581.2(2)
10.155
72.033
2.53Ϫ29.04
1440
Ϫ5 Յ h Յ 5, Ϫ17 Յ k Յ 21,
Ϫ12 Յ l Յ 12
2972
905 (Rint ϭ 0.0692)
905 /38
Space group, Z
˚
a /A
˚
b /A
˚
c /A
Four compositions in the ScϪAuϪTe system, Sc6Au3Te, Sc6Au6Te,
Sc6Au3Te4, and Sc17Au6Te3, all produced the same unknown
phase(s) at 1100 °C. The 40 % unknown from the first does not
change for the Sc6Au6Te composition, but vanishes at Sc6Au3Te4,
and increases for Sc17Au6Te3. Some of the products in the last
sample decrease on reactions at 1200 °C. In addition, ScAu is found
in the Sc6Au3Te and Sc6Au6Te samples, and ScTe occurs in 75 %
yield for the Te-richer Sc6Au3Te4. YϪAuϪTe samples gave 95 % of
an unknown phase from a Y6Au3Te sample after 1000 °C and
1100 °C reactions, but this decreased to only 25 % at 1200 °C along
with 15 % of a second unknown phase. The composition Y6Au6Te
showed 35 % of another unknown after reactions over a range of
1000 °C to 1200 °C. Finally, test experiments with the lighter
lanthanide praseodymium plus nickel, ruthenium, rhodium or
platinum in 7:2:2 stoichiometries gave no like targets after arc melt-
ing or subsequent annealing at 1100 °C for 12 days. Reactant PrTe
was the main phase plus a presumed alloy of praseodymium and
Z that showed only a systematic shift with the radii of the Z. A
summary of all of these results appears later.
3
˚
V /A
dcalcd. /g·cmϪ3
Absorp. coeff. (Mo-Kα) μ /mmϪ1
Θ range /deg.
F(000)
Range in h k l
Reflections collected
Independent reflections
Data /parameters
Goodness-of-fit on F2
Final R1, wR2 indices
[I > 2σ(I)]
1.140
0.0416, 0.0905
0.0476, 0.0920
4.734, Ϫ3.382
(all data)
Largest diff. peak and hole /e·A
Ϫ3
˚
Results and Discussion
These explorations of ternary rare-earth-metal-rich tellu-
ride systems have led to a good variety of new compounds,
as identified mainly with the aid of powder patterns calcu-
lated for known RϪZϪTe structure types. These results
give better insights into the breadth of possible phases. The
most general new feature is the common but not uniform
occurrence of orthorhombic R7Z2Te2-type phases. Hereto-
fore, only a few examples [13, 14] were known, namely,
Er7Ni2Te2 and Lu7Z2Te2, Z ϭ Ni, Pd, Ru, in contrast to
the seemingly more common orthorhombic and hexagonal
R6ZTe2-type compounds. However, the present study has
turned up 10 more R7Z2Te2 examples, Y with Au, Dy with
Pd, Ir or Pt, five more for Er with Z ϭ Rh, Pd, Ag, Ir, or
Pt, and Lu with Rh. These are inevitably evident in the
powder patterns obtained directly after arc-melting, indicat-
ing their appreciable thermal stabilities as well. Unit cell
data for the new members of this family are listed in
Powder X-ray Diffraction
Powder diffraction patterns were recorded in the 2Θ range of
5Ϫ100° over 30 min with the aid of a Huber Guinier 670 image-
plate diffractometer and Cu-Kα1 radiation. The samples were
ground to a fine powder in the glove box and evenly distributed
between two Mylar films, which were then mounted between Al-
rings. Diffraction data were collected immediately. Estimated yields
in vol % were achieved by comparison of the observed patterns
with those calculated from structures of known, neighboring com-
pounds
Single-Crystal Diffraction Studies
Several well-formed single crystals of Dy7Ir2Te2 were selected and
protected with a layer of silicone oil. Their qualities were checked
with a single-crystal X-ray diffractometer (Stoe Image Plate Dif-
fraction System, IPDS II) and a complete intensity data set was
collected at ambient temperature with the aid of graphite-mono- Table 2.
850 www.zaac.wiley-vch.de
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2009, 848Ϫ854

Reduced Ternary Rare-EarthϪTransition Metal Tellurides
Table 2. New compounds with the orthorhombic Er7Ni2Te2-type
structure (Imm2).
3
˚
˚
˚
˚
Compound a /A
b /A
c /A
Volume /A Lines indexed
a)
Y7Au2Te2
4.041(2) 15.53(3) 9.687(9) 607.9(1)
13
4
a)
Dy7Pd2Te2 3.935(2) 15.541(1) 9.541(6) 588.8(7)
b)
Dy7Ir2Te2
3.9552(8) 15.645(3) 9.392(2) 581.2(2)
a)
Dy7Pt2Te2 3.9569(5) 15.607(3) 9.539(2) 589.1(1)
10
8
8
a)
Er7Rh2Te2 3.8877(7) 15.50(1) 9.363(3) 564.1(3)
a)
Er7Pd2Te2 3.9166(6) 15.452(7) 9.513(2) 575.7(2)
b)
Er7Ag2Te2 3.949(6) 15.676(2) 9.409(2) 582.4 (4)
a)
Er7Ir2Te2
Er7Pt2Te2
3.9399(7) 15.435(3) 9.328(3) 567.3(1)
3.9301(9) 15.458(9) 9.476(3) 575.6(3)
13
11
16
a)
a)
Lu7Rh2Te2 3.854(1) 15.290(5) 9.289(4) 547.4(2)
a) Lattice constants refined from Guinier powder data. b) Lattice
constants refined from IPDS-II single crystal data.
Table 3 summarizes seven other new examples found
among the eight R and the eight Z (Ru, Rh, Pd, Ag, Os, Ir,
Pt, Au) combinations studied that are isostructural with
other known ternary compositions, as well as 14 instances
in which new patterns could not be assigned. (only negative
results are omitted.) The number of unknown structures
with gold in the last group is appealing. All of the Z mem-
bers tested except Os acted as interstitials frequently. The R
dependences generally followed the observations that in
Figure 1. Off-[100] projection of the orthorhombic Dy7Ir2Te2 struc-
ture along the short a axis.
Structural Description
A slightly-off-[100] section of the structure of Dy7Ir2Te2
part prompted this study (below). An exception might be is shown in Figure 1. The principal building units outlined
that Dy, Er and Lu are more frequent participants than Tm consist of distorted, Ir-centered, tricapped trigonal prisms
or Yb although the last two have been less well studied. of Dy that share basal trigonal faces to form columns along
However, comparisons when both Z and R are varied are the a axis. Adjacent columns along c are displaced along the
not that simple. Major complications in such homologous projection axis by a/2 by means of a simply and common
series may arise from irregular changes in the competitive regularity. As depicted in Figure 2, the interconnections
phases that define stability. Thus, series of comparative into puckered layers in this direction take place via bifunc-
RϪZ binary phases often exhibit irregular variations in tional Dy1 and Dy3 that are members of a trigonal prism
composition and structure with changes in Z. The electron in one column and also cap a rectangular face in its
counts may also be important although these last two ef- neighbor. This type of function is common in many metal-
fects have not been studied. Finally, the four experiments rich structures, often showing staggered bridging in three
with Pr (above) were undertaken to test the proposition that directions via tri-capped trigonal prisms, but this structure
the light lanthanides were uncommon, even unknown, par- features only the bicapped motifs here. Finally, the three-
ticipants in this ternary telluride chemistry, and the negative dimensional structure is generated by Dy4 atoms that face-
results appear to reaffirm that postulate.
cap on adjoining layers along the b direction, Figure 1. In
Table 3. Additional literature and experimental results for ternary RϪZϪTe phasesa),b)
.
Sc
Y
Pr
Dy
Er
Tm
Yb
Lu
ternary
tellurides Sc6ZTe2 (H)
Sc5Ni2Te2 [20],
Y5Z2Te2
(Z ϭ Fe, Co,
Dy6ZTe2 (H),
(Z ϭ FeϪNi) [16], (Z ϭ Co, Ni),
Dy6RuTe2 (H),
Dy17Ru6Te3
Er4ZTe2,
Tm6RuTe2 (H),
Tm6RhTe2 (H)
Lu6RuTe2 (H),
Lu6AgTe2 (O) [10],
Lu7Z2Te2 (Z ϭ Ni,
Pd, Ru) [14]
(Z ϭ Mn, Fe, Co, Ni) [22, 23],
Ni, Ru, Os, Rh, Y6ZTe2 (O)
Pd, Ir, Ag, Cu, Cd) (Z ϭ Rh, Pd,
[15, 17], Sc14Z3Te8 Ag, Y) [22]
(Z ϭ Ru, Os) [21]
Er5Z2Te2
(Z ϭ Co, Ni) [24],
Er6ZTe2 (H)
(Z ϭ Co, Ni, Ru)
[10], Er6AgTe2 (O),
Er6OsTe2 (H)
Er7Ni2Te2 [11],
Er17Ru6Te3 [12],
Er21Ir8Te4 [13]
unknown Au
tellurides
Au (3 phases)
Ru, Ir
Pd
Ru, Rh, Pd, Au Rh, Au
Ru
a) All known compounds are listed in italics with the corresponding references. All new R7Z2Te2type phases are listed only in Table 3. b) H and
¯
O distinguish between hexagonal (P62m, No. 189) and orthorhombic R6ZTe2 phases (Pnma, No. 62), respectively.
Z. Anorg. Allg. Chem. 2009, 848Ϫ854
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
851

N. Herzmann, S. Gupta, J. D. Corbett
ARTICLE
addition, Te1 is surrounded by a bicapped trigonal prism neighbours even though it also plays both prismatic and
of Dy, and Te2, by a monocapped trigonal prism of Dy, capping roles in the structure. Finally the relatively large
˚
common ways of accommodating the separate anions.
size of Te is what fixes both the short a axis at 3.995 A, and
the prismatic heights as well, Figure 2. These fixed separ-
ations for all atoms have been omitted from the d(DyϪDy)
considerations above.
Distinctions Among Rare-Earth Elements in Ternary Metal-
rich Tellurides
Tellurides of the sort described here exhibit some remark-
able differences among the rare-earth elements in their
metal-richest phases, namely, that substantially all examples
occur with scandium, yttrium, and at least the heavier lan-
thanide members that have been better studied, dysprosium
through lutetium. Extensive earlier investigations among
the condensed cluster halides of the rare-earth elements that
were likewise stabilized by diverse centered transition
metals (e.g., R12(Z)X14) also revealed a few similar but less
extreme differences wherein unusual distortions occurred
for particular combinations of R and Z. (Recall that the
degrees of cluster condensation are less in the anion-richer
halides, and the metalϪmetal interactions are correspond-
ingly less numerous.) Thus, certain cluster chain structure
types show significant distortions, Y3I3Ir but not Pr3I3Ru,
for example [26]. Other distortions are found among the
Figure 2. ϳ [010]-view of a portion of the chains of condensed
trigonal prismatic dysprosium atoms (black), each centered by an
iridium atom (smaller spheres), that lie along c (Figure 1) in
Dy7Ir2Te2. Note that prismatic Dy1 and Dy3 atoms in one column
constitute the face-capping atoms in its neighbours.
Appreciable distortions accompany the formation of this tetrameric R16Z4I20 oligomers and their differentiation from
structure type with its reduced symmetry and stoichiometry the chain structure of the isomeric R4ZI5 is similarly related
change. The simplest measures are the DyϪIr distances, [27]. In all cases, the distortions appear to lead to more and
˚
which range over only 2.84 to 2.91 A within the trigonal shorter RϪR, RϪZ and ZϪZ distances. Köckerling and
˚
prisms, but these increase to 3.10 and 3.80 A for the face- Martin [28] produced an attractive interpretation of all of
capping dysprosium atoms. The different functionalities of, these distortions in terms of those R and Z that had the
and the nearest neighbors to, the four types of dysprosium more similar orbital energies and thence better mixing be-
atoms are reflected in much larger and more diverse varia- tween the (mainly) valence d orbitals on R and Z. They
tions among DyϪDy distances. In more regular cases, compared the last effects in terms of the differences in first
prismaticϪprismatic atom distances are usually somewhat ionization energies of the respective R and Z elements.
smaller than for prismaticϪface-capping separations, de- Orbital energies tabulated for EHTB calculations work
pending somewhat on the relative height of the trigonal similarly [26, 27].
prisms (along c here). In Dy7Ir2Te2, the Dy1 and Dy3
The tellurides appear to exhibit more extreme versions of
atoms play both prismatic and face-capping roles (Figure 1, these differences, presumably because of the greater degrees
˚
Figure 2), and show d(DyϪDy) ranges of 3.47Ϫ3.81 A of condensation and thence increased numbers of contacts
˚
around Dy1 and a contrasting 3.47Ϫ3.69 A about Dy3. The among R and Z. Moreover, certain R elements evidently
˚
former is the same as a 3.47Ϫ3.81 A range about Dy2, even form no such materials, and the stable condensed metal-
though this has only a prismatic role in the structure, rich tellurides are heavily concentrated among Sc, Y and
˚
whereas an intermediate 3.64Ϫ3.77 A is found around Dy4, (so far) Dy through Lu. (However, it would appear that Gd
which is only face-capping between chains. However, the and Yb have not been well investigated with respect to the
numbers of Te bound to each type of R atom is also known latter region.) Since there was a concern that our R sam-
to have a substantial effect on its RϪR distances (and bond pling had not been very uniform (you fish in places that
populations) [9, 20]. Thus appreciable (polar) covalent yield fish!), four additional reactions were run under the
DyϪTe bonding naturally raises the higher lying 5d (and standard protocol with Pr7Z2Te2 compositions, but nothing
other) levels on Dy and leads to weaker DyϪDy interac- pertinent was seen (text in Experimental Section; Table 3).
tions with its less affected neighbors. Figure 1 shows that
An extension of the above ideas regarding unusual distor-
Dy3 is indeed more shielded from Te than the others, with tions observed in the analogous condensed cluster halides
only one Te neighbour versus two (Dy1) or three (Dy2, qualitatively fits changes in telluride stabilities as well, and
Dy4) for the others. (Dy3 also has one more Ir neighbour.) more broadly. Figure 3 shows on the left a plot of the I1 ϩ
˚
This is presumably the reason that Dy3 has about a 0.13 A I2 energies [29] for the rare-earth elements with energies
smaller value for its upper limit of bond lengths to its like increasing downward. These are considered to afford some
852
www.zaac.wiley-vch.de
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2009, 848Ϫ854

Reduced Ternary Rare-EarthϪTransition Metal Tellurides
Figure 3. Left: The variation of the sum of the first and the second ionization energies within the lanthanide series of elements and those
for three typical transition metals below (dashed lines). Elements below the thick black line generally form ternary rare-earth-metal-rich
tellurides. Right: Representations of variation of the degrees of orbital mixing that accompany different relative energies of R and
Z orbitals.
readily available measures of the relative orbital energy
levels (and atom electronegativities) for R. Also marked be-
References
low are the analogous values for three fairly extreme Z
members: Mn, Rh and Pt. The panel on the right reminds
[1] A. Simon, Hj. Mattausch, G. J. Miller, W. Bauhofer, R. K.
Kremer in: K. A. Gschneidner Jr., L. Eyring (Eds.), Handbook
us that, in the simplest picture, the magnitudes of bonding/
mixing of orbitals on different atoms vary inversely with
the energy difference between their valence (nd) orbitals (all
other things being equal). We take this parallel between the
ionization energy properties of R and Z to reflect by some
crude measure the observation that the RϪZ interactions,
and thence bond energies and relative stabilities follow the
differences in rare-earth orbital energies as measured by
ionization parameters fairly well. Of course, there are many
other features that enter into bond strengths and phase
stabilities (and always with respect to those of the most
stable decomposition products), but this simple measure
gives a striking parallel. Again, the relative properties and
stabilities of the Gd and Tb analogues have not been stud-
ied well, although it is noteworthy that only Sc, Gd and Dy
have so far yielded the unusual binary Sc2Te-type struc-
ture [30].
on the Physics and Chemistry of Rare Earths, vol. 15, Elsevier-
Science Publishers, Amsterdam, 1991, p. 191.
[2] J. D. Corbett, J. Alloys Compd. 1995, 229, 10.
[3] H. Schäfer, H.-G. Schnering, Angew. Chem. 1964, 76, 833.
[4] T. Hughbanks, J. D. Corbett, Inorg. Chem. 1988, 27, 2022.
[5] J. D. Corbett, J. Alloys Compd. 2006, 418, 1.
[6] L. Brewer, P. R. Wengert, Metall. Trans. 1973, 4, 83.
[7] R. G. Pearson, Inorg. Chem. 1988, 27, 734.
[8] J. D. Corbett in Inorganic Chemistry in Focus II, G. Meyer,
D. Naumann, L. Wesemann, (Eds.), Wiley-VCH, Weinheim,
Germany, Vol. 2, 2005, ch. 8.
[9] P. A. Maggard, J. D. Corbett, J. Am. Chem. Soc. 2000, 122,
838.
[10] L. M. Castro-Castro, L. Chen, J. D. Corbett, J. Solid State
Chem. 2007, 180, 3172.
[11] A. Mehta, J. D. Corbett, J. Solid State Chem. 2008, 181,
871.
[12] S. Gupta, J. D. Corbett, unpublished research.
[13] F. Meng, C. Magliocchi, T. Hughbanks, Inorg. Chem. 2001,
40, 2482.
[14] L. Chen, J. D. Corbett, Inorg. Chem. 2004, 43, 3371.
[15] P. A. Maggard, J. D. Corbett, Inorg. Chem. 2000, 39, 4143.
[16] N. Bestaoui, S. Herle, J. D. Corbett, J. Solid State Chem. 2000,
155, 9.
[17] L. Chen, J. D. Corbett, Inorg. Chem. 2004, 43, 436.
[18] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837.
[19] G. M. Sheldrick, SHELXL-97 and SHELXS-97; Programs for
Crystal Structure Analysis, Universität Göttingen, 1997.
[20] P. A. Maggard, J. D. Corbett, Inorg. Chem. 1999, 38,
1945.
[21] L. Chen, J. D. Corbett, J. Am. Chem. Soc. 2003, 125, 1170.
[22] P. A. Maggard, J. D. Corbett, J. Am. Chem. Soc. 2000, 122,
10740.
Clearly the results presented herein are nowhere near the
final answers. Some of the more attractive problems are a
clearer positioning of Gd and Tb in these systems and,
especially, the many unidentified and doubtlessly novel
telluride products implied by the unknown phases listed in
Table 3.
Acknowledgement
We thank Professor Gerd Meyer for the support of N. H. provided
by the University of Köln. Partial support (S. G.) came from the
Office of Basic Energy Sciences, Materials Sciences Division, U.S.
Department of Energy (DOE). The work was performed in the
Ames Laboratory, which is operated for DOE by Iowa State Uni-
versity under Contract No. DE-AC02Ϫ07Ch11358.
[23] P. A. Maggard, J. D. Corbett, Inorg. Chem. 2004, 43,
2556.
[24] C. Magliocchi, F. Meng, T. Hughbanks, J. Solid State Chem.
2004, 177, 3896.
Z. Anorg. Allg. Chem. 2009, 848Ϫ854
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
853

N. Herzmann, S. Gupta, J. D. Corbett
ARTICLE
[25] F. Meng, C. Magliocchi, T. Hughbanks, J. Alloys Compd. 2003,
358, 98.
[26] M. W. Payne, P. K. Dorhout, S.-J. Kim, T. R. Hughbanks, J.
D. Corbett, Inorg. Chem. 1992, 31, 1389.
[29] D. A. Johnson, Some thermodynamic aspects of inorganic chem-
istry, 2nd. ed.; Cambridge University Press, Cambridge, 1982,
pp. 255Ϫ257.
[30] P. S. Herle, J. D. Corbett, Inorg. Chem. 2001, 40, 1858.
[27] S. J. Steinwand, J. D. Corbett, J. D. Martin, Inorg. Chem. 1997,
36, 6413.
Received: December 19, 2008
[28] M. Köckerling, J. D. Martin, Inorg. Chem. 2001, 40, 389.
854
www.zaac.wiley-vch.de
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2009, 848Ϫ854