Synthesis and structure of Ta4SI11: Disorder and mixed valency in the first tantalum sulfide iodide

Article abstract of DOI:10.1021/ic034075q

The new compound Ta₄SI₁₁ has been prepared by direct reaction of the elements at 430 °C for 2 weeks in evacuated Pyrex ampules and characterized by single-crystal x-ray diffraction, x-ray photoelectron spectroscopy, magnetic susceptibility measurements, and semiempirical electronic structure calculations. Ta₄SI₁₁ crystallizes with orthorhombic symmetry in space group Pmmn; a = 16.135(3) A, b = 3.813(1) A, c = 8.131(2) A, and Z = 1. The disordered structure involves two crystallographically distinct sites for Ta atoms, both of which are 50% occupied as well as a bridging anion site that is 50% S and 50% I. Magnetic susceptibility above 100 K gives μ_eff = 1.53 μ_B to suggest one unpaired electron per formula unit. x-ray photoelectron spectroscopy and extended Hueckel calculations suggest that the structure consists of Ta₃ triangles and isolated Ta atoms, leading to the formulation (Ta₃)⁹⁺(Ta⁴⁺)(S^2-) (I^-)₁₁ and we hypothesize that each Ta₃ is capped by a sulfur atom.

Full text of DOI:10.1021/ic034075q

Inorg. Chem. 2003, 42, 4165−4170
Synthesis and Structure of Ta4SI11: Disorder and Mixed Valency in the
First Tantalum Sulfide Iodide
Mark D. Smith† and Gordon J. Miller*
Department of Chemistry, Iowa State UniVersity, Ames, Iowa 50011
Received January 23, 2003
The new compound Ta4SI11 has been prepared by direct reaction of the elements at 430 °C for 2 weeks in
evacuated Pyrex ampules and characterized by single-crystal X-ray diffraction, X-ray photoelectron spectroscopy,
magnetic susceptibility measurements, and semiempirical electronic structure calculations. Ta4SI11 crystallizes with
orthorhombic symmetry in space group Pmmn; a ) 16.135(3) Å, b ) 3.813(1) Å, c ) 8.131(2) Å, and Z ) 1. The
disordered structure involves two crystallographically distinct sites for Ta atoms, both of which are 50% occupied
as well as a bridging anion site that is 50% S and 50% I. Magnetic susceptibility above 100 K gives µeff ) 1.53
µB to suggest one unpaired electron per formula unit. X-ray photoelectron spectroscopy and extended Hu¨ckel
calculations suggest that the structure consists of Ta3 triangles and “isolated” Ta atoms, leading to the formulation
2-
(Ta3)9+(Ta4+)(S )(I-)11 and we hypothesize that each Ta3 is capped by a sulfur atom.
19
Introduction
Nb3QX7,11-16 NbSeI,17,18 and Nb6SI9 display remarkable
structural diversity, adopting structures with isolated niobium
atoms, Nb-Nb dimers, Nb3 triangular clusters, Nb4 tetra-
hedral clusters, and Nb6 octahedral clusters, respectively.
Additionally, centered Nb6 trigonal prismatic clusters in the
case of the alkali-metal salts A3Nb6SBr17,20,21 and tetranuclear
butterfly clusters in CsNb4Cl11,22 have been characterized.
Generally, the metal atoms in these compounds can be
assigned a single oxidation state (though occasionally a
The number of well-characterized ternary tantalum chal-
cogenide halides is meager, limited to pseudo one-dimen-
sional chalcogen-rich compounds such as TaTe4I,1 (TaSe4)nI
(n ) 2, 3),2,3 and Ta4Se16Brn (n ) 1, 2)3,4 and to the layered
halide-rich compounds Ta3SBr75 and Ta3QI7 (Q ) Se, Te),6
recently discovered in our laboratory. A few others have been
mentioned in the literature, with only cursory structural
characterization. TaS2Cl2, TaSCl2, TaSCl3, TaSBr3, and
TaSeBr37 have been assigned their various structural features
on the basis of chemical analysis and metal-chalcogen
stretching frequencies only. No tantalum sulfide iodide, to
the best of our knowledge, has ever been reported. Certainly,
the tally of ternary tantalum chalcogenide halide compounds
lags well behind that of its lighter relative niobium, which
has yielded a rich ternary chemistry. NbTe4I,8 NbSeX3,9,10
(8) Deng, S.-Q.; Zhuang, H.-H.; Lu, C.-Z.; Huang, J.-S.; Huang, J.-L.
Acta Crystallogr. 1993, C49, 1135.
(9) Franzen, H. F.; Ho¨nle, W.; von Schnering, H. G. Z. Anorg. Allg. Chem.
1983, 497, 13.
(10) Schmidt, P. J.; Thiele, G. Z. Anorg. Allg. Chem. 1999, 625, 1055.
(11) Miller, G. J. J. Alloys Compd. 1995, 229, 93.
(12) Furuseth, S.; Ho¨nle, W.; Miller, G. J.; von Schnering, H. G. 9th
International Conference on Solid Compounds of Transition Elements,
Abstracts; Royal Society of Chemistry: London, 1988.
(13) Miller, G. J. J. Alloys Compd. 1995, 217, 5.
* Author to whom correspondence should be addressed. E-mail:
gmiller@iastate.edu.
(14) Khvorykh, G. V.; Shevelkov, A. V.; Dolgikh, V. A.; Popovkin, B. A.
J. Solid State Chem. 1995, 120, 311.
(15) Schmidt, P. J.; Thiele, G.; Whangbo, M.-H. Eur. J. Inorg. Chem. 1999,
785.
† Current address: Department of Chemistry and Biochemistry, Univer-
sity of South Carolina, Columbia, SC 29210.
(1) Tremel, W. Chem. Ber. 1992, 125, 2165.
(2) Gressier, P.; Geumas, L.; Meerschaut, A. Acta Crystallogr. 1982, B38,
2877.
(16) Schmidt, P. J.; Thiele, G. Acta Crystallogr. Sect. C 1997, C53, 1743.
(17) Fedorov, V. E.; Evstaf’ev, V. K.; Kirik, S. D.; Mishchenko, A. V.
Zh. Neorg. Khim. 1981, 26, 2701.
(3) Gressier, P.; Meerschaut, A.; Guemas, L.; Rouxel, J.; Monceau, P. J.
Solid State Chem. 1984, 51, 141.
(4) Grenouilleau, P.; Meerschaut, A.; Guemas, L.; Rouxel, J. J. Solid State
Chem. 1987, 66, 293.
(5) Smith, M. D.; Miller, G. J. J. Solid State Chem. 1998, 140, 226.
(6) Smith, M. D.; Miller, G. J. J. Am. Chem. Soc. 1996, 118, 12238.
(7) Fenner, J.; Rabinau, A.; Trageser, G. AdV. Inorg. Chem. Radiochem.
1980, 23, 364.
(18) Ben-Yaich, H.; Jegaden, J. C.; Potel, M.; Sergent, M.; Rostogi, A.
K.; Tournier, R. J. Less-Common Met. 1984, 102, 9.
(19) Meyer, J. H.; Corbett, J. D. Inorg. Chem. 1991, 30, 963.
(20) Womelsdorf, H.; Meyer, H. J. Angew. Chem. 1994, 106, 2022; Angew.
Chem., Int. Ed. Engl. 1994, 33 (19), 1943.
(21) Womelsdorf, H.; Meyer, H.-J. Z. Anorg. Allg. Chem. 1996, 622, 2083.
(22) Broll, A.; Simon, A.; von Schnering H.-G.; Scha¨fer, H. Z. Anorg. Allg.
Chem. 1969, 367, 1.
10.1021/ic034075q CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/29/2003
Inorganic Chemistry, Vol. 42, No. 13, 2003 4165

Smith and Miller
nonintegral one). Compounds made at higher temperatures
seldom exhibit mixed-valent behavior, where the same metal
exists in two different formal oxidation states in the same
compound. A few examples from niobium chemistry are
known, however. The binary halide Nb3Cl8 actually forms
in the composition range Nb3-xCl8 (x e 0.44)23 and has
been described as a mixed (Nb4+)2Cl8/(Nb2.67+)3Cl8 system.12
Nb3Se5Cl7 has been formulated (Nb5+)(Nb4+)2(Se22-)2(Se2-)-
(Cl-)7.24 Another example is the peculiar compound Nb7S2I19,
where trigonal bipyramidal (Nb5+)I5 molecules are encap-
sulated inside a layered (Nb3+)3SI7 framework.25 In this
compound, however, the NbI5 units can almost be considered
a kind of molecule of crystallization since they are encap-
sulated in large channels and are not directly joined to the
extended Nb3SI7 framework. Such compounds, besides
providing structural novelty, are also interesting examples
of the delicate balance possible between competing redox
forces within a structure.26
Figure 1. SEM images of (left) a typical spiny ball formation of Ta4SI11;
(right) higher magnification of the end of a crystal.
decomposing to an uncharacterized white powder after several days.
In inert atmospheres, the compound appears to be indefinitely stable
toward decomposition up to temperatures of ≈525 °C, above which
only Ta6I14, TaI5, and TaS2 are observed by X-ray powder
diffraction.
Scanning Electron Microscopy. Verification of the presence
of tantalum, sulfur, and iodine in multiple single-crystal samples
of Ta4SI11 was carried out using a JEOL 6100 scanning electron
microscope operating at 15 kV.
During the course of our studies of the ternary chalco-
genide halide chemistry of tantalum, we have focused on
halide-rich compositions, usually near Ta3QX7. As part of
an attempt to synthesize Ta3SI7, we have discovered the novel
mixed-valent compound Ta4SI11, containing both trinuclear
X-ray Photoelectron Spectroscopy. XPS spectra were measured
using a PHI 5500 multitechnique surface analysis equipment
instrument, using Mg KR radiation (29.35 eV). Samples were
prepared by finely powdering several larger single crystals of Ta4-
SI11 in the glovebox immediately before use. The samples were
pressed onto indium foil in the glovebox and transferred to the XPS
instrument via an inert atmosphere carrier. Binding energies were
corrected for charging using the C 1s peak of adventitious carbon
(284.8 eV) as an internal reference.27 However, no significant
charging effects were observed, as C 1s always appeared at 284.8
eV. Additionally, large intact single crystals were mounted and
etched by sputtered Ar ions to remove any surface impurities
introduced during handling, to probe the presumably pristine interior
of the sample.
3+
Ta3 clusters and isolated Ta4+ ions amid undulating,
disordered sulfur and iodine layers. Herein, we report the
synthesis as well as structural and theoretical characterization
of this new structure type.
Experimental Section
Synthesis. Ta4SI11 was first observed as an abundant (ca. 50%)
side product from reactions of the elements in the molar ratio 3:1:7
Ta:S:I at 450 °C in sealed, evacuated borosilicate glass tubes,
designed to grow Ta3SI7 phases. Because of the inability to identify
the compound from its Guinier powder diffraction pattern, a single
crystal was selected and its structure solved (see below), yielding
the composition “Ta4SI11”. Tubes were then loaded at this stoichi-
ometry and heated at various temperatures to determine optimum
synthetic conditions. Ta4SI11 is best made by stoichiometric reaction
of the elements at 430 °C, for a duration of 2 weeks. Growth of
the compound is sensitive to temperature gradients in the tube:
highest (nearly quantitative) yields were obtained when the tube
furnace was packed tightly with asbestos, to smooth out uneven
temperature regions as much as possible. Heating a Ta4SI11 mixture
with an imposed temperature gradient leads to a mixture of phases,
including hexagonal and orthorhombic Ta3SI7, and Ta6I14 in addition
to Ta4SI11. The reagents used were tantalum “turnings” (Aesar,
99.99%, cleaned with an HF/HNO3/H2SO4 solution to remove
surface impurities, then rinsed with ethanol, and dried in vacuo at
500 °C), sulfur powder (Alfa, 99.9%, sublimed twice before use),
and iodine (Alfa, 99.9%, resublimed). All purified reagents and
products were handled in an Ar-filled glovebox.
Magnetic Susceptibility. Temperature-dependent magnetic sus-
ceptibility measurements were performed with a Quantum Design
SQUID magnetometer. Aggregations of Ta4SI11 bars were loaded
and sealed under an inert atmosphere in fused silica tubes. The
samples were kept in place by means of two tightly fitting fused
silica rods on either side of the sample. The samples were chosen
so that only larger single-crystal aggregations were used, to avoid
unwanted powder impurities and to minimize the surface area
susceptible to oxidation. Measurements were taken from 4 to 300
K, at a field strength of 3 T. The results are shown in Figure 2. A
room-temperature moment of 1.53 µB was measured, reproducible
over three separate runs.
X-ray Crystallography. A small silver bar (0.02 × 0.03 × 0.12
mm3) was sealed inside a glass capillary under argon and mounted
on a Siemens P4 diffractometer. A primitive orthorhombic unit cell
with dimensions a ) 16.135(3) Å, b ) 3.813(1) Å, and c ) 8.131-
(2) Å was indexed and refined on the basis of 45 reflections in the
range 6° e 2θ e 25°. Axial photographs confirmed these lattice
metrics. 2629 reflections were collected to 2θmax ) 55°, of which
685 were unique, 533 unique observed (I > 2σI), and Rint ) 0.0498.
An empirical absorption correction was applied to the data, using
the average of six complete “psi-scans” measured in well-separated
regions of reciprocal space. The structure was solved using direct
methods (SHELX-86) and refined with the SHELXL-93 crystal-
Ta4SI11 crystallizes as long silver bars, which often aggregate in
dense thickets (Figure 1). The brittle crystals splinter easily along
the long axis. Ta4SI11 is sensitive to oxygen and moisture,
(23) Hulliger, F. In Structural Chemistry of Layer-Type Phases; Le´vy, F.,
Ed.; Reidel: Dordrecht, 1976.
(24) Rijnsdorp, J.; Jellinek, F. J. Solid State Chem. 1979, 28, 149.
(25) Miller, G. J.; Lin, J. Angew. Chem. 1994, 106, 357; Angew. Chem.,
Int. Ed. Engl. 1994, 33, 334.
(27) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook
of X-ray Photoelectron Spectroscopy; Chastain, J., Ed.; Perkin-Elmer
Corp.: Eden Prairie, MN, 1992.
(26) Rouxel, J. Comments Inorg. Chem. 1993, 14, 207.
4166 Inorganic Chemistry, Vol. 42, No. 13, 2003

Synthesis and Structure of Ta4SI11
Figure 2. Temperature-dependent magnetic susceptibility for Ta4SI11
Figure 3. Displacement ellipsoid view (80% probability) and atom labeling
scheme showing the Ta3 clusters and isolated Ta centers hypothesized to
be disordered throughout the structure of Ta4SI11. Independent atomic
positions are labeled in bold; symmetry-equivalent positions are smaller
and gray.
(molar susceptibility ø, emu mol-1, and inverse molar susceptibility ø-1
,
mol emu-1). A fit to Curie-Weiss behavior for temperatures exceeding
100 K gives µeff ) 1.53 µB.
Table 1. Summary of Crystallographic Data for Ta4SI11
fw
2151.76
space group
lattice params
a (Å)
Pmmn (No. 59)
16.135(3)
3.813(1)
b (Å)
c (Å)
8.131(2)
500.2(2)
1
vol (Å3)
Z
d
calc (g cm-3
)
7.143
radiation
Mo KR, λ ) 0.71071 Å
38.854 mm-1
23 °C
linear abs coeff
temp
Figure 4. (a) [010] projection of Ta4SI11. Larger black circles are Ta atoms;
smaller black circles are S atoms; large white circles are I atoms. (b) [001]
projection of Ta4SI11. Both figures illustrate all fully and partially occupied
positions. In the [001] projection, it is clear that two Ta sites cannot be
simultaneously occupied.
residualsa
R1 (observed/all data)
wR2 (observed/all data)
0.0322/0.0503
0.0695/0.0762
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|; wR2 ) [∑w(Fo2 - Fc )2/∑w(Fo )2]1/2; w )
of composition “SI11” stack parallel to the (3h02) family of
lattice planes in cubic close-packed ((...ABC...) or (...c...))
fashion, at an angle of 37.25° from the a axis.
1/σ2(Fo).
Table 2. Important Internuclear Distances (Å) in Ta4SI11
However, instead of a pattern of interstitial site occupation
by the metal atoms that generates flat parallel slabs as in
the layered M3QX7 compounds, in Ta4SI11 the occupation
of these sites defines an undulating sheet structure with a
period of 16.135 Å. Ta4SI11 can still be viewed as a “layered”
structure since the undulating Ta4SI11 layers are separated
by a van der Waals gap.
atom-atom
distance (Å)
2.998(2)
Ta1-Ta1
Ta1-Ta2
Ta1-I1
Ta1-I2
Ta1-I4
Ta1-S
Ta2-I1
Ta2-I3
Ta2-I4
Ta2-S
3.157(1)
2.769(1) ×2
2.583(2); 3.127(2)
2.892(2)
2.523(2)
2.744(1) ×2
2.780(1) ×2; 3.112(2)
2.286(1)
The nature of the distibution of tantalum atoms within the
layers is not straightforward. The structure solution gives
rise to two crystallographically distinct tantalum positions.
Full occupation of these sites would lead to butterfly Ta4
clusters that bend at the turns of the undulating anion layers.
However, the site occupation factors of both metal sites each
refine to near 50%. An additional complication is the atomic
position at the cradle or elbow of the apparent bent cluster,
seemingly bridging all four metals. This site refines as a 50:
50 sulfur-iodine split position (S/I4). Figure 4b shows a
[001] view of what the crystallographic experiment sees,
which is an average over the entire crystal. The question
now becomes how the metal atoms are distributed throughout
the structure and how the metal atom distribution correlates
with the split S/I4 position. We can attribute the various site
occupancies to structurally ordered ribbons of Ta clusters,
shown in Figure 5, with the distribution of ribbons along
[100] as well as along different layers in the layer stacking
2.324(5)
lographic package.28,29 Further data collection details and important
internuclear distances are given in Tables 1 and 2, respectively.
Results and Discussion
Figure 3 shows a 80% probability displacement ellipsoid
plot of the local Ta environments in Ta4SI11, representing
the (Ta3)9+/Ta4+ disorder model discussed below. Figure 4a
shows Ta4SI11 viewed parallel to the short b axis. The
compound can be viewed as a close-packed structure similar
to the M3QX7 compounds, but with a different metal atom
distribution. In Ta4SI11, approximately close-packed layers
(28) Sheldrick, G. M. SHELX-86; University of Go¨ttingen: Go¨ttingen,
Germany, 1986.
(29) Sheldrick, G. M. SHELXL93 Program for the Refinement of Crystal
Structures; University of Go¨ttingen: Go¨ttingen, Germany, 1993.
Inorganic Chemistry, Vol. 42, No. 13, 2003 4167

Smith and Miller
Table 3. Tantalum 4f7/2 Binding Energies (eV) Measured by X-ray
Photoelectron Spectroscopy
compound
Ta metal
binding energy (eV)
21.6-21.9
23.8
(NEt4)2[Ta6Cl12]Cl6
[Ta6Cl12]Cl2(H2O)4‚4H2O
K3TaO4
TaS2
25.8
25.9
26.7
Ta2O5
TaBr5
K2TaF7
hc-Ta3SeI7
hc-Ta3TeI7
26.7
26.9
29.4
23.4
6
6
23.4
metal center and suggests the presence of triangular clusters
having an average oxidation state of +3. The binding
energies of Ta in hc-Ta3SeI7 and hc-Ta3TeI7, which contain
Ta3 clusters, are 23.4 eV.6 For comparison, literature XPS
values27 for tantalum in other compounds are listed in Table
3. It is also possible to assign the peak at 26.3 eV to oxygen
contamination of the sample, and not to a mixed-valence
effect within the compound. However, oxygen contamination
of the crystalline samples used would be limited to surface
oxidation since all products were handled under an inert
atmosphere. To explore this possibility, samples of single
crystals were etched in the XPS instrument by sputtering
with Ar ions. Etching removes contaminated surface layers
and exposes the nonoxidized, presumably pristine interior
of the sample. After etching, the peak at 26.3 eV persisted,
which suggests this peak is characteristic of the sample, and
not due to surface oxidation. Additionally, the Ta 4f7/2 peak
of Ta2O5 appears at slightly higher binding energy (26.7 eV,
Table 3).
Figure 5. Three possible distributions of Ta atoms that may exist in Ta4-
SI11. (a) “Butterfly” clusters; (b) triangles alternating with isolated Ta atoms;
(c) herringbone pattern of Ta2 dimers.
The two binding energies suggest a mixture of Ta3
trinuclear clusters and tantalum atoms in a higher oxidation
state. It is not likely that the two peaks result from chemical
inequivalencies within the trinuclear cluster itself (local
symmetry Cs), as the 2.8-eV difference between the two is
much too large to support this contention. The distortion of
the Ta3 clusters away from perfect equilateral geometry is
moderate, with Ta-Ta distances of 3.157(2) and 2.999(2)
Å and angles of 56.70(1)° and 61.65(1)°. A formulation of
Ta4SI11 consistent with the crystallographic, magnetic, and
X-ray photoelectron spectroscopic data is “(Ta3+)3(Ta4+)-
(S2-)(I-)11”; a 1:1 mixture of Ta3 clusters and isolated Ta
atoms in the +4 oxidation state. These two metal species
must be randomly distributed throughout the three-dimen-
sional structure to give the apparent averaged structure solved
from the X-ray diffraction data. While it is possible (but, as
yet, unverifiable) that some kind of ordering of the trinuclear
clusters and isolated atoms exists within each undulating
layer, certainly no ordering or registry can exist between the
Ta4SI11 layers; otherwise, different symmetry, unit cell, or
superstructure reflections would be found. This formulation
is also bolstered by the magnetic data, where the derived
magnetic moment of 1.53 µB indicates one unpaired electron
per formula unit, close to the spin-only value for isolated d1
Ta4+ centers.
([001]) direction being completely disordered. Within a
ribbon along [010], there are three possibilities for the
distribution of Ta atoms (see Figure 5): (1) Ta4 “butterfly”
clusters linked by bridging S/I4 ions; (2) alternating Ta3
triangular clusters and isolated Ta atoms; and (3) a “her-
ringbone” pattern of Ta2 dimers. The distribution of S and
I4 atoms should be closely correlated with the patterns of
Ta clusters in possibilities (1) and (2), but less so for
possibility (3). The solution to this problem cannot be
conclusively determined based on the X-ray diffraction
experiment, which measures average electron density over
the entire crystal and cannot elucidate the precise local
situation: independent characterization methods are required.
X-ray Photoelectron Spectroscopy. To help clarify more
details about the structure, X-ray photoelectron spectroscopy
spectra were measured. The XPS spectrum of any tantalum
atom produces a doublet signal, corresponding to the 4f7/2
(lower binding energy) and 4f5/2 (higher binding energy).
These peaks are always separated by 1.91 eV. This constant
separation facilitates deconvolution of overlapping peaks due
to inequivalent atoms. The binding energies reported below
are for the Ta 4f7/2 peaks. The Ta4SI11 spectrum showed two
peaks at 23.8 and 26.3 eV with visibly different integrated
intensities (the intensity of the 23.8-eV peak is noticeably
greater than that of the 26.3-eV peak), which suggest two
different oxidation states for Ta in different concentrations.
The Ta 4f7/2 binding energy of 23.8 eV indicates a reduced
The assignment of the split sulfur-iodine position remains,
but we hypothesize an arrangement commensurate with the
pattern of trinuclear clusters given our experience with
4168 Inorganic Chemistry, Vol. 42, No. 13, 2003

Synthesis and Structure of Ta4SI11
Table 4. Atomic Orbital Parameters Used in Extended Hu¨ckel
Calculations on Different Models of Ta4SI11
atom
Ta
orbital
Hii (eV)
ú1
c1
ú2
c2
6s
6p
5d
-8.96
-4.99
-9.83
2.28
2.24
4.76
0.6104
1.94
0.6104
S
I
3s
3p
-20.00
-13.30
2.12
1.83
5s
5p
-18.00
-12.72
2.66
2.32
the [010] direction of each undulating Ta4SI11 layer, a Ta4
cluster is fully occupied, with the adjacent Ta4 set unoc-
cupied. For this model, the sulfur was positioned in the crook
of the cluster, coordinated to all four Ta atoms.
Models 2 and 3: A 50:50 mixture of Ta3 clusters and
isolated Ta atoms. There are two possible unique distributions
in this case: (Figure 6b) such that the Ta3 clusters are all
parallel throughout the structure when viewed along [010],
or (Figure 6c), the Ta3 clusters alternate orientation, following
the undulating layers.
Models 4 and 5: A “herringbone” pattern of Ta2 dimers
along [010], with each dimer bridged by a sulfur atom and
an iodine atom (Figure 6d,e).
The calculations showed a significant energetic separation
among the three broad cases: the total energies of the two
Ta3 + Ta cases (Figure 6b,c) were found to be quite similar,
differing by only 0.07 eV (per formula unit) from one
another. However, both of these cases were clearly stabilized
relative to the butterfly clusters by 0.27 eV per formula unit
and to the dimers by 1.57 eV.
Figure 6. Five different structural models of Ta atom distribution used
for extended Hu¨ckel calculations of total energies.
various M3YX7 compounds. Therefore, at each trinuclear
cluster, sulfur is the capping atom. Conversely, then, when
the single tantalum atom occurs, it is surrounded by an
octahedral coordination environment of entirely iodine. This
assignment accounts for the amount of each species present
in this structure, as well as being the most reasonable
placement of the sulfur atoms in light of the behavior of the
chalcogen in the similar compounds Ta3QX7,6 that is, always
capping a trinuclear cluster. While most Ta-I and Ta-S
bond distances are typical,5,6 there is, however, an extremely
short Ta2-I4 bond length of 2.286(1) Å, around the Ta4+
ions. However, with the disorder and fractional occupancy
present throughout the structure, such a distance could be a
false artifact forced upon the structure as a consequence of
averaging. No unusual motion is manifest in the anisotropic
displacement parameters for S or I4 (see Supporting Infor-
mation). The only noticeably large Uij value is U33 for I2,
apparently elongated into the van der Waals gap.
Extended Hu1ckel Calculations. Extended Hu¨ckel calcu-
lations were carried out on five structural models (illustrated
in Figure 6a-e) of the Ta4SI11 structure based upon the three
models enumerated above, using atomic orbital parameters
given in Table 4. The purpose of these calculations was to
determine the total energies of various possible metal atom
distributions. The unit cells for all five models utilized
doubled b axis lengths to account for the Ta positions along
the ribbons but kept the observed values of a and c axes
lengths. The five models are (view normal to the layers in
all cases with additional [010] views for 6b and 6c):
Ta4SI11 bears a structural resemblance to the novel niobium
sulfide iodide o-Nb3SI7.11 Both structures contain undulating
mixed sulfur-iodine layers, with the metal atoms defining
a similar undulating layer structure. Within both undulating
layer structures, triangular clusters are capped by a sulfur
atom, generating local M3X13-type clusters. In o-Nb3SI7, the
Nb3 cluster coordination environment involves an additional
sulfur atom; this second coordinating S provides a bridge to
another Nb3 cluster. Coordination of the Nb3 units is then
“Nb3S2I11”. In Ta4SI11, the sulfur atom is not shared between
separate clusters, and the Ta3 coordination is “Ta3SI12”.
Because of these structural similarities, and considering the
crystallographic difficulties encountered in the solutions of
both structures, one might raise the possibility of whether
Ta4SI11 might actually be “o-Ta3SI7”, isostructural with the
Nb case, but with a disorder mechanism falsely but repro-
ducibly generating two different solutions. In fact, we have
also synthesized a compound whose powder pattern is
virtually the same as that of o-Nb3SI7, though diffraction
quality single crystals have not been grown.30 Examination
of Guinier X-ray powder diffraction patterns of pure samples
of Ta4SI11 and “o-Ta3SI7” as well as the powder pattern from
the published o-Nb3SI7 structure was definitive in establishing
the uniqueness of the two structure types. It is clear that the
patterns are significantly different, reflecting the different
structures that emerge from the single-crystal structure
determinations.
Model 1: “Butterfly” Ta4 clusters, distributed over the
three-dimensional structure as shown in Figure 6a. Along
(30) Smith, M. D.; Miller, G. J. Unpublished results.
Inorganic Chemistry, Vol. 42, No. 13, 2003 4169

Smith and Miller
The available data support the contention of a mixed-
valent, disordered sulfide iodide. The problem of the short
Ta2-I4 distance remains and is a disturbing one. Other
attempts at characterizing the sample have been attempted:
Various mass spectrometry techniques (electron impact
ionization and chemical ionization in positive and negative
ion modes) have been employed, but the compound is not
sufficiently volatile to produce any vapor-phase fragments
that could illuminate structural details. The compound is not
soluble without decomposition, and therefore other ionization
techniques (electrospray, etc.) cannot be used. The potential
of an undetected superstructure exists. In particular, the split
position might be “frozen” or resolved into one position,
giving rise to an ordered structure. Examination of several
crystals at low temperature, using both a rotating anode X-ray
source and CCD area detector data, did not reveal either a
different unit cell or superstructure reflections that would
indicate a structural change from that reported herein.
Supporting Information Available: X-ray crystallographic data
in CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
IC034075Q
4170 Inorganic Chemistry, Vol. 42, No. 13, 2003