Crystal Growth & Design
Article
(vw), 918 (s), 867 (m), 563 (s). Raman, νmax, cm−1: 3105 (w), 3012
(s), 3000 (vs), 2967 (s), 2906 (m), 2844 (vw), 2787 (w), 2742 (vw),
1627 (vw), 1549 (w), 1535 (m), 1472 (m), 1462 (m), 1426 (vw),
1402 (m), 1376 (w), 1303 (vs), 1202 (s), 1150 (m), 1054 (s), 1038
(vs), 826 (m), 812 (vs), 462 (m), 438 (m), 387 (s), 320 (vw), 248
(w), 195 (vw), 141 (vw). Anal. Found: C 18.03, H 5.84, N 10.29.
C4H14Br2N2O requires: C 18.06, H 5.31, N 10.53%.
The crystals of 1 and 2, stored at room temperature, were
reinvestigated a few months after the first elemental analysis
investigations and were found to be unchanged. This suggests
that they are stable at ambient conditions over long time scales,
in contrast to the other compounds of this class.23 Further,
compound 1 was found to easily and repeatedly crystallize at
different experimental conditions involving different (i) molar
ratios of starting materials (piperazine to rhodium(III) chloride
trihydrate molar ratios of 1:10, 6:1, and 10:1), (ii)
concentrations of hydrobromic acid (ca. 9, 4.5, and 3 mol/L
HBr for the same 6:1 molar ratio of starting materials), and (iii)
method/temperature of crystallization−hydrothermal24 and
diffusion-controlled8,11 processes (for the substrates molar
ratio of 2:1). The single crystal products always show the same
dark red color and the shape of flat needles. Some very few tiny
red plates, unfortunately not suitable for the further character-
ization, e.g., by X-ray diffraction analysis, were also found (in
particular, in the process of decreasing the concentration of
HBr).
Thermogravimetric analyses show a stepwise thermal
decomposition for all four compounds (Figure S1, Supporting
Information). The thermal decomposition of 1 and 3 proceeds,
in general, in two main steps, whereas in the case of 2 and 4, in
principle, three steps were found suggesting the lower thermal
stability of the iodide-containing compounds. The first steps,
thermolyses, for 1−3, are associated with dehydration starting
at ca. 365 K for 1 and 2, and at ca. 340 K for 3. Then the loss of
HX (X = Br or I) followed by the pyrolysis of organic
components occurs. The loss of 4 and 2 equivalents of HI at ca.
585 and 520 K for 2 and 4, respectively, is indicated. This
suggests the formation, at this stage, of a relatively stable
[RhI2(C4H10N2)2]I semiproduct in the case of 2.
Structures of 1 and 2. The compounds 1 and 2 are clearly
isostructural; i.e., they crystallize in the same orthorhombic
centrosymmetric Pnma space group, their unit cell parameters
are similar, the positions of corresponding atoms are
approximately the same, and they are characterized by similar
crystal packing (Tables 1, S1 and S2, Figure 1). The high
degree of isostructurality is further confirmed by the calculated
unit cell similarity factor of 0.0438 and the isostructurality index
of 98.8%25,26 and 95.3%.27 It is interesting to note that the axial
dimensions of 2 are almost proportionally larger than those of
1; however the largest relative change (5.8%) was found for the
smallest unit cell c parameter. Examination of the molecular
packing, shown in Figure 1, indicates that the larger [RhI6]3−
ions together with the longer distances between the structural
components in 2 and a greater volume of voids17 seem to be
responsible for the larger lattice expansion in the c direction.
As stated above, the principal arrangement of the structural
components in both crystals (the [RhX6]3− octahedral
complexes, isolated X− ions, piperazine-1,4-diium cations, and
H2O molecules) that all have crystallographically imposed
mirror symmetry is similar in 1 and 2. However, the inspection
of the system of noncovalent interactions shows clear
differences between them.
4, IR, νmax, cm−1: 3406 (s, br), 3071 (s), 2999 (vs), 2985 (vs), 2968
(vs), 2946 (vs), 2926 (vs), 2795 (m), 2771 (m), 2746 (vs), 2727 (s),
2714 (s), 2682 (m), 2581 (m), 2565 (m), 2503 (w), 2484 (w), 2449
(m), 2357 (m), 2096 (w), 1928 (vw), 1742 (vw), 1539 (vs), 1451
(m), 1443 (w), 1427 (s), 1383 (s), 1373 (vs), 1315 (w), 1292 (m),
1192 (m), 1069 (m), 1046 (m), 1019 (vw), 987 (m), 916 (s), 855
(m), 556 (w). Raman, νmax, cm−1: 2999 (vw), 2984 (vw), 2947 (w),
1535 (w), 1456 (vw), 1439 (vw), 1407 (vw), 1390 (vw), 1376 (vw),
1305 (w), 1294 (w), 1132 (vw), 1040 (vw), 1032 (vw), 826 (vw), 816
(vw), 798 (vw), 456 (vw), 434 (vw), 390 (vw), 217 (w), 199 (m), 167
(vs), 139 (m). Anal. Found: C 8.14, H 2.01, N 4.59. C4H12I4N2
requires: C 8.06, H 2.03, N 4.70%.
Crystal Structure Determinations. Intensity data were collected
at 295(2) K on STADI4 CCD and Xcalibur Eos four circle area
detector single crystal diffractometers, with graphite monochromated
MoKα radiation from crystals of all four compounds. The KUMA
Diffraction Instruments and Oxford Diffraction software was used
during the data collection, unit cell parameters determinations, and
data-reduction processes. All data were corrected for Lorentz,
polarization, and absorption corrections.14,15 All the structures were
solved by the Patterson method and refined by the full-matrix least-
squares method against F2 using SHELX.16
In all of the structures the non-hydrogen atoms were refined using
anisotropic displacement parameters. All of the hydrogen atom
positions were located in subsequent difference Fourier maps. The
riding model was applied to the hydrogen atoms attached to nitrogen
and carbon atoms, in all structures, whereas the hydrogen atoms
bonded to oxygen atoms were refined using appropriate geometrical
restraints (DFIX command of SHELXL).16 The isotropic displace-
ment parameters of hydrogen atoms were taken with coefficients being
1.2 and 1.5 (for hydrogen atoms bonded to carbon, nitrogen, and
oxygen atoms, respectively) times larger than the respective
parameters of their parent atoms. Selected crystal data and the
structure determination details for all structures are listed in Table 1
(Table S1, Supporting Information). The structure drawings were
prepared using Mercury.17 The interactions were compared using the
Hirshfeld surface analysis provided by CrystalExplorer.18−21
RESULTS AND DISCUSSION
■
Syntheses and Characterization. The diffusion-con-
trolled ligand exchange crystal growth procedure, using the
same basic “parent” inorganic starting material and an organic
linker, i.e., rhodium(III) chloride trihydrate dissolved in
hydrochloric acid and solutions of piperazine in hydrobromic
and hydroiodic acid, that were additionally separated by a
“neutral” layer of the hydrohalic acid, yielded the needle-shaped
single crystals of 1 and 2, respectively. No mixed chlorido-
bromido- nor chlorido-iodiodorhodate(III) ions were found in
the crystal samples. The efficiency of ligand exchange processes
along with the purity of compounds 1 and 2 were confirmed by
both energy dispersive X-ray fluorescence analysis investigating
selected single crystals and FT-Raman spectroscopy using the
crystalline samples.
The geometry of [RhX6]3− octahedra, in both structures,
slightly deviates from the ideal octahedron reflecting the
different radii of Br and I28−30 and the differences in their
structural environment (Figure 2). It is of note that the
structural environment should be considered as a decisive
factor, especially in the case of 2, responsible for formation of
the thermodynamically stable species that tend to be lost of
heating.
The present work is the first report of the synthesis and
crystal growth of a compound containing isolated [RhI6]3−
octahedra. In contrast to the previous multistep complicated
procedure used for obtaining Cs3[IrI6],22 the liquid−liquid
diffusion controlled ligand exchange crystal growth method was
found to be very simple and efficient in terms of time, yield, and
purity of the single product of 2.
C
Cryst. Growth Des. XXXX, XXX, XXX−XXX