[UCl4(HCN)4]-a hydrogen cyanide complex of uranium tetrachloride

Article abstract of DOI:10.1039/c7cc09401a

The reaction of uranium tetrachloride with anhydrous liquid hydrogen cyanide yields a turquoise microcrystalline powder of tetrachloridotetraformonitrileuranium(iv), [UCl₄(HCN)₄]. We determined the crystal structure of this compound by powder neutron diffraction. The compound was further characterized by IR spectroscopy and thermogravimetric analysis as well as by magnetic measurements. The paramagnetic compound crystallizes in the tetragonal space group type I4. To the best of our knowledge this compound represents the first structurally elucidated uranium(iv) complex with HCN as a ligand.

Full text of DOI:10.1039/c7cc09401a

ChemComm
View Article Online
View Journal
COMMUNICATION
[UCl4(HCN)4] – a hydrogen cyanide complex of
uranium tetrachloride†
Cite this:DOI: 10.1039/c7cc09401a
ab
S. S. Rudel,ab C. Pietzonka,ab M. Hoelzelc and F. Kraus
*
Received 7th December 2017,
Accepted 2nd January 2018
DOI: 10.1039/c7cc09401a
rsc.li/chemcomm
The reaction of uranium tetrachloride with anhydrous liquid hydrogen
cyanide yields a turquoise microcrystalline powder of tetrachlorido-
tetraformonitrileuranium(IV), [UCl4(HCN)4]. We determined the crystal
structure of this compound by powder neutron diffraction. The
compound was further characterized by IR spectroscopy and
thermogravimetric analysis as well as by magnetic measurements.
The paramagnetic compound crystallizes in the tetragonal space
%
group type I4. To the best of our knowledge this compound
represents the first structurally elucidated uranium(IV) complex with
HCN as a ligand.
Fig. 1 The structure of the [UCl4(HCN)4] molecule in its solid state
compound. Isotropic displacement parameters at 70% probability at 50 K
for all atoms.
Compared to the multitude of known cyanide complexes of the
transition metals, only a few examples have been reported for
uranium, and no pseudobinary uranium cyanides have been
synthesized to date.1–8 In contrast, however, coordination com-
pounds of the free prussic acid are extremely rare in comparison,
irrespective of the central atom.9–23 We tend to ascribe this to the
difficulties that arise during handling the highly toxic hydrogen
cyanide, a liquid, with a boiling point just above room temperature
that additionally tends to polymerize. Depending on the conditions,
the latter can even lead to explosions.24,25 While cyanide salts of
monovalent cations, for example of the alkali metals, are slightly
soluble in liquid anhydrous HCN (aHCN), ionic compounds of
higher valent metals are practically insoluble in it.24,25 However,
with Lewis acidic compounds of more covalent character, like BeCl2,
BCl3, VCl4, or NbCl5, hydrogen cyanide acts as a Lewis base and
forms adducts. These are supposedly dissolved in the form of the
respective molecular formonitrile complexes.13,16,22,23 The sub-
limeable TiCl4ꢀ2HCN, which was the first described HCN
adduct,9 was reported to be partially reduced to Ti(III) by
warming to 60 1C in liquid aHCN,19 and to form TiN upon
pyrolysis of the gaseous compound.26 Stable HCN adducts thus
might be promising starting compounds for hard or refractory
nitride coatings.
Bagnall and coworkers observed a reaction of UCl4 with
liquid aHCN in the context of their research on uranium
cyanides. Besides the position of the IR bands of the obtained
compound, only its decomposition to UCl4 and HCN under
vacuum was reported.2 These authors synthesized the HCN
adduct according to eqn (1) in prospect of the synthesis of
uranium cyanides by the elimination of hydrogen chloride from
the intermediate compound by addition of a base (eqn (2)), in
which they were unsuccessful.
UCl4 + xHCN - [UCl4(HCN)x]
(1)
[UCl4(HCN)x] + B - [UCl4ꢁx(CN)x] + HB+ + Clꢁ (B = Base)
a
¨
Fachbereich Chemie, Philipps-Universitat Marburg, Hans-Meerweinstraße 4,
(2)
35032 Marburg, Germany. E-mail: florian.kraus@chemie.uni-marburg.de;
Web: https://www.uni-marburg.de/fb15/ag-kraus
b WZMW, Hans-Meerwein-Straße, 35032 Marburg, Germany
However, when they reacted UCl4 with NH4CN in liquid
ammonia, they were able to substitute one chloride ion by a
cyanide ion, forming [UCl3(CN)]ꢀ4NH3. Having reported on the
HCN adduct of UF5 recently,27 we wondered what the nature of
the HCN adduct of UCl4 should be.
c Forschungs-Neutronenquelle Heinz Maier-Leibnitz, Lichtenbergstraße 1,
85747 Garching, Germany. E-mail: markus.hoelzel@frm2.tum.de
† Electronic supplementary information (ESI) available: For experimental proce-
dures and details on analytical data, see the supporting information.39–44 CSD
433858. See DOI: 10.1039/c7cc09401a
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 Assignment of IR bands of [UCl4(HCN)4], solid,34 and gaseous
After sublimation of aHCN onto a sample of powdered grass-
green UCl4, a turquoise colored powder under a colorless liquid
was obtained after melting of the HCN. Removing the excess of
HCN after one week under cooling led to a highly moisture
sensitive blueish green powder of the composition [UCl4(HCN)4].
HCN.35 Numbering according to the literature35
[UCl4(HCN)4]
HCN(gas)
HCN(solid)
n~/cmꢁ1
n~/cmꢁ1
~n/cmꢁ1
Assignment
3159
3311
2097
712
3132
2097
838
n3: n(C–H)
The compound is relatively stable at room temperature, if free 2119
n1: n(CRN)
n2: d(H–CRN)
n2: d(H–CRN)
785
762
from residual HCN.
828
After two weeks of storage at 2 1C under liquid HCN we
observed signs of decomposition of the solvent. Namely, a
brown tint occurred and flocculation of a brown substance
started. So, neither crystallization in HCN nor growing crystals expected. The U–Cl distance is observed to be 2.623(9) Å, the
by application of heat was successful, because of the thermal U–N distance to be 2.603(8) Å. Both are in the range of atomic
instability of [UCl4(HCN)4]. The powder X-ray diffraction pattern distances reported for other compounds of bisphenoidal shape,
obtained at room temperature could be indexed in the tetragonal such as the analogous acetonitrile complex [UCl4(CH3CN)4]
crystal system and the unit cell parameters could be determined. (U–Cl: 2.620(4)–2.627(4),28 resp. 2.614(2)–2.624(2) Å,29 U–N:
Mass spectrometry coupled thermal analysis indicated a mass 2.577(14)–2.601(13),28 resp. 2.567(6)–2.599(6) Å29) which however
loss of four equivalents of HCN per UCl4. The IR spectra (see does not crystallize isotypically, or in the bipyridine complex
below) indicated U–N bonding as the H–C stretch was observed. [UCl4(Me2bpy)2] (U–Cl: 2.622(2)–2.641(1) Å, U–N: 2.617(3)–
A pycnometric density measurement led to approximately 2.674(3) Å).30 As opposed to these reports the U–N distance is
r = 2.76 g cmꢁ3 and thus indicated Z = 2. The extinction longer in the amine complex, [UCl4(1,2-bis(dimethylamino)-
conditions only pointed to I centering. So, the structure could ethane)2], which is likely due to steric reasons (U–Cl: 2.600(4)–
be solved by crystallographic reasoning. However, only the 2.617(3) Å, U–N: 2.769(9)–2.810(8) Å).31 The molecular structure
uranium atom could be refined using the powder X-ray data. is quite similar as well to the nitrile adducts reported in the
The position of the C, Cl, and N atoms could not be located as literature.45,46 Finally, the uranium(III) complex [UCl3(pyridine)4]2
well as the H atom positions. Using powder neutron diffraction shows slightly longer U–N as well as U–Clterminal atomic distances
data (Fig. 2), the structure solution and refinement succeeded (U–Cl: 2.734(2)–2.768(2) Å, U–N: 2.690(8)–2.772(7) Å) as may be
in the space group I4 with cell parameters a = 9.0137(2) Å, expected.32 The N–C bond length observed by us is 1.13(1) Å, the
%
c = 7.4037(3) Å, V = 601.53(2) Å3, and Z = 2 at T = 295 K, and C–H bond length is 1.05(3) Å. Both bond lengths are the same as
a = 8.9122(4) Å, c = 7.3255(7) Å, and V = 581.84(7) Å3 at T = 50 K. those of pure HCN within the three sigma criterion when compared
All atoms could be located and their displacement parameters to atomic distances obtained from the microwave spectroscopic
refined isotropically.
data of gaseous HCN (C–N: 1.156(1) Å, C–H: 1.064(4) Å),33 as well as
The uranium atom is coordinated by four Cl atoms and by for estimations of bond lengths for solid HCN with reference to
four HCN molecules in a bisphenoidal shape (Fig. 1). The HCN X-ray and IR spectroscopic data (C–H: 1.07(2) Å, C–N: 1.160(5) Å).21,34
molecules coordinate via the N atom to the U atom, as This observation is in accordance with other HCN coordination
Fig. 2 Neutron diffraction pattern of [UCl4(HCN)4] at 50 K: experimental data (black crosses), calculated Rietveld profile (red) and difference profile
(bottom, black). The calculated reflection positions are shown as black ticks. The range from 0 to ca. 71 2y has been excluded from the refinement.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018

View Article Online
ChemComm
Communication
0
Fig. 3 Molar susceptibility and effective magnetic moment of [UCl4(HCN)4 ].
compounds like [TiCl4(HCN)2] (C–N: 1.11(1) Å),10 [NbCl5(HCN)]
(C–N: 1.090(4) Å),22 and [VCl4(NCH)2] (C–N: 1.12(1)–1.14(1) Å, C–H:
1.11(9) Å).23 In the IR spectrum of the title compound (Fig. S1, ESI†)
we observed bands at 3159, 2119, 785, and 762 cmꢁ1 which
correspond well to the reported ones (3150, 2115, 785 and
760 cmꢁ1).2 For their assignment see Table 1. The bands are
only slightly shifted compared to gaseous HCN (3311, 2097, and
712 cmꢁ1).35 We undertook an attempt to obtain Raman
spectra but despite several attenuations of the laser beam only
sample decomposition was observed.
Fig. 4 Temperature dependent scan at 0.1 T and fitting (red).
The resulting values are 3.82 BM f.u.ꢁ1 for the effective
magnetic moment (neff), and ꢁ181 K for the Weiss constant.
The Weiss constant does not indicate an antiferromagnetic
interaction between the uranium(IV) ions. It can be interpreted as
a parameter that compensates for the overestimated neff value.
The temperature-independent part as well as the deviation
of the susceptibility from the Curie-law gives reason to a more
complex view. For several uranium(IV) compounds similar
magnetic properties were reported.36–38 An admixture of excited
states to the 3H4 ground-state depends naturally on the influence
of the ligands due to the orbital overlap.47,48 These phenomena
seem to be responsible for the observed behavior of the compound.
In conclusion, [UCl4(HCN)4], which is only the second example
of a hydrogen cyanide complex of a uranium compound, has been
synthesized and structurally characterized by a combination of
powder neutron and X-ray diffraction, as well as IR spectroscopic,
magnetic and thermal analyses. The paramagnetic compound
As reported,2 HCN can be removed from [UCl4(HCN)4] by the
application of fine vacuum whereby pure UCl4 is recovered. At
atmospheric pressure, thermal decomposition occurs significantly
between 70 and 150 1C. In this range, a mass loss of approximately
19% (22.16% in theory) was detected that was assigned to HCN
via simultaneous mass spectrometry of the liberated gas. The
deviation between the observed and calculated mass loss is
ascribed to the transfer of the sample into the thermogravi-
metric analysis instrument in air and subsequent evacuation
and purging of the device. This handling may cause partial
hydrolysis and loss of some HCN in the vacuum.
A color change to a more greenish color was visible after
storing the compound in a vanadium cuvette for one year.
Powder X-ray diffraction hints to the presence of a tiny amount
of UCl4. We attribute this to the decomposition of the compound
due to its slight HCN vapor pressure even at room temperature.
In closed vessels, HCN would be continuously removed from the
gas phase due to its polymerization.
%
crystallizes in the tetragonal space group I4. It loses HCN upon
heating to 100 1C or in a vacuum at room temperature, whereby
UCl4 is recovered. Such compounds may be useful for the chemical
vapor deposition of nitride films and potentially for the formation
of uranium nitrides. Investigations are ongoing.
We thank Dr M. Fichtl for TG-MS measurements and our
Master’s student J. Preinl for synthetic assistance. F. K. and
S. R. thank M. Karppinen and O. Mustonen for additional
SQUID measurements. F. K. thanks the DFG for a Heisenberg-
Investigation of the magnetic properties of the compound
with a physical property measurement system shows an almost
temperature independent susceptibility in the range from 1.8 K
to 80 K. Above this range it changes to a Curie-like paramagnetic
behavior. Fig. 3 shows the molar susceptibility as well as the
effective magnetic moment in number of Bohr magnetons per
¨
professorship. We thank Prof. H. Kohlmann, Universitat
formula unit. The value at 300 K is 3.0 BM f.u.ꢁ1
.
ˇ´ˇ
Leipzig, for helpful discussions. We thank Dr Petrıcek for the
quick removal of bugs from the Jana2006 software.
Fig. 4 shows the data of the temperature dependent scan in
different representations as well as the corresponding Curie–
Weiss-fitting curves. The fitting results should be regarded as a
rough approximation. The Curie–Weiss fits were performed in a
range from 200 to 300 K as we assume that the magnetic properties
agree better with a Curie-like behavior in this temperature range.
Conflicts of interest
None to declare.
This journal is ©The Royal Society of Chemistry 2018
Chem. Commun.

View Article Online
Communication
ChemComm
Chemische Gesellschaft, Kohlenstoff-Stickstoff-Verbindungen, Verlag
Chemie, Berlin, 1971.
26 G. Jander and B. Gru¨ttner, Chem. Ber., 1948, 81, 107.
27 B. Scheibe, S. S. Rudel, M. R. Buchner, A. J. Karttunen and F. Kraus,
Chem. – Eur. J., 2017, 23, 291.
28 F. A. Cotton, D. O. Marler and W. Schwotzer, Acta Crystallogr., 1984,
C40, 1186.
29 G. Van Den Bossche, J. Rebizant, M. R. Spirlet and J. Goffart, Acta
Crystallogr., 1986, C42, 1478.
30 R. E. Jilek, N. C. Tomson, R. L. Shook, B. L. Scott and J. M. Boncella,
Inorg. Chem., 2014, 53, 9818.
31 A. Zalkin, P. G. Edwards, D. Zhang and R. A. Andersen, Acta
Crystallogr., 1986, C42, 1480.
32 H. S. La Pierre, F. W. Heinemann and K. Meyer, Chem. Commun.,
2014, 50, 3962.
33 J. W. Simmons, W. E. Anderson and W. Gordy, Phys. Rev., 1950,
77, 77. Erratum: Phys. Rev., 1952, 86, 1056.
34 R. E. Hoffman and D. F. Hornig, J. Chem. Phys., 1949, 17, 1163.
35 H. C. Allen, E. D. Tidwell and E. K. Plyler, J. Chem. Phys., 1956,
25, 302.
36 E. J. Schelter, P. Yang, B. L. Scott, J. D. Thompson, R. L. Martin, P. J. Hay,
D. E. Morris and J. L. Kiplinger, Inorg. Chem., 2007, 46, 7477.
37 A. Bronova, T. Droß, R. Glaum, H. Lueken, M. Speldrich and
W. Urland, Inorg. Chem., 2016, 55, 6848.
Notes and references
1 B. Kanellakopulos, E. Dornberger and H. Billich, J. Organomet.
Chem., 1974, 76, C42.
2 K. W. Bagnall and J. L. Baptista, J. Inorg. Nucl. Chem., 1970, 32, 2283.
´
3 J.-C. Berthet, P. Thuery and M. Ephritikhine, Chem. Commun., 2007, 606.
´
´
4 J. Maynadie, J.-C. Berthet, P. Thuery and M. Ephritikhine, Organo-
metallics, 2007, 26, 2623.
5 W. J. Evans, T. J. Mueller and J. W. Ziller, Chem. – Eur. J., 2010, 16, 964.
´
´
6 A. Herve, Y. Bouzidi, J.-C. Berthet, L. Belkhiri, P. Thuery,
A. Boucekkine and M. Ephritikhine, Inorg. Chem., 2014, 53, 6995.
´
´
7 A. Herve, Y. Bouzidi, J.-C. Berthet, L. Belkhiri, P. Thuery, A. Boucekkine
and M. Ephritikhine, Inorg. Chem., 2015, 54, 2474.
´
8 J.-C. Berthet, P. Thuery and M. Ephritikhine, Dalton Trans., 2015,
44, 7727.
¨
9 F. Wohler, Ann. Chem. Pharm., 1850, 73, 226.
10 G. Constant, J. C. Daran and Y. Jeannin, Acta Crystallogr., 1971, B27, 2388.
11 B. Corain, Coord. Chem. Rev., 1982, 47, 165.
12 A. A. A. Emara and G. J. Schrobilgen, Inorg. Chem., 1992, 31, 1323.
13 R. Fricke and O. Rode, Z. Anorg. Allg. Chem., 1926, 152, 347.
¨
14 I. C. Tornieporth-Oetting, T. M. Klapotke, T. S. Cameron, J. Valkonen,
P. Rademacher and K. Kowski, J. Chem. Soc., Dalton Trans., 1992, 0, 537.
15 W. A. Burns and K. R. Leopold, J. Am. Chem. Soc., 1993, 115, 11622.
16 K. Kawai and I. Kanesaka, Spectrochim. Acta, Part A, 1969, 25, 1265.
17 I. Kanesaka, Y. Hase and K. Kawai, Bull. Chem. Soc. Jpn., 1972,
45, 1595.
38 A. Bronova, T. Bredow, R. Glaum and W. Urland, Inorg. Chem., 2016,
55, 6853.
18 G. Constant, J. C. Daran and Y. Jeannin, J. Solid State Chem., 1970,
2, 421.
19 G. Constant, J.-C. Daran, Y. Jeannin and B. Saperas, C. R. Acad. Sci.,
Paris, 1969, 269, 975.
20 G. Brauer and H. Walz, Z. Anorg. Allg. Chem., 1963, 319, 236.
21 W. J. Dulmage and W. N. Lipscomb, Acta Crystallogr., 1951, 4, 330.
22 C. Chavant, G. Constant, Y. Jeannin and R. Morancho, Acta Crystal-
logr., 1975, B31, 1823.
23 G. Constant, J.-C. Daran, Y. Jeannin and R. Morancho, J. Coord.
Chem., 1973, 2, 303.
39 http://gestis-en.itrust.de/, search Entry: ‘‘hydrogen cyanide’’, 29.03.2017.
40 S. S. Rudel and F. Kraus, J. Chem. Soc., Dalton Trans., 2017, 46, 5835.
41 K. H. Slotta, Ber. Dtsch. Chem. Ges. B, 1934, 67, 1028.
42 V. Petricek, M. Dusek and L. Palatinus, Jana 2006 – The crystal-
lographic computing system, Institute of Physics, Praha, Czeck Repub-
lic, 2017.
43 M. Hoelzel, A. Senvshyn, N. Juenke, H. Boysen, W. Schmahl and
H. Fuess, Nucl. Instrum. Methods Phys. Res., Sect. A, 2012, 667, 32.
44 OPUS, Bruker Optik GmbH, Ettlingen, Germany, 2012.
´
45 J.-C. Berthet, G. Siffredi, P. Thuery and M. Ephritikhine, Eur. J. Inorg.
24 B. Gru¨ttner, in Inorganic Chemistry in Liquid Hydrogen Cyanide and Liquid
Hydrogen Fluoride, ed. G. Jander, H. Spandau and C. C. Addison,
Akademie-Verlag, Berlin, 1971, vol. 2, part 1, pp. 3–118.
Chem., 2007, 4017.
46 A. E. Enriquez, B. L. Scott and M. P. Neu, Inorg. Chem., 2005,
44, 7403.
47 D. R. Kindra and W. J. Evans, Chem. Rev., 2014, 114, 8865.
48 S. T. Liddle, Angew. Chem., Int. Ed., 2015, 54, 8604.
¨
25 Gmelin-Institut fur Anorganische Chemie und Grenzgebiete, ed.
D. Koschel, L. Gmelin, R. J. Meyer, E. H. E. Pietsch and E. Fluck, Deutsche
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2018