The Existence of Tricyanomethane

Article abstract of DOI:10.1002/anie.201506753

Calcium tricyanomethanide reacts with hydrogen fluoride under formation of tricyanomethane and Ca(HF₂)₂. Tricyanomethane is stable below -40 °C and was characterized by IR, Raman, and NMR spectroscopy. The vibrational spectra were compared to the quantum-chemical frequencies at the PBE1PBE/6-311G(3df,3dp) level of theory and confirm the predicted C_3v symmetry of the molecule with regular C-H (109.8pm), C-C (146.7pm), and C-N (114.7pm) bonds.

Full text of DOI:10.1002/anie.201506753

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201506753
German Edition: DOI: 10.1002/ange.201506753
Cyanoform
The Existence of Tricyanomethane
Theresa Soltner, Jonas Häusler, and Andreas J. Kornath*
Dedicated to Professor Rolf Minkwitz on the occasion of his 75th birthday
Abstract: Calcium tricyanomethanide reacts with hydrogen
fluoride under formation of tricyanomethane and Ca(HF2)2.
Tricyanomethane is stable below À408C and was characterized
by IR, Raman, and NMR spectroscopy. The vibrational spectra
were compared to the quantum-chemical frequencies at the
PBE1PBE/6-311G(3df,3dp) level of theory and confirm the
Under strictly anhydrous conditions, Ca(C(CN)3)2 dis-
solves in hydrogen fluoride at À508C under formation of
a colorless homogenous solution. The NMR spectra provided
a first evidence for the formation of 1 according to Eq. (1):
CaðCðCNÞ3Þ2 þ 4 HF ! 2 HCðCNÞ3 þ CaðHF2Þ2
ð1Þ
À
predicted C3v symmetry of the molecule with regular C H
The 13C NMR spectrum displays a singlet at 106.1 ppm,
typical for the cyanide group, and a doublet at 16.9 ppm,
which is due to coupling with the proton (1J(C,H) = 147 Hz).
À
ꢀ
(109.8 pm), C C (146.7 pm), and C N (114.7 pm) bonds.
T
ricyanomethane (cyanoform; 1) is a textbook example of
1
The H NMR spectrum shows a singlet at 5.79 ppm for the
one of the strongest organic acids (pKa = À5.1 in water), but
the molecule has previously only been identified by micro-
wave spectroscopy in the gas phase at very low pressures.[1–3]
The isolation of 1 was first attempted more than one century
ago by Schmidtmann, who used the salts of the corresponding
base, (NC)3CÀ, and sulfuric acid as the starting materials.[4]
Since then, numerous attempts to isolate 1 have been
reported, but none of them were successful. These were
well described by Dunitz et al., who reinvestigated most of
these attempts.[5] It has been assumed that the corresponding
acid of tricyanomethanide might be the tautomeric dicyano-
ketenimine (2; Figure 1).
acidic proton of 1, and the 14N NMR spectrum displays only
a broad signal at À127.1 ppm for the cyanide groups.
À
Furthermore, the 19F NMR spectrum confirmed the HF2
formation by the typical resonance at À150 ppm.[7]
Further evidence for the formation of 1 is given by the
vibrational spectra of the colorless 1/Ca(HF2)2 mixture that
was obtained by removal of HF at À788C. The experimental
Raman frequencies of 1, those of deuterated DC(CN)3 (1a),
and the quantum-chemically calculated frequencies together
with their assignments are summarized in Table 1. The
Raman spectra are shown in Figure 2.
For the assignment of the vibrational modes of 1, C3v
symmetry was assumed in agreement with the quantum-
chemical calculations. Therefore, the molecule should exhibit
18 fundamental vibrations (Gvib(C3v) = 5A1 + A2 + 6E);
whereas the A1 and E modes should be Raman and IR
active, the A2 mode is inactive in both spectroscopic methods.
The presence of Ca(HF2)2 does not cause any difficulties, as
only one vibration of the HF2À anion (at ca. 600 cmÀ1) is active
Figure 1. Tautomers of cyanoform.
Table 1: Observed and calculated Raman frequencies [cmÀ1] for
HC(CN)3 (1) and DC(CN)3 (1a) at À1968C.
Quantum-chemical calculations have shown that 1 is more
stable than 2 by 7.4 kJmolÀ1,[6] but the stabilities could change
in the condensed phase owing to strong hydrogen bonds, and
the presence of 2 would thereby explain the remarkable
reactivity of cyanoform. The previously reported unsuccess-
ful syntheses prompted us to attempt the isolation of either
1 or 2 from Ca(C(CN)3)2 in hydrogen fluoride at low
temperatures.
HC(CN)3 (1)
Raman
DC(CN)3 (1a)
Raman
Assignment
X=H, D
calcd[a]
calcd[a]
2885 (38)
2922 (57)
2126 (66)
2145 (30)
n(CX) (A1)
2287 (100) 2316 (100) 2286(100) 2316 (100) ns(CN) (A1)
2259(7)
1253 (5)
1022 (7)
825 (24)
575 (7)
2310 (40)
1232 (4)
1002 (2)
813 (4)
559 (2)
556 (3)
345 (2)
160 (1)
126 (4)
2319 (13)
849 (15)
1116 (6)
806 (27)
570 (15)
561 (7)
2310 (42)
831 (0.1)
1098 (2)
794 (5)
559 (2)
556 (3)
336 (2)
159 (1)
126 (4)
nas(CN) (E)
d(CCX) (E)
nas(CC) (E)
ns(CC) (A1)
d(CCC) (E)
d(CCN) (A1)
d(CCN) (E)
d(C(CN)3) (A1)
d(C(CN)3) (E)
567 (16)
347 (45)
[*] Dr. T. Soltner, M. Sc. J. Häusler, Prof. Dr. A. J. Kornath
Department of Chemistry
339 (67)
Ludwig Maximilian University Munich
Butenandtstrasse 5–13, 81377 Munich (Germany)
E-mail: andreas.kornath@cup.uni-muenchen.de
[a] Calculated at the PBE1PBE/6-311G(3df,3dp) level of theory. Fre-
quencies were scaled by an empirical factor of 0.96. The relative Raman
activities are given in %.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201506753.
Angew. Chem. Int. Ed. 2015, 54, 13775 –13776
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13775

.
Angewandte
Communications
an excess of anhydrous hydrogen fluoride (3.00 g) was condensed by
cooling to À1968C. The reactor was warmed up to À508C for
approximately 10 min until the Ca(C(CN)3)2 had completely dis-
solved. The colorless solution was then cooled to À788C (dry ice), and
the excess hydrogen fluoride was removed in dynamic vacuum over
night. The remaining colorless microcrystalline product (300 mg) was
a 2:1 mixture of HC(CN)3 and Ca(HF2)2. It decomposed above
À408C with a color change from colorless to yellow to red. The
deuterated DC(CN)3 was prepared analogously with DF instead of
HF. 1H NMR (400 MHz, [D6]acetone, À458C, TMS): d = 5.79 ppm
(s). 13C NMR (100.6 MHz, [D6]acetone, À458C, TMS): d = 16.9 (d,
1J(C,H) = 147 Hz, CH), 106.1 ppm (s, CN). 14N NMR (28.9 MHz,
[D6]acetone, À458C, nitromethane): d = À127.1 ppm. IR (neat,
~
À1208C): n ¼2915 (s), 2886 (m), 2310 (vw), 2212 (vw), 1349 (w),
1248 (vw), 1025 (s), 1006 (m), 918 (w), 829 (s), 569 cmÀ1 (vs).
Acknowledgements
Figure 2. Raman spectra of a) HC(CN)3 (1) and b) DC(CN)3 (1a).
We are grateful to Prof. Jack D. Dunitz and Dr. Yi-Lin Wu for
a sample of sodium tricyanomethanide and helpful discus-
sions. Financial support of this work by the Ludwig Max-
imilian University of Munich (LMU) is gratefully acknowl-
edged.
in the Raman spectrum and displays a very low line
intensity.[8] The C H, C N, and C C stretching vibrations
À
ꢀ
À
of 1 are seen in their typical regions and confirm the structure
=
=
of 1, whereas the absence of C C and N N stretching modes
(both usually of high Raman intensity) excludes the forma-
tion of 2.[8] The isotopic H/D shift from 2885 to 2145 cmÀ1
corroborates the CH stretching vibration. Overall, the
experimental and calculated frequencies in Table 1 agree
fairly well, especially considering that the calculated frequen-
cies do not take interactions between molecules into account.
The calculated gas-phase structure of 1 at the PBE1PBE/6-
311G(3df,3dp) level of theory is comparable to that pre-
viously determined at the MP2/6-311 ++ G(2d,2p) level.[9–12]
Keywords: cyanoform · hydrogen fluoride · quantum chemistry ·
tricyanomethane · vibrational spectroscopy
How to cite: Angew. Chem. Int. Ed. 2015, 54, 13775–13776
Angew. Chem. 2015, 127, 13979–13980
[1] R. H. Boyd, J. Phys. Chem. 1963, 67, 737 – 744.
[2] E. Raamat, K. Kaupmees, G. Ovsjannikov, A. Trummal, A. Kütt,
J. Saame, I. Koppel, I. Kaljurand, L. Lipping, T. Rodima, V. Pihl,
I. A. Koppel, I. Leito, J. Phys. Org. Chem. 2013, 26, 162 – 170.
[3] B. Bak, H. Scanholt, J. Mol. Struct. 1977, 37, 153 – 156.
À
À
The calculations predict regular C H (109.8 pm), C C
ꢀ
[4] H. Schmidtmann, Ber. Dtsch. Chem. Ges. 1896, 29, 1168 – 1175.
(146.7 pm), and C N (114.7 pm) bonds with the expected
ˇ
[13]
ˇ
[5] D. Sisak, L. B. McCusker, A. Buckl, G. Wuitschik, Y.-L. Wu,
À ꢀ
linear C C N arrangement and a C-C-C angle of 111.48.
W. B. Schweizer, J. D. Dunitz, Chem. Eur. J. 2010, 16, 7224 –
7230.
[6] S. S. Stoyanov, J. A. Ysenov, I. G. Binev, Asian Chem. Lett. 2009,
13, 65 – 72.
The separation of 1 from the by-product Ca(HF2)2 was
found to be difficult and has not yet been achieved because
1 is only stable below À408C and very moisture-sensitive.
Furthermore, both 1 and Ca(HF2)2 are soluble in hydrogen
fluoride. Attempts to extract 1 with diethyl ether at low
temperatures led to red solutions, which are reminiscent of
the decomposition products already observed during the
unsuccessful attempts to prepare 1 as reported in literature.[5]
The vapor pressure of 1 below À408C is not sufficient for
a high-vacuum sublimation. Even though the purification of
1 remains to be a significant challenge, its existence in the
condensed phase has been confirmed. Considering the
numerous previously reported attempts to prepare 1, we
attribute the success of our strategy mainly to the low reaction
temperature and the choice of hydrogen fluoride as a suitable
acid and solvent.
[7] K. O. Christe, J. Fluorine Chem. 1990, 46, 339 – 342.
[8] K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Vol. A, Wiley, Hoboken, 2009.
[9] Gaussian09, revision a.02: M. J. Frisch et al., Gaussian Inc.,
Pittsburgh (PA, USA), 2003.
[10] a) T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. v. R. Schleyer,
J. Comput. Chem. 1983, 4, 294 – 301; b) R. Krishnan, J. S.
Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980, 72, 650 –
654; c) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72,
5639 – 5648; d) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev.
Lett. 1996, 77, 3865 – 3868.
[11] P. Csµszµr, P. Pulay, J. Mol. Struct. 1984, 114, 31 – 34.
[12] H. Brand, J. F. Lieb, A. Schulz, Eur. J. Org. Chem. 2008, 4665 –
4675.
[13] A. Bondi, J. Phys. Chem. 1964, 68, 441 – 451.
[14] S. Trofimenko, E. L. Little, J. Org. Chem. 1963, 28, 217 – 218.
Experimental Section
Ca(C(CN)3)2 (220 mg, 1 mmol), which was prepared according to
a literature method,[14] was placed into a reactor (FEP tube), and then
Received: July 21, 2015
Published online: September 18, 2015
13776 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 13775 –13776