Does Diisooxocyan (OCN-NCO) Exist?

Full text of DOI:10.1021/ic960049z

Inorg. Chem. 1996, 35, 4791-4793
4791
isooxocyanogen, diisooxocyan). It is well established that
AgOCN reacts with Br2 at temperatures as high as 200 °C to
form bromine isocyanate, BrNCO, in high yield.8 Carbonyl
diisocyanate, CO(NCO)2, is a known side product of this
reaction.8d,9 In the present study, we reacted a large excess of
freshly prepared and thoroughly dried (at 150 °C) AgOCN with
Br2 (distilled and dried over P4O10) at 18 °C in a gas cell which
was especially designed for this experiment.10 All absorptions
in the gas IR spectrum could undoubtedly assigned to carbonyl
diisocyanate as the only stable IR-active product in the gas phase
(no remaining BrNCO).9 Careful analysis of the nonvolatile
solid revealed, besides unreacted AgOCN, the presence of AgBr
and elemental silver. In another experiment, we repeated the
reaction of AgOCN with Br2 in a stainless steel-glass vacuum
line and isolated (volatile at -115 °C) and identified elemental
nitrogen as a further product. The formation of N2 (and not
O2!) was unequivocally established by gas discharge and mass
spectrometry.11 However, traces of CO and of CO2 were also
detected by mass spectrometry. The volumetric analysis
revealed that 3-4 equiv of bromine (Br2) generates 1 equiv of
nitrogen (N2). Therefore, the overall stoichiometry of the
reaction can be approximated according to eq 1. According to
Does Diisooxocyan (OCN-NCO) Exist?
Axel Schulz and Thomas M. Klapo1tke*
Department of Chemistry, University of Glasgow,
Glasgow G12 8QQ, U.K.
ReceiVed January 17, 1996
Introduction
The linear cyanate ion [NCO]- can be regarded as a
pseudohalide ion, and covalently bound species are known
which contain the NCO unit coordinated either via oxygen (H-
OCN, cyanic acid)1 or via nitrogen (H-NCO, isocyanic
acid).1ac,2 However, most of the known NCO compounds exist
in the iso form (R-NdCdO, R ) halogen, alkyl, aryl).3 (N.B.
There are also many trimeric organic OCN derivatives, e.g.
triazines and cyanuric acid.3) The structurally related fulminate
ion [CNO]- is also known (AgONC, silver fulminate; HCNO,
fulminic acid),1c,2d,4 and there are also theoretical and experi-
mental reports on the corresponding dipseudohalogen (ONC-
CNO, cyanogen N,N′-dioxide).5 The existence of dipseudohalo-
gen species containing cyanate units has also been postulated
by several authors;6 however, there are no experimental proofs
for a compound of the type NtC-O-O-CtN (1, oxocyano-
gen, dioxocyan). In 1980, Delgado and Fernandez reported the
formation of oxocyanogen (1) as an intermediate species in the
reaction of AgOCN with Br2.7 In that paper, the authors claimed
to have stabilized “(OCN)2” by subsequent reaction with TiCl4,
resulting in the formation of a polymeric compound of the type
{TiCl4(OCN)2}n.7 However, no direct evidence for the forma-
tion of (OCN)2 was found.
24AgOCN + 11Br2 f
22AgBr + 2Ag + 8OC(NCO)2 + 4N2 (1)
eq 1 2.75 equiv Br2 corresponds to 1 equiv of N2. Experimen-
tally it was found that 3-4 equiv of Br2 generated 1 equiv of
N2. This may indicate either very crude volumetric data or that
eq 1 just gives an approximate overall stoichiometry. Elemental
silver may be formed by reaction of OCN• radicals with AgOCN
(eq 2). This reaction was estimated to be thermodynamically
AgOCN(s) + OCN•(g) f
Results and Discussion
We have now carried out a combined theoretical and
experimental study and investigated the reaction behavior of
AgOCN with Br2 both in solution and in a neat reaction at room
temperature. In this note, we want to report direct evidence
for the intermediate formation of OdCdN-NdCdO (2,
Ag(s) + OdCdN-NdCdO(g) (2)
2
highly favorable: ∆H°(2) ) -53.7 kcal mol-1 12
The formation
.
of AgBr can easily be explained by reaction of AgOCN with
BrNCO (eq 3), which is thermodynamically allowed by ∆H°(3)
(1) (a) Defrees, D. J.; Loew, G. H.; McLean, A. D. Astrophys. J. 1982,
254, 405. (b) Yokoyama, K.; Takane, S.; Fueno, T. Bull. Chem. Soc.
Jpn. 1991, 64, 2230. (c) Pinnavaia, N.; Bramley, M. J.; Su, M. D.;
Green, W. H.; Handy, N. C. J. Mol. Phys. 1993, 78, 319. (d) East, A.
L. L.; Johnson, C. S.; Allen, W. D. J. Chem. Phys. 1993, 98, 1299.
(e) Blanch, R. J.; McCluskey, A. Chem. Phys. Lett. 1995, 241, 116.
(2) (a) Breulet, J.; Lievin, J. Theor. Chim. Acta 1982, 61, 59. (b) Fusina,
L.; Carlotti, M.; Carli, B. Can. J. Phys. 1984, 62, 1452. (c) Spiglanin,
T. A.; Chandler, D. W. J. Chem. Phys. 1987, 87, 1577. (d) Hop, C. E.
C. A.; Vandenberg, K. J.; Holmes, J. L.; Terlouw, J. K. J. Am. Chem.
Soc. 1989, 111, 72. (e) Yamada, K. M. T.; Winnewisser, M.; Johns,
J. W. C. J. Mol. Spectrosc. 1990, 140, 353. (f) Ruscic, B.; Berkowitz,
J. J. Chem. Phys. 1994, 100, 4498. (g) Boyce, C. W.; Gillies, C. W.;
Warner, H.; Gillies, J. Z.; Lovas, F. J.; Suenram, R. D. J. Mol.
Spectrosc. 1995, 171, 533. (h) Brown, S. S.; Berghout, H. J.; Crim,
F. F. J. Chem. Phys. 1995, 102, 8440.
AgOCN(s) + BrNCO(g) f AgBr(s) + 2
(3)
) -11.2 kcal mol-1 16
.
We do stress, however, that reaction 2
(8) (a) Gottardi, W. Angew. Chem. 1971, 83, 445; Angew. Chem., Int.
Ed. Engl. 1971, 10, 416. (b) Gottardi, W. Monatsh. Chem. 1972, 103,
1150. (c) Frost, D. C.; MacDonald, C. B.; McDowell, C. A.;
Westwood, N. P. C. Chem. Phys. 1980, 47, 111. (d) Devore, T. C. J.
Mol. Spectrosc. 1987, 162, 287. (e) Gerke, M.; Schatte, G.; Willner,
H. J. Mol. Spectrosc. 1989, 135, 359.
(9) Balfour, W. J.; Fougere, S. G.; Klapstein, D.; Nau, W. M. Spectrochim.
Acta 1994, 50A, 1039.
(10) Gas IR (10 cm, KBr, 2 Torr, Philips PU9800 FTIR) ν in cm-1: 2275
s, 2242 vs, 1775 s, 1768 sh, 1748 s, 1738 s, 1428 vs, 1405 m, 1075
vs, 1070 vs, 730 m, 725 sh, 618 sh, 609 m; for assignment, see ref 9.
(11) MS (EI, 70 eV, 20 °C) m/e (intensity): 32 (2), 28 (100). Reaction of
AgOC15N with Br2, MS (EI, 70 eV, 20 °C) m/e (intensity): 32 (2),
30 (100), 29 (10), 28 (8).
(3) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Perga-
mon: Oxford, 1984, p 336.
(4) (a) Winnewisser, M.; Bodenseh, H. K. Z. Naturforsch. 1967, 22A,
1724. (b) Winnewisser, B. P.; Jensen, P. J. Mol. Spectrosc. 1983, 101,
408. (c) Quapp, W.; Albert, S.; Winnewisser, B. P.; Winnewisser, M.
J. Mol. Spectrosc. 1993, 160, 540. (d) Wagner, G.; Winnewisser, B.
P.; Winnewisser, M.; Sarka, K. J. Mol. Spectrosc. 1993, 162, 82. (e)
Islami, K.; Jabs, W.; Preusser, J.; Winnewisser, M.; Winnewisser, B.
P. Ber. Bunsen-Ges. Phys. Chem. Chem. Phys. 1995, 99, 565.
(5) (a) Grundmann, C. Angew. Chem., Int. Ed. Engl. 1963, 2, 260. (b)
Maier, G.; Teles, J. H. Angew. Chem. 1987, 99,152; Angew. Chem.,
Int. Ed. Engl. 1987, 26, 155. (c) Pasinszki, T.; Westwood, N. P. C. J.
Am. Chem. Soc. 1995, 117, 8425.
(12) UL(AgOCN) ) 166.7 kcal mol-1 13
(OCN) ) 83.0 kcal mol-1 15 ∆Hatom(Ag) ) 68.0 kcal mol-1 14
(N-N) ) 59.6 kcal mol-1 13
,
IP(Ag) ) 175.8 kcal mol-1 14
,
EA-
BE-
,
,
.
(13) Klapo¨tke, T. M.; Tornieporth-Oetting, I. C. Nichtmetallchemie;
VCH: Weinheim, Germany, 1994; pp 81, 93, and 96-103 and
Appendix.
(14) Johnson, D. A. Some Thermodynamic Aspects of Inorganic Chemistry;
Cambridge University Press: Cambridge, U.K., 1982; Appendix.
(15) Ziegler, T.; Gutsev, G. L. J. Comput. Chem. 1992, 13, 70.
(16) (a) See ref 12; BE(Br-N) ) 46.9 kcal mol-1 13,16b
,
EA(Br) ) 82.7
(6) (a) Hunt, H. J. Am. Chem. Soc. 1932, 54, 907. (b) Powell, P.; Timms,
P. TheChemistry of Non-Metals; Chapman and Hall: London, 1974.
(7) Delgado, M. S.; Fernandez, V. Z. Anorg. Allg. Chem. 1981, 476, 149.
kcal mol-1 13
,
UL(AgBr) ) 179.2 kcal mol-1 13
.
(b) Schulz. A.;
Tornieporth-Oetting, I. C.; Klapo¨tke, T. M. Inorg. Chem. 1995, 34,
4343.
S0020-1669(96)00049-3 CCC: $12.00 © 1996 American Chemical Society

4792 Inorganic Chemistry, Vol. 35, No. 16, 1996
Notes
may not necessarily be valid since reaction 4 followed by
AgOCN + Br2 f AgBr + BrNCO
(4)
reaction 3 also results in OCNNCO. The formation of
substantial amounts of elemental silver, however, can best be
explained by a radical mechanism (i.e. eq 2).
We also studied the reaction of AgOCN with Br2 in solution
(CFCl3, CH2Cl2) at various temperatures (-60, -40, 20 °C)
and followed the reaction by 14N NMR spectroscopy. We have
found that an equimolar reaction of AgOCN with Br2 and with
an excess of bromine (eq 4) led mainly to the formation BrNCO
and AgBr. However, if bromine was reacted with a 10-fold
excess of AgOCN, OC(NCO)2 was observed as the major
product.17 In contrast to an earlier report according to which a
solution of oxocyanogen can be filtered at -60 °C7 we could
not find any evidence for the existence of oxocyanogen (1) at
this temperature. We therefore rule out the formation of
oxocyanogen (1) as a reactive intermediate species in the
reaction of AgOCN with Br2. Moreover, the generation of
elemental nitrogen can only be explained by the formation of
isooxocyanogen, OCN-NCO (2), since atomic nitrogen can be
ruled out as a high-energy intermediate species.
To explain why isooxocyanogen (2) and not oxocyanogen
(1) is formed as a short-lived intermediate by the reaction of
AgOCN with Br2, we carried out ab initio computations at
correlated levels.18 The structures of both isomers were
computed and fully optimized at the MP2(FU)/6-31G(d,p) level
of theory (Figure 1). The N-N bound species 2 turned out to
be favored over the O-O isomer 1 by 89.9 kcal mol-1 at the
HF level and by 82.1 kcal mol-1 at the MP2 level (Table 1).
The very long and weak O-O bond in 1 of 1.62 Å (cf. d(O-
O, H2O2) ) 1.47 Å)3 is in contrast to the relatively short N-N
bond in 2 of 1.39 Å. The latter value of the N-N distance
corresponds to a bond order between a single and a double bond
(typical values: N-N single bond, 1.449 Å; NdN double bond,
1.252 Å).19 With a reported N-N bond dissociation energy
for N2H4 of 59.6 kcal mol-1 (d(NN) ) 1.45 Å)13 and for N2H2
of 122 kcal mol-1 (d(NN) ) 1.25 Å),19a we estimate a N-N
bond dissociation energy for OCNNCO (2) of roughly 79 kcal
mol-1. Although the final dissociation products of the two
isomers 1 and 2 are the same, the difference between the
estimated N-N bond dissociation energy for 2 and the energy
Figure 1. MP2-optimized structures: (top) two different views of
NCO-OCN (1); (bottom) OCN-NCO (2).
Table 1. Ab Initio Data (MP2(FU)/6-31G(d,p) Level) for
Oxocyanogen, NCO-OCN (1), and Isooxocyanogen, OCN--CO
(2)
NtC-O-O-CtN (1) OdCdN-NdCdO (2)
E/au
-335.103 436 5
-335.234 22
0.00
C2h
Erel/kcal mol-1
symmetry
d(OO)/Å
82.9
C2
difference between 1 and 2 (HF 89.9, MP2 82.1 kcal mol-1
)
can be explained by the initial formation of the high-energy
•OCN radical in the dissociation of 1 which has to rearrange to
give the more stable OCN• radical. This rather special bonding
situation can be rationalized in the NBO picture by strong
noncovalent contributions (donor-acceptor interaction, negative
hyperconjugation).18,20,21 In the case of the O-O compound
(1), there are two strong intramolecular donor-acceptor interac-
tions which both weaken the O-O bond and therefore also
explain the long O-O distance. In one interaction electron
density is transferred from the bonding π(CN) orbitals into the
empty and antibonding σ*(OO) orbital (Figure 2 (top)). In the
1.622
d(NN)/Å
d(OC)/Å
d(CN)/Å
(COOC)/deg
(NCOO)/deg
(OCNN)/deg
(CNNC)/deg
(NCO)/deg
(COO)/deg
(CNN)/deg
1.385
1.181
1.238
1.296
1.190
127.9
-175.9
180.0
180.0
168.5
176.5
104.6
123.3
other interaction, electron density flows from the occupied σ-
(OO) orbital into the empty and antibonding π*(CN) orbitals
(Figure 2 (bottom)). In contrast to this situation, the N-N
compound (2) only shows one strong hyperconjugation with a
donation of electron density from the π(C1N1) orbital into the
π*(C2N2) orbital [and vice versa: π(C2N2) f π*(C1N1)]. This
interaction clearly strengthens the N-N bond (Figure 3) and
therefore accounts for a partial-double-bond character of the
N-N bond.
(17) 14N NMR (14.462 MHz, NS ) 40.000, PW ) 48 µs, Bruker SY 200),
δ in ppm relative to MeNO2: BrNCO, -393.1 (∆ν1/2 ) 220 Hz);
OC(NCO)2, -317.7 Hz (∆ν1/2 ) 360 Hz).
(18) Frisch, M. J.; Trucks, G. W.; Head-Gordon, M.; Gill, P. M. W.; Wong,
M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M.
A.; Replogle, E. S.; Gomperts, R.; Andres, K.; Raghavachari, K.;
Binkley, J. S.; Gonzales, C.; Martin, R. L.; Fox, D. J.; DeFrees, D. J.;
Baker, J.; Stewart, J. J. P.; Pople, J. A. GAUSSIAN 92; Gaussian Inc.:
Pittsburgh, PA, 1992.
(19) (a) Janoschek, R. Angew. Chem. 1993, 105, 242; Angew. Chem., Int.
Ed. Engl. 1992, 32, 230. (b) Holleman, A. F.; Wiberg, E; Niberg, N.
Lehrbuch der Anorganischen Chemie; Walter de Gruyter: Berlin, New
York, 1985; pp 562-563.
In VB terms22,23 the preference for the N-N bound isomer 2
over the O-O compound 1 can be explained either by the

Notes
Inorganic Chemistry, Vol. 35, No. 16, 1996 4793
Figure 2. Hyperconjugation in NCO-OCN (1): (top) π(CN) f
σ*(OO), 2 × 20 kcal mol-1; (bottom) σ (OO) f π*(CN), 16 kcal mol-1
.
Figure 3. Hyperconjugation in OCN-NCO (2): π(C1N1) f π*(C2N2)
(same for π(C2N2) f π*(C1N1)), 2 × 11 kcal mol-1
.
preferred localization of the single electron in the radical
intermediate OCN• in a nitrogen AO or by the fact that lone
pair-lone pair repulsion favors compound 2 (one LP per N
atom) over compound 1 (two LPs per O atom) (cf. dissociation
energies: HO-OH, 34.5 kcal mol-1; H2N-NH2, 59.1 kcal
mol-1).13,14 The calculated geometry for the OCN• radical
(d(NC) ) 1.23 Å, d(CO) ) 1.13 Å)24 is also in better accord
Figure 4. Lewis structures for NCO-OCN (1) and OCN-NCO (2).
with a Lewis structure of type A than with one of type B. These
(20) (a) Reed, A.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899.
(b) Reed, A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1987, 109, 7362.
(c) Reed, A.; Schleyer, P. v. R. Inorg. Chem. 1988, 27, 3969. (d)
NBO analysis: In the quantum mechanical computation (subjecting
the HF density matrix as represented in the localized NBOs to a
second-order perturbative analysis), the energy was computed accord-
ing to
considerations clearly suggest that structure 2a should be
preferred over structure 1a, i.e. OCN-NCO (2) rather than
NCO-OCN (1) (Figure 4).
Nicely in agreement with this, MO computations for the NCO
radical clearly indicate much larger odd-electron density for
nitrogen (0.6477) than for oxygen (0.0994).24
2
æ|hF|æ*
(2)
ææ*
E
) -2
Eæ* - Eæ
with hF being the Fock operator.
(21) Klapo¨tke, T. M.; Schulz, A. Quantenchemische Methoden in der
Hauptgruppenchemie; Spektrum: Heidelberg, Germany, 1996; p 101.
(22) Harcourt, R. D. Qualitative Valence-Bond Descriptions of Electron-
Rich Molecules: Pauling “3-Electron Bonds” and “Increased-Valence”
Theorie. In Lecture Notes in Chemistry; Berthier, G., Dewar, M. J.
S., Fischer, H., Fukui, K., Hall, G. G., Hartmann, H., Jaffe´, H. H.,
Jortner, J., Kutzelnigg, W., Ruedenberg, K., Scrocco, E., Eds.;
Springer: Berlin, Heidelberg, New York, 1982.
Acknowledgment. The financial support of this research by
the DFG and the University of Glasgow is gratefully acknowl-
edged. We are also indebted to Jim Gall for recording the 14
N
NMR spectra and to Prof. Richard D. Harcourt for initiating
suggestions and many helpful discussions. We thank both
reviewers for most valuable comments and suggestions.
(23) Harcourt, R. D. In Valence bond theory and chemical structure; Klein,
D. J., Trinajstic, N., Eds.; Elsevier: Amsterdam, 1990; p 251.
(24) Thomson, C.; Wishart, B. J. Theor. Chim. Acta 1974, 35, 361.
IC960049Z