Reaction of AgOCN with NO, NO2, CINO2, CINO, and BrNO: Evidence of the Formation of OCN-NO2 and OCN-NO

Article abstract of DOI:10.1021/ic960707f

Reactions of silver cyanate with binary nitrogen oxides, NO and NO₂, and ternary nitrogen oxohalides, C1NO₂, C1NO, and BrNO, were studied by means of gas phase IR spectroscopy. The experimental data for the neat reactions of AgOCN with NO₂ and AgOCN with C1NO₂ are compatible with the formation of an intermediate, short-lived OCN-NO₂ molecule, which undergoes cleavage-rearrangement to give N₂O and CO₂. In contrast, NO does not react with AgOCN. The reaction of XNO (X = Cl, Br) and AgOCN is much more complex, leading to a mixture of NO, NO₂, N₂O, CO₂, NCNCO, and X(CO)NCO. A reaction mechanism that involves the formation of neutral nitrosyl isocyanate, OCN-NO, has been proposed as the initial step. The two neutral intermediates, OCN-NO₂ and OCN-NO, were computed by ab initio methods at the correlated MP2 level of theory and are discussed in terms of intramolecular stabilization by donor-acceptor interaction (negative hyperconjugation).

Full text of DOI:10.1021/ic960707f

Inorg. Chem. 1996, 35, 7897-7904
7897
Reaction of AgOCN with NO, NO2, ClNO2, ClNO, and BrNO: Evidence of the Formation
of OCN-NO2 and OCN-NO
Thomas M. Klapo1tke* and Axel Schulz*
Department of Chemistry, University of Glasgow, Glasgow G12 8QQ, U.K.
ReceiVed June 14, 1996X
Reactions of silver cyanate with binary nitrogen oxides, NO and NO2, and ternary nitrogen oxohalides, ClNO2,
ClNO, and BrNO, were studied by means of gas phase IR spectroscopy. The experimental data for the neat
reactions of AgOCN with NO2 and AgOCN with ClNO2 are compatible with the formation of an intermediate,
short-lived OCN-NO2 molecule, which undergoes cleavage-rearrangement to give N2O and CO2. In contrast,
NO does not react with AgOCN. The reaction of XNO (X ) Cl, Br) and AgOCN is much more complex,
leading to a mixture of NO, NO2, N2O, CO2, NCNCO, and X(CO)NCO. A reaction mechanism that involves the
formation of neutral nitrosyl isocyanate, OCN-NO, has been proposed as the initial step. The two neutral
intermediates, OCN-NO2 and OCN-NO, were computed by ab initio methods at the correlated MP2 level of
theory and are discussed in terms of intramolecular stabilization by donor-acceptor interaction (negative
hyperconjugation).
Introduction
there are no experimental proofs for a compound of the type
N-C-O-O-C-N (oxocyanogen, dioxocyan).
In a flash light photolysis experiment, Basco investigated the
reaction of cyan radicals with molecular oxygen, finding that a
gaseous compound was formed which might be NCOCN,
NCNCO, or even NCOOCN.5
In 1980 Delgado and Fernandez reported on the formation
of oxocyanogen as an intermediate species in the reaction of
AgOCN with Br2.6 In that paper the authors claimed to have
stabilized the “(OCN)2” by subsequent reaction with TiCl4,
resulting in the formation of a polymeric compound of the type
{TiCl4(OCN)2}n.6 However, no direct evidence for the forma-
tion of (OCN)2 was found.
Cyanogen isocyanate was first prepared by Mayer from the
reaction between silver cyanate and cyanogen chloride, ClCN,
yielding a polymeric cyanogen isocyanate which equilibrates
with its monomer NCNCO (eq 2).7 Shortly thereafter, Gottardi
The unstable bent halogen isocyanates, XNCO (X ) Cl, Br,
I), which are isoelectronic to the bent halogen azides, have been
extensively investigated in the gas phase.1 They can easily be
prepared from the reaction of silver cyanate with the corre-
sponding dihalogen at high temperatures. The halogen isocy-
anates are very unstable as they are very sensitive toward
hydrolysis and they tend to polymerize.1
We have recently been studying the reactions of silver cyanate
with bromine.2 In these studies it has been shown that the final
products and their distribution very much depend on the
stoichiometry. When a large excess of silver cyanate was used,
the only stable, IR-active product in the gas phase was found
to be the carbonyl diisocyanate, OCN(CO)NCO (eq 1).2 The
excess AgOCN + Br2 f AgBr + OC(NCO)2 + N2 +
Ag + ... (1)
nAgOCN + nClCN f nAgCl + (C2N2O)n
(2)
formation of diisooxocyan, OCN-NCO, as a likely intermediate
in reaction 1 has been discussed on the basis of experimentally
observed decomposition products and on the grounds of
energetic considerations.2
There are also theoretical and experimental reports on the
related dipseudohalogen (ONC-CNO, cyanogen di-N-oxide).3
The existence of dipseudohalogen species containing cyanate
units has also been postulated by several authors,4 however,
detected NCNCO as one of the thermal decomposition products
of AgOCN.8 AgOCN decomposes above 335 °C. The only
gaseous decomposition products are CO, N2, CO2, and NCNCO.
A radical mechanism involving OCN radicals has been dis-
cussed. The remaining residue is not pure metallic silver but a
mixture of metallic silver and polymeric substances which
decompose after long strong heating liberating (CN)2 and
NCNCO.8
In earlier studies it was proposed that the intermediate
products formed in the reaction of azide compounds with
nitrosonium and nitronium salts (i.e., ON-N3 and O2N-N3)
decompose according to eqs 3 and 4 in an unimolecular process
X Abstract published in AdVance ACS Abstracts, November 15, 1996.
(1) (a) Nachbaur, E.; Gottardi, W. Monatsh. Chem. 1966, 97, 115. (b)
Eysel, H. H.; Nachbaur, E. Z. Anorg. Chem. 1971, 381, 71. (c) Gottardi,
W. Angew. Chem. 1971, 83, 445; Angew. Chem., Int. Ed. Engl. 1971,
10, 416. (d) Gottardi, W. Monatsh. Chem. 1972, 103, 1150. (e) Frost,
D. C.; MacDonald, C. B.; McDowell, C. A.; Westwood, N. P. C.
Chem. Phys. 1980, 47, 111. (f) Devore, T. C. J. Mol. Spectrosc. 1987,
162, 287. (g) Gerke, M.; Schatte, G.; Willner, H. J. Mol. Spectrosc.
1989, 135, 359. (h) Jemson, H. M.; Lewis-Aevan, W.; Westwood, N.
P. C.; Gerry, M. C. L. J. Mol. Spectrosc. 1986, 119, 22.
(2) (a) Schulz, A.; Klapo¨tke, T. M. Inorg. Chem. 1996, 35, 4791. (b)
Schulz, A.; Klapo¨tke, T. M. Does Diisooxocyan (OCN-NCO) Exists?.
RSC Autumn Meeting, Brunel University, London 9-13 September
1996; Poster 40.
(3) (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.
N4O f N2O + N2
N4O2 f 2N2O
(3)
(4)
via cyclic intermediates (i.e., cyclic N4O and N2O).9,10
(4) (a) Hunt, H. J. Am. Chem. Soc. 1932, 54, 907. (b) Powell, P.; Timms,
P. The Chemistry of Non-Metals; Chapman and Hall: London, 1974.
(5) Basco, N. Proc. R. Soc. 1965, A283, 302.
(6) Delgado, M. S.; Fernandez, V. Z. Anorg. Allg. Chem. 1981, 476, 149.
(7) Mayer, E. Monatsh. Chem. 1970, 101, 834.
(8) Gottardi, W. Monatsh. Chem. 1971, 102, 264.
S0020-1669(96)00707-0 CCC: $12.00 © 1996 American Chemical Society

7898 Inorganic Chemistry, Vol. 35, No. 26, 1996
Klapo¨tke and Schulz
(3) Vibrational Spectroscopy. Infrared spectra were recorded by
using a gas cell equipped with NaCl windows on a Philips PU9800
FTIR spectrophotometer.
Similar to the reaction of AgOCN with X-NO (X ) Cl, Br),
we investigated the reaction of AgSCN with ClNO, which yields
ON-SCN.11 Nitrosyl thiocyanate, ON-SCN, is a very unstable
compound known only in solution, which decomposes into NO
and (SCN)2.
(4) Computational Methods. The calculations were performed
using the program package Gaussian 92.16 The 6-31G* standard basis
sets were employed for the all-electron calculations. All structures were
fully optimized at the MP2 level, and the stationary points were
characterized by frequency analyses. For the open-shell molecules,
the spin-unrestricted Hartree-Fock method (UHF) was applied. In
order to give more reliable estimates of the reaction energies involving
open-shell species, spin projection techniques were used which eliminate
some (but not all) contributions from higher spin states.17 We present
the calculated eigenvalues S2 of the S2 operator and the energies of
the spin-projected wavefunctions at the PMP2/6-31G* level.
Lately, we have been investigating the nitration of hydrogen
cyanide (reaction of NO2+BF4- with HCN) yielding dinitrogen
monoxide, N2O, and carbon monoxide, CO, which is further
+
oxidized by NO2 to give CO2 and NO+. The experimental
data are in accord with the intermediate formation of the neutral
CNNO2.12 Furthermore, the reaction of silver cyanide with
nitrosyl chloride leads to the blue-green gas nitrosyl cyanide,
which is stable at room temperature.13,14
In reactions of AgOCN with halides, as well as in the thermal
decomposition of AgOCN, OCN radicals seem to be in-
volved.2,5,8 This naturally led to the attempted preparation of
the new compounds OCN-NO2 and OCN-NO using NO and
NO2 as reaction partners. To our knowledge there have been
no reports on the reaction of silver cyanate with nitrogen oxides
(NO, NO2) or nitrogen oxohalides. In this paper we report on
OCN-NO2 and OCN-NO formed as likely short-lived inter-
mediates in the reaction of silver cyanate with the corresponding
binary nitrogen oxides, NO and NO2, and ternary nitrogen
oxohalides, Cl-NO2, Cl-NO, and Br-NO.
Results and Discussion
(1) Reaction of NO2/Br2, NO2, and ClNO2 with AgOCN.
In the reaction of bromine with a large excess of silver cyanate,
it was found (eq 1) that besides AgBr a considerable amount
of elemental silver was generated, which was explained by a
radical mechanism (eq 5).2
AgOCN(s) + OCN•(g) f Ag(s) + OdCdNsNdCdO(g)
(5)
In order to quench the OCN• radicals, we reacted AgOCN
with 1:1, 2:1, and 3:1 mixtures of Br2 and NO2 and followed
the reaction by IR spectroscopy. (Pure NO2Br is very unstable.
N.B. very recently it was found that BrNO2 is formed in good
yield in a continuous flow by ClNO2 with a dilute aqueous
solution of NaBr).26 After 10 s reaction time at room temper-
ature CO2, N2O, BrNCO, traces of OCN(CO)NCO, and unre-
acted NO2 (Table 1) were observed in all cases. However, after
a 3 min reaction time only CO2 and N2O (and traces of OCN-
(CO)NCO) remained as IR-active gaseous compounds. More-
over, the analysis of the nonvolatile solid revealed beside
unreacted AgOCN the presence of AgBr and elemental silver.
We do stress that the presence of NO2 radicals has an influence
on the reaction of bromine and silver cyanate since only small
amounts of carbonyl diisocyanate were found. This can best
be explained by a radical reaction according to eq 6 which was
Experimental Section
Caution! Halogen isocyanates, binary nitrogen oxides, and ternary
nitrogen oxohalides are toxic and appropriate safety precautions should
be taken.
(1) Materials. AgOCN was prepared as a white (light-sensitive)
solid by precipitation from an aqueous solution of NaOCN and AgNO3.8
It was thoroughly washed, dried (under vacuum, over P4O10), and
degassed under high vacuum shortly before use. This is an important
step since we found that most X-NCO compounds are very susceptible
to hydrolysis.
(2) Reactions. For the preparation of samples for IR spectroscopy,
a 10 cm gas cell with a directly attached reaction vessel (15 cm3)
separated by a Young PTFE valve was used. In the reaction vessel,
dry degassed AgOCN was prepared (1.00 g, 6.7 mmol, 6-fold excess).
On the calibrated stainless steel vacuum line (SS 316) 1 mmol of the
following gases was condensed onto the silver cyanate at -196 °C:
(i) Br2 (Aldrich)/NO2 (Aldrich) (1:1, 2:1, 3:1); (ii) NO2 (Aldrich); (iii)
NO2Cl;15a (iv) NO (Aldrich); (v) NOCl;15b (vi) NOBr.15c The reaction
mixture was allowed to warm to room temperature, and IR spectra
were measured. The volatile products formed were condensed into
the attached IR gas cell, and IR spectra were recorded without delay at
20 °C and 2 Torr and again after different time intervals (see Results
and Discussion).
•NO2(g) + OCN•(g) f OCN-NO2(g)
(6)
estimated to be thermodynamically favorable ∆RH(298K) )
(16) 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.
(17) (a) Sosa, C.; Schlegel, H. B. Int. J. Quantum Chem. 1986, 29, 1001.
(b) Schlegel, H. B. J. Chem. Phys. 1986, 84, 4530.
(18) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coor-
dination Compounds, 4th ed.; Wiley & Sons: New York, Chichester,
Brisbane, Toronto, Singapore, 1986.
(19) Mecke, R.; Langenbucher, F. Infrared Spectra; Heyden & Son:
London, 1965; Serial No. 6.
(20) Mecke, R.; Langenbucher, F. Infrared Spectra; Heyden & Son:
London, 1965; Serial No. 106.
(21) Laane, J.; Jones, L. H.; Ryan, R. R.; Asprey, L. B. J. Mol. Spectrosc.
1969, 99, 489.
(22) Durig, J. R.; Kim, Y. H.; Guirgis, G. A. Spectrochim. Acta 1994, 50A,
463.
(23) Durig, J. R.; Guirgis, G. A.; Krutules, K. A. J. Mol. Struct. 1995,
354, 1.
(24) (a) Sullivan, J. F.; Nandy, S. K.; Lee, M. J.; Durig, J. R. J. Raman
Spectrosc. 1992, 23, 51. (b) Klapstein, D.; Nau, W. M. J. Mol. Struct.
1994, 317, 59.
(9) (a) Schulz, A.; Tornieporth-Oetting, I. C.; Klapo¨tke, T. M. Angew.
Chem. 1993, 105, 1697; Angew. Chem., Int. Ed. Engl. 1993, 32, 1610.
(b) Schulz, A.; Klapo¨tke, T. M. Chem. Ber. 1995, 128, 201. (c) Schulz,
A. Ph.D. Thesis, TU Berlin, Berlin, 1994. (d) Galbraith, J. M.;
Schaefer, H. F., III J. Am. Chem. Soc. 1996, 118, 4860.
(10) Klapo¨tke, T. M.; Schulz, A.; Tornieporth-Oetting, I. C. Chem. Ber.
1994, 127, 2181.
(11) (a) So¨derba¨ck, E. Ann. Chem. 1919, 217. (b) Lecher, H.; Graf, F. Chem.
Ber. 1926, 59, 2601. (c) Seel, F.; Kiskerri Bolz, A.; Nogardi, J. Z.
Anorg. Allg. Chem. 1951, 264, 298. (d) Addison, C. C.; Lewis, J. Q.
ReV. 1955, 9, 115.
(12) Klapo¨tke, T. M.; McIntyre, G.; Schulz, A. J. Chem. Soc., Dalton Trans.
1996, 3237.
(13) Dickinson, R; Kirby, G. W.; Sweeny, J. G.; Tyler, J. K. J. Chem.
Soc., Chem. Commun. 1973, 241.
(14) Horsewood, P.; Kirby, G. W. J. Chem. Soc., Chem. Commun. 1971,
1139.
(15) (a) Price, C. C.; Sears, C. A. J. Am. Chem. Soc. 1953, 75, 3276. (b)
Brauer, G. Handbook of PreparatiVe Inorganic Chemistry, 2nd ed.;
Academic Press: London, 1965; Vol. 2, p 511. (c) Brauer, G.
Handbook of PreparatiVe Inorganic Chemistry, 2nd ed.; Academic
Press: London, 1965, Vol. 2, p 513.
(25) Balfour, W. J.; Fougere, S. G.; Klapstein, D.; Nau, W. M. Spectrochim.
Acta 1994, 50A, 1039.
(26) (a) Frenzel, A.; Scheer, W.; Behnke, W.; Zetsch, J. Phys. Chem., in
press. (b) Scheffler, D.; Grothe, H.; Willner, H.; Frenzel, A.; Zetzsch,
C. Inorg. Chem., in press.

Reaction of AgOCN with Binary Nitrogen Oxides
Inorganic Chemistry, Vol. 35, No. 26, 1996 7899
Table 1. Vibrational Frequencies (cm-1) of the Experimentally Observed Moleculesa
species
frequencies
ref
18
NO
CO2
N2O
1880 (s)
3716 (w), 3609 (w), 2326 (vs), 741 (w), 721 (m), 667 (vs)
3891 (w), 3480 (m), 2809 (w), 2591 (m), 2488 (m), 2457 (m), 2217 (vs), 1890 (w),
1302 (vs), 1275 (vs), 1183 (m), 1155 (m), 694 (w)
2900 (w), 1750 (s), 1610 (vs), 1260 (s), 750 (m)
18, 19
18, 20
NO2
18
18, 21
26
Br-NO
Br-NO2
Cl-NO
Cl-NO2
Br-CN
Cl-CN
1799 (vs), 542 (s) 266 (m)
b
1667 (s), 1293 (vs), 1200 (w), 787 (s), 605 (w)
1800 (vs), 595 (s), 332 (m)
18
18, 22
18
18
1f, 7
23
24
25
1685 (vs), 1268 (vs), 792 (s), 652 (m)
2200 (vs), 574 (s), 342 (m)
2219 (vs), 714 (s), 380 (m)
NC-NCO
2280 (vs), 2233 (vs), 1750 (m), 1073 (s), 727 (w), 610 (w)
2255 (vs), 1178 (m), 1816 (s), 1420 (s), 1399 (m), 987 (sh), 965 (s), 775 (w), 613 (m)
2260 (vs), 1818 (s), 1780 (m), 1426 (m) 1411 (m), 992 (s), 789 (w), 651 (w), 634 (w)
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)
Br(CO)NCO
Cl(CO)NCO
OCN(CO)NCO
a In our studies we could reproduce and confirm all frequencies reported in the literature.18-25 b The reported values for Br-NO2 were taken
from the literature. No attempts were made to reproduce these data.26
Table 2. Calculated Reaction Energies and Enthalpies (kcal mol-1
)
AgOCN the presence of elemental silver. It is known from the
literature that excited oxygen atoms, O(1D), react in a similar
way with HNCO leading to CO2 and NH(a1∆) in a strongly
exothermic reaction (∆H ) -92 kcal mol-1).30 Moreover, the
mechanism in that reaction is thought to involve the electrophilic
attack of the excited oxygen on the carbon atom of the NCO
chain.31
eq
E(MP2/6-31G*) E(PMP2/6-31G*) ∆RH(298K,zpe)a
6
-46.8
-83.0
-5.9
-39.0
-38.9
-82.9
-4.0
7
9
10
X ) Cl
X ) Br
-1.7
-3.6
-116.5
-30.1
-1.6
-2.8
-115.6
-29.8
We also studied the neat reaction of ClNO2 and AgOCN at
room temperature. We found that independently from the molar
ratio of the reactants, this exothermic reaction always led to
11
12
15
X ) Cl
X ) Br
-15.6
-22.1
-14.9
-21.0
N2O, CO2, and AgCl [∆RH(298K) ) -4.0 kcal mol-1 32
cf.
;
Table 2; eqs 7 and 9].
17
18
19
-176.5
-46.2
-175.6
-38.2
-38.2
-173.0
-172.9
AgOCN(s) + ClNO2(g) f AgCl(s) + OCNNO2(g) (9)
a Reaction enthalpy after correction for zpe contributions, rotational
and translational degrees of freedom, and the work term at room
temperature.
Due to the only experimentally observed products, CO2 and
N2O, the proposed reactive short-lived OCN-NO2 thus would
belong to the group of nitrogen-containing compounds, which
like nitrosyl azide (f N2 + N2O)9 and nitryl azide (f 2N2O)10
completely decompose at low temperature yielding only gases.
Tetranitromethane (f 2NO2 + 2NO + CO2 for T > 550 K or
f 3NO2 + NO + CO),27a azodicarbonitrile (f N2 + NC-
CN),27b methyl azide (f N2 + H2 + HCN),33 or ammonium
nitrate (f N2O + 2H2O)34 completely decompose on heating
into thermodynamically favorable, gaseous compounds (eq 7).
(2) Reaction of NO, ClNO, and BrNO with AgOCN. NO
is a stable free radical, which in contrast to NO2 does not react
with silver cyanate at room temperature, not even when the
reaction mixture was gently heated or activated by ultrasound.
-38.9 kcal mol-1 (Table 2). The short-lived intermediate
OCN-NO2, can then undergo a cleavage-rearrangement to give
the experimentally observed N2O and CO2 (eq 7, Table 1; see
OCN-NO2(g) f CO2(g) + N2O(g)
(7)
below, (3) Computations). This decomposition reaction was
calculated to be highly exothermic ∆RH(298K) ) -82.9 kcal
mol-1 (Table 2). A similar thermal decomposition reaction has
been reported in literature.27a The thermal decomposition of
gaseous tetranitromethane, C(NO2)4, in contact with silver wool
was investigated by Bock and Zanathy, yielding NO2, CO2, NO,
and CO traces.27a Carbonyl dinitrite, (O2N)2CO, has been
discussed as an intermediate decomposing into NO2, NO, and
CO2 (or in the absence of silver, depending on the temperature,
into 2NO2 and CO), indicating a preceding chemisorption at
the silver surface.27a
(28) UL(AgOCN) ) 166.7 kcal mol-1 29c
∆Hatom.(Ag) ) 68.0 kcal mol-1 29b Ip(NO2) ) 232.7 kcal mol-1 29a
∆RH°(298K,eq17) ) -175.6 kcal mol-1 (see Table 2).
;
Ip(Ag) ) 175.8 kcal mol-1 29b
;
;
;
(29) (a) Pearson, R. G. J. Am. Chem. Soc. 1988, 110, 7684. (b) Johnson,
D. A. Some Thermodynamic Aspects of Inorganic Chemistry; Cam-
bridge University Press: Cambridge, U.K., 1982; appendix. (c)
Klapo¨tke, T. M.; Tornieporth-Oetting, I. C. Nichtmetallchemie;
VCH: Weinheim, 1994; pp 81, 93, 96-103, appendix. (d) U.S.
Department of Commerce, National Bureau of Standards; Selected
Values of chemical thermodynamic properties; United States Govern-
ment Printing Office: Washington, DC, 1982.
On the other hand, NO2 directly reacts with AgOCN at room
temperature forming OCN-NO2, which decomposes into CO2
and N2O as the only IR-active gaseous compounds (eqs 7 and
8, Table 1). This reaction was also predicted to be exothermic
(30) (a) Ongstad, A. P.; Liu, X.; Coombe, R. D. J. Phys. Chem. 1988, 92,
5578. (b) Singleton, S. M.; Coombe, R. D. J. Phys. Chem. 1992, 96,
9865.
•NO2(g) + AgOCN(s) f Ag(s) + OCN-NO2(g) (8)
(31) Although much of the energy released in this reaction comes from
the formation of the new C-O bond, angular momentum contraints
force the system to produce NH in its excited a1∆ state.
(32) ∆RH° was calculated for the following reaction: OCN-(g) + NO2-
Cl(g) f Cl-(g) + OCNNO2(g).
[∆RH(298K) ) -20.0 kcal mol-1]. The thermodynamic driving
force is the formation of CO2 and N2O (eq 7).28,29 Furthermore,
analysis of the nonvolatile solid revealed beside unreacted
(33) Bock, H.; Dammel, R. Angew. Chem. 1987, 99, 518; Angew. Chem.,
Int. Ed. Engl. 1987, 26, 504.
(27) (a) Bock, H.; Zanathy, L. Angew. Chem., Int. Ed. Engl. 1992, 31,
900. (b) Bock, H.; Dammel, R.; Lorenzak, P.; Wentrup, C. Z.
Naturforsch. B 1990, 45, 59.
(34) Holleman, A. F.; Wiberg, E. Lehrbuch der Anorganischen Chemie,
101 ed.; Walter de Gruyter: Berlin, 1995; p 579.

7900 Inorganic Chemistry, Vol. 35, No. 26, 1996
Klapo¨tke and Schulz
This behavior can partly be attributed to the much lower boiling
point of NO (-152 °C) compared to that of NO2.51 Therefore,
in contrast to the reaction with NO2, one would not expect a
liquid phase to be present in the reaction of NO with AgOCN.
Alternatively, the different reactivity of NO2 and NO can be
explained by the fact that the EA of NO2 greatly exceeds that
of NO.35 One could expect the initial formation of the radial
Combining the exothermicities of reactions 11 and 12
suggests that oxygen transfer from N2O to CO is 85.8 kcal mol-1
exothermic. The experimental value of 87.1 kcal mol-1 is nicely
in agreement with this calculation.29d Both dissociation reac-
tions are exothermic [∆RH(298K,eq11) ) -115.6 kcal mol-1
;
∆RH(298K,eq12) ) -29.8 kcal mol-1; Table 2], but as the
transition state for the decomposition according to eq 12 is much
higher in energy than the transition state for the decomposition
into CO2 and N2 (see Computations, Table 6) only the products
for reaction 11 were observed (small amounts of N2)39 as well
as those of the more likely reaction with XNO.
•
anion salt, Ag+(NO2 OCN)-, which then collapses by electron
transfer, while the analogous Ag+(NO•OCN)- salt would be
much less stable. On the other hand, NO reacts rapidly with
other free radicals or halogens (forming nitrosyl halides) while
it is not very reactive toward most stable molecules.36
The formation of X-CN, NO2, and NO can best be explained
by the following mechanism (eq 13, Figure 1).
In contrast to NO, nitrosyl chloride and nitrosyl bromide react
with silver cyanate spontaneously. The IR spectrum obtained
after a 10 s reaction time at room temperature clearly showed
in both cases (Cl-NO and Br-NO) the presence of NO, NO2,
X-CN (X ) Cl, Br), NC-NCO, X(CO)NCO, and small
amounts of CO2 and N2O (Table 1). The difference in band
shapes and band widths was indicative for the presence of at
least two different NCO- and CN-containing molecules (in the
region 2000-2300 cm-1). Cyanogen isocyanate, NC-NCO,1f,7
was unequivocally identified from the bands at 1775 (ν3 + ν8),
1743 (ν2 - ν7), 1430 (ν3), and 1070 cm-1 (ν4). The halogen-
carbonyl isocyanate, X(CO)NCO (X ) Cl,24 Br23), was deduced
from the bands at 1820 (ν2) and 990 cm-1 (ν4) for Cl(CO)-
NCO and 1815 (ν2) and 980 cm-1 (ν4) for Br(CO)NCO (Table
1). By noting and comparing differences in the relative
intensities of the bands as well as band shapes and band widths
in duplicate experiments but after different reaction times, it
was possible to determine the infrared bands arising from each
molecule. While after a 3 min reaction time the amount of CO2
and N2O had increased the amount of X(CO)NCO and the
amount of NO2 had decreased; neither ClCN nor BrNC could
be detected. After 10 min, no halogencarbonyl isocyanate,
X(CO)NCO (X ) Cl, Br), remained, while increasing amounts
of CO2 and N2O were observed. The volume of NO, NO2, and
NC-NCO decreased slowly. Eventually, after 24 h only CO2
and N2O remained as IR-active gaseous compounds. Further-
more, analysis of the nonvolatile solid revealed, beside unreacted
AgOCN, the presence of AgX (X ) Cl, Br), elemental silver,
and oligomeric NCO compounds.
OCN-NO(g) + X-NO(g) f
{X-CN-NO•}(g) + •NO2(g) (13)
Due to the polarity of the O-N moiety in OCN-NO and Cl-N
in Cl-NO (Figure 1), nitrogen dioxide is formed as well as
X-CN-NO• radicals which can decompose rapidly to give
•
X-CN and NO (eq 14).40 The thermodynamic driving force
{X-CN-NO•} f X-CN(g) + •NO(g)
(14)
for both reactions (eqs 13 and 14), the formation of the unstable
radical X-CN-NO• (which does not represent a minimum on
the PES) and the decomposition into the experimentally
observed X-CN and NO, is the formation of CO2 and N2O.
The formation of CO2 and N2O can easily be explained by the
reaction of nitrogen dioxide with an excess of silver cyanate as
described before (eqs 7 and 8). The experimental result
(formation of NO2 and NO) suggests that the stability of the
OCN-NO molecule may be determined by the activation barrier
for breaking the OCN-NO bond.
The reaction of cyanogen chloride, ClCN, with silver cyanate
has been known for a long time, yielding monomeric NCNCO
([∆RH(298K, eq15, XdCl) ) -14.9 kcal mol-1; ∆RH(298K,
eq15, XdBr) ) -21.0 kcal mol-1; Table 2)] (eq 15) which
X-CN(g) + AgOCN(s) f AgX(s) + NC-NCO(g) (15)
The initial step of the nitrosyl halide/silver cyanate reaction
seems to be the exothermic formation [∆RH(298K,X)Cl) )
-1.6 kcal mol-1; ∆RH°(298K,X)Br) ) -2.8 kcal mol-1; Table
2]37 of the intermediate OCN-NO (eq 10) Since the highly
nNC-NCO a (C2N2O)n
(16)
is in equilibrium with polymeric cyanogen isocyanate (eq 16).7,41
The polymerization as well as its slow decomposition explains
the disappearance of NC-NCO after 24 h.
The explanation for the formation of the halogencarbonyl
isocyanate is not as straightforward as with NC-NCO. We
assume it is formed in a much more complex reaction of NC-
NCO with X-NC (X ) Cl, Br) and solid AgOCN similar to
the formation of carbonyl diisocyanate (eq 1). The correspond-
ing halogencarbonyl isocyanate is not stable in the mixture of
NC-NCO, NO2, NO, CO2, N2O, and solid AgOCN and could
not be observed anymore after 10 min.
AgOCN(s) + X-NO(g) f
AgX(s) + OCN-NO(g)
X ) Cl, Br (10)
polarized OCN-NO molecule is very reactive,38 this short-lived
intermediate is expected to react immediately with XNO (X )
Cl, Br) or decompose into either molecular N2 and CO2 via
cleavage-rearrangement (eq 11) or into N2O and CO
(eq 12).
(3) Computations. Structural parameters obtained from
MP2/6-31G* calculations for isocyanates, nitryl, and nitrosyl
compounds as well as azides are in good agreement with those
obtained from experimental studies.42,43 It is generally agreed
that this level is sufficient to predict the relative stabilities of
OCN-NO(g) f CO2(g) + N2(g)
OCN-NO(g) f CO(g) + N2O(g)
(11)
(12)
(35) Ip(NO) ) 215.0 kcal mol-1 29b
(see Table 2).
;
∆RH(298K,eq19) ) -172.9 kcal mol-1
(39) The formation of N2 was established by gas discharge.
(40) XCN is more stable than XNC, X ) Cl, Br: Schulz, A.; Klapo¨tke, T.
M., unpublished results.
(36) (a) Czapski, G.; Holcman, J.; Bielski, B. H. J. Am. Chem. Soc. 1994,
116, 11465. (b) Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992,
258, 1898. (c) Blough, N.; Zafiriou, O. C. Inorg. Chem. 1985, 24,
3502.
(41) ∆RH was calculated for the following reaction: X-CN(g) + OCN-(g)
f X-(g) + NC-NCO(g).
(37) ∆RH was calculated for the following reaction: OCN-(g) + XNO(g)
(42) (a) Otto, M.; Lotz, S. D.; Frenking, G. Inorg. Chem. 1992, 31, 3647.
(b) Schulz, A.; Tornieporth-Oetting, I. C.; Klapo¨tke, T. Inorg. Chem.
1995, 34, 4343. (c) Cradock, S.; Durig, J. R.; Sullivan, J. F. J. Mol.
Struct. 1985, 131, 121. (d) Durig, J. R.; Little, T. S. Gounev, T. K.;
f Cl-(g) + OCN-NO(g).
(38) The N-N bond of OCN-NO is about as weak (d(NN) ) 1.522 Å)
as the weak halogen-nitrogen bond in nitrosyl halides.

Reaction of AgOCN with Binary Nitrogen Oxides
Inorganic Chemistry, Vol. 35, No. 26, 1996 7901
Figure 1. Polarity of the O-N moiety in OCN-NO and Cl-N in
Cl-NO.
the isomers and will give reasonably reliable results for the
equilibrium structures.44,45
Table 7 shows the calculated structures at MP2/6-31G* level
as well as the total energies and zero point vibrational energies
for all discussed species. The theoretically predicted vibrational
frequencies for all species have been calculated with the
harmonic approximation. The deviation from experimentally
obtained frequencies may partly be compensated by using
scaling factors. They may also be different for various
vibrational modes present in the molecule.42a,46 Some of the
calculated molecules were calculated before by other groups at
different levels so that we want to focus only on the OCN-
NO2 and OCN-NO molecules.47
Figure 2. MP2/6-31G* optimized minimum structure of OCN-NO2
(Cs symmetry) and the structure of the transition state (Cs symmetry)
for the intrinsic decomposition into CO2 and N2O (bond distances in
Å, angles in deg).
OCN-NO2. We discussed two possibilities of the formation
•
of OCN-NO2: (i) the reaction of OCN• radicals with NO2
radicals according to eq 6 and (ii) the reaction of ClNO2 with
AgOCN (eq 9). Both reactions are exothermic; however, for
the radical reaction the energy liberated is much higher (Table
2). Another possible mechanism would involve the ionic
Table 3. MP2/6-31G* Level Fully Optimized Structures (Distances
in Å, Angles in deg) of Y-NO2 and Y-NO (Y ) OCN, NNN)
OCN-NO2 NaNbNc-NdO2 OCN-NO NaNbNc-NdO
d(NO)
+
reaction of NO2 cations (e.g., NO2BF4) with OCN- anions
trans
1.226
1.235
1.462
1.274
1.151
1.223
1.232
1.451
1.253
1.173
1.191
1.204
cis
(e.g., NaOCN) in solution. The reaction of the ions in the gas
phase has been estimated to be strongly exothermic [∆RH(298K,eq
17) ) -175.6 kcal mol-1; Table 2].
d(NN), d(NcNd)
d(CN), d(NbNc)
d(OC), d(NaNb)
1.522
1.245
1.181
1.484
1.270
1.154
(OCN),
(NaNbNc)
(ONN)
trans
172.7
171.8
173.4
174.3
+
OCN- (g) + NO2 (g) f OCN-NO2(g)
(17)
112.5
118.5
108.0
113.7
117.8
115.4
110.9
117.0
110.5
107.1
To answer the question why nitryl isocyanate, OCN-NO2,
and not nitryl cyanate, NCO-NO2, is formed as a short-lived
intermediate by the reaction of AgOCN with NO2 (as well as
AgOCN and Cl-NO2), we carried out ab initio computations
at a correlated level. Only for the N-N bound species was a
minimum ground state structure found whereas for the O-N
bound species neither a singlet nor a triplet ground state exists
at the applied level of theory. At this point, the question of
whether there is a minimum for the O-N bound species must
still be regarded as being open. It is not unreasonable to assume
that with even larger basis sets and applying higher levels of
theory NCO-NO2 may possess a minimum. In any case, one
would expect the N-O bound species to be less stable than the
N-N bound species because of a strong intramolecular σ(ON)/
π(CN) interaction which weakens the N-O single bond in
NCO-NO2.
cis
(CNN),
(NbNcNd)
(ONO)
129.0
128.5
Nicely in agreement with this, the MO computations for the
•NCO radical clearly indicate much larger odd-electron density
for nitrogen (0.6477) than for oxygen (0.0994).48 Similarly, it
•
can be expected that also for NO2 the odd electron is more
localized on the nitrogen so that for a radical mechanism the
formation of the N-N bound species should be favored.
Table 3 summarizes the computed structural parameters of
OCN-NO2. The molecular structure of OCN-NO2 was fully
optimized at the MP2/6-31G* level and is shown in Figure 2.
All structural parameters are in good agreement with those of
other covalent isocyanates and nitryl compounds.9,10,12,23,24,42c,d,44,49
Therefore, the interpretation of OCN-NO2 as nitryl isocyanate
seems to be justified. As expected for covalently bound
isocyanates, OCN-NO2 displays a bent configuration with an
N-C-O bond angle of 171.8°. The calculated structure
reminds one of the situation for the isoelectronic nitryl azide,
NNN-NO2 (Table 3).10 The computed N-O bond distances
in OCN-NO2 of 1.232 and 1.223 Å are slightly longer than in
N2O (1.186 Å)50 and in ClNO2 (1.21 Å),51 corresponding to a
bond order slightly less than 2. The N-N bond is rather long
Gardner, J. K.; Sullivan, J. F. J. Mol. Struct. 1996, 375, 83. (e) Durig,
J. R.; Guirgis, K. A.; Krutules, Phan, H. V.; Stidham, H. D. J. Raman
Spectrosc. 1994, 25, 221. (f) Klatt, G.; Willetts, A.; Handy, N. C.;
Esposti, C. D. Chem. Phys. Lett. 1995, 237, 273.
(43) (a) Yu, D.; Rauk, A.; Armstrong, D. A. J. Phys. Chem. 1992, 96,
6031. (b) Ervin, K. M.; Ho, J.; Lineberger, W. C. J. Phys. Chem.
1988, 92, 5402.
(44) Pasinszki, T.; Westwood, P. C. J. Chem. Soc., Faraday Trans. 1996,
92, 333.
(45) MP2 calculations with a 6-31G* basis set give bond length of ∼0.01
Å, generally too long, especially for multiple bonds, and angles in
the range 1-2°.46
(46) (a) Hehre, W.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab initio
Molecular Orbital Theory; Wiley & Sons: New York, 1986; p 236.
(b) Klapo¨tke, T. M.; Schulz, A. Quantenchemische Methoden in der
Hauptgruppenchemie; Spektrum: Heidelberg, 1996.
(47) References are included in Table 7.
(48) Thomson, C.; Wishart, B. J. Theor. Chim. Acta 1974, 35, 361.
(49) (a) Aljuaid, S. S.; Alnasr, A. A. K.; Eaborn, C.; Hitchcock, P. J.
Organomet. Chem. 1995, 488, 155. (b) McAllister, M. A.; Tidwell,
T. T. J. Chem. Soc., Perkin Trans. 1994, 10, 2239. (c) Pasinszki, T.;
Westwood, N. P. C. J. Chem. Soc., Faraday Trans. 1996, 92, 333.
(50) Lacy, M; Whiffen, D. H. Mol. Phys. 1982, 45, 241.

7902 Inorganic Chemistry, Vol. 35, No. 26, 1996
Klapo¨tke and Schulz
Table 4. Computed NPA Partial Charges (e) of Y-NdO2 and
Y-NdO (Y ) OaCbNc, NaNbNc)
according to eq 7 was calculated to be strongly exothermic
[∆RH(298K) ) -82.9 kcal mol-1, Table 2]. The decomposition
of the short-lived OCN-NO2 could go either in an unimolecular
process via cleavage-rearrangement to give N2O and CO2 (cf.
NNN-NO2 decomposes intrinsically into two N2O)10 or in a
bimolecular process. Exploration of the PES led to the Cs
symmetry transition state (TS) for the intrinsic decomposition
(Figure 2). The transition structure (Figure 2) is basically a
distorted O-N-N-C ring with a rather long O-C bond
distance of 1.779 Å and still resembles more the nitryl isocyanate
structure than the decomposition products. This is in agreement
with Hammond’s postulates that the TS lies more toward the
higher energy side.56 Mechanistically, this cleavage-rearrange-
ment is thought to be driven by a nucleophilic attack of one of
the nitryl oxygens on the carbon of the OCN chain. Energeti-
cally, the barrier to dissociation is at 14.4 kcal mol-1, indicating
that the OCN-NO2 is intrinsically stable, but this small barrier
could severely impede further attempts to isolate OCN-NO2.
OCN-NO. Recently, the heavier analogues, XCN(NY) [X,
Y ) S,O; X * Y], have been investigated by using ab initio
MP2/6-31G* methods.44 This study indicates that the NS-
NCO isomer is more stable than the SN-NCO isomer and that
the ON-SCN isomer is more stable than the ON-NCS isomer.
In contrast to the sulfur analogues but in agreement with the
nitrosyl azide, the ab initio computation shows a planar trans
trans structure for (Cs) OCN-NO (Figure 3, Table 3) as the
most likely minimum structure in the gas phase (N.B. The O-N
bound species, NCO-NO, is higher in energy and possesses a
rather weak and long N-O bond of 1.69 Å at MP2/6-31+G*
level). The remarkable difference to the bond situation in
nitrosyl azide and nitryl isocyanate is the fact that the N-N
bond of 1.522 Å is rather long and comparable with 1.49 Å in
F2N-NF2.57 The predicted bond length is longer than the
experimental nitrogen-nitrogen bond length in hydrazine H2N-
NH2 (1.449 Å)52 but much shorter than the nitrogen-nitrogen
bond length in the weakly bound dinitrogen tetraoxide O2N-
NO2 (1.782 Å)58 and resembles the weak halogen-nitrogen
bond situation in XNO molecules. It has been known for a
long time that all species of the form XNO (X ) F, Cl, Br)
have a bent structure (Cs) with tight N-O bonds and weak bonds
between the halogen and the NO moiety.59 The weak halogen-
nitrogen bond in nitrosyl halides stems from an unfavorable
overlap of a singly-occupied halogen valence p orbital with a
singly occupied antibonding 2π orbital of NO.21,59d In addition,
the heavier halides are better electron donors and therefore
contribute more electron density to the NO π* orbital,59d a trend
that is consistent with the decreased X-N bond strength from
F to Cl to Br. Since the OCN unit can be regarded as a pseudo-
halogen, these rather special bonding situations can be rational-
ized in the picture of localized bond orbitals (NBO) by strong
noncovalent contributions (donor-acceptor interaction). There
is one strong p-LP(O) f σ*(N-N) intramolecular donor-
acceptor interaction (besides π electron density delocalization)
which weakens the N-N bond and therefore also explains the
fairly long N-N distance (E2 ) 42.3 kcal mol-1).55 There are
several possible ways of drawing reasonable Lewis structures
for the trans-chain nitrosyl isocyanate, a few of which are
represented in Figure 4 with structure A corresponding to the
most likely one due to NBO analysis.
Na/Oa Nb/Cb Nc/Nc Nd/Nd trans O cis O
NNN-NO2 +0.17 +0.29 -0.32 +0.76 -0.40 -0.49
OCN-NO2 -0.51 +1.12 -0.56 +0.77 -0.37 -0.45
NNN-NO
OCN-NO
+0.07 +0.25 -0.34 +0.34 -0.30
-0.54 +1.09 -0.62 +0.35 -0.28
Table 5. Computed IR Data (MP2/6-31G*); Harmonic Vibrational
1
Frequencies (cm-1) and Infrared Intensities (km mol-1) of A′
1
OCN-NO2 and A′ OCN-NOa
approx
approx
OCN-NO2
assignment
OCN-NO
assignment
ω1
ω2
ω3
ω4
ω5
ω6
ω7
ω8
ω9
ω10
ω11
ω12
2285(692)
1843(150)
1319(20)
1299(442)
858(192)
750(19)
731(11)
563(22)
535(16)
515(8)
υas(OCN)
2280(941)
1614(398)
1296(206)
768(196)
640(18)
υas(OCN)
υ(NO)
υas(NO2)
υ(OCNNO2)
υ(OCNNO2)
υ(OCN-NO2)
δ(OCNNO2)
γ(OCNNO2)
γ(OCNNO2)
δ(OCNNO2)
δ(OCNNO2)
δ(OCNNO2)
γ(OCNNO2)
υs(OCN)
υ(OCN-NO)
δ(OCN)
591(20)
γ(OCN)
359(378)
159(4)
110(0)
δ(OCNNO)
δ(OCNNO)
γ(OCNNO)
144(1)
70(0)
a Unscaled.
with 1.451 Å, comparable to 1.462 Å in NNN-NO2 (Table 3).
The value of the N-N distance corresponds to a bond order
less than 1 (typical values: NsN single bond, 1.447 Å in
N2H4,52 NdN double bond, 1.252 Å in N2H2).53 As indicated
by NBO analysis,54 there is a significant interaction of the lone
pairs on both oxygen atoms with the unoccupied, antibonding
σ* orbital of the N-N bond. This LP(O) f σ*(N-N) donor-
acceptor interaction (hyperconjugation)46b accounts for the rather
long N-N bond [LP(cis-O) E2 ) 29.9 kcal mol-1; LP(trans-
O) E2 ) 32.5 kcal mol-1].55 Moreover, the N-N bond has a
polarity OCNδ--Nδ+O2 that resembles the situation of the
polarized halogen isocyanates and covalent azides with the N-N
moiety being even more polarized than in NNN-NO2 (Table
4).
The theoretically predicted harmonic vibrational frequencies
and IR intensities of OCN-NO2 are presented in Table 5.
According to Table 5, ω1 is almost purely an asymmetric
stretching mode of the OCN moiety, while ω2 represents the
asymmetric stretch for the NO2 group. The assignments of ω3
and ω4 are not as straightforward as with ω1 and ω2. The ω3
and ω4 motions can be described as a mixture of both symmetric
stretch of the OCN and the NO2 units, whereas ω5 represents
an N-N stretching along with O-N-O and O-C-N bending.
The remaining normal modes are either in-plane deformation
or out-of-plane deformation vibrations.
The dissociation of isolated, gaseous OCN-NO2 molecules
into the experimentally observed products, CO2 and N2O,
(51) Klapo¨tke, T. M.; Tornieporth-Oetting, I. C. Nichtmetallchemie;
VCH: Weinheim, 1994; p 298.
(52) Kuhata, K.; Fukuyama, T.; Kuchitsu, K. J. Phys. Chem. 1982, 86,
602.
(53) Carlotti, N.; Johns, J. W. C.; Tromeetti, A. Can. J. Phys. 1974, 52,
340.
(54) (a) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.
NBO Version 3.1. (b) Carpenter, J. E.; Weinhold, F. J. Mol. Struct.
(THEOCHEM) 1988, 41, 169. (c) Foster, J. P.; Weinhold, F. J. Am.
Chem. Soc. 1980, 102, 7211. (d) Reed, A. E.; Weinhold, F. J. Chem.
Phys. 1983, 78, 4066. (e) Reed, A. E.; Weinstock, R. B.; Weinhold,
F. J. Chem. Phys. 1985, 83, 735. (f) Reed, A.; Curtiss, L. A.; Weinhold,
F. Chem. ReV. 1988, 88, 899. (g) Reed, A.; Schleyer, P. v. R. J. Am.
Chem. Soc. 1987, 109, 7362. (h) Reed, A.; Schleyer, P. v. R. Inorg.
Chem. 1988, 27, 3969. (i) Weinhold, F.; Carpenter, J. E. The Structure
of Small Molecules and Ions; Plenum: New York, 1988; p 227.
(55) NBO analysis: The second-order pertubative energy was computed
(56) Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334.
(57) Peters, N. J. S.; Allen, L. C. Mol. Struct. Energ. 1988, 8, 199.
(58) McClelland, B. W.; Gundersen, G.; Hedberg, K. J. Chem. Phys. 1972,
56, 4541.
(59) (a) Burns, W. G.; Bernstein, H. J. J. Chem. Phys. 1950, 18, 1669. (b)
Millen, D. J.; Mitra, D. Trans. Faraday Soc. 1970, 66, 2414. (c)
Takeuchi, T. J. Electron Spectros. Relat. Phenom. 1986, 40, 27. (d)
Meredith, C.; Quelch, G. E.; Schaefer, H. F., III J. Chem. Phys. 1992,
96, 480. (e) Obermeyer, A.; Borrmann, H.; Simon, A. J. Am. Chem.
Soc. 1995, 117, 7887.
2
according to E2 ) -2( φ|hFφ* /Eφ* - Eφ), with hF being the Fock
operator.

Reaction of AgOCN with Binary Nitrogen Oxides
Inorganic Chemistry, Vol. 35, No. 26, 1996 7903
Figure 3. MP2/6-31G* optimized minimum structure (bond distances
in Å, angles in deg) of trans- and cis-OCN-NO2 (Cs symmetry).
Figure 5. MP2/6-31G* optimized transition state structures (bond
distances in Å, angles in deg) of OCN-NO; TS1 (f CO + N2O);
TS2 (rotation f cis-isomer), TS3 (ring f CO2 + N2).
Figure 4. Lewis structures for OCN-NO.
Table 6. Energies (kcal mol-1) Relative to CO2 + N2 of the
Stationary Points on the OCN-NO PES and the Likely
Decomposition Products
Table 5 shows the calculated unscaled MP2 frequencies for
the nitrosyl isocyanate. Analogous to OCN-NO2, ω1 is almost
purely an asymmetric stretching mode of the OCN moiety,
whereas ω2 represents the N-O stretch. The ω3 motion can be
described as the asymmetric O-C-N stretch, and ω4 is best
described as an N-N stretching along with O-N-O and
O-C-N bending. The remaining normal modes correspond
either to in-plane or to out-of-plane deformation vibrations.
Nitrosyl isocyanate can be formed in three different reac-
tions: (i) in a radical reaction (eq 18), (ii) in an ionic process
CO + CO2 +
trans
cis
TS1a TS2b TS3c N2O
N2
MP2 116.5 119.8 175.3 121.1 136.1 86.4
MP2 + zpe 115.6 118.9 175.1 120.5 135.1 87.0
0.0
0.0
a TS1 (f CO + N2O). b TS2 (rotation f cis-isomer). c TS3 (ring
f CO2 + N2).
According to our experiments, presumably OCN-NO is
initially formed in the reaction of anhydrous silver isocyanate
with XNO (X ) Cl, Br) but is not stable in the presence of
either XNO and AgOCN. On the other hand, there are two
possible ways for an intrinsic decomposition (eqs 9 and 10).
The driving force is the formation of the much more thermo-
dynamically stable CO2 and N2 or CO and N2O.
•NO(g) + OCN•(g) f
OCN-NO(g), ∆RH(298K) ) -38.2 kcal mol-1 (18)
(eq 19), and (iii) in the reaction of MOCN and XNO (M ) Na,
NO+(g) + OCN-(g) f
Similar to the decomposition of trans-N4O, the direct
decomposition into CO and N2O (eq 12) is impeded since the
barrier to overcome is ∼58.1 kcal mol-1 (breaking of the strong
N-C bond, Cs-TS1, Figure 5), and in accordance with this, CO
was not observed in the experiment. On the other hand, the
reaction via rotation (C1-TS2, Figure 5) to cis-OCN-NO (Figure
3)60 followed by intrinsic decomposition via a four-membered-
ring transition state structure into CO2 and N2 is favored (eq
11, Table 6). This is in contrast to the decomposition of cis-
N4O, where first a cyclic N4O isomer is formed which then
dissociates. No minimum structure for a five-membered OCN-
NO ring was found. Energetically, the breaking of the extended
OCN-NO(g), ∆RH(298K) ) -172.9 kcal mol-1 (19)
Ag; X ) Cl, Br; eq 10). While all these reactions were predicted
to be exothermic, the reaction of the ions is the most exothermic
one. These ∆RH(298K) values are estimates for the gas phase
reactions (lattice or solvation energies were not included).
Combining the data used in the reactions 18 and 19 suggests
an electron affinity for OCN of EA(OCN) ) 65.8 kcal mol-1
and an ionization potential for NO of Ip(NO) ) 201.2 kcal mol-1
(cf. Table 7). These values are in reasonable agreement with
the experimental data of EA(OCN) ) 83.0 kcal mol-1 and an
ionization potential for NO of Ip(NO) ) 215.0 kcal mol-1 29b,62
.
(61) d(NaCb) ) 1.136 Å, d(CcNd) ) 1.315 Å, d(NeCf))1.206 Å, and d(CfO)
) 1.136 Å; (NCN) ) 177.3°, (CNC) ) 133.8°, and (NCO) )
173.7°.
(60) 2266(773), 1555(107), 1269(274), 802(172), 716(177), 595(24), 358-
(228), 132(1), and 114(0) cm-1
.
(62) Ziegler, T.; Gutsev, G. L. J. Comput. Chem. 1992, 13, 70.

7904 Inorganic Chemistry, Vol. 35, No. 26, 1996
Klapo¨tke and Schulz
MP2
tot
Table 7. Calculated Total Energies E
(au), Interatomic Distances (Å) and Angles (deg), and Zero-Point Vibrational Energies zpe (kcal
mol-1
)
species
MP2
tot
PMP2
tot
a
E
S2
E
structural datab
zpe
4.9
BrCN
(1Σ)
-2662.56668
d(BrC) ) 1.791
d(CN) ) 1.185
BrNO
ClCN
ClNO
Cl-NO2)
CO
(1A′)
(1Σ)
-2699.56541
-552.17622
-589.16773
-664.16328
-113.02122
-188.10775
-260.20812
-109.25528
-184.20414
-129.24259
-129.55890
-204.23933
-204.56025
-167.65908
-167.54486
-297.17740
-297.17209
-297.08359
-297.17010
-297.14621
-372.17969
-372.15667
d(BrN) ) 2.211, d(NO) ) 1.157, BrNO ) 114.559d
3.8
5.2
d(ClC) ) 1.638, d(CN) ) 1.185
(1A′)
(1A1)
(1Σ)
d(ClN))2.033, d(NO) ) 1.156, (ClNO ) 114.059d
3.9
d(NO) ) 1.207, (ONCl) ) 113.422
d(CO ) 1.151
7.7
3.0
CO2
(1Σg)
(1A′)
(1Σg)
(1Σ)
d(CO) ) 1.18043
ref 61
d(NN) ) 1.131
7.2
NC-NCO
13.1
3.1
NN
N2O
d(NN) ) 1.172, d(NO) ) 1.193
d(NO) ) 1.103
6.7
3.0
NO+
(1Σ)
•NO
(2Π)
0.78 (0.75)
0.771 (0.75)
0.833 (0.75)
-129.56321
-204.56420
-167.55333
d(NO) ) 1.143
d(NO) ) 1.156
2.58
7.2
+
NO2
(1Σg)
(2A1)
(1Σ)
•NO2
d(NO) ) 1.217,43,62 (ONO) ) 133.7
d(CO) ) 1.242, d(CN) ) 1.21143
d(CN) ) 1.255, d(CO) ) 1.16743,62
Figure 3
6.3
OCN-
6.6
OCN•
(2A1)
(1A′)
(1A′)60
(1A′)64
(1A′)65
(1A′)66
(1A′)
(1A′)67
7.1
tt-OCN-NO
ct-OCN-NO
TS1-OCN-NO
TS2-OCN-NO
TS3-OCN-NO
OCN-NO2
TS-OCN-NO2
11.2
11.2
10.5
10.9
11.3
15.6
Figure 3
Figure 5
Figure 5
Figure 5
Figure 2
Figure 2
a Spin-projected energies. b Structural data at MP2/6-31G* level; references concerning different levels of theory.
techniques and have been compared with the results previously
published for the isoelectronic azides, NNN-NO2 and NNN-
NO. For both molecules, OCN-NO2 and OCN-NO, the
structural parameters and vibrational frequencies have been
characterized in order to aid future experimental attempts at
isolation. The intrinsic decomposition of OCN-NO was
calculated to be different from that of NNN-NO since no cyclic
isomer was found and the decomposition products CO2 and N2
are more likely than N2O and CO (the latter would correspond
to the decomposition products N2O and N2 for N4O).
Both molecules, nitryl isocyanate and nitrosyl cyanate, are
intrinsically stable. Although their intramolecular dissociation
is accompanied by a release of -82.9 (OCN-NO2, eq 7) and
-115.6 kcal mol-1 (OCN-NO, eq 11), it should be possible
to isolate these molecules using low-temperature techniques as
their barriers to decomposition are high enough (for OCN-
NO2 14.4 kcal mol-1; for OCN-NO 19.6 kcal mol-1).
Figure 6. Qualitative description of the OCN-NO potential energy
hypersurface.
π system by rotation about the C-N bond is accompanied by
a smaller energy barrier (4.6 kcal mol-1) and cyclic decomposi-
tion transition state (19.6 kcal mol-1, Table 6). Moreover, the
rotation into the cis structure results in an increase in the O-N
and N-C bond lengths and a decrease in the N-N and C-O
bond lengths (Figure 5, Table 6). With the molecule again
contained in a single plane, delocalized π interaction over the
entire molecule is possible and the energy of the cis chain isomer
(Figure 3) drops to near that of the trans chain isomer (Table
6). A qualitative description of the OCN-NO PES is shown
in Figure 6.
Note Added in Proof. While this paper was with the
referees, in an independent theoretical and experimental study
G. Maier et al. could for the first time isolate the predicted
NCO-OCN molecule from the photolysis reaction of carbolyl
azide isocyanate (eq 20) in an argon matrix at 10 K.63
Hν, 10 K
N3-C(O)-NdCdO
8 OdCdN-NdCdO + N2
(20)
Conclusions
Acknowledgment. The financial support of this research by
the DFG (KL 636/4-1 and -5-1) and the University of Glasgow
is gratefully acknowledged. We thank both references for most
valuable comments and helpful suggestions.
We have carried out a combined theoretical and experimental
study and investigated the reaction behavior of silver cyanate
toward NO2, NO, ClNO2, BrNO, and ClNO. On the basis of
the observed decomposition products, we propose nitryl isocy-
anate, OCN-NO2, as a likely intermediate for the NO2/ClNO2
reaction, which decomposes at room temperature forming CO2
and N2O. The AgOCN/XNO reaction (X ) Cl, Br) is a highly
complex reaction where X-CN and X(CO)NCO were experi-
mentally observed as long-lived intermediates, in addition to
NO, NO2, N2O, and CO2.
IC960707F
(63) Maier, G.; Naumann, M.; Reisenauer, H. P.; Eckwert, J. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1696.
(64) 1969, 1752, 1241, 850, 559, 493, 311, 150, and -i1172 cm-1
.
(65) 2293, 1662, 1301, 726, 642, 551, 276, 153, and -i102 cm-1
(66) 2120, 1418, 1293,1038, 777, 711, 324, 233, and -i568 cm-1
(67) 2144, 1734, 1360, 1082, 1049, 770, 759, 639, 609, 391, 179, and -i459
cm-1
.
.
Moreover, the OCN-NO2 and the OCN-NO singlet potential
energy hypersurfaces have been studied in detail using ab initio
.