Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2008 Dec 8.
Published in final edited form as: J Phys Chem A. 2006 Mar 23;110(11):4089–4092. doi: 10.1021/jp0563540

Stability of Carbon-Nitrogen Cages in Fourfold Symmetry

Douglas L Strout 1
PMCID: PMC2597175  NIHMSID: NIHMS63075  PMID: 16539433

Abstract

Molecules consisting entirely of nitrogen have been studied extensively for their potential as high energy density materials (HEDM). Many nitrogen molecules previously studied have low-energy dissociation routes and are therefore too unstable to serve as practical HEDM. However, the incorporation of heteroatoms into a nitrogen structure can have stabilizing effects. Theoretical calculations are carried out on a sequence of high-energy cages with carbon and nitrogen. Density functional theory (B3LYP), perturbation theory (MP2 and MP4), and coupled-cluster theory (CCSD(T)) are used in conjunction with the correlation-consistent basis sets of Dunning. Stability trends as a function of molecule size are calculated and discussed.

Introduction

Nitrogen molecules have been the subjects of many recent studies because of their potential as high energy density materials (HEDM). An all-nitrogen molecule Nx can undergo the reaction Nx → (x/2) N2, a reaction that can be exothermic by 50 kcal/mol or more per nitrogen atom1,2. To be a practical energy source, however, a molecule Nx would have to resist dissociation well enough to be a stable fuel. Theoretical studies37 have shown that numerous Nx molecules are not sufficiently stable to be practical HEDM, including cyclic and acyclic isomers with eight to twelve atoms. Cage isomers of N8 and N12 have also been shown710 by theoretical calculations to be unstable. Experimental progress in the synthesis of nitrogen molecules has been very encouraging, with the N5+ and N5 ions having been recently produced11,12 in the laboratory. More recently, a network polymer of nitrogen has been produced13 under very high pressure conditions. Experimental successes have sparked theoretical studies1,14,15 on other potential all-nitrogen molecules. More recent developments include the experimental synthesis of high energy molecules consisting predominantly of nitrogen, including azides16,17 of various heteroatoms and polyazido isomers18 of compounds such as 1,3,5-triazine. Future developments in experiment and theory will further broaden the horizons of high energy nitrogen research.

The stability properties of Nx molecules have also been extensively studied in a computational survey19 of various structural forms with up to 20 atoms. Cyclic, acyclic, and cage isomers have been examined to determine the bonding properties and energetics over a wide range of molecules. A more recent computational study20 of cage isomers of N12 examined the specific structural features that lead to the most stable molecules among the three-coordinate nitrogen cages. Those results showed that molecules with the most pentagons in the nitrogen network tend to be the most stable, with a secondary stabilizing effect due to triangles in the cage structure. A recent study21 of larger nitrogen molecules N24, N30, and N36 showed significant deviations from the pentagon-favoring trend. Each of these molecule sizes has fullerene-like cages consisting solely of pentagons and hexagons, but a large stability advantage was found for molecules with fewer pentagons, more triangles, and an overall structure more cylindrical than spheroidal. Studies22,23 of intermediate-sized molecules N14, N16, and N18 also showed that the cage isomer with the most pentagons was not the most stable cage, even when compared to isomer(s) containing triangles (which have 60° angles that should have significant angle strain). For each of these molecule sizes, spheroidally-shaped molecules proved to be less stable than elongated, cylindrical ones.

However, while it is possible to identify in relative terms which nitrogen cages are the most stable, it has been shown7 in the case of N12 that even the most stable N12 cage is unstable with respect to dissociation. The number of studies demonstrating the instability of various all-nitrogen molecules has resulted in considerable attention toward compounds that are predominantly nitrogen but contain heteroatoms that stabilize the structure. In addition to the experimental studies1618 cited above, theoretical studies have been carried out that show, for example, that nitrogen cages can be stabilized by oxygen insertion24,25 or phosphorus substitution26.

A study27 of carbon-nitrogen cages showed that carbon substitution into an N12 cage results in a stable N6C6H6, but larger, nitrogen-richer molecules were predicted to be less stable than N6C6H6. Replacement of the nitrogen triangles in N12 by carbon triangles led to the stable N6C6H6, but it is possible that angle strain in the triangles still influenced the stability of the molecules. In the current study, carbon-nitrogen cages are examined in fourfold symmetry rather than threefold, with carbon squares rather than triangles. Starting with a cage N8C8H8, a fourfold-symmetric counterpart to N6C6H6, a sequence of molecules will be tested to determine if this further relief of ring strain enhances the stability of the molecules. The stability of each carbon-nitrogen cage is determined by theoretical calculations of the energies of various dissociation pathways.

Computational Details

Geometries are optimized with density functional theory28,29 (B3LYP) and second-order perturbation theory30 (MP2). Single energy points are calculated with fourth-order perturbation theory30 (MP4(SDQ)) and coupled-cluster theory31 (CCSD(T)). Molecules are optimized in the singlet state, and dissociation intermediates are optimized in the triplet state, which is the ground state for all dissociations in this study. (Open-shell calculations have been carried out using UHF. Spin contamination is not a significant problem. For example, for the MP2/cc-pVDZ calculations on the three intermediates of N8C8H8, the S2 values are all between 2.03 and 2.06.) Vibrational frequencies have been calculated at the MP2/cc-pVDZ level of theory for N8C8H8 and its dissociation intermediates. B3LYP/cc-pVDZ frequencies have been calculated for all intact molecules in this study. The basis sets are the double-zeta (cc-pVDZ), augmented double-zeta (aug-cc-pVDZ), and triple-zeta (cc-pVTZ) sets of Dunning32. The Gaussian03 computational chemistry software33 has been used for all calculations in this study.

Results and Discussion

N8C8H8

The smallest molecule in this study is the N8C8H8 molecule shown in Figure 1. The molecule has D4d point group symmetry and three symmetry-independent bonds. The bond dissociation energies for these bonds have been tabulated in Table 1. As with the previously-studied N6C6H6, the NN bond is the weakest, but its dissociation energy is over 30 kcal/mol at the CCSD(T)/cc-pVDZ level of theory, which should be considered the most reliable calculation in Table 1. Basis set effects, calculated with MP2 theory, indicate that larger basis sets increase the bond-breaking energies slightly. Inclusion of zero-point energy decreases the bond energies, but the decrease is less than the increasing effect of the larger basis sets for the NN bond. Therefore, the actual dissociation energy for the NN bond is most likely even higher than the best cc-pVDZ calculations would indicate. The other bonds have even higher energies, so the N8C8H8 molecule should resist dissociation well enough to be a stable high energy density material.

Figure 1.

Figure 1

N8C8H8 molecule, with three symmetry-independent bonds labeled. The molecule has D4d point group symmetry. Nitrogen atoms are shown in white, carbon atoms in gray, and hydrogen atoms are in white on the ends of the molecule.

TABLE 1.

Bond dissociation energies for the N8C8H8 cage (energies in kcal/mol).

Bonds (see Figure 1)
Energy Geometry CC CN NN
B3LYP/cc-pVDZ B3LYP/cc-pVDZ +65.8 +72.9 +25.0
MP2/cc-pVDZ MP2/cc-pVDZ +78.1 +92.3 +44.4
MP2(+ZPE)/cc-pVDZ MP2/cc-pVDZ +75.5 +89.7 +43.7
MP4/cc-pVDZ MP2/cc-pVDZ +72.3 +83.6 +35.5
CCSD(T)/cc-pVDZ MP2/cc-pVDZ +71.0 +80.3 +34.2
MP2/aug-cc-pVDZ MP2/aug-cc-pVDZ +78.2 +93.5 +46.8
MP2/cc-pVTZ MP2/cc-pVTZ +80.0 +95.5 +46.8
MP4/aug-cc-pVDZ MP2/aug-cc-pVDZ +72.3 +84.4 +37.4

Larger molecules

Since it is the nitrogen content of the molecules that provides the energy release, it is favorable to study molecules richer in nitrogen than the N8C8H8, which is only 52% nitrogen by mass. The basic design of the N8C8H8 has been extended to the larger molecules N16C12H8 and N24C16H8. These larger molecules are shown in Figures 2 and 3 and are composed of 60% and 63% nitrogen by mass, respectively. However, in the threefold symmetry study, lengthening the molecule resulted in decreasing stability. Are these larger fourfold-symmetric molecules as stable as the smaller N8C8H8? Bond-breaking energies for symmetry-independent bonds (five in N16C12H8 and seven in N24C16H8) are tabulated in Tables 2 and 3. For these molecules, as with N8C8H8, the weakest bonds are the N-N bonds. However, the bond-breaking energies are about the same as for N8C8H8. The stability of these fourfold-symmetric molecules is not dependent on molecule size, which stands in contrast with the previously-studied threefold-symmetric molecules. The molecules in the current study are all stable enough to serve as high energy density materials (HEDM).

Figure 2.

Figure 2

N16C12H8 molecule, with five symmetry-independent bonds labeled. The molecule has D4h point group symmetry. Nitrogen atoms are shown in white, carbon atoms in gray, and hydrogen atoms are in white on the ends of the molecule.

Figure 3.

Figure 3

N24C16H8 molecule, with seven symmetry-independent bonds labeled. The molecule has D4d point group symmetry. Nitrogen atoms are shown in white, carbon atoms in gray, and hydrogen atoms are in white on the ends of the molecule.

TABLE 2.

Bond dissociation energies for the N16C12H8 cage (energies in kcal/mol).

Bonds (see Figure 5)
Energy Geometry CC1 CN1 NN CN2 CC2
B3LYP/cc-pVDZ B3LYP/cc-pVDZ a a +25.5 a +72.8
MP2/cc-pVDZ MP2/cc-pVDZ +79.2 +96.6 +45.3 +109.8 +92.3
MP4/cc-pVDZ MP2/cc-pVDZ +73.2 +87.1 b +101.1 +84.7
a

Geometry optimization not successful

b

Calculation exceeds resources, due to C1 symmetry of intermediate

TABLE 3.

Bond dissociation energies for the N24C16H8 (energies in kcal/mol)

Bonds (see Figure 6) B3LYP/cc-pVDZ MP2/cc-pVDZ
CC1 a +79.2
CN1 a +96.7
NN1 +25.6 +45.1
CN2 a +108.2
CC2 +72.6 +92.5
CN3 a +114.7
NN2 +25.1 +45.1
a

Geometry optimization not successful

Comparison to azido compounds

Although these results indicate the possibility of an entire class of stable high-energy molecules, the general stoichiometry of these molecules in the limit of large sizes is N2C. This means the class of fourfold-symmetric molecules has a theoretical upper limit of 70% nitrogen by mass. As stated previously, much research effort is focused on the development of high-energy azido or polyazido compounds, some of which have nitrogen composition well above 70%. The azido functionality is a straightforward way to enrich a molecular structure with nitrogen. It is therefore reasonable to ask how the molecules in Figures 13 compare with azido compounds in terms of energy release. Table 4 shows the energy release data for unimolecular dissociation of the three molecules in this study along with an appropriate azido compound for reference, namely triazidomethane (N9CH). (Note: For unimolecular dissociation, it is assumed that the carbon becomes graphitic in nature. Graphitic compounds, namely benzene (C6H6) and coronene (C24H12) are chosen as products.) Triazidomethane is 91% nitrogen by mass, far richer than any molecule in this study, but it is no more energetic than N16C12H8 or N24C16H8, which are both less than 65% nitrogen. The reason for this is a straightforward comparison between the N-N single bonds in the molecules in the current study and the N=N double bonds in the N=N=N azido bonding group, which release less energy34 upon conversion to the triple bonds in molecular N2. Therefore, the molecules in this study have energy release properties comparable even to other high-energy molecules that are richer in nitrogen.

TABLE 4.

Free energies of unimolecular dissociation of the compounds in this study, with triazidomethane (N9CH) included for comparison. Energies calculated using B3LYP/cc-pVDZ method. Mass percentages of nitrogen are shown for each molecule. Benzene (C6H6) and coronene (C24H12) are graphitic products that are chosen to account for carbon and hydrogen in the decomposition.

Molecule % N Reaction kcal/mol kcal/g
N8C8H8 52 N8C8H8 →4 N2+ (4/3) C6H6 −339.6 −1.6
N16C12H8 60 N16C12H8 →8 N2+ (2/3) C6H6+ (1/3) C24H12 −683.0 −1.8
N24C16H8 63 N24C16H8 →12 N2+ (2/3) C24H12 −1028.6 −1.9
N9CH 91 N9CH →(9/2) N2+ (1/6) C6H6 −243.3 −1.8

Table 5 shows the results of a similar energetic comparison, the difference being the assumption of an oxygen atmosphere. Such an atmosphere would permit not only the decomposition of the nitrogen into N2 but also combustion of the hydrocarbon content. The energies in Table 5, therefore, include the energies from Table 4 and an additional contribution from the carbon. Not surprisingly, the data in Table 5 vary strongly by the carbon content of each molecule. The triazidomethane had the smallest hydrocarbon contribution to the energy because it has the smallest percentage of hydrocarbon content. The energies of the fourfold-symmetric molecules in this study decrease with increasing size because of decreasing hydrocarbon content. In an oxygen atmosphere, the hydrocarbon content of the molecules in this study provide an even greater energy release relative to azido compounds that are richer in nitrogen.

TABLE 5.

Free energies of dissociation of the compounds in this study, with triazidomethane (N9CH) included for comparison. The presence of an oxygen atmosphere is assumed. Energies calculated using B3LYP/cc-pVDZ method. Mass percentages of nitrogen are shown for each molecule.

Molecule % N Reaction kcal/mol kcal/g
N8C8H8 52 N8C8H8+ 10 O2 →4 N2+ 8 CO2+ 4 H2O −1644.2 −7.6
N16C12H8 60 N16C12H8+ 14 O2 →8 N2+ 12 CO2+ 4 H2O −2514.1 −6.7
N24C16H8 63 N24C16H8+ 18 O2 →12 N2+ 16 CO2+ 4 H2O −3386.3 −6.3
N9CH 91 N9CH + (5/4) O2 →(9/2) N2+ CO2+ (1/2) H2O −406.4 −2.9

Conclusion

These fourfold-symmetric molecules represent an entire class of high-energy materials with no significant variations in stability with respect to molecule size. Successive stacking of eight-membered rings of nitrogen with four-membered rings of carbon results in molecules of arbitrary size, as well as substantial energy release properties. Also, the conversion of N-N single bonds to triple bonds is sufficiently energetic to overcome the limitations of the N2C stoichiometry of the molecules in the current study. N8C8H8, N16C12H8, and N24C16H8, as well as their larger analogues, all have the ability to serve as high energy density materials (HEDM).

Supplementary Material

1si20051222_04

Acknowledgments

The Alabama Supercomputer Authority is gratefully acknowledged for a grant of computer time on the SGI Altix operated in Huntsville, AL. This work is also supported by the National Institutes of Health (NIH/NCMHD grant 1P20MD000547-01). This work was partially supported by the National Computational Science Alliance under grant number CHE050022N and utilized the IBM p690 cluster in Champaign, Illinois. The taxpayers of the state of Alabama in particular and the United States in general are also gratefully acknowledged.

References and Notes

  • 1.Fau S, Bartlett RJ. J Phys Chem A. 2001;105:4096. [Google Scholar]
  • 2.Tian A, Ding F, Zhang L, Xie Y, Schaefer HF., III J Phys Chem A. 1997;101:1946. [Google Scholar]
  • 3.Chung G, Schmidt MW, Gordon MS. J Phys Chem A. 2000;104:5647. [Google Scholar]
  • 4.Strout DL. J Phys Chem A. 2002;106:816. [Google Scholar]
  • 5.Thompson MD, Bledson TM, Strout DL. J Phys Chem A. 2002;106:6880. [Google Scholar]
  • 6.Li QS, Liu YD. Chem Phys Lett. 2002;353:204. [Google Scholar]; Li QS, Qu H, Zhu HS. Chin Sci Bull. 1996;41:1184. [Google Scholar]
  • 7.Li QS, Zhao JF. J Phys Chem A. 2002;106:5367. [Google Scholar]; Qu H, Li QS, Zhu HS. Chin Sci Bull. 1997;42:462. [Google Scholar]
  • 8.Gagliardi L, Evangelisti S, Widmark PO, Roos BO. Theor Chem Acc. 1997;97:136. [Google Scholar]
  • 9.Gagliardi L, Evangelisti S, Bernhardsson A, Lindh R, Roos BO. Int J Quantum Chem. 2000;77:311. [Google Scholar]
  • 10.Schmidt MW, Gordon MS, Boatz JA. Int J Quantum Chem. 2000;76:434. [Google Scholar]
  • 11.Christe KO, Wilson WW, Sheehy JA, Boatz JA. Angew Chem Int Ed. 1999;38:2004. doi: 10.1002/(SICI)1521-3773(19990712)38:13/14<2004::AID-ANIE2004>3.0.CO;2-7. [DOI] [PubMed] [Google Scholar]
  • 12.Vij A, Pavlovich JG, Wilson WW, Vij V, Christe KO. Angew Chem Int Ed. 2002;41:3051. doi: 10.1002/1521-3773(20020816)41:16<3051::AID-ANIE3051>3.0.CO;2-T. [DOI] [PubMed] [Google Scholar]; Butler RN, Stephens JC, Burke LA. Chem Commun. 2003;8:1016. doi: 10.1039/b301491f. [DOI] [PubMed] [Google Scholar]
  • 13.Eremets MI, Gavriliuk AG, Trojan IA, Dzivenko DA, Boehler R. Nature Materials. 2004;3:558. doi: 10.1038/nmat1146. [DOI] [PubMed] [Google Scholar]
  • 14.Fau S, Wilson KJ, Bartlett RJ. J Phys Chem A. 2002;106:4639. [Google Scholar]
  • 15.Dixon DA, Feller D, Christe KO, Wilson WW, Vij A, Vij V, Jenkins HDB, Olson RM, Gordon MS. J Am Chem Soc. 2004;126:834. doi: 10.1021/ja0303182. [DOI] [PubMed] [Google Scholar]
  • 16.Knapp C, Passmore J. Angew Chem Int Ed. 2004;43:4834. doi: 10.1002/anie.200401748. [DOI] [PubMed] [Google Scholar]
  • 17.Haiges R, Schneider S, Schroer T, Christe KO. Angew Chem Int Ed. 2004;43:4919. doi: 10.1002/anie.200454242. [DOI] [PubMed] [Google Scholar]
  • 18.Huynh MV, Hiskey MA, Hartline EL, Montoya DP, Gilardi R. Angew Chem Int Ed. 2004;43:4924. doi: 10.1002/anie.200460366. [DOI] [PubMed] [Google Scholar]
  • 19.Glukhovtsev MN, Jiao H, Schleyer PvR. Inorg Chem. 1996;35:7124. doi: 10.1021/ic9606237. [DOI] [PubMed] [Google Scholar]
  • 20.Bruney LY, Bledson TM, Strout DL. Inorg Chem. 2003;42:8117. doi: 10.1021/ic034696j. [DOI] [PubMed] [Google Scholar]
  • 21.Strout DL. J Phys Chem A. 2004;108:2555. [Google Scholar]
  • 22.Sturdivant SE, Nelson FA, Strout DL. J Phys Chem A. 2004;108:7087. [Google Scholar]
  • 23.Strout DL. J Phys Chem A. 2004;108:10911. [Google Scholar]
  • 24.Strout DL. J Phys Chem A. 2003;107:1647. [Google Scholar]
  • 25.Sturdivant SE, Strout DL. J Phys Chem A. 2004;108:4773. [Google Scholar]
  • 26.Strout DL. J Chem Theory Comput. 2005;1:561. doi: 10.1021/ct050067i. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Colvin KD, Cottrell R, Strout DL. J Chem Theory Comput. (published ASAP October 29, 2005) [Google Scholar]
  • 28.Becke AD. J Chem Phys. 1993;98:5648. [Google Scholar]
  • 29.Lee C, Yang W, Parr RG. Phys Rev B. 1988;37:785. doi: 10.1103/physrevb.37.785. [DOI] [PubMed] [Google Scholar]
  • 30.Moller C, Plesset MS. Phys Rev. 1934;46:618. [Google Scholar]
  • 31.Purvis GD, Bartlett RJ. J Chem Phys. 1982;76:1910. [Google Scholar]; Scuseria GE, Janssen CL, Schaefer HF., III J Chem Phys. 1988;89:7382. [Google Scholar]
  • 32.Dunning TH., Jr J Chem Phys. 1989;90:1007. [Google Scholar]
  • 33.Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Montgomery JA, Jr, Vreven T, Kudin KN, Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross JB, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P, Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, Strain MC, Farkas O, Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cioslowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW, Gonzalez C, Pople JA. Gaussian 03, Revision B.01. Gaussian, Inc; Pittsburgh PA: 2003. [Google Scholar]
  • 34.Bond energies from Physical Chemistry, 7th ed., by Atkins and de Paula (p. 1096), for N-N bonds are 163, 409, and 946 kJ/mol, respectively, for single, double and triple bonds. Therefore, conversion of a double bond to triple (537 kJ/mol) is over 30% less energetic than conversion of a single bond to a triple bond (783 kJ/mol).

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1si20051222_04

RESOURCES