Ti(N5)4 as a Potential Nitrogen-Rich Stable High-Energy Density

Subscriber access provided by Weizmann Institute of Science

Article 5

4

Ti(N) as a Potential Nitrogen Rich Stable High Energy Density Material Changhyeok Choi, Hae-Wook Yoo, EunMee Goh, Soo Gyeong Cho, and Yousung Jung J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.6b04226 • Publication Date (Web): 06 Jun 2016 Downloaded from http://pubs.acs.org on June 11, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry A is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Ti(N5)4 as a Potential Nitrogen Rich Stable High Energy Density Material

Changhyeok Choi,† Hae-Wook Yoo,‡ Eun Mee Goh, †‡ Soo Gyeong Cho,‡* Yousung Jung†* †

Graduate school of Energy Environment Water and Sustainability (EEWS), Korea

Advanced Institute of Science and Technology (KAIST), 291 Daehakro, Daejeon 305-701, Korea ‡

Agency for Defense Development, Yuseong P.O Box 35-42, Daejeon 34186, Korea

Abstract We have studied molecular structures and kinetic stability of M(N5)3 (M=Sc,Y) and M(N5)4 (M=Ti,Zr,Hf) complexes theoretically. All of these compounds are found to be stable with more than a 13 kcal/mol of kinetic barrier. In particular, Ti(N5)4 showed the largest dissociation energy of 173.0 kcal/mol and thermodynamic stability. This complex had a high nitrogen content (85% by weight), and a significantly high nitrogen to metal ratio (20:1) among the neutral M(N5)n species studied here and in literature. Ti(N5)4 is thus forecasted to be a good candidate for a nitrogen rich high energy density material (HEDM). We reveal in further detail using ab initio molecular dynamics simulations that the dissociation pathways of M(N5)n involve the rearrangements of the bonding configurations before dissociation.

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

1. Introduction Nitrogen-rich compounds and polynitrogens which contain a large number of N-N and N=N bonds have been pursued as a potential high performance high energy density material (HEDM) both theoretically and experimentally.1-2 Due to significantly smaller bond dissociation energies of the N-N (38.4 kcal/mol) and N=N (100 kcal/mol) bonds than that of the N≡N triple bond (229 kcal/mol), converting these two types of bonds of low order to the N≡N bonds releases a large amount of energy.1 In addition, the main product of explosion or rapid combustion of nitrogen-rich compounds without consuming additional oxygen is an environmentally benign nitrogen molecule (N2), and for this reason, nitrogen-rich compounds including polynitrogens are sometimes considered as green HEDMs. Extensive synthetic efforts have been made to make nitrogen-rich compounds, particularly utilizing the aromatic nitrogen heterocycles. The synthesis of polynitrogen compounds has been extremely challenging although many theoretical studies have shown that some polynitrogen compounds can have local minimum structures and thus have a possibility of synthesis.3-10 However, the actual experimental observations of these compounds have been only scarcely reported to a very limited extent. For example, the synthesis of N5+ cation was reported in 1999,5 and in 2002, the N4 molecule was detected experimentally but its lifetime -

barely exceeded 1 μs.4 In the same year, the pentazole anion (N5 ring) was also detected.3 -

Among the polynitrogen anions, N5 shows an acceptable theoretical kinetic stability of -

27.8 kcal/mol toward the dissociation into N3 and N2 that comes from an additional stability -

-

of 6π electrons aromaticity in N5 .2 It is notable that N5 has the same geometry (D5h -

symmetry) as, and is also isoelectronic with the cyclopentadienyl anion (C5H5 ) that easily 2


Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


coordinates with a metal cation as in the well-known metallocene compounds. Thus, similarly -

to C5H5 ,

-

N5 can also be expected to form coordination complexes with the metal cations;

earlier studies indeed reported stable structures for ScN7 and N5MN7 (M=Ti, Zr, Hf, Th) by 3-

3-

-

-

noting the similarity between C7H7 and N7 , and C5H5 and N5 that are all aromatic.11-12 Following these results, M(N5)n complexes have been studied. For example, alkali metals (Li, Na, K, Rb) and alkali earth metals (Mg, Ca, Sr, Ba) were considered to form M(N5) and +

M(N5)2 compounds, respectively, and cationic MN5 complexes with the latter divalent alkali earth metals were also suggested.13-15 As transition metal analogues, as described above, q

ferrocene-like Fe(N5)2 and Ni(N5)2 structures and charged M(N5)n species were also proposed theoretically.16-18 As briefly summarized above, previous studies to find the metal-containing polynitrogen compounds have mainly been focused on M(N5)n complexes with n < 3.14, 16-18 If there are -

stable species that contain more N5 units than previously suggested, it would of course be -

more desirable as a practical HEDM since the more N5 contents would mean the higher energetic performances. In this study, we investigated the structures and stability of M(N5)3 and M(N5)4 species. In -

-

particular, based on the similarity between N5 and C5H5 described above, we identified the transition metal elements that have been experimentally proven to exist as M(C5H5)3 and M(C5H5)4 as a possible M(N5)3 and M(N5)4 HEDM candidate. So far, for group 3 metals, M(C5H5)3 complexes with M = Sc, Y were reported experimentally, and for group 4 metals, M(C5H5)4 complexes with M = Ti, Zr, Hf exist in experimental literatures.19-21 Thus in this paper, the structures and stability of M(N5)3 (M=Sc, Y) and M(N5)4 (M=Ti, Zr, Hf) were 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

investigated as a potential high nitrogen containing energetic material using density functional calculations. In addition, using ab initio molecular dynamics (AIMD) simulations, the long-believed dissociation pathways of M(N5)n species were also verified for the first time which indicate that the dissociation of these compounds often requires the rearrangements of the ground-state bonding configurations before dissociation. 2. Computational details All calculations were performed using Q-CHEM quantum chemistry package.22 Geometry optimization and frequency calculations were performed using the B3LYP functional with D3-dispersion correction (B3LYP-D3).23-26 We used the 6-31G* basis set for N, Sc, Ti and LANL2DZ effective core-potential (ECP) basis for the second and third row transition metal elements (Y, Zr, Hf).27-29 The calculated energy difference between with (6-31+G*) and without diffuse function (6-31G*) was small (~0.5 kcal/mol) for the first dissociation kinetic barrier of Ti(N5)4, thus we used 6-31G* throughout this study to minimize the computational cost. Minimum energy structures were confirmed by all positive Harmonic frequencies. Transition states were characterized by having one imaginary frequency and confirmed by Intrinsic Reaction Coordinates (IRC) calculations. Natural Bond Orbital (NBO) analysis implemented in Q-CHEM was used for atomic population analysis.30 Born-Oppenheimer molecular dynamics (BOMD) calculations at the same level with static DFT calculations (B3LYP-D3/6-31G*) were used to follow the dissociation pathways.

3. Results and discussions 3.1. Geometry 4


Page 4 of 22

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


-

Since N5 has five coordination sites, it can form ƞ5 bonding at maximum. In previous studies, ƞ1, ƞ2, and ƞ5 types of M-N5 bonds were reported in M(N5)n species.11-18 For n=2, for example, ƞ5 (that resemble ferrocene) and ƞ2 type species were reported.15-18 For n>2, ƞ1 and ƞ2 bonding configurations were reported.16 Thus, we here considered all possible combinations of ƞ1, ƞ2 and ƞ5 bonding configurations in M(N5)3 and M(N5)4 structures as initial geometries in order to find the most stable isomers. The lowest energy structures thus found for all M(N5)3 and M(N5)4 compounds had the ƞ2 bonding configurations only. For Ti(N5)4, however, another local minimum structure (5.9 kcal/mol higher in energy than the lowest energy structure) that consists of two ƞ1 bondings and two ƞ2 bondings was also found. The lowest energy M(N5)3 and M(N5)4 structures shown in Fig. 1 have a D3h and D2d point group symmetry, respectively. In Table 1, typical bond lengths for the optimized structures are summarized (See Fig. 1 for atomic labeling). By symmetry, the M-N and N-N bond lengths in each ring are the same. In M(N5)3 structures, two M-N bond lengths are the same with the ƞ2 type bonding (Fig. 1). Going from Sc to Y, the M-N bond length is increased by 0.19 Å. In both Sc(N5)3 and Y(N5)3, the N1-N2 bond (1.346Å and 1.350 Å, respectively) is the longest bond among the N-N -

bonds in N5 ring. Since the bond strength is often qualitatively inversely proportional to bond -

length, this N1-N2 bond might be the weakest bond in N5 ring which breaks first at an initial stage of chemical reaction (e.g., explosion). The second longest bond is N3-N4 (or N4-N5). Using the Mayer bond order analysis, the calculated bond orders for N1-N2 and N3-N4 are 1.1 and 1.3, respectively, for both Sc(N5)3 and Y(N5)3. For comparison, in case of an isolated -

N5 , the N-N bond order is 1.3 and its length is 1.330 Å, suggesting that the N1-N2 bond in Sc(N5)3 and Y(N5)3 is weaker and the dissociation will occur mainly through the elongation 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of this bond. In M(N5)4 structures, the M-N bonds are also ƞ2 type. Going from Ti to Zr, the M-N1 and M-N2 bond lengths are increased by 0.18 Å and 0.14 Å, respectively, whereas they are decreased slightly by 0.028 Å and 0.033 Å, respectively, going from Zr to Hf. Contrary to M(N5)3, the optimized M(N5)4 structure shows different bond lengths for N3-N4 and N4-N5. For Ti(N5)4, N3-N4 is the longest N-N bond and N1-N2 is the second longest N-N bond. Other M(N5)4 species show that N1-N2 and N3-N4 are the longest and second longest bonds, respectively. The Mayer bond order for the N1-N2 and N3-N4 bonds are 1.1 and 1.3, respectively, for all M(N5)4 species, consistent with the trend in bond lengths. Interestingly, however, in Ti(N5)4, the bond order of N1-N2 is smaller than that of N3-N4 although the N1-N2 bond (1.322 Å) is shorter than N3-N4 (1.333 Å). It can be understood from a low electron occupancy in the N1-N2 bonding orbital in Ti(N5)4; based on the Natural Bond Orbital (NBO) analysis, the occupation number (ON) of the N1-N2 bonding orbital in Ti(N5)4 is 1.82, smaller than ON = 1.98 for all other M(N5)4 species, yielding an approximate bond order of (1.82-0.082)/2 = 0.87. As in the M(N5)3 structures described above, therefore, the weakest N1-N2 and N3-N4 bonds are likely to undergo dissociation first in M(N5)4. Based on these bond lengths and bond orders, possible dissociation pathways are considered in the next section.

6


Page 6 of 22

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Optimized structures of M(N5)3 (M = Sc, Y) and M(N5)4 (M = Ti, Zr, Hf) species. Color codes: blue = nitrogen, gray = central metal.

Table 1. Bond lengths for the optimized M(N5)3 and M(N5)4 structures. All values are in Å. Species

M-N1

M-N2

N1-N2

N2-N3

N3-N4

N4-N5

N5-N1

Sc(N5)3

2.136

2.136

1.346

1.309

1.326

1.326

1.309

Y(N5)3

2.326

2.326

1.350

1.309

1.326

1.326

1.309

Ti(N5)4

2.023

2.114

1.322

1.309

1.333

1.321

1.315

Zr(N5)4

2.203

2.256

1.337

1.307

1.330

1.323

1.311

Hf(N5)4

2.175

2.223

1.339

1.307

1.329

1.323

1.310

3.2. Dissociation pathways and kinetic stability Many previous researches have shown that the dissociation of M(N5)n is accompanied by -

-

the breaking of N5 rings.13-18 We thus considered the same dissociation pathways that N5 is -

dissociated into N2 molecule and N3 bonded to a metal cation. In particular, we studied the -

sequential ring opening mechanism in which each N5 ring is dissociated one by one starting 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

from the lowest energy M(ƞ2-N5)n structures, namely in a sequence of M(ƞ2-N5)3 → M(ƞ2N5)2(N3) + N2 → M(ƞ2-N5)(N3)2 + 2N2 → M(N3)3 + 3N2, for example. On the basis of bond lengths and bond orders discussed in the previous section (Table 1), the N1-N2 and N3-N4 -

bonds in N5 will likely break for all metals. Thus, we considered the transition states (TS) that show elongations in the N1-N2 and N3-N4 bonds for each ring dissociation reaction. The dissociation barrier determines the life time and feasibility for synthesis, detection, and further applications and usages. Kinetic barriers of each dissociation step (Ea1, Ea2, Ea3, Ea4) are thus summarized in Table 2. In the same table, we also compare thermodynamic -

stabilities of these compounds, in terms of total dissociation energy (∆Er), metal-N5 binding energy (∆Eb), and the formation energy (∆Ef). All three quantities are related but defined slightly differently as: ∆Er = E(N(N3)n) + nE(N2) - E(M(N5)n), -

∆Eb = -(E(M(N5)n) – E(Mn+) -nE(N5 )), and ∆Ef = E(M(N5)n) – E(M) – 5n/2E(N2). In addition to total dissociation energy (∆Er), reaction energies of each dissociation step (∆Ern) are summarized in SI (Table S1). We also note that, since the reaction energy of a single polynitrogen ring dissociation is usually much higher than the kinetic barrier for each ring opening (also shown here in Table 2),2, 11-16, 18 as long as the first dissociation occurs it can induce sequential dissociation of other polynitrogen rings. So, the kinetic barrier for the first ring opening is often used as a reasonable measure of the kinetic stability of these compounds.11-12, 16 We followed the same convention in this study to estimate the kinetic 8


Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


stability. We first consider the dissociation pathways for M(N5)3 (Fig. 2a). As expected, dissociation of the M(N5)3 species begins with the simultaneous elongation of N1-N2 and N3-N4 with the corresponding transition state shown in Fig. 2a. Compared to the N1-N2 bond in the optimized reactant structures (Table 1), the same bond in TS significantly elongates by 0.594Å and 0.581Å for Sc(N5)3 and Y(N5)3, respectively. Similarly, the N3-N4 bond elongates by 0.149Å and 0.163Å, respectively. The result of this concerted first dissociation -

is thus N2 (N2-N3) gas and N3 (N1-N5-N4) bonded to metal cation. The subsequent -

decompositions of the remaining N5 rings also occur in the same manner as the first dissociation as in Fig. 3, with the final products being M(N5)3 + 3N2. We note an interesting tendency of gradually increased kinetic barriers (Table 2) as the dissociation proceeds; M(N5)3 (15.1 for Sc, 16.1 for Y) vs. M(N5)2(N3) (16.1 for Sc, 16.9 for Y) vs. M(N5)(N3)2 (16.8 for Sc, 17.6 for Y).

Figure 2. Transition states for the first dissociation step of M(N5)3 and M(N5)4. For M(N5)3, the TS has the ƞ2 bonding configuration, but for M(N5)4, there are two types of TS, (b) one with ƞ1 and (c) the other with ƞ2. Numbers indicate the bond lengths in Å (color codes in 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

numbers are the same as those of metallic labeling at the bottom).

Figure 3. Dissociation pathways for Sc(N5)3. Dissociated dinitrogen is omitted for clarity. The TS and intermediate structures shown in this figure correspond to Sc(N5)3, but those for Y(N5)3 are similar (See Fig. S1 in SI). Unlike M(N5)3 species, there are two different dissociation mechanisms for a dissociation of M(N5)4. One is the dissociation of the N1-N2 and N3-N4 after isomerization (Path 1), the other is the direct and concerted dissociation of N1-N2 and N3-N4 (Path 2) as in M(N5)3. In Path 1 (Fig. 2b), the M(ƞ2-N5)4 species first isomerizes to M(ƞ1-N5)(ƞ2-N5)3, then dissociates. The latter intermediate isomer (confirmed as a minimum by the IRC calculations followed by Hessian), however, is almost metastable (extremely flat potential energy surfaces) and we were not able to find the transition state that connects this structure to the reactant minimum with ƞ2 configuration. By contrast, The Path 2 (Fig. 2c) is similar to the first TS of M(N5)3 shown in Fig. 2a, in which the TS exhibits the simultaneous elongation of the N1-N2 -

and N3-N4 bonds in ƞ2-N5 . The full energy diagrams for both pathways are depicted in Fig. 4 10


Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in which the final dissociation products are M(N3)4 + 4N2. In order to distinguish intermediates (denoted as I), TS, and Ea for Path 1 (Fig. 4a) vs. Path 2 (Fig. 4b), we specify the bonding configuration in the notations in addition. For example, the first transition sate for Path 1 that occurs stepwise is denoted as ƞ1-TS1, and when needed to further specify the metal species such as Ti(N5)4, we use ƞ1-TS1(Ti). For Ti, the stepwise bond-rearranged mechanism (Path 1) is preferred with ƞ1-Ea1 = 13.2 kcal/mol, while for Zr and Hf, the concerted mechanism (Path 2) is preferred with ƞ2-Ea1 = 16.6 and 16.3 kcal/mol, respectively. In all cases, ƞ2-I is more stable than ƞ1-I although the energy difference between these two intermediates are relatively small, < 8 kcal/mol. Although Ti(N5)4 seems to have a lower kinetic stability (13.2 kcal/mol) compared to Zr and Hf-analogues (~16 kcal/mol), Ti shows the highest total released energy of 173.0 kcal/mol upon dissociation (∆Er in Table 2). The total dissociation energies for M(N5)4 are -

generally higher than those of M(N5)3 owing to the larger number of coordinated N5 rings. If -

one normalizes these values per N5 ring, Ti(N5)4 (-43.3 kcalmol), Zr(N5)4 (-40.5 kcal/mol) and Hf(N5)4 (-40.5 kcal/mol) still show higher unit dissociation energy than Sc(N5)3 (-30.5 -

kcal/mol) and Y(N5)3 (-29.6 kcal/mol). Ti(N5)4 also shows the highest N5 binding energy (∆Eb) of 2158.8 kcal/mol, compared to other M(N5)4 species having less than 1900 kcal/mol of ∆Eb. In terms of the formation energy (a similar quantity to Eb), Ti(N5)4 again shows the best HDEM performance among the species in this study from a thermodynamic point of view. Compared with several polyazides that have been previously predicted and later synthesized, Ti(N5)4 also shows a higher energetic performance. We calculated the formation energies of some experimentally observed neutral M(N3)n complexes such as Ti(N3)4, Nb(N3)5 and Ta(N3)5.31-33 The formation energies, defined by E(M(N3)n) – E(M) – 3n/2 E(N2), 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

for Ti(N3)4, Nb(N3)5 and Ta(N3)5 are 32.3, 97.3 and 52.1 kcal/mol, respectively. Converting these values to per-nitrogen basis, the formation energies per N for Ti(N3)4, Nb(N3)5, Ta(N3)5 are 2.7, 6.5 and 3.5 (in kcal/mol), respectively, which are lower than the formation energy per N for Ti(N5)4, 10.3 kcal/mol. This comparison suggests that Ti(N5)4 would have a higher energetic performance compared to these previous azide-based compounds for a given number of nitrogen contents. In addition, we note that Ti(N5)4 has a high metal to nitrogen ratio (1:20) with a significant nitrogen content of 85% by weight, the highest values among the reported neutral M(N5)n complexes in literature.11-18 Laser ablation methods have often been used to synthesize MNn species using N2 as a nitrogen source after theoretical predictions.34-44 For Sc and group 4 metals (Ti, Zr, Hf) considered in this paper, several MNn species including TiN12 have been observed.39-40, 44 Considering a large formation energy of Ti(N5)4 using N2 as a reactant (Table 2), however, a direct synthesis from Ti and N2 appears rather difficult. Alternatively, since Ti4+ can bind N5

-

very strongly (Table 2), the synthesis of Ti(N5)4 from a reaction between Ti4+ and N5

-

-

derivatives (e.g. C6H5-N5 derivatives) may be potentially possible.3, 45-46

12


Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Two possible dissociation pathways for Zr(N5)4: (a) Path 1 that involves a change of bonding type from ƞ2 in the reactant to ƞ1 in TS, and (b) Path 2 that involves a direct N1-N2 dissociation. Dinitrogen is omitted for clarity. The TS and intermediate structures shown in this figure correspond to Zr(N5)4, but those for other M(N5)4 are similar (See Fig. S2 and Fig. S3 in SI).

Table 2. Summary of the kinetic barriers (Ea1, Ea2, Ea3, Ea4), total dissociation energy (∆Er), -

the metal-N5 binding energy (∆Eb), and the formation energy (∆Ef), in unit of kcal/mol. Path 1 (ƞ1-Ea)

Path2 (ƞ2-Ea) ∆Er

∆Ef

∆Eb

16.8

-91.5

94.6

1094.2

16.9

17.6

-88.7

97.0

964.6

13.2

13.7

14.4

14.8

18.5

Zr(N5)4

19.5

18.9

19.5

19.3

16.6

Hf(N5)4

20.7

19.2

19.5

18.8

16.3

species

Ea1

Ea2

Ea3

Sc(N5)3

15.1

16.1

Y(N5)3

16.1

Ti(N5)4a

Ea4

Ea1

Ea2

Ea3

Ea4

17.9

19.5

-173.0

205.3

2158.8

16.1

16.5

18.7

-161.9

172.4

1839.8

15.7

18.1

17.8

-161.9

155.2

1838.1

13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

a

Page 14 of 22

We were not able to locate ƞ2-TS2 of Path2 for Ti despite an extensive search.

3.3. Ab initio Molecular Dynamics According to the previous theoretical studies, the dissociation of M(N5)n species occurs via a sequential M(N5)n → M(N5)n-1 + N2 pathway.13-18 However, the dynamical aspects of this proposed dissociation pathway have not been addressed in literature. To comprehend the dissociation pathways, we performed the AIMD simulations for Ti(N5)4 since it shows the higher energetic performance than any other species considered in this study. In order to overcome the kinetic barriers within a short simulation time (due to a high computational cost of long time simulations) the AIMD simulation was performed in high temperature (2000 K). The time step of simulation is 42 a.u. (~1 fs), and the simulation was run for 2.4 ps. Major snapshots of the AIMD trajectory are shown in Fig. 5. The dissociation pathway of Ti(N5)4 is shown in Fig. 5. As expected from the previous sections, elongation of the N1-N2 bond is first observed (5a). After the N1-N2 bond -

-

lengthening occurs, the ƞ2-N5 configuration changes to ƞ1-N5 (5b), consistent with Path 1 -

(Fig. 4a) predicted in static DFT calculations. Interestingly, two N5 rings change their bonding configuration from ƞ2 to ƞ1 before the first dissociation actually occurs, a picture not seen in static DFT calculations that assumed the sequential ring opening one by one. Indeed, this Ti(ƞ1-N5)2(ƞ2-N5)2 configuration turns out to be a true minimum upon a separate further geometry optimization and frequency calculation (as briefly mentioned in section 3.1), whose energy level is 5.9 kcal/mol above the initial Ti(ƞ2-N5)4. The structure 5b then converts to 5d -

via 5c by dissociating one ƞ1-N5 ring, in which 5d upon further geometry optimization also 14


Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


corresponds to ƞ1-I2 in Fig. 4. We also located the true transition state corresponding to 5c, -

whose energy is almost the same as ƞ1-TS1. Dissociation of the remaining N5 rings (from 5e -

to 5i) then occurs with all three N5 rings changed to ƞ1 type (5e); structure 5h corresponds to ƞ1-I4, but the other intermediates species (5e, 5f, and 5g) are different from those found -

(assumed) in our previous static calculations in that many N5 rings change their bonding configurations simultaneously. AIMD snapshots before the third and fourth dissociation steps are shown in SI (Fig. S4). The third and fourth dissociations occur via Fig. S4b and Fig. S4c in the same manner as the previous dissociations via the elongation of N1-N2 and N3-N4 bonds. We found the transition state corresponding to the third dissociation in AIMD by static DFT calculations (Fig. S4a). This transition state is 1.7 kcal/mol above ƞ1-TS3. The TS of the fourth dissociation step in Path 1 (ƞ1-TS4 in Fig. S2a) is well matched with the AIMD snapshot (Fig. S4c), thus, all of the intermediates in the fourth dissociation step are confirmed in AIMD simulations. To summarize the AIMD trajectories, dissociation reaction of Ti(N5)4 indeed occurs via the stepwise bond-rearranged transition states (Path 1), but the mechanism revealed from AIMD is clearly complementary in details to that using static DFT calculations, suggesting the necessity of AIMD simulations for a more complete understanding.

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

Figure 5. AIMD snapshots for dissociation (Ti(N5)4 → Ti(N3)4 + 4N2) at different times. Some elongated bond lengths are shown in Å. Dinitrogen produced from the previous dissociation step is omitted for clarity.

4. Conclusion We have studied the lowest energy structures and stability of M(N5)3 for group 3 metals (M = Sc, Y), and M(N5)4 for group 4 metals (M = Ti, Zr, Hf) as nitrogen rich potential high energy density materials. Among the species considered here, Ti(N5)4 shows the best thermodynamic stability with a large formation and dissociation energy, as well as a reasonable first dissociation barrier for kinetic stability. Ti(N5)4 also has a high metal to nitrogen ratio of 1:20 with 85 % nitrogen contents by weight, the largest values among the reported neutral M(N5)n species in literature, thus proposed as a promising candidate for a high performance HEDM. The ab initio molecular dynamics simulations demonstrate that the 16


Page 17 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dissociation pathways of M(N5)n involves the rearrangement of the bonding configurations before dissociation into a final state M(N3)4 + 4N2, the results not seen in the limited static DFT calculations. Consideration of isomerization reactions of M(N5)n species would thus be clearly important to understand the dissociation pathway accurately in related -future studies.

Author Information Corresponding Authors *Y. Jung, E-mail: [email protected] Tel: +82-42-350-1712 *S. Cho, E-mail: [email protected] Tel: +82-42-821-3704

Acknowledgment This work was supported by the Agency for Defense Development (6111F5-911127201) funded by Defense Acquisition Program Administration of Korea.

Supporting Information Reaction pathways for Y(N5)3, Ti(N5)4 and Zr(N5)4, reaction energy of each step and additional AIMD snapshots.

17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

5. References (1) Zarko, V. E. Searching for ways to create energetic materials based on polynitrogen compounds (review). Combustion, Explosion, and Shock Waves 2010, 46, 121-131. (2) Rahm, M.; Brinck, T. Kinetic stability and propellant performance of green energetic materials.

Chem. Eur. J. 2010, 16, 6590-6600. (3) Vij, A.; Pavlovich, J. G.; Wilson, W. W.; Vij, V.; Christe, K. O. Experimental detection of the pentaazacyclopentadienide (pentazolate) anion, cyclo‐N5− . Angew. Chem. 2002, 114, 3177-3180. (4) Cacace, F.; De Petris, G.; Troiani, A. Experimental detection of tetranitrogen. Science 2002, 295, 480-481. (5) Christe, K. O.; Wilson, W. W.; Sheehy, J. A.; Boatz, J. A. N 5+: a novel homoleptic polynitrogen ion as a high energy density material. Angew. Chem. Int. Ed. 1999, 38, 2004-2009. (6) Gagliardi, L.; Orlandi, G.; Evangelisti, S.; Roos, B. O. A theoretical study of the nitrogen clusters formed from the ions N3-, N5+, and N5. J. Chem. Phys. 2001, 114, 10733-10737. (7) Dixon, D. A.; Feller, D.; Christe, K. O.; Wilson, W. W.; Vij, A.; Vij, V.; Jenkins, H. D. B.; Olson, R. M.; Gordon, M. S. Enthalpies of formation of gas-phase N3, N3-, N5+, and N5-from ab initio molecular orbital theory, stability predictions for N5+ N3-and N5+ N5-, and experimental evidence for the instability of N5+ N3. J. Am. Chem. Soc. 2004, 126, 834-843. (8) Li, Q. S.; Wang, L. J.; Xu, W. G. Structures and stability of N9, N9− and N9+ clusters. Theor. Chem.

Acc. 2000, 104, 67-77. (9) Tobita, M.; Bartlett, R. J. Structure and stability of N 6 isomers and their spectroscopic characteristics. J. Phys. Chem. A 2001, 105, 4107-4113. (10) Liu, Y. D.; Zhao, J. F.; Li, Q. S. Structures and stability of N 7+ and N7− clusters. Theor. Chem.

Acc. 2002, 107, 140-146. (11) Gagliardi, L.; Pyykkö, P. η5-N5--Metal-η7-N73-: A new cass of compounds. J. Phys. Chem. A 2002, 106, 4690-4694. (12) Gagliardi, L.; Pyykkö, P. Scandium Cycloheptanitride, ScN 7: A predicted high-energy molecule containing an [η7-N7]3-ligand. J. Am. Chem. Soc. 2001, 123, 9700-9701. (13) Zhao, J. F.; Li, Q. S. A kinetic stability study of MN 5+ (M = Be, Mg, Ca, Sr, and Ba). Int. J.

Quantum Chem 2004, 98, 485-494. (14) Zhao, J. F.; Li, N.; Li, Q. S. A kinetic stability study of MN 5 (M= Li, Na, K, and Rb). Theor. Chem.

Acc. 2003, 110, 10-18. (15) Zhang, X.; Yang, J.; Lu, M.; Gong, X. Structure, stability and intramolecular interaction of M(N5)2 (M= Mg, Ca, Sr and Ba): a theoretical study. RSC Advances 2015, 5, 21823-21830. (16) Straka, M.; Pyykkö, P. One metal and forty nitrogens. Ab initio predictions for possible new high-energy pentazolides. Inorg. Chem. 2003, 42, 8241-8249. (17) Lein, M.; Frunzke, J.; Timoshkin, A.; Frenking, G. Iron Bispentazole Fe(η 5‐N5)2, a theoretically predicted high‐energy compound: Structure, bonding analysis, metal–ligand bond strength and a 18


Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Ccomparison with theisoelectronic ferrocene. Chem. Eur. J. 2001, 7, 4155-4163. (18) Zhang, X.; Yang, J.; Lu, M.; Gong, X. Theoretical studies on stability and pyrolysis mechanism of salts formed by N5− and metallic cations Na+, Fe2+ and Ni2+. Struct. Chem. 2015, 26, 785-792. (19) Atwood, J.; Smith, K. Nature of the scandium-carbon bond. II. Crystal and molecular structure of tricyclopentadienylscandium. J. Am. Chem. Soc. 1973, 95, 1488-1491. (20) Calderazzo, F.; Englert, U.; Pampaloni, G.; Tripepi, G. The tetracyclopentadienyls of titanium, zirconium and hafnium: new synthetic procedures and reactivity. J. Organomet. Chem. 1998, 555, 49-56. (21) Adam, M.; Behrens, U.; Fischer, R. Structure of tris (η5-cyclopentadienyl) yttrium (III). Acta

Crystallogr. Sect. C: Cryst. Struct. Commun. 1991, 47, 968-971. (22) Shao, Y.; Gan, Z.; Epifanovsky, E.; Gilbert, A. T.; Wormit, M.; Kussmann, J.; Lange, A. W.; Behn, A.; Deng, J.; Feng, X. Advances in molecular quantum chemistry contained in the Q-Chem 4 program package. Mol. Phys. 2015, 113, 184-215. (23) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J.

Chem. Phys. 2010, 132, 154104. (24) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 2011, 32, 1456-1465. (25) Becke, A. D. Density‐functional thermochemistry. III. The role of exact exchange. J. Chem.

Phys. 1993, 98, 5648-5652. (26) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Physical review B 1988, 37, 785. (27) Hariharan, P. C.; Pople, J. A. The influence of polarization functions on molecular orbital hydrogenation energies. Theoretica chimica acta 1973, 28, 213-222. (28) Hay, P. J.; Wadt, W. R. Ab initio effective core potentials for molecular calculations. Potentials for K to Au including the outermost core orbitals. J. Chem. Phys. 1985, 82, 299-310. (29) Rassolov, V. A.; Pople, J. A.; Ratner, M. A.; Windus, T. L. 6-31G* basis set for atoms K through Zn. J. Chem. Phys. 1998, 109, 1223-1229. (30) Glendening, E.; Badenhoop, J.; Reed, A.; Carpenter, J.; Bohmann, J.; Morales, C.; Weinhold, F. NBO, version 5.0. Theoretical Chemistry Institute, University of Wisconsin: Madison, WI 2001. (31) Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. The binary group 4 azides [Ti(N3)4], [P(C6H5)4][Ti(N3)5], and [P(C6H5)4]2[Ti(N3)6] and on linear Ti-N-NN coordination. Angew. Chem. 2004, 116, 3210-3214. (32) Haiges, R.; Boatz, J. A.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Experimental Eevidence for linear metal–azido coordination: The binary group 5 azides [Nb(N3)5],[Ta(N3)5],[Nb(N3)6]− , and [Ta(N3)6]− , and 1:1 acetonitrile adducts [Nb(N3)5(CH3CN)] and [Ta(N3)5(CH3CN)]. Angew. Chem. 2006,

118, 4948-4953. 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

(33) Gagliardi, L.; Pyykkö, P. Predicted Group 4 Tetra-azides M (N3)4 (M= Ti-Hf, Th): The first examples of linear M-NNN coordination. Inorg. Chem. 2003, 42, 3074-3078. (34) Andrews, L.; Zhou, M.; Chertihin, G. V.; Bare, W. D.; Hannachi, Y. Reactions of laser-ablated aluminum atoms with nitrogen atoms and molecules. Infrared spectra and density functional calculations for the AlN2, Al2N, Al2N2, AlN3, and Al3N molecules. J. Phys. Chem. A 2000, 104, 16561661. (35) Zhou, M.; Andrews, L. Reactions of laser-ablated Ga, In, and Tl atoms with nitrogen atoms and molecules. infrared spectra and density functional calculations of GaN, NGaN, NInN, and the M3N and MN3 molecules. J. Phys. Chem. A 2000, 104, 1648-1655. (36) Andrews, L.; Citra, A.; Chertihin, G. V.; Bare, W. D.; Neurock, M. Reactions of laser-ablated Co and Ni atoms with nitrogen atoms and molecules. Infrared spectra and DFT calculations of metal nitride molecular species and complexes. J. Phys. Chem. A 1998, 102, 2561-2571. (37) Andrews, L.; Bare, W. D.; Chertihin, G. V. Reactions of laser-ablated V, Cr, and Mn atoms with nitrogen atoms and molecules. Infrared spectra and density functional calculations on metal nitrides and dinitrogen complexes. J. Phys. Chem. A 1997, 101, 8417-8427. (38) Chertihin, G. V.; Andrews, L.; Neurock, M. Reactions of laser-ablated iron atoms with nitrogen atoms and molecules. Matrix infrared spectra and density functional calculations of novel iron nitride molecules. J. Phys. Chem. 1996, 100, 14609-14617. (39) Kushto, G. P.; Souter, P. F.; Chertihin, G. V.; Andrews, L. An infrared spectroscopic and density functional investigation of dinitrogen activation by group IV metal atoms. Journal of Chemical

Physics 1999, 110, 9020-9031. (40) Chertihin, G. V.; Andrews, L.; Bauschlicher, C. W. Reactions of laser-ablated scandium atoms with nitrogen: Matrix infrared spectra and DFT calculations for scandium nitrides and the fixation of nitrogen by two scandium atoms. J. Am. Chem. Soc. 1998, 120, 3205-3212. (41) Andrews, L.; Wang, X.; Gong, Y.; Kushto, G. P.; Vlaisavljevich, B.; Gagliardi, L. Infrared spectra and electronic structure calculations for NN complexes with U, UN, and NUN in solid argon, neon, and nitrogen. J. Phys. Chem. A 2014, 118, 5289-5303. (42) Andrews, L.; Wang, X.; Gong, Y.; Vlaisavljevich, B.; Gagliardi, L. Infrared Sspectra and electronic structure calculations for the NUN (NN)1–5 and NU (NN)1–6 complexes insolid argon. Inorg. Chem. 2013, 52, 9989-9993. (43) Lu, Z.-H.; Jiang, L.; Xu, Q. A combined experimental and theoretical study of iron Ddinitrogen complexes: Fe(N2), Fe(NN)x (x= 1− 5), and Fe(NN)3− . J. Phys. Chem. A 2010, 114, 2157-2163. (44) Ding, K.-W.; Li, X.-W.; Xu, H.-G.; Li, T.-Q.; Ge, Z.-X.; Wang, Q.; Zheng, W.-J. Experimental observation of TiN12+ cluster and theoretical investigation of its stable and metastable isomers.

Chem. Sci. 2015, 6, 4723-4729. (45) Bazanov, B.; Haas, Y. Solution Photochemistry of [p-(Dimethylamino) phenyl] pentazole (DMAPP) at 193 and 300 nm. J. Phys. Chem. A 2014, 119, 2661-2671. 20


Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(46) Portius, P.; Davis, M.; Campbell, R.; Hartl, F. e.; Zeng, Q.; Meijer, A. J.; Towrie, M. Dinitrogen release from arylpentazole: a picosecond time-resolved infrared, spectroelectrochemical, and DFT computational study. J. Phys. Chem. A 2013, 117, 12759-12769.

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of contents (TOC) image

22


Page 22 of 22

Ti(N5)4 as a Potential Nitrogen-Rich Stable High-Energy Density

Recommend Documents