AlH3 and Ammonia

Dec 23, 2014 - Demchuk , A.; Simpson , S.; Koplitz , B. Exploration of the Laser-Assisted Clustering and Reactivity of Trimethylaluminum with and with...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/JPCA

Initial Gas Phase Reactions between Al(CH3)3/AlH3 and Ammonia: Theoretical Study Anna S. Lisovenko,† Keiji Morokuma,‡ and Alexey Y. Timoshkin*,† †

Inorganic Chemistry Group, Institute of Chemistry, St. Petersburg State University, University Pr. 26, Old Peterhof, St. Petersburg, 198504, Russia ‡ Fukui Institute for Fundamental Chemistry, Kyoto University, 34-4 Takano Nishihiraki-cho, Kyoto 606-8103, Japan S Supporting Information *

ABSTRACT: Mechanisms of initial stages of gas phase reactions between trimethylaluminum and ammonia have been explored by DFT studies. Subsequent substitution of CH3 groups in AlMe3 by amido groups and substitution of hydrogen atoms in ammonia by AlMe2 groups have been considered. Structures of Al(CH3) x(NH 2) 3−x, NH x(Al(CH3)2)3−x (x = 0−3) and related donor−acceptor complexes, dimerization products, and reaction pathways for the methane elimination have been obtained. The transition state for the first methane elimination from Al(CH3)3NH3 adduct is the highest point on the reaction pathway; subsequent processes are exothermic and do not require additional activation energy. In excess ammonia, subsequent methane elimination reactions may lead to formation of [Al(NH2)3]2, while in excess trimethylaluminum, formation of N(AlMe2)3 is feasible. Formation of [AlMe2NH2]2 dimer is very favorable thermodynamically. Studies on model reactions between AlH3 and NH3 indicate that reaction barriers obtained for hydrogen-substituted species may serve as an upper estimate in studying the reactivity of methylsubstituted analogues in more complex systems.

■

of methane elimination is 107.5 kJ mol−1.13 Structural studies revealed trimeric [Me2AlNH2]3 in the solid state with a skewboat conformation of the Al3N3 ring.10 The monomeric form Me2AlNH2 has been proposed as an intermediate in the decomposition of AlMe3NH3 in solution11 and has also been suggested to exist in the vapor phase at MOCVD conditions.14 Monomeric Me2AlNH2 was experimentally characterized by Müller in 1996 as a product of UV irradiation of matrix-isolated ammonia trimethylalane.15 Since the Me2AlNH2 molecule possesses coordinatively unsaturated donor and acceptor centers, it can be involved in reactions with Lewis acids and bases (including self-interaction to form dimers and higher oligomers). In the present report we limited ourselves to theoretical study on mechanisms of AlMe2NH2 formation and its subsequent gas phase reactions, including methane elimination, dimerization, and interactions with either excess AlMe3 or excess NH3, eventually leading to Al(NH2)3 and N(AlMe2)3 (Scheme 1). In order to avoid nitrogen loss, CVD processes of nitride deposition are usually carried out in excess ammonia (typical NH3/group 13 precursor ratio > 1000:1).16 It should be noted that, although the parent Al(NH2)3 is not known, its organic derivatives Al(NR2)3 are well-known laboratory species.17−19 The substituent R plays an important

INTRODUCTION The chemical vapor deposition (CVD) method1,2 is one of the leading ways for production of group 13−15 binary materials, such as aluminum and gallium nitrides. Gas phase chemistry involved in the metalorganic chemical vapor deposition (MOCVD) of AlN is not yet fully understood. Experimental mass-spectrometry observations reveal formation of cluster compounds in the gas phase upon laser irradiation of AlMe3− NH3 mixtures at low temperatures.3 In order to understand the gas phase chemistry, theoretical studies of reaction mechanisms are very helpful. The first stage of the decomposition process, methane elimination from the AlMe3NH3 adduct, was the subject of earlier theoretical works.4−7 Subsequent reactivity is much less explored. Reaction between AlMe3 and NH3 produces the donor− acceptor complex AlMe3NH3, which was structurally characterized in the solid state by X-ray powder diffraction (XRD) data.8 Its pyrolysis in the condensed phase yields the amidoand imidoalanes and, finally, aluminum nitride, as was pointed out by Wiberg.9 Detailed studies on AlMe3NH3 formation and thermal decomposition in solution were performed by Sauls and Interrante.10,11 Dissociation enthalpy of the donor− acceptor bond breaking in the AlMe3NH3 adduct was measured by solution calorimetry in hexane (115 ± 1 kJ mol−1)12 and in benzene (95 ± 5 kJ mol−1).11 According to Fourier transform infrared (FTIR) spectroscopic studies on the kinetics of the gas phase reaction between AlMe3 and NH3, the activation energy © 2014 American Chemical Society

Received: July 31, 2014 Revised: November 11, 2014 Published: December 23, 2014 744

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A Scheme 1. Considered Reaction Pathways and Numbering of Compounds

■

RESULTS AND DISCUSSION Structural Data. Optimized geometries of reactants, products, and transition states (TSs) are presented in Figure 1. Comparison of calculated and experimental bond distances and bond angles for Me3AlNH3 is given in Table 1. The computed Al−N bond distance for the gas phase complex is longer than the experimental value in the solid state. The AlC3 skeleton of 2 deviates only slightly from planarity (C−Al−C = 117.3°). The Al−N distance of 2.106 Å is relatively long compared to the common range of simple amine adducts, which are known to have AlN bond lengths between 1.94 and 2.10 Å.28 Our findings are in accord with earlier MP2(fc)/631G* results.15 Computed Al−N bond lengths in coordinatively unsaturated amides 3, 5, 11, and 14 are 1.78−1.80 Å, significantly shorter than the 1.958 Å observed in the saturated [Me2AlNH2]3 trimer.10 Complex formation with NH3 lengthens Al−NH2 distances in 10, 8, and 13 to 1.817−1.836 Å. Note that addition of AlMe3 does not lead to simple donor−acceptor complexes; cyclic forms 4, 6, and 12 with CH3 bridging groups are formed instead. This clearly shows the tendency of Al-rich compounds to adopt methyl-bridged structures. All the transition states TS1−TS9 for the unimolecular methane elimination are qualitatively very similar; the Al−N bond lengths in the transition states are elongated to 1.92−1.94 Å. Thermodynamic Data. Thermodynamic data for the considered reactions are summarized in Table 2. The standard enthalpy for the gaseous AlMe3·NH3 complex formation from

role in the association degree of the compound in the solid state. For R = CH3, dimeric Al2(NMe2)6 structure with two NMe2 bridges is observed,17 while bulkier iPr18 and SiMe319 substituents favor monomeric Al(NR2)3 structures in the solid state and in benzene solution.20 Considered reaction processes and labeling of the compounds and transition states are shown in Scheme 1. Since for the study of large oligomer species the hydrogen-substituted analogues are often used as model systems,21 we have also considered model reactions, in which a methyl group on aluminum is replaced by a hydrogen atom. These model compounds and corresponding transition states for hydrogen elimination will be denoted by addition of the label m, for example, 1m. Thus, we will compare reaction profiles and energetics of reactions of AlMe3 and AlH3 with ammonia.

■

COMPUTATIONAL DETAILS Density functional theory (DFT) in the framework of the hybrid three-parameter exchange functional of Becke22 with the gradient corrected correlation functional of Lee, Yang, and Parr23 (B3LYP) with all electron def2-TZVPP basic set24 was used throughout. All structures have been fully optimized and verified to be minima or transition states (TSs) by subsequent vibrational analysis. Intrinsic reaction coordinate scans confirmed that obtained TSs are connecting reactant and products. Single point energy computations have been performed at M06-2X25 and B3LYP-D326 levels of theory. All computations have been carried out using the standard Gaussian 03 program package.27 745

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A

Figure 1. Optimized geometries of considered molecules and transition states. Distances in Å; angles in degrees. B3LYP/def2-TZVPP level of theory.

AlMe3 and NH3 is −79.8 kJ mol−1; this value is smaller than experimental solution calorimetry values of −95 ± 5 kJ mol−1 in benzene11 and −115 ± 1 kJ mol−1 in hexane.12 Reaction of NH3 with unsaturated 3, 5, and 11 leads to formation of donor−acceptor (DA) complexes 10, 8, and 13; standard enthalpies for these reactions are in the range −59 to

−68 kJ mol−1. Further elimination of methane from DA complexes 10, 8, and 13 is exothermic by −26 to −30 kJ mol−1. In contrast, reaction of 3, 5, and 11 with AlMe3 leads to cyclic structures 4, 6, and 12 with bridging μ-NH2 and μ-CH3 groups. Because of this additional stabilization, further elimination of methane becomes less exothermic or slightly endothermic; the 746

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A

Table 3. Activation Energies, kJ mol−1, at the B3LYP/def2TZVPP Level of Theory

Table 1. Computed Gas Phase and Experimental Solid State Bond Lengths (Å) and Angles (deg) of Me3AlNH3 solid state

transition state

R = CH3

R=H

param

B3LYP/TZVPP

gas phase MP2(fc)/6-31G*15

XRD8

Al−N Al−C N−Al−C C−Al−C

2.103 1.988 100.0 117.0

2.106 1.993 99.6 117.3

2.004 1.945−1.976 101.3−103.0 115.2−115.7

TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9 TS10 TS11 TS12 TS13

125.3 167.5 160.4 129.0 132.7 152.2 135.8 304.8 297.7 125.8 83.1 118.1 60.6

126.6 203.5 205.7 123.7 123.6 183.2 119.9

standard enthalpies for methane elimination reactions from compounds 4, 6, and 12 are in the range −4 to +9 kJ mol−1. A similar situation is observed for model systems as well. Exothermic complexation with AlH3 results in cyclic structures 4m, 6m, and 12m with bridging μ-NH2 and μ-H groups. However, unlike the methane elimination, all reactions of hydrogen elimination are endothermic. Activation Energies. Activation energies for the subsequent steps of methane/hydrogen elimination are summarized in Table 3. Note that activation energies for the methane elimination in excess ammonia (about 130 kJmol−1) at TS4, TS5, and TS7 are significantly lower than the activation energies in excess trimethylaluminum (about 160 kJ mol−1) at TS2, TS3, and TS6; the difference between these values is on average 30 kJ mol−1. In model systems the difference between activation energies for the hydrogen elimination in excess ammonia (about 120 kJ mol−1) and that in AlH3 excess (about 200 kJ mol−1) increases to 80 kJ mol−1. This increase mostly comes from the higher stabilization of the H-bridged structure 6m compared to CH3-bridged 6 due to small steric hindrance of the H atom. Computed activation energies for the methane elimination from 1 (125.3, 125.4, and 125.8 kJ mol−1 at B3LYP, B3LYP-D3, and M06-2X levels of theory, respectively) are by about 18 kJ mol−1 overestimated compared to the experimental value of

107.5 kJ mol−1 derived from FTIR measurements.13 Note that the activation energies for methane and hydrogen elimination are very close: 125.3 and 126.6 kJ mol−1, respectively. Relative energies of considered structures computed by different DFT methods are given in Table 4. Obtained results indicate that differences in reaction energies, computed at B3LYP and B3LYP-D3 levels of theory on B3LYP/def2TZVPP optimized geometries, are small. Values obtained at the M06-2X level are slightly more exothermic. This suggests that B3LYP/def2-TZVPP optimized geometries are reliable. Reaction profiles for the subsequent hydrogen and methane elimination are given in Figures 2 and 3, respectively. The initial methane elimination step is colored in black. Reactions of 3 and 3m with ammonia are plotted to the left side and colored in blue; reactions with AlR3 are plotted to the right side and colored in red. Points near the transition state correspond to energies of structures on the intrinsic reaction coordinate. Both

Table 2. Thermodynamic Data (kJ mol−1 and kJ mol−1deg−1) for the Considered Gas Phase Processes, at B3LYP/def2-TZVPP Level of Theory R = CH3

R=H

process

ΔH°298

ΔS°298

ΔG°298

ΔH°298

ΔS°298

ΔG°298

AlR3 + NH3 = 2 2 = 3 + RH 3 + AlR3 = 4 4 = 5 + RH 5 + AlR3 = 6 6 = 7 + RH 5 + NH3 = 8 8 = 9 + RH 3 + NH3 = 10 10 = 11 + RH 11 + AlR3 = 12 12 = 9 + RH 11 + NH3 = 13 13 = 14 + RH 3 = 15 + CH4 10 = 16 + CH4 2 + 2 = 17 + CH4 17 = 18 + CH4 2 + 3 = 12 + CH4 12 = 18 3 + 3 = 18

−79.8 −27.7 −108.1 8.9 −86.9 4.1 −68.1 −30.6 −64.9 −29.1 −100.2 −3.7 −59.1 −26.2 233.1 196.1 −120.2 −130.0 −114.4 −107.6 −194.9

−166.3 127.3 −224.8 185.3 −260.3 204.7 −148.5 123.0 −137.2 125.8 −218.9 165.2 −132.5 113.7 139.6 117.3 −57.8 95.8 −63.9 −24.3 −216.6

−30.2 −65.6 −41.1 −46.4 −9.4 −56.9 −23.8 −67.2 −24.0 −66.6 −34.9 −53.0 −19.7 −60.1 191.5 161.2 −103.0 −158.6 −95.3 −100.4 −130.3

−103.8 15.2 −164.3 82.7 −167.9 97.3 −84.1 6.8 −76.6 3.9 −152.5 66.3 −63.0 1.6

−120.2 102.1 −150.9 127.7 −169.0 162.2 −122.4 104.1 −121.5 98.5 −148.0 129.4 −113.0 99.2

−68.0 −15.2 −119.4 44.7 −117.5 49.0 −47.6 −24.2 −40.4 −25.5 −108.4 27.8 −29.3 −27.9

747

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A Table 4. Relative Energies with Respect to Free AlR3 and NH3, kJ mol−1 B3LYP-D3/def2-TZVPP //B3LYP/def2TZVPP

B3LYP/def2-TZVPP

a

compd

R = CH3

2 TS1 3 4 TS2 5 6 TS3 7 8 TS4 9 10 TS5 11 12 TS6 13 TS7 14

−88.5 (−84.6) 36.8 (29.7)a −109.2 (−110.1)a −225.2 −57.8 −211.8 −305.0 −144.6 −298.0 −287.8 −158.8 −311.1 −181.4 −48.7 −203.5 −311.1 −158.9 −269.4 −133.5 −288.6 a

M06-2X/def2-TZVPP// B3LYP/def2TZVPP

R=H

R = CH3

R=H

R = CH3

R=H

−114.6 12.1 −79.4 −255.6 −52.1 −153.2 −332.2 −126.5 −216.6 −246.6 −122.9 −220.0 −164.8 −41.2 −141.3 −305.2 −121.9 −211.6 −91.7 −191.2

−99.8 25.2 −109.0 −250.8 −77.3 −217.1 −348.2 −181.8 −316.1 −304.8 −176.0 −315.9 −191.9 −59.4 −202.9 −335.4 −177.8 −279.1 −143.6 −287.4

−119.1 8.0 −80.9 −263.6 −59.2 −157.0 −345.1 −138.1 −224.9 −257.5 −132.7 −226.1 −172.3 −48.0 −144.8 −316.9 −131.5 −223.1 −102.2 −197.5

−111.0 14.8 −110.9 −278.3 −90.0 −222.4 −388.6 −203.1 −329.1 −320.7 −191.5 −323.7 −205.8 −72.1 −207.5 −365.2 −192.6 −295.9 −159.5 −295.2

−131.1 −3.3 −95.9 −302.8 −88.5 −190.6 −412.8 −195.3 −283.1 −302.7 −176.6 −275.3 −200.0 −73.9 −175.9 −370.3 −176.0 −267.1 −143.4 −245.1

B3LYP/LanL2DZ* level of theory, ref 7.

Figure 2. Reaction profiles for hydrogen elimination. See text for details.

reaction profiles are qualitatively similar. Methane elimination reactions are exothermic, while reactions of hydrogen elimination are endothermic. Aluminum-rich compounds prefer H or CH3 bridging structures over simple adducts, which energetically strongly stabilize compounds 6, 6m, and 12, 12m and increase the activation energies for hydrogen/methane elimination. Reactions of subsequent hydrogen elimination are slightly endothermic, while reactions for subsequent methane elimination are slightly exothermic. Overall, energy barriers for hydrogen elimination are close to or slightly larger than those of methane elimination (Figure S5, Supporting Information). Thus, data obtained for the hydrogen-substituted model compounds may serve as an upper-limit estimate for the reactivity of methyl-substituted analogues. It is expected that, if

the hydrogen elimination is feasible, then methane elimination should be also favorable both thermodynamically and kinetically. Dimerization of monomeric Al(NH2)3 is exothermic by 180.6 kJ mol−1, but unfavorable by entropy by 136.4 J mol−1 K−1; formation of this compound is favorable at low temperatures. Monomeric AlMe2NH2 (3) formed after the first methane elimination from Al(CH3)3NH3 adduct (limiting reaction step), appears to be the key intermediate. Therefore, different reaction pathways starting from 3 have been considered, and results are shown in Figure 4. Methane elimination from 3 with formation of MeAlNH (15) is highly endothermic, and the activation energy (TS8) exceeds 300 kJ mol−1. Thus, this 748

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A

Figure 3. Reaction profiles for methane elimination. See text for details.

were also unsuccessful; upon optimization such a chain structure converges to cyclic dimer 18. Formation of 18 is the most energetically favorable process. However, since the concentration of 3 in the gas phase is very small (it is a product of the limiting step of the methane elimination from AlMe3NH3), the probability of self-interaction is much smaller than the probability of exothermic barrierless reactions with much more abundant ammonia or AlMe3. If 3 is formed, it supposed to exothermically react with AlMe3 or NH3 (both with zero barrier) with formation of 4 and 10 (Figure 4), which will follow the reaction pathways described above (Figure 3). Note that for 10 there are two possible methane elimination processes, leading to 11 via TS5 and to 16 via TS9 (Scheme 1). The second process is endothermic and has a much larger activation energy (Figure S6, Supporting Information). We have also considered alternative pathway for the generation of dimer 18, starting from interaction of 3 with 2, or by direct interaction between the two adduct molecules 2 + 2. These reaction profiles are presented in Figure 5, and relative energies of related species are given in Table 5. Methane eliminations via TS10 and TS12 are limiting steps for these pathways. Note that energetically TS10 and TS12 lie below the isolated AlMe3 and NH3, analogously with TS2 and TS5 for the reaction pathways considered earlier (Figure 3). We conclude that, although formation of the dimer 18 is the most energetically favorable process, different reaction pathways are competitive kinetically. Excess of one of the reagents (AlMe3 or

Figure 4. Reaction profiles for reactivity of 3. See text for details.

pathway is not feasible both thermodynamically and kinetically. Since 3 carries unsaturated Al and N centers, it is capable of donor−acceptor interaction with Lewis acids and bases, including self-interaction (dimerization). We have found that reactions of 3 with NH3, AlMe3, and self-dimerization are highly exothermic and apparently proceed without barriers (we were not able to locate transition states for these reactions). Our attempts to optimize the chain Me2AlNH2AlMe2NH2 dimer, formed by head-to-tail donor−acceptor interactions,

Figure 5. Alternative reaction pathways for dimer [Me2AlNH2]2 (18) formation. Relative energies with respect to 1 and NH3. 749

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A Table 5. Relative Energies Erel, kJ mol−1, with Respect to Free AlMe3 and NH3, at the B3LYP/def2-TZVPP Level of Theory compd

Erel

compd

Erel

compd

Erel

TS8 15 TS9 16

195.6 129.7 116.3 19.2

2+2 TS10 17 TS11

−177.0 −51.2 −298.1 −215.1

18 3+2 TS12 TS13

−422.0 −197.7 −79.6 −251.1

of SPbU” and Fukui Institute for Fundamental Chemistry, Kyoto University.

■

(1) Pierson, H. O. Handbook of Chemical Vapor Deposition: Principles, Technology and Applications, 2nd ed.; William Andrew Publishing LLC: New York, 2000. (2) Stringfellow, G. B. Organometallic Vapor Phase Epitaxy, 2nd ed.; Academic Press: New York, 1998. (3) Demchuk, A.; Simpson, S.; Koplitz, B. Exploration of the LaserAssisted Clustering and Reactivity of Trimethylaluminum with and without NH3. J. Phys. Chem. A 2003, 107, 1727−1733. (4) Tachibana, A.; Makino, O.; Tanimura, S.; Tokunaga, H.; Akutsu, N.; Matsumoto, K. Quantum Chemical Studies of Gas Phase Reactions between TMA, TMG, TMI and NH3. Phys. Status Solidi A 1999, 176, 699−703. (5) Makino, O.; Nakamura, K.; Tachibana, A.; Tokunaga, H.; Akutsu, N.; Matsumoto, K. Quantum Chemical Mechanism in Parasitic Reaction of AlGaN Alloys Formation. Appl. Surf. Sci. 2000, 159−160, 374−379. (6) Nakamura, K.; Makino, O.; Tachibana, A.; Matsumoto, K. Quantum Chemical Study of Parasitic Reaction in III−V Nitride Semiconductor Crystal Growth. J. Organomet. Chem. 2000, 611, 514− 524. (7) Ikenaga, M.; Nakamura, K.; Tachibana, A.; Matsumoto, K. Quantum Chemical Study of Gas-Phase Reactions of Trimethylaluminium and Triethylaluminium with Ammonia in III−V Nitride Semiconductor Crystal Growth. J. Cryst. Growth 2002, 237−239, 936− 941. (8) Muller, J.; Ruschewitz, U.; Indris, O.; Hartwig, H.; Stahl, W. Structure of Ammonia Trimethylalane (Me3Al-NH3): Microwave Spectroscopy, X-ray Powder Diffraction, and ab Initio Calculations. J. Am. Chem. Soc. 1999, 121, 4647−4652. (9) Bahr, G. In FIAT Review of WWII German Science, 1939−1946; Klemm, W., Ed.; Dieterichsche Verlagsbuchhandlung: Wiesbaden, FRG, 1948; Vol. 24, Inorganic Chemistry, Part II, p 155. (10) Sauls, F. C.; Interrante, L. V. Coordination Compounds of Aluminum as Precursors to Aluminum Nitride. Coord. Chem. Rev. 1993, 128, 193−207. (11) Sauls, F. C.; Interrante, L. V.; Jiang, Z. Me3Al·NH3 Formation and Pyrolytic Methane Loss: Thermodynamics, Kinetics, and Mechanism. Inorg. Chem. 1990, 29, 2989−2996. (12) Henrickson, C. H.; Duffy, D.; Eyman, D. P. Lewis Acidity of Alanes. Interactions of Trimethylalane with Amines, Ethers, and Phosphines. Inorg. Chem. 1968, 7, 1047−1051. (13) Creighton, J. R.; Wang, G. T. Kinetics of Metal OrganicAmmonia Adduct Decomposition: Implications for Group-III Nitride MOCVD. J. Phys. Chem. A 2005, 109, 10554−10562. (14) Amato, C. C.; Hudson, J. B.; Interrante, L. V. Identification of the Gas-Phase Products Which Occur During the Deposition of Aluminum Nitride Using the Organometallic Precursor: Tris(Dimethylaluminum Amide) ([(CH3)2AlNH2]3. Appl. Surf. Sci. 1992, 54, 18−24. (15) Muller, J. Aminodimethylalane (Me2AlNH2): Matrix Isolation and Ab Initio Calculations. J. Am. Chem. Soc. 1996, 118, 6370−6376. (16) Jones, A. C.; O’Brien, P. CVD of Compound Semiconductors: Precursor Synthesis, Developments and Applications; VCH: Weinheim, 1997. (17) Waggoner, K. M.; Olmstead, M. M.; Power, P. P. Structural and Spectroscopic Characterization of the Compounds [Al(NMe2)3]2, [Ga(NMe2)2]2, [(Me2N)2Al{μ-N(H)1-Ad}]2 (1-Ad = 1-Adamantanyl) and [{Me(μ-NPh2)Al}2NPh(μ-C6H4)]. Polyhedron 1990, 9, 257− 263. (18) Brothers, P. J.; Wehmschulte, R. J.; Olmstead, M. M.; RuhlandSenge, K.; Parkin, S. R.; Power, P. P. Synthesis, Structure, and Spectroscopic Characterization of Unassociated Mono-, Di-, and Triamido Derivatives of Aluminum and Gallium. Organometallics 1994, 13, 2792−2799.

NH3) will favor the reaction pathway shown in Figure 3 with formation of 7 or 14.

■

CONCLUSIONS The initial reactivity between AlR3 and NH3 and the reactions of R2AlNH2 with excess AlR3 and ammonia have been studied (R = H, CH3). Reaction pathways leading to [Me2AlNH2]2 dimer have been also considered. Structures of reactants, products, and transition states for the unimolecular hydrogen/ methane eliminations have been optimized, and thermodynamic characteristics of corresponding reactions have been obtained. It is shown that the first methane elimination from the AlR3NH3 complex is the highest point on the reaction pathway, and the only one lying above the isolated AlR3 and NH3. Subsequent sequential substitution of methyl groups in AlMe3 by amido groups is exothermic, and despite the larger activation energies of the intermediate steps, the overall reaction profile is rather downhill. Thus, we may speculate that the first methane elimination is a rate-limiting step of the reaction. Reaction pathways leading to [Me2AlNH2]2 dimer are competitive with reaction pathways of subsequent substitution reactions; several processes may operate simultaneously. Higher vapor pressures are expected to favor the dimerization pathway, while low vapor pressures and excess of one of the reagents will favor substitution reactions. Comparison of hydrogen- and methyl-substituted analogues reveals that these reaction pathways are similar. It is suggested that data obtained for the hydrogen-substituted model compounds may serve as an upper-limit estimate for the reactivity of methyl-substituted analogues. This may be utilized for the estimation of the reactivity of more complex gas phase reactions leading to oligomer species.

■

ASSOCIATED CONTENT

S Supporting Information *

Tables with total energies, enthalpies, and entropies, and xyz coordinates for the optimized structures of considered compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: +7(812)4284071. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by SPbSU Grants 12.38.255.2014 and 12.50.1563.2013. Research was carried out using computational resources provided by the Resource Center “Computer Center 750

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751

Article

The Journal of Physical Chemistry A (19) Sheldrick, G. M.; Sheldrick, W. S. Crystal and Molecular Structure of Tris[bis(trimethylsilyl)amino]aluminium, Al[N(SiMe3)2]3. J. Chem. Soc. A 1969, 2279−2282. (20) Ruff, J. K. The Preparation and Reactions of Dialkylamino Derivatives of Aluminum. J. Am. Chem. Soc. 1961, 83, 2835−2839. (21) Kormos, B. L.; Jegier, J. A.; Ewbank, P. C.; Pernisz, U.; Young, V. G.; Cramer, C. J.; Gladfelter, W. L. Oligomeric Rods of Alkyl- and Hydridogallium Imides. J. Am. Chem. Soc. 2005, 127, 1493−1503. (22) Becke, A. D. A New Mixing of Hartree-Fock and Local DensityFunctional Theories. J. Chem. Phys. 1993, 98, 1372−1377. (23) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 1988, 37, 785−789. (24) Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence, Triple Zeta Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297−3305. (25) Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group Thermochemistry, Thermochemical Kinetics, NoncovalentInteractions, Excited States, and Transition Elements: Two New Functionals and Systematic Testing of Four M06-Class Functionals and 12 Other Functional. Theor. Chem. Acc. 2008, 120, 215−241. (26) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the Damping Function in Dispersion Corrected Density Functional Theory. J. Comput. Chem. 2011, 32, 1456−1465. (27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C, et al. Gaussian 03, revision B.05; Gaussian, Inc.: Wallingford, CT, 2004. (28) Haaland, A. In Coordination Chemistry of Aluminum; Robinson, G. H., Ed.; VCH: Weinheim, 1993.

751

DOI: 10.1021/jp507713b J. Phys. Chem. A 2015, 119, 744−751