Multiple Mechanisms for the Thermal Decomposition of

Subscriber access provided by UNIV OF DURHAM

Article

Multiple Mechanisms for the Thermal Decompositions of Metallaisoxazolin-5-ones from Computational Investigations Chen-Chen Zhou, M. Frederick Hawthorne, Kendall N. Houk, and Gonzalo Jiménez-Osés J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.7b01169 • Publication Date (Web): 13 Jul 2017 Downloaded from http://pubs.acs.org on July 14, 2017

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 Organic Chemistry 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 Organic Chemistry

Multiple Mechanisms for the Thermal Decompositions of Metallaisoxazolin-5-ones from Computational Investigations Chen-Chen Zhou,a M. Frederick Hawthorneb, K. N. Houk a* Gonzalo Jiménez-Osésc*

a

Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569, USA

b

International Institute of Nano and Molecular Medicine, University of Missouri, 1514 Research Park Dr., Columbia, MO 65211-3450, USA

c

Departamento de Química, Centro de Investigación en Síntesis Química, Universidad de La Rioja, 26006 Logroño, La Rioja, Spain

Table of Contents

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

ABSTRACT The thermal decompositions of metallaisoxazolin-5-ones containing Ir, Rh or Co are investigated using density functional theory. The experimentally observed decarboxylations of these molecules are found to proceed through retro-(3+2)-cycloaddition reactions, generating the experimentally reported η2 side-bonded nitrile complexes. These intermediates can isomerize in situ to yield a η1 nitrile complex. A competitive alternative pathway is also found where the decarboxylation happens concertedly with an aryl migration process, producing a η1 isonitrile complex. Despite their comparable stability, these η1 bonded species were not detected experimentally The experimentally detected η2 side bound species are likely involved in the subsequent C–H activation reactions with hydrocarbon solvents reported for some of these metallaisoxazolin-5-ones.

INTRODUCTION Some decades ago, Hawthorne and co-workers synthesized 5-membered metallaisoxazolin5-one metallacycles (1) via 1,3-dipolar cycloadditions of aryl nitrile oxides to the M –C bond of low-valent iridium or rhodium carbonyl complexes (Scheme 1).1

Scheme 1. Synthesis of 5-membered metallaisoxazolin-5-one metallacycles by 1,3-dipolar cycloaddition


Page 2 of 22

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


Metallaisoxazolin-5-ones decompose under mild thermal (Scheme 2 and Table 1, entry 1) or photochemical conditions. When the reaction is performed in hydrocarbon solvents (i.e. benzene or toluene, see Table 1, entries 2-3), products resulting from solvent C–H functionalization reactions are observed.2 An intermediate of this transformation, 2, was isolated and characterized by X-ray crystallography. The structure is an uncommon η2 sidebonded nitrile complex, with the carbon and nitrogen atoms both bonded to the metal center. Such a structure was first reported in the 1980s,3 with molybdenum as the metal center. Similar tungsten complexes4,5 are also well known, but complexes with a more electron-rich metal center like iridium or rhodium are less common. To account for the observed C–H functionalization products (see for instance IIa in Scheme 2), it was proposed that the η2 nitrile complex releases the nitrile, generating an active 16-electron species (Ia), which subsequently inserts into a C–H bond of the hydrocarbon solvent molecule.2 The dimer of the proposed 16-electron species (IIIa) was observed in the products, supporting the above proposition. The mechanism of this reaction has been revealed to be rather complicated, and other intermediates might be involved.6 (Scheme 2)

Scheme 2. The thermal decomposition of a 5-membered metallaisoxazolin-5-one (1b), subsequent solvent C–H functionalization (product IIa) and dimerization (product IIIa),



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 4 of 22

and the observed (2a) and proposed (Ia) intermediates.

The generation of the η2 nitrile complex involves extrusion of carbon dioxide. Similar processes has been reported both in organometallic7,8 and organic systems,9 where CO2 is cleaved from the complex while the five-membered ring skeleton is contracted to a threemembered ring. The decomposition reaction of γ-butyrolactone happens at only high temperature, and computational studies have shown that the barriers of these reactions are quite high.9 However, metallaisoxazolin-5-ones decompose at relatively low temperatures. Experimental results show that the metal center largely influences the reactivity of metallacycles. Thus, the iridium complex bearing a triphenylphosphine ligand 1b slowly decomposes at 110 ℃ over 24 h (Table 1, entry 4), while the rhodium analogue 1e is less stable (Table 1, entry 5). On the other hand, first-row cobalt analogue 1f decomposes almost immediately in solution at room temperature (Table 1, entry 6).1,6 The ligands attached to the metal also influence the stability of the complex: Ir complex with PPh3 (Table 1, entry 4) is significantly more stable than those with CO (Table 1, entries 1-3). These observations agree with recent theoretical studies on the synthesis of these metallacycles.10

Table 1. Thermal decomposition of 5-membered metallaisoxazolin-5-ones in the absence and presence of aromatic hydrocarbon solvents.2 Entry

Reactant

Reaction conditions


Product(s)

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


1

THF, 50 ℃, 14 h

2

toluene, 50 ℃, 35 d

3

benzene, hν, 50 ℃, 36 h

4

toluene, 110 ℃, 24 h,

5

toluene, 25 ℃, 6 h

6

toluene, 25 ℃, 5 min

Cp = cyclopentadienyl ligand; Cp* = pentamethylcyclopentadienyl ligand; Tol = tolyl (i.e. p-methylphenyl) ligand.

We have now investigated the mechanism of the thermal decarboxylation of the metallaisoxazolin-5-ones using density functional theory (DFT) methods and report the factors influencing the reaction rates and products.

Computational details. All calculations have been carried out using Gaussian 0911. Hybrid M0612 and B3LYP13,14 corrected with Grimme’s D3 dispersion with Becke-Johnson



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

damping15 (B3LYP-D3) were employed for geometry optimizations. Both methods gave similar results in terms of geometries and energies (see Supporting Information), although B3LYP-D3 consistently produced a larger number of undesired small negative frequencies, affecting the appropriate estimation of vibrational contribution to entropy and thus to free energy. Hence, M06 was selected as the most appropriate method for geometry optimization. The 6-31G(d) basis set was used in geometry optimizations for all the nonmetal atoms, and subsequent single point energy calculations were performed with the 6311+G(2d,p) basis set. The Stuttgart/Dresden effective core potential (SDD) was used for Ir, Rh and Co in both geometry optimization and single point energy calculations. An ultrafine integral grid was used for single point energy calculations. The possibility of different conformations was taken into account for all structures. Frequency analyses were carried out at the same level used in the geometry optimizations to obtain thermal and entropic corrections at 298 K. The nature of the stationary points (minima or TS) was determined in each case according to the appropriate number of negative eigenvalues of the Hessian matrix. The quasiharmonic approximation reported by Truhlar et al. was used to replace the harmonic oscillator approximation for the calculation of the vibrational contribution to enthalpy and entropy.16 Frequencies were not scaled. Mass-weighted intrinsic reaction coordinate (IRC) calculations were performed from the transition state geometries by using the Gonzalez and Schlegel scheme17,18 to check the connectivity between the different stationary points in the potential energy surface; the Hessian were recalculated every 10 steps to ensure the accuracy of the reaction path. The calculated Gibbs free energies are electronic energies derived from single point energy calculations with M06/6-311+G(2d,p)+SDD(Co,Rh,Ir) modified with thermal and entropic corrections from the M06/6-31G(d)+SDD(Co,Rh,Ir) geometry optimization calculations. The SMD


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


solvation model19 with solvation parameters for benzene was used throughout geometry optimization and in the calculation of the single point energies. Atomic charges analyses were performed using the NBO 5.0 program20 at the same level used for single-point energy calculations. Gibbs free energies (∆G) were used for the discussion on the relative stabilities of the considered structures. Cartesian coordinates, electronic energies, entropies, enthalpies, Gibbs free energies, and lowest frequencies of the calculated structures are available as supporting information.

RESULTS AND DISCUSSION Six model complexes containing different metals (iridium, rhodium or cobalt) and cyclopentadienyl (Cp), PMe3 or CO ligands (1a’–1f’, Figure 1) were chosen for the systematic calculation of the factors influencing the reactivities of these systems. Additionally, two real systems that were studied experimentally (1a and 1b)6 were included in the study for verification purposes.

Figure 1. Metallaisoxazolin-5-ones chosen for computation.

Decarboxylation mechanisms. Each of the starting metallaisoxazolin-5-ones was calculated to undergo decarboxylation, in most cases, via two different transition states



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

(TS), shown in Scheme 3. A concerted transition state was found in the singlet state for the decarboxylation process (TS1, pathway a), in which the formation of the M–N bond and the cleavage of M–C and N–O bonds happen simultaneously to give a η2 side-bonded nitrile complex (2, Scheme 3). A second pathway (TS2, pathway b) has been found for the decarboxylation process, leading to a different product. This is a concerted process where the aryl group migrates from carbon to nitrogen while the CO2 moiety is cleaved. This pathway leads to a linear η1 isonitrile complex, 3, which has not been reported experimentally. Both calculated pathways differ from those reported previously for the Pd/Cu/Ag-catalyzed decarboxylation reaction of aryl carboxylic acids,21,22 or the Cu/Pdcatalyzed decarboxylative cross-couplings of benzoates with aryl halides.23 In such mechanisms, the carbonyl carbon is not directly attached to the metal center. Instead, the carboxyl moiety is bound to the metal through a M–O bond, and the metal inserts into the Cipso–CO2 bond through a concerted four-membered transition state with concomitant release of CO2. Activation energies for the calculated pathways are listed in Table 2, and TS1a’-f’ structures are given in Figure 2.

Scheme

3.

The

calculated

mechanisms

for

decarboxylation


of

5-membered

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


metallaisoxazolin-5-ones: TS1 leads to formation of the η2 complex 2 and TS2 leads to a rearranged isonitrile complex 3.

Both pathways are highly exothermic and, as expected, their relative activation energies depend on substitution. The competing transition states (TS1 and TS2) have very similar activation energies for iridium complexes (1a’,b’ and 1b) and in some cases TS2 could not be calculated (as for carbonyl-ligated 1c’ and 1e’). This suggests that linear isonitriles 3, despite being generally more stable than their η2 side-bonded nitrile isomers, will not be observed as a main product in most cases. Another possible reason for the fact that 3 has not been described experimentally is that this species may be a transient intermediate that quickly undergoes subsequent C–H activation reactions with the solvent. Indeed, rhodium isonitrile complexes have been reported to be capable of C–H activation.24 . This possibility was examined by calculating the thermodynamics of both the substitution of the isonitrile ligand by benzene –yielding a η2 coordination complex– and the subsequent oxidative addition of the metal into a benzene C–H bond (see Supporting Information). In all cases, these processes were calculated to be highly endergonic (∆G ~ 27 to 42 kcal mol-1). However, the same chemical steps were found to be much more thermodynamically feasible from the corresponding η2 side-bonded nitrile isomers 2, particularly for Ir complexes (∆G ~ -2 to 5 kcal mol-1) for which such C–H insertion adducts were experimentally isolated (Table 1, entries 2 and 3). The higher stability of the exchanged benzonitrile ligand with respect to isocyanobenzene is likely the thermodynamic driving force for this C–H insertion reaction from the η2 side-bonded nitrile isomers.

Table 2. Gibbs free energies (∆G) (in kcal mol-1) for the decarboxylation transition states



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

and the derived products, with respect to the reactants (metallaisoxazolin-5-ones, 1’) calculated

with

SMD(benzene)/M06/6-311+G(2d,p)+SDD//M06/6-

31G(d)+SDD(Ir,Rh,Co). Calculations from real metallaisoxazolin-5-ones (1) are also show for comparison.

Initial complex

Pathway a ∆G‡

∆G

Pathway b ∆G‡

∆G

model systems 1a’-f’

TS1a’-f’ 2a’-f’ TS2a’-f’ 3a’-f’

a’

30.1

-46.4

29.6

-53.3

b’

32.2

-34.6

32.8

-41.8

c’

21.7

-56.3

d’

24.8

-45.0

e’

20.1

-59.2

f’

24.6

-49.1

path not found 27.4

-46.9

path not found 26.4

-48.7

real systems 1a,b

TS1a,b

2a,b

TS2a,b

3a,b

a

28.0

-43.0

31.1

-46.7

b

31.8

-38.0

31.6

-44.5


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


Figure

2:

Structures

for

TS1a’-f’

calculated

with

SMD(benzene)/M06/6-

311+G(2d,p)+SDD//M06/6-31G(d)+SDD(Ir,Rh,Co). Activation free energies are given in kcal mol-1 and distances in angstroms.

In organic systems, a concerted pericyclic process like TS1 is thermally forbidden by the Woodward-Hoffmann rules25,26. However, when a transition metal is present in the ring, symmetry rules can be overcome by the existence of extra d and f orbitals on the metal. Accordingly, the participation of a metal lone pair in the formation of a transient M=Cβ bond, has been proposed to facilitate the N–Cβ cleavage in the stepwise osmium-mediated



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

fragmentation of β-lactams, which has a prohibitively large activation barrier in the absence of the catalyst.27 The varying electron density around the metal center, and subsequently the metal-carbon bond strength, are related to the stabilities of the different metallacycles. The electronic density is both influenced by ligands and the metal itself. Phosphines are better σ donors and increase the electron density on the metal center while CO is a strong back-bonding acceptor, which reduces the metal electronic density. As a consequence, complexes with PMe3 are generally more stable than those with a CO ligand, i.e. show a higher barrier for the decomposition reaction (1b’, 1d’ and 1f’ vs. 1a’, 1c’ and 1e’, respectively).These general trends were verified through an NBO charge analysis, which can be used as a quantitative indicator of the electronic density around the metal center (Table 3). The values of the calculated activation barriers (Table 1) agree with the trends expected from the metal electron density in the reactants, and correlate well with the experimental observations: Ir complex 1b decomposes at the highest temperature (110 ℃) and has the highest calculated barrier (calculated ∆G‡ = 32 kcal mol-1 for 1a’), while Rh and Co complexes 1c and 1d (calculated ∆G‡ = 25 kcal mol-1 for both 1c’ and 1e’) rearrange at lower temperatures and shorter times.

Table 3. NBO charges at the metal center of the starting metallaisoxazolin-5-ones. System Charge

1a’

1b’

1c’

1d’

1e’

1f’

+0.40 +0.27 +0.24 +0.15 +0.22 +0.20


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


The pathway through TS2 is a more complicated process. The calculated structures of two transitions states are given for comparison (Figure 3). In TS1 the 5-membered ring is slightly non-planar, and the phenyl group remains coplanar with the metallacycle. However, in TS2 the 5-membered ring is planar, but the phenyl group rotates to be approximately perpendicular to it at the saddle point. Since decarboxylation and phenyl migration happen together in TS2, the cleavage of the C–O bond is quite advanced, and migration of the phenyl occurs to ensure a chemically reasonable bonding for the nitrogen atom. In TS1, this was supplied by bonding of N to the metal. Bond lengths, bond angles and dihedrals involved in the reaction are represented along the intrinsic reaction coordinates (IRC) of TS1 and TS2 of the reaction of 9a to further look into the reaction process (Supporting Information).

Figure 3. Geometries of the two competing transition states (TS1 and TS2) for the thermal decomposition of 1b’ calculated with SMD(benzene)/M06/6-311+G(2d,p)+SDD//M06/6-



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

31G(d)+SDD(Ir,Rh,Co). Activation free energies are given in kcal mol-1 and distances in angstroms

Since the reaction can also be carried out photochemically (Table 1, entry 3)6 we have also investigated a triplet excited state process. For the triplet state, a transition state was found where the N–O bond breaks homolytically. The overall activation energy of such a process can be evaluated by the relative energy difference of the triplet transition state and the singlet reactant, giving a 56 kcal mol-1 barrier, which is obviously too high to be practical thermally, but readily achievable photochemically. IRC calculations (Figure 4) trace bonding changes from the transition state to the reactant and product. In the triplet surface, the reaction proceeds in a stepwise manner where the N–O bond breaks first, generating a biradical which extrudes CO2. This process is almost identical to the mechanism of CO2 extrusion from non-metallic five-membered rings reported in the literature.28


Page 14 of 22

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


Figure 4. Decarboxylation of 1b’ calculated in the triplet state with SMD(benzene)/M06/6311+G(2d,p)+SDD//M06/6-31G(d)+SDD(Ir,Rh,Co). Electronic energy and the bond lengths of involved bonds plotted along the reaction progress represented by intrinsic reaction coordinates (IRC) showing the reaction progress. The zero of such reaction progress is arbitrarily set to be at the transition state TS1b’-triplet.

Stability of plausible isomer products Beside the η1 isonitrile complex 3 derived from TS2, the side-bonded nitrile 2 was also found computationally to undergo a ring-opening isomerization process to generate the η1 nitrile complex 4 (Figure 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

Cp

L M

L

Ph Cp N

2

Ph

M N

Cp

TS3

Page 16 of 22

L M N 4

Ph

Figure 5. Isomerization of side-bonded nitrile 2.

The calculated energies of this process are given in Table 4. In all the systems studied, the barrier for this process is generally lower than TS1, from which the side-bonded nitrile complex is generated. Therefore, in a reaction where η2 nitrile complex 2 is observed, isomerization should also be kinetically possible. Theoretically, the two isomers could be in a thermal equilibrium. In that case the thermodynamically more stable species would be the main product. The Gibbs free energies of the two isomers were calculated and compared (Table 4).

Table 4. Gibbs free energies (∆G) (in kcal mol-1) for the isomerization of side-bonded nitrile 2, with respect to the reactants (metallaisoxazolin-5-ones, 1’) calculated with SMD(benzene)/M06/6-311+G(2d,p)+SDD//M06/6-31G(d)+SDD(Ir,Rh,Co). The intrinsic activation barriers for the isomerization process with respect to each η2 nitrile complex (∆∆G‡) are shown in parentheses. Calculations from real metallaisoxazolin-5-ones (1) are also show for comparison.


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


Initial complex

Isomerization ∆G

∆G‡ (∆∆G‡)

∆G

model systems 1a’-f’

2a’-f’

TS3a’-f’

4a’-f’

a’

-46.4

-32.5 (13.9)

-48.3

b’

-34.6

-10.6 (24.0)

-30.0

c’

-56.3

-49.2 (7.1)

-60.5

d’

-45.0

-29.4 (15.5)

-43.4

e’

-59.2

-50.4 (8.7)

-65.7

f’

-49.1

-31.5 (17.7)

-48.9

real systems 1a,b

2a,b

TS3a,b

4a,b

a

-43.0

-26.7 (16.4)

-41.7

b

-38.0

-15.9 (22.1)

-34.9

According to NBO charge analysis (Table 5), CO, as a back-bonding acceptor, makes the metal less electron-rich in intermediates 2a’ and 2c’ than in those bearing PMe3 as a ligand (2b’ and 2d’, respectively). This ligand effect weakens the metal-carbon bond in the sidebonded nitrile complex and makes it less stable. Here, the η1-nitrile complex becomes more stable, whereas with phosphines there is a clear preference towards the η2 side-bonded nitrile complexes, in good line with experiments (Table 1, entries 4-6).

Table 5. NBO charges at the metal center of the η2 nitrile complexes. System

2a’

2b’

2c’

2d’

2e’

2f’

Charge

+0.42

0.26

0.26

+0.15

+0.33

+0.30



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

For the experimental systems, the more π donor of Cp* ligand in 1a, overcompensate the trend exhibited by the model complexes bearing Cp. Hence, the experimentally isolated and characterized η2 side-bonded nitrile complexes (2a and 2b) are confirmed to be the more stable species, in good agreement with the experimental results. As a representative example, Figure 6 shows the complete energy landscape for complex 1a, including all the calculated species and the transition structures connecting them.

Figure 6. Energy landscape for the decarboxylation (pathways a and b) and subsequent isonitrile

isomerization

of

metallaisoxazolin-5-one

1a

calculated

SMD(benzene)/M06/6-311+G(2d,p)+SDD//M06/6-31G(d)+SDD(Ir,Rh,Co).

with

Shadowed

circle represents the predicted major species.

CONCLUSIONS The decarboxylation of metallaisoxazolin-5-ones occurs in a retro-[2+2]-cycloaddition manner to generate the experimentally observed η2 side-bonded nitrile complexes. The


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


reaction is highly exothermic and the barrier is generally low enough to be overcome under normal laboratory conditions. An alternative decarboxylation pathway in which decarboxylation and aryl migration happen concertedly to generate linear isonitrile isomers, has been found for many of the studied systems. The activation barrier of this process is generally higher, but it becomes competitive in some iridium complexes, which are consistently less reactive than Rh and Co analogues. The electronic density on the metal center is found to be crucial to the reactivity of the metallacycles as well as the relative stability of the products. The η1bonded isonitrile or nitrile complexes are suggested to be theoretically possible, and the η2 side-bonded complexes

are likely transient species involved in the experimentally

observed subsequent C–H functionalization of solvent molecules.

ASSOCIATED CONTENT Supporting Information. Additional figures, Cartesian coordinates, electronic energies, entropies, enthalpies, Gibbs free energies and lowest frequencies of the calculated structures. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Authors *K.N.H: e-mail, [email protected] *G.J-O: e-mail, [email protected] Notes The authors declare no competing financial interest.



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

ACKNOWLEDGEMENTS We are grateful to the National Science Foundation (CHE-1361104) and D.G.I. MINECO/FEDER (projects CTQ2015-70524-R and RYC-2013-14706 to G.J.O.) for financial support. Computing resources used in this work were provided CESGA, UR (Beronia cluster), UCLA Academic Technology Services/Institute for Digital Research and Education and IDRE and XSEDE (NSF OCI-1053575).

REFERENCES (1)

Chetcuti, P. A.; Walker, J. A.; Knobler, C. B.; Hawthorne, M. F. Organometallics

1988, 7, 641-650. (2)

Chetcuti, P. A.; Hawthorne, M. F. J. Am. Chem. Soc. 1987, 109, 942-943.

(3)

Wright, T. C.; Wilkinson, G.; Motevalli, M.; Hursthouse, M. B. J. Chem. Soc.,

Dalton Trans. 1986, 2017-2019. (4)

Jackson, A. B.; Schauer, C. K.; White, P. S.; Templeton, J. L. J. Am. Chem. Soc.

2007, 129, 10628-10629. (5)

Khosla, C.; Jackson, A. B.; White, P. S.; Templeton, J. L. Organometallics 2012, 31,

987-994. (6)

Chetcuti, P. A.; Knobler, C. B.; Hawthorne, M. F. Organometallics 1988, 7, 650-

660. (7)

Tafesh, A. M.; Weiguny, J. Chem. Rev. 1996, 96, 2035-2052.

(8)

Mooibroek, T. J.; Bouwman, E.; Drent, E. Organometallics 2012, 31, 4142-4156.

(9)

Rzepa, H. S.; Wentrup, C. J. Org. Chem. 2013, 78, 7565-7574.

(10)

Ugur, I.; Cinar, S. A.; Dedeoglu, B.; Aviyente, V.; Hawthorne, M. F.; Liu, P.; Liu, F.;

Houk, K. N.; Jiménez-Osés, G. J. Org. Chem. 2017, 82, 5096–5101.


Page 20 of 22

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


(11)

Gaussian 09, R. A.; Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; J. R. Cheeseman; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; J. A. Montgomery, J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.; 2009; Gaussian, I., Wallingford CT. (12)

Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.

(13)

Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.

(14)

Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.

(15)

Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011, 32, 1456-1465.

(16)

Ribeiro, R. F.; Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2011,

115, 14556-14562. (17)

Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154-2161.

(18)

Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523-5527.

(19)

Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378-

6396. (20)

Glendening, E.; Badenhoop, J.; Reed, A.; Carpenter, J.; Bohmann, J.; Morales, C.;

Weinhold, F. Theoretical Chemistry Institute, University of Wisconsin: Madison, WI 2001.



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

(21)

Xue, L.; Su, W.; Lin, Z. Dalton Trans. 2010, 39, 9815-9822.

(22)

Xue, L.; Su, W.; Lin, Z. Dalton Trans. 2011, 40, 11926-11936.

(23)

Fromm, A.; van Wüllen, C.; Hackenberger, D.; Gooßen, L. J. J. Am. Chem. Soc.

2014, 136, 10007-10023. (24)

Jones, W. D.; Hessell, E. T. Organometallics 1990, 9, 718-727.

(25)

Hoffmann, R.; Woodward, R. B. J. Am. Chem. Soc. 1965, 87, 2046-2048.

(26)

Woodward, R. B.; Hoffmann, R. Angew. Chem., Int. Ed. Engl. 1969, 8, 781-853.

(27)

Casarrubios, L.; Esteruelas, M. A.; Larramona, C.; Lledós, A.; Muntaner, J. G.;

Oñate, E.; Ortuño, M. A.; Sierra, M. A. Chem.-Eur. J. 2015, 21, 16781-16785. (28)

Rzepa, H. S.; Wentrup, C. J Org Chem 2013, 78, 7565-7574.


Page 22 of 22

Multiple Mechanisms for the Thermal Decomposition of

Recommend Documents