Kinetic Study of the Oxidative Addition Reaction between Methyl Iodide and [Rh(imino-β-diketonato)(CO)(PPh)3] Complexes, Utilizing UV–Vis, IR Spectrophotometry, NMR Spectroscopy and DFT Calculations

The oxidative addition of methyl iodide to [Rh(β-diketonato)(CO)(PPh)3] complexes, as modal catalysts of the first step during the Monsanto process, are well-studied. The β-diketonato ligand is a bidentate (BID) ligand that bonds, through two O donor atoms (O,O-BID ligand), to rhodium. Imino-β-diketones are similar to β-diketones, though the donor atoms are N and O, referred to as an N,O-BID ligand. In this study, the oxidative addition of methyl iodide to [Rh(imino-β-diketonato)(CO)(PPh)3] complexes, as observed on UV–Vis spectrophotometry, IR spectrophotometry and NMR spectrometry, are presented. Experimentally, one isomer of [Rh(CH3COCHCNPhCH3)(CO)(PPh3)] and two isomers of [Rh(CH3COCHCNHCH3)(CO)(PPh3)] are observed—in agreement with density functional theory (DFT) calculations. Experimentally the [Rh(CH3COCHCNPhCH3)(CO)(PPh3)] + CH3I reaction proceeds through one reaction step, with a rhodium(III)-alkyl as the final reaction product. However, the [Rh(CH3COCHCNHCH3)(CO)(PPh3)] + CH3I reaction proceeds through two reaction steps, with a rhodium(III)-acyl as the final reaction product. DFT calculations of all the possible reaction products and transition states agree with experimental findings. Due to the smaller electronegativity of N, compared to O, the oxidative addition reaction rate of CH3I to the two [Rh(imino-β-diketonato)(CO)(PPh)3] complexes of this study was 7–11 times faster than the oxidative addition reaction rate of CH3I to [Rh(CH3COCHCOCH3)(CO)(PPh3)].


Introduction
Rhodium complexes are used in several industrial processes as catalysts [1][2][3]. Rhodium complexes, as catalysts, are used, for example, in the hydrosilylation crosslinking of silicone rubber [4], for making nitric acid [5] and in the hydrogenation of benzene to cyclohexane [6]. The [Rh(β-diketonato)(CO)(PPh) 3 ] complex, with (β-diketonato = acetylacetono (acac)) as a hex-1-ene hydroformylation catalyst, produced 68% hex-2-ene [7]. Probably the most wellknown example of a rhodium process used in the industry is the rhodium catalyzed carbonylation of methanol to produce acetic acid by the Monsanto process [8][9][10], where the first organometallic step involves the oxidative addition of methyl iodide to cis-[Rh(CO) 2 I 2 ] − . [Rh(β-diketonato)(CO)(PPh) 3 ] complexes, as model catalysts of the first step of the Monsanto process, are well-studied [11,12]. It is found that more electronegative groups attached to the backbone of the β-diketonato ligand and slowed down the oxidative addition reaction rate of methyl iodide to [Rh(β-diketonato)(CO)(PPh) 3 ] complexes [11]. The β-diketonato ligand is a bidentate (BID) ligand that bonds through two O donor atoms (O,O-BID ligand) to rhodium. Imino-β-diketones are similar to β-diketones, though the donor atoms are N and O, hereby referred to as a N,O-BID ligand. The influence of the electronegativity of

Synthesis of Rhodium(I) Complexes
The Rh-dicarbonyl complexes [Rh(CH 3 COCHCNHCH 3 )(CO) 2 ] 2A and [Rh(CH 3 CO CHCNPhCH 3 )(CO) 2 ] 2B were synthesized from µ-dichloro-tetracarbonyldirhodium (obtained from Sigma Aldrich, Dramstadt, Germany) and the appropriate ligands, (CH 3 COCH 2 CNHCH 3 ) 1A and (CH 3 COCH 2 CNPhCH 3 ) 1B, respectively, in a methanol solution, see Scheme 1. The red methanol solution of the dirhodium complex turned to a yellow colour after the addition of the ligands. The extraction of the dicarbonyl complexes was performed with n-hexane, until the organic solvent that was used to extract it was clear. The collected solvent containing the product was then concentrated by evaporation. After a fine powder crystallized, the solution was filtered again, and the complex crystallized slowly from the n-hexane solution. Varshavsky reported the synthesis and properties of 2A [13] and related Rh-dicarbonyl complexes containing β-aminivinyl ketones [14]. The structure of 2B was confirmed by solid state crystal analysis [15].
Synthesizing the triphenylphosphine, containing Rh complexes [Rh(CH 3 COCHCNHCH 3 ) (CO)(PPh 3 )] 3A and [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B, was performed similarly to the method of various researchers, when synthesizing Rh-monocarbonyl-triphenylphosphineβ-diketonato complexes from Rh-dicarbonyl-β-diketonato complexes [16][17][18]. The slow addition of the triphenylphosphine hexane solution to the dicarbonyl Rh complex hexane solution yielded tiny bubbles of CO gas, formed during the reaction. The reaction was left approximately 5 min, until no more bubbles were formed and the flask was then left at room temperature (with a slight draft in open air). The crystallization of the final products was performed in a manner similar to the dicarbonyl complexes. Varshavsky reported the synthesis and properties of many related Rh-monocarbonyl-β-ketoiminato complexes containing tertiary phosphines [19]. Two isomers are possible for each of 3A and 3B-one with the N β-ketoiminato trans to CO (N-trans-CO) and one with O β-ketoiminato trans to CO (Otrans-CO), as shown in Scheme 1. On the 1 H-NMR 8.6% 3A isomer, N-trans-CO and 91.4% 3A isomer O-trans-CO is observed, while 100% 3B isomer O-trans-CO and no 3B isomer N-trans-CO is observed. The major isomer of 3A and 3B, observed as the O-trans-CO form, agrees with the fact that the 3A isomer O-trans-CO [20] and molecules related to 3B isomer O-trans-CO [21][22][23][24], could be isolated and characterized by solid state crystal structures.
On IR, it was not possible to distinguish between the different isomers of 3A and 3B-only one peak, corresponding to the CO stretching, was observed, at 1952.7 (3A) and 1966.6 (3B) cm −1 . The IR stretching frequencies of the carbonyl group in 3A and 3B are at lower wavenumbers than that of the dicarbonyl complexes 2A (2044. 1 and 1971.0 cm −1 ) and 2B (2059.0 and 1997.7 cm −1 ), respectively. This is indicative of a strengthening of the CO bonds, due to the addition of the more electron-rich triphenylphosphine ligand to the complex. The more electronegative groups or atoms will attract electrons in a bond more towards themselves, therefore slightly shortening and, thus, strengthening the bonds-which shifts the stretching frequency to smaller values.

DFT Study of Rhodium(I) and Rhodium(III) Complexes
The density functional theory (DFT) calculated optimized structures of the two possible isomers each of 3A and 3B are shown in Figure 1. The complexes have a square planar geometry, though the 3B isomer N-trans-CO exhibits a distorted square planar geometrydue to the steric stress between the phenyl ring that is attached to the N atom and the phenyl rings attached to the P atom. This stress causes the P atom to be pushed out of the plane and, therefore, 3B isomer N-trans-CO has a higher energy than 3B isomer O-trans-CO and is less stable.  Figure 2. The highest occupied molecular orbitals (HOMOs) of both 3A and 3B are mainly of rhodium d z 2 character, see Figure 2. The d z 2 HOMO plays an important role during the oxidative addition reaction between rhodium(I) complexes with CH 3 I [25,26], as will be discussed in Section 2.3.4 (below).

Rhodium(III)
Oxidative addition to the two possible isomers of [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A, can theoretically lead to 12 possible Rh(III)-alkyl products (indicated with 1a-12a in Table 1, as well as their enantiomers 1b-12b) and 6 possible Rh(III)-acyl products (indicated with 13a-18a, as well as their enantiomers 13b-18b, in Table 1). The relative electronic energies of the optimized geometries of all the possible Rh(III) isomers that could be optimized are also given in Table 1. Some of the Rh(III)-acyl geometries were converted during the optimization to the geometries of the lowest energy Rh(III)-acyl isomers. The structures and DFT-calculated energies of the possible alkyl and acyl reaction products of the [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B + CH 3 I reaction are given in Table 2. Since enantiomers have the same chemical properties and energies, we will only discuss the 12 possible Rh(III)-alkyl products (1a-12a) and 6 possible Rh(III)-acyl products (13a-18a). The energies obtained for the different Rh(III) products 1a-18a in Tables 1 and 2 lead to the following insights:

•
Products of [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A + CH 3 I reaction (Table 1): Evaluating the energies of the different Rh(III)-alkyl isomers in Table 1, it can be seen that the most preferred orientation of the product geometry is with the O β-ketoiminato atom O atom of the ligand trans to CO (O-trans-CO) and the CH 3 and I atoms bonded at the apical positions (top and bottom, with respect to the square planar geometry), since it has the lowest relative energy (2a, reaction product of CH 3 I and 3A isomer O-trans-CO). The Rh(III)-alkyl isomer with the N β-ketoiminato atom of the ligand trans to CO, and the CH 3 and I atoms bonded at the apical positions (1a, reaction product of CH 3 I and 3A isomer N-trans-CO) is only slightly higher in energy (0.07 eV). The energetically-preferred Rh(III)-alkyl isomer to form is, thus, 2a, while 1a may form in very small amounts. Evaluation of the energies of the Rh(III)-acyl isomers in Table 1, indicates that the two Rh(III)-acyl isomers 13a and 14a with the COCH 3 ligand in the apical position have the lowest energy. Furthermore, the energy of Rh(III)-acyl isomers 13a and 14a are ca. 0.3 eV lower than the energy of the lowest energy Rh(III)-alkyl isomers 1a and 2a. This indicates that the Rh(III)-alkyl product of oxidative addition could convert, via CO insertion, to the lower more stable Rh(III)-acyl product as the final reaction product. Rh(III)-acyl 13a has slightly higher energy (0.04 eV) than 14a, therefore, the amount of 13a will be less than 14a.

Kinetics of Iodomethane Oxidative Addition to Rhodium(I)
The oxidative addition reactions between the two [Rh(N,O-BID)(CO)(PPh 3 )] complexes 3A and 3B and CH 3 I are presented. The reaction is followed by IR, UV-Vis, and NMR. Additionally, a DFT study of the reaction will be presented to compliment the experimental findings. In each subsection, the results of 3B will be presented first, followed by the results for 3A, since it was experimentally found that 3B exhibited a single reaction step, while the 3A + CH 3 I reaction showed two steps.

IR Study
The oxidative addition reaction was first followed on IR, since this technique is ideal to distinguish between CO bonds in metal-CO complexes of Rh(I)-carbonyl and Rh(III)complexes. CO groups in Rh(I)-carbonyl complexes resonate at~1980-2000 cm −1 , CO groups in Rh(III)-alkyl-carbonyl complexes resonate at~2050-2100 cm −1 and CO groups in Rh(III)-acyl complexes resonate at~1700-1750 cm −1 [27].
In Figure 3a, which displays the IR spectra of the CH 3 I reaction with [Rh(CH 3 COCHCN PhCH 3 )(CO)(PPh 3 )] 3B, it can be seen that there are two main peaks that vary during the reaction. The first peak, at 1973.62 cm −1 , is assigned to the disappearance of the Rh(I) complex as the reaction progresses and a product is formed, see Figure 3c. The second peak at 2059.72 cm −1 is assigned to the formation of the Rh(III)-alkyl reaction product. The rate constants determined from the IR kinetic experiments for different CH 3 I concentrations are given in Table 3. From Table 3, it can be seen that the rate of disappearance of [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B and rate of the formation of the Rh(III)-alkyl product show very good correlation, since the values are so close to each other. This trend is seen across all three concentrations. The increase in the observed kinetic rate constant with the increase in CH 3 I concentration indicates that this reaction step is concentration-dependent. From the results obtained from the IR study, the reaction step of the [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B + CH 3 I reaction can, thus, be presented as Scheme 2a.   In Figure 3b, which displays the IR spectra of the CH 3 I reaction with [Rh(CH 3 COCHCN HCH 3 )(CO)(PPh 3 )] 3A, it can be seen that there are three main peaks that vary during the reaction. The Rh(I) complex displays a CO stretching frequency peak at 1967.66 cm −1 . This absorbance peak decreases as the reaction progresses, which is explained by the disappearance of the Rh(I) complex, as it reacts with CH 3 I, in order to form a Rh(III) complex. The first product that forms is seen at the increase of the absorbance peak position at 2061.30 cm −1 . This is attributed to the formation of a Rh(III)-alkyl complex after the direct addition of the CH 3 I to the Rh centre. This peak shows an increase at first, which shows the formation of the Rh(III)-alkyl product, up to a maximum absorbance level, after which it starts to decrease in intensity, see Figure 3d. This is attributed to the formation of a second Rh(III) product. When the Rh(III)-alkyl product starts to decrease in intensity, the formation of a second Rh(III) product can be seen at an absorbance peak of 1713.74 cm −1 , see Figure 3b. The peak at 1713.74 cm −1 is designated to the Rh(III)-acyl complex. The intensity of the peak at 1713.74 cm −1 increases, up to a maximum amount, Figure 3d. The formation of the Rh(III)-acyl is due to CO insertion of the Rh(III)-alkyl product, and this step is called the second reaction step. Rh(III)-acyl is the final product observed for the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A + CH 3 I reaction. The observed pseudo first-order reaction constant calculated at the indicated wavelengths for each reaction step is given in Table 3. From the data presented in Table 3 it can be seen that the observed rate constants for the loss of the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A complex and the formation of the Rh(III)-alkyl product (the first reaction step), indicates a CH 3 I concentration-dependent step. The small differences between the rate constants for the Rh(I) loss and the Rh(III)-alkyl gain are due to the fact that the Rh(III)-alkyl is already converted to Rh(III)-acyl, while the first reaction step proceeds. Thus, the Rh(III)-alkyl increase during the first reaction step is, thus, virtually smaller [18]. The observed rate constants for the loss of the Rh(III)alkyl product and formation of the Rh(III)-acyl product show good correlation with each other. There is barely any significant change in the rate values over the concentration range for the second step, which indicates that it is independent of the CH 3 I concentration. This is expected, since the conversion between the two products is not dependent on the concentration of CH 3 I, but on the individual properties of the complexes themselves, i.e., the type of ligand system employed. From the results obtained from the IR study, the reaction steps of the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A + CH 3 I reaction can, thus, be presented as given in Scheme 2b.

UV-Vis Study
The in situ analysis of the oxidative addition reaction between the [Rh(N,O-BID)(CO) (PPh 3 )], with N,O-BID = (CH 3 COCHCNHCH 3 ) − 1A and (CH 3 COCHCNPhCH 3 ) − 1B, complexes and CH 3 I are also measured by UV-Vis spectroscopy. Only the first step of both reactions, the [CH 3 I] dependent step, was followed on UV-Vis.
The reaction between the [Rh(CH 3 COCHCNHCH 3 )(CO(PPh 3 )] 3A complex and CH 3 I yields spectra, as seen in Figure 4a. From Figure 4a, it is seen that there is a decrease of the absorption around a wavelength of 330 nm and increase of the absorption around 408 nm, as the reaction progresses. The decrease in the absorption (330 nm) is theorized to be the decrease in the concentration of the Rh(I) complex as it reacts with CH 3 I to form a Rh(III) complex. The increase in the absorption (408 nm) is identified as the Rh(III) product complex; however, the identification of the type of product is not possible. The Rh(III) products that are formed could either be the Rh(III)-alkyl, Rh(III)-acyl, or a mixture of both [17,28]. The absorbance changes, at both 330 nm and 408 nm, with time, are shown in Figure 4b.  Table 4, for the oxidative addition reaction of 3A and 3B. The activation enthalpy and entropy for the 3B and CH 3 I reaction in Figure 4c can be obtained from the Eyring plot in Figure 4d. The large and negative activation entropy for the first oxidative addition step is indicative of an associative mechanism, as commonly found for oxidative addition of CH 3 I to square planar d 8 metal complexes, which is proposed to proceed via an S N 2 mechanism [26,[29][30][31].

NMR Study
The UV-Vis spectrophotometric analysis yielded the second-order rate constants for the two phosphine-containing Rh complexes, and the IR spectrophotometric analysis yielded information regarding the types of products formed (alkyl and/or acyl) during the reaction between the Rh(I) complexes and CH 3 I. In order to gather more information, regarding the types of products and possible isomers involved in the reaction, an in situ analysis of the reaction was performed through the use of 1 H-NMR spectroscopy. All possible product structures, of which there are multiples for each product type, will be presented and discussed in this section.
The [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B and CH 3 I reaction yielded similar results, as with the IR study, in that there was only one product formed during the reaction, namely the Rh(III)-alkyl product. Only one Rh(I) reactant 3B and one Rh(III)-alkyl product isomer was observed on NMR. The methine and methyl proton signals were tabulated in Table 5, with the 1 H-NMR region presented in Figure 5. From Figure 5 (left), it can be seen that the resonance peak for the methine proton of the starting [Rh(CH 3 COCHCNPhCH 3 ) (CO)(PPh 3 )] 3B complex (5.018 ppm) decreased as the reaction progressed, and a resonance peak, representing the methine proton signal for the Rh(III)-alkyl product (5.004 ppm), increased during the course of the reaction [17,18,28]. Figure 5 (right) illustrates the change in the resonance peak intensities as the reaction progressed, with respect to the methyl proton signals. The [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B methyl resonances (blue arrows, 1.688 ppm and 1.567 ppm) decreased as the reaction proceeded, and the Rh(III)alkyl product that formed has methyl resonances (green arrows, 1.869 ppm and 1.655 ppm) that increased as the reaction proceeded. The doublet of doublets that formed (~1.370 ppm) is attributed to the CH 3 of the CH 3 I, attached to the Rh central atom (formation of an alkyl reaction product). The doublet of doublets signal of the CH 3 group of the alkyl was due to coupling with Rh (spin 1 2 ) and P (spin 1 2 ) [18,28]. This doublet of doublets signal also increased as the reaction proceeded. Figure 5 show that only one isomer of the Rh(III)-alkyl product formed during the reaction from a single isomer of 3B. Table 5 contains the observed pseudo first-order reaction rate constants, obtained from the change in concentration with time, of both the methine and methyl resonance signals. The observed kinetic rate constants, obtained from the methine and methyl peaks, when compared with those obtained through UV-Vis and IR spectroscopy, show good correlation between the values, see Tables 3-5.   3 2.157 (a) and 2.133 (b) 0.00150(1) N-CH 3 2.861 (a) and 2.555 (b) 0.00150 (1) a Unable to obtain sufficient statistical data. b Two Rh(III)-acyl isomers observed-the peaks of Rh(III)-acyl B were too small to obtain sufficient statistical data.
The [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A and CH 3 I reaction yielded similar behaviour to those observed in the IR study, in that there were two reaction steps observed during the reaction on NMR, according to Scheme 2b. The methine and methyl proton signals are tabulated in Table 5, with the 1 H-NMR region presented in Figure 6. Figure 6 (left) illustrates the methine proton signals as the reaction progressed. The pure [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A isomer A (5.002 ppm) and isomer B (4.990 ppm) showed a decrease in the proton signal intensity as the reaction progressed, with the formation of Rh(III)-alkyl appearing at 4.881 ppm. Only one Rh(III)-alkyl isomer was observed, but the existence of a second Rh(III) isomer cannot be excluded. The second Rh(I) isomer B is minute; therefore, the alkyl that could formed from the small Rh(I) isomer B could not be observed, since the peak was too small, or possibly due to overlap with the Rh(III)-alkyl isomer observed. The resonance peaks of the Rh(III)-alkyl isomers increased for a short time during the reaction, up to a maximum intensity, after which it starts to decrease. This is due to the formation of the Rh(III)-acyl product isomers (4.976 ppm and 5.014 ppm), which caused a decrease in the Rh(III)-alkyl isomers' signal intensity. It was observed that the intensity of the signals, due to the second Rh(III)-acyl isomer B, was very small and, although observed, the formation of the second Rh(III)-acyl isomer B could not be followed kinetically [26,27,32]. Figure 6 (right) displays the change in the methyl proton signals between the two isomers of the starting material, [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A, and alkyl product that form as the reaction progresses. The Rh(I) starting material isomers (blue arrows, 2.056 ppm and 1.583 ppm, main isomer) showed a decrease in the two methyl proton signals, and Rh(III)-alkyl product (green arrows, 2.017 ppm and 1.862 ppm) showed an increase in the signal strength at first. As the reaction progressed, signals of the Rh(III)-alkyl product isomers decreased, and the Rh(III)-acyl product isomers (main isomer A, pink arrows, 2.861 ppm and 2.157 ppm and minor isomer B, black arrows, 2.555 ppm and 2.133 ppm) showed an increase in the signal strength as time progresses. The reason for the Rh(III)-alkyl methyl proton resonances showing an increase, and then a decrease, is the same as for the methine proton focussed analysis that is: the Rh(III)-alkyl product starts forming up to a maximum, after which it starts to convert to the Rh(III)-acyl product [27,32]. Table 5 contains the observed pseudo first-order reaction rate constants of both the methine and methyl resonance signals. The observed kinetic rate constants, obtained from the methine and methyl peaks, when compared with those obtained through UV-Vis and IR spectrophotometry, showed a good correlation between the values, as seen in Figure 7b.

DFT Study of Reaction Mechanism
A DFT study of the reactants, transition state, and the products of the [Rh(CH 3 COCH CNPhCH 3 )(CO)(PPh 3 )] 3B + CH 3 I oxidative addition reaction is presented in Figure 8, with the results summarized in Table 6 (results for the most stable O-trans-CO isomer of 3B is presented). Figure 8a contains the optimized geometries of the reactant molecule, [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B, complex's reaction with CH 3 I, before the reaction starts. There was a large separation between the reactant molecules (d = 32.731 Å) at the moment of the mixture. The reactant molecules then move through the bulk solution, until the distance between them is short enough and orientation of the molecules is sufficient to facilitate a reaction to occur [26,32]. The oxidative addition reaction between CH 3 I and square planar Rh(I) complexes can theoretically proceed via three different transition state structures [33][34][35]. The three methods of addition are a linear/back addition (where there is a near 180 • angle formed between the approaching CH 3 I and Rh (trans addition)), bent addition (where the angle is between 90 • and 180 • (trans addition)), and front addition (where the angle is less than 90 • (cis addition)). DFT computations showed that the most preferred addition method was a linear/back approach of the CH 3 I to the Rh(I) metal centre. This method had an activation barrier that was~7 times lower in energy than the other two methods [26]. Figure 8b contains the optimized geometry of the linear oxidative addition of the methyl group to the Rh(I) metal centre of the [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B complex. The CH 3 I molecule, when it is close enough to the Rh complex molecule, approaches linearly from the apical position (top as illustrated), with the methyl group closest to the Rh complex. This approach of the CH 3 I molecule, and subsequent bond formation between the methyl group and Rh metal centre, cause an increase in the total system energy, up to a maximum value, the activation energy. After the Rh-C CH3I bond is formed, the energy of the system decreases again, as the bond formed stabilizes and the I − ion diffuses away [26,32]. The I − ion is then free to diffuse away into the bulk solution to react with another five-coordinate Rh complex. The bond formation between the fivecoordinate cationic intermediate and a I − ion is a barrierless process, with the formation of the Rh(III)-alkyl 2a product with the O β-ketoiminat atom of the ligand trans to CO, and the CH 3 and I atoms bonded at the apical positions (top and bottom with respect to the square planar geometry), see Figure 8c. The energies of the reaction steps calculated by DFT are summarized in Table 6. The energies are given relative to the final product in the reaction mechanism. The energy barrier of the transition state (TS) is 0.29 eV and 0.83 eV higher than the reactants and [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )(CH 3 )(I)]-alkyl 2a product, respectively. The energy profile is, thus, favourable for the product formation. All DFT attempts at determining a TS for formation of an acyl product through CO insertion at the Rh(III)-alkyl 2a failed. This fact, as well as the fact that Rh(III)-alkyl 2a and Rh(III)-acyl 13a are equi-energetic (see energies in Table 2), with no driving force for CO insertion to occur in 2a to form 13a, compliment the experimental findings that a Rh(III)-alkyl is the final reaction product of the [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B + CH 3 I reaction. The DFT-calculated results for the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A + CH 3 I oxidative addition reaction are presented in Figure 9, with the results summarized in Table 6. The results related to the most stable reactant isomer, the isomer with O β-ketoiminato trans to CO (O-trans-CO), is presented. Figure 9a contains the optimized geometries of the reactant molecules, [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A and CH 3 I, before the oxidative addition reaction proceeds. There is a large separation between the two molecules. The molecules merely move through the bulk solution until both the distance between the two is short enough and the orientation of the molecules is correct in order to facilitate a reaction. There is a distance between the two molecules >21.825 Å. As the molecules move through the bulk solution this distance decreases.  (CO)(PPh 3 )] 3A complex from the apical position linearly with the methyl group closest to the Rh metal centre. As the CH 3 I molecule approaches, the distance between the methyl group and I increases from 2.1736 Å (DFT optimized geometry of CH 3 I) to 2.610 Å, and the distance between the methyl group and Rh metal centre decreases, until a bond is formed. When the distance between the methyl group and the I atom reaches a certain length the bond between the two breaks, and the I − ion is then free to diffuse back into the bulk solution in order to react with another five-coordinate Rh intermediate that is formed. As the Rh-C CH3I distance decreases, the total energy of the system increases up to a maximum point. At this maximum the transition state occurs, where the CH 3 I molecule breaks up into a methyl group and I − ion, and the methyl group then forms a bond with the Rh metal centre to form a five-coordinate intermediate complex as seen in Figure 9b. After the bond is formed, the total system energy then starts to decrease. As the I − ion approaches the Rh metal centre, the total energy of the system decreases, up until the bond formation occurs, after which the system reaches a point of stability, as shown in Figure 9c, with the formation of the Rh(III)-alkyl complex.
The next reaction step is the CO insertion of the CH 3 migration step. The CH 3 group, thus, starts moving towards the C atom of the CO group. As the CH 3 migrates towards the CO group, the CO group gets lifted out of the plane formed between the N and O atoms of the ligand and P atom of the PPh 3 ligand by 15.50 • -effectively decreasing the C methyl -Rh-C co bond angle even more. During the migration the Rh-C methyl bond length increases, and the C methyl -C co distance decreases. These changes continue up to a certain point (C methyl -Rh-C co = 56.92 • ), where the Rh-C methyl bond breaks and the C methyl -C co bond forms [26,32]. As can be seen from Figure 9d, when the moment of the transition state is reached, the ligand starts shifting or flipping, in order to position the COCH 3 group in an apical position. Figure 9e illustrates the final product's geometry, optimized by DFT calculations, of the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )]-acyl product 14a. It can be seen that the acyl group is now in an apical position and I atom has taken the place of the PPh 3 group from the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )]-alkyl product 2a, with the PPh 3 group taking the place of the CO group of alkyl product 2a. This is due to this orientation having a smaller total system energy than any of the other orientations. After the bond is formed between the C methyl and the C CO atoms, the orientation starts to change. The six-membered ring, formed between the Rh metal centre and the L,L -BID ligand, flips. The O atom moves in such a way as to allow for the I atom to be positioned trans to the N atom. During this shift the energy of the system decreases up to a minimum where the I atom is trans and no longer in an apical position until the final, stable geometry of the five-coordinate Rh(III) complex was obtained [26,32].
The energies of the two reaction steps (calculated by DFT) are summarized in Table 6. The energies are given relative to the final product in the reaction mechanism. From Table 6 it can be seen that the first transition state, the linear addition of CH 3 I, to the structure has an energy barrier of only 0.26 eV. This energy barrier calculated for the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A + CH 3 I reaction, is favourable for the reaction to occur and to proceed to the Rh(III)-alkyl reaction product that is 0.82 eV lower in energy than the TS1. The energy barrier of the second step is higher than the first step (0.91 eV vs. 0.26 eV), indicating that the second reaction is much slower than the first reaction step, in agreement with experimental results. The Rh(III)-acyl product has the lowest energy and, thus, most stable and constitutes the final reaction product. These computational results are, thus, corroborated by both the IR and NMR spectra illustrating the formation of an alkyl and acyl product, with the acyl product being the final product.  (14a and 13a). The reaction scheme for this complex can be seen in Scheme 3b. it is found that for more electronegative the groups R and R attached to the backbone of the ligand, a slower oxidative addition reaction rate is obtained [11,36]. For example, the reaction rate for the oxidation addition step of the [Rh(CF 3 COCHCOCF 3 )(CO)(PPh 3 )] + CH 3 I reaction (0.00013(1) mol −1 dm 3 s −1 [37]) is more than 100 times smaller the [Rh(CH 3 COCHCOCH 3 ) (CO)(PPh 3 )] + CH 3 I reaction (0.024(3) mol −1 dm 3 s −1 [38]). This is because the two CF 3 groups attached to the ligand in [Rh(CF 3 COCHCOCF 3 )(CO)(PPh 3 )] are more electronegative than the two CH 3 groups (χ CF 3 = 3.01 and χ CH 3 = 3.34 [25] [38]) is slower than the oxidative addition reaction rate of CH 3 I to 3A and 3B, 0.144(1) mol −1 dm 3 s −1 and 0.213(5) mol −1 dm 3 s −1 , respectively. The more electronegative O atom in [Rh(CH 3 CO CHCOCH 3 )(CO)(PPh 3 )] withdraws more electron density from it's d z 2 , HOMO, than the N atom in 3A and 3B (Figure 2). The nucleophilic attack of the rhodium(I) centre on the CH 3 of the CH 3 I group, thus, proceeds slower in the case of [Rh(CH 3 COCHCOCH 3 )(CO)(PPh 3 )] than for 3A and 3B.

Instrumentation
Characterization of the complexes were performed on a Bruker Avance DPX 300 NMR The ADF program (Software for Chemistry & Materials, Amsterdam, The Netherlands) [39] was utilized in the calculations of the complexes. A TZP basis set with the PW91 [40] functional was utilized in the optimization of the geometries. The theoretical IR frequencies were calculated with the selected basis set and functional to obtain zero point energy (ZPE) corrected electronic energies. Solvent effects were taken into account using chloroform as solvent, using the COSMO [41][42][43] model of solvation with an Esurf cavity [44]. Electronic energies for the rhodium(III) products were also calculated, using the optimized PW91 geometries and the TPSS [45,46] and PBE [47] functionals, respectively, both including Grimme's D3 [48] dispersion correction and the TZ2P basis set. The optimized coordinates of the DFT calculations are provided in Supplementary Materials.

Synthesis
The synthesis of the ligands 1A and 1B was performed as published in literature [15].

Kinetics
Oxidative addition reactions between the triphenylphosphine-containing Rh complexes, 3A and 3B, and iodomethane, CH 3 I, were monitored on UV-Vis spectrophotometry, IR spectrophotometry and 1 H-NMR spectrometry, as described in our previous publications [11,36]. All kinetic measurements were performed under the indicated pseudo firstorder conditions (CH 3 I being in excess, with respect to the concentration of the rhodium complex). The observed first-order rate constants k obs were obtained from the slope of the graph of ln(A inf − A t ) vs. time data, according to the first-order kinetic equation: ln(A inf − A t ) = −k obs t + ln(A inf − A 0 ) where A t = absorbance of a complex at time t, A inf and A 0 is absorbance of complex at time infinite and zero, respectively. The second-order rate constants, k 2 (in dm 3 mol −1 s −1 ), were obtained from the slope of the graph of k obs against the concentration of CH 3 I: k obs = k 2 [CH 3 I] + k −2 where k 2 (k −2 ) is the second-order rate constant of the forward (backward) reaction of the kinetic reaction between the Rh complexes 3A and 3B and CH 3 I.
Activation parameters were obtained from the Eyring relationship: where ∆H # = activation enthalpy, ∆S # = activation entropy, T = temperature, k 2 = secondorder rate constant of the first kinetic step at temperature T, k B = Boltzmann's constant, h = Planck's constant, R = universal gas constant.

Conclusions
Experimentally, one isomer of [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] 3B and two isomers of [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] 3A were observed. DFT calculations showed that the 3B isomer N-trans-CO exhibited a distorted square planar geometry, due to the steric stress between the phenyl ring, which is attached to the N atom and phenyl rings attached to the P atom, causing the 3B isomer N-trans-CO to be less stable. The electronic energies of the optimized molecules, in the Boltzmann equation, predicted 100% of 3B isomer O-trans-CO (no 3B isomer N-trans-CO) and 96.5% of 3A isomer O-trans-CO, in agreement with experimental observation.
Experimentally, the [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] (3B) + CH 3 I reaction proceeds through one reaction step, with a rhodium(III)-alkyl as final reaction product. The DFT calculations of all the possible reaction products and transition states, in agreement with experimental findings, showed that, for the experimentally observed one-step reaction [Rh(CH 3 COCHCNPhCH 3 )(CO)(PPh 3 )] (3B) + CH 3 I, the lowest energy Rh(III)-alkyl 2a and Rh(III)-acyl 13a reaction products are equi-energetic. There is, thus, no driving force for CO insertion to occur in 2a to form 13a.
Experimentally, the [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] (3A) + CH 3 I reaction proceeds through two reaction steps, with rhodium(III)-acyl as the final reaction product. The DFT calculations of all the possible reaction products and transition states, in agreement with experimental findings, showed that, for the experimentally observed two-step reaction [Rh(CH 3 COCHCNHCH 3 )(CO)(PPh 3 )] (3A) + CH 3 I, the lowest energy Rh(III)-acyl isomers, 13a and 14a, were ca. 0.3 eV lower in energy than the energy of the lowest energy Rh(III)alkyl isomers, 1a and 2a, providing a driving force for CO insertion to occur in the 1a and 2a to form 13a and 14a, respectively.
The kinetic rate of the oxidative addition of methyl iodide to the two [Rh(imino-βdiketonato)(CO)(PPh) 3