Synthesis and Determination of [1,3,5-C6H3(CH3)3]Mo(CO)3
The reaction of mesitylene with Mo(CO)6 under reflux yields [1,3,5-C6H3(CH3)3]Mo(CO)3 in low percent yield (around 1%). 1H NMR of [1,3,5-C6H3(CH3)3]Mo(CO)3 shows singlets at δ 2.25 and 5.23 with absorption ratios of 9:3, respectively. 1H NMR of mesitylene shows singlets at δ 2.25 and 6.78, also with absorption ratios of 9:3, respectively. This suggests addition of the metal complex to mesitylene causes downfield shifting of the signal for protons attached directly to the ring as they are unshielded from the backbonding of carbonyl groups. The IR spectrum of [1,3,5-C6H3(CH3)3]Mo(CO)3 shows a strong antisymmetric C-O stretch at 1852 cm-1 and a medium symmetric C-O stretch at 1942 cm-1 with peak areas of 64.462 cm-1 and 9.111 cm-1 respectively. The calculated OC-Mo-CO bond angle is 108.32°.
The reaction of mesitylene with Mo(CO)6 under reflux yields [1,3,5-C6H3(CH3)3]Mo(CO)3. The reaction specifically takes place in the following manner:
This is compound of interest because it a metal-arene complex and can be considered to be an octahedral rather than tetrahedral complex. This is because the OC-Mo-CO bond angles are close to 90° instead of the expect 109.5° for tetrahedrals.1 In order to determine the structure of said substance from its 1H NMR, the peaks must be compared to the same spectrum of mesitylene for indication of identical methyl group peaks and downfield shifting a peak indicative of protons attached directly to the ring. The IR spectrum of [1,3,5-C6H3(CH3)3]Mo(CO)3 can be analyzed for peaks indicative of symmetrical and antisymmetrical carbonyl stretching, whose areas can be used to calculate the bond angle between the carbonyl groups attached the to metal.
All syntheses were carried out in nitrogen and the reagents and solvents were purchased from commercial sources and used as received unless otherwise noted. The synthesis of [1,3,5-C6H3(CH3)3]Mo(CO)3 (1) was based on reports published previously.1
[1,3,5-C6H3(CH3)3]Mo(CO)3 (1). Mo(CO)6 (2.083 g, 7.92 mmol) and mesitylene (10 mL, 72 mmol) were added subsequently to a 100 mL 3 neck round-bottom flask along with a small magnetic stir bar. A sand bath was constructed and set to 50% power. A greased sidearm, stopcock, and 30 cm cold water condenser were attached to the round-bottom flask. A greased gas inlet was then attached to the condenser and connected to a bubbler. The condenser was not connected to a cold water source; it was used only to allow air to circulate. The sidearm was connected to a nitrogen source, and the system was allowed to degas for 5 minutes. The system was then put on the sand bath and the stir bar was spun at a moderate speed via a magnetic stirring instrument.
After 5 minutes, the solution in the round-bottom flask was not boiling as outlined, so the sand bath was turned up to 70% power. The sand bath was turned up to 85% another 5 minutes later. A rigorous boil was achieved when the sand bath was set to 95% 5 minutes after that. It was then set to 85% power in efforts to obtain a less extreme boil. After a total of 0.33 h of reflux, the solution was taken off the sand bath and allowed to cool to room temperature. When the apparatus was removed from the sand bath, it was dropped and roughly more than 60% of the solution was lost. The remaining solution cooled to a blackish yellow color.
The following and final procedures took place in the presence of air. Once cool, the solution was washed with 15 mL of hexane via suction filtration in a 15 mL frit. The solution was then washed with another 5 mL of hexane. About 10 mL CH2Cl2 was added to the blackish yellow powder precipitate remaining in the frit. The powder was washed with 25 to 30 mL of hexane and vacuum dried. This powder was discarded and the collected yellowish washings were rotovapped for about 0.33 h to obtain the desired product. The product was vacuum filtered, as it would not completely dry under the rotovap. The resulting yellowish powder was determined to be 1 (0.028 g, 1.18% yield based on the amount of Mo(CO)6 used). 1H NMR (CH2Cl2): δ 2.25 (s, -CH3), 5.23 (s, C-H). FTIR (ATR) ν(C-O) 1852 cm-1 (s, C-O linkage), ν(C-O) 1942 cm-1 (m, C-O linkage).
1,3,5-C6H3(CH3)3 (2). The 1H NMR spectrum of 2 was extrapolated from the literature.1 1H NMR (CHCl3): δ 2.25 (s, -CH3), 6.78 (s, C-H).
The reaction of Mo(CO)6 with mesitylene yielded 0.028 g of the product, [1,3,5-C6H3(CH3)3]Mo(CO)3. This translated to 0.09328 mmol, and thus a 1.18% yield based on the amount of Mo(CO)6 used, which was the limiting reagent in the reaction. Mo(CO)6 reacted to form the product in a 1:1 ratio, and 7.92 mmol of Mo(CO)6 was used to start, so that proportion was taken into account when calculating the percent yield. Proton NMR spectroscopy yielded a two peaks of interest. A peak found at δ 2.25 was indicative of methylhydrogens and a peak noted at δ 5.23 was suggestive of hydrogens attached directly to the aromatic ring.1 These peaks were noted with relative intensities of 9 to 3, respectively. Peaks seen at δ 7.25 and 1.54 were attributed to solvent and hexane, respectively. The 1H NMR spectrum of mesitylene in CHCl3 showed absorptions at δ 2.25 and 6.78 with relative intensities of 9 to 3, respectively.1 The peak at δ 2.25 hinted of methyl protons and the peak at δ 6.78 was suggestive of protons bonded directly the ring. The mass spectrum of [1,3,5-C6H3(CH3)3]Mo(CO)3 showed a peak at m/z = 302.0, which is nearly equal to the molar mass of said substance.1 The IR spectrum showed a strong peak around 1852 cm-1 of area 64.462 cm-1 indicative of antisymmetrical C-O stretching and a medium peak around 2942 cm-1 of area 9.111 cm-1 indicative of symmetrical C-O stretching. These areas were used to calculate a OC-Mo-CO bond angle of 108.32°.
The percent yield of [1,3,5-C6H3(CH3)3]Mo(CO)3 is very poor. Much of this quantitative shortcoming can be attributed to clumsiness, as much of the solution containing future product was lost when the reaction vessel was dropped. This not only resulted in a direct loss of solution, but also exposed the solution to air. The solution had been kept under nitrogen as to prevent decomposition of the products. This exposure to air undoubtedly had a contribution to the poor percent yield. The solution was also not heated as desired because it was difficult to control the sand bath. The solution was to be brought to a moderate boil, but could not be controlled to do so. It would not boil, then boiled rigorously a moment later. Attempts were made to subdue the boiling, but were unsuccessful. This overheating was probably not favorable for the reaction. More adept control of the sand bath would have resulted in a better yield of product.
Washing the product with excess amounts of hexane and CH2Cl2 also may have added to the loss of product. Excess washing would make it difficult to extract the product from the solution, as there would have been a relatively small amount of product compared to the amount of solution it was dissolved in. The solution could not be completely dried with the rotovap, which means there was an excess of hexane and/or mesitylene in the solution. Vacuum filtration then had to be used to collect the product, which was not ideal. Best case scenario, the rotovap would have completely dried the product and it would have been scraped out of the flask. Vacuum filtration gives a better chance for loss of product.
The product seemed pure as it produced clear 1H NMR and IR spectra readings. The 1H NMR spectrum shows a methyl peak at δ 2.25 and a C-H peak at δ 5.23, and the reagent in the reaction, mesitylene, also gives a peak at δ 2.25. This seems to confirm the structure and addition of the metal complex, as only the C-H peak was shifted downfield. The methyl protons are too shielded to be affected by the metal. The downfield shift is caused by backbonding of the carbonyls. IR spectroscopy revealed 3 C-O stretches, 2 of which were accounted for by a strong peak at 1852 cm-1 and the 3rd of which was accounted for by a medium peak at 1942 cm-1. Two peaks were seen because of symmetrical and antisymmetrical stretching of the carbonyls.1 The two antisymmetical modes have exactly identical absorption frequencies, and the symmetrical mode has a different absorption frequency than them, which means that a total of two peaks should be seen.1 The areas of these peaks, 64.462 cm-1 for antisymmetical stretch and 9.111 cm-1 for symmetrical stretch, allowed for discovery of the bond angle between the CO ligands. The calculated angle was 108.32°. This seems to make sense as [1,3,5-C6H3(CH3)3]Mo(CO)3 is predicted to be a tetrahedral complex and the expected bond angle for tetrahedral complexes is 109.5°, but in reality [1,3,5-C6H3(CH3)3]Mo(CO)3 acts more like an octahedral complex and the bond angle should be slightly less than 90°.1 This error could be due to excess solvent or impure an sample, which resulted in skewed IR spectrum readings and thus incorrect peak areas. The oxidation state and electron count of Mo in [1,3,5-C6H3(CH3)3]Mo(CO)3 are 0 and 6 electrons, so it is an 18 electron complex.
The main purpose of the experiment was to interpret the 1H NMR and IR spectra of [1,3,5-C6H3(CH3)3]Mo(CO)3 to confirm the product and to decipher the bond angle between the carbonyls. The 1H NMR spectrum of the product showed peaks at δ 2.25 and 5.23, while the 1H NMR spectrum of mesitylene gave peaks at δ 2.25 and 6.78, both in ratios of 9:3, respectively. This ratio seems to confirm methyl groups and single protons attached to the ring. The downfield shifting of the second peak is attributed to the coordination of mesitylene to the metal complex. The protons attached directly to the ring are affected by backbonding of the carbonyl groups. The methyl hydrogens are shielded, and thus are not affected by the metal complex. The IR spectrum yielded two peaks near 2000 cm-1. These two peaks account for 3 C-O stretches, 2 of which are accounted for by a strong peak at 1852 cm-1 and the 3rd of which are accounted for by a medium peak at 1942 cm-1. The strong peak accounts for antisymmetrical stretching of the carbonyls and the medium peak accounts for symmetrical stretching of the carbonyls. The areas of these peaks, 62.462 cm-1 and 9.111 cm-1 respectively, provide for a theoretical angle between the carbonyls of 108.32°. In reality, this angle should be only nearly 90°. The percent yield for the reaction was poor and could have improved with a more steady hand, more precise heating of the reagents, less exposure of the solution to air, and less solvent used for washing.
(1) Angelici, R. J.; Girolami, G. S.; Rachufuss T. B. Synthesis and Technique in Inorganic Chemistry: A Laboratory Manual; University Science Books: Sausilito, CA, 1999; pp 161-170.