Tinkering with Tin: Synthesis of SnCl(CH2C6H5)3 and SnCl4[OS(CH3)2]2
The reaction of tin with benzyl chloride under reflux yields SnCl(CH2C6H5)3 in low percent yield (around 15%) due to various factors. 1H NMR of SnCl(CH2C6H5)3 shows a singlet at δ 3.15 with satellites containing coupling constant between Sn-H of 78.1 Hz. The reaction of SnCl4 with OS(CH3)2 yields SnCl4[OS(CH3)2]2 with poor percent yield (around 140%) due to various factors. The IR spectrum of SnCl4[OS(CH3)2]2 shows a large S-O stretch at 899.7 cm-1, which is a lower frequency than the S-O stretch of 1042.5 cm-1 from the IR spectrum of OS(CH3)2, suggesting coordination on the metal complex at oxygen.
The reaction of tin with benzyl chloride yields the metal complex SnCl(CH2C6H5)3. The reaction specifically takes place in the following manner:
This is compound of interest because of the different isotopes of tin that naturally occur. In order to determine the structure of said substance from a 1H NMR spectrum, the peaks must be analyzed for satellites occurring from coupling of hydrogen to isotopes of tin. The observed spectrum is a combination of SnCl(CH2C6H5)3 molecules containing different isotopes of Sn, namely 117Sn and 119Sn, among others. The value for JSn-H between satellites can confirm the presence of SnCl(CH2C6H5)3. Mass spectrum can also be used to confirm the presence of SnCl(CH2C6H5)3 in the product by looking for a peak at m/z equal to the molecular mass of the substance (427 g/mol).
SnCl4[OS(CH3)2]2 can be synthesized from the following reaction:
It is a reaction of interest because it can be used to determine where OS(CH3)2 coordinates to Sn. Coordination of S to Sn would result in a higher frequency of S-O stretch on an IR spectrum than the frequency of the S-O stretch on the IR spectrum of OS(CH3)2 due to a strengthening of the S-O bond. Coordination of O to Sn would result in a lower frequency of S-O stretch because of a weakened S-O bond from the resonance form needed to secure that coordination.1
All syntheses were carried out in air and the reagents and solvents were purchased from commercial sources and used as received unless otherwise noted. The synthesis of SnCl(CH2C6H5)3 (1) and SnCl4[OS(CH3)2]2 (2) were based on reports published previously.1
SnCl(CH2C6H5)3 (1). 325 mash-Aldrich Sn powder (1.939 g, 16.34 mmol), 99% Aldrich benzyl chloride (6.0 mL, 52 mmol), and deionized H2O (4.0 mL) were added subsequently to a 50 mL round-bottom flask. A small stir bar was added then added to this solution. A sand bath was placed over a stir plate and a cold water condenser with greased joint was inserted into the round-bottom flask containing the solution. This connection was further secured with a keck clip. The round bottom flask was placed in the sand bath and the condenser was connected to the cold water source.
Once secure, the sand bath was set to 50% power and the stir bar was spun a shade over moderate speed. The solution was allowed to reflux for 2.75 h. Once reflux was complete, the 50 mL round-bottom flask was removed from the condenser and allowed to cool in an ice bath until it was cold. The liquid was decanted from the white precipitate and discarded. The white precipitate was saved. Ethyl acetate (10 mL) was added to the precipitate and this solution was heated using a sand bath at 65% power and stirred at moderate speed until the white solid had completely dissolved. The round-bottom flask was then taken off the heat and submerged in an ice bath for approximately 0.5 h to allow for recrystallization.
The stir bar was removed and a high vacuum was used for about 10 minutes to extract the extraneous liquid and leave a white powder. Several mL of ether were added to the round-bottom flask and a glass stir rod was used to break up the chunks of powder and dissolve it in the ether. This solution was then suction filtered with a 30 mL glass frit and allowed the dry. The precipitate collected was 1 (0.471 g, 13.49% based on the amount of Sn powder used). 1H NMR (CDCl3): δ 7.25 (s, CDCl3), 7.19 (s, Ph-H), 7.17 (s, Ph-H), 7.05 (s, Ph-H), 3.15 (s, JSn-H = 78.1 Hz, CH2).
SnCl4[OS(CH3)2]2 (2). SnCl4 (2.25 mL, 19 mmol), anhydrous diethyl ether (45 mL), DMSO (2.9 mL, 41 mmol), and ether (5 mL) were subsequently added to a 125 mL Erlenmeyer flask. The DMSO was added using a syringe for safety purposes while the other reagents were measured and added using graduated cylinders. This solution yielded a precipitate which was isolated by suction filtering the solution through a 50 mL glass frit. Several extra mL of ether were added to the Erlenmeyer flask to help aid in transfer of all the precipitate to the filter. Once dry, the white powder precipitate 2 was collected and weighed. 11.04 g (26.49 mmol) were recovered, giving a yield of 139.4%. FTIR (ATR) ν(S-O) 899.7 cm-1 (s S-O coordination to Sn).
OS(CH3)2 (3). The IR spectrum of (3) was taken by Dr. Graham. FTIR (ATR) ν(S-O) 1042.5 cm-1 (s, S-O linkage).
The reaction of Sn, benzyl chloride, and H2O yielded 0.471 g of the product, SnCl(CH2C6H5)3. This translated to 1.102 mmol, and thus was a 13.49% yield based on the amount of Sn used, which was the limiting reagent in the reaction. Sn reacted to form the product in a 2:1 ratio, and 16.34 mmol of Sn was used to start, so that proportion was taken into account when calculating the percent yield. Proton NMR yielded several peaks. A set of peaks were found at δ 7.19, 7.17, and 7.05 were indicative of phenyl resonances derived from protons on the phenyl ring of the product.1 A sharp peak located at δ 7.25 was due to the chloroform solvent 1. Coupling between Sn and H produced a singlet at δ 3.15 with satellites 1JSn-H equal to 78.1 Hz due to the presence of isotopes 117Sn and 119Sn.1 The mass spectrum of SnCl(CH2C6H5)3 showed a peak at m/z = 427.0,1 which is the molar mass of said substance.
The reaction of SnCl4 and DMSO yielded 11.04 g of product, SnCl4[OS(CH3)2]2. This translated to 26.49 mmol, and thus was a 139.4% yield. IR spectrum of SnCl4[OS(CH3)2]2 showed its largest peak at 899.7 cm-1 while the IR spectrum of OS(CH3)2 gave its largest peak at 1042.51 cm-1.
The percent yields for each product are less than stellar. During the synthesis of SnCl(CH2C6H5)3, letting the reagents reflux for a longer time period, closer to 3 hours, may have been beneficiary to resulting in more product. The reagents may not have all completely reacted. Some of the precipitate may have been accidentally removed during decanting after reflux, and much ethyl acetate may have been added to that precipitate. A minimal amount should have been used while trying to dissolve the white solid precipitate in the sand bath. The less ethyl acetate needed and used would have resulted in a better percent yield.
The fact that recrystallization did not seem take place as detailed1 and that a vacuum had to be used to dry the product most likely did not help the yield of product either. Product may have been lost during the vacuuming process. The glass frit used for filtering was not of the utmost quality, and so product may have escaped during that process, also. The product was also not completely dry, and gave a misleading mass measurement, meaning the percent yield is even lower than recorded. The product obtained did seem to give a clear 1H NMR spectra, meaning it was fairly pure. The coupling between Sn and H due to isotopes 117Sn and 119Sn 1 in the product is distinctly visible around δ 3.15 with a JSn-H of 78.1 Hz. The NMR chemical shifts of the Sn-CH2 protons and the C6H5 protons of SnCl(CH2C6H5)3 are so different because of the 117 and 119 isotopes of Sn. They affect the coupling with H, producing satellites. The C6H5 protons are not affected by this coupling. The literature states that peaks for phenyl resonance appear around δ 7 and chloroform appears around δ 7.24, which seems to validate the experimental values obtained (δ 7.05 to 7.19 for phenyl resonances ad δ 7.25 for chloroform).1
The percent yield of SnCl4[OS(CH3)2]2 is likely thrown off because the product was not dry when it was weighed. Several extra mL of ether were used for transport of the precipitate to the glass frit for filtering, which would most likely lead to loss of product. Not all of the precipitate was able to be transferred from the Erlenmeyer flask to the frit. Product was distinctly lost due to a seemingly defective frit, too. Attempts to refilter the filtrate were unsuccessful. The fact that the frit quality was poor meant that the product was not able to be dried well and thus carried extra weight. The product seemed to smoke away as it was allowed to dry further in exposure to air, and lost mass over time. This conundrum could not be adequately explained. IR spectrum of the product SnCl4[OS(CH3)2]2 shows its largest stretch at 899.7 cm-1, hinting that the coordination to Sn occurs at oxygen, as this frequency is lower than that of the largest stretch from the OS(CH3)2 spectrum (1042.51 cm-1). This is due to resonance and weakening of the S-O bond.1 The two spectra are similar save for the shifting of that one peak. The stretch at 1042.51 cm-1 is consistent with the value stated in the literature for S-O stretching (approximately 1100 cm-1).1
The main purposes of the experiments were to decipher the 1H NMR for SnCl(CH2C6H5)3 in order to determine JSn-H and to interpret the IR spectra of OS(CH3)2 and SnCl4[OS(CH3)2]2 to tell whether O or S coordinates to Sn. The JSn-H was found to be 78.1 Hz. The observed spectrum is a combination of SnCl(CH2C6H5)3 molecules containing different isotopes of Sn, namely 117Sn and 119Sn, among others. This is what makes the 1H NMR difficult to interpret, but the JSn-H value is validation of the presence of SnCl(CH2C6H5)3. The percent yield for that reaction was poor, due to varying factors. Longer reflux time, less ethyl acetate used during recrystallization, and better filtration techniques all would have contributed to an improved percent yield.
The IR spectrum of OS(CH3)2 showed its largest stretch around 1050 cm-1 while the IR spectrum of SnCl4[OS(CH3)2]2 gave its largest stretch around 900 cm-1. This lower frequency of stretch suggests that O and not S coordinates to the Sn. This is due in part to the resonance form needed to form the complex, which weakens the S-O bonding, and thus lowers the stretching frequency. The yield for the reaction to form SnCl4[OS(CH3)2]2 was also poor, which can be mostly attributed to poor filtering and drying techniques, along with the overuse of ether.
(1) “Synthesis and Techniques in Inorganic Chemistry,” Third Edition, G. S. Girolami, T.R. Rauchfuss, and R. Angelici, University Science Books, 1999.